Official ARM version: v5.4.0

    Add CMSIS V5.4.0, please refer to index.html available under \docs folder.

    Note: content of \CMSIS\Core\Include has been copied under \Include to keep the same structure
         used in existing projects, and thus avoid projects mass update
    Note: the following components have been removed from ARM original delivery (as not used in ST packages)
        - CMSIS_EW2018.pdf
        - .gitattributes
        - .gitignore
        - \Device
        - \CMSIS
           - \CoreValidation
		   - \DAP
           - \Documentation
           - \DoxyGen
           - \Driver
           - \Pack
           - \RTOS\CMSIS_RTOS_Tutorial.pdf
           - \RTOS\RTX
           - \RTOS\Template
           - \RTOS2\RTX
           - \Utilities
           - All ARM/GCC projects files are deleted from \DSP, \RTOS and \RTOS2

Change-Id: Ia026c3f0f0d016627a4fb5a9032852c33d24b4d3

25 files changed, 6776 insertions, 0 deletions

diff --git a/DSP/Source/MatrixFunctions/arm_mat_add_f32.c b/DSP/Source/MatrixFunctions/arm_mat_add_f32.c
new file mode 100644
index 0000000..4a54049
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_add_f32.c
@@ -0,0 +1,196 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_add_f32.c
+ * Description:  Floating-point matrix addition
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @defgroup MatrixAdd Matrix Addition
+ *
+ * Adds two matrices.
+ * \image html MatrixAddition.gif "Addition of two 3 x 3 matrices"
+ *
+ * The functions check to make sure that
+ * <code>pSrcA</code>, <code>pSrcB</code>, and <code>pDst</code> have the same
+ * number of rows and columns.
+ */
+
+/**
+ * @addtogroup MatrixAdd
+ * @{
+ */
+
+
+/**
+ * @brief Floating-point matrix addition.
+ * @param[in]       *pSrcA points to the first input matrix structure
+ * @param[in]       *pSrcB points to the second input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ */
+
+arm_status arm_mat_add_f32(
+  const arm_matrix_instance_f32 * pSrcA,
+  const arm_matrix_instance_f32 * pSrcB,
+  arm_matrix_instance_f32 * pDst)
+{
+  float32_t *pIn1 = pSrcA->pData;                /* input data matrix pointer A  */
+  float32_t *pIn2 = pSrcB->pData;                /* input data matrix pointer B  */
+  float32_t *pOut = pDst->pData;                 /* output data matrix pointer   */
+
+#if defined (ARM_MATH_DSP)
+
+  float32_t inA1, inA2, inB1, inB2, out1, out2;  /* temporary variables */
+
+#endif //      #if defined (ARM_MATH_DSP)
+
+  uint32_t numSamples;                           /* total number of elements in the matrix  */
+  uint32_t blkCnt;                               /* loop counters */
+  arm_status status;                             /* status of matrix addition */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numRows != pSrcB->numRows) ||
+     (pSrcA->numCols != pSrcB->numCols) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif
+  {
+
+    /* Total number of samples in the input matrix */
+    numSamples = (uint32_t) pSrcA->numRows * pSrcA->numCols;
+
+#if defined (ARM_MATH_DSP)
+
+    /* Loop unrolling */
+    blkCnt = numSamples >> 2U;
+
+    /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+     ** a second loop below computes the remaining 1 to 3 samples. */
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) + B(m,n) */
+      /* Add and then store the results in the destination buffer. */
+      /* Read values from source A */
+      inA1 = pIn1[0];
+
+      /* Read values from source B */
+      inB1 = pIn2[0];
+
+      /* Read values from source A */
+      inA2 = pIn1[1];
+
+      /* out = sourceA + sourceB */
+      out1 = inA1 + inB1;
+
+      /* Read values from source B */
+      inB2 = pIn2[1];
+
+      /* Read values from source A */
+      inA1 = pIn1[2];
+
+      /* out = sourceA + sourceB */
+      out2 = inA2 + inB2;
+
+      /* Read values from source B */
+      inB1 = pIn2[2];
+
+      /* Store result in destination */
+      pOut[0] = out1;
+      pOut[1] = out2;
+
+      /* Read values from source A */
+      inA2 = pIn1[3];
+
+      /* Read values from source B */
+      inB2 = pIn2[3];
+
+      /* out = sourceA + sourceB */
+      out1 = inA1 + inB1;
+
+      /* out = sourceA + sourceB */
+      out2 = inA2 + inB2;
+
+      /* Store result in destination */
+      pOut[2] = out1;
+
+      /* Store result in destination */
+      pOut[3] = out2;
+
+
+      /* update pointers to process next sampels */
+      pIn1 += 4U;
+      pIn2 += 4U;
+      pOut += 4U;
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+    /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
+     ** No loop unrolling is used. */
+    blkCnt = numSamples % 0x4U;
+
+#else
+
+    /* Run the below code for Cortex-M0 */
+
+    /* Initialize blkCnt with number of samples */
+    blkCnt = numSamples;
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) + B(m,n) */
+      /* Add and then store the results in the destination buffer. */
+      *pOut++ = (*pIn1++) + (*pIn2++);
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+    /* set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixAdd group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_add_q15.c b/DSP/Source/MatrixFunctions/arm_mat_add_q15.c
new file mode 100644
index 0000000..896e60c
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_add_q15.c
@@ -0,0 +1,151 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_add_q15.c
+ * Description:  Q15 matrix addition
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixAdd
+ * @{
+ */
+
+/**
+ * @brief Q15 matrix addition.
+ * @param[in]       *pSrcA points to the first input matrix structure
+ * @param[in]       *pSrcB points to the second input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * <b>Scaling and Overflow Behavior:</b>
+ * \par
+ * The function uses saturating arithmetic.
+ * Results outside of the allowable Q15 range [0x8000 0x7FFF] will be saturated.
+ */
+
+arm_status arm_mat_add_q15(
+  const arm_matrix_instance_q15 * pSrcA,
+  const arm_matrix_instance_q15 * pSrcB,
+  arm_matrix_instance_q15 * pDst)
+{
+  q15_t *pInA = pSrcA->pData;                    /* input data matrix pointer A  */
+  q15_t *pInB = pSrcB->pData;                    /* input data matrix pointer B */
+  q15_t *pOut = pDst->pData;                     /* output data matrix pointer */
+  uint16_t numSamples;                           /* total number of elements in the matrix  */
+  uint32_t blkCnt;                               /* loop counters  */
+  arm_status status;                             /* status of matrix addition  */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numRows != pSrcB->numRows) ||
+     (pSrcA->numCols != pSrcB->numCols) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* Total number of samples in the input matrix */
+    numSamples = (uint16_t) (pSrcA->numRows * pSrcA->numCols);
+
+#if defined (ARM_MATH_DSP)
+
+    /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+    /* Loop unrolling */
+    blkCnt = (uint32_t) numSamples >> 2U;
+
+    /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+     ** a second loop below computes the remaining 1 to 3 samples. */
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) + B(m,n) */
+      /* Add, Saturate and then store the results in the destination buffer. */
+      *__SIMD32(pOut)++ = __QADD16(*__SIMD32(pInA)++, *__SIMD32(pInB)++);
+      *__SIMD32(pOut)++ = __QADD16(*__SIMD32(pInA)++, *__SIMD32(pInB)++);
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+    /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
+     ** No loop unrolling is used. */
+    blkCnt = (uint32_t) numSamples % 0x4U;
+
+    /* q15 pointers of input and output are initialized */
+
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) + B(m,n) */
+      /* Add, Saturate and then store the results in the destination buffer. */
+      *pOut++ = (q15_t) __QADD16(*pInA++, *pInB++);
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+#else
+
+    /* Run the below code for Cortex-M0 */
+
+    /* Initialize blkCnt with number of samples */
+    blkCnt = (uint32_t) numSamples;
+
+
+    /* q15 pointers of input and output are initialized */
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) + B(m,n) */
+      /* Add, Saturate and then store the results in the destination buffer. */
+      *pOut++ = (q15_t) __SSAT(((q31_t) * pInA++ + *pInB++), 16);
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    /* set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixAdd group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_add_q31.c b/DSP/Source/MatrixFunctions/arm_mat_add_q31.c
new file mode 100644
index 0000000..f230ad2
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_add_q31.c
@@ -0,0 +1,195 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_add_q31.c
+ * Description:  Q31 matrix addition
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixAdd
+ * @{
+ */
+
+/**
+ * @brief Q31 matrix addition.
+ * @param[in]       *pSrcA points to the first input matrix structure
+ * @param[in]       *pSrcB points to the second input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * <b>Scaling and Overflow Behavior:</b>
+ * \par
+ * The function uses saturating arithmetic.
+ * Results outside of the allowable Q31 range [0x80000000 0x7FFFFFFF] will be saturated.
+ */
+
+arm_status arm_mat_add_q31(
+  const arm_matrix_instance_q31 * pSrcA,
+  const arm_matrix_instance_q31 * pSrcB,
+  arm_matrix_instance_q31 * pDst)
+{
+  q31_t *pIn1 = pSrcA->pData;                    /* input data matrix pointer A */
+  q31_t *pIn2 = pSrcB->pData;                    /* input data matrix pointer B */
+  q31_t *pOut = pDst->pData;                     /* output data matrix pointer */
+  q31_t inA1, inB1;                              /* temporary variables */
+
+#if defined (ARM_MATH_DSP)
+
+  q31_t inA2, inB2;                              /* temporary variables */
+  q31_t out1, out2;                              /* temporary variables */
+
+#endif //      #if defined (ARM_MATH_DSP)
+
+  uint32_t numSamples;                           /* total number of elements in the matrix  */
+  uint32_t blkCnt;                               /* loop counters */
+  arm_status status;                             /* status of matrix addition */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numRows != pSrcB->numRows) ||
+     (pSrcA->numCols != pSrcB->numCols) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif
+  {
+    /* Total number of samples in the input matrix */
+    numSamples = (uint32_t) pSrcA->numRows * pSrcA->numCols;
+
+#if defined (ARM_MATH_DSP)
+
+    /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+    /* Loop Unrolling */
+    blkCnt = numSamples >> 2U;
+
+
+    /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+     ** a second loop below computes the remaining 1 to 3 samples. */
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) + B(m,n) */
+      /* Add, saturate and then store the results in the destination buffer. */
+      /* Read values from source A */
+      inA1 = pIn1[0];
+
+      /* Read values from source B */
+      inB1 = pIn2[0];
+
+      /* Read values from source A */
+      inA2 = pIn1[1];
+
+      /* Add and saturate */
+      out1 = __QADD(inA1, inB1);
+
+      /* Read values from source B */
+      inB2 = pIn2[1];
+
+      /* Read values from source A */
+      inA1 = pIn1[2];
+
+      /* Add and saturate */
+      out2 = __QADD(inA2, inB2);
+
+      /* Read values from source B */
+      inB1 = pIn2[2];
+
+      /* Store result in destination */
+      pOut[0] = out1;
+      pOut[1] = out2;
+
+      /* Read values from source A */
+      inA2 = pIn1[3];
+
+      /* Read values from source B */
+      inB2 = pIn2[3];
+
+      /* Add and saturate */
+      out1 = __QADD(inA1, inB1);
+      out2 = __QADD(inA2, inB2);
+
+      /* Store result in destination */
+      pOut[2] = out1;
+      pOut[3] = out2;
+
+      /* update pointers to process next sampels */
+      pIn1 += 4U;
+      pIn2 += 4U;
+      pOut += 4U;
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+    /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
+     ** No loop unrolling is used. */
+    blkCnt = numSamples % 0x4U;
+
+#else
+
+    /* Run the below code for Cortex-M0 */
+
+    /* Initialize blkCnt with number of samples */
+    blkCnt = numSamples;
+
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) + B(m,n) */
+      /* Add, saturate and then store the results in the destination buffer. */
+      inA1 = *pIn1++;
+      inB1 = *pIn2++;
+
+      inA1 = __QADD(inA1, inB1);
+
+      /* Decrement the loop counter */
+      blkCnt--;
+
+      *pOut++ = inA1;
+
+    }
+
+    /* set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixAdd group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_cmplx_mult_f32.c b/DSP/Source/MatrixFunctions/arm_mat_cmplx_mult_f32.c
new file mode 100644
index 0000000..bb8341e
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_cmplx_mult_f32.c
@@ -0,0 +1,272 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_cmplx_mult_f32.c
+ * Description:  Floating-point matrix multiplication
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @defgroup CmplxMatrixMult  Complex Matrix Multiplication
+ *
+ * Complex Matrix multiplication is only defined if the number of columns of the
+ * first matrix equals the number of rows of the second matrix.
+ * Multiplying an <code>M x N</code> matrix with an <code>N x P</code> matrix results
+ * in an <code>M x P</code> matrix.
+ * When matrix size checking is enabled, the functions check: (1) that the inner dimensions of
+ * <code>pSrcA</code> and <code>pSrcB</code> are equal; and (2) that the size of the output
+ * matrix equals the outer dimensions of <code>pSrcA</code> and <code>pSrcB</code>.
+ */
+
+
+/**
+ * @addtogroup CmplxMatrixMult
+ * @{
+ */
+
+/**
+ * @brief Floating-point Complex matrix multiplication.
+ * @param[in]       *pSrcA points to the first input complex matrix structure
+ * @param[in]       *pSrcB points to the second input complex matrix structure
+ * @param[out]      *pDst points to output complex matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ */
+
+arm_status arm_mat_cmplx_mult_f32(
+  const arm_matrix_instance_f32 * pSrcA,
+  const arm_matrix_instance_f32 * pSrcB,
+  arm_matrix_instance_f32 * pDst)
+{
+  float32_t *pIn1 = pSrcA->pData;                /* input data matrix pointer A */
+  float32_t *pIn2 = pSrcB->pData;                /* input data matrix pointer B */
+  float32_t *pInA = pSrcA->pData;                /* input data matrix pointer A  */
+  float32_t *pOut = pDst->pData;                 /* output data matrix pointer */
+  float32_t *px;                                 /* Temporary output data matrix pointer */
+  uint16_t numRowsA = pSrcA->numRows;            /* number of rows of input matrix A */
+  uint16_t numColsB = pSrcB->numCols;            /* number of columns of input matrix B */
+  uint16_t numColsA = pSrcA->numCols;            /* number of columns of input matrix A */
+  float32_t sumReal1, sumImag1;                  /* accumulator */
+  float32_t a0, b0, c0, d0;
+  float32_t a1, b1, c1, d1;
+  float32_t sumReal2, sumImag2;                  /* accumulator */
+
+
+  /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+  uint16_t col, i = 0U, j, row = numRowsA, colCnt;      /* loop counters */
+  arm_status status;                             /* status of matrix multiplication */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numCols != pSrcB->numRows) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
+  {
+
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*      #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
+    /* row loop */
+    do
+    {
+      /* Output pointer is set to starting address of the row being processed */
+      px = pOut + 2 * i;
+
+      /* For every row wise process, the column loop counter is to be initiated */
+      col = numColsB;
+
+      /* For every row wise process, the pIn2 pointer is set
+       ** to the starting address of the pSrcB data */
+      pIn2 = pSrcB->pData;
+
+      j = 0U;
+
+      /* column loop */
+      do
+      {
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sumReal1 = 0.0f;
+        sumImag1 = 0.0f;
+
+        sumReal2 = 0.0f;
+        sumImag2 = 0.0f;
+
+        /* Initiate the pointer pIn1 to point to the starting address of the column being processed */
+        pIn1 = pInA;
+
+        /* Apply loop unrolling and compute 4 MACs simultaneously. */
+        colCnt = numColsA >> 2;
+
+        /* matrix multiplication        */
+        while (colCnt > 0U)
+        {
+
+          /* Reading real part of complex matrix A */
+          a0 = *pIn1;
+
+          /* Reading real part of complex matrix B */
+          c0 = *pIn2;
+
+          /* Reading imaginary part of complex matrix A */
+          b0 = *(pIn1 + 1U);
+
+          /* Reading imaginary part of complex matrix B */
+          d0 = *(pIn2 + 1U);
+
+          sumReal1 += a0 * c0;
+          sumImag1 += b0 * c0;
+
+          pIn1 += 2U;
+          pIn2 += 2 * numColsB;
+
+          sumReal2 -= b0 * d0;
+          sumImag2 += a0 * d0;
+
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+
+          a1 = *pIn1;
+          c1 = *pIn2;
+
+          b1 = *(pIn1 + 1U);
+          d1 = *(pIn2 + 1U);
+
+          sumReal1 += a1 * c1;
+          sumImag1 += b1 * c1;
+
+          pIn1 += 2U;
+          pIn2 += 2 * numColsB;
+
+          sumReal2 -= b1 * d1;
+          sumImag2 += a1 * d1;
+
+          a0 = *pIn1;
+          c0 = *pIn2;
+
+          b0 = *(pIn1 + 1U);
+          d0 = *(pIn2 + 1U);
+
+          sumReal1 += a0 * c0;
+          sumImag1 += b0 * c0;
+
+          pIn1 += 2U;
+          pIn2 += 2 * numColsB;
+
+          sumReal2 -= b0 * d0;
+          sumImag2 += a0 * d0;
+
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+
+          a1 = *pIn1;
+          c1 = *pIn2;
+
+          b1 = *(pIn1 + 1U);
+          d1 = *(pIn2 + 1U);
+
+          sumReal1 += a1 * c1;
+          sumImag1 += b1 * c1;
+
+          pIn1 += 2U;
+          pIn2 += 2 * numColsB;
+
+          sumReal2 -= b1 * d1;
+          sumImag2 += a1 * d1;
+
+          /* Decrement the loop count */
+          colCnt--;
+        }
+
+        /* If the columns of pSrcA is not a multiple of 4, compute any remaining MACs here.
+         ** No loop unrolling is used. */
+        colCnt = numColsA % 0x4U;
+
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          a1 = *pIn1;
+          c1 = *pIn2;
+
+          b1 = *(pIn1 + 1U);
+          d1 = *(pIn2 + 1U);
+
+          sumReal1 += a1 * c1;
+          sumImag1 += b1 * c1;
+
+          pIn1 += 2U;
+          pIn2 += 2 * numColsB;
+
+          sumReal2 -= b1 * d1;
+          sumImag2 += a1 * d1;
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        sumReal1 += sumReal2;
+        sumImag1 += sumImag2;
+
+        /* Store the result in the destination buffer */
+        *px++ = sumReal1;
+        *px++ = sumImag1;
+
+        /* Update the pointer pIn2 to point to the  starting address of the next column */
+        j++;
+        pIn2 = pSrcB->pData + 2U * j;
+
+        /* Decrement the column loop counter */
+        col--;
+
+      } while (col > 0U);
+
+      /* Update the pointer pInA to point to the  starting address of the next row */
+      i = i + numColsB;
+      pInA = pInA + 2 * numColsA;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);
+
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixMult group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_cmplx_mult_q15.c b/DSP/Source/MatrixFunctions/arm_mat_cmplx_mult_q15.c
new file mode 100644
index 0000000..5dee79c
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_cmplx_mult_q15.c
@@ -0,0 +1,413 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_cmplx_mat_mult_q15.c
+ * Description:  Q15 complex matrix multiplication
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup CmplxMatrixMult
+ * @{
+ */
+
+
+/**
+ * @brief Q15 Complex matrix multiplication
+ * @param[in]       *pSrcA points to the first input complex matrix structure
+ * @param[in]       *pSrcB points to the second input complex matrix structure
+ * @param[out]      *pDst points to output complex matrix structure
+ * @param[in]		*pScratch points to the array for storing intermediate results
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * \par Conditions for optimum performance
+ *  Input, output and state buffers should be aligned by 32-bit
+ *
+ * \par Restrictions
+ *  If the silicon does not support unaligned memory access enable the macro UNALIGNED_SUPPORT_DISABLE
+ *	In this case input, output, scratch buffers should be aligned by 32-bit
+ *
+ * @details
+ * <b>Scaling and Overflow Behavior:</b>
+ *
+ * \par
+ * The function is implemented using a 64-bit internal accumulator. The inputs to the
+ * multiplications are in 1.15 format and multiplications yield a 2.30 result.
+ * The 2.30 intermediate
+ * results are accumulated in a 64-bit accumulator in 34.30 format. This approach
+ * provides 33 guard bits and there is no risk of overflow. The 34.30 result is then
+ * truncated to 34.15 format by discarding the low 15 bits and then saturated to
+ * 1.15 format.
+ *
+ * \par
+ * Refer to <code>arm_mat_mult_fast_q15()</code> for a faster but less precise version of this function.
+ *
+ */
+
+
+
+
+arm_status arm_mat_cmplx_mult_q15(
+  const arm_matrix_instance_q15 * pSrcA,
+  const arm_matrix_instance_q15 * pSrcB,
+  arm_matrix_instance_q15 * pDst,
+  q15_t * pScratch)
+{
+  /* accumulator */
+  q15_t *pSrcBT = pScratch;                      /* input data matrix pointer for transpose */
+  q15_t *pInA = pSrcA->pData;                    /* input data matrix pointer A of Q15 type */
+  q15_t *pInB = pSrcB->pData;                    /* input data matrix pointer B of Q15 type */
+  q15_t *px;                                     /* Temporary output data matrix pointer */
+  uint16_t numRowsA = pSrcA->numRows;            /* number of rows of input matrix A    */
+  uint16_t numColsB = pSrcB->numCols;            /* number of columns of input matrix B */
+  uint16_t numColsA = pSrcA->numCols;            /* number of columns of input matrix A */
+  uint16_t numRowsB = pSrcB->numRows;            /* number of rows of input matrix A    */
+  uint16_t col, i = 0U, row = numRowsB, colCnt;  /* loop counters */
+  arm_status status;                             /* status of matrix multiplication */
+  q63_t sumReal, sumImag;
+
+#ifdef UNALIGNED_SUPPORT_DISABLE
+  q15_t in;                                      /* Temporary variable to hold the input value */
+  q15_t a, b, c, d;
+#else
+  q31_t in;                                      /* Temporary variable to hold the input value */
+  q31_t prod1, prod2;
+  q31_t pSourceA, pSourceB;
+#endif
+
+#ifdef ARM_MATH_MATRIX_CHECK
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numCols != pSrcB->numRows) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif
+  {
+    /* Matrix transpose */
+    do
+    {
+      /* Apply loop unrolling and exchange the columns with row elements */
+      col = numColsB >> 2;
+
+      /* The pointer px is set to starting address of the column being processed */
+      px = pSrcBT + i;
+
+      /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+       ** a second loop below computes the remaining 1 to 3 samples. */
+      while (col > 0U)
+      {
+#ifdef UNALIGNED_SUPPORT_DISABLE
+        /* Read two elements from the row */
+        in = *pInB++;
+        *px = in;
+        in = *pInB++;
+        px[1] = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB * 2;
+
+        /* Read two elements from the row */
+        in = *pInB++;
+        *px = in;
+        in = *pInB++;
+        px[1] = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB * 2;
+
+        /* Read two elements from the row */
+        in = *pInB++;
+        *px = in;
+        in = *pInB++;
+        px[1] = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB * 2;
+
+        /* Read two elements from the row */
+        in = *pInB++;
+        *px = in;
+        in = *pInB++;
+        px[1] = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB * 2;
+
+        /* Decrement the column loop counter */
+        col--;
+      }
+
+      /* If the columns of pSrcB is not a multiple of 4, compute any remaining output samples here.
+       ** No loop unrolling is used. */
+      col = numColsB % 0x4U;
+
+      while (col > 0U)
+      {
+        /* Read two elements from the row */
+        in = *pInB++;
+        *px = in;
+        in = *pInB++;
+        px[1] = in;
+#else
+
+        /* Read two elements from the row */
+        in = *__SIMD32(pInB)++;
+
+        *__SIMD32(px) = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB * 2;
+
+
+        /* Read two elements from the row */
+        in = *__SIMD32(pInB)++;
+
+        *__SIMD32(px) = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB * 2;
+
+        /* Read two elements from the row */
+        in = *__SIMD32(pInB)++;
+
+        *__SIMD32(px) = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB * 2;
+
+        /* Read two elements from the row */
+        in = *__SIMD32(pInB)++;
+
+        *__SIMD32(px) = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB * 2;
+
+        /* Decrement the column loop counter */
+        col--;
+      }
+
+      /* If the columns of pSrcB is not a multiple of 4, compute any remaining output samples here.
+       ** No loop unrolling is used. */
+      col = numColsB % 0x4U;
+
+      while (col > 0U)
+      {
+        /* Read two elements from the row */
+        in = *__SIMD32(pInB)++;
+
+        *__SIMD32(px) = in;
+#endif
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB * 2;
+
+        /* Decrement the column loop counter */
+        col--;
+      }
+
+      i = i + 2U;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);
+
+    /* Reset the variables for the usage in the following multiplication process */
+    row = numRowsA;
+    i = 0U;
+    px = pDst->pData;
+
+    /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
+    /* row loop */
+    do
+    {
+      /* For every row wise process, the column loop counter is to be initiated */
+      col = numColsB;
+
+      /* For every row wise process, the pIn2 pointer is set
+       ** to the starting address of the transposed pSrcB data */
+      pInB = pSrcBT;
+
+      /* column loop */
+      do
+      {
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sumReal = 0;
+        sumImag = 0;
+
+        /* Apply loop unrolling and compute 2 MACs simultaneously. */
+        colCnt = numColsA >> 1;
+
+        /* Initiate the pointer pIn1 to point to the starting address of the column being processed */
+        pInA = pSrcA->pData + i * 2;
+
+
+        /* matrix multiplication */
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+
+#ifdef UNALIGNED_SUPPORT_DISABLE
+
+          /* read real and imag values from pSrcA buffer */
+          a = *pInA;
+          b = *(pInA + 1U);
+          /* read real and imag values from pSrcB buffer */
+          c = *pInB;
+          d = *(pInB + 1U);
+
+          /* Multiply and Accumlates */
+          sumReal += (q31_t) a *c;
+          sumImag += (q31_t) a *d;
+          sumReal -= (q31_t) b *d;
+          sumImag += (q31_t) b *c;
+
+          /* read next real and imag values from pSrcA buffer */
+          a = *(pInA + 2U);
+          b = *(pInA + 3U);
+          /* read next real and imag values from pSrcB buffer */
+          c = *(pInB + 2U);
+          d = *(pInB + 3U);
+
+          /* update pointer */
+          pInA += 4U;
+
+          /* Multiply and Accumlates */
+          sumReal += (q31_t) a *c;
+          sumImag += (q31_t) a *d;
+          sumReal -= (q31_t) b *d;
+          sumImag += (q31_t) b *c;
+          /* update pointer */
+          pInB += 4U;
+#else
+          /* read real and imag values from pSrcA and pSrcB buffer */
+          pSourceA = *__SIMD32(pInA)++;
+          pSourceB = *__SIMD32(pInB)++;
+
+          /* Multiply and Accumlates */
+#ifdef ARM_MATH_BIG_ENDIAN
+          prod1 = -__SMUSD(pSourceA, pSourceB);
+#else
+          prod1 = __SMUSD(pSourceA, pSourceB);
+#endif
+          prod2 = __SMUADX(pSourceA, pSourceB);
+          sumReal += (q63_t) prod1;
+          sumImag += (q63_t) prod2;
+
+          /* read real and imag values from pSrcA and pSrcB buffer */
+          pSourceA = *__SIMD32(pInA)++;
+          pSourceB = *__SIMD32(pInB)++;
+
+          /* Multiply and Accumlates */
+#ifdef ARM_MATH_BIG_ENDIAN
+          prod1 = -__SMUSD(pSourceA, pSourceB);
+#else
+          prod1 = __SMUSD(pSourceA, pSourceB);
+#endif
+          prod2 = __SMUADX(pSourceA, pSourceB);
+          sumReal += (q63_t) prod1;
+          sumImag += (q63_t) prod2;
+
+#endif /*      #ifdef UNALIGNED_SUPPORT_DISABLE */
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        /* process odd column samples */
+        if ((numColsA & 0x1U) > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+
+#ifdef UNALIGNED_SUPPORT_DISABLE
+
+          /* read real and imag values from pSrcA and pSrcB buffer */
+          a = *pInA++;
+          b = *pInA++;
+          c = *pInB++;
+          d = *pInB++;
+
+          /* Multiply and Accumlates */
+          sumReal += (q31_t) a *c;
+          sumImag += (q31_t) a *d;
+          sumReal -= (q31_t) b *d;
+          sumImag += (q31_t) b *c;
+
+#else
+          /* read real and imag values from pSrcA and pSrcB buffer */
+          pSourceA = *__SIMD32(pInA)++;
+          pSourceB = *__SIMD32(pInB)++;
+
+          /* Multiply and Accumlates */
+#ifdef ARM_MATH_BIG_ENDIAN
+          prod1 = -__SMUSD(pSourceA, pSourceB);
+#else
+          prod1 = __SMUSD(pSourceA, pSourceB);
+#endif
+          prod2 = __SMUADX(pSourceA, pSourceB);
+          sumReal += (q63_t) prod1;
+          sumImag += (q63_t) prod2;
+
+#endif /*      #ifdef UNALIGNED_SUPPORT_DISABLE */
+
+        }
+
+        /* Saturate and store the result in the destination buffer */
+
+        *px++ = (q15_t) (__SSAT(sumReal >> 15, 16));
+        *px++ = (q15_t) (__SSAT(sumImag >> 15, 16));
+
+        /* Decrement the column loop counter */
+        col--;
+
+      } while (col > 0U);
+
+      i = i + numColsA;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);
+
+    /* set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixMult group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_cmplx_mult_q31.c b/DSP/Source/MatrixFunctions/arm_mat_cmplx_mult_q31.c
new file mode 100644
index 0000000..65cbb66
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_cmplx_mult_q31.c
@@ -0,0 +1,282 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_cmplx_mult_q31.c
+ * Description:  Floating-point matrix multiplication
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup CmplxMatrixMult
+ * @{
+ */
+
+/**
+ * @brief Q31 Complex matrix multiplication
+ * @param[in]       *pSrcA points to the first input complex matrix structure
+ * @param[in]       *pSrcB points to the second input complex matrix structure
+ * @param[out]      *pDst points to output complex matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * @details
+ * <b>Scaling and Overflow Behavior:</b>
+ *
+ * \par
+ * The function is implemented using an internal 64-bit accumulator.
+ * The accumulator has a 2.62 format and maintains full precision of the intermediate
+ * multiplication results but provides only a single guard bit. There is no saturation
+ * on intermediate additions. Thus, if the accumulator overflows it wraps around and
+ * distorts the result. The input signals should be scaled down to avoid intermediate
+ * overflows. The input is thus scaled down by log2(numColsA) bits
+ * to avoid overflows, as a total of numColsA additions are performed internally.
+ * The 2.62 accumulator is right shifted by 31 bits and saturated to 1.31 format to yield the final result.
+ *
+ *
+ */
+
+arm_status arm_mat_cmplx_mult_q31(
+  const arm_matrix_instance_q31 * pSrcA,
+  const arm_matrix_instance_q31 * pSrcB,
+  arm_matrix_instance_q31 * pDst)
+{
+  q31_t *pIn1 = pSrcA->pData;                    /* input data matrix pointer A */
+  q31_t *pIn2 = pSrcB->pData;                    /* input data matrix pointer B */
+  q31_t *pInA = pSrcA->pData;                    /* input data matrix pointer A  */
+  q31_t *pOut = pDst->pData;                     /* output data matrix pointer */
+  q31_t *px;                                     /* Temporary output data matrix pointer */
+  uint16_t numRowsA = pSrcA->numRows;            /* number of rows of input matrix A */
+  uint16_t numColsB = pSrcB->numCols;            /* number of columns of input matrix B */
+  uint16_t numColsA = pSrcA->numCols;            /* number of columns of input matrix A */
+  q63_t sumReal1, sumImag1;                      /* accumulator */
+  q31_t a0, b0, c0, d0;
+  q31_t a1, b1, c1, d1;
+
+
+  /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+  uint16_t col, i = 0U, j, row = numRowsA, colCnt;      /* loop counters */
+  arm_status status;                             /* status of matrix multiplication */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numCols != pSrcB->numRows) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
+  {
+
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*      #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
+    /* row loop */
+    do
+    {
+      /* Output pointer is set to starting address of the row being processed */
+      px = pOut + 2 * i;
+
+      /* For every row wise process, the column loop counter is to be initiated */
+      col = numColsB;
+
+      /* For every row wise process, the pIn2 pointer is set
+       ** to the starting address of the pSrcB data */
+      pIn2 = pSrcB->pData;
+
+      j = 0U;
+
+      /* column loop */
+      do
+      {
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sumReal1 = 0.0;
+        sumImag1 = 0.0;
+
+        /* Initiate the pointer pIn1 to point to the starting address of the column being processed */
+        pIn1 = pInA;
+
+        /* Apply loop unrolling and compute 4 MACs simultaneously. */
+        colCnt = numColsA >> 2;
+
+        /* matrix multiplication        */
+        while (colCnt > 0U)
+        {
+
+          /* Reading real part of complex matrix A */
+          a0 = *pIn1;
+
+          /* Reading real part of complex matrix B */
+          c0 = *pIn2;
+
+          /* Reading imaginary part of complex matrix A */
+          b0 = *(pIn1 + 1U);
+
+          /* Reading imaginary part of complex matrix B */
+          d0 = *(pIn2 + 1U);
+
+          /* Multiply and Accumlates */
+          sumReal1 += (q63_t) a0 *c0;
+          sumImag1 += (q63_t) b0 *c0;
+
+          /* update pointers */
+          pIn1 += 2U;
+          pIn2 += 2 * numColsB;
+
+          /* Multiply and Accumlates */
+          sumReal1 -= (q63_t) b0 *d0;
+          sumImag1 += (q63_t) a0 *d0;
+
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+
+          /* read real and imag values from pSrcA and pSrcB buffer */
+          a1 = *pIn1;
+          c1 = *pIn2;
+          b1 = *(pIn1 + 1U);
+          d1 = *(pIn2 + 1U);
+
+          /* Multiply and Accumlates */
+          sumReal1 += (q63_t) a1 *c1;
+          sumImag1 += (q63_t) b1 *c1;
+
+          /* update pointers */
+          pIn1 += 2U;
+          pIn2 += 2 * numColsB;
+
+          /* Multiply and Accumlates */
+          sumReal1 -= (q63_t) b1 *d1;
+          sumImag1 += (q63_t) a1 *d1;
+
+          a0 = *pIn1;
+          c0 = *pIn2;
+
+          b0 = *(pIn1 + 1U);
+          d0 = *(pIn2 + 1U);
+
+          /* Multiply and Accumlates */
+          sumReal1 += (q63_t) a0 *c0;
+          sumImag1 += (q63_t) b0 *c0;
+
+          /* update pointers */
+          pIn1 += 2U;
+          pIn2 += 2 * numColsB;
+
+          /* Multiply and Accumlates */
+          sumReal1 -= (q63_t) b0 *d0;
+          sumImag1 += (q63_t) a0 *d0;
+
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+
+          a1 = *pIn1;
+          c1 = *pIn2;
+
+          b1 = *(pIn1 + 1U);
+          d1 = *(pIn2 + 1U);
+
+          /* Multiply and Accumlates */
+          sumReal1 += (q63_t) a1 *c1;
+          sumImag1 += (q63_t) b1 *c1;
+
+          /* update pointers */
+          pIn1 += 2U;
+          pIn2 += 2 * numColsB;
+
+          /* Multiply and Accumlates */
+          sumReal1 -= (q63_t) b1 *d1;
+          sumImag1 += (q63_t) a1 *d1;
+
+          /* Decrement the loop count */
+          colCnt--;
+        }
+
+        /* If the columns of pSrcA is not a multiple of 4, compute any remaining MACs here.
+         ** No loop unrolling is used. */
+        colCnt = numColsA % 0x4U;
+
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          a1 = *pIn1;
+          c1 = *pIn2;
+
+          b1 = *(pIn1 + 1U);
+          d1 = *(pIn2 + 1U);
+
+          /* Multiply and Accumlates */
+          sumReal1 += (q63_t) a1 *c1;
+          sumImag1 += (q63_t) b1 *c1;
+
+          /* update pointers */
+          pIn1 += 2U;
+          pIn2 += 2 * numColsB;
+
+          /* Multiply and Accumlates */
+          sumReal1 -= (q63_t) b1 *d1;
+          sumImag1 += (q63_t) a1 *d1;
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        /* Store the result in the destination buffer */
+        *px++ = (q31_t) clip_q63_to_q31(sumReal1 >> 31);
+        *px++ = (q31_t) clip_q63_to_q31(sumImag1 >> 31);
+
+        /* Update the pointer pIn2 to point to the  starting address of the next column */
+        j++;
+        pIn2 = pSrcB->pData + 2U * j;
+
+        /* Decrement the column loop counter */
+        col--;
+
+      } while (col > 0U);
+
+      /* Update the pointer pInA to point to the  starting address of the next row */
+      i = i + numColsB;
+      pInA = pInA + 2 * numColsA;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);
+
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixMult group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_init_f32.c b/DSP/Source/MatrixFunctions/arm_mat_init_f32.c
new file mode 100644
index 0000000..783f7be
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_init_f32.c
@@ -0,0 +1,76 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_init_f32.c
+ * Description:  Floating-point matrix initialization
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @defgroup MatrixInit Matrix Initialization
+ *
+ * Initializes the underlying matrix data structure.
+ * The functions set the <code>numRows</code>,
+ * <code>numCols</code>, and <code>pData</code> fields
+ * of the matrix data structure.
+ */
+
+/**
+ * @addtogroup MatrixInit
+ * @{
+ */
+
+/**
+   * @brief  Floating-point matrix initialization.
+   * @param[in,out] *S             points to an instance of the floating-point matrix structure.
+   * @param[in]     nRows          number of rows in the matrix.
+   * @param[in]     nColumns       number of columns in the matrix.
+   * @param[in]     *pData	   points to the matrix data array.
+   * @return        none
+   */
+
+void arm_mat_init_f32(
+  arm_matrix_instance_f32 * S,
+  uint16_t nRows,
+  uint16_t nColumns,
+  float32_t * pData)
+{
+  /* Assign Number of Rows */
+  S->numRows = nRows;
+
+  /* Assign Number of Columns */
+  S->numCols = nColumns;
+
+  /* Assign Data pointer */
+  S->pData = pData;
+}
+
+/**
+ * @} end of MatrixInit group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_init_q15.c b/DSP/Source/MatrixFunctions/arm_mat_init_q15.c
new file mode 100644
index 0000000..08da19f
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_init_q15.c
@@ -0,0 +1,67 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_init_q15.c
+ * Description:  Q15 matrix initialization
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixInit
+ * @{
+ */
+
+  /**
+   * @brief  Q15 matrix initialization.
+   * @param[in,out] *S             points to an instance of the floating-point matrix structure.
+   * @param[in]     nRows          number of rows in the matrix.
+   * @param[in]     nColumns       number of columns in the matrix.
+   * @param[in]     *pData	   points to the matrix data array.
+   * @return        none
+   */
+
+void arm_mat_init_q15(
+  arm_matrix_instance_q15 * S,
+  uint16_t nRows,
+  uint16_t nColumns,
+  q15_t * pData)
+{
+  /* Assign Number of Rows */
+  S->numRows = nRows;
+
+  /* Assign Number of Columns */
+  S->numCols = nColumns;
+
+  /* Assign Data pointer */
+  S->pData = pData;
+}
+
+/**
+ * @} end of MatrixInit group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_init_q31.c b/DSP/Source/MatrixFunctions/arm_mat_init_q31.c
new file mode 100644
index 0000000..22e6f6d
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_init_q31.c
@@ -0,0 +1,72 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_init_q31.c
+ * Description:  Q31 matrix initialization
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @defgroup MatrixInit Matrix Initialization
+ *
+ */
+
+/**
+ * @addtogroup MatrixInit
+ * @{
+ */
+
+  /**
+   * @brief  Q31 matrix initialization.
+   * @param[in,out] *S             points to an instance of the floating-point matrix structure.
+   * @param[in]     nRows          number of rows in the matrix.
+   * @param[in]     nColumns       number of columns in the matrix.
+   * @param[in]     *pData	   points to the matrix data array.
+   * @return        none
+   */
+
+void arm_mat_init_q31(
+  arm_matrix_instance_q31 * S,
+  uint16_t nRows,
+  uint16_t nColumns,
+  q31_t * pData)
+{
+  /* Assign Number of Rows */
+  S->numRows = nRows;
+
+  /* Assign Number of Columns */
+  S->numCols = nColumns;
+
+  /* Assign Data pointer */
+  S->pData = pData;
+}
+
+/**
+ * @} end of MatrixInit group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_inverse_f32.c b/DSP/Source/MatrixFunctions/arm_mat_inverse_f32.c
new file mode 100644
index 0000000..b82373a
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_inverse_f32.c
@@ -0,0 +1,691 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_inverse_f32.c
+ * Description:  Floating-point matrix inverse
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @defgroup MatrixInv Matrix Inverse
+ *
+ * Computes the inverse of a matrix.
+ *
+ * The inverse is defined only if the input matrix is square and non-singular (the determinant
+ * is non-zero). The function checks that the input and output matrices are square and of the
+ * same size.
+ *
+ * Matrix inversion is numerically sensitive and the CMSIS DSP library only supports matrix
+ * inversion of floating-point matrices.
+ *
+ * \par Algorithm
+ * The Gauss-Jordan method is used to find the inverse.
+ * The algorithm performs a sequence of elementary row-operations until it
+ * reduces the input matrix to an identity matrix. Applying the same sequence
+ * of elementary row-operations to an identity matrix yields the inverse matrix.
+ * If the input matrix is singular, then the algorithm terminates and returns error status
+ * <code>ARM_MATH_SINGULAR</code>.
+ * \image html MatrixInverse.gif "Matrix Inverse of a 3 x 3 matrix using Gauss-Jordan Method"
+ */
+
+/**
+ * @addtogroup MatrixInv
+ * @{
+ */
+
+/**
+ * @brief Floating-point matrix inverse.
+ * @param[in]       *pSrc points to input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns
+ * <code>ARM_MATH_SIZE_MISMATCH</code> if the input matrix is not square or if the size
+ * of the output matrix does not match the size of the input matrix.
+ * If the input matrix is found to be singular (non-invertible), then the function returns
+ * <code>ARM_MATH_SINGULAR</code>.  Otherwise, the function returns <code>ARM_MATH_SUCCESS</code>.
+ */
+
+arm_status arm_mat_inverse_f32(
+  const arm_matrix_instance_f32 * pSrc,
+  arm_matrix_instance_f32 * pDst)
+{
+  float32_t *pIn = pSrc->pData;                  /* input data matrix pointer */
+  float32_t *pOut = pDst->pData;                 /* output data matrix pointer */
+  float32_t *pInT1, *pInT2;                      /* Temporary input data matrix pointer */
+  float32_t *pOutT1, *pOutT2;                    /* Temporary output data matrix pointer */
+  float32_t *pPivotRowIn, *pPRT_in, *pPivotRowDst, *pPRT_pDst;  /* Temporary input and output data matrix pointer */
+  uint32_t numRows = pSrc->numRows;              /* Number of rows in the matrix  */
+  uint32_t numCols = pSrc->numCols;              /* Number of Cols in the matrix  */
+
+#if defined (ARM_MATH_DSP)
+  float32_t maxC;                                /* maximum value in the column */
+
+  /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+  float32_t Xchg, in = 0.0f, in1;                /* Temporary input values  */
+  uint32_t i, rowCnt, flag = 0U, j, loopCnt, k, l;      /* loop counters */
+  arm_status status;                             /* status of matrix inverse */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+
+  /* Check for matrix mismatch condition */
+  if ((pSrc->numRows != pSrc->numCols) || (pDst->numRows != pDst->numCols)
+     || (pSrc->numRows != pDst->numRows))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+
+    /*--------------------------------------------------------------------------------------------------------------
+	 * Matrix Inverse can be solved using elementary row operations.
+	 *
+	 *	Gauss-Jordan Method:
+	 *
+	 *	   1. First combine the identity matrix and the input matrix separated by a bar to form an
+	 *        augmented matrix as follows:
+	 *				        _ 	      	       _         _	       _
+	 *					   |  a11  a12 | 1   0  |       |  X11 X12  |
+	 *					   |           |        |   =   |           |
+	 *					   |_ a21  a22 | 0   1 _|       |_ X21 X21 _|
+	 *
+	 *		2. In our implementation, pDst Matrix is used as identity matrix.
+	 *
+	 *		3. Begin with the first row. Let i = 1.
+	 *
+	 *	    4. Check to see if the pivot for column i is the greatest of the column.
+	 *		   The pivot is the element of the main diagonal that is on the current row.
+	 *		   For instance, if working with row i, then the pivot element is aii.
+	 *		   If the pivot is not the most significant of the columns, exchange that row with a row
+	 *		   below it that does contain the most significant value in column i. If the most
+	 *         significant value of the column is zero, then an inverse to that matrix does not exist.
+	 *		   The most significant value of the column is the absolute maximum.
+	 *
+	 *	    5. Divide every element of row i by the pivot.
+	 *
+	 *	    6. For every row below and  row i, replace that row with the sum of that row and
+	 *		   a multiple of row i so that each new element in column i below row i is zero.
+	 *
+	 *	    7. Move to the next row and column and repeat steps 2 through 5 until you have zeros
+	 *		   for every element below and above the main diagonal.
+	 *
+	 *		8. Now an identical matrix is formed to the left of the bar(input matrix, pSrc).
+	 *		   Therefore, the matrix to the right of the bar is our solution(pDst matrix, pDst).
+	 *----------------------------------------------------------------------------------------------------------------*/
+
+    /* Working pointer for destination matrix */
+    pOutT1 = pOut;
+
+    /* Loop over the number of rows */
+    rowCnt = numRows;
+
+    /* Making the destination matrix as identity matrix */
+    while (rowCnt > 0U)
+    {
+      /* Writing all zeroes in lower triangle of the destination matrix */
+      j = numRows - rowCnt;
+      while (j > 0U)
+      {
+        *pOutT1++ = 0.0f;
+        j--;
+      }
+
+      /* Writing all ones in the diagonal of the destination matrix */
+      *pOutT1++ = 1.0f;
+
+      /* Writing all zeroes in upper triangle of the destination matrix */
+      j = rowCnt - 1U;
+      while (j > 0U)
+      {
+        *pOutT1++ = 0.0f;
+        j--;
+      }
+
+      /* Decrement the loop counter */
+      rowCnt--;
+    }
+
+    /* Loop over the number of columns of the input matrix.
+       All the elements in each column are processed by the row operations */
+    loopCnt = numCols;
+
+    /* Index modifier to navigate through the columns */
+    l = 0U;
+
+    while (loopCnt > 0U)
+    {
+      /* Check if the pivot element is zero..
+       * If it is zero then interchange the row with non zero row below.
+       * If there is no non zero element to replace in the rows below,
+       * then the matrix is Singular. */
+
+      /* Working pointer for the input matrix that points
+       * to the pivot element of the particular row  */
+      pInT1 = pIn + (l * numCols);
+
+      /* Working pointer for the destination matrix that points
+       * to the pivot element of the particular row  */
+      pOutT1 = pOut + (l * numCols);
+
+      /* Temporary variable to hold the pivot value */
+      in = *pInT1;
+
+      /* Grab the most significant value from column l */
+      maxC = 0;
+      for (i = l; i < numRows; i++)
+      {
+        maxC = *pInT1 > 0 ? (*pInT1 > maxC ? *pInT1 : maxC) : (-*pInT1 > maxC ? -*pInT1 : maxC);
+        pInT1 += numCols;
+      }
+
+      /* Update the status if the matrix is singular */
+      if (maxC == 0.0f)
+      {
+        return ARM_MATH_SINGULAR;
+      }
+
+      /* Restore pInT1  */
+      pInT1 = pIn;
+
+      /* Destination pointer modifier */
+      k = 1U;
+
+      /* Check if the pivot element is the most significant of the column */
+      if ( (in > 0.0f ? in : -in) != maxC)
+      {
+        /* Loop over the number rows present below */
+        i = numRows - (l + 1U);
+
+        while (i > 0U)
+        {
+          /* Update the input and destination pointers */
+          pInT2 = pInT1 + (numCols * l);
+          pOutT2 = pOutT1 + (numCols * k);
+
+          /* Look for the most significant element to
+           * replace in the rows below */
+          if ((*pInT2 > 0.0f ? *pInT2: -*pInT2) == maxC)
+          {
+            /* Loop over number of columns
+             * to the right of the pilot element */
+            j = numCols - l;
+
+            while (j > 0U)
+            {
+              /* Exchange the row elements of the input matrix */
+              Xchg = *pInT2;
+              *pInT2++ = *pInT1;
+              *pInT1++ = Xchg;
+
+              /* Decrement the loop counter */
+              j--;
+            }
+
+            /* Loop over number of columns of the destination matrix */
+            j = numCols;
+
+            while (j > 0U)
+            {
+              /* Exchange the row elements of the destination matrix */
+              Xchg = *pOutT2;
+              *pOutT2++ = *pOutT1;
+              *pOutT1++ = Xchg;
+
+              /* Decrement the loop counter */
+              j--;
+            }
+
+            /* Flag to indicate whether exchange is done or not */
+            flag = 1U;
+
+            /* Break after exchange is done */
+            break;
+          }
+
+          /* Update the destination pointer modifier */
+          k++;
+
+          /* Decrement the loop counter */
+          i--;
+        }
+      }
+
+      /* Update the status if the matrix is singular */
+      if ((flag != 1U) && (in == 0.0f))
+      {
+        return ARM_MATH_SINGULAR;
+      }
+
+      /* Points to the pivot row of input and destination matrices */
+      pPivotRowIn = pIn + (l * numCols);
+      pPivotRowDst = pOut + (l * numCols);
+
+      /* Temporary pointers to the pivot row pointers */
+      pInT1 = pPivotRowIn;
+      pInT2 = pPivotRowDst;
+
+      /* Pivot element of the row */
+      in = *pPivotRowIn;
+
+      /* Loop over number of columns
+       * to the right of the pilot element */
+      j = (numCols - l);
+
+      while (j > 0U)
+      {
+        /* Divide each element of the row of the input matrix
+         * by the pivot element */
+        in1 = *pInT1;
+        *pInT1++ = in1 / in;
+
+        /* Decrement the loop counter */
+        j--;
+      }
+
+      /* Loop over number of columns of the destination matrix */
+      j = numCols;
+
+      while (j > 0U)
+      {
+        /* Divide each element of the row of the destination matrix
+         * by the pivot element */
+        in1 = *pInT2;
+        *pInT2++ = in1 / in;
+
+        /* Decrement the loop counter */
+        j--;
+      }
+
+      /* Replace the rows with the sum of that row and a multiple of row i
+       * so that each new element in column i above row i is zero.*/
+
+      /* Temporary pointers for input and destination matrices */
+      pInT1 = pIn;
+      pInT2 = pOut;
+
+      /* index used to check for pivot element */
+      i = 0U;
+
+      /* Loop over number of rows */
+      /*  to be replaced by the sum of that row and a multiple of row i */
+      k = numRows;
+
+      while (k > 0U)
+      {
+        /* Check for the pivot element */
+        if (i == l)
+        {
+          /* If the processing element is the pivot element,
+             only the columns to the right are to be processed */
+          pInT1 += numCols - l;
+
+          pInT2 += numCols;
+        }
+        else
+        {
+          /* Element of the reference row */
+          in = *pInT1;
+
+          /* Working pointers for input and destination pivot rows */
+          pPRT_in = pPivotRowIn;
+          pPRT_pDst = pPivotRowDst;
+
+          /* Loop over the number of columns to the right of the pivot element,
+             to replace the elements in the input matrix */
+          j = (numCols - l);
+
+          while (j > 0U)
+          {
+            /* Replace the element by the sum of that row
+               and a multiple of the reference row  */
+            in1 = *pInT1;
+            *pInT1++ = in1 - (in * *pPRT_in++);
+
+            /* Decrement the loop counter */
+            j--;
+          }
+
+          /* Loop over the number of columns to
+             replace the elements in the destination matrix */
+          j = numCols;
+
+          while (j > 0U)
+          {
+            /* Replace the element by the sum of that row
+               and a multiple of the reference row  */
+            in1 = *pInT2;
+            *pInT2++ = in1 - (in * *pPRT_pDst++);
+
+            /* Decrement the loop counter */
+            j--;
+          }
+
+        }
+
+        /* Increment the temporary input pointer */
+        pInT1 = pInT1 + l;
+
+        /* Decrement the loop counter */
+        k--;
+
+        /* Increment the pivot index */
+        i++;
+      }
+
+      /* Increment the input pointer */
+      pIn++;
+
+      /* Decrement the loop counter */
+      loopCnt--;
+
+      /* Increment the index modifier */
+      l++;
+    }
+
+
+#else
+
+  /* Run the below code for Cortex-M0 */
+
+  float32_t Xchg, in = 0.0f;                     /* Temporary input values  */
+  uint32_t i, rowCnt, flag = 0U, j, loopCnt, k, l;      /* loop counters */
+  arm_status status;                             /* status of matrix inverse */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+  /* Check for matrix mismatch condition */
+  if ((pSrc->numRows != pSrc->numCols) || (pDst->numRows != pDst->numCols)
+     || (pSrc->numRows != pDst->numRows))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*      #ifdef ARM_MATH_MATRIX_CHECK    */
+  {
+
+    /*--------------------------------------------------------------------------------------------------------------
+	 * Matrix Inverse can be solved using elementary row operations.
+	 *
+	 *	Gauss-Jordan Method:
+	 *
+	 *	   1. First combine the identity matrix and the input matrix separated by a bar to form an
+	 *        augmented matrix as follows:
+	 *				        _  _	      _	    _	   _   _         _	       _
+	 *					   |  |  a11  a12  | | | 1   0  |   |       |  X11 X12  |
+	 *					   |  |            | | |        |   |   =   |           |
+	 *					   |_ |_ a21  a22 _| | |_0   1 _|  _|       |_ X21 X21 _|
+	 *
+	 *		2. In our implementation, pDst Matrix is used as identity matrix.
+	 *
+	 *		3. Begin with the first row. Let i = 1.
+	 *
+	 *	    4. Check to see if the pivot for row i is zero.
+	 *		   The pivot is the element of the main diagonal that is on the current row.
+	 *		   For instance, if working with row i, then the pivot element is aii.
+	 *		   If the pivot is zero, exchange that row with a row below it that does not
+	 *		   contain a zero in column i. If this is not possible, then an inverse
+	 *		   to that matrix does not exist.
+	 *
+	 *	    5. Divide every element of row i by the pivot.
+	 *
+	 *	    6. For every row below and  row i, replace that row with the sum of that row and
+	 *		   a multiple of row i so that each new element in column i below row i is zero.
+	 *
+	 *	    7. Move to the next row and column and repeat steps 2 through 5 until you have zeros
+	 *		   for every element below and above the main diagonal.
+	 *
+	 *		8. Now an identical matrix is formed to the left of the bar(input matrix, src).
+	 *		   Therefore, the matrix to the right of the bar is our solution(dst matrix, dst).
+	 *----------------------------------------------------------------------------------------------------------------*/
+
+    /* Working pointer for destination matrix */
+    pOutT1 = pOut;
+
+    /* Loop over the number of rows */
+    rowCnt = numRows;
+
+    /* Making the destination matrix as identity matrix */
+    while (rowCnt > 0U)
+    {
+      /* Writing all zeroes in lower triangle of the destination matrix */
+      j = numRows - rowCnt;
+      while (j > 0U)
+      {
+        *pOutT1++ = 0.0f;
+        j--;
+      }
+
+      /* Writing all ones in the diagonal of the destination matrix */
+      *pOutT1++ = 1.0f;
+
+      /* Writing all zeroes in upper triangle of the destination matrix */
+      j = rowCnt - 1U;
+      while (j > 0U)
+      {
+        *pOutT1++ = 0.0f;
+        j--;
+      }
+
+      /* Decrement the loop counter */
+      rowCnt--;
+    }
+
+    /* Loop over the number of columns of the input matrix.
+       All the elements in each column are processed by the row operations */
+    loopCnt = numCols;
+
+    /* Index modifier to navigate through the columns */
+    l = 0U;
+    //for(loopCnt = 0U; loopCnt < numCols; loopCnt++)
+    while (loopCnt > 0U)
+    {
+      /* Check if the pivot element is zero..
+       * If it is zero then interchange the row with non zero row below.
+       * If there is no non zero element to replace in the rows below,
+       * then the matrix is Singular. */
+
+      /* Working pointer for the input matrix that points
+       * to the pivot element of the particular row  */
+      pInT1 = pIn + (l * numCols);
+
+      /* Working pointer for the destination matrix that points
+       * to the pivot element of the particular row  */
+      pOutT1 = pOut + (l * numCols);
+
+      /* Temporary variable to hold the pivot value */
+      in = *pInT1;
+
+      /* Destination pointer modifier */
+      k = 1U;
+
+      /* Check if the pivot element is zero */
+      if (*pInT1 == 0.0f)
+      {
+        /* Loop over the number rows present below */
+        for (i = (l + 1U); i < numRows; i++)
+        {
+          /* Update the input and destination pointers */
+          pInT2 = pInT1 + (numCols * l);
+          pOutT2 = pOutT1 + (numCols * k);
+
+          /* Check if there is a non zero pivot element to
+           * replace in the rows below */
+          if (*pInT2 != 0.0f)
+          {
+            /* Loop over number of columns
+             * to the right of the pilot element */
+            for (j = 0U; j < (numCols - l); j++)
+            {
+              /* Exchange the row elements of the input matrix */
+              Xchg = *pInT2;
+              *pInT2++ = *pInT1;
+              *pInT1++ = Xchg;
+            }
+
+            for (j = 0U; j < numCols; j++)
+            {
+              Xchg = *pOutT2;
+              *pOutT2++ = *pOutT1;
+              *pOutT1++ = Xchg;
+            }
+
+            /* Flag to indicate whether exchange is done or not */
+            flag = 1U;
+
+            /* Break after exchange is done */
+            break;
+          }
+
+          /* Update the destination pointer modifier */
+          k++;
+        }
+      }
+
+      /* Update the status if the matrix is singular */
+      if ((flag != 1U) && (in == 0.0f))
+      {
+        return ARM_MATH_SINGULAR;
+      }
+
+      /* Points to the pivot row of input and destination matrices */
+      pPivotRowIn = pIn + (l * numCols);
+      pPivotRowDst = pOut + (l * numCols);
+
+      /* Temporary pointers to the pivot row pointers */
+      pInT1 = pPivotRowIn;
+      pOutT1 = pPivotRowDst;
+
+      /* Pivot element of the row */
+      in = *(pIn + (l * numCols));
+
+      /* Loop over number of columns
+       * to the right of the pilot element */
+      for (j = 0U; j < (numCols - l); j++)
+      {
+        /* Divide each element of the row of the input matrix
+         * by the pivot element */
+        *pInT1 = *pInT1 / in;
+        pInT1++;
+      }
+      for (j = 0U; j < numCols; j++)
+      {
+        /* Divide each element of the row of the destination matrix
+         * by the pivot element */
+        *pOutT1 = *pOutT1 / in;
+        pOutT1++;
+      }
+
+      /* Replace the rows with the sum of that row and a multiple of row i
+       * so that each new element in column i above row i is zero.*/
+
+      /* Temporary pointers for input and destination matrices */
+      pInT1 = pIn;
+      pOutT1 = pOut;
+
+      for (i = 0U; i < numRows; i++)
+      {
+        /* Check for the pivot element */
+        if (i == l)
+        {
+          /* If the processing element is the pivot element,
+             only the columns to the right are to be processed */
+          pInT1 += numCols - l;
+          pOutT1 += numCols;
+        }
+        else
+        {
+          /* Element of the reference row */
+          in = *pInT1;
+
+          /* Working pointers for input and destination pivot rows */
+          pPRT_in = pPivotRowIn;
+          pPRT_pDst = pPivotRowDst;
+
+          /* Loop over the number of columns to the right of the pivot element,
+             to replace the elements in the input matrix */
+          for (j = 0U; j < (numCols - l); j++)
+          {
+            /* Replace the element by the sum of that row
+               and a multiple of the reference row  */
+            *pInT1 = *pInT1 - (in * *pPRT_in++);
+            pInT1++;
+          }
+          /* Loop over the number of columns to
+             replace the elements in the destination matrix */
+          for (j = 0U; j < numCols; j++)
+          {
+            /* Replace the element by the sum of that row
+               and a multiple of the reference row  */
+            *pOutT1 = *pOutT1 - (in * *pPRT_pDst++);
+            pOutT1++;
+          }
+
+        }
+        /* Increment the temporary input pointer */
+        pInT1 = pInT1 + l;
+      }
+      /* Increment the input pointer */
+      pIn++;
+
+      /* Decrement the loop counter */
+      loopCnt--;
+      /* Increment the index modifier */
+      l++;
+    }
+
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+
+    if ((flag != 1U) && (in == 0.0f))
+    {
+      pIn = pSrc->pData;
+      for (i = 0; i < numRows * numCols; i++)
+      {
+        if (pIn[i] != 0.0f)
+            break;
+      }
+
+      if (i == numRows * numCols)
+        status = ARM_MATH_SINGULAR;
+    }
+  }
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixInv group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_inverse_f64.c b/DSP/Source/MatrixFunctions/arm_mat_inverse_f64.c
new file mode 100644
index 0000000..54e5982
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_inverse_f64.c
@@ -0,0 +1,691 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_inverse_f64.c
+ * Description:  Floating-point matrix inverse
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @defgroup MatrixInv Matrix Inverse
+ *
+ * Computes the inverse of a matrix.
+ *
+ * The inverse is defined only if the input matrix is square and non-singular (the determinant
+ * is non-zero). The function checks that the input and output matrices are square and of the
+ * same size.
+ *
+ * Matrix inversion is numerically sensitive and the CMSIS DSP library only supports matrix
+ * inversion of floating-point matrices.
+ *
+ * \par Algorithm
+ * The Gauss-Jordan method is used to find the inverse.
+ * The algorithm performs a sequence of elementary row-operations until it
+ * reduces the input matrix to an identity matrix. Applying the same sequence
+ * of elementary row-operations to an identity matrix yields the inverse matrix.
+ * If the input matrix is singular, then the algorithm terminates and returns error status
+ * <code>ARM_MATH_SINGULAR</code>.
+ * \image html MatrixInverse.gif "Matrix Inverse of a 3 x 3 matrix using Gauss-Jordan Method"
+ */
+
+/**
+ * @addtogroup MatrixInv
+ * @{
+ */
+
+/**
+ * @brief Floating-point matrix inverse.
+ * @param[in]       *pSrc points to input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns
+ * <code>ARM_MATH_SIZE_MISMATCH</code> if the input matrix is not square or if the size
+ * of the output matrix does not match the size of the input matrix.
+ * If the input matrix is found to be singular (non-invertible), then the function returns
+ * <code>ARM_MATH_SINGULAR</code>.  Otherwise, the function returns <code>ARM_MATH_SUCCESS</code>.
+ */
+
+arm_status arm_mat_inverse_f64(
+  const arm_matrix_instance_f64 * pSrc,
+  arm_matrix_instance_f64 * pDst)
+{
+  float64_t *pIn = pSrc->pData;                  /* input data matrix pointer */
+  float64_t *pOut = pDst->pData;                 /* output data matrix pointer */
+  float64_t *pInT1, *pInT2;                      /* Temporary input data matrix pointer */
+  float64_t *pOutT1, *pOutT2;                    /* Temporary output data matrix pointer */
+  float64_t *pPivotRowIn, *pPRT_in, *pPivotRowDst, *pPRT_pDst;  /* Temporary input and output data matrix pointer */
+  uint32_t numRows = pSrc->numRows;              /* Number of rows in the matrix  */
+  uint32_t numCols = pSrc->numCols;              /* Number of Cols in the matrix  */
+
+#if defined (ARM_MATH_DSP)
+  float64_t maxC;                                /* maximum value in the column */
+
+  /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+  float64_t Xchg, in = 0.0f, in1;                /* Temporary input values  */
+  uint32_t i, rowCnt, flag = 0U, j, loopCnt, k, l;      /* loop counters */
+  arm_status status;                             /* status of matrix inverse */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+
+  /* Check for matrix mismatch condition */
+  if ((pSrc->numRows != pSrc->numCols) || (pDst->numRows != pDst->numCols)
+     || (pSrc->numRows != pDst->numRows))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+
+    /*--------------------------------------------------------------------------------------------------------------
+	 * Matrix Inverse can be solved using elementary row operations.
+	 *
+	 *	Gauss-Jordan Method:
+	 *
+	 *	   1. First combine the identity matrix and the input matrix separated by a bar to form an
+	 *        augmented matrix as follows:
+	 *				        _ 	      	       _         _	       _
+	 *					   |  a11  a12 | 1   0  |       |  X11 X12  |
+	 *					   |           |        |   =   |           |
+	 *					   |_ a21  a22 | 0   1 _|       |_ X21 X21 _|
+	 *
+	 *		2. In our implementation, pDst Matrix is used as identity matrix.
+	 *
+	 *		3. Begin with the first row. Let i = 1.
+	 *
+	 *	    4. Check to see if the pivot for column i is the greatest of the column.
+	 *		   The pivot is the element of the main diagonal that is on the current row.
+	 *		   For instance, if working with row i, then the pivot element is aii.
+	 *		   If the pivot is not the most significant of the columns, exchange that row with a row
+	 *		   below it that does contain the most significant value in column i. If the most
+	 *         significant value of the column is zero, then an inverse to that matrix does not exist.
+	 *		   The most significant value of the column is the absolute maximum.
+	 *
+	 *	    5. Divide every element of row i by the pivot.
+	 *
+	 *	    6. For every row below and  row i, replace that row with the sum of that row and
+	 *		   a multiple of row i so that each new element in column i below row i is zero.
+	 *
+	 *	    7. Move to the next row and column and repeat steps 2 through 5 until you have zeros
+	 *		   for every element below and above the main diagonal.
+	 *
+	 *		8. Now an identical matrix is formed to the left of the bar(input matrix, pSrc).
+	 *		   Therefore, the matrix to the right of the bar is our solution(pDst matrix, pDst).
+	 *----------------------------------------------------------------------------------------------------------------*/
+
+    /* Working pointer for destination matrix */
+    pOutT1 = pOut;
+
+    /* Loop over the number of rows */
+    rowCnt = numRows;
+
+    /* Making the destination matrix as identity matrix */
+    while (rowCnt > 0U)
+    {
+      /* Writing all zeroes in lower triangle of the destination matrix */
+      j = numRows - rowCnt;
+      while (j > 0U)
+      {
+        *pOutT1++ = 0.0f;
+        j--;
+      }
+
+      /* Writing all ones in the diagonal of the destination matrix */
+      *pOutT1++ = 1.0f;
+
+      /* Writing all zeroes in upper triangle of the destination matrix */
+      j = rowCnt - 1U;
+      while (j > 0U)
+      {
+        *pOutT1++ = 0.0f;
+        j--;
+      }
+
+      /* Decrement the loop counter */
+      rowCnt--;
+    }
+
+    /* Loop over the number of columns of the input matrix.
+       All the elements in each column are processed by the row operations */
+    loopCnt = numCols;
+
+    /* Index modifier to navigate through the columns */
+    l = 0U;
+
+    while (loopCnt > 0U)
+    {
+      /* Check if the pivot element is zero..
+       * If it is zero then interchange the row with non zero row below.
+       * If there is no non zero element to replace in the rows below,
+       * then the matrix is Singular. */
+
+      /* Working pointer for the input matrix that points
+       * to the pivot element of the particular row  */
+      pInT1 = pIn + (l * numCols);
+
+      /* Working pointer for the destination matrix that points
+       * to the pivot element of the particular row  */
+      pOutT1 = pOut + (l * numCols);
+
+      /* Temporary variable to hold the pivot value */
+      in = *pInT1;
+
+      /* Grab the most significant value from column l */
+      maxC = 0;
+      for (i = l; i < numRows; i++)
+      {
+        maxC = *pInT1 > 0 ? (*pInT1 > maxC ? *pInT1 : maxC) : (-*pInT1 > maxC ? -*pInT1 : maxC);
+        pInT1 += numCols;
+      }
+
+      /* Update the status if the matrix is singular */
+      if (maxC == 0.0f)
+      {
+        return ARM_MATH_SINGULAR;
+      }
+
+      /* Restore pInT1  */
+      pInT1 = pIn;
+
+      /* Destination pointer modifier */
+      k = 1U;
+
+      /* Check if the pivot element is the most significant of the column */
+      if ( (in > 0.0f ? in : -in) != maxC)
+      {
+        /* Loop over the number rows present below */
+        i = numRows - (l + 1U);
+
+        while (i > 0U)
+        {
+          /* Update the input and destination pointers */
+          pInT2 = pInT1 + (numCols * l);
+          pOutT2 = pOutT1 + (numCols * k);
+
+          /* Look for the most significant element to
+           * replace in the rows below */
+          if ((*pInT2 > 0.0f ? *pInT2: -*pInT2) == maxC)
+          {
+            /* Loop over number of columns
+             * to the right of the pilot element */
+            j = numCols - l;
+
+            while (j > 0U)
+            {
+              /* Exchange the row elements of the input matrix */
+              Xchg = *pInT2;
+              *pInT2++ = *pInT1;
+              *pInT1++ = Xchg;
+
+              /* Decrement the loop counter */
+              j--;
+            }
+
+            /* Loop over number of columns of the destination matrix */
+            j = numCols;
+
+            while (j > 0U)
+            {
+              /* Exchange the row elements of the destination matrix */
+              Xchg = *pOutT2;
+              *pOutT2++ = *pOutT1;
+              *pOutT1++ = Xchg;
+
+              /* Decrement the loop counter */
+              j--;
+            }
+
+            /* Flag to indicate whether exchange is done or not */
+            flag = 1U;
+
+            /* Break after exchange is done */
+            break;
+          }
+
+          /* Update the destination pointer modifier */
+          k++;
+
+          /* Decrement the loop counter */
+          i--;
+        }
+      }
+
+      /* Update the status if the matrix is singular */
+      if ((flag != 1U) && (in == 0.0f))
+      {
+        return ARM_MATH_SINGULAR;
+      }
+
+      /* Points to the pivot row of input and destination matrices */
+      pPivotRowIn = pIn + (l * numCols);
+      pPivotRowDst = pOut + (l * numCols);
+
+      /* Temporary pointers to the pivot row pointers */
+      pInT1 = pPivotRowIn;
+      pInT2 = pPivotRowDst;
+
+      /* Pivot element of the row */
+      in = *pPivotRowIn;
+
+      /* Loop over number of columns
+       * to the right of the pilot element */
+      j = (numCols - l);
+
+      while (j > 0U)
+      {
+        /* Divide each element of the row of the input matrix
+         * by the pivot element */
+        in1 = *pInT1;
+        *pInT1++ = in1 / in;
+
+        /* Decrement the loop counter */
+        j--;
+      }
+
+      /* Loop over number of columns of the destination matrix */
+      j = numCols;
+
+      while (j > 0U)
+      {
+        /* Divide each element of the row of the destination matrix
+         * by the pivot element */
+        in1 = *pInT2;
+        *pInT2++ = in1 / in;
+
+        /* Decrement the loop counter */
+        j--;
+      }
+
+      /* Replace the rows with the sum of that row and a multiple of row i
+       * so that each new element in column i above row i is zero.*/
+
+      /* Temporary pointers for input and destination matrices */
+      pInT1 = pIn;
+      pInT2 = pOut;
+
+      /* index used to check for pivot element */
+      i = 0U;
+
+      /* Loop over number of rows */
+      /*  to be replaced by the sum of that row and a multiple of row i */
+      k = numRows;
+
+      while (k > 0U)
+      {
+        /* Check for the pivot element */
+        if (i == l)
+        {
+          /* If the processing element is the pivot element,
+             only the columns to the right are to be processed */
+          pInT1 += numCols - l;
+
+          pInT2 += numCols;
+        }
+        else
+        {
+          /* Element of the reference row */
+          in = *pInT1;
+
+          /* Working pointers for input and destination pivot rows */
+          pPRT_in = pPivotRowIn;
+          pPRT_pDst = pPivotRowDst;
+
+          /* Loop over the number of columns to the right of the pivot element,
+             to replace the elements in the input matrix */
+          j = (numCols - l);
+
+          while (j > 0U)
+          {
+            /* Replace the element by the sum of that row
+               and a multiple of the reference row  */
+            in1 = *pInT1;
+            *pInT1++ = in1 - (in * *pPRT_in++);
+
+            /* Decrement the loop counter */
+            j--;
+          }
+
+          /* Loop over the number of columns to
+             replace the elements in the destination matrix */
+          j = numCols;
+
+          while (j > 0U)
+          {
+            /* Replace the element by the sum of that row
+               and a multiple of the reference row  */
+            in1 = *pInT2;
+            *pInT2++ = in1 - (in * *pPRT_pDst++);
+
+            /* Decrement the loop counter */
+            j--;
+          }
+
+        }
+
+        /* Increment the temporary input pointer */
+        pInT1 = pInT1 + l;
+
+        /* Decrement the loop counter */
+        k--;
+
+        /* Increment the pivot index */
+        i++;
+      }
+
+      /* Increment the input pointer */
+      pIn++;
+
+      /* Decrement the loop counter */
+      loopCnt--;
+
+      /* Increment the index modifier */
+      l++;
+    }
+
+
+#else
+
+  /* Run the below code for Cortex-M0 */
+
+  float64_t Xchg, in = 0.0f;                     /* Temporary input values  */
+  uint32_t i, rowCnt, flag = 0U, j, loopCnt, k, l;      /* loop counters */
+  arm_status status;                             /* status of matrix inverse */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+  /* Check for matrix mismatch condition */
+  if ((pSrc->numRows != pSrc->numCols) || (pDst->numRows != pDst->numCols)
+     || (pSrc->numRows != pDst->numRows))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*      #ifdef ARM_MATH_MATRIX_CHECK    */
+  {
+
+    /*--------------------------------------------------------------------------------------------------------------
+	 * Matrix Inverse can be solved using elementary row operations.
+	 *
+	 *	Gauss-Jordan Method:
+	 *
+	 *	   1. First combine the identity matrix and the input matrix separated by a bar to form an
+	 *        augmented matrix as follows:
+	 *				        _  _	      _	    _	   _   _         _	       _
+	 *					   |  |  a11  a12  | | | 1   0  |   |       |  X11 X12  |
+	 *					   |  |            | | |        |   |   =   |           |
+	 *					   |_ |_ a21  a22 _| | |_0   1 _|  _|       |_ X21 X21 _|
+	 *
+	 *		2. In our implementation, pDst Matrix is used as identity matrix.
+	 *
+	 *		3. Begin with the first row. Let i = 1.
+	 *
+	 *	    4. Check to see if the pivot for row i is zero.
+	 *		   The pivot is the element of the main diagonal that is on the current row.
+	 *		   For instance, if working with row i, then the pivot element is aii.
+	 *		   If the pivot is zero, exchange that row with a row below it that does not
+	 *		   contain a zero in column i. If this is not possible, then an inverse
+	 *		   to that matrix does not exist.
+	 *
+	 *	    5. Divide every element of row i by the pivot.
+	 *
+	 *	    6. For every row below and  row i, replace that row with the sum of that row and
+	 *		   a multiple of row i so that each new element in column i below row i is zero.
+	 *
+	 *	    7. Move to the next row and column and repeat steps 2 through 5 until you have zeros
+	 *		   for every element below and above the main diagonal.
+	 *
+	 *		8. Now an identical matrix is formed to the left of the bar(input matrix, src).
+	 *		   Therefore, the matrix to the right of the bar is our solution(dst matrix, dst).
+	 *----------------------------------------------------------------------------------------------------------------*/
+
+    /* Working pointer for destination matrix */
+    pOutT1 = pOut;
+
+    /* Loop over the number of rows */
+    rowCnt = numRows;
+
+    /* Making the destination matrix as identity matrix */
+    while (rowCnt > 0U)
+    {
+      /* Writing all zeroes in lower triangle of the destination matrix */
+      j = numRows - rowCnt;
+      while (j > 0U)
+      {
+        *pOutT1++ = 0.0f;
+        j--;
+      }
+
+      /* Writing all ones in the diagonal of the destination matrix */
+      *pOutT1++ = 1.0f;
+
+      /* Writing all zeroes in upper triangle of the destination matrix */
+      j = rowCnt - 1U;
+      while (j > 0U)
+      {
+        *pOutT1++ = 0.0f;
+        j--;
+      }
+
+      /* Decrement the loop counter */
+      rowCnt--;
+    }
+
+    /* Loop over the number of columns of the input matrix.
+       All the elements in each column are processed by the row operations */
+    loopCnt = numCols;
+
+    /* Index modifier to navigate through the columns */
+    l = 0U;
+    //for(loopCnt = 0U; loopCnt < numCols; loopCnt++)
+    while (loopCnt > 0U)
+    {
+      /* Check if the pivot element is zero..
+       * If it is zero then interchange the row with non zero row below.
+       * If there is no non zero element to replace in the rows below,
+       * then the matrix is Singular. */
+
+      /* Working pointer for the input matrix that points
+       * to the pivot element of the particular row  */
+      pInT1 = pIn + (l * numCols);
+
+      /* Working pointer for the destination matrix that points
+       * to the pivot element of the particular row  */
+      pOutT1 = pOut + (l * numCols);
+
+      /* Temporary variable to hold the pivot value */
+      in = *pInT1;
+
+      /* Destination pointer modifier */
+      k = 1U;
+
+      /* Check if the pivot element is zero */
+      if (*pInT1 == 0.0f)
+      {
+        /* Loop over the number rows present below */
+        for (i = (l + 1U); i < numRows; i++)
+        {
+          /* Update the input and destination pointers */
+          pInT2 = pInT1 + (numCols * l);
+          pOutT2 = pOutT1 + (numCols * k);
+
+          /* Check if there is a non zero pivot element to
+           * replace in the rows below */
+          if (*pInT2 != 0.0f)
+          {
+            /* Loop over number of columns
+             * to the right of the pilot element */
+            for (j = 0U; j < (numCols - l); j++)
+            {
+              /* Exchange the row elements of the input matrix */
+              Xchg = *pInT2;
+              *pInT2++ = *pInT1;
+              *pInT1++ = Xchg;
+            }
+
+            for (j = 0U; j < numCols; j++)
+            {
+              Xchg = *pOutT2;
+              *pOutT2++ = *pOutT1;
+              *pOutT1++ = Xchg;
+            }
+
+            /* Flag to indicate whether exchange is done or not */
+            flag = 1U;
+
+            /* Break after exchange is done */
+            break;
+          }
+
+          /* Update the destination pointer modifier */
+          k++;
+        }
+      }
+
+      /* Update the status if the matrix is singular */
+      if ((flag != 1U) && (in == 0.0f))
+      {
+        return ARM_MATH_SINGULAR;
+      }
+
+      /* Points to the pivot row of input and destination matrices */
+      pPivotRowIn = pIn + (l * numCols);
+      pPivotRowDst = pOut + (l * numCols);
+
+      /* Temporary pointers to the pivot row pointers */
+      pInT1 = pPivotRowIn;
+      pOutT1 = pPivotRowDst;
+
+      /* Pivot element of the row */
+      in = *(pIn + (l * numCols));
+
+      /* Loop over number of columns
+       * to the right of the pilot element */
+      for (j = 0U; j < (numCols - l); j++)
+      {
+        /* Divide each element of the row of the input matrix
+         * by the pivot element */
+        *pInT1 = *pInT1 / in;
+        pInT1++;
+      }
+      for (j = 0U; j < numCols; j++)
+      {
+        /* Divide each element of the row of the destination matrix
+         * by the pivot element */
+        *pOutT1 = *pOutT1 / in;
+        pOutT1++;
+      }
+
+      /* Replace the rows with the sum of that row and a multiple of row i
+       * so that each new element in column i above row i is zero.*/
+
+      /* Temporary pointers for input and destination matrices */
+      pInT1 = pIn;
+      pOutT1 = pOut;
+
+      for (i = 0U; i < numRows; i++)
+      {
+        /* Check for the pivot element */
+        if (i == l)
+        {
+          /* If the processing element is the pivot element,
+             only the columns to the right are to be processed */
+          pInT1 += numCols - l;
+          pOutT1 += numCols;
+        }
+        else
+        {
+          /* Element of the reference row */
+          in = *pInT1;
+
+          /* Working pointers for input and destination pivot rows */
+          pPRT_in = pPivotRowIn;
+          pPRT_pDst = pPivotRowDst;
+
+          /* Loop over the number of columns to the right of the pivot element,
+             to replace the elements in the input matrix */
+          for (j = 0U; j < (numCols - l); j++)
+          {
+            /* Replace the element by the sum of that row
+               and a multiple of the reference row  */
+            *pInT1 = *pInT1 - (in * *pPRT_in++);
+            pInT1++;
+          }
+          /* Loop over the number of columns to
+             replace the elements in the destination matrix */
+          for (j = 0U; j < numCols; j++)
+          {
+            /* Replace the element by the sum of that row
+               and a multiple of the reference row  */
+            *pOutT1 = *pOutT1 - (in * *pPRT_pDst++);
+            pOutT1++;
+          }
+
+        }
+        /* Increment the temporary input pointer */
+        pInT1 = pInT1 + l;
+      }
+      /* Increment the input pointer */
+      pIn++;
+
+      /* Decrement the loop counter */
+      loopCnt--;
+      /* Increment the index modifier */
+      l++;
+    }
+
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+
+    if ((flag != 1U) && (in == 0.0f))
+    {
+      pIn = pSrc->pData;
+      for (i = 0; i < numRows * numCols; i++)
+      {
+        if (pIn[i] != 0.0f)
+            break;
+      }
+
+      if (i == numRows * numCols)
+        status = ARM_MATH_SINGULAR;
+    }
+  }
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixInv group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_mult_f32.c b/DSP/Source/MatrixFunctions/arm_mat_mult_f32.c
new file mode 100644
index 0000000..a038f2f
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_mult_f32.c
@@ -0,0 +1,274 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_mult_f32.c
+ * Description:  Floating-point matrix multiplication
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @defgroup MatrixMult Matrix Multiplication
+ *
+ * Multiplies two matrices.
+ *
+ * \image html MatrixMultiplication.gif "Multiplication of two 3 x 3 matrices"
+
+ * Matrix multiplication is only defined if the number of columns of the
+ * first matrix equals the number of rows of the second matrix.
+ * Multiplying an <code>M x N</code> matrix with an <code>N x P</code> matrix results
+ * in an <code>M x P</code> matrix.
+ * When matrix size checking is enabled, the functions check: (1) that the inner dimensions of
+ * <code>pSrcA</code> and <code>pSrcB</code> are equal; and (2) that the size of the output
+ * matrix equals the outer dimensions of <code>pSrcA</code> and <code>pSrcB</code>.
+ */
+
+
+/**
+ * @addtogroup MatrixMult
+ * @{
+ */
+
+/**
+ * @brief Floating-point matrix multiplication.
+ * @param[in]       *pSrcA points to the first input matrix structure
+ * @param[in]       *pSrcB points to the second input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ */
+
+arm_status arm_mat_mult_f32(
+  const arm_matrix_instance_f32 * pSrcA,
+  const arm_matrix_instance_f32 * pSrcB,
+  arm_matrix_instance_f32 * pDst)
+{
+  float32_t *pIn1 = pSrcA->pData;                /* input data matrix pointer A */
+  float32_t *pIn2 = pSrcB->pData;                /* input data matrix pointer B */
+  float32_t *pInA = pSrcA->pData;                /* input data matrix pointer A  */
+  float32_t *pOut = pDst->pData;                 /* output data matrix pointer */
+  float32_t *px;                                 /* Temporary output data matrix pointer */
+  float32_t sum;                                 /* Accumulator */
+  uint16_t numRowsA = pSrcA->numRows;            /* number of rows of input matrix A */
+  uint16_t numColsB = pSrcB->numCols;            /* number of columns of input matrix B */
+  uint16_t numColsA = pSrcA->numCols;            /* number of columns of input matrix A */
+
+#if defined (ARM_MATH_DSP)
+
+  /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+  float32_t in1, in2, in3, in4;
+  uint16_t col, i = 0U, j, row = numRowsA, colCnt;      /* loop counters */
+  arm_status status;                             /* status of matrix multiplication */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numCols != pSrcB->numRows) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
+  {
+
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*      #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
+    /* row loop */
+    do
+    {
+      /* Output pointer is set to starting address of the row being processed */
+      px = pOut + i;
+
+      /* For every row wise process, the column loop counter is to be initiated */
+      col = numColsB;
+
+      /* For every row wise process, the pIn2 pointer is set
+       ** to the starting address of the pSrcB data */
+      pIn2 = pSrcB->pData;
+
+      j = 0U;
+
+      /* column loop */
+      do
+      {
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum = 0.0f;
+
+        /* Initiate the pointer pIn1 to point to the starting address of the column being processed */
+        pIn1 = pInA;
+
+        /* Apply loop unrolling and compute 4 MACs simultaneously. */
+        colCnt = numColsA >> 2U;
+
+        /* matrix multiplication        */
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          in3 = *pIn2;
+          pIn2 += numColsB;
+          in1 = pIn1[0];
+          in2 = pIn1[1];
+          sum += in1 * in3;
+          in4 = *pIn2;
+          pIn2 += numColsB;
+          sum += in2 * in4;
+
+          in3 = *pIn2;
+          pIn2 += numColsB;
+          in1 = pIn1[2];
+          in2 = pIn1[3];
+          sum += in1 * in3;
+          in4 = *pIn2;
+          pIn2 += numColsB;
+          sum += in2 * in4;
+          pIn1 += 4U;
+
+          /* Decrement the loop count */
+          colCnt--;
+        }
+
+        /* If the columns of pSrcA is not a multiple of 4, compute any remaining MACs here.
+         ** No loop unrolling is used. */
+        colCnt = numColsA % 0x4U;
+
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          sum += *pIn1++ * (*pIn2);
+          pIn2 += numColsB;
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        /* Store the result in the destination buffer */
+        *px++ = sum;
+
+        /* Update the pointer pIn2 to point to the  starting address of the next column */
+        j++;
+        pIn2 = pSrcB->pData + j;
+
+        /* Decrement the column loop counter */
+        col--;
+
+      } while (col > 0U);
+
+#else
+
+  /* Run the below code for Cortex-M0 */
+
+  float32_t *pInB = pSrcB->pData;                /* input data matrix pointer B */
+  uint16_t col, i = 0U, row = numRowsA, colCnt;  /* loop counters */
+  arm_status status;                             /* status of matrix multiplication */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numCols != pSrcB->numRows) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
+  {
+
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*      #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* The following loop performs the dot-product of each row in pInA with each column in pInB */
+    /* row loop */
+    do
+    {
+      /* Output pointer is set to starting address of the row being processed */
+      px = pOut + i;
+
+      /* For every row wise process, the column loop counter is to be initiated */
+      col = numColsB;
+
+      /* For every row wise process, the pIn2 pointer is set
+       ** to the starting address of the pSrcB data */
+      pIn2 = pSrcB->pData;
+
+      /* column loop */
+      do
+      {
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum = 0.0f;
+
+        /* Initialize the pointer pIn1 to point to the starting address of the row being processed */
+        pIn1 = pInA;
+
+        /* Matrix A columns number of MAC operations are to be performed */
+        colCnt = numColsA;
+
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          sum += *pIn1++ * (*pIn2);
+          pIn2 += numColsB;
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        /* Store the result in the destination buffer */
+        *px++ = sum;
+
+        /* Decrement the column loop counter */
+        col--;
+
+        /* Update the pointer pIn2 to point to the  starting address of the next column */
+        pIn2 = pInB + (numColsB - col);
+
+      } while (col > 0U);
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+      /* Update the pointer pInA to point to the  starting address of the next row */
+      i = i + numColsB;
+      pInA = pInA + numColsA;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixMult group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_mult_fast_q15.c b/DSP/Source/MatrixFunctions/arm_mat_mult_fast_q15.c
new file mode 100644
index 0000000..8d720c7
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_mult_fast_q15.c
@@ -0,0 +1,525 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_mult_fast_q15.c
+ * Description:  Q15 matrix multiplication (fast variant)
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixMult
+ * @{
+ */
+
+
+/**
+ * @brief Q15 matrix multiplication (fast variant) for Cortex-M3 and Cortex-M4
+ * @param[in]       *pSrcA points to the first input matrix structure
+ * @param[in]       *pSrcB points to the second input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @param[in]       *pState points to the array for storing intermediate results
+ * @return          The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * @details
+ * <b>Scaling and Overflow Behavior:</b>
+ *
+ * \par
+ * The difference between the function arm_mat_mult_q15() and this fast variant is that
+ * the fast variant use a 32-bit rather than a 64-bit accumulator.
+ * The result of each 1.15 x 1.15 multiplication is truncated to
+ * 2.30 format. These intermediate results are accumulated in a 32-bit register in 2.30
+ * format. Finally, the accumulator is saturated and converted to a 1.15 result.
+ *
+ * \par
+ * The fast version has the same overflow behavior as the standard version but provides
+ * less precision since it discards the low 16 bits of each multiplication result.
+ * In order to avoid overflows completely the input signals must be scaled down.
+ * Scale down one of the input matrices by log2(numColsA) bits to
+ * avoid overflows, as a total of numColsA additions are computed internally for each
+ * output element.
+ *
+ * \par
+ * See <code>arm_mat_mult_q15()</code> for a slower implementation of this function
+ * which uses 64-bit accumulation to provide higher precision.
+ */
+
+arm_status arm_mat_mult_fast_q15(
+  const arm_matrix_instance_q15 * pSrcA,
+  const arm_matrix_instance_q15 * pSrcB,
+  arm_matrix_instance_q15 * pDst,
+  q15_t * pState)
+{
+  q31_t sum;                                     /* accumulator */
+  q15_t *pSrcBT = pState;                        /* input data matrix pointer for transpose */
+  q15_t *pInA = pSrcA->pData;                    /* input data matrix pointer A of Q15 type */
+  q15_t *pInB = pSrcB->pData;                    /* input data matrix pointer B of Q15 type */
+  q15_t *px;                                     /* Temporary output data matrix pointer */
+  uint16_t numRowsA = pSrcA->numRows;            /* number of rows of input matrix A    */
+  uint16_t numColsB = pSrcB->numCols;            /* number of columns of input matrix B */
+  uint16_t numColsA = pSrcA->numCols;            /* number of columns of input matrix A */
+  uint16_t numRowsB = pSrcB->numRows;            /* number of rows of input matrix A    */
+  uint32_t col, i = 0U, row = numRowsB, colCnt;  /* loop counters */
+  arm_status status;                             /* status of matrix multiplication */
+
+#ifndef UNALIGNED_SUPPORT_DISABLE
+
+  q31_t in;                                      /* Temporary variable to hold the input value */
+  q31_t inA1, inA2, inB1, inB2;
+  q31_t sum2, sum3, sum4;
+  q15_t *pInA2, *pInB2, *px2;
+  uint32_t j = 0;
+
+#else
+
+  q15_t in;                                      /* Temporary variable to hold the input value */
+  q15_t inA1, inA2, inB1, inB2;
+
+#endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numCols != pSrcB->numRows) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif
+  {
+    /* Matrix transpose */
+    do
+    {
+      /* Apply loop unrolling and exchange the columns with row elements */
+      col = numColsB >> 2;
+
+      /* The pointer px is set to starting address of the column being processed */
+      px = pSrcBT + i;
+
+      /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+       ** a second loop below computes the remaining 1 to 3 samples. */
+      while (col > 0U)
+      {
+#ifndef UNALIGNED_SUPPORT_DISABLE
+        /* Read two elements from the row */
+        in = *__SIMD32(pInB)++;
+
+        /* Unpack and store one element in the destination */
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *px = (q15_t) in;
+
+#else
+
+        *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#endif /*    #ifndef ARM_MATH_BIG_ENDIAN    */
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Unpack and store the second element in the destination */
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#else
+
+        *px = (q15_t) in;
+
+#endif /*    #ifndef ARM_MATH_BIG_ENDIAN    */
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Read two elements from the row */
+        in = *__SIMD32(pInB)++;
+
+        /* Unpack and store one element in the destination */
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *px = (q15_t) in;
+
+#else
+
+        *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#endif /*    #ifndef ARM_MATH_BIG_ENDIAN    */
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Unpack and store the second element in the destination */
+
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#else
+
+        *px = (q15_t) in;
+
+#endif /*    #ifndef ARM_MATH_BIG_ENDIAN    */
+
+#else
+
+        /* Read one element from the row */
+        in = *pInB++;
+
+        /* Store one element in the destination */
+        *px = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Read one element from the row */
+        in = *pInB++;
+
+        /* Store one element in the destination */
+        *px = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Read one element from the row */
+        in = *pInB++;
+
+        /* Store one element in the destination */
+        *px = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Read one element from the row */
+        in = *pInB++;
+
+        /* Store one element in the destination */
+        *px = in;
+
+#endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Decrement the column loop counter */
+        col--;
+      }
+
+      /* If the columns of pSrcB is not a multiple of 4, compute any remaining output samples here.
+       ** No loop unrolling is used. */
+      col = numColsB % 0x4U;
+
+      while (col > 0U)
+      {
+        /* Read and store the input element in the destination */
+        *px = *pInB++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Decrement the column loop counter */
+        col--;
+      }
+
+      i++;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);
+
+    /* Reset the variables for the usage in the following multiplication process */
+    row = numRowsA;
+    i = 0U;
+    px = pDst->pData;
+
+#ifndef UNALIGNED_SUPPORT_DISABLE
+    /* Process two rows from matrix A at a time and output two rows at a time */
+    row = row >> 1;
+    px2 = px + numColsB;
+#endif
+
+    /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
+    /* row loop */
+    while (row > 0U)
+    {
+      /* For every row wise process, the column loop counter is to be initiated */
+      col = numColsB;
+
+      /* For every row wise process, the pIn2 pointer is set
+       ** to the starting address of the transposed pSrcB data */
+      pInB = pSrcBT;
+
+#ifndef UNALIGNED_SUPPORT_DISABLE
+      /* Process two (transposed) columns from matrix B at a time */
+      col = col >> 1;
+      j = 0;
+#endif
+
+      /* column loop */
+      while (col > 0U)
+      {
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum = 0;
+
+        /* Initiate the pointer pInA to point to the starting address of the column being processed */
+        pInA = pSrcA->pData + i;
+
+#ifndef UNALIGNED_SUPPORT_DISABLE
+        sum2 = 0;
+        sum3 = 0;
+        sum4 = 0;
+        pInB  = pSrcBT + j;
+        pInA2 = pInA + numColsA;
+        pInB2 = pInB + numRowsB;
+
+        /* Read in two elements at once - alows dual MAC instruction */
+        colCnt = numColsA >> 1;
+#else
+        colCnt = numColsA >> 2;
+#endif
+
+        /* matrix multiplication */
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+#ifndef UNALIGNED_SUPPORT_DISABLE
+
+          inA1 = *__SIMD32(pInA)++;
+          inB1 = *__SIMD32(pInB)++;
+          inA2 = *__SIMD32(pInA2)++;
+          inB2 = *__SIMD32(pInB2)++;
+
+          sum  = __SMLAD(inA1, inB1, sum);
+          sum2 = __SMLAD(inA1, inB2, sum2);
+          sum3 = __SMLAD(inA2, inB1, sum3);
+          sum4 = __SMLAD(inA2, inB2, sum4);
+
+#else
+
+          inA1 = *pInA;
+          inB1 = *pInB;
+          sum += inA1 * inB1;
+
+          inA2 = pInA[1];
+          inB2 = pInB[1];
+          sum += inA2 * inB2;
+
+          inA1 = pInA[2];
+          inB1 = pInB[2];
+          sum += inA1 * inB1;
+
+          inA2 = pInA[3];
+          inB2 = pInB[3];
+          sum += inA2 * inB2;
+
+          pInA += 4;
+          pInB += 4;
+
+#endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        /* process odd column samples */
+#ifndef UNALIGNED_SUPPORT_DISABLE
+        if (numColsA & 1U) {
+          inA1 = *pInA++;
+          inB1 = *pInB++;
+          inA2 = *pInA2++;
+          inB2 = *pInB2++;
+          sum  += inA1 * inB1;
+          sum2 += inA1 * inB2;
+          sum3 += inA2 * inB1;
+          sum4 += inA2 * inB2;
+        }
+#else
+        colCnt = numColsA % 0x4U;
+
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          sum += (q31_t) (*pInA++) * (*pInB++);
+
+          colCnt--;
+        }
+#endif
+
+        /* Saturate and store the result in the destination buffer */
+        *px++  = (q15_t) (sum >> 15);
+
+#ifndef UNALIGNED_SUPPORT_DISABLE
+        *px++  = (q15_t) (sum2 >> 15);
+        *px2++ = (q15_t) (sum3 >> 15);
+        *px2++ = (q15_t) (sum4 >> 15);
+        j += numRowsB * 2;
+#endif
+
+        /* Decrement the column loop counter */
+        col--;
+
+      }
+
+      i = i + numColsA;
+
+#ifndef UNALIGNED_SUPPORT_DISABLE
+      i = i + numColsA;
+      px = px2 + (numColsB & 1U);
+      px2 = px + numColsB;
+#endif
+
+      /* Decrement the row loop counter */
+      row--;
+
+    }
+
+    /* Compute any remaining odd row/column below */
+
+#ifndef UNALIGNED_SUPPORT_DISABLE
+
+    /* Compute remaining output column */
+    if (numColsB & 1U) {
+
+      /* Avoid redundant computation of last element */
+      row = numRowsA & (~0x1);
+
+      /* Point to remaining unfilled column in output matrix */
+      px = pDst->pData+numColsB-1;
+      pInA = pSrcA->pData;
+
+      /* row loop */
+      while (row > 0)
+      {
+
+        /* point to last column in matrix B */
+        pInB  = pSrcBT + numRowsB*(numColsB-1);
+
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum  = 0;
+
+        /* Compute 4 columns at once */
+        colCnt = numColsA >> 2;
+
+        /* matrix multiplication */
+        while (colCnt > 0U)
+        {
+          inA1 = *__SIMD32(pInA)++;
+          inA2 = *__SIMD32(pInA)++;
+          inB1 = *__SIMD32(pInB)++;
+          inB2 = *__SIMD32(pInB)++;
+
+          sum  = __SMLAD(inA1, inB1, sum);
+          sum  = __SMLAD(inA2, inB2, sum);
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        colCnt = numColsA & 3U;
+        while (colCnt > 0U) {
+          sum += (q31_t) (*pInA++) * (*pInB++);
+          colCnt--;
+        }
+
+        /* Store the result in the destination buffer */
+        *px  = (q15_t) (sum  >> 15);
+        px += numColsB;
+
+        /* Decrement the row loop counter */
+        row--;
+      }
+    }
+
+    /* Compute remaining output row */
+    if (numRowsA & 1U) {
+
+      /* point to last row in output matrix */
+      px = pDst->pData+(numColsB)*(numRowsA-1);
+
+      pInB  = pSrcBT;
+      col = numColsB;
+      i = 0U;
+
+      /* col loop */
+      while (col > 0)
+      {
+
+        /* point to last row in matrix A */
+        pInA = pSrcA->pData + (numRowsA-1)*numColsA;
+
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum  = 0;
+
+        /* Compute 4 columns at once */
+        colCnt = numColsA >> 2;
+
+        /* matrix multiplication */
+        while (colCnt > 0U)
+        {
+          inA1 = *__SIMD32(pInA)++;
+          inA2 = *__SIMD32(pInA)++;
+          inB1 = *__SIMD32(pInB)++;
+          inB2 = *__SIMD32(pInB)++;
+
+          sum  = __SMLAD(inA1, inB1, sum);
+          sum  = __SMLAD(inA2, inB2, sum);
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        colCnt = numColsA & 3U;
+        while (colCnt > 0U) {
+          sum += (q31_t) (*pInA++) * (*pInB++);
+          colCnt--;
+        }
+
+        /* Store the result in the destination buffer */
+        *px++  = (q15_t) (sum  >> 15);
+
+        /* Decrement the col loop counter */
+        col--;
+      }
+    }
+
+#endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */
+
+    /* set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixMult group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_mult_fast_q31.c b/DSP/Source/MatrixFunctions/arm_mat_mult_fast_q31.c
new file mode 100644
index 0000000..78b33ef
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_mult_fast_q31.c
@@ -0,0 +1,384 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_mult_fast_q31.c
+ * Description:  Q31 matrix multiplication (fast variant)
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixMult
+ * @{
+ */
+
+/**
+ * @brief Q31 matrix multiplication (fast variant) for Cortex-M3 and Cortex-M4
+ * @param[in]       *pSrcA points to the first input matrix structure
+ * @param[in]       *pSrcB points to the second input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @return          The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * @details
+ * <b>Scaling and Overflow Behavior:</b>
+ *
+ * \par
+ * The difference between the function arm_mat_mult_q31() and this fast variant is that
+ * the fast variant use a 32-bit rather than a 64-bit accumulator.
+ * The result of each 1.31 x 1.31 multiplication is truncated to
+ * 2.30 format. These intermediate results are accumulated in a 32-bit register in 2.30
+ * format. Finally, the accumulator is saturated and converted to a 1.31 result.
+ *
+ * \par
+ * The fast version has the same overflow behavior as the standard version but provides
+ * less precision since it discards the low 32 bits of each multiplication result.
+ * In order to avoid overflows completely the input signals must be scaled down.
+ * Scale down one of the input matrices by log2(numColsA) bits to
+ * avoid overflows, as a total of numColsA additions are computed internally for each
+ * output element.
+ *
+ * \par
+ * See <code>arm_mat_mult_q31()</code> for a slower implementation of this function
+ * which uses 64-bit accumulation to provide higher precision.
+ */
+
+arm_status arm_mat_mult_fast_q31(
+  const arm_matrix_instance_q31 * pSrcA,
+  const arm_matrix_instance_q31 * pSrcB,
+  arm_matrix_instance_q31 * pDst)
+{
+  q31_t *pInA = pSrcA->pData;                    /* input data matrix pointer A */
+  q31_t *pInB = pSrcB->pData;                    /* input data matrix pointer B */
+  q31_t *px;                                     /* Temporary output data matrix pointer */
+  q31_t sum;                                     /* Accumulator */
+  uint16_t numRowsA = pSrcA->numRows;            /* number of rows of input matrix A    */
+  uint16_t numColsB = pSrcB->numCols;            /* number of columns of input matrix B */
+  uint16_t numColsA = pSrcA->numCols;            /* number of columns of input matrix A */
+  uint32_t col, i = 0U, j, row = numRowsA, colCnt;  /* loop counters */
+  arm_status status;                             /* status of matrix multiplication */
+  q31_t inA1, inB1;
+
+#if defined (ARM_MATH_DSP)
+
+  q31_t sum2, sum3, sum4;
+  q31_t inA2, inB2;
+  q31_t *pInA2;
+  q31_t *px2;
+
+#endif
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numCols != pSrcB->numRows) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*      #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+
+    px = pDst->pData;
+
+#if defined (ARM_MATH_DSP)
+    row = row >> 1;
+    px2 = px + numColsB;
+#endif
+
+    /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
+    /* row loop */
+    while (row > 0U)
+    {
+
+      /* For every row wise process, the column loop counter is to be initiated */
+      col = numColsB;
+
+      /* For every row wise process, the pIn2 pointer is set
+       ** to the starting address of the pSrcB data */
+      pInB = pSrcB->pData;
+
+      j = 0U;
+
+#if defined (ARM_MATH_DSP)
+      col = col >> 1;
+#endif
+
+      /* column loop */
+      while (col > 0U)
+      {
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum = 0;
+
+        /* Initiate data pointers */
+        pInA = pSrcA->pData + i;
+        pInB  = pSrcB->pData + j;
+
+#if defined (ARM_MATH_DSP)
+        sum2 = 0;
+        sum3 = 0;
+        sum4 = 0;
+        pInA2 = pInA + numColsA;
+        colCnt = numColsA;
+#else
+        colCnt = numColsA >> 2;
+#endif
+
+        /* matrix multiplication */
+        while (colCnt > 0U)
+        {
+
+#if defined (ARM_MATH_DSP)
+          inA1 = *pInA++;
+          inB1 = pInB[0];
+          inA2 = *pInA2++;
+          inB2 = pInB[1];
+          pInB += numColsB;
+
+          sum  = __SMMLA(inA1, inB1, sum);
+          sum2 = __SMMLA(inA1, inB2, sum2);
+          sum3 = __SMMLA(inA2, inB1, sum3);
+          sum4 = __SMMLA(inA2, inB2, sum4);
+#else
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          /* Perform the multiply-accumulates */
+          inB1 = *pInB;
+          pInB += numColsB;
+          inA1 = pInA[0];
+          sum = __SMMLA(inA1, inB1, sum);
+
+          inB1 = *pInB;
+          pInB += numColsB;
+          inA1 = pInA[1];
+          sum = __SMMLA(inA1, inB1, sum);
+
+          inB1 = *pInB;
+          pInB += numColsB;
+          inA1 = pInA[2];
+          sum = __SMMLA(inA1, inB1, sum);
+
+          inB1 = *pInB;
+          pInB += numColsB;
+          inA1 = pInA[3];
+          sum = __SMMLA(inA1, inB1, sum);
+
+          pInA += 4U;
+#endif
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+#ifdef ARM_MATH_CM0_FAMILY
+        /* If the columns of pSrcA is not a multiple of 4, compute any remaining output samples here. */
+        colCnt = numColsA % 0x4U;
+        while (colCnt > 0U)
+        {
+          sum = __SMMLA(*pInA++, *pInB, sum);
+          pInB += numColsB;
+          colCnt--;
+        }
+        j++;
+#endif
+
+        /* Convert the result from 2.30 to 1.31 format and store in destination buffer */
+        *px++  = sum << 1;
+
+#if defined (ARM_MATH_DSP)
+        *px++  = sum2 << 1;
+        *px2++ = sum3 << 1;
+        *px2++ = sum4 << 1;
+        j += 2;
+#endif
+
+        /* Decrement the column loop counter */
+        col--;
+
+      }
+
+      i = i + numColsA;
+
+#if defined (ARM_MATH_DSP)
+      i = i + numColsA;
+      px = px2 + (numColsB & 1U);
+      px2 = px + numColsB;
+#endif
+
+      /* Decrement the row loop counter */
+      row--;
+
+    }
+
+    /* Compute any remaining odd row/column below */
+
+#if defined (ARM_MATH_DSP)
+
+    /* Compute remaining output column */
+    if (numColsB & 1U) {
+
+      /* Avoid redundant computation of last element */
+      row = numRowsA & (~0x1);
+
+      /* Point to remaining unfilled column in output matrix */
+      px = pDst->pData+numColsB-1;
+      pInA = pSrcA->pData;
+
+      /* row loop */
+      while (row > 0)
+      {
+
+        /* point to last column in matrix B */
+        pInB  = pSrcB->pData + numColsB-1;
+
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum  = 0;
+
+        /* Compute 4 columns at once */
+        colCnt = numColsA >> 2;
+
+        /* matrix multiplication */
+        while (colCnt > 0U)
+        {
+          inA1 = *pInA++;
+          inA2 = *pInA++;
+          inB1 = *pInB;
+          pInB += numColsB;
+          inB2 = *pInB;
+          pInB += numColsB;
+          sum = __SMMLA(inA1, inB1, sum);
+          sum = __SMMLA(inA2, inB2, sum);
+
+          inA1 = *pInA++;
+          inA2 = *pInA++;
+          inB1 = *pInB;
+          pInB += numColsB;
+          inB2 = *pInB;
+          pInB += numColsB;
+          sum = __SMMLA(inA1, inB1, sum);
+          sum = __SMMLA(inA2, inB2, sum);
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        colCnt = numColsA & 3U;
+        while (colCnt > 0U) {
+          sum = __SMMLA(*pInA++, *pInB, sum);
+          pInB += numColsB;
+          colCnt--;
+        }
+
+        /* Convert the result from 2.30 to 1.31 format and store in destination buffer */
+        *px = sum << 1;
+        px += numColsB;
+
+        /* Decrement the row loop counter */
+        row--;
+      }
+    }
+
+    /* Compute remaining output row */
+    if (numRowsA & 1U) {
+
+      /* point to last row in output matrix */
+      px = pDst->pData+(numColsB)*(numRowsA-1);
+
+      col = numColsB;
+      i = 0U;
+
+      /* col loop */
+      while (col > 0)
+      {
+
+        /* point to last row in matrix A */
+        pInA = pSrcA->pData + (numRowsA-1)*numColsA;
+        pInB  = pSrcB->pData + i;
+
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum  = 0;
+
+        /* Compute 4 columns at once */
+        colCnt = numColsA >> 2;
+
+        /* matrix multiplication */
+        while (colCnt > 0U)
+        {
+          inA1 = *pInA++;
+          inA2 = *pInA++;
+          inB1 = *pInB;
+          pInB += numColsB;
+          inB2 = *pInB;
+          pInB += numColsB;
+          sum = __SMMLA(inA1, inB1, sum);
+          sum = __SMMLA(inA2, inB2, sum);
+
+          inA1 = *pInA++;
+          inA2 = *pInA++;
+          inB1 = *pInB;
+          pInB += numColsB;
+          inB2 = *pInB;
+          pInB += numColsB;
+          sum = __SMMLA(inA1, inB1, sum);
+          sum = __SMMLA(inA2, inB2, sum);
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        colCnt = numColsA & 3U;
+        while (colCnt > 0U) {
+          sum = __SMMLA(*pInA++, *pInB, sum);
+          pInB += numColsB;
+          colCnt--;
+        }
+
+        /* Saturate and store the result in the destination buffer */
+        *px++ = sum << 1;
+        i++;
+
+        /* Decrement the col loop counter */
+        col--;
+      }
+    }
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    /* set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixMult group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_mult_q15.c b/DSP/Source/MatrixFunctions/arm_mat_mult_q15.c
new file mode 100644
index 0000000..3244f47
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_mult_q15.c
@@ -0,0 +1,457 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_mult_q15.c
+ * Description:  Q15 matrix multiplication
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixMult
+ * @{
+ */
+
+
+/**
+ * @brief Q15 matrix multiplication
+ * @param[in]       *pSrcA points to the first input matrix structure
+ * @param[in]       *pSrcB points to the second input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @param[in]       *pState points to the array for storing intermediate results (Unused)
+ * @return          The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * @details
+ * <b>Scaling and Overflow Behavior:</b>
+ *
+ * \par
+ * The function is implemented using a 64-bit internal accumulator. The inputs to the
+ * multiplications are in 1.15 format and multiplications yield a 2.30 result.
+ * The 2.30 intermediate
+ * results are accumulated in a 64-bit accumulator in 34.30 format. This approach
+ * provides 33 guard bits and there is no risk of overflow. The 34.30 result is then
+ * truncated to 34.15 format by discarding the low 15 bits and then saturated to
+ * 1.15 format.
+ *
+ * \par
+ * Refer to <code>arm_mat_mult_fast_q15()</code> for a faster but less precise version of this function for Cortex-M3 and Cortex-M4.
+ *
+ */
+
+arm_status arm_mat_mult_q15(
+  const arm_matrix_instance_q15 * pSrcA,
+  const arm_matrix_instance_q15 * pSrcB,
+  arm_matrix_instance_q15 * pDst,
+  q15_t * pState)
+{
+  q63_t sum;                                     /* accumulator */
+
+#if defined (ARM_MATH_DSP)
+
+  /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+  q15_t *pSrcBT = pState;                        /* input data matrix pointer for transpose */
+  q15_t *pInA = pSrcA->pData;                    /* input data matrix pointer A of Q15 type */
+  q15_t *pInB = pSrcB->pData;                    /* input data matrix pointer B of Q15 type */
+  q15_t *px;                                     /* Temporary output data matrix pointer */
+  uint16_t numRowsA = pSrcA->numRows;            /* number of rows of input matrix A    */
+  uint16_t numColsB = pSrcB->numCols;            /* number of columns of input matrix B */
+  uint16_t numColsA = pSrcA->numCols;            /* number of columns of input matrix A */
+  uint16_t numRowsB = pSrcB->numRows;            /* number of rows of input matrix A    */
+  uint16_t col, i = 0U, row = numRowsB, colCnt;  /* loop counters */
+  arm_status status;                             /* status of matrix multiplication */
+
+#ifndef UNALIGNED_SUPPORT_DISABLE
+
+  q31_t in;                                      /* Temporary variable to hold the input value */
+  q31_t pSourceA1, pSourceB1, pSourceA2, pSourceB2;
+
+#else
+
+  q15_t in;                                      /* Temporary variable to hold the input value */
+  q15_t inA1, inB1, inA2, inB2;
+
+#endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numCols != pSrcB->numRows) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+  {
+    /* Matrix transpose */
+    do
+    {
+      /* Apply loop unrolling and exchange the columns with row elements */
+      col = numColsB >> 2;
+
+      /* The pointer px is set to starting address of the column being processed */
+      px = pSrcBT + i;
+
+      /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+       ** a second loop below computes the remaining 1 to 3 samples. */
+      while (col > 0U)
+      {
+#ifndef UNALIGNED_SUPPORT_DISABLE
+
+        /* Read two elements from the row */
+        in = *__SIMD32(pInB)++;
+
+        /* Unpack and store one element in the destination */
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *px = (q15_t) in;
+
+#else
+
+        *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#endif /* #ifndef ARM_MATH_BIG_ENDIAN */
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Unpack and store the second element in the destination */
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#else
+
+        *px = (q15_t) in;
+
+#endif /* #ifndef ARM_MATH_BIG_ENDIAN */
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Read two elements from the row */
+        in = *__SIMD32(pInB)++;
+
+        /* Unpack and store one element in the destination */
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *px = (q15_t) in;
+
+#else
+
+        *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#endif /* #ifndef ARM_MATH_BIG_ENDIAN */
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Unpack and store the second element in the destination */
+
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *px = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#else
+
+        *px = (q15_t) in;
+
+#endif /* #ifndef ARM_MATH_BIG_ENDIAN */
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+#else
+
+        /* Read one element from the row */
+        in = *pInB++;
+
+        /* Store one element in the destination */
+        *px = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Read one element from the row */
+        in = *pInB++;
+
+        /* Store one element in the destination */
+        *px = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Read one element from the row */
+        in = *pInB++;
+
+        /* Store one element in the destination */
+        *px = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Read one element from the row */
+        in = *pInB++;
+
+        /* Store one element in the destination */
+        *px = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+#endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */
+
+       /* Decrement the column loop counter */
+        col--;
+      }
+
+      /* If the columns of pSrcB is not a multiple of 4, compute any remaining output samples here.
+       ** No loop unrolling is used. */
+      col = numColsB % 0x4U;
+
+      while (col > 0U)
+      {
+        /* Read and store the input element in the destination */
+        *px = *pInB++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += numRowsB;
+
+        /* Decrement the column loop counter */
+        col--;
+      }
+
+      i++;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);
+
+    /* Reset the variables for the usage in the following multiplication process */
+    row = numRowsA;
+    i = 0U;
+    px = pDst->pData;
+
+    /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
+    /* row loop */
+    do
+    {
+      /* For every row wise process, the column loop counter is to be initiated */
+      col = numColsB;
+
+      /* For every row wise process, the pIn2 pointer is set
+       ** to the starting address of the transposed pSrcB data */
+      pInB = pSrcBT;
+
+      /* column loop */
+      do
+      {
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum = 0;
+
+        /* Apply loop unrolling and compute 2 MACs simultaneously. */
+        colCnt = numColsA >> 2;
+
+        /* Initiate the pointer pIn1 to point to the starting address of the column being processed */
+        pInA = pSrcA->pData + i;
+
+
+        /* matrix multiplication */
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+#ifndef UNALIGNED_SUPPORT_DISABLE
+
+          /* read real and imag values from pSrcA and pSrcB buffer */
+          pSourceA1 = *__SIMD32(pInA)++;
+          pSourceB1 = *__SIMD32(pInB)++;
+
+          pSourceA2 = *__SIMD32(pInA)++;
+          pSourceB2 = *__SIMD32(pInB)++;
+
+          /* Multiply and Accumlates */
+          sum = __SMLALD(pSourceA1, pSourceB1, sum);
+          sum = __SMLALD(pSourceA2, pSourceB2, sum);
+
+#else
+          /* read real and imag values from pSrcA and pSrcB buffer */
+          inA1 = *pInA++;
+          inB1 = *pInB++;
+          inA2 = *pInA++;
+          /* Multiply and Accumlates */
+          sum += inA1 * inB1;
+          inB2 = *pInB++;
+
+          inA1 = *pInA++;
+          inB1 = *pInB++;
+          /* Multiply and Accumlates */
+          sum += inA2 * inB2;
+          inA2 = *pInA++;
+          inB2 = *pInB++;
+
+          /* Multiply and Accumlates */
+          sum += inA1 * inB1;
+          sum += inA2 * inB2;
+
+#endif /* #ifndef UNALIGNED_SUPPORT_DISABLE */
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        /* process remaining column samples */
+        colCnt = numColsA & 3U;
+
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          sum += *pInA++ * *pInB++;
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        /* Saturate and store the result in the destination buffer */
+        *px = (q15_t) (__SSAT((sum >> 15), 16));
+        px++;
+
+        /* Decrement the column loop counter */
+        col--;
+
+      } while (col > 0U);
+
+      i = i + numColsA;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);
+
+#else
+
+  /* Run the below code for Cortex-M0 */
+
+  q15_t *pIn1 = pSrcA->pData;                    /* input data matrix pointer A */
+  q15_t *pIn2 = pSrcB->pData;                    /* input data matrix pointer B */
+  q15_t *pInA = pSrcA->pData;                    /* input data matrix pointer A of Q15 type */
+  q15_t *pInB = pSrcB->pData;                    /* input data matrix pointer B of Q15 type */
+  q15_t *pOut = pDst->pData;                     /* output data matrix pointer */
+  q15_t *px;                                     /* Temporary output data matrix pointer */
+  uint16_t numColsB = pSrcB->numCols;            /* number of columns of input matrix B */
+  uint16_t numColsA = pSrcA->numCols;            /* number of columns of input matrix A */
+  uint16_t numRowsA = pSrcA->numRows;            /* number of rows of input matrix A    */
+  uint16_t col, i = 0U, row = numRowsA, colCnt;  /* loop counters */
+  arm_status status;                             /* status of matrix multiplication */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numCols != pSrcB->numRows) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /* #ifdef ARM_MATH_MATRIX_CHECK */
+
+  {
+    /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
+    /* row loop */
+    do
+    {
+      /* Output pointer is set to starting address of the row being processed */
+      px = pOut + i;
+
+      /* For every row wise process, the column loop counter is to be initiated */
+      col = numColsB;
+
+      /* For every row wise process, the pIn2 pointer is set
+       ** to the starting address of the pSrcB data */
+      pIn2 = pSrcB->pData;
+
+      /* column loop */
+      do
+      {
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum = 0;
+
+        /* Initiate the pointer pIn1 to point to the starting address of pSrcA */
+        pIn1 = pInA;
+
+        /* Matrix A columns number of MAC operations are to be performed */
+        colCnt = numColsA;
+
+        /* matrix multiplication */
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          /* Perform the multiply-accumulates */
+          sum += (q31_t) * pIn1++ * *pIn2;
+          pIn2 += numColsB;
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        /* Convert the result from 34.30 to 1.15 format and store the saturated value in destination buffer */
+        /* Saturate and store the result in the destination buffer */
+        *px++ = (q15_t) __SSAT((sum >> 15), 16);
+
+        /* Decrement the column loop counter */
+        col--;
+
+        /* Update the pointer pIn2 to point to the  starting address of the next column */
+        pIn2 = pInB + (numColsB - col);
+
+      } while (col > 0U);
+
+      /* Update the pointer pSrcA to point to the  starting address of the next row */
+      i = i + numColsB;
+      pInA = pInA + numColsA;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);
+
+#endif /* #if defined (ARM_MATH_DSP) */
+    /* set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixMult group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_mult_q31.c b/DSP/Source/MatrixFunctions/arm_mat_mult_q31.c
new file mode 100644
index 0000000..9bd2b97
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_mult_q31.c
@@ -0,0 +1,282 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_mult_q31.c
+ * Description:  Q31 matrix multiplication
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixMult
+ * @{
+ */
+
+/**
+ * @brief Q31 matrix multiplication
+ * @param[in]       *pSrcA points to the first input matrix structure
+ * @param[in]       *pSrcB points to the second input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * @details
+ * <b>Scaling and Overflow Behavior:</b>
+ *
+ * \par
+ * The function is implemented using an internal 64-bit accumulator.
+ * The accumulator has a 2.62 format and maintains full precision of the intermediate
+ * multiplication results but provides only a single guard bit. There is no saturation
+ * on intermediate additions. Thus, if the accumulator overflows it wraps around and
+ * distorts the result. The input signals should be scaled down to avoid intermediate
+ * overflows. The input is thus scaled down by log2(numColsA) bits
+ * to avoid overflows, as a total of numColsA additions are performed internally.
+ * The 2.62 accumulator is right shifted by 31 bits and saturated to 1.31 format to yield the final result.
+ *
+ * \par
+ * See <code>arm_mat_mult_fast_q31()</code> for a faster but less precise implementation of this function for Cortex-M3 and Cortex-M4.
+ *
+ */
+
+arm_status arm_mat_mult_q31(
+  const arm_matrix_instance_q31 * pSrcA,
+  const arm_matrix_instance_q31 * pSrcB,
+  arm_matrix_instance_q31 * pDst)
+{
+  q31_t *pIn1 = pSrcA->pData;                    /* input data matrix pointer A */
+  q31_t *pIn2 = pSrcB->pData;                    /* input data matrix pointer B */
+  q31_t *pInA = pSrcA->pData;                    /* input data matrix pointer A */
+  q31_t *pOut = pDst->pData;                     /* output data matrix pointer */
+  q31_t *px;                                     /* Temporary output data matrix pointer */
+  q63_t sum;                                     /* Accumulator */
+  uint16_t numRowsA = pSrcA->numRows;            /* number of rows of input matrix A    */
+  uint16_t numColsB = pSrcB->numCols;            /* number of columns of input matrix B */
+  uint16_t numColsA = pSrcA->numCols;            /* number of columns of input matrix A */
+
+#if defined (ARM_MATH_DSP)
+
+  /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+  uint16_t col, i = 0U, j, row = numRowsA, colCnt;      /* loop counters */
+  arm_status status;                             /* status of matrix multiplication */
+  q31_t a0, a1, a2, a3, b0, b1, b2, b3;
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numCols != pSrcB->numRows) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
+    /* row loop */
+    do
+    {
+      /* Output pointer is set to starting address of the row being processed */
+      px = pOut + i;
+
+      /* For every row wise process, the column loop counter is to be initiated */
+      col = numColsB;
+
+      /* For every row wise process, the pIn2 pointer is set
+       ** to the starting address of the pSrcB data */
+      pIn2 = pSrcB->pData;
+
+      j = 0U;
+
+      /* column loop */
+      do
+      {
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum = 0;
+
+        /* Initiate the pointer pIn1 to point to the starting address of pInA */
+        pIn1 = pInA;
+
+        /* Apply loop unrolling and compute 4 MACs simultaneously. */
+        colCnt = numColsA >> 2;
+
+
+        /* matrix multiplication */
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          /* Perform the multiply-accumulates */
+          b0 = *pIn2;
+          pIn2 += numColsB;
+
+          a0 = *pIn1++;
+          a1 = *pIn1++;
+
+          b1 = *pIn2;
+          pIn2 += numColsB;
+          b2 = *pIn2;
+          pIn2 += numColsB;
+
+          sum += (q63_t) a0 *b0;
+          sum += (q63_t) a1 *b1;
+
+          a2 = *pIn1++;
+          a3 = *pIn1++;
+
+          b3 = *pIn2;
+          pIn2 += numColsB;
+
+          sum += (q63_t) a2 *b2;
+          sum += (q63_t) a3 *b3;
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        /* If the columns of pSrcA is not a multiple of 4, compute any remaining output samples here.
+         ** No loop unrolling is used. */
+        colCnt = numColsA % 0x4U;
+
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          /* Perform the multiply-accumulates */
+          sum += (q63_t) * pIn1++ * *pIn2;
+          pIn2 += numColsB;
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        /* Convert the result from 2.62 to 1.31 format and store in destination buffer */
+        *px++ = (q31_t) (sum >> 31);
+
+        /* Update the pointer pIn2 to point to the  starting address of the next column */
+        j++;
+        pIn2 = (pSrcB->pData) + j;
+
+        /* Decrement the column loop counter */
+        col--;
+
+      } while (col > 0U);
+
+#else
+
+  /* Run the below code for Cortex-M0 */
+
+  q31_t *pInB = pSrcB->pData;                    /* input data matrix pointer B */
+  uint16_t col, i = 0U, row = numRowsA, colCnt;  /* loop counters */
+  arm_status status;                             /* status of matrix multiplication */
+
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numCols != pSrcB->numRows) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcB->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* The following loop performs the dot-product of each row in pSrcA with each column in pSrcB */
+    /* row loop */
+    do
+    {
+      /* Output pointer is set to starting address of the row being processed */
+      px = pOut + i;
+
+      /* For every row wise process, the column loop counter is to be initiated */
+      col = numColsB;
+
+      /* For every row wise process, the pIn2 pointer is set
+       ** to the starting address of the pSrcB data */
+      pIn2 = pSrcB->pData;
+
+      /* column loop */
+      do
+      {
+        /* Set the variable sum, that acts as accumulator, to zero */
+        sum = 0;
+
+        /* Initiate the pointer pIn1 to point to the starting address of pInA */
+        pIn1 = pInA;
+
+        /* Matrix A columns number of MAC operations are to be performed */
+        colCnt = numColsA;
+
+        /* matrix multiplication */
+        while (colCnt > 0U)
+        {
+          /* c(m,n) = a(1,1)*b(1,1) + a(1,2) * b(2,1) + .... + a(m,p)*b(p,n) */
+          /* Perform the multiply-accumulates */
+          sum += (q63_t) * pIn1++ * *pIn2;
+          pIn2 += numColsB;
+
+          /* Decrement the loop counter */
+          colCnt--;
+        }
+
+        /* Convert the result from 2.62 to 1.31 format and store in destination buffer */
+        *px++ = (q31_t) clip_q63_to_q31(sum >> 31);
+
+        /* Decrement the column loop counter */
+        col--;
+
+        /* Update the pointer pIn2 to point to the  starting address of the next column */
+        pIn2 = pInB + (numColsB - col);
+
+      } while (col > 0U);
+
+#endif
+
+      /* Update the pointer pInA to point to the  starting address of the next row */
+      i = i + numColsB;
+      pInA = pInA + numColsA;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);
+
+    /* set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixMult group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_scale_f32.c b/DSP/Source/MatrixFunctions/arm_mat_scale_f32.c
new file mode 100644
index 0000000..dbc385a
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_scale_f32.c
@@ -0,0 +1,169 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_scale_f32.c
+ * Description:  Multiplies a floating-point matrix by a scalar
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @defgroup MatrixScale Matrix Scale
+ *
+ * Multiplies a matrix by a scalar.  This is accomplished by multiplying each element in the
+ * matrix by the scalar.  For example:
+ * \image html MatrixScale.gif "Matrix Scaling of a 3 x 3 matrix"
+ *
+ * The function checks to make sure that the input and output matrices are of the same size.
+ *
+ * In the fixed-point Q15 and Q31 functions, <code>scale</code> is represented by
+ * a fractional multiplication <code>scaleFract</code> and an arithmetic shift <code>shift</code>.
+ * The shift allows the gain of the scaling operation to exceed 1.0.
+ * The overall scale factor applied to the fixed-point data is
+ * <pre>
+ *     scale = scaleFract * 2^shift.
+ * </pre>
+ */
+
+/**
+ * @addtogroup MatrixScale
+ * @{
+ */
+
+/**
+ * @brief Floating-point matrix scaling.
+ * @param[in]       *pSrc points to input matrix structure
+ * @param[in]       scale scale factor to be applied
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns either <code>ARM_MATH_SIZE_MISMATCH</code>
+ * or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ */
+
+arm_status arm_mat_scale_f32(
+  const arm_matrix_instance_f32 * pSrc,
+  float32_t scale,
+  arm_matrix_instance_f32 * pDst)
+{
+  float32_t *pIn = pSrc->pData;                  /* input data matrix pointer */
+  float32_t *pOut = pDst->pData;                 /* output data matrix pointer */
+  uint32_t numSamples;                           /* total number of elements in the matrix */
+  uint32_t blkCnt;                               /* loop counters */
+  arm_status status;                             /* status of matrix scaling     */
+
+#if defined (ARM_MATH_DSP)
+
+  float32_t in1, in2, in3, in4;                  /* temporary variables */
+  float32_t out1, out2, out3, out4;              /* temporary variables */
+
+#endif //      #if defined (ARM_MATH_DSP)
+
+#ifdef ARM_MATH_MATRIX_CHECK
+  /* Check for matrix mismatch condition */
+  if ((pSrc->numRows != pDst->numRows) || (pSrc->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+  {
+    /* Total number of samples in the input matrix */
+    numSamples = (uint32_t) pSrc->numRows * pSrc->numCols;
+
+#if defined (ARM_MATH_DSP)
+
+    /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+    /* Loop Unrolling */
+    blkCnt = numSamples >> 2;
+
+    /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+     ** a second loop below computes the remaining 1 to 3 samples. */
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) * scale */
+      /* Scaling and results are stored in the destination buffer. */
+      in1 = pIn[0];
+      in2 = pIn[1];
+      in3 = pIn[2];
+      in4 = pIn[3];
+
+      out1 = in1 * scale;
+      out2 = in2 * scale;
+      out3 = in3 * scale;
+      out4 = in4 * scale;
+
+
+      pOut[0] = out1;
+      pOut[1] = out2;
+      pOut[2] = out3;
+      pOut[3] = out4;
+
+      /* update pointers to process next sampels */
+      pIn += 4U;
+      pOut += 4U;
+
+      /* Decrement the numSamples loop counter */
+      blkCnt--;
+    }
+
+    /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
+     ** No loop unrolling is used. */
+    blkCnt = numSamples % 0x4U;
+
+#else
+
+    /* Run the below code for Cortex-M0 */
+
+    /* Initialize blkCnt with number of samples */
+    blkCnt = numSamples;
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) * scale */
+      /* The results are stored in the destination buffer. */
+      *pOut++ = (*pIn++) * scale;
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixScale group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_scale_q15.c b/DSP/Source/MatrixFunctions/arm_mat_scale_q15.c
new file mode 100644
index 0000000..af664ca
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_scale_q15.c
@@ -0,0 +1,171 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_scale_q15.c
+ * Description:  Multiplies a Q15 matrix by a scalar
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixScale
+ * @{
+ */
+
+/**
+ * @brief Q15 matrix scaling.
+ * @param[in]       *pSrc points to input matrix
+ * @param[in]       scaleFract fractional portion of the scale factor
+ * @param[in]       shift number of bits to shift the result by
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * @details
+ * <b>Scaling and Overflow Behavior:</b>
+ * \par
+ * The input data <code>*pSrc</code> and <code>scaleFract</code> are in 1.15 format.
+ * These are multiplied to yield a 2.30 intermediate result and this is shifted with saturation to 1.15 format.
+ */
+
+arm_status arm_mat_scale_q15(
+  const arm_matrix_instance_q15 * pSrc,
+  q15_t scaleFract,
+  int32_t shift,
+  arm_matrix_instance_q15 * pDst)
+{
+  q15_t *pIn = pSrc->pData;                      /* input data matrix pointer */
+  q15_t *pOut = pDst->pData;                     /* output data matrix pointer */
+  uint32_t numSamples;                           /* total number of elements in the matrix */
+  int32_t totShift = 15 - shift;                 /* total shift to apply after scaling */
+  uint32_t blkCnt;                               /* loop counters */
+  arm_status status;                             /* status of matrix scaling     */
+
+#if defined (ARM_MATH_DSP)
+
+  q15_t in1, in2, in3, in4;
+  q31_t out1, out2, out3, out4;
+  q31_t inA1, inA2;
+
+#endif //     #if defined (ARM_MATH_DSP)
+
+#ifdef ARM_MATH_MATRIX_CHECK
+  /* Check for matrix mismatch */
+  if ((pSrc->numRows != pDst->numRows) || (pSrc->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif //    #ifdef ARM_MATH_MATRIX_CHECK
+  {
+    /* Total number of samples in the input matrix */
+    numSamples = (uint32_t) pSrc->numRows * pSrc->numCols;
+
+#if defined (ARM_MATH_DSP)
+
+    /* Run the below code for Cortex-M4 and Cortex-M3 */
+    /* Loop Unrolling */
+    blkCnt = numSamples >> 2;
+
+    /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+     ** a second loop below computes the remaining 1 to 3 samples. */
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) * k */
+      /* Scale, saturate and then store the results in the destination buffer. */
+      /* Reading 2 inputs from memory */
+      inA1 = _SIMD32_OFFSET(pIn);
+      inA2 = _SIMD32_OFFSET(pIn + 2);
+
+      /* C = A * scale */
+      /* Scale the inputs and then store the 2 results in the destination buffer
+       * in single cycle by packing the outputs */
+      out1 = (q31_t) ((q15_t) (inA1 >> 16) * scaleFract);
+      out2 = (q31_t) ((q15_t) inA1 * scaleFract);
+      out3 = (q31_t) ((q15_t) (inA2 >> 16) * scaleFract);
+      out4 = (q31_t) ((q15_t) inA2 * scaleFract);
+
+      out1 = out1 >> totShift;
+      inA1 = _SIMD32_OFFSET(pIn + 4);
+      out2 = out2 >> totShift;
+      inA2 = _SIMD32_OFFSET(pIn + 6);
+      out3 = out3 >> totShift;
+      out4 = out4 >> totShift;
+
+      in1 = (q15_t) (__SSAT(out1, 16));
+      in2 = (q15_t) (__SSAT(out2, 16));
+      in3 = (q15_t) (__SSAT(out3, 16));
+      in4 = (q15_t) (__SSAT(out4, 16));
+
+      _SIMD32_OFFSET(pOut) = __PKHBT(in2, in1, 16);
+      _SIMD32_OFFSET(pOut + 2) = __PKHBT(in4, in3, 16);
+
+      /* update pointers to process next sampels */
+      pIn += 4U;
+      pOut += 4U;
+
+
+      /* Decrement the numSamples loop counter */
+      blkCnt--;
+    }
+
+    /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
+     ** No loop unrolling is used. */
+    blkCnt = numSamples % 0x4U;
+
+#else
+
+    /* Run the below code for Cortex-M0 */
+
+    /* Initialize blkCnt with number of samples */
+    blkCnt = numSamples;
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) * k */
+      /* Scale, saturate and then store the results in the destination buffer. */
+      *pOut++ =
+        (q15_t) (__SSAT(((q31_t) (*pIn++) * scaleFract) >> totShift, 16));
+
+      /* Decrement the numSamples loop counter */
+      blkCnt--;
+    }
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixScale group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_scale_q31.c b/DSP/Source/MatrixFunctions/arm_mat_scale_q31.c
new file mode 100644
index 0000000..d190cf1
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_scale_q31.c
@@ -0,0 +1,191 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_scale_q31.c
+ * Description:  Multiplies a Q31 matrix by a scalar
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixScale
+ * @{
+ */
+
+/**
+ * @brief Q31 matrix scaling.
+ * @param[in]       *pSrc points to input matrix
+ * @param[in]       scaleFract fractional portion of the scale factor
+ * @param[in]       shift number of bits to shift the result by
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * @details
+ * <b>Scaling and Overflow Behavior:</b>
+ * \par
+ * The input data <code>*pSrc</code> and <code>scaleFract</code> are in 1.31 format.
+ * These are multiplied to yield a 2.62 intermediate result and this is shifted with saturation to 1.31 format.
+ */
+
+arm_status arm_mat_scale_q31(
+  const arm_matrix_instance_q31 * pSrc,
+  q31_t scaleFract,
+  int32_t shift,
+  arm_matrix_instance_q31 * pDst)
+{
+  q31_t *pIn = pSrc->pData;                      /* input data matrix pointer */
+  q31_t *pOut = pDst->pData;                     /* output data matrix pointer */
+  uint32_t numSamples;                           /* total number of elements in the matrix */
+  int32_t totShift = shift + 1;                  /* shift to apply after scaling */
+  uint32_t blkCnt;                               /* loop counters  */
+  arm_status status;                             /* status of matrix scaling      */
+  q31_t in1, in2, out1;                          /* temporary variabels */
+
+#if defined (ARM_MATH_DSP)
+
+  q31_t in3, in4, out2, out3, out4;              /* temporary variables */
+
+#endif //      #ifndef ARM_MAT_CM0
+
+#ifdef ARM_MATH_MATRIX_CHECK
+  /* Check for matrix mismatch  */
+  if ((pSrc->numRows != pDst->numRows) || (pSrc->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif //    #ifdef ARM_MATH_MATRIX_CHECK
+  {
+    /* Total number of samples in the input matrix */
+    numSamples = (uint32_t) pSrc->numRows * pSrc->numCols;
+
+#if defined (ARM_MATH_DSP)
+
+    /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+    /* Loop Unrolling */
+    blkCnt = numSamples >> 2U;
+
+    /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+     ** a second loop below computes the remaining 1 to 3 samples. */
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) * k */
+      /* Read values from input */
+      in1 = *pIn;
+      in2 = *(pIn + 1);
+      in3 = *(pIn + 2);
+      in4 = *(pIn + 3);
+
+      /* multiply input with scaler value */
+      in1 = ((q63_t) in1 * scaleFract) >> 32;
+      in2 = ((q63_t) in2 * scaleFract) >> 32;
+      in3 = ((q63_t) in3 * scaleFract) >> 32;
+      in4 = ((q63_t) in4 * scaleFract) >> 32;
+
+      /* apply shifting */
+      out1 = in1 << totShift;
+      out2 = in2 << totShift;
+
+      /* saturate the results. */
+      if (in1 != (out1 >> totShift))
+        out1 = 0x7FFFFFFF ^ (in1 >> 31);
+
+      if (in2 != (out2 >> totShift))
+        out2 = 0x7FFFFFFF ^ (in2 >> 31);
+
+      out3 = in3 << totShift;
+      out4 = in4 << totShift;
+
+      *pOut = out1;
+      *(pOut + 1) = out2;
+
+      if (in3 != (out3 >> totShift))
+        out3 = 0x7FFFFFFF ^ (in3 >> 31);
+
+      if (in4 != (out4 >> totShift))
+        out4 = 0x7FFFFFFF ^ (in4 >> 31);
+
+
+      *(pOut + 2) = out3;
+      *(pOut + 3) = out4;
+
+      /* update pointers to process next sampels */
+      pIn += 4U;
+      pOut += 4U;
+
+
+      /* Decrement the numSamples loop counter */
+      blkCnt--;
+    }
+
+    /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
+     ** No loop unrolling is used. */
+    blkCnt = numSamples % 0x4U;
+
+#else
+
+    /* Run the below code for Cortex-M0 */
+
+    /* Initialize blkCnt with number of samples */
+    blkCnt = numSamples;
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) * k */
+      /* Scale, saturate and then store the results in the destination buffer. */
+      in1 = *pIn++;
+
+      in2 = ((q63_t) in1 * scaleFract) >> 32;
+
+      out1 = in2 << totShift;
+
+      if (in2 != (out1 >> totShift))
+        out1 = 0x7FFFFFFF ^ (in2 >> 31);
+
+      *pOut++ = out1;
+
+      /* Decrement the numSamples loop counter */
+      blkCnt--;
+    }
+
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixScale group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_sub_f32.c b/DSP/Source/MatrixFunctions/arm_mat_sub_f32.c
new file mode 100644
index 0000000..7c0b54e
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_sub_f32.c
@@ -0,0 +1,197 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_sub_f32.c
+ * Description:  Floating-point matrix subtraction
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @defgroup MatrixSub Matrix Subtraction
+ *
+ * Subtract two matrices.
+ * \image html MatrixSubtraction.gif "Subraction of two 3 x 3 matrices"
+ *
+ * The functions check to make sure that
+ * <code>pSrcA</code>, <code>pSrcB</code>, and <code>pDst</code> have the same
+ * number of rows and columns.
+ */
+
+/**
+ * @addtogroup MatrixSub
+ * @{
+ */
+
+/**
+ * @brief Floating-point matrix subtraction
+ * @param[in]       *pSrcA points to the first input matrix structure
+ * @param[in]       *pSrcB points to the second input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ */
+
+arm_status arm_mat_sub_f32(
+  const arm_matrix_instance_f32 * pSrcA,
+  const arm_matrix_instance_f32 * pSrcB,
+  arm_matrix_instance_f32 * pDst)
+{
+  float32_t *pIn1 = pSrcA->pData;                /* input data matrix pointer A */
+  float32_t *pIn2 = pSrcB->pData;                /* input data matrix pointer B */
+  float32_t *pOut = pDst->pData;                 /* output data matrix pointer  */
+
+#if defined (ARM_MATH_DSP)
+
+  float32_t inA1, inA2, inB1, inB2, out1, out2;  /* temporary variables */
+
+#endif //      #if defined (ARM_MATH_DSP)
+
+  uint32_t numSamples;                           /* total number of elements in the matrix  */
+  uint32_t blkCnt;                               /* loop counters */
+  arm_status status;                             /* status of matrix subtraction */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numRows != pSrcB->numRows) ||
+     (pSrcA->numCols != pSrcB->numCols) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+  {
+    /* Total number of samples in the input matrix */
+    numSamples = (uint32_t) pSrcA->numRows * pSrcA->numCols;
+
+#if defined (ARM_MATH_DSP)
+
+    /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+    /* Loop Unrolling */
+    blkCnt = numSamples >> 2U;
+
+    /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+     ** a second loop below computes the remaining 1 to 3 samples. */
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) - B(m,n) */
+      /* Subtract and then store the results in the destination buffer. */
+      /* Read values from source A */
+      inA1 = pIn1[0];
+
+      /* Read values from source B */
+      inB1 = pIn2[0];
+
+      /* Read values from source A */
+      inA2 = pIn1[1];
+
+      /* out = sourceA - sourceB */
+      out1 = inA1 - inB1;
+
+      /* Read values from source B */
+      inB2 = pIn2[1];
+
+      /* Read values from source A */
+      inA1 = pIn1[2];
+
+      /* out = sourceA - sourceB */
+      out2 = inA2 - inB2;
+
+      /* Read values from source B */
+      inB1 = pIn2[2];
+
+      /* Store result in destination */
+      pOut[0] = out1;
+      pOut[1] = out2;
+
+      /* Read values from source A */
+      inA2 = pIn1[3];
+
+      /* Read values from source B */
+      inB2 = pIn2[3];
+
+      /* out = sourceA - sourceB */
+      out1 = inA1 - inB1;
+
+
+      /* out = sourceA - sourceB */
+      out2 = inA2 - inB2;
+
+      /* Store result in destination */
+      pOut[2] = out1;
+
+      /* Store result in destination */
+      pOut[3] = out2;
+
+
+      /* update pointers to process next sampels */
+      pIn1 += 4U;
+      pIn2 += 4U;
+      pOut += 4U;
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+    /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
+     ** No loop unrolling is used. */
+    blkCnt = numSamples % 0x4U;
+
+#else
+
+    /* Run the below code for Cortex-M0 */
+
+    /* Initialize blkCnt with number of samples */
+    blkCnt = numSamples;
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) - B(m,n) */
+      /* Subtract and then store the results in the destination buffer. */
+      *pOut++ = (*pIn1++) - (*pIn2++);
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixSub group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_sub_q15.c b/DSP/Source/MatrixFunctions/arm_mat_sub_q15.c
new file mode 100644
index 0000000..28e659f
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_sub_q15.c
@@ -0,0 +1,148 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_sub_q15.c
+ * Description:  Q15 Matrix subtraction
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixSub
+ * @{
+ */
+
+/**
+ * @brief Q15 matrix subtraction.
+ * @param[in]       *pSrcA points to the first input matrix structure
+ * @param[in]       *pSrcB points to the second input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * <b>Scaling and Overflow Behavior:</b>
+ * \par
+ * The function uses saturating arithmetic.
+ * Results outside of the allowable Q15 range [0x8000 0x7FFF] will be saturated.
+ */
+
+arm_status arm_mat_sub_q15(
+  const arm_matrix_instance_q15 * pSrcA,
+  const arm_matrix_instance_q15 * pSrcB,
+  arm_matrix_instance_q15 * pDst)
+{
+  q15_t *pInA = pSrcA->pData;                    /* input data matrix pointer A */
+  q15_t *pInB = pSrcB->pData;                    /* input data matrix pointer B */
+  q15_t *pOut = pDst->pData;                     /* output data matrix pointer */
+  uint32_t numSamples;                           /* total number of elements in the matrix */
+  uint32_t blkCnt;                               /* loop counters  */
+  arm_status status;                             /* status of matrix subtraction  */
+
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+
+  /* Check for matrix mismatch condition */
+  if ((pSrcA->numRows != pSrcB->numRows) ||
+     (pSrcA->numCols != pSrcB->numCols) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* Total number of samples in the input matrix */
+    numSamples = (uint32_t) pSrcA->numRows * pSrcA->numCols;
+
+#if defined (ARM_MATH_DSP)
+
+    /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+    /* Apply loop unrolling */
+    blkCnt = numSamples >> 2U;
+
+    /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+     ** a second loop below computes the remaining 1 to 3 samples. */
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) - B(m,n) */
+      /* Subtract, Saturate and then store the results in the destination buffer. */
+      *__SIMD32(pOut)++ = __QSUB16(*__SIMD32(pInA)++, *__SIMD32(pInB)++);
+      *__SIMD32(pOut)++ = __QSUB16(*__SIMD32(pInA)++, *__SIMD32(pInB)++);
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+    /* If the blockSize is not a multiple of 4, compute any remaining output samples here.
+     ** No loop unrolling is used. */
+    blkCnt = numSamples % 0x4U;
+
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) - B(m,n) */
+      /* Subtract and then store the results in the destination buffer. */
+      *pOut++ = (q15_t) __QSUB16(*pInA++, *pInB++);
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+#else
+
+    /* Run the below code for Cortex-M0 */
+
+    /* Initialize blkCnt with number of samples */
+    blkCnt = numSamples;
+
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) - B(m,n) */
+      /* Subtract and then store the results in the destination buffer. */
+      *pOut++ = (q15_t) __SSAT(((q31_t) * pInA++ - *pInB++), 16);
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixSub group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_sub_q31.c b/DSP/Source/MatrixFunctions/arm_mat_sub_q31.c
new file mode 100644
index 0000000..3bf5508
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_sub_q31.c
@@ -0,0 +1,196 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_sub_q31.c
+ * Description:  Q31 matrix subtraction
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixSub
+ * @{
+ */
+
+/**
+ * @brief Q31 matrix subtraction.
+ * @param[in]       *pSrcA points to the first input matrix structure
+ * @param[in]       *pSrcB points to the second input matrix structure
+ * @param[out]      *pDst points to output matrix structure
+ * @return     		The function returns either
+ * <code>ARM_MATH_SIZE_MISMATCH</code> or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ *
+ * <b>Scaling and Overflow Behavior:</b>
+ * \par
+ * The function uses saturating arithmetic.
+ * Results outside of the allowable Q31 range [0x80000000 0x7FFFFFFF] will be saturated.
+ */
+
+
+arm_status arm_mat_sub_q31(
+  const arm_matrix_instance_q31 * pSrcA,
+  const arm_matrix_instance_q31 * pSrcB,
+  arm_matrix_instance_q31 * pDst)
+{
+  q31_t *pIn1 = pSrcA->pData;                    /* input data matrix pointer A */
+  q31_t *pIn2 = pSrcB->pData;                    /* input data matrix pointer B */
+  q31_t *pOut = pDst->pData;                     /* output data matrix pointer */
+  q31_t inA1, inB1;                              /* temporary variables */
+
+#if defined (ARM_MATH_DSP)
+
+  q31_t inA2, inB2;                              /* temporary variables */
+  q31_t out1, out2;                              /* temporary variables */
+
+#endif //      #if defined (ARM_MATH_DSP)
+
+  uint32_t numSamples;                           /* total number of elements in the matrix  */
+  uint32_t blkCnt;                               /* loop counters */
+  arm_status status;                             /* status of matrix subtraction */
+
+
+#ifdef ARM_MATH_MATRIX_CHECK
+  /* Check for matrix mismatch condition  */
+  if ((pSrcA->numRows != pSrcB->numRows) ||
+     (pSrcA->numCols != pSrcB->numCols) ||
+     (pSrcA->numRows != pDst->numRows) || (pSrcA->numCols != pDst->numCols))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif
+  {
+    /* Total number of samples in the input matrix */
+    numSamples = (uint32_t) pSrcA->numRows * pSrcA->numCols;
+
+#if defined (ARM_MATH_DSP)
+
+    /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+    /* Loop Unrolling */
+    blkCnt = numSamples >> 2U;
+
+    /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+     ** a second loop below computes the remaining 1 to 3 samples. */
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) - B(m,n) */
+      /* Subtract, saturate and then store the results in the destination buffer. */
+      /* Read values from source A */
+      inA1 = pIn1[0];
+
+      /* Read values from source B */
+      inB1 = pIn2[0];
+
+      /* Read values from source A */
+      inA2 = pIn1[1];
+
+      /* Subtract and saturate */
+      out1 = __QSUB(inA1, inB1);
+
+      /* Read values from source B */
+      inB2 = pIn2[1];
+
+      /* Read values from source A */
+      inA1 = pIn1[2];
+
+      /* Subtract and saturate */
+      out2 = __QSUB(inA2, inB2);
+
+      /* Read values from source B */
+      inB1 = pIn2[2];
+
+      /* Store result in destination */
+      pOut[0] = out1;
+      pOut[1] = out2;
+
+      /* Read values from source A */
+      inA2 = pIn1[3];
+
+      /* Read values from source B */
+      inB2 = pIn2[3];
+
+      /* Subtract and saturate */
+      out1 = __QSUB(inA1, inB1);
+
+      /* Subtract and saturate */
+      out2 = __QSUB(inA2, inB2);
+
+      /* Store result in destination */
+      pOut[2] = out1;
+      pOut[3] = out2;
+
+      /* update pointers to process next samples */
+      pIn1 += 4U;
+      pIn2 += 4U;
+      pOut += 4U;
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+    /* If the numSamples is not a multiple of 4, compute any remaining output samples here.
+     ** No loop unrolling is used. */
+    blkCnt = numSamples % 0x4U;
+
+#else
+
+    /* Run the below code for Cortex-M0 */
+
+    /* Initialize blkCnt with number of samples */
+    blkCnt = numSamples;
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+    while (blkCnt > 0U)
+    {
+      /* C(m,n) = A(m,n) - B(m,n) */
+      /* Subtract, saturate and then store the results in the destination buffer. */
+      inA1 = *pIn1++;
+      inB1 = *pIn2++;
+
+      inA1 = __QSUB(inA1, inB1);
+
+      *pOut++ = inA1;
+
+      /* Decrement the loop counter */
+      blkCnt--;
+    }
+
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixSub group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_trans_f32.c b/DSP/Source/MatrixFunctions/arm_mat_trans_f32.c
new file mode 100644
index 0000000..84165ce
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_trans_f32.c
@@ -0,0 +1,206 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_trans_f32.c
+ * Description:  Floating-point matrix transpose
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+/**
+ * @defgroup MatrixTrans Matrix Transpose
+ *
+ * Tranposes a matrix.
+ * Transposing an <code>M x N</code> matrix flips it around the center diagonal and results in an <code>N x M</code> matrix.
+ * \image html MatrixTranspose.gif "Transpose of a 3 x 3 matrix"
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixTrans
+ * @{
+ */
+
+/**
+  * @brief Floating-point matrix transpose.
+  * @param[in]  *pSrc points to the input matrix
+  * @param[out] *pDst points to the output matrix
+  * @return 	The function returns either  <code>ARM_MATH_SIZE_MISMATCH</code>
+  * or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+  */
+
+
+arm_status arm_mat_trans_f32(
+  const arm_matrix_instance_f32 * pSrc,
+  arm_matrix_instance_f32 * pDst)
+{
+  float32_t *pIn = pSrc->pData;                  /* input data matrix pointer */
+  float32_t *pOut = pDst->pData;                 /* output data matrix pointer */
+  float32_t *px;                                 /* Temporary output data matrix pointer */
+  uint16_t nRows = pSrc->numRows;                /* number of rows */
+  uint16_t nColumns = pSrc->numCols;             /* number of columns */
+
+#if defined (ARM_MATH_DSP)
+
+  /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+  uint16_t blkCnt, i = 0U, row = nRows;          /* loop counters */
+  arm_status status;                             /* status of matrix transpose  */
+
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+
+  /* Check for matrix mismatch condition */
+  if ((pSrc->numRows != pDst->numCols) || (pSrc->numCols != pDst->numRows))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* Matrix transpose by exchanging the rows with columns */
+    /* row loop     */
+    do
+    {
+      /* Loop Unrolling */
+      blkCnt = nColumns >> 2;
+
+      /* The pointer px is set to starting address of the column being processed */
+      px = pOut + i;
+
+      /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+       ** a second loop below computes the remaining 1 to 3 samples. */
+      while (blkCnt > 0U)        /* column loop */
+      {
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Decrement the column loop counter */
+        blkCnt--;
+      }
+
+      /* Perform matrix transpose for last 3 samples here. */
+      blkCnt = nColumns % 0x4U;
+
+      while (blkCnt > 0U)
+      {
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Decrement the column loop counter */
+        blkCnt--;
+      }
+
+#else
+
+  /* Run the below code for Cortex-M0 */
+
+  uint16_t col, i = 0U, row = nRows;             /* loop counters */
+  arm_status status;                             /* status of matrix transpose  */
+
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+  /* Check for matrix mismatch condition */
+  if ((pSrc->numRows != pDst->numCols) || (pSrc->numCols != pDst->numRows))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*      #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* Matrix transpose by exchanging the rows with columns */
+    /* row loop     */
+    do
+    {
+      /* The pointer px is set to starting address of the column being processed */
+      px = pOut + i;
+
+      /* Initialize column loop counter */
+      col = nColumns;
+
+      while (col > 0U)
+      {
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Decrement the column loop counter */
+        col--;
+      }
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+      i++;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);          /* row loop end  */
+
+    /* Set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixTrans group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_trans_q15.c b/DSP/Source/MatrixFunctions/arm_mat_trans_q15.c
new file mode 100644
index 0000000..6ba0904
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_trans_q15.c
@@ -0,0 +1,272 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_trans_q15.c
+ * Description:  Q15 matrix transpose
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixTrans
+ * @{
+ */
+
+/*
+ * @brief Q15 matrix transpose.
+ * @param[in]  *pSrc points to the input matrix
+ * @param[out] *pDst points to the output matrix
+ * @return 	The function returns either  <code>ARM_MATH_SIZE_MISMATCH</code>
+ * or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ */
+
+arm_status arm_mat_trans_q15(
+  const arm_matrix_instance_q15 * pSrc,
+  arm_matrix_instance_q15 * pDst)
+{
+  q15_t *pSrcA = pSrc->pData;                    /* input data matrix pointer */
+  q15_t *pOut = pDst->pData;                     /* output data matrix pointer */
+  uint16_t nRows = pSrc->numRows;                /* number of nRows */
+  uint16_t nColumns = pSrc->numCols;             /* number of nColumns */
+  uint16_t col, row = nRows, i = 0U;             /* row and column loop counters */
+  arm_status status;                             /* status of matrix transpose */
+
+#if defined (ARM_MATH_DSP)
+
+  /* Run the below code for Cortex-M4 and Cortex-M3 */
+#ifndef UNALIGNED_SUPPORT_DISABLE
+
+  q31_t in;                                      /* variable to hold temporary output  */
+
+#else
+
+  q15_t in;
+
+#endif	/*	#ifndef UNALIGNED_SUPPORT_DISABLE	*/
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+
+  /* Check for matrix mismatch condition */
+  if ((pSrc->numRows != pDst->numCols) || (pSrc->numCols != pDst->numRows))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*      #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* Matrix transpose by exchanging the rows with columns */
+    /* row loop     */
+    do
+    {
+
+      /* Apply loop unrolling and exchange the columns with row elements */
+      col = nColumns >> 2U;
+
+      /* The pointer pOut is set to starting address of the column being processed */
+      pOut = pDst->pData + i;
+
+      /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+       ** a second loop below computes the remaining 1 to 3 samples. */
+      while (col > 0U)
+      {
+#ifndef UNALIGNED_SUPPORT_DISABLE
+
+        /* Read two elements from the row */
+        in = *__SIMD32(pSrcA)++;
+
+        /* Unpack and store one element in the destination */
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *pOut = (q15_t) in;
+
+#else
+
+        *pOut = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#endif /*    #ifndef ARM_MATH_BIG_ENDIAN    */
+
+        /* Update the pointer pOut to point to the next row of the transposed matrix */
+        pOut += nRows;
+
+        /* Unpack and store the second element in the destination */
+
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *pOut = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#else
+
+        *pOut = (q15_t) in;
+
+#endif /*    #ifndef ARM_MATH_BIG_ENDIAN    */
+
+        /* Update the pointer pOut to point to the next row of the transposed matrix */
+        pOut += nRows;
+
+        /* Read two elements from the row */
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        in = *__SIMD32(pSrcA)++;
+
+#else
+
+        in = *__SIMD32(pSrcA)++;
+
+#endif /*    #ifndef ARM_MATH_BIG_ENDIAN    */
+
+        /* Unpack and store one element in the destination */
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *pOut = (q15_t) in;
+
+#else
+
+        *pOut = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#endif /*    #ifndef ARM_MATH_BIG_ENDIAN    */
+
+        /* Update the pointer pOut to point to the next row of the transposed matrix */
+        pOut += nRows;
+
+        /* Unpack and store the second element in the destination */
+#ifndef ARM_MATH_BIG_ENDIAN
+
+        *pOut = (q15_t) ((in & (q31_t) 0xffff0000) >> 16);
+
+#else
+
+        *pOut = (q15_t) in;
+
+#endif /*    #ifndef ARM_MATH_BIG_ENDIAN    */
+
+#else
+        /* Read one element from the row */
+        in = *pSrcA++;
+
+        /* Store one element in the destination */
+        *pOut = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        pOut += nRows;
+
+        /* Read one element from the row */
+        in = *pSrcA++;
+
+        /* Store one element in the destination */
+        *pOut = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        pOut += nRows;
+
+        /* Read one element from the row */
+        in = *pSrcA++;
+
+        /* Store one element in the destination */
+        *pOut = in;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        pOut += nRows;
+
+        /* Read one element from the row */
+        in = *pSrcA++;
+
+        /* Store one element in the destination */
+        *pOut = in;
+
+#endif	/*	#ifndef UNALIGNED_SUPPORT_DISABLE	*/
+
+        /* Update the pointer pOut to point to the next row of the transposed matrix */
+        pOut += nRows;
+
+        /* Decrement the column loop counter */
+        col--;
+      }
+
+      /* Perform matrix transpose for last 3 samples here. */
+      col = nColumns % 0x4U;
+
+#else
+
+  /* Run the below code for Cortex-M0 */
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+  /* Check for matrix mismatch condition */
+  if ((pSrc->numRows != pDst->numCols) || (pSrc->numCols != pDst->numRows))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* Matrix transpose by exchanging the rows with columns */
+    /* row loop     */
+    do
+    {
+      /* The pointer pOut is set to starting address of the column being processed */
+      pOut = pDst->pData + i;
+
+      /* Initialize column loop counter */
+      col = nColumns;
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+      while (col > 0U)
+      {
+        /* Read and store the input element in the destination */
+        *pOut = *pSrcA++;
+
+        /* Update the pointer pOut to point to the next row of the transposed matrix */
+        pOut += nRows;
+
+        /* Decrement the column loop counter */
+        col--;
+      }
+
+      i++;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    } while (row > 0U);
+
+    /* set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixTrans group
+ */
diff --git a/DSP/Source/MatrixFunctions/arm_mat_trans_q31.c b/DSP/Source/MatrixFunctions/arm_mat_trans_q31.c
new file mode 100644
index 0000000..6f698e0
--- /dev/null
+++ b/DSP/Source/MatrixFunctions/arm_mat_trans_q31.c
@@ -0,0 +1,198 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mat_trans_q31.c
+ * Description:  Q31 matrix transpose
+ *
+ * $Date:        27. January 2017
+ * $Revision:    V.1.5.1
+ *
+ * Target Processor: Cortex-M cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2017 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "arm_math.h"
+
+/**
+ * @ingroup groupMatrix
+ */
+
+/**
+ * @addtogroup MatrixTrans
+ * @{
+ */
+
+/*
+  * @brief Q31 matrix transpose.
+  * @param[in]  *pSrc points to the input matrix
+  * @param[out] *pDst points to the output matrix
+  * @return 	The function returns either  <code>ARM_MATH_SIZE_MISMATCH</code>
+  * or <code>ARM_MATH_SUCCESS</code> based on the outcome of size checking.
+ */
+
+arm_status arm_mat_trans_q31(
+  const arm_matrix_instance_q31 * pSrc,
+  arm_matrix_instance_q31 * pDst)
+{
+  q31_t *pIn = pSrc->pData;                      /* input data matrix pointer  */
+  q31_t *pOut = pDst->pData;                     /* output data matrix pointer  */
+  q31_t *px;                                     /* Temporary output data matrix pointer */
+  uint16_t nRows = pSrc->numRows;                /* number of nRows */
+  uint16_t nColumns = pSrc->numCols;             /* number of nColumns  */
+
+#if defined (ARM_MATH_DSP)
+
+  /* Run the below code for Cortex-M4 and Cortex-M3 */
+
+  uint16_t blkCnt, i = 0U, row = nRows;          /* loop counters */
+  arm_status status;                             /* status of matrix transpose */
+
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+
+  /* Check for matrix mismatch condition */
+  if ((pSrc->numRows != pDst->numCols) || (pSrc->numCols != pDst->numRows))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* Matrix transpose by exchanging the rows with columns */
+    /* row loop     */
+    do
+    {
+      /* Apply loop unrolling and exchange the columns with row elements */
+      blkCnt = nColumns >> 2U;
+
+      /* The pointer px is set to starting address of the column being processed */
+      px = pOut + i;
+
+      /* First part of the processing with loop unrolling.  Compute 4 outputs at a time.
+       ** a second loop below computes the remaining 1 to 3 samples. */
+      while (blkCnt > 0U)
+      {
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Decrement the column loop counter */
+        blkCnt--;
+      }
+
+      /* Perform matrix transpose for last 3 samples here. */
+      blkCnt = nColumns % 0x4U;
+
+      while (blkCnt > 0U)
+      {
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Decrement the column loop counter */
+        blkCnt--;
+      }
+
+#else
+
+  /* Run the below code for Cortex-M0 */
+
+  uint16_t col, i = 0U, row = nRows;             /* loop counters */
+  arm_status status;                             /* status of matrix transpose */
+
+
+#ifdef ARM_MATH_MATRIX_CHECK
+
+  /* Check for matrix mismatch condition */
+  if ((pSrc->numRows != pDst->numCols) || (pSrc->numCols != pDst->numRows))
+  {
+    /* Set status as ARM_MATH_SIZE_MISMATCH */
+    status = ARM_MATH_SIZE_MISMATCH;
+  }
+  else
+#endif /*    #ifdef ARM_MATH_MATRIX_CHECK    */
+
+  {
+    /* Matrix transpose by exchanging the rows with columns */
+    /* row loop     */
+    do
+    {
+      /* The pointer px is set to starting address of the column being processed */
+      px = pOut + i;
+
+      /* Initialize column loop counter */
+      col = nColumns;
+
+      while (col > 0U)
+      {
+        /* Read and store the input element in the destination */
+        *px = *pIn++;
+
+        /* Update the pointer px to point to the next row of the transposed matrix */
+        px += nRows;
+
+        /* Decrement the column loop counter */
+        col--;
+      }
+
+#endif /* #if defined (ARM_MATH_DSP) */
+
+      i++;
+
+      /* Decrement the row loop counter */
+      row--;
+
+    }
+    while (row > 0U);            /* row loop end */
+
+    /* set status as ARM_MATH_SUCCESS */
+    status = ARM_MATH_SUCCESS;
+  }
+
+  /* Return to application */
+  return (status);
+}
+
+/**
+ * @} end of MatrixTrans group
+ */