summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rw-r--r--lab-windows/dsss_experiments-ber.ipynb505
-rw-r--r--lab-windows/dsss_experiments.ipynb86
2 files changed, 556 insertions, 35 deletions
diff --git a/lab-windows/dsss_experiments-ber.ipynb b/lab-windows/dsss_experiments-ber.ipynb
new file mode 100644
index 0000000..eda21d7
--- /dev/null
+++ b/lab-windows/dsss_experiments-ber.ipynb
@@ -0,0 +1,505 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": 68,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from matplotlib import pyplot as plt\n",
+ "import numpy as np\n",
+ "from scipy import signal as sig\n",
+ "import struct\n",
+ "import random\n",
+ "import ipywidgets\n",
+ "import itertools\n",
+ "from multiprocessing import Pool\n",
+ "\n",
+ "import colorednoise\n",
+ "\n",
+ "np.set_printoptions(linewidth=240)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "%matplotlib widget"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "sampling_rate = 10 # sp/s"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 59,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# From https://github.com/mubeta06/python/blob/master/signal_processing/sp/gold.py\n",
+ "preferred_pairs = {5:[[2],[1,2,3]], 6:[[5],[1,4,5]], 7:[[4],[4,5,6]],\n",
+ " 8:[[1,2,3,6,7],[1,2,7]], 9:[[5],[3,5,6]], \n",
+ " 10:[[2,5,9],[3,4,6,8,9]], 11:[[9],[3,6,9]]}\n",
+ "\n",
+ "def gen_gold(seq1, seq2):\n",
+ " gold = [seq1, seq2]\n",
+ " for shift in range(len(seq1)):\n",
+ " gold.append(seq1 ^ np.roll(seq2, -shift))\n",
+ " return gold\n",
+ "\n",
+ "def gold(n):\n",
+ " n = int(n)\n",
+ " if not n in preferred_pairs:\n",
+ " raise KeyError('preferred pairs for %s bits unknown' % str(n))\n",
+ " t0, t1 = preferred_pairs[n]\n",
+ " (seq0, _st0), (seq1, _st1) = sig.max_len_seq(n, taps=t0), sig.max_len_seq(n, taps=t1)\n",
+ " return gen_gold(seq0, seq1)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def modulate(data, nbits=5):\n",
+ " # 0, 1 -> -1, 1\n",
+ " mask = np.array(gold(nbits))*2 - 1\n",
+ " \n",
+ " sel = mask[data>>1]\n",
+ " data_lsb_centered = ((data&1)*2 - 1)\n",
+ "\n",
+ " return (np.multiply(sel, np.tile(data_lsb_centered, (2**nbits-1, 1)).T).flatten() + 1) // 2"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def correlate(sequence, nbits=5, decimation=1, mask_filter=lambda x: x):\n",
+ " mask = np.tile(np.array(gold(nbits))[:,:,np.newaxis]*2 - 1, (1, 1, decimation)).reshape((2**nbits + 1, (2**nbits-1) * decimation))\n",
+ "\n",
+ " sequence -= np.mean(sequence)\n",
+ " \n",
+ " return np.array([np.correlate(sequence, row, mode='full') for row in mask])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "mean: 49.98625\n"
+ ]
+ }
+ ],
+ "source": [
+ "with open('/mnt/c/Users/jaseg/shared/raw_freq.bin', 'rb') as f:\n",
+ " mains_noise = np.copy(np.frombuffer(f.read(), dtype='float32'))\n",
+ " print('mean:', np.mean(mains_noise))\n",
+ " mains_noise -= np.mean(mains_noise)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 72,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def generate_test_signal(duration, nbits=6, signal_amplitude=2.0e-3, decimation=10, seed=0):\n",
+ " test_data = np.random.RandomState(seed=seed).randint(0, 2 * (2**nbits), duration)\n",
+ " \n",
+ " signal = np.repeat(modulate(test_data, nbits) * 2.0 - 1, decimation) * signal_amplitude\n",
+ " noise = np.resize(mains_noise, len(signal))\n",
+ " \n",
+ " return test_data, signal + noise"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "nonlinear_distance = lambda x: 100**(2*np.abs(0.5-x%1)) / (np.abs(x)+3)**2\n",
+ "\n",
+ "def plot_distance_func():\n",
+ " fig, ax = plt.subplots()\n",
+ " x = np.linspace(-1.5, 5.5, 10000)\n",
+ " ax.plot(x, nonlinear_distance(x))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 38,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "noprint = lambda *args, **kwargs: None"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 74,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def run_ser_test(sample_duration=128, nbits=6, signal_amplitude=2.0e-3, decimation=10, threshold_factor=4.0, power_avg_width=2.5, max_lookahead=6.5, seed=0, ax=None, print=print):\n",
+ "\n",
+ " test_data, signal = generate_test_signal(sample_duration, nbits, signal_amplitude, decimation, seed)\n",
+ " cor_an = correlate(signal, nbits=nbits, decimation=decimation)\n",
+ "\n",
+ " power_avg_width = int(power_avg_width * (2**nbits - 1) * decimation)\n",
+ "\n",
+ " bit_period = (2**nbits) * decimation\n",
+ " peak_group_threshold = 0.1 * bit_period\n",
+ " \n",
+ " cwt_res = np.array([ sig.cwt(row, sig.ricker, [0.73 * decimation]).flatten() for row in cor_an ])\n",
+ " if ax:\n",
+ " ax.grid()\n",
+ " ax.plot(cwt_res.T)\n",
+ " \n",
+ " th = np.array([ np.convolve(np.abs(row), np.ones((power_avg_width,))/power_avg_width, mode='same') for row in cwt_res ])\n",
+ "\n",
+ " def compare_th(elem):\n",
+ " idx, (th, val) = elem\n",
+ " #print('compare_th:', th.shape, val.shape)\n",
+ " return np.any(np.abs(val) > th*threshold_factor)\n",
+ "\n",
+ " peaks = [ list(group) for val, group in itertools.groupby(enumerate(zip(th.T, cwt_res.T)), compare_th) if val ]\n",
+ " peak_group = []\n",
+ " for group in peaks:\n",
+ " pos = np.mean([idx for idx, _val in group])\n",
+ " pol = np.mean([max(val.min(), val.max(), key=abs) for _idx, (_th, val) in group])\n",
+ " pol_idx = np.argmax(np.bincount([ np.argmax(np.abs(val)) for _idx, (_th, val) in group ]))\n",
+ " #print(f'group', pos, pol, pol_idx)\n",
+ " #for pol, (_idx, (_th, val)) in zip([max(val.min(), val.max(), key=abs) for _idx, (_th, val) in group], group):\n",
+ " # print(' ', pol, val)\n",
+ " if ax:\n",
+ " ax.axvline(pos, color='cyan', alpha=0.3)\n",
+ "\n",
+ " if not peak_group or pos - peak_group[-1][1] > peak_group_threshold:\n",
+ " if peak_group:\n",
+ " peak_pos = peak_group[-1][3]\n",
+ " if ax:\n",
+ " ax.axvline(peak_pos, color='red', alpha=0.6)\n",
+ " #ax3.text(peak_pos-20, 2.0, f'{0 if pol < 0 else 1}', horizontalalignment='right', verticalalignment='center', color='black')\n",
+ "\n",
+ " peak_group.append((pos, pos, pol, pos, pol_idx))\n",
+ " #ax3.axvline(pos, color='cyan', alpha=0.5)\n",
+ "\n",
+ " else:\n",
+ " group_start, last_pos, last_pol, peak_pos, last_pol_idx = peak_group[-1]\n",
+ "\n",
+ " if abs(pol) > abs(last_pol):\n",
+ " #ax3.axvline(pos, color='magenta', alpha=0.5)\n",
+ " peak_group[-1] = (group_start, pos, pol, pos, pol_idx)\n",
+ " else:\n",
+ " #ax3.axvline(pos, color='blue', alpha=0.5)\n",
+ " peak_group[-1] = (group_start, pos, last_pol, peak_pos, last_pol_idx)\n",
+ "\n",
+ " avg_peak = np.mean(np.abs(np.array([last_pol for _1, _2, last_pol, _3, _4 in peak_group])))\n",
+ " print('avg_peak', avg_peak)\n",
+ "\n",
+ " noprint = lambda *args, **kwargs: None\n",
+ " def mle_decode(peak_groups, print=print):\n",
+ " peak_groups = [ (pos, pol, idx) for _1, _2, pol, pos, idx in peak_groups ]\n",
+ " candidates = [ (0, [(pos, pol, idx)]) for pos, pol, idx in peak_groups if pos < bit_period*2.5 ]\n",
+ "\n",
+ " while candidates:\n",
+ " chain_candidates = []\n",
+ " for chain_score, chain in candidates:\n",
+ " pos, ampl, _idx = chain[-1]\n",
+ " score_fun = lambda pos, npos, npol: abs(npol)/avg_peak + nonlinear_distance((npos-pos)/bit_period)\n",
+ " next_candidates = sorted([ (score_fun(pos, npos, npol), npos, npol, nidx) for npos, npol, nidx in peak_groups if pos < npos < pos + bit_period*max_lookahead ], reverse=True)\n",
+ "\n",
+ " print(f' candidates for {pos}, {ampl}:')\n",
+ " for score, npos, npol, nidx in next_candidates:\n",
+ " print(f' {score:.4f} {npos:.2f} {npol:.2f} {nidx:.2f}')\n",
+ "\n",
+ " nch, cor_len = cor_an.shape\n",
+ " if cor_len - pos < 1.5*bit_period or not next_candidates:\n",
+ " score = sum(score_fun(opos, npos, npol) for (opos, _opol, _oidx), (npos, npol, _nidx) in zip(chain[:-1], chain[1:])) / len(chain)\n",
+ " yield score, chain\n",
+ "\n",
+ " else:\n",
+ " print('extending')\n",
+ " for score, npos, npol, nidx in next_candidates[:3]:\n",
+ " if score > 0.5:\n",
+ " new_chain_score = chain_score * 0.9 + score * 0.1\n",
+ " chain_candidates.append((new_chain_score, chain + [(npos, npol, nidx)]))\n",
+ " print('chain candidates:')\n",
+ " for score, chain in sorted(chain_candidates, reverse=True):\n",
+ " print(' ', [(score, [(f'{pos:.2f}', f'{pol:.2f}') for pos, pol, _idx in chain])])\n",
+ " candidates = [ (chain_score, chain) for chain_score, chain in sorted(chain_candidates, reverse=True)[:10] ]\n",
+ "\n",
+ " res = sorted(mle_decode(peak_group, print=noprint), reverse=True)\n",
+ " #for i, (score, chain) in enumerate(res):\n",
+ " # print(f'Chain {i}@{score:.4f}: {chain}')\n",
+ " (_score, chain), *_ = res\n",
+ "\n",
+ " def viz(chain):\n",
+ " last_pos = None\n",
+ " for pos, pol, nidx in chain:\n",
+ " if last_pos:\n",
+ " delta = int(round((pos - last_pos) / bit_period))\n",
+ " if delta > 1:\n",
+ " print(f'skipped {delta} symbols at {pos}')\n",
+ " for i in range(delta-1):\n",
+ " yield None\n",
+ " decoded = nidx*2 + (0 if pol < 0 else 1)\n",
+ " yield decoded\n",
+ " if ax:\n",
+ " ax.axvline(pos, color='blue', alpha=0.5)\n",
+ " ax.text(pos-20, 0.0, f'{decoded}', horizontalalignment='right', verticalalignment='center', color='black')\n",
+ "\n",
+ " last_pos = pos\n",
+ "\n",
+ " decoded = list(viz(chain))\n",
+ " print('decoding [ref|dec]:')\n",
+ " failures = 0\n",
+ " for i, (ref, found) in enumerate(itertools.zip_longest(test_data, decoded)):\n",
+ " print(f'{ref or -1:>3d}|{found or -1:>3d} {\"✔\" if ref==found else \"✘\" if found else \" \"}', end=' ')\n",
+ " if ref != found:\n",
+ " failures += 1\n",
+ " if i%8 == 7:\n",
+ " print()\n",
+ " ser = failures/len(test_data)\n",
+ " print(f'Symbol error rate e={ser}')\n",
+ " br = sampling_rate / decimation / (2**nbits) * nbits * (1 - ser) * 3600\n",
+ " print(f'maximum bitrate r={br} b/h')\n",
+ " return ser, br"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 39,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "application/vnd.jupyter.widget-view+json": {
+ "model_id": "9b7546a233fb4b6cb6e8f809250ba768",
+ "version_major": 2,
+ "version_minor": 0
+ },
+ "text/plain": [
+ "Canvas(toolbar=Toolbar(toolitems=[('Home', 'Reset original view', 'home', 'home'), ('Back', 'Back to previous …"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "(63,) (63,)\n",
+ "(63,) (63,)\n",
+ "avg_peak 1.6845488102985742\n",
+ "skipped 2 symbols at 10079.0\n",
+ "skipped 2 symbols at 30870.0\n",
+ "skipped 2 symbols at 42209.5\n",
+ "decoding [ref|dec]:\n",
+ " 44| 44 ✔ 47| 47 ✔ 117|117 ✔ 64| 64 ✔ 67| 67 ✔ 123|123 ✔ 67| 67 ✔ 103|103 ✔ \n",
+ " 9| 9 ✔ 83| 83 ✔ 21| 21 ✔ 114|114 ✔ 36| 36 ✔ 87| 87 ✔ 70| -1 88| 88 ✔ \n",
+ " 88| 88 ✔ 12| 12 ✔ 58| 58 ✔ 65| 65 ✔ 102|102 ✔ 39| 39 ✔ 87| 87 ✔ 46| 46 ✔ \n",
+ " 88| 88 ✔ 81| 81 ✔ 37| 37 ✔ 25| 25 ✔ 77| 77 ✔ 72| 72 ✔ 9| 9 ✔ 20| 20 ✔ \n",
+ "115|115 ✔ 80| 80 ✔ 115|115 ✔ 69| 69 ✔ 126|126 ✔ 79| 79 ✔ 47| 47 ✔ 64| 64 ✔ \n",
+ " 82| 82 ✔ 99| 99 ✔ 88| 88 ✔ 49| 49 ✔ 115|115 ✔ 29| 29 ✔ 19| 19 ✔ 19| -1 \n",
+ " 14| 14 ✔ 39| 39 ✔ 32| 32 ✔ 65| 65 ✔ 9| 9 ✔ 57| 57 ✔ 127|127 ✔ 32| 32 ✔ \n",
+ " 31| 31 ✔ 74| 74 ✔ 116|116 ✔ 23| 23 ✔ 35| 35 ✔ 126|126 ✔ 75| 75 ✔ 114|114 ✔ \n",
+ " 55| 55 ✔ 28| -1 34| 34 ✔ -1| -1 ✔ -1| -1 ✔ 36| 36 ✔ 53| 53 ✔ 5| 5 ✔ \n",
+ " 38| 38 ✔ 104|104 ✔ 116|116 ✔ 17| 17 ✔ 79| 79 ✔ 4| 4 ✔ 105|105 ✔ 42| 42 ✔ \n",
+ " 58| 58 ✔ 31| 31 ✔ 120|120 ✔ 1| 1 ✔ 65| 65 ✔ 103|103 ✔ 41| 41 ✔ 57| 57 ✔ \n",
+ " 35| 35 ✔ 102|103 ✘ 119|119 ✔ 11| 11 ✔ 46| 46 ✔ 82| 82 ✔ 91| 91 ✔ -1| -1 ✔ \n",
+ " 14| 14 ✔ 99| 99 ✔ 53| 53 ✔ 12| 12 ✔ 121|121 ✔ 42| 42 ✔ 84| 84 ✔ 75| 75 ✔ \n",
+ " 68| 68 ✔ 6| 6 ✔ 68| 68 ✔ 47| 47 ✔ 127|127 ✔ 116|116 ✔ 3| 3 ✔ 76| 76 ✔ \n",
+ "100|100 ✔ 52| 52 ✔ 104|104 ✔ 78| 78 ✔ 15| 15 ✔ 20| 20 ✔ 99| 99 ✔ 58| 58 ✔ \n",
+ " 23| 23 ✔ 79| 79 ✔ 13| 13 ✔ 117|117 ✔ 85| 85 ✔ 48| 48 ✔ 49| 49 ✔ 69| 69 ✔ \n",
+ "Symbol error rate e=0.03125\n",
+ "maximum bitrate r=326.953125 b/h\n"
+ ]
+ }
+ ],
+ "source": [
+ "fig, ax = plt.subplots(figsize=(12,5))\n",
+ "run_ser_test(ax=ax)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 84,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "<ipython-input-84-f36c7b0ffb61>:13: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`).\n",
+ " fig, ax = plt.subplots(figsize=(12, 9))\n"
+ ]
+ },
+ {
+ "data": {
+ "application/vnd.jupyter.widget-view+json": {
+ "model_id": "b5eb2cd4f4224f75bc3dc73b6143d849",
+ "version_major": 2,
+ "version_minor": 0
+ },
+ "text/plain": [
+ "Canvas(toolbar=Toolbar(toolitems=[('Home', 'Reset original view', 'home', 'home'), ('Back', 'Back to previous …"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "nbits=5\n",
+ "signal_amplitude=0.00029: ser=1.01000 ±0.012207515615390381, br=-5.62500\n",
+ "signal_amplitude=0.00020: ser=1.01469 ±0.013678792892649557, br=-8.26172\n",
+ "signal_amplitude=0.00052: ser=1.00406 ±0.018895270572288715, br=-2.28516\n",
+ "signal_amplitude=0.00043: ser=1.01000 ±0.018817586521655747, br=-5.62500\n",
+ "signal_amplitude=0.00024: ser=1.01156 ±0.014510233457804875, br=-6.50391\n",
+ "signal_amplitude=0.00036: ser=1.00687 ±0.014875787794264881, br=-3.86719\n",
+ "signal_amplitude=0.00032: ser=1.01156 ±0.01837117307087384, br=-6.50391\n",
+ "signal_amplitude=0.00022: ser=1.01000 ±0.013535254892317322, br=-5.62500\n",
+ "signal_amplitude=0.00057: ser=1.00281 ±0.012476540486048206, br=-1.58203\n",
+ "signal_amplitude=0.00039: ser=1.00812 ±0.014402148277253642, br=-4.57031\n",
+ "signal_amplitude=0.00047: ser=1.00438 ±0.012899854650343935, br=-2.46094\n",
+ "signal_amplitude=0.00027: ser=1.00937 ±0.014657549249448218, br=-5.27344\n",
+ "signal_amplitude=0.00063: ser=0.99938 ±0.019253652250936705, br=0.35156\n",
+ "signal_amplitude=0.00077: ser=0.99156 ±0.03231920868461974, br=4.74609\n",
+ "signal_amplitude=0.00093: ser=0.95156 ±0.06625442202223185, br=27.24609\n",
+ "signal_amplitude=0.00112: ser=0.76000 ±0.2099632594348354, br=135.00000\n",
+ "signal_amplitude=0.00136: ser=0.51375 ±0.30673813139223494, br=273.51562\n",
+ "signal_amplitude=0.00165: ser=0.39844 ±0.38814210912370745, br=338.37891\n",
+ "signal_amplitude=0.00070: ser=0.99281 ±0.023688242072809035, br=4.04297\n",
+ "signal_amplitude=0.00084: ser=0.96375 ±0.050769469787461836, br=20.39062\n",
+ "signal_amplitude=0.00102: ser=0.91063 ±0.10310321739645179, br=50.27344\n",
+ "signal_amplitude=0.00124: ser=0.72500 ±0.23567348639059932, br=154.68750\n",
+ "signal_amplitude=0.00150: ser=0.40969 ±0.3064419041596629, br=332.05078\n",
+ "signal_amplitude=0.00182: ser=0.32531 ±0.38085840544748384, br=379.51172\n",
+ "signal_amplitude=0.00200: ser=0.29000 ±0.3885339029608613, br=399.37500\n",
+ "nbits=6\n",
+ "signal_amplitude=0.00052: ser=1.00375 ±0.027432445434193427, br=-1.26562\n",
+ "signal_amplitude=0.00029: ser=1.01531 ±0.013528038013695853, br=-5.16797\n",
+ "signal_amplitude=0.00020: ser=1.02000 ±0.01698459780212649, br=-6.75000\n",
+ "signal_amplitude=0.00024: ser=1.01844 ±0.0197494066366562, br=-6.22266\n",
+ "signal_amplitude=0.00043: ser=1.01000 ±0.013535254892317322, br=-3.37500\n",
+ "signal_amplitude=0.00036: ser=1.01500 ±0.01860884366369926, br=-5.06250\n",
+ "signal_amplitude=0.00032: ser=1.00906 ±0.01443601182806387, br=-3.05859\n",
+ "signal_amplitude=0.00022: ser=1.01656 ±0.015200483133769137, br=-5.58984\n",
+ "signal_amplitude=0.00057: ser=0.98281 ±0.04926213365760764, br=5.80078\n",
+ "signal_amplitude=0.00027: ser=1.02000 ±0.015946688527716343, br=-6.75000\n",
+ "signal_amplitude=0.00047: ser=1.00687 ±0.02815276407388802, br=-2.32031\n",
+ "signal_amplitude=0.00039: ser=1.00906 ±0.016189792308735775, br=-3.05859\n",
+ "signal_amplitude=0.00077: ser=0.76906 ±0.23454244018940368, br=77.94141\n",
+ "signal_amplitude=0.00063: ser=0.94031 ±0.08557822627572974, br=20.14453\n",
+ "signal_amplitude=0.00112: ser=0.29750 ±0.347296478171029, br=237.09375\n",
+ "signal_amplitude=0.00093: ser=0.50125 ±0.3293776683952632, br=168.32812\n",
+ "signal_amplitude=0.00136: ser=0.37250 ±0.42536588111001566, br=211.78125\n",
+ "signal_amplitude=0.00165: ser=0.51000 ±0.46215950303980546, br=165.37500\n",
+ "signal_amplitude=0.00070: ser=0.90063 ±0.1848975645458858, br=33.53906\n",
+ "signal_amplitude=0.00084: ser=0.64687 ±0.26652421325275494, br=119.17969\n",
+ "signal_amplitude=0.00124: ser=0.38500 ±0.39889079606767064, br=207.56250\n",
+ "signal_amplitude=0.00102: ser=0.40875 ±0.3467111099315971, br=199.54688\n",
+ "signal_amplitude=0.00150: ser=0.40375 ±0.4435118198819508, br=201.23438\n",
+ "signal_amplitude=0.00182: ser=0.58531 ±0.46179168734397985, br=139.95703\n",
+ "signal_amplitude=0.00200: ser=0.61594 ±0.4584529436730666, br=129.62109\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "<matplotlib.legend.Legend at 0x7fe2bdd64430>"
+ ]
+ },
+ "execution_count": 84,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "sample_duration=128\n",
+ "sample_reps = 25\n",
+ "sweep_points = 25\n",
+ "\n",
+ "default_params = dict(\n",
+ " nbits=6,\n",
+ " signal_amplitude=2.0e-3,\n",
+ " decimation=10,\n",
+ " threshold_factor=4.0,\n",
+ " power_avg_width=2.5,\n",
+ " max_lookahead=6.5)\n",
+ "\n",
+ "fig, ax = plt.subplots(figsize=(12, 9))\n",
+ "\n",
+ "for nbits in [5, 6]: # FIXME make sim stable for higher bit counts\n",
+ " print(f'nbits={nbits}')\n",
+ " \n",
+ " def calculate_ser(v):\n",
+ " params = dict(default_params)\n",
+ " params['signal_amplitude'] = v\n",
+ " params['nbits'] = nbits\n",
+ " sers, brs = [], []\n",
+ " for i in range(sample_reps):\n",
+ " ser, br = run_ser_test(**params, sample_duration=sample_duration, print=noprint, seed=np.random.randint(0xffffffff))\n",
+ " sers.append(ser)\n",
+ " brs.append(br)\n",
+ " sers, brs = np.array(sers), np.array(brs)\n",
+ " ser = np.mean(sers)\n",
+ " print(f'signal_amplitude={v:<.5f}: ser={ser:<.5f} ±{np.std(sers):<.5f}, br={np.mean(brs):<.5f}')\n",
+ " return ser\n",
+ " \n",
+ " vs = 0.2e-3 * 10 ** np.linspace(0, 1.0, sweep_points)\n",
+ " with Pool(6) as p:\n",
+ " data = p.map(calculate_ser, vs)\n",
+ " \n",
+ " ax.plot(vs, data, label=f'{nbits} bit')\n",
+ "ax.grid()\n",
+ "ax.set_xlabel('Amplitude in mHz')\n",
+ "ax.set_ylabel('Symbol error rate')\n",
+ "ax.legend()"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "labenv",
+ "language": "python",
+ "name": "labenv"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.8.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/lab-windows/dsss_experiments.ipynb b/lab-windows/dsss_experiments.ipynb
index be41b59..f2bbc0b 100644
--- a/lab-windows/dsss_experiments.ipynb
+++ b/lab-windows/dsss_experiments.ipynb
@@ -30,6 +30,15 @@
},
{
"cell_type": "code",
+ "execution_count": 105,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "sampling_rate = 10 # sp/s"
+ ]
+ },
+ {
+ "cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
@@ -410,29 +419,29 @@
},
{
"cell_type": "code",
- "execution_count": 96,
+ "execution_count": 145,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "(31,) (31,)\n",
- "(31,) (31,)\n"
+ "(63,) (63,)\n",
+ "(63,) (63,)\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
- "<ipython-input-96-b3aae757ccad>:33: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`).\n",
+ "<ipython-input-145-babcf8a4e867>:33: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`).\n",
" fig, ((ax1, ax3), (ax2, ax4)) = plt.subplots(2, 2, figsize=(16, 9))\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "246c19a3c1424e7eb0b675ff060ea5b3",
+ "model_id": "10aa67d294304f2ba26c8e6d5555d5e6",
"version_major": 2,
"version_minor": 0
},
@@ -446,10 +455,10 @@
{
"data": {
"text/plain": [
- "(0.002, 0.013899708)"
+ "(0.002, 0.014074279)"
]
},
- "execution_count": 96,
+ "execution_count": 145,
"metadata": {},
"output_type": "execute_result"
}
@@ -457,15 +466,15 @@
"source": [
"decimation = 10\n",
"signal_amplitude = 2.0e-3\n",
- "nbits = 5\n",
+ "nbits = 6\n",
"\n",
"#test_data = np.random.randint(0, 2, 100)\n",
"#test_data = np.array([0, 1, 0, 0, 1, 1, 1, 0])\n",
- "#test_data = np.random.RandomState(seed=0).randint(0, 2 * (2**nbits), 64)\n",
+ "test_data = np.random.RandomState(seed=0).randint(0, 2 * (2**nbits), 64)\n",
"#test_data = np.random.RandomState(seed=0).randint(0, 8, 64)\n",
"#test_data = np.array(list(range(8)) * 8)\n",
"#test_data = np.array([0, 1] * 32)\n",
- "test_data = np.array(list(range(64)))\n",
+ "#test_data = np.array(list(range(64)))\n",
"\n",
"foo = np.repeat(modulate(test_data, nbits) * 2.0 - 1, decimation) * signal_amplitude\n",
"noise = np.resize(mains_noise, len(foo))\n",
@@ -493,12 +502,20 @@
"ax1.plot(foo + noise)\n",
"ax1.plot(foo)\n",
"ax1.set_title('raw')\n",
+ "ax1.grid(axis='y')\n",
"\n",
"ax2.plot(filtered)\n",
"ax2.plot(foo)\n",
"ax2.set_title('filtered')\n",
+ "ax2.grid(axis='y')\n",
+ "\n",
+ "for i in range(0, len(foo) + 1, decimation*(2**nbits - 1)):\n",
+ " ax1.axvline(i, color='gray', alpha=0.5, lw=1)\n",
+ " ax2.axvline(i, color='gray', alpha=0.5, lw=1)\n",
"\n",
- "ax3.plot(cor1.T)\n",
+ "for i, (color, trace) in enumerate(zip(plt.cm.winter(np.linspace(0, 1, cor1.shape[0])), cor1.T)):\n",
+ " if i%3 == 0:\n",
+ " ax3.plot(trace + 0.5 * i, alpha=1.0, color=color)\n",
"ax3.set_title('corr raw')\n",
"ax3.grid()\n",
"\n",
@@ -678,21 +695,21 @@
},
{
"cell_type": "code",
- "execution_count": 97,
+ "execution_count": 146,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
- "<ipython-input-97-2d2c2f814215>:11: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`).\n",
+ "<ipython-input-146-badd40342f73>:11: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`).\n",
" fig, (ax1, ax3) = plt.subplots(2, figsize=(12, 5))\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "b083c661b5b441d6b7fc45201faa0576",
+ "model_id": "509bf67d93b74741b48bca58529c4b9d",
"version_major": 2,
"version_minor": 0
},
@@ -707,26 +724,24 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "cor_an (33, 20149)\n",
- "cwt_res (33, 20149)\n",
- "th (33, 20149)\n",
- "[((33,), (33,)), ((33,), (33,)), ((33,), (33,)), ((33,), (33,)), ((33,), (33,))]\n",
- "peaks: 180\n",
- "avg_peak 1.058897833824206\n",
- "skipped 3 symbols at 5889.0\n",
- "skipped 2 symbols at 8369.0\n",
- "skipped 2 symbols at 14568.5\n",
- "skipped 2 symbols at 16739.0\n",
+ "cor_an (65, 40949)\n",
+ "cwt_res (65, 40949)\n",
+ "th (65, 40949)\n",
+ "[((65,), (65,)), ((65,), (65,)), ((65,), (65,)), ((65,), (65,)), ((65,), (65,))]\n",
+ "peaks: 982\n",
+ "avg_peak 1.6673786030736735\n",
+ "skipped 2 symbols at 30238.5\n",
"decoding [ref|dec]:\n",
- " -1| -1 ✔ 1| 1 ✔ 2| 2 ✔ 3| 3 ✔ 4| 4 ✔ 5| 5 ✔ 6| 6 ✔ 7| 7 ✔ \n",
- " 8| 8 ✔ 9| 9 ✔ 10| 10 ✔ 11| 11 ✔ 12| 12 ✔ 13| 13 ✔ 14| 14 ✔ 15| 15 ✔ \n",
- " 16| -1 ✘ 17| -1 ✘ 18| 18 ✔ 19| 19 ✔ 20| 20 ✔ 21| 21 ✔ 22| 22 ✔ 23| 23 ✔ \n",
- " 24| 24 ✔ 25| -1 ✘ 26| 26 ✔ 27| 27 ✔ 28| 28 ✔ 29| 29 ✔ 30| 30 ✔ 31| 31 ✔ \n",
- " 32| 32 ✔ 33| 33 ✔ 34| 34 ✔ 35| 35 ✔ 36| 36 ✔ 37| 37 ✔ 38| 38 ✔ 39| 39 ✔ \n",
- " 40| 40 ✔ 41| 41 ✔ 42| 42 ✔ 43| 43 ✔ 44| 44 ✔ 45| -1 ✘ 46| 46 ✔ 47| 47 ✔ \n",
- " 48| 48 ✔ 49| 49 ✔ 50| 50 ✔ 51| 51 ✔ 52| -1 ✘ 53| 53 ✔ 54| 54 ✔ 55| 55 ✔ \n",
- " 56| 56 ✔ 57| 57 ✔ 58| 58 ✔ 59| 59 ✔ 60| 60 ✔ 61| 61 ✔ 62| 62 ✔ 63| 56 ✘ \n",
- "Symbol error rate r=0.09375\n"
+ " 44| 44 ✔ 47| 47 ✔ 117|117 ✔ 64| 64 ✔ 67| 67 ✔ 123|123 ✔ 67| 67 ✔ 103|103 ✔ \n",
+ " 9| 9 ✔ 83| 83 ✔ 21| 21 ✔ 114|114 ✔ 36| 36 ✔ 87| 87 ✔ 70| 70 ✔ 88| 88 ✔ \n",
+ " 88| 88 ✔ 12| 12 ✔ 58| 58 ✔ 65| 65 ✔ 102|102 ✔ 39| 39 ✔ 87| 87 ✔ 46| 46 ✔ \n",
+ " 88| 88 ✔ 81| 81 ✔ 37| 37 ✔ 25| 25 ✔ 77| 77 ✔ 72| 72 ✔ 9| 9 ✔ 20| 20 ✔ \n",
+ "115|115 ✔ 80| 80 ✔ 115|115 ✔ 69| 69 ✔ 126|126 ✔ 79| 79 ✔ 47| 47 ✔ 64| 64 ✔ \n",
+ " 82| 82 ✔ 99| 99 ✔ 88| 88 ✔ 49| 49 ✔ 115|115 ✔ 29| 29 ✔ 19| -1 19| 19 ✔ \n",
+ " 14| 14 ✔ 39| 39 ✔ 32| 32 ✔ 65| 64 ✘ 9| 9 ✔ 57| 57 ✔ 127|127 ✔ 32| 32 ✔ \n",
+ " 31| 31 ✔ 74| 74 ✔ 116|116 ✔ 23| 23 ✔ 35| 35 ✔ 126|126 ✔ 75| 75 ✔ 114| 26 ✘ \n",
+ "Symbol error rate e=0.046875\n",
+ "maximum bitrate r=321.6796875 b/h\n"
]
}
],
@@ -869,12 +884,13 @@
"print('decoding [ref|dec]:')\n",
"failures = 0\n",
"for i, (ref, found) in enumerate(itertools.zip_longest(test_data, decoded)):\n",
- " print(f'{ref or -1:>3d}|{found or -1:>3d} {\"✔\" if ref==found else \"✘\"}', end=' ')\n",
+ " print(f'{ref or -1:>3d}|{found or -1:>3d} {\"✔\" if ref==found else \"✘\" if found else \" \"}', end=' ')\n",
" if ref != found:\n",
" failures += 1\n",
" if i%8 == 7:\n",
" print()\n",
- "print(f'Symbol error rate r={failures/len(test_data)}')\n",
+ "print(f'Symbol error rate e={failures/len(test_data)}')\n",
+ "print(f'maximum bitrate r={sampling_rate / decimation / (2**nbits) * nbits * (1 - failures/len(test_data)) * 3600} b/h')\n",
"#ax3.plot(th)"
]
},