Skip to content

Commit

Permalink
Added python script used in PlutoSDR spectral analysis page
Browse files Browse the repository at this point in the history
  • Loading branch information
@SrikanthPagadarai
SrikanthPagadarai committed Dec 5, 2024
1 parent 24d73de commit 9fe13a0
Showing 1 changed file with 206 additions and 0 deletions.
206 changes: 206 additions & 0 deletions bindings/python/examples/gn_doc_spectral_analysis_plutosdr.py
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,206 @@
import adi
import matplotlib.pyplot as pl
import genalyzer as gn
import numpy as np
from scipy import signal
from tabulate import tabulate
import pprint

# Create Pluto object and Configure properties
sdr = adi.Pluto(uri="ip:192.168.2.1")
sdr.rx_rf_bandwidth = 4000000
sdr.tx_lo = 1000000000
sdr.rx_lo = sdr.tx_lo
sdr.tx_cyclic_buffer = True
sdr.tx_hardwaregain_chan0 = -30 # Reduce transmit gain when PlutoSDR ports are connected by a loopback cable
sdr.gain_control_mode_chan0 = "slow_attack"
sdr.rx_buffer_size = 2**15
sdr.sample_rate = 4000000

# Sinusoidal tone settings
freq = 350000 # tone frequency
ampl_dbfs = 0.0 # amplitude of the tone in dBFS
fsr = 2.0 # full-scale range of I/Q components of the complex tone
ampl = (fsr / 2) * 10 ** (ampl_dbfs / 20) # amplitude of the tone in linear scale

# Trasmit complex sinusoidal tone using the DDS
sdr.dds_single_tone(freq, ampl)

# Get data from radio (allow a few buffer pulls for AGC to settle)
for n in range(10):
x = sdr.rx()

# FFT configuration
navg = 1 # number of FFT averages
nfft = int(sdr.rx_buffer_size/navg) # FFT-order
window = gn.Window.BLACKMAN_HARRIS # window function to apply
axis_type = gn.FreqAxisType.DC_CENTER # axis type
axis_fmt = gn.FreqAxisFormat.FREQ # axis-format
qres = 12 # PlutoSDR data resolution in bits
code_fmt = gn.CodeFormat.TWOS_COMPLEMENT # integer data format

# Compute FFT
x_re = np.array(x.real).astype(np.int16)
x_im = np.array(x.imag).astype(np.int16)
fft_cplx = gn.fft(x_re, x_im, qres, navg, nfft, window, code_fmt)
freq_axis = gn.freq_axis(nfft, axis_type, sdr.sample_rate, axis_fmt)
fft_db = gn.db(fft_cplx)
if gn.FreqAxisType.DC_CENTER == axis_type:
fft_db = gn.fftshift(fft_db)

# Fourier analysis configuration
test_label = "fa"
gn.fa_create(test_label)

num_harmonics = 3 # number of harmonics to analyze
ssb_fund = 6 # number of single-side bins for the signal component
ssb_rest = 4 # default number of single-side bins for non-signal components
ssb_dc = 2 # number of single-side bins for the DC-component
signal_component_label = 'A'
gn.fa_ssb(test_label, gn.FaSsb.DEFAULT, ssb_rest)
gn.fa_max_tone(test_label, signal_component_label, gn.FaCompTag.SIGNAL, ssb_fund)
gn.fa_fsample(test_label, sdr.sample_rate)
gn.fa_hd(test_label, num_harmonics)
gn.fa_ssb(test_label, gn.FaSsb.DC, ssb_dc)

# Fourier analysis execution
results = gn.fft_analysis(test_label, fft_cplx, nfft, axis_type)

annots = gn.fa_annotations(results, axis_type, axis_fmt)
print('annots["labels"]: ')
labels_head = ('frequency (Hz)', 'magnitude (dBFs)', 'component label')
labels_table = tabulate(annots["labels"], headers=labels_head, tablefmt="grid")
print(labels_table, "\n")

print('annots["tone_boxes"]: ')
c1 = [x[0] for x in annots["tone_boxes"]]
c2 = [x[2] for x in annots["tone_boxes"]]
tone_boxes_head = ('box left boundary (Hz)', 'width (Hz)')
tone_boxes_table = tabulate(map(list, zip(*(c1, c2))), headers=tone_boxes_head, tablefmt="grid")
print(tone_boxes_table, "\n")

print('+----------------+')
print("results dictionary")
print('+----------------+')
pprint.pprint(results)

# plot
toneDC_Hz = annots["labels"][0][0]
toneDC_bin = toneDC_Hz/(sdr.sample_rate/nfft)
toneDC_mag = fft_db[int(toneDC_bin+0.5*nfft)]
toneA_Hz = annots["labels"][1][0]
toneA_bin = toneA_Hz/(sdr.sample_rate/nfft)
toneA_mag = fft_db[int(toneA_bin+0.5*nfft)]
toneA_im_Hz = annots["labels"][2][0]
toneA_im_bin = toneA_im_Hz/(sdr.sample_rate/nfft)
toneA_im_mag = fft_db[int(toneA_im_bin+0.5*nfft)]
tone2A_Hz = annots["labels"][3][0]
tone2A_bin = tone2A_Hz/(sdr.sample_rate/nfft)
tone2A_mag = fft_db[int(tone2A_bin+0.5*nfft)]
tone2A_im_Hz = annots["labels"][4][0]
tone2A_im_bin = tone2A_im_Hz/(sdr.sample_rate/nfft)
tone2A_im_mag = fft_db[int(tone2A_im_bin+0.5*nfft)]
tone3A_im_Hz = annots["labels"][5][0]
tone3A_im_bin = tone3A_im_Hz/(sdr.sample_rate/nfft)
tone3A_im_mag = fft_db[int(tone3A_im_bin+0.5*nfft)]
toneWO_Hz = annots["labels"][6][0]
toneWO_bin = toneWO_Hz/(sdr.sample_rate/nfft)
toneWO_mag = fft_db[int(toneWO_bin+0.5*nfft)]

sfdr = results["sfdr"]
nsd = results["nsd"]
abn = results["abn"]
snr = results["snr"]
fsnr = results["fsnr"]
sinad = results["sinad"]
scale_MHz = 1e-6
fig, ax = pl.subplots()
fig.clf()
pl.plot(freq_axis*scale_MHz, fft_db)
pl.grid(True)
pl.xlabel('frequency (MHz)')
pl.ylabel('magnitude (dBFs)')
pl.xlim(freq_axis[0]*scale_MHz, freq_axis[-1]*scale_MHz)
pl.ylim(-140.0, 20.0)
for x, y, label in annots["labels"]:
if label == 'dc':
pl.annotate(label+": ["+f"{toneDC_Hz:.2f}"+" ,"+f"{toneDC_mag:.2f}"+"]",
xy=(x*scale_MHz, toneDC_mag),
xytext=(x*scale_MHz, -10),
color = 'red',
horizontalalignment="center",
arrowprops=dict(arrowstyle='->',color='red',lw=1))
elif label == 'A':
pl.annotate(label+": ["+f"{toneA_Hz:.2f}"+" ,"+f"{toneA_mag:.2f}"+"]",
xy=(x*scale_MHz, toneA_mag),
xytext=(x*scale_MHz, 10),
color = 'black',
horizontalalignment="center",
arrowprops=dict(arrowstyle='->',color='black',lw=1))
elif label == '-A':
pl.annotate(label+": ["+f"{toneA_im_Hz:.2f}"+" ,"+f"{toneA_im_mag:.2f}"+"]",
xy=(x*scale_MHz, toneA_im_mag),
xytext=(x*scale_MHz, -20),
color = 'black',
horizontalalignment="center",
arrowprops=dict(arrowstyle='->',color='black',lw=1))
elif label == '2A':
pl.annotate(label+": ["+f"{tone2A_Hz:.2f}"+" ,"+f"{tone2A_mag:.2f}"+"]",
xy=(x*scale_MHz, tone2A_mag),
xytext=(x*scale_MHz, -40),
color = 'green',
horizontalalignment="center",
arrowprops=dict(arrowstyle='->',color='green',lw=1))
elif label == '-2A':
pl.annotate(label+": ["+f"{tone2A_im_Hz:.2f}"+" ,"+f"{tone2A_im_mag:.2f}"+"]",
xy=(x*scale_MHz, tone2A_im_mag),
xytext=(x*scale_MHz, -40),
color = 'green',
horizontalalignment="center",
arrowprops=dict(arrowstyle='->',color='green',lw=1))
elif label == '-3A':
pl.annotate(label+": ["+f"{tone3A_im_Hz:.2f}"+" ,"+f"{tone3A_im_mag:.2f}"+"]",
xy=(x*scale_MHz, tone3A_im_mag),
xytext=(x*scale_MHz, -60),
color = 'magenta',
horizontalalignment="center",
arrowprops=dict(arrowstyle='->',color='magenta',lw=1))
elif label == 'wo':
pl.annotate(label+": ["+f"{toneWO_Hz:.2f}"+" ,"+f"{toneWO_mag:.2f}"+"]",
xy=(x*scale_MHz, toneWO_mag),
xytext=(x*scale_MHz, -80),
color = 'magenta',
horizontalalignment="center",
arrowprops=dict(arrowstyle='->',color='magenta',lw=1))
else:
pl.annotate(label, xy=(x*scale_MHz, y), ha="center", va="bottom")
pl.axhline(y = toneA_mag, color = 'k', linestyle = '-')
pl.axhline(y = toneWO_mag, color = 'k', linestyle = '-')
pl.annotate('',
xy=(1.25,toneWO_mag),
xytext=(1.25,toneA_mag),
arrowprops=dict(arrowstyle='<->',color='black',lw=1))
pl.annotate('SFDR'+": "+f"{sfdr:.2f}"+' dB',
xy=(1.25, -40),
xytext=(1.25, -40),
verticalalignment="center",
rotation=270)
pl.axhline(y = abn, color = 'r', linestyle = '-')
pl.annotate('ABN'+": "+f"{abn:.2f}"+' dB',
xy=(0.5, -100),
xytext=(0.5, -100),
color = 'red',
ha="center")
pl.axhline(y = nsd, color = 'r', linestyle = '-')
pl.annotate('NSD'+": "+f"{nsd:.2f}"+' dB',
xy=(0.5, -120),
xytext=(0.5, -120),
color = 'red',
ha="center")
textstr = '\n'.join((
'SNR'+": "+f"{snr:.2f}"+' dB',
'FSNR'+": "+f"{fsnr:.2f}"+' dB',
'SINAD'+": "+f"{sinad:.2f}"+' dB'))
props = dict(boxstyle='round', facecolor='wheat', alpha=0)
ax.text(0.5, 0.5, textstr, fontsize=14, bbox=props)
pl.savefig('spectral_analysis_summary_pluto.png')

0 comments on commit 9fe13a0

Please sign in to comment.