spectr.py

import subprocess, sys, getopt
#import matplotlib.pyplot as plt
import numpy as np
from datetime import datetime
from matplotlib import cm
from matplotlib.ticker import EngFormatter
from matplotlib.widgets import Cursor

def read_dev(time):
	dt = datetime.now()
	outfile = "scan_"+ dt.strftime("%Y-%m-%d_%Hh%Mm%Ss")
	gain = 496 # /10dB
	freq_arg = '-f '+str(898e6)+':'+str(920e6)
	gain_arg = '-g '+str(gain)
	time_arg = '-e '+str(time)+'m'
	file_arg = '-m '+str(outfile)
	print ("executando fftw por " + str(time) +" minutos com ganho de "+str(gain/10)+" dB")
	subprocess.run(["rtl_power_fftw", freq_arg, "-b 8192", gain_arg,time_arg,"-o 70" ,file_arg])

def fileparse(filename): #analise dos metadados do scan 
	with open(filename, 'rb') as f:
		data = np.fromfile(f, dtype=np.float32)


	#ler o arquivo de metadados e criar o dicionário de propriedades
	metafile = filename.replace(".bin",".met")
	f= open(metafile, "r")

	propval = []
	propname = ["nbins","nscans","freqstart","freqend","freqstep","int_time","duration","timestart","timeend"]

	for line in f:
		field = line.split(" # ")
		try:
			propval.append(int(field[0]))
		except ValueError:
			try:
				propval.append(float(field[0]))
			except ValueError:
				propval.append(field[0])
	f.close()
	props = dict(zip(propname,propval))
	data.shape=(-1,props["nbins"])
	print(props)


	binstep = (props['freqend'] - props['freqstart'])/props['nbins']
	x = np.arange(props["freqstart"], props["freqend"],binstep)
	y = range(props['nscans'])
	xmesh,ymesh = np.meshgrid(x,y, sparse=True)
	return data
	

# def render_graph(xmesh,ymesh,z):