4. Tools

The psymo.tools package provides some extra tools that do not involve usage of SAC files.

4.2. Code Examples

4.2.1. Noise

In this example two realistic random noise signals are generated and their PSDs are calculated.

_images/peterson.png 
#!/usr/bin/env python

# Example script for pysmo.tools.noise
import pysmo.tools.noise as noise
import numpy as np
import matplotlib.pyplot as plt
from scipy import signal

npts = 10000
delta = 0.5
fs = 1 / delta
nperseg = npts/4
nfft = npts/2
time = np.linspace(0, npts*delta, npts)

# Generate random noise
high_noise = noise.genNoiseNHNM(delta, npts)
low_noise = noise.genNoiseNLNM(delta, npts)

# Calculuate power spectral density
f_high, Pxx_dens_high = signal.welch(high_noise, fs, nperseg=nperseg,
                                     nfft=nfft, scaling='density')
f_low, Pxx_dens_low = signal.welch(low_noise, fs, nperseg=nperseg,
                                   nfft=nfft, scaling='density')

fig = plt.figure(figsize=(13, 6))
# Plot random high and low noise
plt.subplot(221)
plt.plot(time, high_noise, 'c')
plt.ylabel('Ground Accelaration')

plt.subplot(223)
plt.plot(time, low_noise, 'm')
plt.ylabel('Ground Accelaration')
plt.xlabel('Time [s]')

# Plot PSD of noise
plt.subplot(122)
plt.plot(1/f_high, 10*np.log10(Pxx_dens_high), 'c',
         label='generated high noise')
plt.plot(1/f_low, 10*np.log10(Pxx_dens_low), 'm', label='generated low noise')
plt.plot(noise.NLNM['T'], noise.NLNM['dB'], 'k', linewidth=2)
plt.plot(noise.NHNM['T'], noise.NHNM['dB'], 'k', linewidth=2)
plt.gca().set_xscale("log")
plt.xlabel('Period [s]')
plt.ylabel('Power Spectral Density (dB/Hz)')
plt.legend()
plt.savefig('peterson.png')
plt.show()