DSP Module (stx.dsp)

Digital signal processing tools for filtering, spectral analysis, and time-frequency decomposition.

Quick Reference

import scitex as stx
import numpy as np

# Generate test signal
fs = 1000  # Hz
t = np.arange(0, 1, 1/fs)
signal = np.sin(2 * np.pi * 10 * t) + 0.5 * np.sin(2 * np.pi * 50 * t)

# Filtering
filtered = stx.dsp.filt.bandpass(signal, fs=fs, low=5, high=30)
low_pass = stx.dsp.filt.lowpass(signal, fs=fs, cutoff=20)
high_pass = stx.dsp.filt.highpass(signal, fs=fs, cutoff=5)

# Power spectral density
freqs, psd = stx.dsp.psd(signal, fs=fs)

# Band powers
powers = stx.dsp.band_powers(signal, fs=fs, bands={
    "delta": (1, 4),
    "theta": (4, 8),
    "alpha": (8, 13),
    "beta": (13, 30),
})

# Hilbert transform (analytic signal)
analytic = stx.dsp.hilbert(signal)
amplitude = np.abs(analytic)
phase = np.angle(analytic)

# Wavelet decomposition
coeffs = stx.dsp.wavelet(signal, fs=fs, freqs=np.arange(1, 50))

Available Functions

Filtering (stx.dsp.filt)

• bandpass(signal, fs, low, high)

• lowpass(signal, fs, cutoff)

• highpass(signal, fs, cutoff)

• notch(signal, fs, freq)

Spectral Analysis

• psd(signal, fs) – Power spectral density

• band_powers(signal, fs, bands) – Power in frequency bands

• wavelet(signal, fs, freqs) – Continuous wavelet transform

Time-Frequency

• hilbert(signal) – Analytic signal via Hilbert transform

• pac(signal, fs) – Phase-amplitude coupling

Utilities

• demo_sig(fs, duration) – Generate demo signals for testing

• crop(signal, start, end, fs) – Crop signal to time range

• detect_ripples(signal, fs) – Detect sharp-wave ripples

API Reference