magnes-signal-generation-utility 0.1.0

Utility to generate signals for algorithm development and verification.

Magnes Signal Generation Utility

Utility to generate signals for algorithm development and verification.

Quickstart

Install the package using pip:

pip install magnes-signal-generation-utility

Generate Synthetic Signal with Artifacts

from msgu.mixture import generate_synthetic_wave_with_cardiac_artifacts

# Generate synthetic signal with cardiac artifacts
data = generate_synthetic_wave_with_cardiac_artifacts(
    fs=250.0,
    freqs=[10.0, 30.0],       # Peak frequencies
    widths=[2.0, 5.0],        # Frequency widths
    gamma=3.0,                # Disturbance gain
    sigma=0.2,                # Noise level
    nbeats=10
)

# Access the components
t = data.t                   # Time array
signal = data.x              # Ground truth signal
disturbance = data.d         # Cardiac artifacts
noise = data.n               # Pink noise
measurement = data.z         # Composite measurement

Generate White and Pink Noise

from msgu.noise import generate_white_noise, generate_pink_noise, PINK_NOISE_ALGO

# Generate white noise
white_noise = generate_white_noise(n=1000)

# Generate pink noise using FFT algorithm
pink_noise = generate_pink_noise(n=1000, algo=PINK_NOISE_ALGO.FFT)

Basic Example: Generate ECG Signal

import numpy as np
import matplotlib.pyplot as plt
from msgu.cardio import generate_ecg

# Generate a 5-beat ECG signal at 250 Hz
fs = 250.0
t, ecg_signal = generate_ecg(fs, nbeats=5, sigma=0.01)

# Plot the signal
plt.plot(t, ecg_signal)
plt.xlabel('Time [s]')
plt.ylabel('Signal [mV]')
plt.title('Pseudo-ECG Signal')
plt.show()

Submodules

Cardio

Pseudo-ECG signals generation utility. Defines a (heart) Beat object, which can be used as a generator of a pseudo-random, single-beat signal. A utility function allows to concatenate n beats into a singe time series.

Gauss

Helper definitions of the Gaussian bell and its derivatives.

Mixture

Signal synthesizers: combining various signal generators into composite signals.

Pseudo-Random Waves Synth prsynth

Superposing blurred-peak, random-phase sine-waves, with additive disturbances (ECG artifacts) and pink noise.

Noise

Noise generation utility.

Periodic

(Quasi) periodic signals generation utility.

Tidal - Pseudo-random sine waves

Includes the implementation of blurred-peak, randomly-phased sine-waves generation

Trajectory

Trajectory generation utility.

S-Curves

Helper utility for the generation of normed, polynomial (order 3 and 5) S-curves.

Usage Examples

Scripts illustrating the intended usage of the msgu package are available in scripts/. To run the script foo.py call

uv run scripts/foo.py

from project root.

Testing

Unittests can be run with uv and pytest from root as

uv run -m pytest [OPTIONS]