| name | ix-signal |
| description | Signal processing — FFT, filtering, wavelets, Kalman, spectral analysis |
| disable-model-invocation | true |
Signal Processing
Analyze and transform time-domain and frequency-domain signals.
When to Use
When the user has time series, audio, sensor data, or any signal that needs frequency analysis, filtering, or noise reduction.
Capabilities
- FFT/IFFT — Frequency decomposition, power spectrum
- Filtering — Low-pass, high-pass, band-pass FIR/IIR filters
- Wavelets — Haar wavelet transform, multi-resolution analysis
- Kalman Filter — State estimation for noisy dynamic systems
- Spectral Analysis — Power spectral density, spectrogram
- Windows — Hamming, Hanning, Blackman for spectral leakage reduction
- Convolution/Correlation — Cross-correlation, autocorrelation
- DCT — Discrete cosine transform (compression, feature extraction)
Programmatic Usage
use ix_signal::fft::{fft, ifft, power_spectrum};
use ix_signal::filter::{low_pass, high_pass};
use ix_signal::kalman::KalmanFilter;
use ix_signal::wavelet::haar_wavelet_transform;
use ix_signal::spectral::spectrogram;
use ix_signal::window::{hamming, hanning};
Tips
- Apply a window function before FFT to reduce spectral leakage
- Kalman filter requires a state model — help user define F, H, Q, R matrices
- Use power spectrum to identify dominant frequencies