ezmsg.sigproc#

Modules

activation

Activation functions (ReLU, sigmoid, logit, etc.) applied element-wise.

adaptive_lattice_notch

Adaptive lattice notch filter for narrow-band interference removal.

affinetransform

Affine transformations via matrix multiplication: y = Ax or y = Ax + B.

aggregate

Aggregation operations over arrays.

asarray

Convert AxisArray data to a target array backend.

bandpower

Spectral band power estimation via windowed FFT and aggregation.

base

Backwards-compatible re-exports from ezmsg.baseproc.

butterworthfilter

Butterworth IIR filter design and application.

butterworthzerophase

Streaming zero-phase Butterworth filter implemented as a two-stage composite processor.

cheby

Chebyshev Type I IIR filter design and application.

combfilter

Comb filter for suppressing a fundamental frequency and its harmonics.

coordinatespaces

Coordinate space transformations for streaming data.

decimate

Decimation (downsample with anti-alias filtering).

denormalize

Denormalize signals by applying random gains and offsets.

detrend

Remove linear or constant trends from data along an axis.

diff

Compute differences along an axis.

downsample

Integer downsampling by selecting every Nth sample along an axis.

ewma

Exponentially weighted moving average (EWMA) utilities and parameter conversion.

ewmfilter

Exponentially weighted moving average filter for streaming normalization.

extract_axis

Extract a named axis from an AxisArray as a standalone array.

fbcca

Filter-bank canonical correlation analysis (FBCCA) for frequency detection.

filter

Core IIR/FIR filtering infrastructure with BA and SOS coefficient support.

filterbank

Parallel filterbank processing with convolution, FFT, or automatic mode selection.

filterbankdesign

Kaiser-window filterbank design with configurable bands and processing mode.

fir_hilbert

FIR Hilbert transform filter for analytic signal and envelope extraction.

fir_pmc

Parks-McClellan (Remez) optimal equiripple FIR filter design and application.

firfilter

FIR filter design and application using scipy.signal.firwin.

gaussiansmoothing

Gaussian kernel smoothing filter.

kaiser

Kaiser window FIR filter design and application.

linear

Apply a linear transformation: output = scale * input + offset.

materialize

Materialize (evaluate) lazy array data.

math

merge

Time-aligned merge of two AxisArray streams along a non-time axis.

messages

Deprecated message types; use ezmsg.util.messages.axisarray.AxisArray instead.

quantize

Quantize signal data to a fixed number of bits.

resample

Resample signals to a target rate using interpolation.

rollingscaler

Rolling z-score normalization using a sliding window of recent samples.

sampler

Triggered windowed sampling from a buffered signal stream.

scaler

Adaptive standard scaling using exponentially weighted moving statistics.

signalinjector

Inject synthetic sinusoidal signals into a data stream.

singlebandpow

Time-domain single-band power estimation.

slicer

Select a subset of data along a named axis using slice notation.

spectral

Backwards-compatible re-exports from spectrum.

spectrogram

Time-frequency spectrogram via windowed short-time Fourier transform.

spectrum

FFT-based power spectrum estimation with configurable window functions.

transpose

Transpose or permute array dimensions.

util

wavelets

Continuous wavelet transform (CWT) for streaming time-frequency analysis.

window

Sliding and tumbling window segmentation of streaming data.