sgwt_forward: Forward SGWT transform (single or batch)

In BioGSP: Biological Graph Signal Processing for Spatial Data Analysis

Forward SGWT transform (single or batch)

Description

Transform signal(s) to spectral domain and apply SGWT filters. Handles both single signals (vector) and multiple signals (matrix) efficiently. Stores original and filtered Fourier coefficients for analysis.

Usage

sgwt_forward(
  signal,
  eigenvectors,
  eigenvalues,
  scales,
  lmax = NULL,
  kernel_type = "heat"
)

Arguments

signal	Input signal vector OR matrix where each column is a signal (n_vertices x n_signals)
eigenvectors	Eigenvectors of the graph Laplacian
eigenvalues	Eigenvalues of the graph Laplacian
scales	Vector of scales for the wavelets
lmax	Maximum eigenvalue (optional)
kernel_type	Kernel family that defines both scaling and wavelet filters (default: "heat")

Value

List containing:

fourier_coefficients

List with original and filtered Fourier coefficients

filters

Filter bank used

Examples

# Create example data and compute graph
data <- data.frame(x = runif(50), y = runif(50), signal = rnorm(50))
SG <- initSGWT(data, signals = "signal", J = 3)
SG <- runSpecGraph(SG, k = 10)
eigenvectors <- SG$Graph$eigenvectors
eigenvalues <- SG$Graph$eigenvalues
scales <- SG$Parameters$scales

# Single signal
signal <- data$signal
result <- sgwt_forward(signal, eigenvectors, eigenvalues, scales)

# Multiple signals (batch processing)
signals_matrix <- cbind(data$signal, data$signal * 2, data$signal * 0.5)
result <- sgwt_forward(signals_matrix, eigenvectors, eigenvalues, scales)

BioGSP documentation built on Feb. 2, 2026, 5:06 p.m.