Robust Sparse Walsh-Hadamard Transform: The SPRIGHT Algorithm

Abstract

We consider the problem of stably computing the K-sparse N -point Walsh-Hadamard Transform (WHT) of a noisy input vector of length N , where K = O(N δ) scales sub-linearly in the signal dimension N for some δ ∈ (0, 1). The most efficient way known by far for computing the WHT of an arbitraryN -length signal is the Fast Walsh-Hadamard Transform (FWHT), which… (More)

Topics

8 Figures and Tables

Cite this paper

@inproceedings{Li2015RobustSW, title={Robust Sparse Walsh-Hadamard Transform: The SPRIGHT Algorithm}, author={Xiao Li and Joseph K. Bradley and Sameer Pawar and Kannan Ramchandran}, year={2015} }