A Note on Sparsification by Frames

TL;DR

This paper establishes a new generalized D-RIP sparsity bound constant for compressed sensing, enabling the reconstruction of signals with sparse D-representations under certain conditions.

Contribution

It introduces a novel D-RIP bound constant, specifically proving signals with k-sparse D-representations can be reconstructed if elta_{2k} < 2/3, extending previous bounds.

Findings

Reconstruction is possible if elta_{2k} < 2/3.

The approach can be extended to other D-RIP bounds like elta_{tk}.

Provides a generalized sparsity bound for compressed sensing.

Abstract

The purpose of this note is to establish a new generalized Dictionary-Restricted Isometry Property (D-RIP) sparsity bound constant for compressed sensing. For fulfilling D-RIP, the constant $\delta_k$ is used in the definition: $(1 -\delta_k)\|D v\|_2^2 \le \|\Phi D v\|_2^2 \le (1 + \delta_k)\|D v\|^2$. We prove that signals with $k$-sparse $D$-representation can be reconstructed if $\delta_{2k} < \frac{2}3$. The approach in this note can be extended to obtain other D-RIP bounds (i.e., $\delta_{tk}$).