Title :
Improving the Bound on the RIP Constant in Generalized Orthogonal Matching Pursuit
Author :
Satpathi, Siddhartha ; Das, Rajib Lochan ; Chakraborty, Manali
Author_Institution :
Dept. of Electron. & Electr. Commun. Eng, Indian Inst. of Technol., Kharagpur, Kharagpur, India
Abstract :
The generalized Orthogonal Matching Pursuit (gOMP) is a recently proposed compressive sensing greedy recovery algorithm which generalizes the OMP algorithm by selecting N( ≥ 1) atoms in each iteration. In this letter, we demonstrate that the gOMP can successfully reconstruct a K-sparse signal from a compressed measurement y=Φx by a maximum of K iterations if the sensing matrix Φ satisfies the Restricted Isometry Property (RIP) of order NK, with the RIP constant δNK satisfying δNK <; √N/√K+2√N. The proposed bound is an improvement over the existing bound on δNK. We also show that by increasing the RIP order just by one (i.e., NK+1 from NK), it is possible to refine the bound further to δNK+1 <; √N/√K+√N, which is consistent (for N=1) with the near optimal bound on δK+1 in OMP.
Keywords :
compressed sensing; greedy algorithms; iterative methods; signal reconstruction; K-sparse signal reconstruction; OMP algorithm; RIP constant; compressed measurement; compressive sensing greedy recovery algorithm; gOMP; generalized orthogonal matching pursuit; iteration method; restricted isometry property; sensing matrix; Compressed sensing; Convergence; Indexes; Matching pursuit algorithms; Sensors; Upper bound; Vectors; Compressive sensing; orthogonal matching pursuit; restricted isometry property; sensing matrix;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2013.2279977