Publication Type : Book Chapter
Publisher : Artificial Intelligence and Evolutionary Computations in Engineering Systems, Springer
Source : Artificial Intelligence and Evolutionary Computations in Engineering Systems, Springer, Volume 394, p.471–480 (2016)
Url : http://link.springer.com/chapter/10.1007/978-81-322-2656-7_43
Campus : Coimbatore
School : School of Engineering
Center : Electronics Communication and Instrumentation Forum (ECIF)
Department : Electronics and Communication
Year : 2015, 2016
Abstract : Compressed sensing (CS) is an emerging field which enables the undersampling of sparse signals rather than at the Nyquist rate. But the main computational challenge involved is in the reconstruction process as it is nonlinear in nature and the solution is obtained by solving a set of under determined linear equations. Greedy algorithms offer the solution to these kinds of problems with less computational complexity than the convex relaxations or linear programming methods. The approximate message passing algorithm offers accurate reconstruction of even the approximately sparse signals with reasonable computational intensity. In this paper, we have implemented a modified version of AMP algorithm and obtained a 50 % reduction in mean squared error and an improvement in signal-to-noise ratio
Cite this Research Publication : N. Sara Mathai and Gandhiraj R., “Performance Improvement of Compressed Sensing Reconstruction Using Modified-AMP Algorithm”, in Artificial Intelligence and Evolutionary Computations in Engineering Systems, vol. 394, Springer, 2016, pp. 471–480.