Sparse Matrix based Computational Overhead Reduction in UMRT for N a power of 2

Abstract

Unique Mapped Real Transform (UMRT) is a transform which helps in frequency domain analysis of signals in the real domain. Different algorithms are developed for the computation of the unique MRT coefficients for N a power of 2 and for N an even number. They identify and place the UMRT coefficients in the form of an N N UMRT matrix.The basis matrices of this transform are observed to be sparse in nature. In this paper a new technique is proposed to reduce the computational overhead in UMRT, the size N being a power of two, exploiting the sparse nature of the basis matrices. General TermsFrequency domain analysis, Sparse representation.