Effective VLSI architecture for Compressed Sensing based channel estimation in ISDB-T system

Abstract

Compressed Sensing (CS) based channel estimation has been proven to give a better reconstruction results compared to conventional method. Despite promising a better result, CS based channel estimation still suffers from high complexity computations due to huge matrix multiplication and matrix inversion operations. This paper will propose an effective CS based channel estimation VLSI architecture for ISDB-T system. Since ISDB-T system is using orthogonal frequency division multiplexing (OFDM) technique for its transmission, its CS channel estimation will utilize a measurement matrix which is formed by parts of Discrete Fourier Transform (DFT) matrix. By exploiting the fact that the measurement matrix will inherit some symmetrical properties from DFT matrix, we can achieve a low complexity CS based channel estimation for practical implementation.