Performance comparison of compressive sparse channel estimation and feedback in 5G for hyper MIMO system

Abstract

Substantial MIMO (large-scale antenna systems, massive or hyper MIMO) differed from the current 4G technology because of preference in utilization of greater number of tune antennas over active terminals, Time-Division and Frequency-Division based Duplexing procedures. Generally Antennas are dependent on estimation accuracy of the channel matrix coefficients. In this work the Frequency division duplexing is compared with the Time division duplexing (TDD) type with feedback. Message passing methods convey multiple measurements is adopted to study the Sparsity of the hyper MIMO channels under frequency selective conditions. Initially the identification of active non-zero taps is done, followed by the channel coefficients estimation at these taps using M-AMP proposed technique. The output in FDD domain clearly manifests that the efficiency of the system is improved in terms of normalized error reduction for the parameters of SNR number of pilots and level of Sparsity.