Channel State Information Feedback Scheme Based on Non-Convex Compressed Sensing for Massive MIMO Systems

Abstract

In this paper, we propose a non-convex compressed sensing(NCCS)-based channel state information(CSI) feedback scheme for massive multiple-input multiple-output(MIMO) systems. Combining the random vector quantization(RVQ), the proposed scheme permits a transmitter to obtain CSI with acceptable accuracy under substantially reduced feedback load. Furthermore, it recovers CSI from fewer measurements than that of existing convex compressed sensing(CCS)-based schemes even if the measurements are inaccurate and incomplete.Simulation results show that the proposed scheme achieves higher throughput than both existing CCS-based feedback scheme and random vector quantization(RVQ) feedback scheme with the same feedback load.

Beamforming Based CSI Reference Signal Transmission for FDD Massive MIMO Systems

“…되고 있다 [5][6][7] . 이처럼 거대 다중 안테나 시스템의 채 널 정보 피드백 양을 줄이기 위한 기법으로 압축센싱 (compressed sensing) [8,9] 기반 피드백 기법이 제안되 었다 [10][11][12][13][14] . (1)…”

Sparse Channel Estimation Based on Combined Measurements in OFDM Systems

“…압축센싱은 넌컨백스 기반의     최소화 방법 ≤   과 컨백스 기 반으로   -최소화 방법과 greedy algorithm이 대표적 이며, 이 방법 모두 센싱행렬의 성질과 측정벡터의 원 소의 개수가 성능에 영향을 미친다 [9][10][11] .…”