VLSI-Implementation-Based Fast Symbol Detector for PQSM MIMO Systems

Abstract

Parallel quadrature spatial modulation (PQSM) multiple-input multiple-output (MIMO) technology is a variant of spatial modulation (SM) technology and has shown significant performance gain. However, few low-complexity detection algorithms with optimal or near-optimal performance have been proposed. In this paper, a near-optimal fast detection scheme is presented along with a corresponding hardware architecture. The proposed algorithm first uses the features of PQSM MIMO systems to further merge the channel matrices of transmit antenna combinations (TACs) into corresponding smaller matrices. Then, the remaining steps of the proposed algorithm include conducting Givens rotations (GRs) to QR decompose the merged matrices, sorting and selecting TAC candidates with higher transmission possibilities for complexity reduction, performing symbol estimation for each TAC candidate, and computing Euclidean distances (EDs) to find the final solution. Based on the steps of the proposed algorithm, five corresponding pipeline architectures are designed for a PQSM MIMO system with four transmit antennas, four receive antennas, and 16-QAM symbol modulation. Finally, the results of simulations, complexity analysis, and hardware implementation show that the proposed detector can provide a near-optimal bit error rate performance as well as a high throughput rate, but with lower complexity.