Parallel low-complexity M-PSK detector for large-scale MIMO systems

MIMO systems have become a de-facto standard in modern wireless communication, and with the upcoming of large-scale MIMO systems, efficient symbol detection techniques become more significant than ever. In this paper, we propose a new symbol detection technique for M-PSK modulated symbols. In order to implement the proposed technique efficiently, we develop an iterative, low-complexity algorithm. Each iteration of the developed algorithm comprises of independent subproblems, which can be solved independently and concurrently on a parallel-computing platform, making it ideal for advanced wireless systems. Our complexity analysis reveals that the computational complexity of the proposed algorithm scales well with the number of transmit/receive antennas of the MIMO systems, and it is independent of constellation size. Simulation results show that the bit error rate (BER) of the proposed detection method is lower than those of conventional suboptimal detectors, and the BER decays faster as the size of the MIMO system increases.