Reduced-Rank Detection Algorithm Based on MMSE-SQRD for OFDM-V-BLAST Systems

Traditional detection algorithms for frequency selective Orthogonal Frequency Division Multiplexing Vertical-Bell Labs Layered Space-Time system (OFDM-V-BLAST) suffer from high computational complexity. So it´s necessary to develop a detection algorithm to minimize the computational complexity and improve BER performance. This paper presents a new reduced-rank detection algorithm based on MMSE Sorted QR Decomposition, called Sorted QR Reduced-Rank Detection (SQRRRD). This new scheme firstly reduces the rank of receive signal vector and channel matrix with orthogonal unitary matrix. Then the reduced-rank channel matrix is decomposed by sorted QR algorithm to overcome the error propagation. In this way, we obtain well Bit Error Rate (BER) performance with lower computational complexity. Theoretical analysis and simulation results show that this proposed reduced-rank detection algorithm based on MMSE-SQRD has better performance. So we can obtain similar BER performance with much lower computational complexity compared with the traditional ones. In the OFDM-V-BLAST system with 4 transmit antennas and 8 receive antennas, using QPSK modulation, compared with traditional MMSE-V-BLAST algorithm, our new method achieves the same BER performance at higher SNR with only 0.6 times that of computational effort.