Improved Markov Chain MBER Detection for Steered Linear Dispersion Coded MIMO Systems

In this paper, we present an iterative Markov chain Minimum Bit Error Rate (MC-MBER) detection aided steered Linear Dispersion Coded (LDC) structure, which amalgamates three different Multiple-Input Multiple-Output (MIMO) functions, namely Space-Time Coding (STC), Spatial Division Multiplexing (SDM) and beamforming. Furthermore, the concept of using a novel a priori Log Likelihood Ratio (LLR) threshold based technique is invoked for the sake of further reducing the computational complexity imposed. Both the EXtrinsic Information Transfer (EXIT) chart analysis and our BER performance results demonstrate that the achievable performance is substantially improved upon increasing the number of iterations I. At BER = 1times10^-4, the required E_b/N₀ value is about 2.4 dB apart from that of the maximum achievable rate in conjunction with l = 10 in a rank-deficient system transmitting Q = 6 QPSK modulated streams in Q = 2 symbol durations with the aid of N = 2 receive antennas and M = 2 transmit antenna arrays, where each of the arrays is composed of L = 2 antenna elements. It is demonstrated that the computational complexity of the a priori LLR threshold assisted MC-MBER detector is reduced by 38% at BER = 1.4times10^-5 in comparison to that of the MC-MBER detector dispensing with the thresholding technique.