Near optimum Sequence Detection for coded OFDM-CMM systems

Orthogonal frequency division multiplexing convolutional multi-code multiplexing (OFDM-CMM) has been proposed to achieve notable coding gain and diversity without loss of spectral efficiency. However, the complexity of maximum likelihood sequence estimation (MLSE) grows exponentially in the number of parallel streams. This detection algorithm is prohibitive when the number of parallel streams is large. The interference cancellation based sequence detection (ICSD) is proposed in this paper to reduce the detection complexity without performance loss. The heart of this method is to combine the interference cancellation and the MLSE into one iteration. The complexity of ICSD grows linearly in the number of parallel streams. The simulation results have shown that the proposed method can efficiently approximate the optimum MAP performance at a moderate signal to noise ratio (SNR) in coded OFDM-CMM systems.