Low Complexity Signal Detection Technique for SFBC-OFDM Systems

In highly frequency selective channel, the performance of SFBC-OFDM system is severely degraded due to adjacent sub-carrier interference (ASI) effects. The ASI effects occur as the channel frequency response (CFR) does not remain constant over two adjacent OFDM subcarriers. As a result, the Alamouti method is not sufficient to separate the received signals perfectly at the receiver side. In order to cancel the ASI effects, several detection techniques are addressed namely parallel interference cancellation (PIC), successive interference cancellation (SIC), diagonalized maximum likelihood detection (DMLD), decision feedback detection (DFD) and maximum likelihood (ML). The ML method provides the optimum performance but the computational complexity is very high. Therefore, we propose a suboptimal modified decision feedback (MDF) detection method which achieves nearly same performance as the ML method with very less computational complexity compared to ML method. Finally, the performances of the above detection methods are compared in terms of symbol error rate (SER) and computational complexity.