Reduced-complexity I/Q turbo detector for convolutional-coding assisted space-time trellis coding

Abstract

This paper presents a reduced-complexity turbo detector for convolutional-coding aided space-time trellis coded (STTC) schemes. The proposed technique is based on the separate in-phase/quadrature-phase (I/Q) detection principle developed for single-transmitter and single-receiver systems. Essentially, the reduced-complexity I/Q turbo detector (R-TD) decomposes the received signal into its constituent I and Q components and detects these components separately, hence reducing the number of possible signal combinations to be 'tested' by the detector. It is observed that the R-TD is capable of approaching the performance of the full-complexity turbo detector, while achieving a complexity reduction factor of 4.2 for a R = 1/2 , convolutional-coded 4-PSK 4-state STTC system communicating over two-path equal-weight Rayleigh fading channels. The performance of the convolutional-coded STTC schemes is also compared to that of the STTC benchmarker refraining from using channel coding, while maintaining a particular throughput. It was found that the convolutional-coded STTC system achieved a gain of 1.2 dB at BER=10/sup -5/, although at the expense of a significant increase in complexity.