Abstract :
In this contribution the performance of multiple-input multiple-output space-division, code-division multiple-access (MIMO SCDMA) is investigated, when the low-complexity reduced-rank detection is employed. In L.L. Yang (2006) a MIMO SCDMA system is proposed, aiming at improving the capacity of the existing direct-sequence (DS) CDMA (DS-CDMA) systems. The performance of the SCDMA system in L.L. Yang (2006) is investigated, when the full-rank multiuser detectors (MUDs) are considered. However, since the rank of the detection-related matrices in SCDMA is usually high, the complexity is extremely high, even linear detectors, such as decorrelating, minimum mean-square error (MMSE), etc. are employed. In this paper the reduced-rank detection scheme, which is derived based on the Taylor polynomial approximation (TPA), is proposed for detecting the SCDMA signal. The bit error rate (BER) performance of the MIMO SCDMA is investigated, when communicating over frequency-selective fading channels. It can be shown that the considered reduced-rank detection is capable of achieving a BER performance that is close to that of using full-rank detection, but with significantly lower detection complexity.
Keywords :
MIMO communication; code division multiple access; error statistics; fading channels; matrix algebra; polynomial approximation; signal detection; space division multiple access; spread spectrum communication; BER; DS-CDMA; MIMO SCDMA; Taylor polynomial approximation; bit error rate; code-division multiple-access; detection-related matrices; direct-sequence CDMA; frequency-selective fading channels; full-rank multiuser detectors; multiple-input multiple-output space-division; reduced-rank detection; space-code-division multiple-access systems; Bit error rate; Decorrelation; Detectors; MIMO; Multiaccess communication; Polynomials; Receiving antennas; Transmitters; Transmitting antennas; Wireless communication;
