Name of Author : Yue Wu Title of Thesis : Low - Complexity Detection for Multiple Antenna Wireless

Multiple-input multiple-output (MIMO) wireless systems use antenna arrays at both the transmitter and receiver to achieve high spectral efficiency. Low-complexity detection is essential for the implementation of MIMO systems. In this thesis, a symbol detector for wireless systems using space division multiple access (SDMA) and orthogonal frequency division multiplexing (OFDM) is derived. The detector uses a sphere decoder (SD) and has much less computational complexity than the naive maximum likelihood (ML) detector in high SNR. Two new detectors for a hybrid system with the combination of spatial multiplexing (SM) and space-time block codes (STBC) are also derived. The new optimal detectors utilize the block structure of STBC and the SD. An optimal detector based on an ordering SD for MIMO OFDM systems by taking advantage of the high correlation among neighboring subcarriers in the system is also derived. The ML detection problem in MIMO systems is then relaxed and a family of constrained detectors is developed. Real constrained detectors and decision feedback detectors are proposed by forcing the relaxed solution to be real. Modulus constrained subgroup detectors are developed for both unitary and non-unitary constellations. A new ordering scheme using these constrained detectors is proposed to achieve a tradeoff between interference suppression and noise enhancement. Moreover, to mitigate the error propagation inherent in decision feedback detectors, a combined constrained and decision feedback detector is introduced. These constrained detectors are suboptimal but they have much less complexity than the SD based optimal detectors in low SNR.