A review: detection techniques for LTE system

Abstract

Long term evolution (LTE) has become the fastest developing mobile system technology and is considered as the fourth generation of wireless communication systems. Multiple-input multiple-output spatial multiplexing is a key technique employed in LTE system to boost the system capacity. With this gain comes the challenge of designing efficient detectors at the receiver side to decouple the transmitted data streams. The exhaustive search detector is optimum in terms of performance but it has a high computational load in terms of time and power. Owing to its high computational load, many detection techniques have been proposed in order to achieve an efficient detection performance with low complexity in terms of computational power. This paper reviews the detection techniques used in LTE including the maximum likelihood (ML) detection, zero forcing (ZF) detection, minimum mean square error (MMSE) detection, successive interference cancellation (SIC) detection, sphere decoding detection, lattice reduction and list decoding. It is shown that Maximum Likelihood detection has the best performance, but the complexity is too high. Other detectors like ZF or MMSE and nonlinear detectors employing SIC have much lower complexity, however, they do not provide detection performance close to that of the ML detection. The sphere detection algorithm has been shown to have low average computational complexity and achieves a quasi-ML performance which gives it an advantage over to other types of detection techniques.