Research On Parallel Detection Algorithms For MIMO Systems

Multiple-Input and Multiple-Output (MIMO) is the technology that can enhance the channel capacity and transmission reliability without increasing the transmission power or bandwidth. Because of this property, MIMO has become a key technology in the next generation communication standard. However, how to complete the detection of the MIMO signal with low complexity and meet the high-speed transmission system requirements has become a challenge issue in the field of MIMO technology.Based on three kinds of low complexity Parallel Detection (PD) schemes which are Parallel Detection for MIMO system with two transmitter antennas and receiver antennas, General Parallel Interference Cancellation (GPIC) and PD with Noise Predictive Partial Decision Feedback (NP-PDF-PDF), this paper improves the MIMO detection scheme in the following aspects:1) An enhanced Zero-Forcing Parallel Detection (ZF-PD) scheme is introduced based on the pre-detection method. This scheme solves the problem of high complexity brought by treating all the constellation points in the mapping space as the candidate decision symbol for the first detection layer signal. Theoretic analysis and simulation results show that ZF-PD algorithm can achieve the same performance of PD algorithm with lower complexity.2) With the increase of the number of antennas, the performance of NP-PDF algorithm decreases. This paper raises an improved NF-PDF algorithm which could enhance the detection efficiency with the same computing complexity.3) Meanwhile, this paper provides a new parallel detection algorithm which could achieve the detection performance of the GPIC scheme with lower complexity.