Research On Antenna Selection And Optimal Algorithm For MIMO System

The wireless communication technology has become more and more important in socio-economic and daily life, and the users also put higher demands on quality of service of communication systems. But limited spectral has been obstacle of the development of wireless communication. The multi-input multi-output (MIMO) technology is an efficient scheme to overcome the problem. Because of its excellent capacity and spectral efficiency, multiple-input multiple-output (MIMO) system has become the direction of future development of wireless communication systems.In this thesis, we analyze the factors that may affect the capacity, rate and reliability of MIMO system in practical applications, and introduce the related theory and key technologies. At the same time, we make detailed study on the antenna selection technology, the diversity-multiplexing tradeoff and the precoding and correlated optimal schemes of multi-user MIMO (MU-MIMO) system.Antenna selection technology which is based on bit error rate (BER) criterion can improve transmission performance of spatial multiplexing system and reduce the system cost. The real-time and rapidity of algorithm is very important. Therefore, we propose a suboptimal antenna selection algorithm based on BER. Utilizing the approximate equivalent relation of BER, post-processing signal-to-noise ratio (SNR) and channel matrix, we can get the antenna subset to improve the performance of system. For reducing the cost of computation, we give out a fast antenna selection algorithm based on a complex recursive Householder QR factorization. The proposed algorithm has lower computational complexity than traditional suboptimal algorithm. Therefore, it meets the real-time and fast requirement of system. Moreover, the proposed algorithm provides good reliability and enjoys significant performance no matter correlated channel or uncorrelated.The MU-MIMO spatial multiplexing system can get good diversity gain with transmitting selection diversity. But it makes the hardware cost too high. For reducing the hardware cost, we give out a joint antenna selection algorithm based on error rate. The proposed antenna selection algorithm can achieve similar error rate probability as the preprocessing optimal method, and get satisfactory tradeoff between the performance and hardware cost.This thesis proposes a switched decision algorithm based on closed loop MIMO system. The proposed algorithm adopts the better mode according to the information of feedback, gets optimal tradeoff of the two gains, and improves the performance of system. At the same time, a fast diversity-multiplexing switching algorithm is proposed in this thesis. The proposed algorithm can get the instantaneous diversity gain by a low complexity judgment criterion. Compared with the traditional algorithm, the proposed algorithm has much lower computational cost and is suitable for real-time communication systems.Aiming at the reliability and data rate of system, a novel viewpoint of diversity and multiplexing switching algorithm is proposed based on adaptive link technique, and a low computational complexity tradeoff algorithm is give out. Based on bit error rate, the proposed scheme achieves adaptive bit and power allocation according to real-time channel state information (CSI), and the system mode is dynamically adjusted based on limited feedback bits from receiver. The proposed algorithm can obtain the diversity and multiplexing gains simultaneously, has low cost and good robustness.The processing is more challenging due to the lack of coordination among the users in a downlink MU-MIMO system where a base station (BS) transmits to different users simultaneously and in the same frequency band. To solve this problem, this thesis proposes an effective precoding algorithm and preprocessing scheme which has low complexity. The proposed algorithm can cancel the multi-user interference (MUI) and make full use of the spectral of MU-MIMO system. Compared with the traditional precoding algorithm, the proposed algorithm can get the similar BER, has lower computational complexity, and meets the requirement of real-time and rapidity of wireless communication system well.Considering the lack of coordination among the users and the influence of residual error, we propose a computationally efficient match filter algorithms for MU-MIMO system. This algorithm extends the channel matrix of ZF to the MMSE solution to minimize the mean square error (MSE), thereby achieves good performance with reduced complexity.In previous works, the multi-user spatial multiplexing system has no special strategies for the computational cost and worst channel-case user. In order to overcome the shortages, we propose a suboptimal iterative user selection method for the MU-MIMO system. The proposed algorithm achieves good sum capacity performance, has lower complexity because of avoiding to calculating the inverse of users'channels, and is suitable for practical communication system.