The Research Of Antenna Selection Algorithm In MU-MIMO System

As one of the technologies that can effectively improve the transmission efficiency of wireless communication,the Multi-User Multiple Input Multiple Output(MU-MIMO)technology can construct multiple independently fading transmission channels in space.Thus it has the potential to greatly increase the channel capacity of the Wi-Fi system.An important indicator that affects the final performance of MU-MIMO transmission is the condition number of the channel matrix: the smaller the condition number,the better the orthogonality between the channel vectors,and the weaker the interference between data streams.Thus,the transmission rate of the system can be higher.However,in practical Wi-Fi systems,the condition number of the channel matrix corresponding to MU-MIMO transmission is often not ideal.This is mainly because as the total number of streams of MU-MIMO transmission increases,the condition number of the corresponding channel matrix also approximately linearly increases.Meanwhile,the space for user selection in the Wi-Fi system is limited.Therefore,this thesis employed the antenna selection on the AP side to improve the condition number of channel matrix.This thesis focuses on the antenna selection algorithm for MU-MIMO transmission in Wi-Fi systems.The main work includes:(1)For single-carrier system,the influence of the condition number of channel matrix on MIMO channel capacity is theoretically analyzed;And based on the principle of minimizing the condition number of channel matrix,an antenna selection algorithm is designed based on the greedy algorithm;(2)For multi-carrier system,a suboptimal antenna selection algorithm is proposed.The biggest problem encountered in antenna selection in multi-carrier(OFDM) systems is that the optimal antenna combination corresponding to each sub-carrier is not consistent.However,there is ultimately only one antenna combination in the system can be selected,so balance among many combinations is needed.To solve this problem,a greedy antenna selection algorithm named the sum of orthogonal projections maximization is proposed.Its performance is close to the optimal algorithm.As a supplement,two algorithms named the amplitude average algorithm and the strongest time domain component algorithm are also introduced.The performance of the three algorithms mentioned above are compared;(3)For multi-link system with multiple APs in parallel transmission,an antenna selection criterion with the minimization of the quotient of the condition number under interference channel scenario is proposed,and a corresponding greedy antenna selection algorithm is proposed based on this criterion.The simulation results show that the performance gain achieved by the proposed antenna selection algorithm in a multi-link system is more significant than that of a single-link system.For comparison,an antenna selection algorithm based on maximizing the signal-leakage-to-noise ratio criterion is also introduced.The performance of this algorithm is inferior to the antenna selection algorithm based on the minimizing the quotient of the condition number criterion,but it has certain advantages in terms of complexity.The research in this thesis is helpful to provide theoretical guidance and algorithm support for improving the performance of MU-MIMO transmission in the next-generation WLAN.