Research On Channel Equalization And Detection Technology Of Broadband Multi-antenna System

Due to the multipath effect of the wireless channel,inter-symbol interference and interlayer interference exist in the MIMO received signal,which must be eliminated at the receiving to achieve signal recovery.The traditional linear equalization detection algorithm can eliminate these two kinds of interference well in the flat fading channel.But in practice,it is mostly the frequency selective fading channel.In this case,the linear algorithm can no longer meet the requirements of the system,so non-linear algorithm must be studied.With the number of users growthing,the number of receiving and transmitting antennas can reach tens to hundreds.Inter-layer interference in the received signal has become the most important interference.However,traditional detection algorithm cannot be applicable due to performance and complexity.Therefore,the detection algorithm based on heuristic search come into being and become a new research direction.In this thesis,the equalization detection technology in the traditional wideband MIMO system is studied.First,analyze the MIMO received signal to study the causes and effects of the two kinds of interference.Later,the linear and non-linear equalization detection algorithms are analyzed in the MIMO SC-FDE system,focusing on the research of two feedback equalizers,FD-DFE and NP-DFE under deep fading.Next,the structure of the NP-DFE nonlinear equalizer is improved,and sequencing and successive interference cancellation are introduced in the feedforward part,combined with the OSIC algorithm.The MMSE-OSIC and NP-DFE joint equalization detector is proposed and compared by simulation.Simulation results show that the improved structure can more effectively eliminate inter-symbol interference and inter-layer interference in the received signal,and improve system performance.Another focus of this thesis is the research on signal detection technology in Massive MIMO system.Due to the large number of communication users,the number of transmitting and receiving antennas is equal,but the traditional linear detection performance has poor performance and the non-linear detection is very complicated.Therefore,LAS detection will be focused on.First,the traditional LAS algorithm is analyzed theoretically.Since it only accepts better solution than the current one,it is easy to enter the local optimal solution.Therefore,the derivative form of detection technology is researched on the premise that the complexity is not greatly improved.In order to further eliminate interference,the QR decomposition and LAS algorithm are combined for joint detection,and orthogonalized sorting is introduced for improvement and compared by simulation.The simulation results show that the joint detection performance is better than other forms of LAS detection,and the system has a lower bit error rate.