Research On Algorithm Of Complexity Reduced Signal Detection For Massive MIMO Coded System

For the past few years,due to the increase of mobile communication traffic and the improvement of people's requirements for the quality of communication,more and more researchers are turning their attention to the research of communication systems with a large number of transmitting and receiving antenna arrays.We call them massive MIMO communication system.This system enables the communication process to obtain a higher data rate through diversity technology,and the efficiency of spectrum utilization is also higher.However,the use of many antenna arrays will produce much computational complexity,affect the efficiency of communication system and bring about problems of balance between communication performance and complexity.Signal detection is an important technology used to achieve better performance,which eliminates a certain amount of interference and performance error by detecting and processing the received signal,thereby improving communication performance.Previous researchers have proposed some detection algorithms,such as ML detection,ZF detection,MMSE detection and so on,which can improve the performance of communication to a certain extent.However,these methods suffer from high complexity or insufficient performance improvement.This paper focuses on finding a compromise between performance and complexity.Performance and complexity of communication system are two aspects of contradiction,so this paper uses some improved algorithms to achieve some trade-off algorithms between the complexity and performance,that is,reducing the complexity as much as possible on the basis of ensuring performance.Firstly,this paper introduces the massive MIMO system and the coding methods applicable to this system.At the same time,it analyzes and simulates the conventional detection algorithms and explains the limitations of these algorithms.Then,we propose two complexity reduced methods.These two methods are based on Neumann series expansion(NSE)and theory of channel hardening,respectively,which can be used to reduce the complexity of calculating soft bit information.In order to further improve the communication performance,we introduce a joint iterative detection and decoding algorithm(JIDD),which will get better performance when the number of antennas is large.However,the result of many iterations will also bring about an increase in complexity.For this problem,we propose two methods.One is to use the channel hardening theory mentioned above in combination with other related algorithms to reduce the unnecessary complexity in the iterative detection and decoding process.The second one is a complexity reduced algorithm based on Newton iterative(NI)method,which uses the nature of Newton iterative method to reduce the complexity of the algorithmThis paper simulates all of these proposed schemes and analyses the computational complexity.The simulations show that,in terms of bit error rate(BER),the improved methods can approach the performance of the conventional ones.Meanwhile,the complexity of our proposed algorithms are far lower than the conventional one.