Research On The Critical Technology Of Cell-free Massive MIMO Systems

In this era of information explosion,the shortage of spectrum resources and communication rate are the main obstacles that hinder the development of communication.Massive MIMO system has been recognized as a promising technology of 5G communication system to solve the problem of rate well.Cell-free massive MIMO technology,developed on the basis of traditional large-scale MIMO and distributed MIMO technology,draws on the advantages of these two technologies.Moreover,it has higher coverage.This paper mainly studies the problems of how to allocate pilot,estimate channel and select antenna in cell-free large-scale MIMO.The specific research contents are as follows:The problems of channel estimation and power allocation based on Rice channel in cell-free massive MIMO system are studied in this paper.With the channel model of the Rice channel,we obtain the channel estimation by the least squares method and use the least mean square error method to improve the channel estimation accuracy.Ideally,the pilots used by the users are completely orthogonal,but non-orthogonal pilots are often used in practice.So there will be pilot pollution problem.We propose a suboptimal pilot allocation scheme based on the idea of maximizing the rate of all users to reduce the pilot pollution problem.Then,the access points(aps)transmit the signal to each user after a series of operations such as precoding.We deduce the closed expression of single user downlink rate by using the received signal.Then we use max-min power allocation algorithm to ensure the consistency of user performance in the coverage area.The problem of system performance based on limited feedback in cell-free massive MIMO systems is studied in this paper.We assume that the channel between the user and the ap in the time division multiplexing mode is reciprocal,and the user side knows the channel coefficient perfectly.Firstly,according to the characteristics of the channel,we design the codebook considering both channel direction information and channel quality information.Secondly,we quantify these two parts of the channel according to the codebook.Assuming that the quantized bits are limited,we study the problem that how to allocate the bits of the two parts to achieve the optimal performance of the system.Then,the performance of the finite feedback system under different precoding conditions is studied.Finally,the computational complexity and performance of different power allocation schemes are compared.The problem of antenna selection based on user grouping in cell-free massive MIMO system is studied in this paper.In order to reduce the actual complexity of cell-free massive MIMO,antenna selection is needed for each user.In cell-free massive MIMO,the number of users and antennas are relatively large,and if we apply antenna selection for each user,the complexity will be very high.Therefore,we assign users with highly similar channel to the same group.One group only need to make one antenna selection for all users.In order to reduce the complexity of antenna selection when performing antenna selection,we propose an incremental antenna selection algorithm.