Performance Optimization Of Cell-free Massive MIMO Systems Under Non-ideal Hardware Conditions

As the density of cellular networks continues to increase,inter-cell interference severely limits the performance of massive multiple-input multiple-output(m MIMO)systems,especially for edge users whose performance is severely affected.In order to solve this problem,Cell-Free(CF)massive MIMO systems have emerged as an important branch of distributed massive antenna systems.In this system,a large number of access points(APs)with single or multiple antennas are randomly deployed within a specified area to replace traditional base stations(BSs),and they serve all users simultaneously on the same time-frequency resource block,providing uniform service quality for all users and solving the problem of poor user experience at the edge.A key problem in implementing the CF massive MIMO system is how to deal with the growing deployment costs,as the constantly increasing demand for wireless communication requires the deployment of more APs.One solution to ensure economic feasibility is to deploy low-cost,non-ideal antenna arrays on transceivers,but the use of low-quality hardware in non-ideal antenna arrays can lead to unavoidable hardware defects.This thesis studies the impact of hardware damage on system performance and related performance optimization schemes in a CF massive MIMO system model based on non-ideal hardware conditions.Two scenarios are considered: one where the system suffers from phase offset and distortion noise,and the other where low-resolution analog to digital converter(ADCs)are used at the AP.The following main work is conducted:To address the problem of phase offset and distortion noise,this thesis uses a CF massive MIMO system model that takes into account phase offset,receiver distortion noise,and Gaussian thermal noise from amplifiers and mixers.Linear minimum mean square error(LMMSE)channel estimator is used to derive closed-form expressions for uplink transmission and rate.A power control problem is proposed to maximize the uplink rate based on the expression.The solution based on the Lagrange multiplier method is a general method for solving this problem,but it is too mathematically complex in design.Therefore,this thesis proposes an optimization scheme based on the particle swarm optimization(PSO)algorithm.The scheme continuously updates the global optimal position and search direction of the particle swarm by setting the initial feasible power coefficient particle swarm and search direction in the feasible domain,and solves the best power control coefficient.Simulations show that the PSO-based scheme improves the uplink transmission and rate compared to the Lagrange multiplier-based scheme,reduces the number of iterations required for solving,saves solving time,and improving the system performance.To address the problem of performance degradation of the CF massive MIMO system caused by deploying low-resolution ADC at the AP,this thesis uses the additive quantization noise model(AQNM)and MMSE channel estimation to derive an approximate expression for uplink transmission and rate.Moreover,the impact of AP antenna number,user antenna number,and ADC accuracy on the total uplink transmission rate is analyzed in-depth.Furthermore,an optimization problem that maximizes the total uplink transmission rate under the constraints of ADC quantization bit number and single-user transmit power is proposed.Due to the direct solution of this problem being relatively difficult,this paper proposes a solution scheme that decomposes the problem into two sub-problems based on two constraints.By embedding the power-constrained sub-problem into the ADC quantization bit-constrained sub-problem and using the solution of one problem to update the solution of the other problem,the optimal power allocation scheme is iteratively obtained through alternating iterations.An algorithm is designed to implement this solution scheme.The simulation results show that the proposed optimization scheme can improve the total uplink transmission rate of the CF massive MIMO system.