Research On Power Allocation Scheme Of Cell-Free Massive MIMO System

According to the development rhythm of "use one generation,build one generation,research one generation " in the mobile communication industry,it is expected that the sixth generation(6G)mobile communication system will be used for commercial use around 2030.Compared with the present mobile communication system,the future 6G mobile communication system will be more flexible,intelligent,safe and reliable.Its larger transmission rate and capacity will meet more application scenarios.At present,there is no unified standard for 6G mobile communication system.According to the development of research at home and abroad,it is foreseeable that cell-free(CF)massive MIMO is expected to become one of the 6G standard technologies.Traditional cellular architecture improves spectrum utilization through frequency reuse and cell partitioning.However,with the continuous reduction of cell area,the development of traditional cellular architecture is limited by the problems of interval interference and frequent handoff,which leads to bottlenecks in the improvement of system performance.In order to solve these bottleneck problems,CF network architecture breaks through the limitations of the traditional cellular network architecture.Instead of dividing network cells or assigning users to a base station,CF network architecture assumes that every access point will provide services for all users in an area,which has become one of the feasible technical solutions concerned by people.China upholds the concept of green development and points out the direction for the sustainable development of mobile communication.In order to save energy consumption,reduce hardware costs and improve system performance to conform to the development concept of green communication,this thesis mainly studies the power optimal allocation of CF massive MIMO systems.This thesis proposes a max-min rate power allocation algorithm based on zero-forcing(ZF)receiver.Based on ZF receiver and low-resolution analog-to-digital converter(ADC),a quantizative strategy is proposed.Based on the quantitative strategy,a total rate maximization algorithm is proposed.The specific contributions of the thesis are as follows:Currently in CF massive MIMO systems,most of the uplink power optimization algorithms have high time complexity.We are trying to find a simple and efficient power optimization algorithm.On the basis of the system equipped with a ZF receiver,we propose a power optimization model that maximizes the minimum average power of the uplink.We use MATLAB to simulate the proposed power optimization algorithm and compare it with the most common full power algorithm.The simulation results show that the 5%-outrage rate of our proposed algorithm is greater than the5%-outrage rate of the equal power coefficient algorithm and can provide a more uniform capacity coverage,and the time complexity of the algorithm is very low.In addition,we weighted the proposed algorithm to ensure that some important users get better quality of service(Qo S).Then,in the uplink of the CF massive MIMO system,we quantify the pilot and uplink data signal based on low-resolution ADC and ZF receivers.We deduced a new rate expression to study the system performance when the system uses low-resolution ADC.We conduct simulation experiments based on this rate expression to verify the performance that the system can achieve under different quantization levels.Experimental results show that,under given conditions,6-bit ADCs can replace perfect ADCs,and the system performance at this time is equivalent to the system performance without quantization error,which greatly reduces the power consumption of the system.Finally,we propose a sum-rate maximization algorithm based on the new rate expression derived.The algorithm is limited by the power constraints and Qo S constraints of each user.Since the original power optimization problem is non-convex,we use the Sequential Convex Approximation(SCA)method to transform it into an iterative geometric programming(GP)problem.We compare the proposed optimization algorithm with the full power strategy.The simulation results show that the total rate maximization algorithm can effectively improve the sum-rate in CF massive MIMO system under the conditions of different access point numbers,user numbers,and antenna numbers.