Research On Power Allocation Algorithm Of MIMO-NOMA System Based On User Experience Quality

Traditional orthogonal multiple access technology has been unable to meet the increasing demand of system access capacity with the rapid development of mobile communication system.Therefore,non-orthogonal multiple access(NOMA)technology is born to improve the utilization rate of unit frequency spectrum.At the same time,multiple-input multiple-output(MIMO)technology can greatly increase the number of users in the system by taking advantage of the multiple antennas,so the combination of MIMO-NOMA technology can further improve the system performance.5G communication networks need to provide high speed service with well quality of experience(Qo E)instead of just improving quality of service(Qo S).Working as a subjective feeling of users to media services,Qo E can reflect users' satisfaction with specific application services which has become a more and more important indicator in5 G communication networks.Therefore,in this study,we established the web browsing business model in MIMO-NOMA system for the first time and studied the power allocation algorithm based on Qo E,which is of great value to the development of 5G communication.A dynamic pricing power allocation algorithm based on Stackelberg game is proposed to optimize the Qo E in MIMO-NOMA system which only considers Qo S without taking the satisfaction of the user terminal into account.In this algorithm,mean opinion score(MOS)value is used to measure the satisfaction of the user terminal in the case of web browsing business which set users as the buyer and the base stations are set to be the seller,all users in the community from the base station with instantaneous unit power purchase power in order to enhance their MOS values.At the same time,the seller base station constantly adjusts the unit price of power to maximize their profits and achieve Stackelberg equilibrium ultimately.The simulation results showed that the proposed algorithm improved the fairness of the power compared with power allocation algorithm based on Qo S.Also,system performance was significantly improved compared with traditional pricing algorithm.Moreover,time complexity is obviously reduced compared with the power allocation algorithm based on convex difference programming.MIMO-NOMA systems only consider energy efficiency optimization based on Qo S.For this problem,we put forward a kind of energy efficiency optimization strategy based on Qo E,which is different from the traditional optimization indicators of energy efficiency-maximization "bits per joule".In this article,the ratio of MOS value ratio and power was regarded as the optimization index of energy efficiency,and this index was established as the utility function in Stackelberg game model.Then non-convex objective function was transformed into equivalent convex function by using Dinkelbach algorithm to get the value of power allocation under equilibrium state.Also,we proved the existence and uniqueness of equilibrium solution.Simulation results showed that the proposed algorithm significantly improves the average Qo E efficiency of the system compared with fixed pricing algorithm and random pricing algorithm.