Research On Signal Detection Method Based On Quantum Bacterial Foraging Optimization In Massive MIMO NOMA Systems

With the continuous development of 5G communication,non-orthogonal multiple access?NOMA?technology is attracting more and more researchers'attention because it can serve more users through non-orthogonal allocation of time-frequency resources.At the same time,another 5G technology,massive multiple input multiple output?MIMO?,has also been widely concerned and studied because of its advantages of multiple system capacity and coverage improvement.Facing the challenge of massive user access in the future,it is of great significance to combine NOMA with massive MIMO.The significance of NOMA is to detect more user signals and expand the scope of user services.Therefore,the research of signal detection in NOMA is particularly important.Most of the existing researches on NOMA signal detection are based on the successful interference canceller?SIC?.Although SIC has simple structure and low computational complexity,it is easy to cause error propagation and low accuracy due to its own structural characteristics.Therefore,this thesis studies signal detection based on quantum bacterial foraging optimization?QBFO?algorithm.QBFO algorithm is a quantum intelligent optimization algorithm proposed by our research group.It has been proved that compared with other swarm intelligence algorithms,QBFO algorithm has the characteristics of faster convergence speed and more accurate optimization solution.In this thesis,QBFO algorithm is used to solve the problem of signal detection in NOMA system.The specific research work is as follows:Firstly,aiming at the massive MIMO NOMA uplink scheduling free system,this thesis proposes a detection scheme based on the data fusion strategy of QBFO algorithm.In the actual scenario,active users are far less than all potential users,so the overall user's active state is sparse.The proposed detection scheme consists of three steps:first,the sparse user detection problem is modeled as L₁-L₂ norm combination optimization problem,and the QBFO algorithm is used to solve the objective function innovatively to get the active user set on each antenna;Second,based on the data fusion strategy,the final active user set of base station multi antenna is obtained;Finally,use the final active user set and then use the QBFO algorithm to detect the user signal.The simulation results show that the proposed detection scheme is feasible and effective at the same time.Compared with the single antenna system,the proposed data fusion based QBFO detection scheme of massive MIMO NOMA can improve the detection performance,and under the same fusion strategy,the detection performance of QBFO algorithm is better.When the number of active users in the system increases,the QBFO algorithm can also obtain good performance.Secondly,for the downlink signal detection of millimeter wave massive MIMO NOMA system,a power domain beam non orthogonal multiple access transmission scheme is designed based on the sparsity of millimeter wave channel.On this basis,a step-by-step detection scheme is proposed.In the millimeter wave massive MIMO NOMA system,after user clustering and analog precoding design to eliminate the interference between clusters,aiming at the situation that the number of antennas at the base station and the number of users in the cluster are large,this thesis regroups the users in the cluster,designs the digital precoding vector,and carries out step-by-step detection.The detection scheme is as follows:firstly,the influence of channel is eliminated for each received signal of each cluster,and the estimation of each group of stacked signals in the cluster is obtained.Then the superimposed signal is detected by QBFO algorithm.The simulation results show that the proposed step-by-step detection scheme has good performance.