Research On The Non-orthogonal Multiple Access Technology And Its Detection Algorithm

With the commercialization of the 5th Generation(5G)mobile communication system,a large number of mobile devices are connected to the system,triggering discussions on how to improve the spectrum efficiency.Compared with the orthogonal multiple access,the nonorthogonal multiple access(NOMA)technology has obvious technical advantages in terms of improving system spectrum efficiency and enhancing user connection capabilities.It is expected to solve the problems of a sudden increase in terminal access and more diversified business scenarios in future mobile communication systems.However,the inherent multiple access interference of the NOMA system will bring difficulties to the signal detection at the receiver.Multi-User Shared Access(MUSA)is one of the NOMA technologies based on code domain multiplexing.This technology uses complex multi-element extension sequences with low correlation as the codebook,which can support high overload and grantfree(GF)access.Due to the above considerations,this thesis studies the MUSA technology and its detection algorithm,so as to provide a reference for the design of future wireless communication systems.The main work is summarized as follows:Firstly,this thesis introduces the related principles of NOMA technology and typical NOMA solutions;secondly,it explains the transmission characteristics and system model of the GF NOMA system;then simulates and analyzes the performance of the existing MUSA system multi-user detection algorithms;finally,the mathematical tools used in this thesis are introduced.Aiming at the problem of poor detection performance of existing multi-user signal detection algorithms in grant-based MUSA systems,a multi-user detection algorithm with hybrid greedy and tabu search(TS)strategies is proposed to overcome the limitation of multiple access interference on the detection performance of NOMA system under high overload.The proposed algorithm first uses the measurement function of Maximum Likelihood(ML)detection as the objective function of the combinatorial optimization problem,applies the greedy algorithm to generate the initial solution,and then employs the TS algorithm in the artificial intelligence field to find the approximately optimal solution of the combinatorial optimization problem.Algorithm complexity analysis shows that the computational complexity of the proposed algorithm is far lower than ML detection.Simulation results show that the proposed algorithm achieves good multi-user detection results under different system overload rates.Aiming at the problem of unknown user activity at the base station when the GF MUSA system performs joint detection of user activity and data,a sparsity and step size adaptive matching pursuit algorithm based on generalized Jaccard coefficients is proposed using the sparsity of signal transmission.The proposed algorithm uses the generalized Jaccard coefficient instead of the traditional inner product matching criterion,which can more accurately measure the correlation between the atom and the residual signal to improve the probability of successful signal reconstruction.At the same time,an adaptive variable step size mechanism is adopted improve the accuracy and speed of signal detection.The simulation results show that the proposed algorithm can realize the joint detection of user activity and data information,and has higher detection performance than traditional algorithms.