Resource allocation in non-orthogonal multiple access

In the Non-Orthogonal Multiple Access (NOMA) system, Resource Allocation (RA) is examined under two headings as User Pairing (UP) and Power Allocation (PA). Resource allocation is accomplished by co-implementation of UP and PA. With UP, the users in the cell are divided into more than one group and NOMA is applied separately for each group. In this study, resources are allocated in the power domain for groups of two users each. Random Pairing, Optimal User Pairing and Adaptive User Pairing algorithms were applied separately to match users. For all these pairing algorithms applied power allocation was made separately within the groups in each algorithm. Power allocation, has been made using the Particle Swarm Optimization (PSO) algorithm, which is one of the heuristic optimization algorithms. The algorithm was run according to the criteria that the data rate of the users in each group should not be lower than the data rates in the OMA system and the sum of the power coefficients assigned to the users should be equal to one. Within the scope of the study, it is aimed to group the users with an optimal match and to increase the total data rate of the system. The results are shown by making comparisons over graphics using the MATLAB program. The user data rates obtained as a result of the power allocation made with the pairing algorithms used and the total data rate of the system were compared with the OMA user data rates and the total data rate of the system. In addition, the system's gain was calculated using the NOMA and OMA speeds of the users and it was shown with a graph. In order to understand the performance of the PSO algorithm on power allocation, the Channel State Information (CSI) based power allocation method, which is frequently used in the literature, has been used. These two methods were compared and the results were interpreted.