The increasing computational complexity in scheduling the large number of users for non-orthogonal multiple access (NOMA) system and future cellular networks lead to the need for scheduling models with relatively lower computational complexity such as heuristic models. The main objective of this paper is to conduct a concise study on ant-colony optimization (ACO) methods and potential nature-inspired heuristic models for NOMA implementation in future high-speed networks. The issues, challenges and future work of ACO and other related heuristic models in NOMA are concisely reviewed. The throughput result of the proposed ACO method is observed to be close to the maximum theoretical value and stands 44% higher than that of the existing method. This result demonstrates the effectiveness of ACO implementation for NOMA user scheduling and grouping.

The increasing demand for wireless network connections requires efficient network resource allocation. The non-orthogonal multiple access (NOMA) technology offers users sharing the same radio bandwidth to increase the bandwidth efficiency. However, the increase in the number of users demanding for the radio bandwidth and network connections will increase the required computational load for grouping the users to share the radio resources. This paper studies a heuristic method for grouping the users based on the discrete particle swarm optimization. The throughput, the average square error and the fitness function values obtained by the proposed method and the existing schemes are measured and observed. It has been demonstrated that the proposed scheme based on discrete particle swarm optimization has produced the throughput close to the upper limit. The convergence of the proposed method is mainly less than 10 iterations at different numbers of resource blocks.

Non-orthogonal multiple access (NOMA) technology meets the increasing demand for high-seed cellular networks such as 5G by offering more users to be accommodated at once in accessing the cellular and wireless network. Moreover, the current demand of cellular networks for enhanced user fairness, greater spectrum efficiency and improved sum capacity further increase the need for NOMA improvement. However, the incurred interference in implementing NOMA user grouping constitutes one of the major barriers in achieving high throughput in NOMA systems. Therefore, this paper presents a computationally lower user grouping approach based on discrete particle swarm intelligence in finding the best user-pairing for 5G NOMA networks and beyond. A discrete particle swarm optimization (DPSO) algorithm is designed and proposed as a promising scheme in performing the user-grouping mechanism. The performance of this proposed approach is measured and demonstrated to have comparable result against the existing state-of-the art approach.