REDUCTION IN COMPUTATIONAL COMPLEXITY AND FAIR ALLOCATION OF RESOURCES IN A 5G HETEROGENEOUS NETWORK USING GLOWWORM SWARM OPTIMIZATION ALGORITHM

The increasing densification and heterogeneous nature of 5G wireless networks pose significant challenges in computational complexity, interference Management, and fair radio resource allocation. This research proposes a Glowworm Swarm Optimization (GSO) based framework for joint subcarrier and power allocation in OFDMA based 5G heterogeneous network (HetNets). The proposed algorithm exploits luciferin driven local search, adaptive neighborhood selection, and a grouping based allocation strategy to reduce the solution space and enhance convergence efficiency. This research adopts a simulation based methodology to investigate the effectiveness of the Glowworm Swarm Optimization (GSO) algorithm in reducing computational complexity and ensuring fair resource allocation in a 5G heterogeneous network. A downlink OFDMA base system model is considered, where a single based station (BS) serves multiple users under quality of service (QoS) constraints. A comprehensive system model is developed, incorporating channel state information (CSI), SINR based objective functions, and proportional rate constraints to ensure quality of service (QOS) provisioning across heterogeneous traffic classes. The GSO algorithm is employed for subcarrier allocation, while optimal power distribution is achieved using a water filling algorithm approach. The computational complexity of the proposed method is analytically derived as O(TPMN), demonstrating improved scalability compared to conventional linear, root finding, and particle swarm optimization (PSO) techniques. Simulation results obtained using Matlab software show that the proposed GSO framework achieves significant reductions in computational overhead (up to 50% compared to benchmark methods) while maintaining high fairness levels (Jain’s Fairness Index approaching unity, approximately 0.95-1.0) and competitive system capacity. Furthermore, the algorithm demonstrates robustness under dynamic channel conditions and varying traffic loads, making it suitable for real time implementation in dense HetNets deployments