Multi-Objective Binomial Scrambled Bumble Bees Mating Optimization for D2D Communication in 5G Networks

Abstract

Device-to-Device (D2D) communication is a gifted component for 5G as it has two innate returns such as traffic reduction and consistent quality of services. Also, D2D is more elastic and efficient and exploits less power than other techniques. Few research works have been designed to provide better D2D communication but, providing maximum throughput is still a demanding one. A novel nature-inspired approach called Multi-Objective Binomial Scrambled Bumble Bees Mating Optimization (MOBSBBMO) technique is proposed to perform D2D communication with higher throughput and lesser loss rate. The MOBSBBMO technique uses multiple objective functions to obtain the optimal mobile device for D2D communication. First, the population (i.e. number of mobile devices) gets initialized arbitrarily. The fitness function is calculated for the number of mobile devices depending on multiple objectives such as energy efficiency, bandwidth capacity, and device cooperativeness. According to the fitness, tournament selection is applied to determine the optimal mobile device. If the fitness criterion is satisfied, then the mobile device is said to be optimal. The MOBSBBMO technique uses the binomial crossover applied to obtain broods by the mating process. Next, scramble mutation is employed to get the best drone bees. Yet, again the mating process is carried out to obtain the optimal solution for the best mobile device. The proposed MOBSBBMO technique is compared with benchmark methods in terms of throughput, data delivery rate, energy efficiency, packet loss rate, and latency. Simulation results show that the proposed MOBSBBMO can notably improve network throughput.