Optimization of Optical Network Units Placement Using Generalized Opposition Based-Learning Starfish Optimization Algorithm

The deployment of Hybrid Wireless-Optical Broadband Access Networks (WOBAN), also known as Fiber-Wireless (FiWi) networks, provides a promising solution to meet the increasing demands of modern, technology-savvy users by integrating the strengths of both wireless and optical access technologies. A critical challenge in this context is the optimal placement of Optical Network Units (ONUs), which is essential for improving network performance in terms of cost efficiency, throughput, and transmission delay. This research proposes a novel approach to solve the ONUs placement problem in FiWi networks, called GOBL-SFOA, based on the integration of Generalized Opposition Based Learning (GOBL) strategy into the original Starfish Optimization Algorithm (SFOA) to enhance its exploration capability. The performance of GOBLSFOA was validated in terms of mean cost, under two different distributions of users: (1) random placement (2) deterministic placement. The experimental results confirm the superiority of the proposed approach over 15 methods such as the original SFOA, Henry Gaz Solubility Optimizer (HGSO), Grey Wolf Optimizer (GWO), Whale Optimization Algorithm (WOA), Black Widow Optimization Algorithm (BWOA), Harris hawks optimization (HHO), Moth Flame Optimization (MFO), Salp Swarm Algorithm (SSA), Puma Optimizar Algorithm (POA), Lizard Optimization Algorithm (DLO), Bonobo optimzer (BO), Chimp Optimization Algorithm (ChOA), Particle Swarm Optimization (PSO), Gorilla Troops Optimizer (GTO), Crown Growth Optimizer (CGO), demonstrating its effectiveness in optimizing ONUs placement for improved network performance.