A Comparative Study of FPGA Implementation of PSO and AOA for Real-Time Optimization

Abstract

The integration of metaheuristic optimization algorithms into Field-Programmable Gate Array (FPGA) architectures offers a promising pathway for enhancing the efficiency and performance of real-time embedded systems. This paper presents a comparative study of two prominent metaheuristic algorithms Particle Swarm Optimization (PSO) and Arithmetic Optimization Algorithm (AOA) focusing on their hardware implementation within FPGA environments. The primary objective is to evaluate and contrast their performance in terms of convergence behavior, hardware resource utilization, and execution time. Experimental evaluations on standard benchmark functions reveal critical trade-offs between optimization accuracy and hardware complexity. The results demonstrate that while PSO exhibits faster convergence in specific optimization tasks, AOA significantly outperforms in terms of hardware efficiency and reduced computational latency. These findings contribute to the design of optimized, resource-hardware metaheuristic implementations and highlight the potential for developing hybrid optimization strategies tailored to FPGA-based real-time applications.