Codesign of an IoT Using a Metaheuristic IP

Abstract

The computational method known as Particle Swarm Optimization (PSO) used when searching for a global minimum of a function, has a number of parameters that determine its behavior and efficiency in optimizing a given problem. Among these parameters, the population topology and the updating technique of particles have an important impact. In a previous work, we have proposed different strategies to enhance the performance of PSO in its software (SW) implementations. In this paper, we target a combined Hardware/Software (HW/SW) implementation of PSO as a representative case study of metaheuristic resolution approaches. Based on a deep comparison of HW/SW methodologies, we find out that codesign is well appropriate for Internet of Things (IoT) design. Besides, it is worth noticing that codesign methodology has attracted attention of researchers and industrials looking for optimizing time and energy consumption of embedded systems and IoT devices.In this work we propose a new approach of implementing PSO on Field Programmable Gate Array (PFGA) which target different architectures using metaheuristic approaches for solving optimization problems.