انجمن کامپیوتر ایران

The complexity of the modern chips is rising and fundamental changes in systems design are necessary. Systemon- a-Programmable-Chip (SOPC) concept is bringing a major revolution in the design of integrated circuits, due to the flexibility it provides and complexity it caters to. Particle Swarm Optimization (PSO) is a new powerful function optimizer that is used successfully to solve problems in numerous fields. A major downside of PSO and many evolutionary algorithms is that they have essential difficulties in their huge computation time due to sequential execution in software implementation. By implementing a modified particle swarm optimizer in hardware, many of the computations can be preformed simultaneously, significantly reducing computation time compared to software. In this paper, a SOPC-based PSO framework is proposed. The obtained results indicate a speedup of up to 40 times in the elapsed computation time.