Applications of particle swarm optimization for neural network training and digital systems

Particle Swarm Optimization (PSO) is an evolutionary computation technique similar to genetic algorithm, which is a population (swarm) based optimization tool. PSO starts with a population of random solutions called particles. Each particle is given a random velocity and is flown through the problem space. The particles work together to achieve a global task. The best particle of the entire swarm is taken as the final solution to the task. In this thesis, three problems are studied using the PSO; their results are presented, compared and contrasted with results obtained using conventional techniques. --Abstract, page iii