Natural Computing Algorithms for Optimization of High-Order Distributed Raman Amplifiers

A study and performance comparison between two methods for optimizing the design of high-order distributed Raman amplifiers (DRA) are presented. Simulated Annealing (SA) and Particle Swarm optimization (PSO) are both natural computing algorithms used for searching the best solution for a given function. Both methods were applied to optimize pumping wavelengths and power for third-order DRAs. In our comparative analysis, the overall best results were obtained by the PSO algorithm. Nevertheless, SA also presented satisfactory results, being an interesting alternative method for designing high-order DRAs, specially considering its simpler implementation.