Improved intelligent particle swarm optimization algorithm for designing the sampling period profile of the sampled fiber Bragg gratings

We could get the index modulation structure of sampled period for the sampled fiber Bragg gratings (FBGs) by using the inverse Flourier transformation of the target channels. Using this method, the enable channels are identical wavelength operation and the unable channels are almost suppressed completely. The enable and unable channels can be setup at will based on the requirements of different applications. However, the index modulation efficiency of the sampled FBGs is very slow. To get a high efficiency, an intelligent particle swarm optimization algorithm is applied to design the phase of the target spectrum. Combing the two methods together, a high efficient multi-level phase sampled FBG that each channel is suppressed or not at will, can be obtained. We also use the particle swarms optimizations (PSO) algorithm to optimize the pure phase sampling profiles of the sampled FGB. Results showed that a much high efficiency of the FBGs can be obtained compared with that obtained from the simple inverse Fourier transformation technology. If appropriate grating period chirp and sampled period chirp is applied to such a grating, a novel FBGs based OADM or interleavers devices with dispersion or dispersion slope compensation can be designed.