Particle Swarm Optimization of Single-Mode Trench-Assisted Bend-Insensitive Fibers

We present a novel design strategy of single-mode trench-assisted bend-insensitive fibers (SM-TA-BIFs) based on the particle swarm optimization (PSO). A performance index (PI) is defined as the ratio of bending loss to mode-field-diameter to evaluate the fiber performance comprehensively. For the PSO design of SM-TA-BIF, the fitness function is constructed by PI and Gauss weighting function, and cutoff wavelength is applied as the constraint condition. The minimum PI can be achieved by the Gauss weighting fitness function with a target bending radius. With the same trench parameters, but a trench located further away from the center, the average PI of the optimized SM-TA-BIF could be 3.2 times lower than that of a commercial BIF at a bending radius of 5 mm within the entire C-band.