Field programmable gate array acceleration of bio-inspired optimization techniques for phased array design

摘要

This paper investigates the performance improvement in computational time achieved by the use of field programmable gate arrays (FPGAs) in electromagnetic simulations. The amplitudes of a linear phased array antenna are optimized to reduce the interference in multi-beam satellite communication systems by implementing the optimization algorithm and the antenna pattern calculations on a single FPGA. The ant colony optimization algorithm, which is a bio-inspired heuristic search method based on the survival skills of ants, is applied to a phased array antenna such that nulls can be placed in the array factor to reduce interference from beams operating at the same frequency band. Due to the inherently parallel nature of the algorithm, speed improvements in the orders of ten thousands compared to conventional programming using Matlab have been demonstrated by being able to pipeline and parallelize the calculations on the FPGA.