Linear array design using Bayesian parameter estimation

In the course of designing and optimizing antenna arrays, many difficulties are often experienced. Due to the complex nature of modern arrays, finding the optimal configuration from scratch can be a challenging task. Many synthesis options are available for certain array types, however they are not without their problems as well. The paper deals with a technique for optimizing an array using Bayesian sampling and inference techniques. This technique can be applied to any array with any configuration to determine the best fitting parameters for a given radiation pattern. We also discuss how the procedure is implemented in modern programming methods and its applicability to parallelization. Results show that this method allows for the examination of an entire parameter space so that the best parameters for an arbitrary linear array can be quickly and efficiently examined.