Solving antenna array problems by multiscale hybrid fast algorithm

In this paper, Multilevel Accelerated Cartesian Expansions Algorithm (MLACEA) integrated with Multilevel Fast Multipole Algorithm (MLFMA) is applied to analysis the electromagnetic problem from large finite antenna array. Since the minimum side length of the cube in MLFMA is limited, MLFMA is numerically unstable and inefficient when applied to multiscale problems, such as antenna arrays and frequency selective surface (FSS). After combined with MLACEA, the large group cube can be redivided. Speed of near group calculation and matrix vector product will be increased. Memory cost will be decreased. Numerical examples are presented to demonstrate the efficiency and accuracy of the algorithm.