In monopulse radar antennas, the synthesizing process of the sum and difference patterns must be fast enough to achieve good tracking of the targets. At the same time, the feed networks of such antennas must be as simple as possible for efficient implementation. To achieve these two goals, an iterative fast Fourier transform (FFT) algorithm is used to synthesize sum and difference patterns with the main focus on obtaining a maximum allowable sharing percentage in the element excitations. The synthesizing process involves iterative calculations of FFT and its inverse transformations; that is, starting from an initial excitation, the successive improved radiation pattern and its corresponding modified element excitations can be found repeatedly until the required radiation pattern is reached. Here, the constraints are incorporated in both the array factor domain and the element excitation domain. By enforcing some constraints on the element excitations during the synthesizing process, the described method provides a significant reduction in the complexity of the feeding network while achieving the required sum and difference patterns. Unlike the standard optimization approaches such as genetic algorithm (GA), the described algorithm performs repeatedly deterministic transformations on the initial field until the prescribed requirements are satisfied. This property makes the proposed synthesizing method converge much faster than GA.
Optimization of Fast Fourier Transform processor using Genetic Algorithm on Raspberry Pi