Subarray Optimization For Large Digital Arrays

Effective ground clutter suppression is the primary problem of airborne early warning radar systems. Unlike ground-based radars, ground clutter is spread in Doppler frequency because of the platform motion, which lowers the detection probability of low-velocity targets. In recent years, with the developments of component technolo-gies and signal processing technologies, adaptive clutter suppression extends from one dimension to two dimensions. Therefore, space-time adaptive processing (STAP) has become a key technology in clutter suppression and has gradually entered into the stage of engineering practice. By simultaneous processing of the signals from the elements of an array antenna and the multiple pulses of a coherent process interval, STAP can provide improved detection of targets obscured by ground clutter and jamming. How-ever, the structure of antenna arrays is huge, including hundreds or even thousands of elements, which makes STAP in element level impossible. This work focuses on how to reduce the dimension in spatial domain by dividing array elements into subarrays, to achieve best system performance. The main contributions of this work are summarized as follows.1. We summarize some related works about subarray optimization, analyze the mechanism of subarray partition from three aspects:the formation of grating lobes, the rank of clutter and imperfect factors. Then we propose two classes of subarray opti-mization criterions:criterions based on array structure and element weights, criterions based on training samples.2. We formulate mathematical models for subarray optimization problem, then develop a problem-specific genetic algorithm to solve it by turning subarray optimiza-tion criterions into different object functions.3. We compare the system performance based on subarray structures according to different optimization criterions, then analyze the advantages of different optimization criterions.Both of the proposed subarray optimization criterions and algorithms work well in engineering, which could throw some insights in the related fields.