Synthesis of Contiguously Subarrayed Arrays With Geometry Control via Alternating Total Variation Norm Optimization

A novel and effective method based on alternating total variation norm optimization (ATVNO) is proposed to synthesize contiguously subarrayed arrays with geometry control. In this method, the total variation norm (TV-norm) of the element excitation vector, representing the first-order difference between adjacent element excitations, and the weighting vector are alternatively selected as optimization variables to achieve the subarray geometry control. Effective constraints on the maximum subarray count and the minimum element count in each subarray can be appropriately incorporated into the proposed ATVNO. To enable beam scanning, an effective subarray architecture is introduced, which integrates adjustable subarray-level excitations with fixed element-level phases. Additionally, to maintain consistent subarray geometry across various scanning beams, collective excitation difference vectors (CEDVs) and their energy bounds are introduced for each pair of adjacent subarray-level excitations. Consequently, the common subarray geometry required for generating multiple patterns can be determined by minimizing the number of CEDVs under multiple constraints. Three illustrative examples are performed to demonstrate the effectiveness and advantages of the proposed method.