Efficient Pattern Synthesis of Separable Cylindrical Antenna Array Employing Alternating Fast Segmented Cyclic Convolution and FFT

The synthesis of separable uniform cylindrical arrays (SUCAs) remains challenging because decomposing a target pattern into independent uniform linear array (ULA) and uniform circular array (UCA) components is ambiguous and undermines accurate control of the overall pattern. To overcome this, an alternating fast segmented cyclic convolution and fast Fourier transform (AFSCC-FFT) method is proposed. The method models an SUCA as an equivalent UCA with an iterative update element pattern incorporating the ULA factor and the basic element pattern. Thus, the transformations between array pattern and excitations are accelerated by performing inverse FFT (IFFT) and FFT for the ULA part, as well as performing fast segmented cyclic convolution (FSCC) and inverse FSCC for the equivalent UCA, achieving low computational complexity. Pattern constraints are achieved by developing an alternating pattern update strategy that iteratively updates the ULA and equivalent UCA patterns, jointly accounting for both subarray contributions to the total pattern, which effectively avoids amplitude and phase ambiguities used in conventional forced pattern decomposition. Numerical examples and full-wave simulations are conducted to validate the efficiency and accuracy of the proposed method, as well as its superiority over existing approaches.