Multi-Objective Optimization and Efficient Pattern Synthesis for Large-scale Conformal Arrays

Designing modern antennas requires balancing several conflicting factors, such as peak sidelobe level (PSLL), aperture efficiency (AE), beam pointing error (BPE), and beamwidth (BW). These trade-offs are critical for applications like satellite communication, radar, and wireless networks. However, optimizing large conformal array antennas—with hundreds or thousands of elements—comes with major challenges. Key issues include the large number of variables to optimize, the heavy computational demands, and the fact that conformal array optimization problems are typically non-convex, making traditional methods unsuitable. To overcome these challenges, this study introduces a new method that simplifies the process while maintaining high performance. The proposed approach uses controllable surfaces to represent the excitation distribution, significantly reducing the number of variables. Three design criteria are established for these surfaces, which are built using complex exponential and Bernstein polynomial bases. By extending these bases into surfaces with the tensor product method, excitation amplitudes and phases can be calculated quickly and efficiently. This method allows large-scale conformal arrays to be optimized using only a few control parameters. Simulations and experiments confirm its effectiveness, offering a practical solution for designing large arrays.