On Optimization of Feed Network for Monopulse Radar using Sequentially Quadratic Programming

A novel method for simplifying the feed network of reconfigurable antenna arrays designed for monopulse radar applications has been introduced and evaluated. This innovative radiating system enables users to define the radiation characteristics for both sum and difference patterns while sharing the excitation amplitudes for a selected set of radiating elements. Additionally, the proposed design methodology ensures optimal feed network size, desired beamwidth, sidelobe levels, and the appropriate slope in the direction of the generated power patterns. The synthesis problem is formulated and addressed as a series of constrained minimization problems, thereby achieving performance maximization with benefits in computational efficiency and convergence to the global optimum. Theoretical insights are corroborated by both numerical simulations and experimental results, including antenna arrays.