AbstractThis work generalizes the iterative fast Fourier transform (FFT) via least‐square active element pattern expansion method to efficiently synthesize two‐dimensional (2D) patterns with accurate sidelobe and mainlobe shape control for planar arrays with radomes. The 2D active element patterns (2D‐AEPs) of a planar array‐radome structure are utilized to include the mutual coupling and radome effect. The 2D‐LSAEPE method is presented to approximate each 2D‐AEP as the pattern by exciting a virtual rectangular subarray with identical element patterns, and the weighting coefficients of the virtual subarray are found by minimizing the least‐square error of the 2D‐AEP approximation. With the help of the 2D‐LSAEPE, the pattern of a planar array with a radome can be efficiently calculated by FFT. To apply the FFT via 2D‐LSAEPE for planar array pattern computation, a blocked accumulating matrix is introduced which relates the virtual and physical excitation vectors. In addition, an interpolation‐based pre‐processing technique is implemented to obtain the virtual element pattern data in uniform grid of the (u, v)‐space so as to meet the requirement of employing the 2D‐FFT. By integrating the FFT with the 2D‐LSAEPE and the interpolation technique, an efficient iterative method is developed to implement accurate pattern synthesis for planar arrays with radomes. Three examples of synthesizing different planar array‐radome systems are conducted. Synthesis results show that the proposed method achieves much more accurate synthesis result than the traditional FFT‐based technique, while taking much less computation time than some other methods. This verifies the effectiveness of the proposed method.