Combining CS With FEKO for Fast Target Characteristic Acquisition

The compressive sensing (CS) is a new sampling paradigm that promises to accurately reconstruct compressible signals from much fewer measurements than nominal number of data points. To apply the attractive properties illustrated above, a new method named CS-FEKO is proposed in this paper, which introduces the ideas from CS into commercial software field computations involving bodies of arbitrary shape (FEKO) to improve the efficiency of FEKO in solving multistatic scattering problems. In performing a multistatic scattering analysis with CS-FEKO, the far-zone scattered fields in one scattering angle over all excitations are viewed as the signal of interest. The CS is first utilized as a preprocessing of FEKO, which modifies the conventional plane incident waves to the excitations that the CS needed. Then, FEKO is applied to solve the scattering problems under the new set of excitations to get the measurements of the scattered fields. Finally, the CS is secondly introduced as a postprocessing of FEKO, where CS inversion is employed to recover the desired far-zone multistatic scattered fields. To make a full use of the high efficiency and the deep optimization benefits of FEKO in solving wide-angle scattering problems, two forms of measurement matrixes are introduced, which have a more concise expression, and can reduce the simulating runtimes dramatically. The multistatic scattering radar cross section for different objects obtained from CS-FEKO are presented and compared with those derived from FEKO to illustrate the accuracy and efficiency of the proposed method.