Anti-Corner Reflector Jamming of Radar Seeker Based on Extreme Learning Machine

Radar seeker is the key equipment for terminal guidance of precision-guided weapons. It has significant advantages such as long-range detection range and weather protection, and plays a vital role in ensuring missile strike accuracy. In recent years, high-resolution radar seekers have been the main development object to ensure the improvement of identification accuracy and accuracy of missile-borne radar seekers. Corner reflector has the excellent characteristics of high scattering fidelity with real target, high combat cost-efficiency ratio, and has become one of the main deception and jamming means of radar seeker, which seriously affects the detection performance of radar target. Corner reflectors are often placed around armored targets. Because of the complexity of ground clutter, the scattering characteristics of corner reflectors under different ground conditions are complex and changeable, which is a powerful interference used to affect radar detection of ground armored targets. In this paper, based on extreme learning machine (ELM) training algorithm, the proposed framework can use a variety of feature mapping functions or kernels, and the proposed method can be directly applied to regression and multi-classification tasks to classify ground battlefield angular anti-jamming and real armored targets. Using the measured echo data of radar seeker, the effectiveness of the algorithm is verified, and the correct recognition rate is 90% in actual back-end data processing.