Unsupervised Clustering Method to Discriminate Dense Deception Jamming for Surveillance Radar

With the development of digital radiofrequency memory devices, the deception jamming has become a major jamming type in radar electronic counter countermeasures, especially the dense deception jamming. Recent years, some methods using signal features are proposed for specific deception jamming excision scene. It verifies the feasibility that uses signal features to discriminate deception jamming. However, discriminator training and simulation are indeed hard to achievable since no jamming samples can be obtained in advance. In this letter, we focus on the online dense deception jamming discrimination problem without the offline data training. Since deception jamming signals generated by jammers are definitely different from physical backscattering target echoes on signal features and dense jamming signals generated by one jammer has similar signal fingerprint features, an unsupervised clustering method with heuristic feature extraction is proposed. All discrimination processes are data driven and can be executed online without a priori information. To evaluate its effectiveness, the proposed method is applied to a group of experimental data collected from a trial C -band radar. Results corroborate the proposed method and indicate the direction of future research.