基于SVM算法的虚假航迹识别

针对云雨杂波和主被动干扰导致多雷达传感器产生虚假目标航迹的问题，利用支持向量机(SVM)算法的自主学习能力，通过构建基于数据驱动的判别模型进行虚假航迹识别。针对航迹起始得到的目标潜在航迹，利用人工智能数据驱动、自学习的特点，设计了SVM算法。通过对已标记真假的目标航迹样本进行离线学习，形成虚假航迹识别的SVM分类器，实现了基于数据驱动的判别模型代替先验知识规则约束的固定模型，并在工程应用中，利用SVM分类器在线识别虚假航迹，完成实时剔除。通过实测雷达数据实验验证，该算法的目标虚假航迹准确率高达95%以上，完全满足实际的工程应用需求。相比基于阈值或规则进行硬性判断的传统虚假航迹识别方法，所提出的算法不仅提高了准确率，还具有较高的实时性，能够适应复杂多变的杂波环境，在实际应用中具有更强的适应性和实用性。因此，提出的基于SVM算法的虚假航迹识别方法对于密集杂波场景下的虚假航迹剔除问题具有显著的实际应用价值。

False track recognition based on SVM algorithm

Aiming at the problem of false target tracks resulting from cloud and rain clutter, as well as active and passive interference in multi-radar sensor systems, the autonomous learning ability of support vector machine (SVM) algorithm is used to construct a data-driven discriminant model for false track recognition. Based on the characteristics of data-driven and self-learning of artificial intelligence, the SVM model is designed for the target potential track obtained from the initial track. Through offline learning of the target track samples that have been marked as true or false, the SVM classifier for false track recognition is realized, a data-driven discrimination model is implemented to replace the fixed model constrained by the rules of prior knowledge. The false track is identified online by SVM classifier, and real-time elimination is completed. The result of the radar data measurement experiment shows that, the accuracy of the target false track of the algorithm is more than 95%, which fully meets the actual engineering application requirements. Compared with the traditional false track recognition methods which make hard judgments based on threshold or rules, the proposed algorithm not only improves the accuracy, but also has high real-time performance and can adapt to the complex clutter environment, which makes the method more adaptable and practical in practical applications. Therefore, the false track identification method based on SVM algorithm proposed has significant practical application value for the false track elimination problem in dense clutter scenes.