基于K-均值的SVC的雷达辐射源信号识别

Radar Emitter Signal Recognition Using SVC Based on K-means

无监督学习是解决未知雷达辐射源信号识别的有效方法。Support Vector Clustering(SVC)是一种基于支持向量机的无监督聚类方法。SVC不仅时间复杂度高,而且在处理分布复杂、不均匀样本时,识别率较低。结合K-均值与SVC的优点,提出一种基于K-均值的SVC无监督聚类方法。此方法用K-均值聚类法对数据样本作初步的线性划分,将原数据样本划分成若干子样本。再将这些子样本分别映射到高维特征空间,用SVC方法去处理非线性问题。由K-均值聚类法将二次规划问题分解,大大减少SVC的计算量,降低时间消耗。相对于原数据样本,子样本的分布较为简单、均匀,容易找到更为合适的SVC参数值。对雷达辐射源信号进行聚类分析的实验结果表明,此方法处理速度较快,识别率较高。

Unsupervised classifiers are good methods for radar emitter signal recognition. Support Vector Clustering （SVC） is an unsupervised classifier based on Support Vector Machine （SVM）. SVC costs a lot of computational time and gets low correct recognition rate when the sample distribution is complex or uneven. Combining the advantages of SVC and K-means, a novel SVC based on K-means （K-SVC） was proposed First, k-means were used to divide the data set into several subsets. Then, SVC was employed to cluster each subset. K-SVC could decompose a large quadratic-programming problem into smaller ones. So, it could reduce the computational cost. When the data set was divided into several subsets, more suitable parameters could be easily found for each subset. The introduced method was tested by using radar emitter signals, and experimental results show that K-SVC can obtain better clustering performances in terms of computational efforts and correct recognition rates than other methods.