基于One-ClassSVM的机载塔康测距信息异常检测方法研究

针对多源导航信息融合系统中导航传感器数据保障的问题，本文提出了一种基于 One-Class SVM 的机载塔康测距信息异常检测方法。首先，提取机载塔康测距信息的时域参数构成特征样本空间；然后，采用 One-Class SVM 训练出机载塔康测距信息正常状态时的模型，通过发现非正常状态的样本进行异常检测。利用模拟的机载塔康测距数据进行方法验证，实验结果表明：该异常检测方法对机载塔康测距信息中的噪声有一定的鲁棒性，可以满足实际应用的需要。

Anomaly Detection of Ranging Information of Airborne TACAN Based on One-Class SVM Classifier

According to the anomaly detection for ranging information of airborne TACAN in system which fused information of multi-source navigation, this article proposed a detection method based on One-Class SVM. First, the features space is built by extracting time domain features. Then, the model of normal state is built by training the One-Class SVM classifier. Anomaly information is detected by the trained model. The experimental results show that the proposed method has good performance in anomaly detection.