用改进核函数提高SVM的雷达目标识别率

Improvement of Radar Target Recognition Based on an Improved Support Vector Machine

对支持向量机中的高斯核进行了改进,利用改进的高斯核构造了一维高分辨率距离像的雷达目标识别算法,并将幂变换引入预处理过程。该技术提高了识别率,减少了识别时间;同时对所完成的目标识别算法的性能进行了评估,从方位角大小、信噪比和训练数据大小三个方面验证了该算法的稳健性。

An improved Gaussian kernel in support vector machine （SVM） is proposed. A classification algorithm for high resolution range profile （HRRP） based on this improved support vector machine is introduced. The power transformation technique is applied in this algorithm. Experiment results show that the recognition rate is improved and the computation time is reduced by using these techniques. The performance of the algorithm is evaluated in terms of azimuth angle size, signal-noise ratio and training set size, and the stability of the algorithm is verified.