基于SVM的雷达故障预诊断技术研究

The Pre-diagnosis Technology Research on SVM-based Radar Fault

摘要: 本文将支持向量机(SVM)引入雷达故障预诊断，实现了雷达进入潜伏性故障时就提前诊断。很好地解决雷达故障诊断流程复杂、诊断时间长等问题。这是由于SVM可以很好地解决小样本、非线性分类问题，而这正是潜伏性雷达故障的特点。本文最后通过实例充分说明了该算法在排除雷达潜伏性故障方面的能力。

Abstract: This paper will Support Vector Machine (SVM) bring to the pre-diagnosis of radar fault, implementing the advance diagnosis of the latent radar fault. To slove the problems best, such as the process of radar fault diagnosis complexity and long-time diagnosis. This is due to SVM can solve the small sample, nonlinear classification problem, which is the characteristics of the latent radar fault. Finally, the paper through examples fully illustrates that the algorithm has the capacity of excluding the radar latent fault.