基于随机扰动技术的共形阵列失效单元非凸压缩感知近场诊断方法

在基于压缩感知的阵列失效单元近场诊断方法中，使用结构化随机采样策略构造的观测矩阵约束等距特性未知，采用1范数极小化凸优化算法将无法确保阵列失效单元的高概率精确诊断.针对这一不足，本文在深入研究非凸优化算法的基础上提出了一种基于随机扰动技术的非凸压缩感知近场诊断算法.首先在失效单元个数满足稀疏性的前提下构造差异性阵列，其次按照随机欠采样方式获取近场幅相信息，最后利用所提基于随机扰动技术的非凸优化算法对差异性阵列激励进行重构，从而实现对阵列失效单元的高概率精确诊断.数值仿真实验表明，所提算法避免了由于观测矩阵的约束等距特性未知对诊断性能造成的不利影响，并且克服了非凸范数易于陷入局部最优解的弊端，有效提高了阵列失效单元的诊断成功概率.

A Non-Convex Compressed Sensing Based Method for Diagnosis of Defective Elements in Conformal Arrays Using Random Perturbation Technique with Near-Field Measurements

The restricted isometry property of observation matrix in near-field measurements is unknown using random under-sampling strategy in compressed sensing based methods,which has a negative influence on the probability of success rate of diagnosis when adopting 1 norm minimization.In order to overcome this limitation,a hybrid diagnosis algorithm using random perturbation-non convex optimization for identification of impaired sensors in conformal arrays with near-field measurements is investigated.Differential array composed of healthy array and damaged array is constructed in the case of the sparsity of the number of failed elements.Then the near-field data are acquired.Finally,accurate diagnosis with high probability is achieved by recovering the sparse excitation of differential array utilizing proposed algorithm.Numerical simulation results demonstrate that the proposed method avoids the adverse impact on the performance of diagnosis arising from the absence of apriori information on RIP of observation matrix,overcomes the problem of local minima associated to the non-convex norm,and therefore improves the probability of success rate of diagnosis effectively.