Intrusion detection methods based on Channel State Information(CSI) are vulnerable to environment layout and noise interference, resulting in low detection rate. To solve this problem, an indoor intrusion detection method based on the algorithm of Direction-Of-Arrival(DOA) estimation for single snapshot was proposed. Firstly, the CSI data received by the antenna array was mathematically decomposed by combining the feature of spatial selective fading of the wireless signals, and the unknown DOA estimation problem was transformed into an over-complete representation problem. Secondly, the sparsity of the sparse signal was constrained by l1 norm, and the accurate DOA information was obtained by solving the sparse regularized optimization problem, so as to provide the reliable feature parameters for the final detection results at data level. Finally, the Indoor Safety Index Number(ISIN) was evaluated according to the DOA changes before and after the moments, and then indoor intrusion detection was realized. In the experiment, the method was verified by real indoor scenes and compared with traditional data preprocessing methods of principal component analysis and discrete wavelet transform. Experimental results show that the proposed method can accurately detect the occurrence of intrusion in different complex indoor environments, with an average detection rate of more than 98%, and has better performance in robustness compared to comparison algorithms.

The clustering algorithm of Gaussian Mixture Model (GMM) is sensitive to initial value and easy to fall into local minimum. In order to solve the problems, taking advantage of strong global search ability of Density Peaks (DP) algorithm, the initial clustering center of GMM algorithm was optimized, and a new Clustering algorithm of GMM based on DP (DP-GMMC) was proposed. Firstly, the clustering center was searched by the DP algorithm to obtain the initial parameters of mixed model. Then, the Expectation Maximization (EM) algorithm was used to estimate the parameters of mixed model iteratively. Finally, the data points were clustered according to the Bayesian posterior probability criterion. In the Iris data set, the problem of dependence on the initial clustering center is solved, and the clustering accuracy of DP-GMMC can reach 96.67%, which is 33.6 percentage points higher than that of the traditional GMM algorithm. The experimental results show that, the proposd DP-GMMC has better clustering effect on low-dimensional datasets.