Feature Extraction for Electromagnetic Environment Complexity Classification Based on Non-Negative Matrix Factorization

Assessing the complexity degree of electromagnetic environments (EME) is a typical pattern recognition problem. Feature extraction is the most significant step for EME complexity classification tasks. In his work, a novel feature extraction scheme for EME complexity classification scheme based on non-negative matrix factorization (NMF) is presented. EME signals with four complexity degree are simulated to evaluate the effectiveness of the presented method. Experimental results reveal that the NMF based features to be desirable indices for accurate complexity degree evaluation of EME.