Specific Radiation Source Algorithm Based on Modulated Broadband Converter

Abstract

The nonlinear characteristics of the power amplifier of the radiation source are used as the basis for the classification and identification of the radiation source. The algorithm is divided into the following steps. Firstly, the multi-channel modulation broadband converter (MWC) is used to preprocess the signal, and the compression and sampling of the signal are completed in parallel to obtain two-dimensional compressed sampling data. Then the gray-scale co-occurrence matrix is calculated for this data, and the corresponding features are extracted, and the identification and classification of radiation sources are carried out according to this feature. In order to illustrate the reliability of performance, the algorithm is compared with VMD_SF and MWC_IC algorithms with better identification performance. The results show that the proposed method is obviously better than the comparison algorithm. In addition, because the classification task of radiation sources is completed in the compressed domain, the proposed algorithm’s sampling frequency and calculation amount are much lower than the VMD_SF algorithm.