Semi-supervised communication radiation source individual recognition method based on improved TSVM

Abstract

The invention belongs to the field of communication radiation source individual recognition, and discloses a semi-supervised communication radiation source individual recognition method based on an improved TSVM. The method is based on a traditional semi-supervised support vector machine TSVM method. Priori information of various distribution proportions in a training sample and a threshold adjustment factor are utilized, a previous prediction result is optimized through a preset threshold, iteration is carried out according to the confidence coefficient of the prediction result, and a traditional TSVM iteration method based on local combination search is improved. According to the method, in individual recognition of the communication radiation source, when the number of training sampleswith labels is small, compared with a TSVM method, the number of iterations of a classifier can be effectively reduced while the recognition accuracy is guaranteed, and the operand of a model is reduced.

Description

technical field [0001] The present invention relates to the field of individual identification of communication radiation sources, and proposes an improved TSVM iterative algorithm based on the Confidence of Prediction, CP-TSVM, an improved TSVM iterative algorithm based on the Confidence of Prediction, CP-TSVM. Background technique [0002] Communication radiation source individual identification (Special Emitter Identification, SEI) is to extract the unique RF signal fingerprint contained in each radiation source, and use these features to construct a classifier to distinguish the target. In order to fully extract signal fingerprint features, a large number of labeled training sets are required. However, in actual scenarios, the acquisition of labels often requires a lot of manpower and material resources, making it difficult to obtain sufficient labeled training samples. If there are not enough training samples, it will be difficult to cover all possible fingerprint infor...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to provide and propose a semi-supervised communication radiation source individual identification method based on improved TSVM (CP-TSVM) in view of the high computational complexity and long model training time of TSVM in semi-supervised communication radiation source individual identification. )

Images