Air radiation source individual intelligent increment identification method, system, terminal and application

Abstract

The invention belongs to the technical field of radiation source individual identification, and discloses an air radiation source individual intelligent increment identification method, system, terminal and application. Four features of fuzzy function, bispectrum transformation, Hilbert-Huang transformation and short-time Fourier transformation are extracted from received air radiation source ADS-B (broadcast type automatic dependent surveillance) signals respectively; performing linear feature fusion on the features to obtain a new feature map; and classifying and identifying the radiation source individuals of known types through a convolutional neural network to obtain a network model. For untrained category data, an incremental learning mode is adopted for training, and intelligent incremental recognition of aerial radiation source individuals is achieved. According to the method, good recognition accuracy can be achieved under the low signal-to-noise ratio, good recognition capacity can still be achieved under different channels, dependence on a certain single feature is not high, the problem that training data arrives in batches is solved, the time needed by training is greatly shortened, and the space needed by data storage is reduced.

Description

technical field [0001] The invention belongs to the technical field of individual radiation source identification, and in particular relates to an intelligent incremental identification method, system, terminal and application of an air radiation source individual. Background technique [0002] At present: Radiation source individual identification refers to the identification of target individuals by extracting one or more modulation characteristics exhibited by the received signal. With the increasing number of individual air targets, it is necessary to judge whether the radiation source of the air target is working normally and whether the attribute of the abnormal air target is friend or foe, and these judgments need to be based on the individual identification of the air radiation source. How to judge quickly and accurately , is an important means to achieve national security. [0003] The radiation source individual identification technology based on fingerprint infor...

AI Technical Summary

Problems solved by technology

[0006] (1) The existing technology has poor recognition performance in low signal-to-noise ratio environments, insufficient generalization ability in different channel environments, and individual recognition methods are highly dependent on feature extraction methods, and the recognition of different feature extraction methods Accuracy varies widely, etc.

[0007] (2) Traditional training strategies must require that all data categories be known in advance and be trained uniformly

[0009] (1) Due to the influence of the current complex electromagnetic environment and channels, it is difficult for the individual signals of common air radiation sources to achieve a high signal-to-noise ratio. The signal-to-noise ratio is low and the interference is strong. It is normal to find a strong anti-interference ability The feature extraction method with good recognition ability for most channels is more difficult;

[0010] (2) The training of streaming data needs to save the previous data, and the storage of old data has a large demand for memory space. If all of them are saved and retrained, not only the previous training results will be wasted, but also a lot of memory will be wasted. Storage space is not conducive to training speed improvement and resource maximization utilization

Images