A Zero-Sample Recognition Method for Unknown Patterns of New System Radar Targets
A radar target and recognition method technology, applied in character and pattern recognition, neural learning methods, computer components, etc., can solve the problems of difficult fast and accurate recognition of data, high time complexity of online recognition, and low accuracy rate
Active Publication Date: 2020-10-02
中国人民解放军32802部队 +1
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, when the complex electromagnetic environment has strong signal density and advanced radiation sources with agile modes, traditional classification methods often cannot obtain all the feature templates and data, and there will be unknown mode signals of radiation sources that have never been seen. Algorithms such as the Attractor Propagation (Affinity Propagation, AP) algorithm and the Fast Density Peak (Cluster of FastSearching Density Peaks, CFS) algorithm can cluster signals without any prior information, but these are completely driven by data The accuracy of the method is often low and the time complexity of online recognition is very high, that is, it is difficult to quickly and accurately identify unknown data
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0045] The present invention will be described in further detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.
[0046] (1) The scene setting of the experiment
[0047] In this experiment, the radar radiation source mode is defined as a fixed combination of features. Usually, the characteristic description of some radar signals, including frequency range, pulse repetition interval, pulse repetition frequency type and pulse width, can be used to describe a specific radar. In the radiation source mode, these descriptions are used as semantic information to construct a semantic description vector of signal samples.
[0048] This experiment evaluates the method on simulated radar signal datasets in two different settings.
[0049] ① New feature combinations: Although the combination of semantic fe...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention proposes a zero-sample recognition method for the unknown pattern of the new system radar target, improves the traditional radiation source pattern recognition algorithm through the cross-layer structure and the self-encoder mechanism, and provides a one-to-one corresponding semantics for each category The description vector, in the training phase, learns the mapping from the input sample to the semantic description vector, and uses the mapping relationship to predict the semantic description of the test sample, and matches the semantic description with the known semantic description vector of the test set, and finally selects the most The categories represented by similar semantic description vectors only need to perform simple multiplication and addition operations to obtain the results during online recognition. Compared with clustering algorithms, it not only improves the accuracy rate but also eases the time complexity of online recognition.
Description
technical field [0001] The invention relates to the intersecting technical field of radar electronic reconnaissance and artificial intelligence, in particular to a zero-sample recognition method for unknown patterns of radar targets in a new system. Background technique [0002] The main task of the electronic warfare system is to discover the existence of radiation sources and generate countermeasures as soon as possible. The main components of this system are to monitor the radiation sources and identify their current working mode, laying the foundation for the selection of subsequent interference measures. [0003] Traditional radiation source pattern recognition algorithms rely on the matching of existing feature templates, or pre-collecting a sufficient number of samples for training of the recognition model. These known templates or patterns that can be learned offline are called known classes. However, when the complex electromagnetic environment has strong signal de...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/214G06F18/241
Inventor 杨健王沙飞李岩汪生田震张滋林
Owner 中国人民解放军32802部队