Unlock instant, AI-driven research and patent intelligence for your innovation.

Method for selecting reactance characteristic combination according to multi-category radiants

A feature combination, radiation source technology, applied in the field of analysis of multi-type radiation sources, can solve the problems of uneven recognition accuracy rate of multi-category, high category recognition accuracy, low category recognition accuracy, etc., to achieve good engineering application prospects, The effect of low calculation cost and easy engineering calculation

Inactive Publication Date: 2013-07-17
THE 28TH RES INST OF CHINA ELECTRONICS TECH GROUP CORP
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to the above reasons, the above feature selection and combination methods cannot well solve the problem of unbalanced multi-category recognition accuracy—some categories have high recognition accuracy, while others have low recognition accuracy.
Although the wrapper method can theoretically obtain the best feature combination to avoid the above defects, the method has a huge exploration space and a very high computational cost, so it cannot be directly applied to the identification of multi-type radiation sources with high real-time requirements.

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method for selecting reactance characteristic combination according to multi-category radiants
  • Method for selecting reactance characteristic combination according to multi-category radiants
  • Method for selecting reactance characteristic combination according to multi-category radiants

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0038] Embodiment 1: This embodiment is an example about the t test. It is assumed that the carrier frequency signals of two types of radiation sources are detected, and the unit is megahertz (MHz). Class 1 includes carrier frequency signals of five radiation sources, and their values ​​are respectively 1001, 1000, 999, 950, and 1050, category 2 includes carrier frequency signals of 6 radiation sources, and their values ​​are 1500, 1280, 1525, 1290, 1010, 2000, respectively, in terms of carrier frequency characteristics, for category "category 1" - "Category 2" is used for t-test to judge whether the carrier frequency characteristics can distinguish this category from the recommendations.

[0039] The t test can be realized by various methods and tools, such as the TTEST function of Microsoft Excel. The t-test yielded a two-tailed p-value of 0.024073. Since this p-value is less than 0.05, it is concluded that the carrier frequency characteristics can distinguish the class pai...

Embodiment 2

[0040] Embodiment 2: It is assumed that the collected radiation source data comes from five different types of airborne radars: type 1, type 2, type 3, type 4 and type 5. Each radiation source collects four different reactance characteristic data: carrier frequency, repetition frequency, pulse width and pulse interval, which are represented by "a", "b", "c" and "d" respectively. These four reactance characteristics The distributions on the five types of radiation sources are as follows: Figure 2a The normalized values ​​shown in , b, c, and d, where each category is represented by a specific identifier. Assume n top The value of the parameter is 2, that is, two reactance feature combinations need to be constructed, and the analog pair set of each reactance feature and the p-value of the corresponding t-test have been obtained according to step 1. The p-value is marked in brackets, as shown in Table 1 and 2 , 3 and 4.

[0041] The reactive feature, the carrier frequency ("a...

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

No PUM Login to View More

Abstract

The invention discloses a method for selecting a reactance characteristic combination according to multi-category radiants. Each two of multiple types of radiants are paired into category pairs; the category differentiating capability of single reactance characteristic is measured by using the sum of a logarithm and p value obtained by check of corresponding t as an index in a differentiating category; the reactance characteristics are ordered therewith; and previous ntop reactance characteristic combinations not superposed capable of differentiating all the category pairs are selected on thebasis. The method has the advantages that: (1) the unbalance problem of characteristic category differentiating capability selected in a multi-category identifying task is avoided-the selected characteristic can only differentiate some local category; (2) the method is independent of any known experience and knowledge; corresponding computation is low in cost; and a realizing method is an engineering method; Based on the invention, the reactance characteristic combination for differentiating multiple types of radiants can be selected and an multi-category radiant identifying task is better finished.

Description

technical field [0001] The invention relates to the field of analysis of multi-type radiation sources, in particular to a method for selecting reactance feature combinations for identification of multi-type radiation sources. Background technique [0002] At present, the academic community has proposed a variety of methods to measure the category discrimination ability of features, such as information gain, entropy, chi-square test and t test, etc. However, for multi-category recognition tasks (the number of categories is greater than 2), there is a serious flaw in these methods alone - the selected features can only distinguish a part of the categories, usually the category with the majority of samples, but cannot distinguish other categories , especially for categories with a small number of samples, resulting in high recognition accuracy for some categories and low accuracy for other categories. In order to solve this problem, researchers have proposed some solutions, bu...

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
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F19/00
Inventor 徐欣张桂林王菁贺成龙
Owner THE 28TH RES INST OF CHINA ELECTRONICS TECH GROUP CORP