A radar radiation source signal identification efficiency evaluation method and system

Abstract

The invention belongs to the technical field of radar radiation source signal identification and evaluation in electronic countermeasure, and discloses a radar radiation source signal identification efficiency evaluation method and system. The method comprises the following steps: firstly, analyzing attributes with relevance characteristics under an evaluation criterion by adopting an intuitionistic fuzzy network analysis method to obtain corresponding attribute weights; Secondly, on the basis of the interval hesitant fuzzy thought, selecting an identification rate test result more reasonably,determining elements in an interval hesitant fuzzy initial decision matrix, and providing more accurate original data for a subsequent decision process; And finally, on the basis of the interval hesitant fuzzy thought, achieving evaluation of the radar radiation source signal recognition efficiency by using a cancellation and selection conversion method and combining an approximation ideal sorting method. The method can reasonably and effectively solve the problem of mutual independence of attributes, and can avoid low-efficiency assessment caused by doping of human factors and incapability of fully assessing all recognition schemes in a decision-making method.

Description

technical field [0001] The invention belongs to the technical field of identification and evaluation of radar radiation source signals in electronic countermeasures, and in particular relates to a radar radiation source signal identification performance evaluation method and system. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: the two key elements of radar emitter signal sorting and identification are characteristic parameters and classifiers, therefore, the evaluation of the recognition performance composed of these two technologies is called radar emitter signal identification A key part of performance evaluation. At present, there is no complete theoretical system for evaluating the effectiveness of radar emitter signal recognition in the field of electronic warfare. The main reasons for this phenomenon are: there are many types of characteristic parameters and classifiers applied to radar emitter signal...

AI Technical Summary

Problems solved by technology

Existing technology 1 makes an accurate evaluation of the recognition results through the radar emitter signal recognition performance evaluation method based on fuzzy comprehensive evaluation, but this method uses the analytic hierarchy process to solve the weight value, ignoring the possible correlation between attributes, and using The evaluation result obtained by this method is qualitative, and the output is the comment corresponding to the maximum degree of membership, which cannot quantitatively express the size of the evaluation result; in view of the main drawbacks of the existing technology 1, the existing technology 2 is improved based on the analytic hierarchy process The radar emitter signal identification effectiveness evaluation method based on interval approximation to ideal sorting method makes a quantitative evaluation of the identification results, but this method uses interval approximation to ideal sorting method in the decision-making process, and this method needs to determine positive and negative ideals in the process of sorting. point, the positive and negative ideal points mentioned in the existing data are all obtained by experience, which will cause a large error in the judgment result of the recognition performance, and the approximation to the ideal sorting method is not non-compensatory

Existing technology 3 avoids the problems existing in the above-mentioned technologies, and proposes to use the optimal ranking method based on relative entropy weighting to evaluate the recognition performance. However, this technology uses the entropy weight method to solve the attribute weights, ignoring the correlation between attributes. And the data to solve the attribute weight is obtained from the data in the initialization decision matrix obtained by t

Images