Situation evaluation method of reliability transmission algorithm based on root node priority search

Abstract

The invention provides a situation evaluation method of a reliability transmission algorithm based on root node priority search, which comprises the following steps of: 1, inputting key nodes in a battlefield intelligence system, and constructing a key point set N = {n1, n2, n3, n4... N13}; wherein the key nodes comprise an observation node and an evidence node; 2, searching other nodes by utilizing an SADBP algorithm, and if one path of the node can reach any key node, adding the key node into an important node; 3, reestablishing a Bayesian network, namely a simplified network, for the important nodes; 4, according to the initial probability table, obtaining a probability table of a node corresponding to the simplified network; 5, obtaining an evidence node and an observation node of thecorresponding simplified network according to the initial observation node and the evidence node; 6, performing evidence input and reasoning calculation on the simplified Bayesian network by adoptingan SADBP algorithm to obtain a situation evaluation reasoning result of the observation node; and 7, obtaining a situation result, and ending situation evaluation reasoning.

Description

technical field [0001] The invention relates to the technical field of situation assessment, in particular to a method for situation assessment of a battlefield intelligence system. Background technique [0002] The battlefield intelligence system fuses and processes intelligence information from various information sources to form a comprehensive air situation situation, conduct attribute identification and situation assessment, and form the best plan that can reflect the actual situation, and an important part of it is Bayesian reasoning and situation assessment. The mathematical description of the situation assessment model of the battlefield intelligence system is as follows: the situation assessment accepts the input of Bayesian network reasoning and decision support system, that is, the information of various threat units in the current battlefield environment at a specific moment, which can be expressed as [0003] S(t)={P 1 (t),P 2 (t),...,P n (t)} (1) [0004] ...

AI Technical Summary

Problems solved by technology

[0012] For the Bayesian network with tree structure, the number of iterations of the belief transfer algorithm in the Bayesian inference of the belief transfer BP algorithm is large, and the reasoning speed is slow. Information is likely to propagate in repeated paths. On the one hand, the propagation The process becomes redundant, on the other hand, it may also cause information to oscillate back and forth without convergence, which is not suitable for battlefield intelligence systems

Images