Combat State Assessment Based on Improved Structure-Variable Dynamic Bayesian Network

Variable structure dynamic Bayesian network can effectively evaluate the combat state and be able to adapt to the sudden battlefield environment, but there is still a problem in the transformation between stages. Aiming at the problem of redundant information transmission in the variable structure dynamic Bayesian network, an improved variable structure dynamic Bayesian network is proposed to filter the information of the transfer network. The method determines whether the nodes are redundant by the conditional independence of the nodes in the adjacent time slices under the condition of the completion of the known phase, and then filters the redundant information. The experimental results show that the proposed method can effectively eliminate the redundant information and enhance the perceived sensitivity of the effective information.