Multiple Data Sources Fusion Method for Effectiveness Evaluation of Military Communication Network

Abstract

In the evaluation process of Military communication network effectiveness (MCNE), the evaluation data mainly come from expert judgments, simulation results and test bed data, and these data cannot be directly used to evaluate MCNE because these three kinds of data are very different in form and in attribute. This research proposed a novel method which synthesizes expert judgments, simulation results, and test bed data to evaluate MCNE with respect to the characteristics of them. Using Belief map as data expression form; taking test bed data as training example; acquiring simulation data's confidence using test bed data and Support vector regression (SVR) which build regression relational model between network parameters and measures of performance; using an extensive Bayesian algorithm (EBA) to fuse data from multiple sources; we proposed an evaluation framework of MCNE based on multiple data sources. Take MCNE evaluation of a Naval Vessels Fleet as an example, we demonstrate that the fusion process proposed is credible and effective. Keywords-military communication network effectiveness; data fusion; belief map; support vector regression; extensive Bayesian algorithmI.

Risk Analysis in Military Operation Network Effectiveness Evaluation

“…Evaluation data sources. There are kinds of evaluation data sources in the MONE evaluation [8] , such as experts' experience data, infield simulation data, and outfield test data and so on. These data have different characteristics and are complementary with each other.…”