Reliability and relevance in the Thresholded Dempster-Shafer algorithm for ESM data fusion

The volume, and imperfect and heterogeneous nature of data to be processed under time-critical conditions pose significant challenge for design of future Command and Control (C2) Systems for decision support. The effectiveness of a multi-source information fusion process used in such systems highly depends on the quality of information that is received and processed. This paper addresses two attributes of the quality of information, namely, the reliability defined as the relative stability and the relevance, and provides a quantitative assessment of their impact on the Electronic Support Measure (ESM) data fusion process employing the Thresholded Dempster-Shafer algorithm. The fusion algorithm's performance is evaluated using several statistical measures and ground truth scenarios.