Research of Information Fusion Evaluation Methodology Based on Dominance Relation

The theories and methods of performance evaluation for multi-sensor data fusion system based on rough set theory and a fuzzy interval are discussed. A information fusion performance criteria system is founded with C³I applications as background, the evaluation process is similar to the decision process. A function g_j(a) represents an evaluation of object a for attribute j and is called a criteria. Each criterion induces a particular ordering of the object, and it is very significant to find a procedure by which to construct one overall preference ordering. Proposing the idea of translating imprecision or ambiguity in measurement of an alternative for a given point of view with the use of a fuzzy set in R which is a set of ordered pairs {x,mu_j ^a(x)} where mu_j ^a(x) is termed "the grade of membership of x" for alternative a related to j. This offered theoretical foundation for aggregation based on all evaluations from attributes.