A weighted fuzzy aggregation method

When a large number of sensors or people are able to give an opinion on some question, giving a unique answer to this question is quite difficult. Data fusion tries to adapt the computation method (i.e. the fusion rule or the aggregation rule) to the context. This field has been widely studied, in multiple applications and a lot of aggregation rules have been defined. But these rules can be used only in the case when the sensors have the same importance. We study the problem of weighted data aggregation i.e. when the sensor´s opinion does not have the same importance (or contribution) on the final decision. We suggest a way to aggregate directly data with various importance. Because the choice of the rule is determined by the application, the new rule, called a weighted rule, has to be able to compute any given aggregation rule. Therefore we propose an operator that defines a new aggregation rule for any vector of weights and for any given aggregation rule