Research On Reference-nets Based Cmposite Events Detection Method In Active Database And Related Application

Active database system is composed of a database and Event-Condition-Action (ECA) rule set built on the database. Events detection is the first and the most important step for ECA rule execution in active database systems. But composite events detection is not easy for most existing active database systems, the methods for composite events detection have imperfections, such as poor ability to express the semantics of the events, events which have different types of semantics can not be detection together, inaccurately compute the constituent events of a composite event for each of the parameter contexts. In this paper, a reference-nets model is proposed for composite events specification and detection to make up these shortcomings.In this paper, a reference-nets model is proposed to implement the detection of events with different types of semantics and to exactly achieve detection of composite events in different parameter contexts. Logic event which is aimed at the realization of S-A rule model is proposed to avoid Event-Condition matches and to be affluent in semantics of composite events. In order to meet the application, events with fuzzy semantics are used to improve the accuracy of detection results. The parameter contexts which are up to application are proposed for composite events detection. Token queue is used to advance the random trigger mode in reference-nets. Compared Petri-nets model with reference-nets model by experiments, the paper make the analysis of performance by using two models. In the last, improved method for traffic information detection is proposed. Then Implemention of the Highway traffic information alarming system of ShannXi Province which based on improved active database system.