Complex Event Recognition From Discrete Sensor Data With a Discrete Event System Framework

Abstract

Recognizing complex events revealed by sensor readings is an increasingly crucial task that serves as one of the foundations for system monitoring and decision-making. In this paper, we investigate the recognition problem for a class of complex events that can be represented by a sequence of discrete sensor outputs. We view the outputs of discrete sensors as sensor events. We use an automaton to describe all possible sequences of sensor events that can be generated in a given system. The complex events to be recognized is a set of sequences of sensor events and can be represented by the marked language of another automaton. For a given sensor event sequence, we introduce the notion of “partition” to stand for a possible complex event sequence. By constructing an augmented automaton that includes all possible partitions, we find a necessary and sufficient condition for the recognition problem to be solvable. We then find an algorithm to check the condition. Finally, an online complex event recognition procedure is proposed to determine the occurred complex events when the complex event recognition problem is solvable.