Probabilistic Event Processing with Negation Operators in Cyber Physical Systems

Abstract

With the widely used sensor devices in many CPS applications, tremendous amounts of uncertain events are produced, which are too simple to be directly used. So Complex Event Processing (CEP) is introduced to get the complex and meaningful information based on event patterns or rules from the uncertain raw data. Meanwhile, the negation operator for event processing is common and useful in CPS applications, so it is necessary to study the probabilistic event processing method with negation operators. In this paper, we take the mixed-model assembly line as an example for a CPS application. We describe a probabilistic event model and present Conditional Probability Matrix (CPM) to calculate the probability of target complex events. A probabilistic complex event detection algorithm with negation operators (PCED-N) with rNFA structure based on SASE is proposed. The experimental results show that our proposed method is efficient to detect complex events over probabilistic event streams with negation operators and it is sensitive to the pattern lengths.