Information flow and complex event processing of the sensor network communication

Abstract

Research institutes and companies are focusing intensively on Complex Event Processing, Event Stream Processing, Information Flow Processing and Surprise Event Detection methods and algorithms. This phenomenon is intensified by the rapid evolution and spreading in practice of the Internet of Things services. A huge amount of data is created continuously by the physical and logical sensors embedded into various equipment and aimed to monitor the state of physical processes. Time critical query-response mechanisms of the QoS based systems are more and more penetrating in the low bandwidth wireless sensor environment, as well. The behavior of the wireless sensor network requires special evaluation of the state information sent to and processed by the sink node. The items of the status information flow are affected by the low energy and low bandwidth characteristics of sensor networks. Despite of uncertainty of the data transmission in wireless sensor networks the energy constraint permits rarely retransmission of data affected by error. To mitigate the aforementioned issues, cognitive infocommunication algorithms may prove to be an effective solutions. Consequently, this paper presents a method to detect special events and surprises in time processes of the sensor network communication. The proposed method is applied to ZigBee sensor network (based on IEEE 802.15.4 standard) in different congestion scenarios.