Tree-Structured Network Based Hierarchical Complex Event Processing in Wireless Sensor Networks

Complex event processing (CEP) is a technique that handles massive data in the Internet of Things (IoT). A lot of researches have been done in RFID. However, most of the existing CEP engines are centralized which result in the resources of sensors are underutilized and wireless network overload. In this paper, wireless sensor network (WSN) will be organized into a logical tree structure. Tree-structured WSN is combined with automatic machine based CEP to form a novel hierarchical CEP. By query planning, the expression will be decomposed to sub-expressions and sub-expressions are assigned to each sensor and gateway. By processing sub-expressions, sensors and gateways would only need to transmit useful data meanwhile their computing and storage capacity are fully utilized. Experiments show that, in this way, the network bandwidth will be saved significantly and the pressure on the server will be reduced.