Skip to Main Content
To reduce the energy consumption and maximize the lifetime of the wireless sensor network (WSN), different techniques are developed to reduce the communication over WSN. However, it is also critical to discuss the energy cost problems in computing and maintaining the information of aggregate queries. Currently, some in-network temporal data suppression techniques are developed for more efficiently processing data aggregation queries over WSN. With those techniques, a data source node which sends data only forwards newly collected data with its value changing beyond predefined constants, ± ε to a data aggregate node which receives data. Otherwise, an aggregate node can infer the absent data report of a source node from the data history of the same source node. In order to further reduce the data communication, hence the energy consumption of the In-network temporal data suppression techniques, in this paper, a technique which makes use the order relationship of compound sensor data is proposed to compress data size during the In-network prediction. An empirical study is carried out to show the performance gains of the proposed techniques optimized by order compression comparing to the previous In-network prediction technique.