Date Added: Mar 2012
Wireless sensor network applications involve discovering patterns from observed events. Generally, prior knowledge of the patterns is not available. The data collected by the sensors are delivered to the sink and offline analyses on the data to extract patterns are conducted. This large volume of data collected affects the performance of the sensor network negatively due to the large communication overhead. The large overhead is a serious obstacle for deploying long lived and large scale sensor networks. In this paper, data mining techniques like Association mining to discover frequent patterns, and their spatial and temporal properties is studied. As the association mining is applied in-network, patterns and not the raw data streams are forwarded to the sink, thus reducing the communication overhead is reduced significantly.