Adaptive working schedule modeling for wireless sensor networks

A wireless sensor network consists of a group of sensing devices with the capability to collect data and relay it to a centralised processing system. They could be deployed and left unattended. However, one major limitation of wireless sensor networks is that sensor nodes have limited power. In general, sensor nodes can operate over a certain period of time depending on their assigned working schedule. Such schedule may be time-based, event-based or a combination of both. For an application that requires collected data to construct a data model for further analysis (e.g., wave pattern analysis), sensor nodes are ordered to periodically collect the data to ensure that the system will receive accurate data for further processing. However, doing so would shorten the overall system lifetime. To enhance the lifetime of a system, we propose a mechanism using statistical methods to adaptively construct a working schedule for each individual sensor node. With the modeled working schedule, it becomes unnecessary for a sensor node to collect data from the working field in a fixed, periodic fashion; thus, allowing a sensor node to become inactive over certain period of time. The simulation results reveal that the system lifetime is increased while the accuracy of the data collected is maintained