A Data Collection Protocol for Local Mobile Sensor Network

In future pervasive computing environments, applications should be able to automatically adapt to their changing context such as environment status and system conditions. Therefore, context information collection is an ignored yet significant research area. An efficient sensor data collection protocol is studied in this paper for a sensor header (collector) to continuously collect data from mobile nodes in its local access area, where the mobile node refers to a relatively tiny sensor often fixed on a mobile object without high precision antenna to sense carriers emitted by other nodes. The protocol consists of registration phase when one-to-many access model is used to identify the local nodes and data collection phase when the efficient one-to-one access model can be incorporated based on the result of first phase. Furthermore, the critical performance parameters are derived to analyze and control the runtime execution of the protocol. Simulation results show that the protocol performance improves in a wide mobility range by appropriately controlling the system parameters.