Peer-to-peer Wireless Sensor Network Data Acquisition System with Pipelined Time Division Scheduling
Chellury Ram Sastry, Chi Ma, Michael Loiacono, Nazif Cihan Taş, Vladimir Zahorcak · 2006
this paper, we introduce a peer-to-peer sensor network (P2PWSNDAS) architecture with pipelined time division scheduling for sensor data acquisition from an ad hoc wireless sensor network (WSN). P2PWSNDAS takes a service-oriented, data-centric view of the deployed WSN. The wireless sensors that constitute the WSN measure physical attributes of the environment they are deployed in such as temperature, pressure, vibration, toxic chemical, biological agents, etc. In our architecture, we assume that the entire WSN can be divided into sets of 'sensor-clusters'. Each cluster has a gateway or base station that aggregates data from its sensor cluster. The gateways themselves which constitute the middleware layer between the WSN and end point clients are designed to form an IP-based distributed peer-to-peer (P2P) overlay network. Clients query the gateways for real-time sensor data via a publish/subscribe/notify paradigm implemented at the gateways, and they can also concurrently receive notifications on alarms/events of interest. Another contribution we report in this paper is a Pipelined Time- division Model (PTM) scheduling for continuous energy-aware MAC-layer communication between a gateway and its sensors. Our PTM scheduling algorithm is particularly targeted towards wireless sensors such as the MICA2 series [4] that have multichannel radio transceivers.