A reactive role assignment for data routing in event-based wireless sensor networks

• Published in: Computer networks (Amsterdam, Netherlands : 1999) Netherlands : 1999), 2009-08, Vol.53 (12), p.1980-1996
• Main Authors: Nakamura, Eduardo F., Ramos, Heitor S., Villas, Leandro A., de Oliveira, Horacio A.B.F., de Aquino, Andre L.L., Loureiro, Antonio A.F.
• Format: Article
• Language: English
• Subjects: Access methods and protocols, osi model; Algorithms; Applied sciences; Data routing; Exact sciences and technology; In-network aggregation; Information fusion; Miscellaneous; Protocol; Role assignment; Routing; Sensors; Services and terminals of telecommunications; Simulation; Studies; Systems, networks and services of telecommunications; Telecommunications; Telecommunications and information theory; Telemetry. Remote supervision. Telewarning. Remote control; Teleprocessing networks. Isdn; Transmission and modulation (techniques and equipments); Wireless networks; Wireless sensor networks
• ISSN: 1389-1286; 1872-7069
• DOI: 10.1016/j.comnet.2009.03.009

In this work, we show how we can design a routing protocol for wireless sensor networks (WSNs) to support an information-fusion application. Regarding the application, we consider that WSNs apply information fusion techniques to detect events in the sensor field. Particularly, in event-driven scenarios there might be long intervals of inactivity. However, at a given instant, multiple sensor nodes might detect one or more events, resulting in high traffic. To save energy, the network should be able to remain in a latent state until an event occurs, then the network should organize itself to properly detect and notify the event. Based on the premise that we have an information-fusion application for event detection, we propose a role assignment algorithm, called Information-Fusion-based Role Assignment (InFRA), to organize the network by assigning roles to nodes only when events are detected. The InFRA algorithm is a distributed heuristic to the minimal Steiner tree, and it is suitable for networks with severe resource constraints, such as WSNs. Theoretical analysis shows that, in some cases, our algorithm has a O ( 1 ) -approximation ratio. Simulation results show that the InFRA algorithm can use only 70% of the communication resources spent by a reactive version of the Centered-at-Nearest-Source algorithm.