The Community for Technology Leaders
Green Image
Issue No. 09 - September (2009 vol. 58)
ISSN: 0018-9340
pp: 1275-1288
Xu Li , Carleton University and University of Ottawa, Ottawa
Nicola Santoro , Carleton University, Ottawa
Ivan Stojmenovic , University of Ottawa, Ottawa and University of Birmingham, UK
We formalize the distance-sensitive service discovery problem in wireless sensor and actor networks, and propose a novel localized algorithm, iMesh. Unlike existing solutions, iMesh uses no global computation and generates constant per-node storage load. In iMesh, new service providers (i.e., actors) publish their location information in four directions, updating an information mesh. Information propagation for relatively remote services is restricted by a blocking rule, which also updates the mesh structure. Based on an extension rule, nodes along mesh edges may further advertise newly arrived relatively near service by backward distance-limited transmissions, replacing previously closer service location. The final information mesh is a planar structure constituted by the information propagation paths. It stores locations of all the service providers and serves as service directory. Service consumers (i.e., sensors) conduct a lookup process restricted within their home mesh cells to discover nearby services. We analytically study the properties of iMesh including construction cost and distance sensitivity over a static network model. We evaluate its performance in static/dynamic network scenarios through extensive simulation. Simulation results verify our theoretical findings and show that iMesh guarantees nearby (closest) service selection with very high probability, >99 percent (respectively, >95 percent).
Service discovery, distance sensitivity, localized algorithms, sensor networks, wireless networks.

N. Santoro, X. Li and I. Stojmenovic, "Localized Distance-Sensitive Service Discovery in Wireless Sensor and Actor Networks," in IEEE Transactions on Computers, vol. 58, no. , pp. 1275-1288, 2009.
112 ms
(Ver 3.3 (11022016))