Abstract

Recent advances in hardware technology and wireless networking have made it possible to deploy large scale ad hoc networks[1]. As these networks begin to grow in size, an efficient mechanism is needed to locate services distributed within them. Protocols such as the Service Location Protocol (SLP)[2] and Universal Plug and Play (UPnP)[3] exist for service discovery in static networks. These protocols are based on a centralized server where services can register themselves and clients can query for them. However, maintaining such a server in an ad hoc network is difficult as nodes randomly join and leave the network. In this paper we study the problems associated with service discovery by first simulating two well known service discovery techniques and investigating their limitations for large network sizes. We then combine the best features of each approach to present an innovative, scalable mechanism. We simulate and analyze this mechanism and show it to scale well for large network sizes without increasing the latencies in locating a service.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call