Wireless Ad Hoc and Sensor Networks

Optimized Energy and Delay-Based Routing 425account while selecting nodes from a route. Due to the energy level beingconsidered in the routing protocol, there is also a balancing of energyconsumption across the network.Implementation of the OEDSR protocol was shown using the G4-SSNand ISN hardware at UMR. The protocol was shown to provide suitabletraffic rates and short E2E delays. Drop rate and E2E delay are dependenton the packet size that is being transmitted. Drop rate increases and E2Edelay decreases as the packet size decreases. A decrease in E2E delay isexpected due to the larger number of packets required to send the sameinformation; however, higher traffic volume also increases the probabilityof packet collisions on the channel and increases overhead.A series of tests taking a nominal of four hops were performed to showthe capabilities of the OEDSR routing protocol to provide the neededthroughput on the network with dynamic routing capabilities. An averagethroughput of approximately 1 kbps and an E2E delay of 0.7 sec areobserved for a nominal route.In reference to implementation, several issues were confronted. First,the issue of hardware capabilities is of concern. Selection of hardwaremust consider the complexity and memory footprint of an algorithm.The constraints of the 8-bit hardware became known during implementationof the OEDSR protocol. For example, the ISN nodes were designedto minimize the physical size of the node and reduce energy consumption.However, the selected processor does not have enough RAM tosupport the OEDSR routing. Therefore, minimum hardware requirementsin terms of memory size, processing power, energy consumption,physical size, and the corresponding tradeoffs have to be explored beforethe particular protocol is targeted and implemented. Additionally, thelimitations of the off-the-shelf radio modules are limiting current capabilitiesof the proposed solution. In particular, the 38.4-kbps limit on theinterface to the 802.15.4 module reduces the overall throughput andincreases delay at each hop, when compared to theoretical 802.15.4capabilities.ReferencesChee-Yee, C. and Kumar, S.P., Sensor networks: evolution, opportunities, and challenges,Proceedings of the IEEE, Vol. 91, No. 8, August 2003, pp. 1247–1256.Clausen, T. and Jacquet, P., Optimized link state routing protocol, IETF MANETWorking Group, Internet Draft, draft-ietf-manet-olsr-11.txt, July 2003.Esler, M., Hightower, J., Anderson, T., and Borriello, G., Next century challenges:data-centric networking for invisible computing: The Portolano Project at theUniversity of Washington, MobiCom ’99, Seattle, WA, August 1999, pp. 15–20.