| Wireless sensor network (WSN) is a multi-hop and self-organizing wireless networks without fixed infrastructure and can be used for sensing, collecting and processing information of monitored object. WSN can be widely used in military, industrial automation, intelligent transportation, environmental monitoring and it now draws more and more international concern. For the very limited computing power, storage capacity, and communication capability, as well as the energy carried by the nodes, the routing algorithm for wireless sensor networks become particularly important.Be different from proactive and reactive algorithm, geographical routing neither needs to food routing query in the whole network nor to store large routing table. Routing decision can be made by a node only rely on the geographical information of the neighbor nodes and the destination node. There's no need to know the whole network topology. So the geographical routing algorithms perform well on scalability, robustness and overhead. Using the greedy routing algorithm in geographical routing algorithms may encounter the void problem. GPSR combines the greedy routing and the perimeter routing to avoid the void problem. The landmark node is proposed by GLR to optimize the triangular routing problem and the blind detouring problem of GPSR. GLR presents the destination shaded area and increases the utilization of the landmark node and decreases the overhead of the network.Based on GPSR, GLR, ITGR, LGPR (Landmark based Greedy Perimeter Routing) is proposed. The destination shaded area and the source shaded area can be defined by processing landmark based discovery routing one time. This can increase the utilization of the landmark node. And the virtual destination node is proposed to modify ITGR when the destination node doesn't lie in the destination shaded area. Compared with GPSR and ITGR, LGPR has shorter average hops and smaller overhead under same simulation environments in OMNeT++4.0. |
PDF Full Text Request