Abstract: Geographic routing is of interest for sensor networks because a point-to-point primitive is an important building block for data-centric applications. Present a new geographic routing algorithm, Improved Greedy Distributed Spanning Tree Routing (GDSTR), which finds shorter routes and generates less maintenance traffic than previous algorithms. While geographic routing potentially scales well, it faces the problem of what to do at local dead ends where greedy forwarding fails. Existing geographic routing algorithms handle dead ends by planar zing the node connectivity graph and then using the right-hand rule to route around the resulting faces. The proposed system the definition of a new kind of spanning tree, which called hull tree, for use in networks where each node has an assigned coordinate. A hull tree is a spanning tree where each node has an associated convex hull that contains within it the locations of all its descendant nodes in the tree. Hull trees provide a way of aggregating location information and they are built by aggregating convex hull information up the tree. This information is used in routing to avoid paths that will not be productive; instead we are able to traverse a significantly reduced sub tree, consisting of only the nodes with convex hulls that contain the destination point uses new kind of spanning tree, which it calls hull tree, for use in networks where each node has an assigned coordinate. The experimental result show the best performance to reduce routing overhead and less delay compare with existing routing protocol.
Keywords: Geographic Routing, Greedy Distributed Spanning Tree Routing, Hull Tree, Routing protocol.