第四章adhoc网络路由技术.pptVIP

The basic message set includes a route request message, route reply message, route error message, and a hello message. The mechanics of each of these messages will be covered in detail later in the presentation. Briefly, however, a host (node) multicasts a RREQ message when it needs to find a route to a destination (either not already contained in its routing table, or one whose status is invalid). Make sure you explain how an intermediate node distinguishes between copies of the same RREQ. In your graph, the node in the middle receives two copies of the RREQ. In this case, the hop count from the source (A) is the same, so it doesn’t matter how its own routing back to A is updated. If you look at the example graph I included in the overview, you see that node B and the destination node F both receive multiple copies of the RREQ. In both cases, each copy followed a different route to get to the respective node. Therefore, B and F must choose the correct version of the message to use for updating their own routing tables and also for forwarding. The hop count field is the key that allows them to decide which message to keep and which message to throw away. You should point out that the RREQ/RREP mechanism depends on the assumption that links are symmetrical, or bidirectional. Charles, you may not want to include this slide, and leave coverage of the topic to whomever is covering either message types (Visal) or route discovery (Fahd?). My intention was just to introduce the basic message flow, and not get into details. In the example shown, the basic flow is clear, but some explanation is required to show how nodes B and F handle duplicates of the RREQ, i.e., how do they decide to throw away the RREQs from C, E and G? I think the Hop Count field in the message is the key, but describing this starts to encroach on later parts of the presentation. 第四章AdHoc网络路由技术 DSR协议优点 节点不需要周期性地发送路由广播分组 无需维持到全网所有节点的路由信息 节省了电池能量和网络带宽，尤其是当没有节点要发送数据时，网络中没有通信开销 仅需要维护路径上节点之间的联通 能完全消除路由环路 能同时