| Ad Hoc network is a kind of having no fixed infrastructure, temporary autonomous, distributed multiple hops network, by which has the function of wireless communication and routing nodes self-organized formation. Due to the mobility of the nodes in the network, network topology changes constantly. Therefore, knowing the network topology is very important for network optimization, network performance improvement, network robustness enhancement as well as network management implementation.The thesis proposes a topology discovering protocol, which is called Mobile Agent Mixed with Datagram (MAMD) protocol, on the basis of network’s scale, makes the network into clusters. Within clusters, cluster members send topology information to cluster head by the way of datagram; among the cluster heads, utilizes the mobile agent to collect and report the topology information, joining the adaptive heartbeat protocol, to reduce the number of topology discovering message; in order to make the protocol more accurate. For the topology structure among cluster heads, it utilizes DV-Hop positioning technology to locate the topology, and reuses RSSI to modify the nodes’ technology within two jumps. Then it gets the topology among the cluster heads.Emulating the proposed algorithm by using NS2 network simulation software, by changing the nodes’movement speed, the number of mobile agents and mobile agents’ migration frequency to analyze these parameters’ effect on the performance of AHB-MAMD protocol, verifies that the MAMD protocol joined the adaptive heartbeat algorithm is suitable to the clustering Ad Hoc network topology discovering. |
