Research And Implementation Of Grid Fisheye State Routing Protocol With QoS Constraint

Mobile Ad Hoc network is a kind of acentric, multi-hop, temporary, self-organizing wireless network, which is consisted by mobile terminals with wireless communication transceiver devices, and built at any time and any place quickly. Due to its quickness and flexibility, mobile Ad Hoc network is widely used in the areas of rescue and disaster relief, emergency services, military communication, mobile meeting, etc., and it attracts much more focus to research as well, in which the routing protocols satisfying certain QoS demand are becoming a research hot spot.GFSR, which uses the grid structure and technology of fisheye scope to transmit data packets, is a proactive routing protocol in MANET. It is robust and scalable to high mobility network. However, if data packets have certain requirement for bandwidth, the optimal path of GFSR maybe not meet this requirement, as a result, the data packets can not be transmitted to the destination node successfully, and in the meantime, they will take up too much bandwidth resources in network.To address the issues mentioned above, this thesis proposes a kind of Grid Fisheye State Routing with QoS constraint, abbreviated QGFSR. QGFSR makes improvements mainly from the following aspects:first, in order to get information about bandwidth of neighbor nodes, the bandwidth information of node is added to the link state message which is transmitted to the neighbor nodes by using multi-level fisheye scope. Second, QGFSR considers the bandwidth as a factor to select the optimal path If the bandwidth of the route meets the minimum bandwidth requirement of data packets, the route can be selected to forward the data packets. Third, in the routing tables which are established twice by shortest path algorithm in different stages, if there is no one route meeting the demand of bandwidth, then QGFSR will utilize the control strategy to prevent from forwarding the data packets to avoid the waste of network resources.Finally, this paper makes multi-scenario performance comparisons between GFSR and QGFSR based on the NS2network simulation software. The simulation Results show that QGFSR outperforms GFSR in average end-to-end delay, data packets delivering ratio and network life time.