Guard Node Based Ad hoc On-demand Multipath Distance Vector Routing in the Presence of Selfishness

碩士 === 國立臺中教育大學 === 資訊工程學系 === 104 === With the development of Mobile Ad hoc Networks (MANET), and the mobility of nodes, network topologies may change constantly. Ad hoc On-demand Multipath Distance Vector routing (AOMDV) is a common routing protocol, which modified the AODV protocol to reduce the...

Full description

Bibliographic Details
Main Authors: LEE,MIN-CHI, 李敏綺
Other Authors: WANG,TSAN-PIN
Format: Others
Language:zh-TW
Published: 2016
Online Access:http://ndltd.ncl.edu.tw/handle/z4ahet
Description
Summary:碩士 === 國立臺中教育大學 === 資訊工程學系 === 104 === With the development of Mobile Ad hoc Networks (MANET), and the mobility of nodes, network topologies may change constantly. Ad hoc On-demand Multipath Distance Vector routing (AOMDV) is a common routing protocol, which modified the AODV protocol to reduce the delay and consumption while restarting the route discovery by multipath extensions. If the route breaks while transferring, the packet will be send through the alternate paths. With limited resources in MANET, the node may have selfish behaviors to reduce its consumption, but the original AOMDV method does not take into account the presence of selfish nodes. Therefore, to maintain the security of the transmission path, we modified the AOMDV, and proposed Guard node-based Ad hoc On-demand Multipath Distance Vector routing (GAOMDV) method by introducing guard nodes to monitor the transmission path. First, guard nodes around the senders on the transmission path will monitor the transmission. Moreover, the sender may launch a simple agreement mechanism to find out the suspicious selfish and/or malicious nodes. The results observed by the guard nodes will be sent to the sender for making routing decisions. Finally, based on the voting results, the sender determines whether the primary transmission path should be replaced by the other paths or not. We use NS2 simulator to compare AOMDV and GAOMDV, with different ratios of selfish nodes and different voting thresholds, and to observe transfer rate and packet delay. To find other malicious nodes hidden in the voting nodes, we designed a simple game to find the best voting strategy.