PENCARIAN RUTE BERDASARKAN PERGERAKAN NODE TETANGGA PADA PROTOKOL ROUTING AODV UNTUK MENEMUKAN RUTE YANG HANDAL DI JARINGAN MANET

The movement of nodes in a Mobile Ad Hoc Network (MANET) is very dynamic and unpredictable. Each node in this network moves freely at different speeds and directions, which causes continuous changes in the network topology. The movement of nodes affects the communication path between nodes, as connec-tions between nodes are lost or re-established according to the position of the node whether it is still with-in communication range or not. In addition, the movement of nodes also affects the performance of rout-ing protocols used in finding routes, such as AODV (Ad hoc On-Demand Distance Vector). The AODV routing protocol does not take into account the movement of neighboring nodes in finding routes, so the route found can be quickly disconnected because the nodes involved in the route move away, out of range of the communication signal. This research aims to improve the performance of the AODV routing proto-col by considering the movement of nodes in the route finding process using the TimePrediction algorithm. The approach used includes modifications to Hello messages, so that they also include position and ve-locity information of neighboring nodes. Hello messages are packets sent periodically by each node in a MANET network to inform neighboring nodes of its presence. The results show that from a given node speed variation, TimePrediction-AODV routing protocol is able to provide better average throughput im-provement of 8.63% for 20 nodes, 29.10% for 40 nodes, and 60.70% for 60 nodes, better average PDR improvement of 6.06% for 20 nodes, 20.54% for 40 nodes, and 27.50% for 60 nodes, and average end-to-end delay reduction of 31.53% for 20 nodes, 28.86% for 40 nodes, and 43.98% for 60 nodes compared to standard AODV.