Contact Duration Aware Buffer Management Forwarding Policy in Sparse Delay Tolerant Networks

Abstract

Delay-Tolerant Networks (DTNs) are target environments suffering from instability or no complete path between source and destination nodes. The contact duration between pairs of network nodes is limited due to node movement and limited range of peer-to-peer wireless communication. The pairs of node takes a long time to contact again in sparse delay-tolerant networks, which is longer than contact duration. Therefore limited contact duration needs more consideration to design buffer management forwarding policy. Several work has been done in the design of forwarding algorithms, but little work has focused on studying forwarding under the presence of limited contact duration. The pairs of network nodes require waiting till they come into communication range again when do not finish exchanging the message during the contact time. Consequently, the message which is started to transmit may drop when expired before the nodes contact again and destination node waits till to receive complete message. Also, receiver node drops a complete message to create space for incoming partial message items upon buffer overflow. To solve the raised problem, we proposed a novel contact duration aware buffer management forwarding policy. Since all intended buffered messages may not be exchanged between nodes within a single contact duration, our proposed scheme select and forward only complete transmit messages. To evaluate the performance of proposed work we have used Opportunistic Network Environment (ONE) network simulator. The proposed algorithm is integrated with the MaxProp routing protocol buffer management forwarding policy and they were analyzed on two different metrics namely bundle delivery probability and average delivery delay. Also, we considered number of node, buffer size and simulation time to test performance of proposed work. Consequently, across all the experiments the simulation results obtained in this thesis show that our proposed algorithm outperforms than existing forwarding policy.