| dc.description.abstract | Due to the sparse node density and limited transmission range, delay tolerant networks(DTNs) are lack of continuous network connectivity. In such environments, most of the time, end-to-end paths does not exist between any pair of source and destination nodes, thus the traditional end-to-end based MANET routing protocols can not achieve satisfactory performance in DTNs. In order to improve the message delivery ratio in DTN environments, most of the DTN routing protocols apply replication-based routing, however, in resource-constraint environments, large number of message copies will consume great amounts network resource like bandwidth or storage space. Therefore, the primary goal of this study is to design a DTN routing protocol which can achieve good message delivery ratio with low message replications. Previous studies observed that humans mobility patterns have historic properties. Based on this observation, this paper proposes a routing algorithm called Timestamp Routing Scheme (TRS) which used the message delivery delay as an utility metric to operate smart routing. TRS contains two part: message delivery delay estimation scheme and routing strategy. In message delivery delay estimation scheme, each node estimated the message delivery delay by broadcasting timestamp. Through a series of analysis and design, the proposed estimation scheme can let node to estimated the message delivery delay to other nodes under low control message overhead O(n). In routing strategy, TRS takes the real environment constraint into
consideration and design corresponding routing strategies include replicate message selection, transmission queue schedule and buffer management scheme. By those strategies, nodes can use bandwidth and storage resources more efficiently and achieve better message delivery ratio under resource-constraint environments. Finally, the simulation result shows that the proposed Timestamp Routing Scheme can achieve higher delivery ratio and generate less message copies than compared routing protocols in three experiment mobility model.
| en_US |