A Optimal Stopping Theory Based Opportunistic Routing Algorithm In Delay Tolerant Networks

With the development of wireless network, its communication requirements for application scenes have brought new challenges to the current wireless network architecture. Hence, the new type of network architecture, named DTN (delay/disruption tolerant network), has been generated to handle the asynchronous message transmission problem in complex network. Due to the impact of certain factors, such as node’s frequent migration, sparse density, obstruction of barriers, energy consumption, limited resources and so on, DTN network owns the high-delay characteristic in the link level, meanwhile, can not maintain steady point-to-point link rate. Owing to the defects of DTN network, the wireless routing protocols for Ad Hoc network like AODV, DSR, which are based on the assumption of existing intact-point-to-point path, can not be directly applied to the DTN network. Moreover, since on-demand driven routing protocols just can maintain the path for a very short time, its receive/delivery rate is very low, approaching zero.Although frequent migration of nodes in the network will lead to the topology architecture to change continually, it also brings about new communication opportunities. How to effectively take advantage of the meeting chances of nodes to transmit messages has become the hot study topic in DTN routing field. The new "Storage-carrying-transmission" routing mechanism has made opportunistic routing methodology a possible way to handle the problems in DTN network.Based on the study and analysis of the existing opportunistic routing for DTN network, We make the surprising observation that most algorithms have roughly similar performance. Considering that, this paper mainly focuses on how to further decrease the routing redundancy overhead of opportunistic algorithms with using quality value, and has carried out certain research work creatively as follows:This paper has been enlightened by the optimal stopping theory, then applies the strategy to the transmission of node quality value in order to further reduce the routing cost, and finally proposes a optimal stopping theory based opportunistic routing algorithm. The theory study shows that the average complexity overhead of the reported strategy is O((?)) and this result outperforms other epidemic opportunistic routing algorithms. Based on analysis of the Opportunistic Networking Environment (ONE) simulator, we adopt several real-world track data of meeting and campus scenes to compare the simulation results with other relative algorithms, the simulation results show that the OSTBO routing algorithm is able to achieve good delivery statistics with lower overhead than comparable solutionThis paper also optimizes the presented OSTBO routing algorithm according to two special scenes. On one hand, we let several first-meeting nodes inherit the quality value without transmitting the message, aiming to further reduce the routing overhead. The experimental result shows the approach of making the initial nodes quit to transmit messages can bring about more benefit; On the other hand, taking the balance of node energy consumption into account, we add energy probability factor to each transmission opportunity and the theory study shows the optimized strategy can decrease the routing overhead. Moreover, the simulation results also approve the optimized routing algorithm outperforms Epidemic, OSTBO routing algorithms under equal parameter conditions, because it can effectively lower the routing overhead with much small cost of delay and prolong the whole network lifetime.