No abstract

In Delay Tolerant Networks (DTNs), some routing algorithms ignore that most nodes are selfish, i.e., nodes are willing to use their own resources to forward messages to nodes with whom they have a relationship. In view of this phenomenon, we propose a routing algorithm based on Geographic Information and Node Selfishness (GINS). To choose a forwarding node, GINS combines nodes' willingness to forward and their geographic information to maximize the possibility of contacting the destination. GINS formulates the message forwarding process as a 0-1 Knapsack Problem with Assignment Restrictions to satisfy node demands for selfishness.Extensive simulations were conducted, and results show that GINS can achieve a high delivery ratio and a lower hop count compared with GRONE and LPHU. Furthermore, its overhead ratio is 25% and 30% less than that of GRONE and LPHU, respectively.

Due to node's mobility, Delay Tolerant Networks (DTNs) feature the nonexistence of end-toend path between source and destination, frequent topology partitions and extremely high delivery latency, thus posing great challenges to successful message transmission. To improve routing performance and provide high quality communication service, nodes' social characteristics are exploited to routing design recently. Hence, a social popularity based routing algorithm is proposed, named SPBR which takes the inter-contact time and multi-hop neighbor information into consideration. In this paper, we first introduce a method to detect the quality of relation between pair of nodes accurately. Used the reliable relationships, social popularity is proposed to evaluate the social power of node in the network. SPBR makes the routing decisions based on the popularity, leading message closer to destinations with low hops of routing and network resources. Extensive simulations are conducted and the results show that the proposed algorithm significantly improves routing performances compared to Epidemic, Prophet and First Contact (FC), especially SPBR is lower byabout 55.1% in overhead ratio and higher by about 22.2% in delivery rate than Epidemic when there are 40 nodes in the networks.

Abstract-Delay tolerant networks (DTNs) are a type of emerging mobile wireless network that experience long delays, intermittent disruption, buffer space and energy limitations, node irregularities and frequent mobility. To this end, many protocols take advantage of the flooding mechanism to increase the probability of successful message transmission. However, sending too many copies of each message may result in large network overhead and communication cost. Based on the Spray and Wait (S&W) algorithm, we propose an improved Adaptive Spray and Wait scheme, named ASW, for different delay tolerant network scenes. In this paper, we define a hybrid utility metric combined with multiple variables, delivery probability and similarity. Then, we distribute message copies in accordance with the proportion of the utility value to optimize the delivery ratio. Extensive simulations have been conducted to verify the efficiency of our proposed algorithm. In addition to the comparison, the results also show that our proposed algorithm achieves a better performance in terms of the delivery ratio and average latency than the other three protocols.

No abstract