Yuanda Cao

Spray and Wait Routing Based on Average Delivery Probability in Delay Tolerant Network

To avoid random and blindfold forwarding decision in Spray and Wait routing in delay tolerant network, Binary Spray and Wait routing based on average delivery probability is proposed in this paper. Based on delivery predictability of PROPHET, average delivery predictability is proposed and used in Binary Spray and Wait routing to make forwarding decision. Furthermore, we evaluate it through simulations versus Spray and Wait routing protocol. The experimental results show there has slight higher average delivery rates and shorter average delays in Binary Spray and Wait routing based on average delivery probability.

Cooperation in delay tolerant networks

In delay tolerant networks (DTNs), all the networking functions rely on the contribution of the participants to fulfill the store-and-forward fashion. Since wireless nodes are buffer and energy constrained, it may not be in the best interest of a node to always accept relay request. In recent years, incentive mechanisms have been proposed to stimulate nodes to cooperate in packet forwarding. However, the lack of end-to-end paths, high variation and long feedback delay in DTNs imply that existing solutions for mobile ad-hoc networks do not apply to DTNs. In this paper, we address the problem assuming that nodes are rational to optimize self interest. We propose a model, named Pay-for-Gain (PFG), based on game theory and loan-credit theory, to investigate equilibrium point that maximizes their own interest. For comparison, the strategy tit-for-tat (TFT) and PFG are tested in existing routing protocols in DTNs.

Similarity Degree-Based Mobility Pattern Aware Routing in DTNs

Delay-tolerant networks (DTN) have the capacity to interconnect devices in roomy regions unlike traditional networks. Based on common sense, the more similar the mobility pattern of nodes, the more stable their relative positions, which results in a connection between them with a longer duration In this paper, a novel routing scheme using the similarity degree of mobility pattern is proposed to optimize routing performance. In addition, a comprehensive metric called similarity degree is applied to scale the degree of similarity of the mobility pattern which includes deviation of the intermediate node’s speed direction from that of the destination, and distance from the intermediate node to the destination. A message is forwarded to the next hop node with the largest similarity degree.

Buffer scheduling policy in DTN routing protocols

Delay Tolerant Networks are with intermittent connectivity, high latency and without end-to-end paths from the source to the destination. Message delivering in these networks is characterized by store-and-forward. So the sequence of message delivering and discarding is important to the performance of routing algorithms. However, there is little effort concentrating on buffer management optimization, which can effectively utilize the limited resource. In this paper, we employ optimal theory to propose a novel buffer management strategy named Optimal Buffer Management (OBM) to optimize the sequence of message forwarding and message discarding. In OBM strategy, globe optimization considering delivery ratio, delay, and overhead is adopt to improve the overall performance of routing algorithms. The simulation results also show that the OBM policy is much better than the existing ones in delivery ratio, delay, and overhead.

A Novel Single Copy Replication Routing Strategy for Delay Tolerant Networks

Currently, there are two main kinds of routing strategies in delay tolerant networks, including single-copy forward routing and multi-copies replication routing. In this paper, we present a novel routing scheme called Single-Copy Replication routing (SCR) which has higher delivery ratio than single-copy forward scheme as well as lower overhead than multicopies replication strategy. SCR only duplicates bundles to the node selected as next hop. Based on the idea of SCR routing strategy, Time Interval Protocol (TIP) is designed to select the next hop according to the time interval between the sender and the destination. Also, we build mathematic model Dominos to analyze the properties of SCR scheme. In the end, the ONE simulator is applied to verify SCR. The theoretical analysis and simulation results both demonstrate that SCR can get higher delivery ratio and lower overhead comparing with other classic routing algorithms.

The Distributed Spanning-Tree Based Service Discovery in Grid Environment

With the dynamic and heterogeneous characteristics of resources in grid environment, efficient service discovery becomes a challenging issue. In this paper, we propose a service discovery approach based on the distributed spanning-tree architecture. According to this architecture, we arrange the grid resources in such a hierarchical way as IS layer, institution layer, organization layer and domain layer in turn. Any member at the same layer is equivalent. Only IS layer and Institution layer have real resource entities while other layers over them are overlays to cluster the resource information onto their own delegation nodes (DN). These DNs have the resource index and can locate the query route. Hence, query message traffic is sharply decreased depending on the DNs locating of query routes. Every Institution in grid owns a global unique ID according to the coding mechanism. Based on this, the parallel searching process can be synchronously implemented among peer nodes, which improves the efficiency dramatically. Caching and collaboration units techniques are adopted to increase the efficiency of service discovery. The DN can be replaced by the new elected one when it fails which ensures the survivability of the system. Performance evaluation shows that our approach achieves a good efficiency, scalability, survivability and adaptability

A Novel Time Interval Routing Protocol in Delay Tolerant Networks

Delay-tolerant networks (DTNs) have the capacity to interconnect the devices in roomy regions unlike today’s networks. In this paper, we propose a routing scheme named Time Interval Routing (TIR) to increase deliver ratio and decrease deliver latency in DTNs. TIR, only using limited and local information as well as simple algorithm, chooses the node in transmission range with the least time interval among neighbors as the next hop. Due to the limited time duration of connections, in this paper, we propose Bundle-Transfer strategies (BTS) mechanisms to ensure that TIR sends the bundle to the destination with the highest possibility. Through comprehensive simulation, TIR displays its advantage over other classic routing schemes in DTNs.

An Incentive Congestion Control Strategy for DTNs with Mal-behaving Nodes

Delay Tolerant Networks (DTNs) are characterized by intermittent connectivity, high delay and without end-to-end path from the source to the destination. The fundamental way of message routing in DTNs is store-and-forward. Thus, the mal-behaving nodes in the networks, which are uncooperative to relay messages, will significantly undermine the routing performance. In this paper, we propose a novel congestion control solution for DTNs with mal-behaving nodes named Incentive Multi-paths Routing with Alternative Storage (IMRASFC). IMRASFC firstly initializes multi-paths routing to mitigate link load, then it establishes rules to bestir mal-behaving nodes to store and forward messages, at last it selects alternative nodes with available storage to store data which would otherwise be dropped given buffer exhaustion. Simulation results show that IMRASFC improves the performance of routing algorithms significantly, especially when the number of copies in the network is large.

A Scalable Evenly Distributed Replica Location Mechanism

In order to provide better load balancing capability and scalability, a replica location mechanism is designed for data grids based on distributed hash structure which equitably assigns node identifiers. The system is peer-to-peer and hierarchical. It collects the location information of the distributed nodes and assigns optimal identifiers for new node to make the node distribution more even. Global indexes of the system are organized based on the distributed hash structure. Localization of the virtual organization can optimize queries on local indexes. Theoretical analysis and simulations show that it provides good performance, good scalability and load balancing capability for replica locations in data grids.

Grid Service Three-step Matching Model Based on Ontology

Facing a vast number of grid services in grid environment, how to find the best and satisfying services is a challenging issue. In this paper, we first propose a template to describe the grid service fully and extend the WSDL specification. In order to match the best satisfying services we present the accuracy degree function of service matching based on ontology to match the service in three-step. Next we apply this services matching solution to the actual model and make services type and QoS matching on different layers. Experiments prove that this grid services matching model has better matching accuracy degree and high efficiency over other models