This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
In this paper, author proposes a new message scheduling framework for epidemic and two-hop forwarding routing in DTNs, such that the forwarding/dropping decision can be made at a node during each contact for either optimal message delivery ratio or message delivery delay.
The proposed framework incorporates a suite of novel mechanisms for network state estimation and utility derivation, such that a node can obtain the priority for dropping each message in case of full buffer. Using simulations based on two mobility models; a synthetic (RandomWayPoint) and a real trace- model (ZebraNet), the simulation results show that the proposed buffer management policies, named GHP, can signi¬cantly improve the routing performance in terms of the performance metrics of interest under limited network information.
A message issued at a node (termed sourced messages) has the highest priority at the node.
Negligible storage overhead without considering that each node could be with a limited buffer space.
Every node is aware of all messages that has encountered during contacts with other nodes
In this framework the computation is very high and time consuming. So for better improvement the computation cost should reduce.
Title: An Efficient Prediction-Based Routing in Disruption-Tolerant Networks 
Quan Yuan, Ionut Cardei and Jie Wu
In this paper, author proposes predict and relay, an efficient routing algorithm for DTNs, where nodes determine the probability distribution of future contact times and choose a proper next-hop in order to improve the end-to-end delivery probability.
The authors propose the Predict and Relay scheme, an efficient routing scheme in DTNs. They introduce a time- homogeneous semi-Markov process model to predict the future contacts of two specified nodes at a specified time. With this model, a node can select a proper neighbor as the next hop to forward the message. This paper defines three different prediction functions to assist in choosing the proper neighbor for message delivery. Simulation results show that their approach raises the delivery ratio, as well as reduces the delivery latency, compared to other traditional DTN routing protocols.
During a contact, nodes can successfully transfer all messages needed to be forwarded. This requires reliable transport and high bandwidth or long enough contacts.
Each landmark has a unique landmark id in the network, and nodes are aware of which landmark they are located at any time.
The whole network is composed of the neighborhoods of landmarks, which means a node is always associated to a certain landmark in the network. Nodes spend zero time on the transition between landmarks.
Improve the protocol to work properly with low bandwidth network.
Title: Message Drop and Scheduling in DTNs- Theory and Practice 
Amir Krifa, Chadi Barakat and Thrasyvoulos Spyropoulos
The authors propose the joint scheduling and drop policy that can optimize different performance metrics, such as average delay and delivery probability. They first use the theory of encounter-based message dissemination to derive the optimal policy based on global knowledge about the network. Then, introduce a method that estimates all necessary parameters using locally collected statistics. Based on this, we derive a distributed scheduling and drop policy that can approximate the performance of the optimal policy in practice.
The authors have proposed an optimal joint scheduling and buffer management policy and introduced an approximation scheme for the required global knowledge of the optimal algorithm. Using simulations based on a synthetic and real mobility traces they showed that their policy based on statistical learning successfully approximates the performance of the optimal algorithm. Both policies plugged into the Epidemic routing protocol outperform current state-of-the-art protocols like RAPID with respect to both delivery rate and delivery delay, in all considered scenarios.
Each message is assumed to be routed using a replication-based scheme.
A class of mobility models that has the following properties.
Meeting times are exponentially distributed or have at least an exponential tail.
Nodes move independently of each other.
Mobility is homogeneous, that is, all node pairs have the same meeting rate.
The history of the network is necessary to perform routing.
Title: On Exploiting Transient Social Contact Patterns for Data Forwarding in Delay-Tolerant Networks 
Wei Gao, Guohong Cao, Tom La Porta and Jiawei Han
In this paper, author proposes a novel approach to improve the performance of data forwarding with a short time constraint in DTNs by exploiting the transient social contact patterns. These patterns represent the transient characteristics of contact distribution, network connectivity and social community structure in DTNs, and they provide analytical formulations on these patterns based on experimental studies of realistic DTN traces. Then propose appropriate forwarding metrics based on these patterns to improve the effectiveness of data forwarding. When applied to various data forwarding strategies, their proposed forwarding metrics achieve much better performance compared to existing schemes with similar forwarding cost.
In this paper they propose effective forwarding metrics to improve the performance of data forwarding in DTNs, by exploiting the transient social contact patterns. They formulate these patterns based on experimental observations from realistic DTN traces, and exploit these patterns for more accurate prediction on the node contact capability. Through extensive trace-driven experiments, they show that our approach significantly improves the data delivery ratio, while keeping similar forwarding cost with existing schemes.
Node contacts are symmetric; i.e., node j contacts i whenever i contacts j, and the network contact graph is therefore undirected.
The data size is small, so that it can be carried by any node and can be completely transmitted during a contact.
The delivery ratio of the present scheme improves but the forwarding cost is same. So for improvement, we can reduce the forwarding cost.
Title: Efficient Probabilistic Routing in Delay Tolerant Networks 
El Mastapha Sammou
This paper focuses on a routing approach which is to improve the PRoPHET routing protocol based on two routing metrics: History contact between the mobile nodes and the frequency of visiting different zones of the network.
To improve routing in delay tolerant networks, we proposed a routing approach based on two routing metrics, the first routing metric uses the history of contact between the nodes and the second metric uses the frequency of visits to areas network.
The routing metrics are used to take decision about network density and based on it the message is sent to the node that has a high probability of delivery and has a high probability of remission.
The prophet routing protocol is modified and its performance is improved.
Title: Social Network Analysis for Routing in Disconnected Delay-Tolerant MANETs 
Elizabeth Daly and Mads Haahr
Social properties of node as some relay nodes are identified based on their centrality characteristics. Due to the complexity of the centrality metrics in populated networks the concept of ego networks is exploited where locally available information is considered.
SimBet Routing, a novel algorithm for routing in disconnected delay-tolerant MANETs based on social network analysis techniques. They have simulate using real trace data that SimBet Routing achieves delivery performance comparable to Epidemic Routing and succeeds in finding a route where PRoPHET fails due to the low connectivity of the sending and receiving nodes.
Focus on find the relay nodes which give higher probability for delivering message.
For selection of relay node, they used node centrality and similarity value for ego centric network.
In this paper, we can use closeness centrality to check network density by checking distance between nodes, which is further useful to make decision about to forward or spray the message copy into network.
Title: Performance Evaluation of Delay Tolerant Routing Protocol by Variation in Buffer Size 
Ramchandra S. MangruIkar and Dr. Mohammad Atigue
In this paper, author obtains various results by varying the storage capacity of nodes to store and forward message. With the help of results, the authors proved that the performance of MaxProp routing protocol is outstanding than all other routing protocols under consideration.
This paper mostly try to elaborate the result of each routing protocol under consideration by varying the buffer size i.e. the message storing capacity of node. The performance of various routing protocols viz. Direct Delivery, First Contact, Epidemic and MaxProp is evaluate by considering map based model by keeping rest of the parameters as constant. They concluded that even the buffer size i.e. the message storage capacity of the nodes get varied, the performance of Maxprop is significantly increased as compared to other routing protocols under study.
As the buffer sizes increase the messages drop by node decrease. This leads to increase the performance in terms of delivery ratio of the network.
Title: A Dynamic Congestion Control Based Routing for Delay-Tolerant Networks 
Shou-Chih Lo and Chuan-Lung Lu
In this paper, they propose a suite of strategies to defend a DTN routing protocol against different congestion conditions. Their proposed mechanism can be applied to any routing protocols that are based on the concept of quota replication. Experimental results show the superiority of this kind of congestion control over existing types presently in use.
Quota-replication routing to generate many copies of messages to increase their probability of reaching their destinations successfully, and also control the amount of copies generated when the network becomes congested. A series of strategies are adopted to alleviate congestion of different degrees.
The values of many parameters are obtained by statistical analysis of the results.
For better improvement, we can compute threshold values, number of initial copies of message.
Title: Adaptive routing in delay tolerant network 
Ravindra Patel, Hemal Shah
In this thesis author has integrated single copy and multi copy routing based on the node average number of contact occurred in the time interval. Each node computes its average contact with neighbor nodes for given time interval. Than computed average for all nodes and take the routing decision.
The proposed routing technique outperformed in terms of overhead ratio, drop messages and delivery ratio. It is closely monitored for ratio of forwarding and replication technique called by routing engine turns out around 1:4. This ensures that routing engine calls the techniques adaptively and manages network utilization.
The forwarding technique uses knowledge of encounter based routing.
The replication strategy is based on multi period spray and wait.
Authors have attempted to figure out node density and thus sparseness and denseness of challenged network. Also, threshold was selected as baseline for differentiation but there is scope for realistic measurement and technique in DTN.
Title: Characterization & Evaluation of Temporal Metrics and Information Diffusion in Delay Tolerant Network 
Vikrant Patel, Hemal Shah
Delay tolerant network refers to network where connectivity is episodic. Due to such intermittent connectivity traditional routing protocol fails. In order to efficiently route the information in such an environment researchers have proposed various routing protocols: forwarding, replication, control replication etc. This protocols uses local information: Neighbourhood, past encounter, contact duration, node mobility etc. to determine next hop and or destination. Node mobility improves the connectivity of network and can be visualize such scenario of moving nodes as time varying graph temporal graph considering node as vertices and contact opportunity to other node as an edge. The work proposed in report discusses about metrics: temporal distance, temporal diameter and temporal centrality and an efficient routing protocol using ONE Simulator.
Node mobility plays an important role for information diffusion in extreme environment. Dynamic properties of connection are first of all captured and analyzed using various time varying matrices. This has enabled in characterizing the properties: time order, frequency, contact duration, interconnect time etc. It has been further used in routing protocol for efficient delivery ratio, less overhead and lower message dropped. The objectives are achieved during the simulations in presence of real trace data in comparison with epidemic and prophet routing. New FR routing has outperformed.
Author has done extensive work with real trace and its characterization. There is scope analyze the synthetic data and benchmark routing protocol with such data sets. Also, in realistic scenarios network resources are scared and limiting it may give close to realistic picture. Also, security issues, real test bed need further investigation.
2.3 Outcome of Literature Survey
It is observed that within the DTN routing, still there are scopes for:
Exploring social network theories for efficient information diffusion
Optimal copies of messages