1. INTRODUCTION. Saravanan.A 1 and Dr.Sunitha Abburu 2

Transcription

1 365 Computing Conditional Intermeeting Time in Conditional Shortest Path Routing Saravanan.A 1 and Dr.Sunitha Abburu 2 1 Adhiyamaan College of Engineering, Department of Computer Application, Hosur 2 Professor and Director, Adhiyamaan College of Engineering, Department of Computer Application, Hosur ABSTRACT Delay tolerant networks are characterized by the sporadic connectivity between their nodes. The Sporadic connectivity leads to lack of stable end-to-end paths from source to destination. Since the future node connections are mostly unknown in these networks, opportunistic forwarding is used to deliver messages. In Intermeeting time, to making effective forwarding decisions using only the network characteristics (i.e. average intermeeting time between nodes) but contact history extracted from previous node is a challenging problem. In this paper, we introduce a new method called conditional intermeeting time, which computes the average intermeeting time between two nodes relative to a meeting with a third node using only the local knowledge of the past contacts. Advantage of using conditional intermeeting time is time efficiency. Experimental results are discussed at the end. Finally conclusion and future work discussed at the end. Keywords: Intermeeting Time, Conditional Intermeeting Time, Shortest Path Routing, Roller net Traces, Cambridge traces. 1. INTRODUCTION Delay-tolerant networking (DTN) is an approach to computer network architecture that seeks to address the technical issues in heterogeneous networks that may lack continuous network connectivity. Routing in delay tolerant networks (DTN) is a challenging problem because at any given time instance, the probability that there is an end-to-end path from a source to a destination is low. Since the routing algorithms for conventional networks assume that the links between nodes are stable most of the time and do not fail frequently, they do not generally work in DTN s. Therefore, the routing problem is still an active research area in DTN s. In this paper introducing concept which is conditional intermeeting time. We redefine the intermeeting time concept between nodes and introduce a new link metric called conditional intermeeting time. The intermeeting time between two nodes given that one of the nodes has previously met a certain other node.so intermeeting time is not a time efficiency. 2. CURRENT RESEARCH In Delay Tolerant networks, the probability that there is an end to end path from a source to destination is low. Since the routing algorithms for conventional networks assume that the links between nodes are stable most of the time and do not fail frequently, they do not generally work in DTN s. So that, the routing problem is still an active Research area in DTN s [1]. Routing algorithms in DTN s make use of a method called store-carry-and-forward. If a node receives a message from one of its contacts, the node stores the message in its buffer and carries the message until it encounters another node which is at least as useful (in terms of the delivery) as itself. Then the message is forwarded to another node. Based on this example, several routing algorithms with different objectives (high delivery rate etc.) and different routing techniques have been proposed [2][3].. In Intermeeting time, ignored some information readily available at transfer decisions. When two nodes (e.g., A and B) communicate, the message forwarding decision is made according to a delivery metric (encounter frequency, time elapsed since last encounter [4] [5], social similarity etc.) of these two nodes with the destination node (D) of the message. However, all these metrics depend on the separate meeting histories of nodes A and B with destination node D1. Nodes A and B do not consider their transition with each other while computing their

2 366 delivery metrics with D. More formally, if X is the random variable representing the intermeeting time between two nodes, the probability is P(X > s + t X > t) _= P(X > s) for s, t > From above formula, the residual time until the next transition of two nodes can be predicted well if the node knows that it has not communicate the other node for t time units [6]. Two common metrics used to define the link costs are minimum expected delay[7] and minimum estimated expected delay[8]. The two metrics are compute the expected waiting time plus the transmission delay between each pair of nodes. 3. COMPUTING THE CONDITIONAL INTERMEETING TIME 3.1 INTERMEETING TIME In intermeeting time, at the transfer decisions unobserved some information readily available. When two nodes (e.g., A and B) communicate, the message forwarding decision is made according to a delivery metric (encounter frequency, time elapsed since last encounter, social similarity etc.) of these two nodes with the destination node (D) of the message. However, all these measures depend on the separate communication histories of nodes A and B with destination node D1. Nodes A and B do not consider their communication with each other while computing their delivery metrics with D. Recent analysis on real mobility traces comprise verified that models assuming the exponential distribution of intermeeting times between pairs of nodes do not match real data well. Instead up to 99% of intermeeting times in many datasets is log-normal distribution if X is the random variable representing the intermeeting time between two nodes, the probability is, P(X > s + t X > t) _= P(X > s) for s, t > From above formula, the remaining time until the next meeting of two nodes can be predicted well if the node knows that it has not communicate the other node for t time units.to take advantage, we propose a new metric called conditional intermeeting time. 3.2 CONDITIONAL INTERMEETING TIME The measures the intermeeting time between two nodes relative to a meeting with a third node using only the local knowledge of the past contacts. In a DTN, each node can compute the average of its standard and conditional intermeeting times with other nodes using its contact history. Such measure is particularly beneficial if the nodes move in a cyclic so-called MobiSpace in which if two nodes contact frequently at particular time in previous cycles, they will probably be in contact around the same time in the next cycle. The common motion cycle is 12 time units, so the discrete probabilistic contact between A and B happen in every 12 time units (1, 13, 25,) and between B and C in every 6 time units (2, 8, 14,). The average intermeeting time between nodes B and C indicates that node B can forward its message to node C in 6 time units. However, the conditional intermeeting time of B with C relative to prior meeting of node A indicates that the message can be forwarded to node C within one time unit. 3.3 CONDITIONAL SHORTEST PATH ROUTING Routing decisions can be made at three different points in an SP based routing: i) at source, ii) at each hop, and iii) at each contact. In the source routing, SP of the message is decided at the source node and the message follows that path. In the per-hop routing, when a message arrives at an intermediate node, the node determines the next hop for the message towards the destination and the message waits for that node. Finally, in the per-contact routing, the routing table is recomputed at each contact with other nodes and the forwarding decision is made accordingly. We define the CSP from a node n0 to a node nd as follows: CSP (n0, nd) = {n0, n1... nd 1, nd _n0 (n1 t) +d 1_i=1τni (Ni+1 ni 1) is minimized.} Here, t represents the time that has passed since the last meeting of node n0 with n1 and _n0 (n1 t) is the expected residual time for node n0 to meet with node n1 given that they have not met in the last t time units. _n0 (n1 t) can be computed as in with parameters of distribution representing the intermeeting time between n0 and nd. τa(b): Average time between two consecutive meetings of nodes A and B. Obviously when the node connections are bidirectional, τa(b) = τb(a). τa (B C): Average time it takes for node A to meet node B after it meets node C. Note that, τa (B C) and τb(a C) are not necessarily equal.

3 367 S: N N matrix where S(i, j) shows the sum of all samples of conditional intermeeting times with node j relative to the meeting with node i. Here, N is the neighbor count of current node. Data Sent C: N N matrix where C(i, j) shows the total number of conditional intermeeting time samples with node j relative to its meeting with node i. βi: Total meeting count with node i. Algorihm1 is used to find and update the conditional intermeeting time between two nodes. Algorithm 1 update (node m, time t) if m is seen first time then firsttimeat[m] t else increment βm by 1 lasttimeat[m] t end if for each neighbor j N and j _= m do start a timer tmj for each neighbor j N and j _= m do for each timer tjm running do S[j][m] += time on tjm increment C[j][m] by 1 delete all timers tjm for each neighbor i N do for each neighbor j N and j _= i do if S[j][i] _= 0 then τs(i j) S[j][i] / C[j][i] end if τs(i) (lasttimeat[i] firsttimeat[i] ) / βi In above algorithm, each node first adds up times expired between repeating meetings of one neighbor and the meeting of another neighbor. Then it divides this total by the number of times it has communicate the first neighbor prior to communicate the second one. 4. PERFORMANCE EVALUTION In Intermeeting time, When two nodes communicate, the message forwarding decision is depends on the delivery metric(frequency, Time elapsed since last encounter, Social Similarity) of these nodes with the designation node message. Client User Get Source Ipaddress and Port No Find the shortest path between source and designation Computing Conditional intermeeting time between nodes Generating Reports Server Data Received Fig.1 System Architecture of conditional time All the delivery metric depend on the separate transfer histories of nodes A and B with designation node D. Nodes A and B do not consider their communication with each other util computing their delivery metrics with D. All the delivery metric depend on the separate transfer histories of nodes A and B with designation node D. Nodes A and B do not consider their communication with each other util computing their delivery metrics with D. By using Conditional Intermeeting Time, we refine the intermeeting time. In Conditional Intermeeting

4 368 Time, computes the average intermeeting time between two nodes relative to a meeting with third node using only the knowledge of the past contacts. By redefine the intermeeting time, all nodes having histories of contacts. In Conditional Intermeeting time, it is very time efficiency while transferring message from source to designation. 3 : Socket ws=(socket)m_ws[clientnum-1] 4 : ws.send(bydata) 4.2 SIMULATION Prog.1 shows to send message to client. Initially Source wil get Ipaddress and Port Number. Then made a decision for shortest path and computing condional intermeeting time. 4.1 ALGORITHM TO SEND MESSAGE TO CLIENT Table 1 : Algorithm1 GetIpAddress 2 : String strhostname=dns.gethostname() 3 : IPHostEntry iph =dns.gethostname(strhostname) 4 : For each IPAdress ipaddress in iph.addresslist 5 : IPStr = ipaddress 6 : endfor 7 : Print IPStr 8 : Stop From the simulations, we compare the proposed Conditional Shortest Path Routing (CSPR) algorithm with standard Shortest Path Routing (SPR). Moreover, in our results we also show the performance of upper and lower performance limits with Epidemic Routing and Direct Delivery. For a simulation run, we generated 50 messages from a random source node to a random destination node at each t seconds. In Roller Net, since the duration of experiment is short, we set t = 1s, but for Cambridge data set, we set t = 1 min. We assume that the nodes have enough buffer space to store every message they receive, the bandwidth is high and the contact durations of nodes are long enough to allow the exchange of all messages between nodes. The Following figure shows the Roller net Trace and Cambridge traces. Table 2 : Algorithm2 To List out Computers in Network 2 : Networkbrowser nb = new NetworkBrowser() 3 : ArrayList arr= nb.getnetworkcomputers(server.sv_ty _Server) 4 : Print ServerType.ar.count 5 : for each string name in arr 6 : Print name 7 : Endfor 8 : Stop Table 3 : Algorithm3 To send message to Client Time(min) Fig. 2. Message delivery ratio vs. time in RollerNet traces. 2 : byte[] bydata = msg

5 369 mobile networks: The single-copy case, IEEE/ACM Transactions on Networking. [3] T.Spyropoulos,K. Psounis,C. S. Raghavendra, Efficient routing in intermittently connected mobile networks: The multi-copy case, IEEE/ACM Transactions on Networking, [4]. H. Dubois-Ferriere, M. Grossglauser, and M.Vetterli, Age Matters: Efficient Route Discovery in Mobile Ad Hoc Networks Using Encounter Ages, Time (min) Fig. 3. Message delivery ratio vs. time in Cambridge traces. 5. CONCLUSION AND FUTURE WORK In Delay Tolerant Networks are irregular connectivity between their nodes. In this paper, we introduced a new metric called conditional Intermeeting time. By the Result of recent studies showing Delay tolerant networks nodes intermeeting time are not memory less. While forwarding message extracted from contact history is challenging problem. Then, we looked at the effects of this metric on shortest path based routing in DTN s. For this purpose, we updated the shortest path based routing algorithms using conditional intermeeting times and proposed to route the messages over conditional shortest paths. Finally in Experimental result shown the Simulation. [5] T. Spyropoulos, K. Psounis, and C. Raghavendra, Spray and Focus: Efficient Mobility-Assisted Routing for Heterogeneous and Correlated Mobility [6] S. Srinivasa and S. Krishnamurthy, CREST: An Opportunistic Forwarding Protocol Based on Conditional Residual Time, in Proceedings of IEEE SECON, [7] S. Jain, K. Fall, and R. Patra, Routing in a delay tolerant network, in Proceedings of ACM SIGCOMM, Aug [8] E. P. C. Jones, L. Li, and P. A. S. Ward, Practical routing in delay tolerant networks, Networking (WDTN), In future work, we will look at the performance of the proposed algorithm in different data sets to see the effect of conditional intermeeting time in different environments. Moreover, we will consider extending our CSPR algorithm by using more information from the contact history while deciding conditional intermeeting times. 6.REFERENCES [1]. Delay tolerant networking research group, [2] T. Spyropoulos, K. Psounis,C. SRaghavendra, Efficient routing in intermittently connected