VIEWS: 61 PAGES: 8 CATEGORY: Emerging Technologies POSTED ON: 11/2/2010 Public Domain
(IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010 Multicast Routing and Wavelength Assignment for Capacity Improvement in Wavelength Division Multiplexing Networks N.Kaliammal G.Gurusamy Professor, Department of ECE, Prof/Dean/ EEE, FIE N.P.R college of Engineering and Technology, Bannari amman Institute of Technology, Dindugul, Tamil nadu Sathyamangalam,Tamil nadu. Email: kala_gowri@yahoo.co.in E-mail: hodeee@bitsathy.ac.in Tel: +91 9965557267 Tel: +91 9791301662 Abstract—In WDM network, the route decision and wavelength amount of data at high speeds by the users over large distance assignment of light-path connections are based mainly on the [2]. routing and wavelength assignment (RWA). The multicast routing and wavelength assignment (MC-RWA) problem is for For the future generation internet, WDM is considered as a maximizing the number of multicast groups admitted or for backbone which is the most talented technology. The data is minimizing the call blocking probability. In this paper, The routed through optical channels called light paths in WDM all design of multicast routing and wavelength assignment technique optical networks. The light path establishment requires same for capacity improvement in wavelength division multiplexing wavelength and it should be used along the entire route of the (WDM) networks is proposed. In this technique, the incoming light path without wavelength conversion. This is commonly traffic is sent from the multicast source to a set of intermediate considered to the wavelength continuity constraint [3]. junction nodes and then, from the junction nodes to the final destinations. The traffic is distributed to the junction nodes in B. Multicasting in WDM Networks predetermined proportions that depend on the capacities of A network technology which is used for the delivery of intermediate nodes. Then, paths from source node to each of the information to a group of destinations is called as multicast destination nodes and the potential paths are divided into addressing. This simultaneously uses the most efficient fragments by the junction nodes and these junction nodes have strategy to deliver the message over each link of the network the wavelength conversion capability. By using the concept of only once. Moreover, it creates the copies only when the links fragmentation and grouping, the proposed scheme can be to the multiple destinations split [4]. generally applied for the wavelength assignment of multicast in WDM network. By simulation results, it is proved that ther In recent years, multicast communication is turning out to proposed technique achieves higher throughput and bandwidth be vital due to its efficient resources usage and the increasing utilization with reduced delay. popularity of the point-to-multipoint multimedia applications. Usually, a source and a set of destinations are included in a I. INTRODUCTION multicast session. In conventional data networks, in order to A. Wavelength-Division-Multiplexing (WDM) Networks allow a multicast session, a multicast tree which is rooted at the source is constructed with branches spanning all the The need for on-demand provisioning of wavelength destinations [5]. routed channels with service differentiated offerings within the transport layer has become more essential due to the recent Recently, multicast routing in optical networks has been emergence of high bit rate IP network applications. Diverse researched which is related to the design of multicast-capable optical transport network architectures have been proposed in optical switches. For multicast in WDM networks, the concept order to achieve the above requirements. This approach is of light-trees was introduced. Reducing the distance of determined by the fundamental advances in the wavelength network-wide hop and the total number of transceivers used in division multiplexing (WDM) technologies. Due to the the network are the objective of setting up the light trees. availability of ultra long-reach transport and all-optical Nowadays, there are several network applications which switching, the deployment of all-optical networks has been require the support of QoS multicast such as multimedia made possible [1]. conferencing systems, video on demand systems, real-time control systems, etc. [6]. The concurrent transmission of multiple streams of data with the assistance of special properties of fiber optics is C. Routing and Wavelength in WDM called as wavelength division multiplexing (WDM). The In WDM network, the route decision and wavelength WDM network provides the capability of transferring huge assignment of light-path connections are based mainly on the routing and wavelength assignment (RWA). This is the most 175 http://sites.google.com/site/ijcsis/ ISSN 1947-5500 (IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010 important and basic issue in resource management. For are divided into fragments by the junction nodes and these maximizing the number of multicast groups admitted or for junction nodes have the wavelength conversion capability. By minimizing the call blocking probability with certain number using the concept of fragmentation and grouping, the proposed of wavelengths, the multicast routing and wavelength scheme can be generally applied for the wavelength assignment (MC-RWA) problem is studied. [7]. assignment of multicast in WDM network. The Least Influence Group (LIG) approach is used to provide the The problem of finding a multicast tree and allocating wavelength selection. available wavelength for each link of the tree is known as the Multicast Routing and Wavelength Assignment (MC-RWA) II. RELATED WORK problem, which plays a key role in supporting multicasting over WDM networks [8]. The problems involving in the Jingyi He et al [7] have proposed for the first time a routing and wavelength assignment in WDM are as follows: formulation of the MC-RWA problem with the objective to maximize the number of multicast groups admitted, or • Improper wavelength assignment, especially for the equivalently, to minimize the call (or session) blocking multicast connection, will cause wavelength probability given a certain number of wavelengths.. The blocking, whereas the network resources may be still formulation is a nonlinear integer program, which in general is underutilized. complex to solve so a near-optimal solution of the problem is • The wavelength continuity constraint, i.e., that links proposed using a two-step approach based on linear from source to destination shall use the same programming. The drawback in this work is that the focus is wavelength to convey data in the same lightpath, on minimizing the user blocking probability instead of the session blocking probability for single-source applications. always makes the wavelength assignment inflexible and causes wavelength blocking. Anping Wang et al [8] have proposed a new multicast • The available wavelength can be maximized by the wavelength assignment algorithm called NGWA with wavelength converter but this type of device is much complexity of O(N), where N is the number of nodes on a intricate and cost is also high when compared with multicast tree. The whole procedure of NGWA algorithm is the type of device which cannot perform the separated into two phases: the partial wavelength assignment conversion. phase and the complete wavelength assignment phase. The • The signal may also decay during the conversion. drawback of this work is that this method achieves only satisfactory performance in terms of the total number of Therefore, it is not possible to have all network nodes wavelength conversions and the average blocking probability be equipped with wavelength conversion capability. • The problem of the node architecture is that they Nina Skorin-Kapov [10] has addressed the problem of were designed without having into account power multicast routing and wavelength assignment (MC RWA) in efficiency, neither complexity of fabrication [9]. wavelength routed WDM optical networks. Multicast requests • The two sub-problems of the routing and wavelength are facilitated in WDM networks by setting up so-called light- assignment are the routing problem and the trees and assigning wavelengths to them. She has proposed a wavelength assignment problem, which can be either heuristic algorithm based on bin packing methods for the general MC RWA problem, which is NP-complete. These coupled or uncoupled. In the case of uncoupled algorithms can consider unicast, multicast and broadcast situation, initially a route or a tree is obtained which requests with or without QoS demands. Computational tests is then followed by the wavelength assignment where indicate that these algorithms are very efficient, particularly the trees must be kept unchanged and is called as the for dense networks. static RWA. In the coupled case, based on the state of the wavelength assignment, the routes are decided Fen Zhou et al [11] have proposed a routing and which is usually called as dynamic or adaptive RWA wavelength assignment for supporting multicast traffic is [7]. investigated in WDM mesh networks under sparse splitting In previous paper, a resource efficient multicast routing constrain. This problem is generally solved in two phases protocol is developed. In this protocol, the incoming traffic is respectively with the purpose of minimizing the number of sent from the multicast source to a set of intermediate junction wavelengths required. Alternative routing is first proposed to nodes and then, from the junction nodes to the final route each session by pre-computing a set of candidate light- destinations. The traffic is distributed to the junction nodes in forests. Then wavelength assignment is formulated as coloring predetermined proportions that depend on the capacities of problems by constructing a conflict graph. Potential heuristic intermediate nodes. Bandwidth required for these paths algorithms are proposed. The drawback of this work is that depends on the ingress–egress capacities, and the traffic split simulation should be done to assess the verification of the ratios. The traffic split ratio is determined by the arrival rate of proposed methods. ingress traffic and the capacity of intermediate junction nodes Yuan Cao et al [12] have proposed an efficient QoS- [13]. guaranteed Group Multicast RWA solutions, where the In this paper, a multicast routing and wavelength transmission delay from any source to any destination within a assignment technique in wavelength division multiplexing multicast group is within a given bound. They have formulated networks is designed. In this technique, paths from source the QoS-guaranteed GMC-RWA problem as an in-group node to each of the destination nodes and the potential paths traffic grooming and multicasting problem, where traffic 176 http://sites.google.com/site/ijcsis/ ISSN 1947-5500 (IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010 streams from members of the same group are groomed in an D. effective way before being delivered to their common 1.2 End if destinations, subject to the following optical layer constraints. 2. End for 3. If {D} ≠ Null, Then III. MULTICAST TREE FORMATION 3.1 Repeat from1. A. Basic Definitions 4. End if The node which cannot split the incoming message to the outgoing ports is called as Multicast Incapable (MI) nodes. B. Multicast Routing But it can utilize a small amount of optical power from the wavelength channel while forwarding it to only one output A collection of point to multiple point paths from the link. source node to each destination is considered as a multicast tree. Choosing a suitable wavelength for its downlink is The nodes which are capable of splitting the incoming flexible for a path in the WDM network which has sparse message to all the outgoing ports are called as Multicast junction nodes. The main objective is to reduce the affected Capable (MC) nodes. capacity. This can be done by selecting a suitable wavelength for the downlink of the junction nodes which reduces the The set which includes the multicast capable nodes (MC influence on the potential request paths across it. The junction node) and the leaf multicast incapable nodes (leaf MI nodes) is node is considered as an end point of a wavelength within a called as MC_SET. fragment. According to the position of converters within the The set which includes only the non-leaf multicast path, the path can be divided into uni-wavelength fragments. incapable nodes, which are not able to connect a new As a result, paths from source node to each of the destination destination to the multicast tree, is called as MI_SET. nodes and the potential paths are divided into fragments by the junction nodes and these junction nodes have the wavelength The set D includes the unvisited multicast destinations conversion capability. which are not yet joined to the multicast tree. A network G= (N, E) with node set N and (directed) edge A constraint path between a node u and a tree T is a E set is taken,where each node in the network can be a source shortest path from node u to a node v in the MC_SET for T, or destination of traffic. The nodes in N are {N1, N2…Nn}. and this shortest path should not traverse any node in MI_SET for T. And the constraint path with the minimum length is S called the Shortest Constraint Path (SCP). R1 0 For one nearest destination d, MC_SET may have different SCPs to the sub-tree. Let X and Y are the nodes for the sub- tree in MC_SET. Without involving any node in MI_SET for 1 2 the sub-tree, both the shortest paths from X and Y to the R2 nearest destination d have the shortest length among all the nodes in MC_SET. Here, the nodes like X and Y are named as Junction junction nodes in the sub-tree. Node 3 4 Member only Algorithm T = {s} MI_SET = Null Figure 1. Multicast Routing Process MC_SET = {s} D = {D1, D2….Dn} The above diagram (Fig. 1) shows the routing process. A 1. For each Di, where i = 1, 2….n predetermined fraction of the traffic entering the network at any node is distributed to every junction node. The 1.1 If dist (Di, N) = min, where N ∈ corresponding route from the source to the junction node can MC_SET, then be denoted as R1. Then each junction node receives the traffic 1.1.1 Add Di to T to be transmitted for different destinations and it routes to their 1.1.2 Find SCP (Di, T) ∉ M, where respective destinations. The corresponding route from the M ∈ MI_SET junction node to the destination can be denoted as R2. 1.1.3 Add SCP (Di, T) to T 1.1.4 Add all the MC nodes to Let Ii and Ei, be the constraints on the total amount of MC_SET traffic at ingress and egress nodes of the network, respectively. 1.1.5 Add all the leaf MI nodes to The traffic along R1 and R2 must be routed along MC_SET bandwidth-guaranteed paths. Bandwidth required for these 1.1.6 Add all the non-leaf MI nodes paths depends on the ingress–egress capacities, and the traffic to MI_SET split ratios. The traffic split ratio (δ) is determined by the 1.1.7 Delete the non - leaf MI node arrival rate of ingress traffic and the capacity of intermediate from MC_SET junction nodes. 1.1.8 Delete the destination di from 177 http://sites.google.com/site/ijcsis/ ISSN 1947-5500 (IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010 where Pij is the jth fragment of the potential path i, Si is the The bandwidth requirement for the routing paths R1 and source node of the potential path i, and Di is the destination of R2 is derived. Consider a node i with maximum incoming the potential path i. Basically, each fragment can be treated as traffic Ii. Node i sends δjIi amount of this traffic to node j a reassignment domain of wavelength. during R1 routing for each jєN and thus the traffic demand is Fragments of a path are mutually independent from the δjIi. Now, node i has received δiIk traffic from any other node wavelength assignment point of view and may be with k. Out of this, the traffic destined for node j is δirkj since all different fragment capacities. The actual capacity of a path is traffic is initially split without regard to the final destination. basically determined by its fragment(s) with the least capacity. The traffic that needs to be routed from node i to node j at R2 The fragment(s) with the least capacity of a path is named the routing is given below: critical fragment of that path. Let CPi and CPPi be the path ∑δ r capacity (the least fragment capacity) of the path i of the i kj ≤ δi E j . multicast tree, and the path capacity of the potential path i, k∈N respectively,then Thus, the traffic demands from node i to node j at the end (3) of R2 routing is λiEj. CPi = min SCPR i j 1≤ j ≤ ri + 1 Hence, the maximum demand from node i to node j as a and result of R1 and R2 routing is δjIi + δiEj. CPpi = min SCPpi j Let M = [mij] = [δjIi + δiEj] be the fixed matrix which can (4) handle the traffic variation. It depends only on aggregate 1≤ j ≤ ri +1 ingress-egress capacities and the traffic split ratios δ1, δ2 …. δn, Capacity of the path cannot be decreased by decreasing the Thus the routing scheme is unaware to the changes in traffic capacity in a fragment whose capacity is larger than the distribution. critical fragment of that path. A path may have more than one critical fragment. Let Fi = {fi1, fi2 …} be the set of the critical IV. MULTICAST WAVELENGTH ASSIGNMENT fragments in the potential path i. Then Fi can be used to A. Grouping the Paths indicate whether the potential path is affected or not during the wavelength assignment of the multicast tree. So, the critical Assume the set Ri = {Ri1, Ri2 … Rij …} to represent all fragment of a potential path is the fragment traveled by the fragments of the path from source to the ith destination in the multicast tree. The impact on the potential path can be reduced multicast tree. Rij is the jth fragment of the path i. If AWRij is by considering the wavelength assignment of that fragment the set of available wavelengths of the jth fragment of path i, carefully. Fragments which come from multicast tree with then the number of wavelengths in AWRij is regarded as the common links into groups are coupled using the concept of capacity of this fragment. The capacity of the jth fragment of grouping. Within a group, all fragments have common the path i, SCPRij is obtained as wavelengths. As a result a group is composed of fragments ⎧OL ( S , J k ) =| AWR j |, whose links are overlapped. i i k = 1, j = 1 ⎪ j ⎪ k k −1 j G = {G1, G2… Gm… GY} (5) SCPR i = ⎨OL ( J i , J i ) =| AWRi |, 1 < k ≤ M i , j = k where G is the set of all groups in a multicast tree, Gm is ⎪ k j the set of all fragments in the mth group. ⎪OL ( J i , D ) =| AWRi |, ⎩ k = Mi, j = Mi +1 (1) The multicast tree with n destinations is treated as n unicast paths from source to each destination. Paths are where s is the source node of the multicast tree, Di is the ith fragmented with respect to junction nodes. Same group destination of the multicast tree, Jik is the kth wavelength fragments have more than one available wavelength in converter in the path i, and Mi + 1 is the number of fragments common. Let AWGm be the connection set of all fragments of path i if there are Mi junction nodes being traveled by the existing wavelengths in the mth group. The group capacity, path. The Overlap function OL(n1, n2) represents the size of CGm, is defined as the number of wavelengths in AWGm. If the intersection set of all available wavelengths for all links links of a fragment and the links in the mth group are from node n1 to n2. overlapped and no common available wavelength between them, this fragment will be considered as a new group. For the potential request paths, the set Pi = {pi1, pi2 …}is defined to indicate all fragments of the ith potential request B. Total Network Capacity Estimation path and the capacity of the jth fragment of the potential path i, The influence of network capacity is examined by SCPPij, can be stated as following checking whether the links of potential paths overlap with those of the multicast groups. If the overlap occurs at the ⎧OL ( S , J k ), k = 1, j = 1 ⎪ i critical fragments of the potential path and the assigned ⎪ wavelength is the one of the available wavelengths in that SCPPi j = ⎨OL ( J i k , J i k −1 ), 1 < k ≤ M i , j = k (2) critical fragment, the path capacity of the potential path will be ⎪ k affected. Let Cm(pi, λ) be the capacity of pi being influenced ⎪OL ( J i , D ), k = Mi, j = Mi +1 ⎩ when the wavelength λ is assigned in the mth group, and x be a 178 http://sites.google.com/site/ijcsis/ ISSN 1947-5500 (IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010 common link of the mth group and the critical fragment of the potential path i. Then TNC m ,λ = ∑C p b ∈P ' m ( pb , λ ) ⎧1 if (λ ∈ xw ) ∧ x ∈ LS m , Fi 3. Assign λ which TNCm, λ is minimum in group m Cm ( pi , λ ) = ⎨ (6) ⎩0 Otherwise V. SIMULATION RESULTS where LSm,Fi = LGm ∩ LFi, LGm is the set of all links in the A. Simulation Model and Parameters mth group, LFi is the set of all links in the critical fragments of the potential path i. and xw is the set of all available In this section, the performance of multicast routing and wavelengths on link x. wavelength assignment technique is simulated with an extensive simulation study based upon the ns-2 network The network capacity affected when λ is assigned for the simulator [14]. The Optical WDM network simulator (OWNs) mth group, TNCm,λ , can be obtained by the summation of the patch in ns2 is used to simulate a NSF network (Fig.1) of 14 influence of all potential paths as nodes. Various simulation parameters are given in table I. TNCm,λ = ∑ Cm ( pi , λ ) (7) pi ∈P The total network capacity (TNC) gets affected since each group should assign one wavelength, and it can be obtained by the summation as TNC = ∑ ∑ Cm ( pi , λm ) − q (8) All m pi ∈P Figure 1. NSF network of 14 nodes In the mth group, λm is the wavelength assigned and q is the affected capacity that is counted repeatedly. This can be TABLE I: SIMULATION PARAMETERS regained in the first term of (8). When the same wavelength is assigned to the groups it leads to repeated counts and also the Topology Mesh critical fragments of the path travels through the group. For Total no. of nodes 14 example, the available wavelengths of the critical fragment of Link Wavelength Number 8 potential path p1 are (λ1, λ2). G1 and G2 are the groups of the Link Delay 10ms multicast tree. If λ1 is assigned to G1 and G2 and if the critical Wavelength Conversion Factor 1 fragment of potential path p1 travels through G1 and G2, then, according to the first term of (8), the affected capacity of p1 is Wavelength Conversion Distance 8 calculated twice. In fact, the decreased capacity is only one. Wavelength Conversion Time 0.024 The other repeated count happens when the same or a different Link Utilization sample Interval 0.5 wavelength is assigned to the groups and more than one Traffic Arrival Rate 0.5 critical fragment of an individual path goes through these Traffic Holding Time 0.2 groups. Packet Size 200 C. Wavelength Assignment No. of Receivers 4 Max Requests Number 50 By using junction nodes the multicast tree is separated into Rate 2,4,6, 8 and 10 Mb groups, so the wavelength assignments for groups are independent of each other. The wavelength assigned in the Number of Traffic Sources 1,2,3,4 and 5 previous group has no effect on the wavelength assigned in the current group. The wavelength assigned for each group can be In this simulation, a dynamic traffic model is used, in easily selected since all of the available wavelengths for a which connection requests arrive at the network according to group have been collected in AWGm. an exponential process with an arrival rate r (call/seconds). The session holding time is exponentially distributed with The Least Influence Group (LIG) algorithm selects the mean holding time s (seconds). wavelengths for groups to maximize the network capacity. The idea behind LIG algorithm is that the wavelength having The connection requests are distributed randomly on all the the least effect on the potential paths is chosen for that group. network nodes. In all the simulation, the results of MRWA The affected network capacity in (7) examines the influence of with the previous paper “resource efficient multicast routing each wavelength assignment. The LIG algorithm is illustrated (REMR) protocol [13].” Is compared. below: B. Performance Metrics AWGm = {λ1, λ2, λ3….} In this simulation the blocking probability, end-to-end 1. Find all pb whose links overlap in the links of group delay and throughput is measured. m 2. For each λ ∈ AWGm 179 http://sites.google.com/site/ijcsis/ ISSN 1947-5500 (IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010 • Bandwidth Utilization: It is the ratio of bandwidth Rate Vs Utilization received into total available bandwidth for a traffic flow. • Average end-to-end delay: The end-to-end-delay is 0.08 averaged over all surviving data packets from the Utilization 0.06 sources to the destinations. MRWA 0.04 • Throughput: It is the number of packets received REMR successfully. 0.02 C. Results 0 2 4 6 8 10 A. Effects of Varying Traffic In the initial simulation, the traffic rate is varied as 2Mb, Rate(MB) 4Mb, 6Mb, 8Mb and 10Mb and measure the throughput, end- to-end delay and bandwidth utilization. Figure 4. Rate Vs Utilization Rate Vs Throughput Figure. 2 shows the throughput occurred when the rate is increased. From the figure, it is proved that the throughput is 1.5 more in the case of MRWA when compared to REMR. Figure.3 shows the end-to-end delay occurred when the 1 MRWA rate is increased. It shows that the delay of MRWA is REMR significantly less than the REMR. 0.5 Figure.4 shows the bandwidth utilization obtained when 0 the rate is increased. MRWA shows better utilization than the 2 4 6 8 10 REMR scheme. Rate(MB) B. Effect of Varying Traffic Figure 2. Rate Vs Throughput In this simulation , the number of traffic sources is varied as 1, 2, 3, 4 and 5 and measure the throughput, end-to-end delay and bandwidth utilization. Rate Vs Delay Traffic Vs Throughput 2000 1500 3500 Delay(sec) MRWA 3000 1000 Throughput REMR 2500 500 2000 MRWA 1500 REMR 0 1000 2 4 6 8 10 500 Rate(MB) 0 1 2 3 4 5 Figure. Rate Vs Delay Traffic Figure 5. Traffic Vs Throughput 180 http://sites.google.com/site/ijcsis/ ISSN 1947-5500 (IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010 distributed to the junction nodes in predetermined proportions Traffic Vs Delay that depend on the capacities of intermediate nodes. Then, paths from source node to each of the destination nodes and 70 the potential paths are divided into fragments by the junction 60 nodes and these junction nodes have the wavelength 50 conversion capability. In order to select the wavelengths for groups to maximize the network capacity, the Least Influence Delay 40 MRWA 30 REMR Group (LIG) algorithm is used, i.e. the wavelength having the 20 least effect on the potential paths is chosen for that group. So 10 the affected network capacity influences the wavelength 0 assignment. By simulation results, it is proved that the proposed technique achieves higher throughput(22% increase) 1 2 3 4 5 and bandwidth utilization (1% increase)with reduced Traffic delay(420sec decrease)for varying rate & 9.4 sec derease in delay ,0.3% increase in utilization and 480times increase in Figure 6. Traffic Vs Delay throughput for varying traffic. REFERENCE Traffic Vs Utilization [1] A. Rajkumar and N.S.Murthy Sharma, “A Distributed Priority Based Routing Algorithm for Dynamic Traffic in Survivable WDM Networks”, IJCSNS International Journal of Computer Science and Network 1 Security, VOL.8 No.11, November 2008. 0.8 [2] Canhui (Sam) Ou Hui Zang, Narendra K. Singhal, Keyao Zhu, Laxman Utilization H. Sahasrabuddhe, Robert A. Macdonald and Biswanath Mukherjee, 0.6 MRWA “Sub path Protection For Scalability And Fast Recovery In Optical 0.4 REMR WDM Mesh Networks”, IEEE Journal On Selected Areas In Communications, Vol. 22, No. 9, November 2004. 0.2 [3] Vinh Trong Le, Son Hong Ngo, Xiao Hong Jiang, Susumu Horiguchi and Yasushi Inoguchi, “A Hybrid Algorithm for Dynamic Lightpath 0 Protection in Survivable WDM Optical Networks”, IEEE, 2005. 1 2 3 4 5 [4] Multicasting: http://en.wikipedia.org/wiki/Multicasting Traffic [5] Fen Zhou, Miklos Molnar and Bernard Cousin, “Distance Priority Based Multicast Routing in WDM Networks Considering Sparse Light Splitting”, IEEE 11th Singapore International Conference on Figure 7. Traffic Vs Utilization Communication Systems – 2008 [6] Xiao-Hua Jia, Ding-Zhu Du, Xiao-Dong Hu, Man-Kei Lee, and Jun Gu, Figure 5 shows the throughput occurred when varying the “Optimization of Wavelength Assignment for QoS Multicast in WDM number of traffic sources. From the figure it is proved that, the Networks”, IEEE Transactions on Communications, Vol. 49, No. 2, February 2001. throughput is more in the case of MRWA when compared to [7] Jingyi He, S.H. Gary Chan and Danny H.K. Tsang, “Routing and REMR. Wavelength Assignment for WDM Multicast Networks”, In the Figure.6 shows the end-to-end delay occurred when proceedings of the IEEE GLOBECOM 2001. varying the number of traffic sources. It shows that the delay [8] Anping Wang, Qiwu Wu, Xianwei Zhou and Jianping Wang, “A New Multicast Wavelength Assignment Algorithm in Wavelength-Converted of MRWA is significantly less than the REMR. Optical Networks”, International Journal of Communications, Network Figure 7 shows the bandwidth utilization obtained when and System Sciences, 2009 varying the number of traffic sources. MRWA shows better [9] G. M. Fernandez and D. Larrabeiti, “Contributions for All-Optical Multicast Routing in WDM Networks”, 16th International Congress of utilization than the REMR scheme. Electrical Engineering, Electronics and Systems, IEEE INTERCON, 2009 Name of Effects on varying Effects on [10] Nina Skorin-Kapov, “Multicast Routing and Wavelength Assignment in performance rate varying Traffic WDM networks: A Bin Packing Approach”, In the proceedings of optics metrices REMR MRWA REMR MRWA Infobase in the optical networks, 2006. THROUGHPUT 0.49 0.71 1520 2000 [11] Fen Zhou, Miklos Molnar and Bernard Cousin, “Multicast Routing and DELAY 1440sec 1020sec 48.6sec 39.2sec Wavelength Assignment in WDM Mesh Networks with Sparse Splitting”, The 5th International Workshop on Traffic Management and UTILISATION 0.0325 0.041 0.456 0.7 Traffic Engineering for the Future Internet, Dec- 2009. [12] Yuan Cao and Oliver Yu, “QoS-Guaranteed Routing and Wavelength Assignment for Group Multicast in Optical WDM Networks”, VI. CONCLUSION Conference on Optical Network Design and Modeling, 2005 In this paper, a multicast routing and wavelength [13] N.Kaliammal and G. Gurusamy, “Resource Efficient Multicast Routing assignment technique in WDM networks is developed. In this Protocol for Dynamic Traffic in Optical WDM Networks”, European Journal of Scientific Research, 2010 technique, the incoming traffic is sent from the multicast [14] Network Simulator: www.isi.edu/nsnam/ns source to a set of intermediate junction nodes and then, from the junction nodes to the final destinations. The traffic is 181 http://sites.google.com/site/ijcsis/ ISSN 1947-5500 (IJCSIS) International Journal of Computer Science and Information Security, Vol. 8, No. 7, October 2010 N. Kaliammal received the B.E (ECE)., M.E(Applied pursuing the Ph.D. degree in Optical Networking, under the guidance of electronics), degrees from the Department of Electronics Dr.G.Gurusamy, Dean and Head/ EEE Department, Bannariamman Institute and Communication Engineering ,from Madurai Kamaraj of Technology, Sathyamangalam, Tamil Nadu. University , Bharathiar University, Tamilnadu , in 1989, 1998, respectively. From 1990 to 1999, she served in the Dr.G.Gurusamy, received his BE, ME and PhD degree from PGS college of PSNA College of Engineering & Tech, Dindigul, technology-Coimbatore. He has 35 years of teaching experience in PSG Tamilnadu, as Lecturer. From 1999 to 2009, she was in College of technology-Coimbatore. He is currently working as a Prof & RVS College of Engineering & Tech, Dindigul, Tamil Dean/in EEE Department of Bannariamman Institute of Technology- Nadu, as assistant professor and Associate Professor. Currently she is working Sathyamangalam. as Professor in NPR College of Engineering &Technology She is currently 182 http://sites.google.com/site/ijcsis/ ISSN 1947-5500