This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Balancing the load among multiple paths leads to the degradation in the all-optical WDM networks. It is always desirable to deliver the entire load on a single optimal path depending upon the current network status. For this, multiple light paths need to be maintained to provide the choice of selecting the optimal path, based on the changing traffic conditions . The limitation of a no conversion network is the wavelength continuity constraint which requires the same wavelength on all links of the light path . The wavelength continuity constraint can be overcome by the placement of converters at all nodes, but the utilization of converters at all nodes in a path is not a good practice . The Centralized Management system (CMS) controls the entire process of data transmission in a network. The failure of the CMS results in the breakdown of entire network . In this paper, we proposed an efficient centralized optimal path with minimum number of conversion (OMNC) algorithm in which it identifies an optimal and backup path with minimum number of conversions and also the tasks are shared by using two CMS, if one fails the other will take over the entire process. Our approach is self regulating, it automatically adapts to the traffic load variation across the network. By simulation result we showed that our proposed algorithm uses less end-to-end delay and less transmission power consumption. Keywords: RWA, centralized management system, WDM networks, multipath, wavelength continuity constraint, wavelength conversion. 1. Introduction Wavelength Division Multiplexing (WDM) and routing are rapidly becoming the technologies-of-choice in network infrastructure to meet the tremendous bandwidth demand of the new millennium. In wavelength routed WDM networks, a connection is realized by a lightpath. In order to establish a connection between a source and destination pair, a wavelength continuous route to be found between the node pair. An algorithm is used for selecting routes and wavelengths to establish lightpaths is known as routing and wavelength assignment (RWA) algorithm. Many problems in wavelength routed WDM networks have RWA as sub problem. Therefore, it is mandatory to use a good routing and wavelength assignment algorithm to establish lightpaths in an efficient manner. A WR algorithm has two components, route selection and wavelength selection. A WR algorithm is also known as routing and wavelength assignment (RWA) algorithm. The route is chosen based on some cost criterion such as the hop length, and the wavelength is chosen based some criterion such as wavelength usage factor in the entire network. Route selection algorithms can be broadly classified into three types: Fixed routing, Alternate routing, and Exhaust Routing. This classification is based on the restriction (if any) on choosing a route from among all possible routes. The wavelength selection algorithm can be broadly classified into most used, Least used, Fixed order, and Random order, depending upon the order in which the wavelengths are searched. The order of such intern may or may not depend on the usage status of the wavelengths. RWA algorithms assume either centralized or distributed control for selecting routes and wavelengths. In the case of centralized control, a central controller is assumed to be available. It keeps track of the state of the network. It is responsible for selecting routes and wavelengths for the request and sending control signals to appropriate nodes for establishing and releasing light paths. Disadvantage of centralized systems is if the master fails, the entire integrated home system does not work. In the case of distributed control, no central controller is used. Up-to-date knowledge of the network state is not known to any node. The computation of the primary and backup paths for an arriving demand is the responsibility of the source node . As it is impossible for each node to maintain the complete network status information, only limited network status information is maintained by each node. This limited information is used for the path computation.
In view of the fact that each lightpath can carry a huge amount of traffic, failures in such networks may seriously damage end-user applications. In accordance with the scale of their effect, failures in all-optical WDM networks can be classified into two categories . One category is a wavelength-level failure which affects the quality of transmission of every individual lightpath. The other category is a fiber-level failure which impacts all the
lightpaths on an individual fiber. As each lightpath is expected to operate at a rate of several gigabytes per second, a failure can lead to a severe data loss. Therefore, it is very important to have networks that are capable of preventing such failures; these are known as survivable networks . In survivable WDM networks, the lightpath known as primary lightpath carries traffic during normal operations. In case a primary lightpath fails, the traffic is rerouted over a new lightpath known as the backup lightpath. The protection and/or restoration schemes can be used to protect the network against failures. 2. Related work Alvaro L.Barradas and Maria do Carmo R. Medeiros  have presented a traffic engineering approach for path selection with the objective of minimizing contention using only topological information. The main idea was to balance the traffic across the network to reduce congestion without incurring link state dissemination protocol penalties. This protocol also does not consider the dynamic traffic load into consideration and hence not adaptive. Dong-won shin et al.  has developed two heuristic multipath routing schemes for survivable multipath problem: CPMR (conditional penalization multipath routing) and SPMR (Successive penalization multipath routing). To deal with the difficulty that link sharing causes, their schemes use â€•link penalizationâ€- methods to control (but not prohibit) link-sharing. Through the simulation they have shown that their schemes have significantly higher routing success rates than a routing scheme that searches for disjoint paths. Mauro Brunato et al.  developed a Load Balancing algorithm for IP-based Optical Networks. Their algorithm (RSNE - Reverse Sub tree Neighborhood Exploration) performed at each iteration which is a basic change of a single entry in a routing table in order to minimize the disruption of the network. They studied the performance of their algorithm in realistic networks under static and dynamic traffic scenarios. Their simulation results shows a rapid reduction of the congestion for static networks and a performance of the incremental scheme while tracking a changing traffic matrix comparable to the complete re-optimization of the traffic.
P. Rajalakshmi_and Ashok Jhunjhunwala  have proposed two heuristic reassignment algorithms namely, Minimum overlapped to Least congested and Random (MOLC) techniques and have studied their performance on some standard backbone optical networks. Simulation results show that the reassignment algorithm can mostly remove the blocking due to the Wavelength continuity constraint (WCC) and can achieve the wavelength conversion performance without using wavelength converters, but as the reassignment is done link by link, the latency for data transmission will increase and also loss in data occurs due to disturbance of a live call during the reassignment process. 3. RWA mathematical model The RWA problem is an optical networking problem with the goal of maximizing the number of optical connections. The RWA problem can be formally defined in an integer linear programming (ILP). Linear programming is mathematical method for determining a way to achieve a best outcome in a given mathematical model for some list of requirements represented as linear relationship. In our ILP model for optimization we considered the path with the less number of conversions so as to minimize the end-to-end delay and power consumption and notations used are as follows:
ï‚·ï€ D - Denotes the physical node
ï‚·ï€ Cu - Cost per unit length of the fiber
ï‚·ï€ Li - Length of the link i [Total â€-m' links are available in network]
ï‚·ï€ Fg - Total number of light channels
ï‚·ï€ Cb -Cost of convertible nodes
ï‚·ï€ Nn - Total number of nodes
ï‚·ï€ Pj - All possible paths from Source to Destination [ â€-j' is path number and maximum number of paths are k]
ï‚·ï€ Hj - Number of hops in the path j .
ï‚·ï€ Nrj - Maximum number of requests can be processed in path â€-j'
ï‚·ï€ NaÎ»i - Number of available free wavelengths on link â€-i'
ï‚·ï€ NuÎ»i - Number of used wavelengths on link â€-i'
ï‚·ï€ Npci - Number of packets available at a particular instant on channel in link â€-i'
ï‚·ï€ T - Amount of traffic on entire network
ï‚·ï€ Dpi - Propagation delay on link â€-i'
ï‚·ï€ Dcb - Conversion delay at node â€-b'
ï‚·ï€ PN - Node pair
ï‚·ï€ Gi - Link i
ï‚·ï€ Wi - Wavelength i
ï‚·ï€ P - Light path
ï‚·ï€ P[w] - Conversion range of wavelength Î»
ï‚·ï€ Nc - Number of converters at nodes
ï‚·ï€ PL-Number of light paths
ï‚·ï€ RP PN - Number of times route P is used for request of the node pair PN
ï‚·ï€ YPNP,g,W - Number of times wavelength Î» is assigned to link g
ï‚·ï€ Î´PNp,g - The existence of consecutive links (g1, g2) on the route P connecting the node pair PN
ï‚·ï€ Î´PNp,g = 1 if â€-l' belongs to the route
Î´PNp,g = 0 otherwise
ï‚·ï€ Î¦ PNP, (g1,g2) - The existence of consecutive links (g1,g2) on the route P connecting the node pair PN
ï‚·ï€ Î“Dg = 1 if the node D is one of the two end nodes of link g
Network cost The cost of the network can be calculated by adding the cost of all links and nodes. Number of requests Nai = NaÎ»i if link I on the path = 0 otherwise. Nrj = max(Na1, Na2â€¦â€¦.Nam) Above equation gives that the maximum number of request can be processed at a source node. Path Selection P= Min (H1, H2â€¦..Hk) The above equation gives the path with less number of hops present between source node and destination node. Amount of traffic on network T= Min The above equation gives the amount of data present on entire network at that particular time. Delay of path j DPi = DPi if link i on path j 0 other wise DCb =Dcb if node b on path j
Minimization of number of conversions Minimize )(),(),,(,2121ïƒ¥ïƒ¥ïƒ¥ïƒ¥ïƒ¥ï€«nnPPggwPwggPDcYCNï¢ï€ Where ï¢ï€ is extremely small quantity (0 < ï¢
<1) Lightpath routing and wavelength assignment constraints. lrPPPRnï€½ïƒ¥ï€ nPï€¢ï€ (1) ïƒ¥ï€½ï„wPwgPPgPPPnnnYR,,,.gï€¢ï€ (2) gPPPwgPfYnnï‚£ïƒ¥ïƒ¥,, ï€¢g (3) Equation (1) represents the routing of requested lightpaths of the node-pair Pn. Equation (2) represents the wavelength assignment on a link. The left-hand side of the equation is the number of lightpaths of the node pair Pn going through the link g. The right-hand side is the number of wavelengths on the link g used for lightpaths of the node-pair Pn going through that link. So the equation ensures that, to all lightpaths, free wavelengths are assigned. Equation (3) ensures that the number of times a wavelength is used on a link is lower than or equal to the maximum number of light channels on that link so that two wavelengths will not use the same channel. 4. Algorithm Step1 : CMS-I checks all the links in the network and if any link is failed then it inform Repair Processor (RP) about the failure and it removes that particular link for further calculation. Step2 : When request comes at any node it intimates the request details to CMS-I. Step3 : CMS-I checks for all the available paths from source to destination and stores in an array P[n], number of links on each path is stored in L[n]. Step4 : Calculate the number of conversions required for each path and store it in PX[n], arrange the path numbers in incremental order with respect to the number conversions required and store the number of conversion detail in C[n] as per the paths in PXi[n]. Step5 : If the value of C and C is equal check for the lengths and if L[pxi(0)]<L[pxi(1)] then PXi is optimal path and PXi is backup path. If the L[pxi(0)]>L[pxi(1)] then PXi is optimal path and PXi is backup path . Step6: If the values of C and C are not equal then PXi is optimal path and PXi is backup path. Now the data will be transmitted through the primary path. Parallel process by CMS-II Step7: CMS-II gets the updates and details from CMS-I and stores in table. Step8: It repeats the steps 4, 5, 6. If there are any changes in the path, it sends a request to the CMS-I which is currently transferring data in the network otherwise it waits for some time â€•tâ€- sec and then repeats step8. Step9: If there is any request from CMS-II then change the paths according to CMS-2 otherwise transfer the data. Step10: While transferring the data in a primary path if there is any failure in the path then inform the Repair Processor (RP) and get the last received packet details at the destination and intimate to the CMS then transfer the data from that last received packet through the Backup path.
5. Simulation Result 0123456789101100.511.522.533.544.55 no of links per requestDelay time OMNCMOLC Fig.1 Number of links Vs Delay 024681012050100150200250 no of links per requestPower consumed in dBM OMNC powerMOLC power Fig.2 Number of links Vs power Fig 1 shows the comparison between the OMNC and MOLC with respect to number of links Vs end to end delay. As per the simulation result OMNC have better performance compared to the MOLC. Fig 2 shows the comparison between the OMNC and MOLC with respect to number of links Vs power consumption. As per the simulation result OMNC have better performance compared to the MOLC.
Total number of nodes
No of links
Link wavelength number
Wavelength conversion factor
Wavelength conversion time
Table1: Simulation Parameters 6. Conclusion In this paper an optimal path with minimum number of conversions (OMNC) algorithm is developed using an efficient centralized system in which two Central management systems (CMS) share the entire work and also act as backup to each other. As opposed to the reactive algorithm our algorithm is proactive in the sense that it optimizes power and reduces latency furthermore our approach is self regulating, it automatically adapts to the traffic conditions across the network. By simulation results we have shown that our proposed algorithm has reduced delay and power consumption.
Reference  T.K Ramesh and P.R. Vaya, â€• An Adaptive Reliable Multipath Routing Protocol for WDM Networksâ€-,IJCES International Journal of Computer Sciences and Engineering Systems, vol. 4, No. 3, July 2010.  Laxman Sahasrabuddhe, S. Ramamurthy, and Biswanath Mukherjee, "Fault Management in IP-Over-WDM Networks: WDM Protection Versus IP Restoration", IEEE journal on selected areas in communications, vol. 20, no. 1, pp: 21- 23, January 2002, Doi: 10.1109/49.974659.  Optical WDM Networks by Biswanath Mukherjee, University of California, Davis  C.Siva Ram Murthy, Mohan Guruswamy, â€•WDM optical networks- Concepts, Designs and Algorithms.â€-  Lu Ruan, Member, IEEE, Haibo Luo, and Chang Liu,â€- A Dynamic Routing Algorithm With Load Balancing Heuristics for Restorable Connections in WDM Networksâ€-, IEEE journal on selected areas in communications, VOL. 22, NO. 9, November 2004  S. Ramamurthy, Laxman Sahasrabuddhe, and Biswanath Mukherjee, "Survivable WDM Mesh Networks", journal of light wave technology, vol. 21, no. 4, April 2003  Harsha V. Madhyastha and N. Balakrishnan,"An Efficient Algorithm for Virtual-Wavelength-Path Routing Minimizing Average Number of Hops", IEEE Journal on Selected Areas in Communications, Volume 21, Issue 9, Nov. 2003 Page(s): 1433 - 1440.  Alvaro L.Barradas and MAria do Carmo R. Medeiros, "Edge- Node Deployed Routing Strategies for Load Balancing in Optical Burst Switched Networks", ETRI Journal, vol.31, no.1, pp: 31-41, Feb. 2009.  Dong-won Shin, Edwin K.P.Chong and Howard Jay Siegel, â€•Survivable Multipath Routing Using Link Penalizationâ€-, Computing and Communications, 2004 IEEE International Conference.  Mauro Brunato, Roberto Battiti and Elio Salvadori, â€•Load Balancing in WDM Networks through Adaptive Routing Table Changesâ€-, Proceedings of Networking 2002-Lecture Notes in Computer Science, Springer-Verlag1.  P. Rajalakshmi_, Ashok Jhunjhunwala,â€- Wavelength reassignment algorithms for all-optical WDM backbone networksâ€-, Optical Switching and Networking 4 (2007) 147-156.