Comparative Study of Wavelength Assignment Algorithms for WDM Optical Network

DOI : 10.17577/IJERTV2IS100492

Download Full-Text PDF Cite this Publication

Text Only Version

Comparative Study of Wavelength Assignment Algorithms for WDM Optical Network

Namita Nag 1 & Indu Bala 2

ECE Department, CEC Mohali

Abstract

Wavelength assignment in optical networks is a critical designing issue as it affects the overall performance of the network. In this paper, wavelength assignment algorithms have been reviewed namely random, first fit and dynamic (proposed) wavelength assignment algorithm. For given system parameters, the performance of all three algorithms is evaluated in terms of blocking probability. Results are showing that performance of dynamic algorithm is better than other two algorithms discussed.

  1. Introduction

    Communication networks are reaching more and more people every day and providing new means of information exchange. Consequently, data and traffic demand is growing rapidly (exponentially). All these bandwidth hungry applications have increased user data traffic and therefore demand for bandwidth [1]. This exponential growth has triggered the development of new technologies, and the advancement of existing one promising technology that meets the high bandwidth demand of data networks is optical networking technologies. This can be accomplished by a multiplexing technique like time division multiplexing (TDM), code division multiplexing (CDM), and wavelength (or frequency) division multiplexing (WDM) [2].

    WDM network can provide end-to-end optical communication channels through optical fibres and intermediate nodes with optical cross-connects, even the source and destination nodes are not connected by a fibre directly. There are issues related with optical communication system like congestion. The problem of congestion can be handled using wavelength assignment technique.

    In this paper various wavelength assignment techniques have been reviewed and their performance comparison is done in terms of blocking probability. Section 2 briefly describes the routing and wavelength assignment issues and constraints. In section 3 pseudo code for Random, First-Fit and dynamic (Proposed) algorithms are given. In section 4 performance comparison for these algorithms is given for variable system parameters. In the end of this paper, conclusion is given based on simulated results.

  2. Routing and Wavelength Assignment

    RWA is the unique feature of WDM networks in which light path is implemented by selecting the path of a physical link between source and destination edge nodes and reserving a particular wavelength on each of these links for the light path. Routing deals with an optical connection and wavelength assignment policy allocate available wavelengths for the path selected for routing. Collectively it is known as routing and wavelength assignment (RWA) [3].

    There are two constraints that have to be kept in mind by the approaches when trying to solve RWA.

    1. Distinct wavelength assignment constraint: All light paths sharing a common fiber must be assigned distinct wavelengths to avoid interference. This applies not only within the all- optical network but in access links as well [4].

    2. Wavelength continuity constraint: The wavelength assigned to each light path remains the same on all the links it traverses from source end-node to destination end- node [4].

      Wavelength continuity constraints can be eliminated by using wavelength converter. The wavelength routed networks with this capacity is known as wavelength convertible networks.

  3. Wavelength Assignment Algorithms

Wavelength assignment is a unique feature in which wavelength are searched before allocating to the path selected. Many algorithms have been proposed by researchers in past few years. In this section the most popular wavelength assignment algorithms have been reviewed [5]. These are:

  1. Random wavelength assignment algorithm

  2. First-fit wavelength assignment algorithm

  3. Proposed wavelength assignment algorithm

    1. Random wavelength assignment algorithm

      Step1: Initialisation of network parameters. Step2: Select any source destination pair.

      Step3: Select any root out of all possible roots for selected source destination pair.

      Step4: Assign any wavelength out of available wavelengths.

      Step5: Is blocking probability of path is greater than threshold value?

      Step6: If yes, repeat step4 and select new wavelength value if no, establish network connection.

      Step7: Repeat step2.

    2. First Fit Wavelength Assignment Algorithm

      Step1: Initialisation of network parameters. Step2: Select any source destination pair.

      Step3: Select any root out of all possible roots for selected source destination pair.

      Step4: Arrange available wavelengths in ascending order.

      Step5: Assign first/next wavelength to the selected path.

      Step6: Calculate blocking probability for selected route.

      Step7: Is blocking probability of path is greater than threshold value?

      Step8: If yes, repeat step4 and select new wavelength value if no, establish network connection.

      Step 9: Repeat step 2 for particular route.

    3. Proposed wavelength assignment algorithm

      Step1: Initialisation of network parameters. Step2: Pick a request with wavelength.

      Step3: Find the first two paths.

      Step4: If yes (path is returned), then make primary and secondary paths. If No, request has been discarded.

      Step5: Send request for s-d pair.

      Step6: Check the destination frequency (frequency matching).

      Step7: If yes, update path selection matching, If no, then frequency conversion is done.

      Step8: Update path structure in links. Step9: Update load of the path assignment. Step10: Update bandwidth structure of path. Step11: Repeat step 2.

  4. Performance Evaluation of algorithms The system parameters considered in simulation set up are mentioned in Table 1 below.

    Table 1: System Parameters

    Parameter

    Value

    Simulator

    NS2

    Simulation Time

    90 mSec

    No of nodes

    2-10

    Traffic Model

    CBR

    Packet size

    2 Bytes

    No. of Packets

    10

    Wavelength Index

    1550.X nm

    *X is variable

    Table 2 to Table 6 are showing the performance comparison of mentioned algorithms for variable load and nodes. The load values taken in simulation are two, four, six, eight and ten only [6].

    Table2: Blocking Probability with Two Erlangs

    Blocking Probability with Two Erlangs load

    Number of Nodes

    Blocking Probability (%) x 10-6

    Random

    First Fit

    Dynamic

    2

    100

    100

    100

    4

    110

    120

    100

    6

    160

    190

    100

    8

    190

    210

    130

    10

    230

    220

    150

    Table3: Blocking Probability with Four Erlangs

    Blocking Probability with Four Erlangs load

    Number of Nodes

    Blocking Probability (%) x 10-3

    Random

    First Fit

    Dynamic

    2

    1.1

    2

    1.1

    4

    4.0

    4

    4.0

    6

    5.5

    6

    5.5

    8

    7.3

    8

    7.3

    10

    8.3

    10

    8.3

    Table 4: Blocking Probability with Six Erlangs

    Blocking Probability with Six Erlangs load

    Number of Nodes

    Blocking Probability (%) x 10-3

    Random

    First Fit

    Dynamic

    2

    1.3

    2

    1.3

    4

    3.2

    4

    3.2

    6

    5.3

    6

    5.3

    8

    8.6

    8

    8.6

    10

    9.8

    10

    9.8

    Table5: Blocking Probability with Eight Erlangs

    Blocking Probability with Eight Erlangs load

    Number of Nodes

    Blocking Probability (%) x 10-3

    Random

    First Fit

    Dynamic

    2

    4.1

    2

    4.1

    4

    5.9

    4

    5.9

    6

    7.9

    6

    7.9

    8

    9.2

    8

    9.2

    10

    9.9

    10

    9.9

    Table6: Blocking Probability with Ten Erlangs

    Blocking Probability with Ten Erlangs load

    Number of Nodes

    Blocking Probability (%) x 10-3

    Random

    First Fit

    Dynamic

    2

    5.4

    2

    5.4

    4

    6.7

    4

    6.7

    6

    8.7

    6

    8.7

    8

    9.4

    8

    9.4

    10

    9.9

    10

    9.9

    It is clear from the observations made in tables that dynamic wavelength assignment algorithm out performs over first fit and random algorithm. Blocking probability of dynamic algorithm is significantly less than rest two algorithms studied in this paper.

  5. Conclusion

In this paper, three wavelength assignment algorithms have been discussed. These are Random, First-Fit and Dynamic (proposed) wavelength assignment algorithm. For given system parameters (i. e. Table 1), the performance of all three algorithms is evaluated in terms of blocking probability. Results are showing that performance of dynamic algorithm is better than other two algorithms.

References

  1. K. Sivalingam and S. Subramanian, Emerging Optical Network Technologies, Eds. Boston, MA: Springer-Verlag, pp.4-6, 2004.

  2. C. S. R. Murthy and M. Gurusamy, WDM Optical Networks-Concepts, Design, and Algorithms, Singapore: Pearson Education, pp.264, 2003.

  3. H. Zang, J. P. Jue, and B. Mukherjee, A review of routing and wavelength assignment approaches for wavelength- routed optical WDM networks, Opt. Networks Mag., vol. 1, no. 1, pp.146-149, Jan. 2000.

  4. B. P. Pal, Guided Wave Optical Components and Devices, Elsevier academic Press, 2006.

  5. A. Wason and Dr. R. S. Kaler, Wavelength Assignment Problem in Optical WDM Networks, IJCSNS International Journal of Computer Science and Network Security, VOL.7 No.4, pp.28, April 2007.

  6. N. Nag and I. Bala, Dynamic WDM Routing Solutions, International Journal of engineering Research and Technology, vol. 2, Sept 2013.

Leave a Reply