Performance evaluation of assembly mechanisms for jet-based optical burst switching networks | |
| Author | Jenjira Uengsakul |
| Call Number | AIT Thesis no.TC-03-04 |
| Subject(s) | Switching circuits Packet switching (Data transmission) |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering; School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. TC-03-04 |
| Abstract | Switching in Optical Network is based on the Optical Circuit Switching (OCS), Optical Packet Switching (OPS) or Optical Burst Switching (OBS) techniques. The OBS is the technique that trades off the system complexity and required transmission bandwidth between the OPS and OCS technique. One of the challenging issues in the OBS network is the burst assembly mechanism. In this thesis, the characteristic of four burst assembly mechanisms and their influence in the network performance are studied. The considered assembly methods are the threshold-based without time limitation, the time-based with minimum burst length, the fixed length threshold-based with time limitation and the variable length hysteresis-based with time limitation, which are called as the Threshold-based, Timer-counter-based, Fixed-based and Hysteresis-based respectively. The system performance is observed in simulated 14-node network based on Just-Enough-Time (JET) reservation protocol. The point-to-point traffic flows in the network are generated by random Exponential and Pareto ONOFF sources. The burst characteristic is observed at the assembly function of the edge switch. With the adaptable burst size decision value, the Hysteresis-based assembly method provides better data burst utilization when compared to the Fixed-based and Timer-counter- based method. It also gives the lowest average assembly delay time. In the network level, the Threshold-based assembly scheme provides the lowest loss probability but with the highest average assembly delay. The Timer-counter-based method gives the highest loss probability variation. The Fixed-based technique, which includes both burst size decision threshold and maximum delay time to control the burst creation process, provides higher loss probability than the Threshold-based method but shorter average assembly delay. The Hysteresis-based method, which improves the Fixed-based method by using adaptable burst size decision value, can further reduce the loss probability and average assembly delay. The simulation results in this studied network show the loss probability reduction in the range of 0.92% to 37.6% depending on the system criteria. The reduction depends on the value of burst size decision threshold and the maximum assembly delay time, which are the criteria of the burst creation. |
| Year | 2003 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis ; no. TC-03-04 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Telecommunications (TC) |
| Chairperson(s) | Erke, Tapio J. ; |
| Examination Committee(s) | Ahmed, Kazi M. ;Teerapat Sanguankotchakorn; |
| Scholarship Donor(s) | Royal Thai Government Fellowship; |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 2003 |