Performance analysis and improvement of some quality of service parameters in optical packet switching networks

Abstract

In the information technology era, there are relentless needs for networks with extremely high capacity. At this juncture to make the dreams of the network users come true, tremendous research has resulted in optical technology which is a great boon to the world of communication. Optical Packet Switching (OPS) presents new avenues for high speed interconnection networks. Quality of Service (QoS) is the only way through which optimization can be attained in the OPS. The major parameters that affect the performance of QoS in OPS are: Bit Error Rate (BER), Packet Loss Rate (PLR), Bandwidth, Delay, Jitter, Recovery time, Response time, Reliability, Availability, Fault tolerance and Throughput and so on. In order to analyze the performance of QoS, the parameter bit error rate (BER) is considered as the first parameter. The second parameter taken into consideration is packet loss rate (PLR) in the network layer. The performance analysis of bufferless slotted optical packet switching using non-uniform traffic pattern is explained. Further, slotted optical packet switching mechanism employed in non-uniform traffic pattern is analyzed using proposed reservation bit algorithm. Buffers with the size of 5 packets are introduced to implement reservation bit algorithm. The third parameter subjected to analyze is delay in network layer. The fourth parameter reliability is assessed in NSFNET (National Science Foundation Network topology). In order to improve the performance of optical packet switching networks, reliability is examined. In future, QoS can be improved by considering the remaining parameters such as jitter, latency, throughput, availability, fault tolerance, recovery time, response time and etc by implementing reservation bit algorithm. By applying reservation bit algorithm, the networks can be operated with minimum loss in the presence of faults. Even though these parameters are related with packet loss, delay and reliability, by employing reservation bit algorithm, accurate QoS can be measured in the OPS.