Modelling and Simulation of Adaptive Buffer Management Mechanism Using Nested Autoregressive Process-Based Mpeg-Encoded Traffic Model

Abstract

High speed networks provide real-time variable bit rate (VBR) service with diversified traffic flow characteristics and quality requirements. The VBR traffic has stringent delay and packet loss requirements. The burstiness of the correlated traffic makes dynamic management highly desirable to satisfy the Quality of Service (QoS) requirements. Buffer management in queueing systems plays important role towards effective control of QoS for various types of applications. In this paper, ADaptive Partial Buffer Sharing (ADPBS) scheme for priority queueing system is proposed and analyzed. The input to the queue has been taken using auto-regressive (AR) process of order 2 for MPEG-encoded VBR video traffic. The results of simulations indicate that dynamically controlled threshold in ADPBS contributes in reducing the packet loss of different priority classes as compared with static partial buffer sharing queue and first-in first-out queue based on threshold control parameters and the traffic type.

Key Words:

• Priority queueing
• auto-regressive process
• buffer management
• quality of service
• packet loss ratio

Additional information

Notes on contributors

S. Kaushal

Sakshi Kaushal is working as a Lecturer at Panjab University, Chandigarh, India. She has received her Ph.D., M.Tech., and B.Tech. Degrees in Computer Science and Engineering. Her main research interest includes quality of service and traffic control analysis of high speed networks.

R.K. Sharma

Rajendra Kumar Sharma received his Ph.D. degree in mathematics from the University of Roorkee (Now, IIT Roorkee), India in 1993. He is currently working as a Professor at Thapar Institute of Engineering and Technology (TIET), Patiala, India, where he teaches, among other things, queuing models and its usage in computer networks. He has been involved in the organization of a number of conferences and other courses at TIET, Patiala. His main research interests are in traffic analysis of computer networks, neural networks, and pattern recognition.