Fuzzy logic controllers for MPEG VBR video transmission over ATM networks

Abstract

The overall aim of the paper is concerned with transmission of a digital video signal over an asynchronous transfer mode (ATM) network, using a self-organizing fuzzy (SOF) controller. In ATM a traffic-shaping buffer is used to prevent excessive back-to-back cells being generated during the peak transmissions of Moving Picture Expert Group (MPEG) variable bit rate (VBR) data sources. The VBR data sources are unpredictable and have substantial variations in bit rate, which could result in data loss and therefore image quality degradation. In this research, a self-organizing fuzzy controller is applied to digital video to eliminate excessive delay or loss at the user network interface. A self-organizing fuzzy controller adjusts the traffic-shaping buffer output rate frame by frame to enable the VBR encoded video to conform to the leaky bucket's contract. A rule-based fuzzy (RBF) controller regulates the average arrival rate to the traffic-shaping buffer to prevent either overflow or starvation of the buffer on a group of picture (GOP) by GOP basis. The computer simulation results demonstrate that the use of a SOF controller reduces excessive delay and data loss at the user network interface as compared with a conventional policing mechanism in ATM.