Multirate voice coding for load control on CSMA/CD local computer networks

Abstract

Abstract A study of mixed voice and data on a local area network using Carrier Sense Multiple Access with Collision Detection (CSMA/CD) is presented. The study determined the feasibility of applying multirate voice coding to control the traffic intensity on the network. By decreasing the voice coding rate for short periods of time (and thus the voice quality) when network traffic increases, a larger number of voice users can be realized. The premise is that short term voice quality can be traded for increased throughput on a CSMA/CD network. Collisions per millisecond was found to give a good indication of the traffic on the network. A feedback equation is used to set the voice coding rate, based on the measured collisions per millisecond. The coding rate determined from the feedback equation is rounded to one of four possible coding rates; other rates are clearly possible. The voice coding rates chosen were 48, 40, 32, and 24 kilo-bits-per-second. The simulation results show that additional voice users can be added to a 1 mega-bit-per-second network when the multirate algorithm is used. Results are given relating the packet delay to the collisions per millisecond. A comparison between multirate and non-multirate is given. An indication of the voice quality for the multirate system is also discussed. The multirate load control algorithm provides a way to increase the amount of voice traffic on a CSMA/CD network.