A New Error Control Scheme for Packetized Voice over High-Speed Local Area NetworksReport
Distribution of packetized digital speech across high-speed LANs has become increasingly feasible, and packet voice protocols supporting audio conferencing are available. These protocols must provide mechanisms to address errors and delay introduced by the network that threaten the quality of the voice playback at the receiving site. A common mechanism for ensuring continuous voice playback in the presence of delay jitter is the use of a control time for the first packet in each talkspurt. Packet loss is generally not recovered in packet voice protocols, though techniques such as forward-error correction (FEC) and priority channels can be used to reduce losses. In this paper we propose a new error control mechanism for packet voice, referred to as Slack ARQ (S - ARQ). S-ARQ is based on extending the control time for the first packet in each talkspurt to allow for the timely retransmission of lost packets. S-ARQ is orthogonal to the use of preventive error control schemes such as FEC or priority channels. It does not require hardware support, imposes little overhead on network resources, and could be easily integrated into extant voice protocols. Experiments using a simulation model developed in this paper indicate that the extended control time required for S - ARQ can be modestly larger than the control time for jitter control and within the limits imposed by end-to-end delay requirements. The feasibility of the S-ARQ scheme is thus established.
Note: Abstract extracted from PDF file via OCR
All rights reserved (no additional license for public reuse)
Dempsey, Bert, Jorg Liebeherr, and Alfred Weaver. "A New Error Control Scheme for Packetized Voice over High-Speed Local Area Networks." University of Virginia Dept. of Computer Science Tech Report (1993).
University of Virginia, Department of Computer Science