Joe Walker
CS125
12/17/04
avt – Audio/Video Transport Working Group
Technology Involved
By far, the main technology involved in the AVT Working Group is RTP (Real-Time Protocol). RTP is a protocol that provides the functionality necessary for transferring real-time content, specifically audio and video. A timestamp and control mechanisms are used to correctly synchronize different streams with timing properties. RTP is currently unreliable as it defines nothing to aid in the recovery of packet loss. However, it supposedly was designed to eventually be much more reliable than it is.
Many of the RFCs and Internet Drafts in the AVT Working Group have to do with an RTP payload format for some kind of codec. The documents go into excruciating detail on exactly how the appropriate payload is formatted. A few that caught my eye, but never completely made sense to me were "RTP Payload for DTMF Digits, Telephony Tones and Telephony Signals" and "RTP Payload Format for Uncompressed Video."
Mailing List Activities
The mailing list generally sent out a daily issue, with anywhere from 1 to 10 messages within. Unfortunately, I didn’t see any flame wars. There were a few new members trying to post job listings and annoying the group chairs, but nothing too entertaining besides that. Some of the most commonly discussed RFCs were 2429 (RTP Payload Format for the 1998 Version of ITU-T Rec. H.263 Video (H.263+)), 2190(RTP Payload Format for H.263 Video Streams), 2733(An RTP Payload Format for Generic Forward Error Correction), 3711(The Secure Real-time Transport Protocol (SRTP)), and 2833(RTP Payload for DTMF Digits, Telephony Tones and Telephony Signals).
Activities at the Fall IETF Meeting
The AVT Working Group met twice during the Fall IETF Meeting, on November 10 and 11, 2004. Topics covered included the RTCP XR MIB, the RTP Profile for TCP Friendly Rate Control, an RTP header extension for anti-shadow redundancy, and RTP payload formats for telephony tones and events. The chairs of the meeting were Colin Perkins and Magnus Westerlund.
No RFCs had been published for the group since the previous meeting, but at the time of the meeting, there were four drafts with the RFC editor awaiting publication. Publication has been requested for an addition eleven drafts, which are pending review by the area director or the IESG.
The chair noted that the group was behind on some milestones, the most notable being the updated RTP MIB. Detailed discussions ensued on RTP Payload Format for MIDI, RTCP XR MIB, Open Issues in RFC 2833bis, RTP Payload Format for JPEG 2000, Far End Camera Control/H.224, RTP Payload for Anti-Shadow Redundancy, RTP Jitter Calculation, RTP Profile for TCP Friendly Rate Control, RTP Timestamps for Variable Frequency Codecs, RTP Payload Format and File Format for VMR-WB, RTP Payload Format for AMR and AMR-WB, and RTP Payload Format for AMR-WB+.
Where the Technology is Going
It seems like a lot of this technology applies well to cell phones. For example, in the discussion of the RTP Payload Format for AMR-WB+, it was mentioned that there may be an unpublished Nokia patent application that relates to this draft. It was also mentioned that AMR-WB+ is the recommended codec for a few different types of cell phone protocols, and will soon be mandatory in 3GPP MBMS.