Analytical model of stream transported media. Issue 1 (1st January 2017)

• Record Type: Journal Article
• Authors: Herrero, Rolando; St‐Pierre, Claude
• Is Part Of: IET networks. Volume 6:Issue 1(2017)
• Page Start: 14
• Page End: 21
• Publication Date: 2017-01-01
• Subjects: fading channels -- speech codecs -- protocols -- telecommunication traffic -- firewalls -- packet radio networks; state‐of‐the‐art speech codecs -- wireless communications -- fading channels characteristic -- bursty network packet loss -- application layer packet loss -- media frames -- switching transport -- Web friendly traffic -- Internet firewalls -- datagram‐based transport -- RTP traffic -- real‐time protocol traffic -- real‐time communications -- packetised media transmission -- analytical model -- stream transported media; Computer network architectures -- Periodicals; Computer network protocols -- Periodicals; Information networks -- Periodicals; Telecommunication systems -- Periodicals; 004.605
• DOI: 10.1049/iet-net.2016.0076
• Languages: English
• ISSNs: 2047-4954

Abstract

Abstract : For transmission of packetised media, real‐time communications rely on the real‐time protocol (RTP). RTP traffic, however, is transmitted over datagram‐based transport that allows fast delivery of media. Internet firewalls are typically configured to allow web friendly traffic that uses stream‐based transport instead and, simultaneously as security measure, reject any other type of transport. A mechanism to overcome this limitation is by switching transport such that media frames that would be rejected by firewalls are transmitted on top of stream transport. In this study, a novel mathematical model that links application layer packet loss to bursty network packet loss in the context of fading channels characteristic of wireless communications is presented for both, datagram and stream transport. This model is compared with experimental scenarios applied to state‐of‐the‐art speech codecs in order to obtain quality scores that can be correlated against theoretical loss probabilities.