| 摘要 |
Voice over Internet Protocol (VoIP) transmission quality predictor working in real-time provides feedback information regarding transmission impairments, transmission quality and end-user satisfaction to quality enhancement mechanisms along the transmission path. Quality enhancement mechanisms use this feedback information in the process of tuning their control parameters. The transmission quality predictor calculates the transmission rating factor R at each of the outputs based on the information regarding voice codec, packet loss, and mouth-to-ear delay. Information regarding voice codec and packet loss is determined from VoIP packet headers. Information regarding mouth-to-ear delay is determined from an additional time stamp that is inserted into each RTP packet header by the VoIP sender. Finally, the play-out delay of received VoIP packets for the actual time window is set equal to the play-out buffer output in the prior time window which gave the highest R-factor value. |
