The threat to circuit-switched telephone services by IP telephony gets neutralised by a bunch of Quality of Service (QoS) issues unique to packet networks. These issues more often than not camouflage the advantages of the reduced cost and bandwidth savings of carrying voice-over-packet networks.

Traditional public switched telephone networks (PSTNs) have long addressed the voice-quality problem by optimizing their circuits for the dynamic range of the human voice and the rhythms of human conversation. PSTNs have evolved to provide an optimal service for time-sensitive voice applications that require low delay, low jitter and constant, but low, bandwidth. PSTN voice quality is relatively standard and predictable.

IP networks, however, were built to support non-real time applications, such as file transfers or e-mail. These applications are characterized by their bursty traffic and sometimes high bandwidth demand but are not sensitive to delay or delay variation.

Voice quality can be improved in two ways – special quality of service algorithms and more bandwidth. The current research on improving QoS in packet networks is directed at enhancing routing capabilities. Besides, billions are being spent on adding more bandwidth capacity to global data networks. These have the potential to make IP telephony a viable commercial alternative to the PSTN.

Delay

Delay, or latency, is the major challenge to QoS for packet voice networks. It is the time required for transmission of data packets from origin to destination. Delay occurring on IP networks primarily results from bandwidth sharing and processing at routers and endpoints within the networks. Data applications, for which IP networks were originally designed, are more tolerant of delay than voice. Data transmission such as e-mail can have an accepted delivery time of several hours. Even more real time applications such as database queries can tolerate several seconds of delay.

Delay causes two problems: echo and talker overlap. Echo is caused by the signal reflections of the speaker’s voice from the far-end telephone equipment back into the speaker’s ear. Echo becomes a significant problem when the round-trip delay becomes greater than 50 milliseconds. As echo is perceived as a significant quality problem, voice-over-packet systems must address the need for echo control and implement some means of echo cancellation. Talker overlap (or the problem of one talker stepping on the other talker’s speech) becomes significant if the one-way delay becomes greater than 250 milliseconds. The end-to-end delay budget is therefore the major constraint and driving requirement for reducing delay through a packet network.

Jitter

Jitter, or delay variation, is the result of packets arriving at their destination at irregular intervals. Bursts of Internet traffic create jitter problems. This distortion is particularly damaging to the QoS of VoIP applications. Severe jitter in IP voice transmissions causes jittery or shaky voice quality.

Removing jitter requires collecting packets and holding them long enough to allow the slowest packets to arrive in time to be played in the correct sequence. This causes additional delay.

Packet Loss

Packet network applications compensate for packet loss by retransmitting lost packets through the use of transmission control protocol (TCP). Data applications such as file transfers and e-mail are less sensitive to the time it takes for this to occur, but real-time voice traffic cannot tolerate this delay. In addition, VoIP networks use connectionless transfer protocols such as user datagram protocol (UDP) that do not guarantee delivery at all.

Network congestion results in dropped packets. In current IP networks, all voice frames are treated like data. Under peak loads and congestion, voice frames will be dropped equally with data frames. Data frames are not time sensitive and dropped packets can be appropriately corrected through the process of retransmission. Lost voice packets, however, cannot be dealt with in this manner. Lost packets mean lost voice information.

Echo Compensation

Echo in a telephone network is caused by signal reflections generated by the hybrid circuit that converts between a four-wire circuit (a separate transmit and receive pair) and a two-wire circuit (a single transmit and receive pair). These reflections of the speaker’s voice are heard in the speaker’s ear. Echo is present even in a conventional circuit-switched telephone network. However, it is acceptable because the round-trip delays through the network are smaller than 50 milliseconds and the echo is masked by the normal side tone every telephone generates. Echo becomes a problem in VoIP because the round-trip delay through the network is almost always greater than 50 milliseconds.

QoS in IP networks and the public Internet is expected to improve with innovations in routing protocols and improvement in physical networks that carry IP traffic. However, the poor quality of telecommunication infrastructure and congestion in the IP infrastructure in developing countries would continue to haunt voice over packet networks for a long time to come.