Breaking the sound barrier.
The transfer of voice traffic over packet networks, and especially voice over IP (VoIP), is rapidly gaining acceptance. Many industry analysts estimate that the over all VoIP market will become a multibillion dollar business within three years.
While many corporations have long been using voice over Frame Relay to save money by utilizing excess Frame Relay capacity, the dominance of IP has shifted most attention from voice over Frame Relay (VoFR) to VoIP. Voice-over-packet transfer can significantly reduce the per-minute cost, resulting in reduced long-distance bills. In fact, many dial-around-calling schemes available rely on VoIP backbones to transfer voice, passing some of the cost savings to the customer. These high-speed backbones take advantage of the convergence of Internet and voice traffic to form a single managed network.
This network convergence also opens the door to novel applications. Interactive shopping (Web pages incorporating a "click to talk" button) is just one example, while streaming audio, electronic white-boarding, and CD-quality conference calls in stereo are other interesting applications.
Along with the initial excitement, customers are worried over possible degradation in voice quality when voice is carried over these packet networks. Whether these concerns are based on experience with the early Internet telephony applications or based on understanding the nature of packet networks, voice quality is a critical parameter in acceptance of VoIP services. Voice quality is not only in the user's interest. It is also in the service provider's interest. Studies done on cellular networks have shown that as voice quality increases, the time users spend on the network also increases. Because users are often billed per use, quality increase translates directly into bottom-line gains for service providers.
HOW DO YOU MEASURE VOICE QUALITY?
With all the factors affecting voice quality--including latency, jitter, packet size, and packet loss--many people ask how to measure voice quality. Standards bodies like the International Telecommunications Union are continually addressing this issue and have already delivered two important recommendations: P.800 (MOS) and P.861 (PSQM). P.800 deals with defining a method to derive a Mean Opinion Score of voice quality. The test involves recording several preselected voice samples over the desired transmission media and then playing them back to a mixed group of men and women under controlled conditions. The scores given by this group are then weighted to give a single MOS score ranging between one (worst) and five (best). An MOS of four is considered "toll-quality" voice.
P.861 Perceptual Speech Quality Measurement tries to automate this process by defining an algorithm through which a computer can derive scores that have a close correlation to the MOS scores. While PSQM is useful, many people have voiced concerns over the suitability of this recommendation to packetized voice networks. It seems that PSQM was designed for the circuit-switched network and does not take into account important parameters, such as jitter and frame loss, that are only relevant to VoIP.
As a result of the PSQM limitations, researchers are trying to come up with alternative, objective ways to measure voice quality. One proposal is the perceptual analysis/measurement system (PAMS) developed by British Telecom (BT). Tests conducted by BT have shown good correlation between automated PAMS scoring and manual MOS results.
Sometimes, you have to be your own judge. Testing tools are available that extract individual voice calls and then decompress the captured data using the detected compression method to allow listening to the actual voice recording. By repeating this process at different points along the voice path, it is not only possible to get a sense of quality but also to determine at what point along the network voice-quality degradation occurs.
SIMULATING NETWORK EFFECTS
With so many pitfalls, it is often desirable to simulate the target network before deploying actual equipment--or to test products still in the development stage before they ship to customers. Such simulation allows vendors to gauge the effect of a suboptimal communication link on voice quality and configure critical parameters, such as jitter buffers, before going into the field or sending products to market.
One company using VoIP testing equipment to develop quality product lines and decrease time to market is 3Com Corp. According to Bert Davenport, a senior product manager at 3Com, the company used a powerful combination of VoIP testing tools from RADCOM Equipment to develop CommWorks, a carrier-class IP telephony platform that is currently shipping to select customers. RADCOM's AudioPro VoIP Analyzer was combined with RADCOM's PrismLite, a portable and integrated multitechnology protocol analyzer, to simulate, monitor, and analyze VoIP traffic.
Davenport says that the RADCOM equipment has been extremely helpful in a number of areas, especially in terms of eliminating jitter from the 3Com system, which enables service providers to migrate voice, data, and video traffic onto a single platform to reduce costs and simplify management. "AudioPro has really helped us to design a very robust carrier-class system," he says. "AudioPro allows you to see all the packets and the inner-packet delays, and it does a very nice job of showing you--on a per-stream basis--what the audio stream really looks like on the network."
One feature that Davenport particularly likes is AudioPro's ability to play back actual voice streams. "Sometimes it's nice to hear the audio," he says. "The AudioPro can translate what you have into an audible .wav file. In combination with the PrismLite, it's great to see the protocol. If we had a bad protocol or a bad protocol message, that's really nice to see"--and hear.
The PrismLite's portability, combined with AudioPro's simple GUI, has extra advantages.
"We've used it not only for testing but also for customer demonstrations," says Davenport. We've taken it into the field to show customers the quality of our system. It's especially nice when you have a lot of data streams, because if you just have one callup, you can do this stuff with a regular protocol analyzer. But if you bring up one span, or five spans, or 10 spans of calls, you've got a lot of traffic on your network and you can't separate out each stream easily. The AudioPro separates out each stream as an independent entity."
The future of IP telephony is a converged network, says Davenport, where a single network is going to be carrying voice, data, and video traffic. "It's not going to happen overnight," he says. "But, eventually we're going to get to the point where the entire infrastructure, or a significant portion of it, is IP-based."
And that's great news for companies like 3Com that are building next-generation VoIP solutions to provide integrated products and services to users. It's also great news for behind-the-scenes testmakers like RADCOM, who are building not only valuable products but also a loyal customer base.
"At the time we made our [AudioPro] purchase, there was nobody else in the market," says Davenport. "There may be a few other players in the market now, but we haven't really looked at them. We've been too happy with the RADCOM AudioPro."
Circle 268 for more information from RADCOM Equipment, Inc.
Yuval Boger, vice president, voice test and management, RADCOM Equipment, Inc., contributed to this article.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Company Business and Marketing|
|Comment:||Voice over IP is increasingly popular, but voice quality is difficult to measure, although the ITU is addressing the issue with the P.800 and P.861 standards.|
|Date:||Jan 1, 2000|
|Previous Article:||Cache and carry.|
|Next Article:||Shouldn't we be talking about voice over anything?|