In-Depth

Monitoring VoIP Communication: Five Key Measures

Five key measures to monitor to address infrastructure problems.

• By Charles Thompson
• 05/30/2006

VoIP traffic is vulnerable to network delay, jitter, and packet loss, making it a difficult network technology to manage. When VoIP contends for bandwidth with other applications, voice quality can suffer from increased static, jumbled conversations, or dropped calls. Regardless of cost savings, users are not tolerant of these effects. They have a firm expectation of quality for phone communication through the many years they have depended on the traditional land-line telephone service.

Regardless of the fact that VoIP can be tricky to manage, businesses are migrating to VoIP to reduce telecom costs and improve business operations. Since VoIP is susceptible to network conditions and users have high expectations of VoIP performance, VoIP issues are highly visible, as is the IT staff tasked with solving those problems.

A mature analyzer that can provide detailed insight into VoIP communication is needed to successfully manage VoIP. It is not sufficient to rely strictly on the “monitoring tools” that come standard with the VoIP equipment. These tools provide little information about the VoIP communication running over the network. Utilizing an independent, third-party analyzer that can detect and resolve audio problems, provide real-time call detail records, and display VoIP connection dynamics is the best option.

But any analyzer is useless if you don’t understand what to monitor or how to interpret the VoIP activity. The following identifies which network conditions to monitor to predict VoIP quality, how to accurately interpret VoIP performance, and other key analyzer features.

Measure #1: Network Jitter and Delay

Managing VoIP quality is primarily a matter of minimizing network delay and jitter (which is variability in packet arrival time), because codecs require a steady, consistent stream of packets to provide adequate playback quality. Packets arriving too late or out of order will result in jerky, jumbled playback (referred to as jitter).

While the jitter buffer on a VoIP phone can mask mild delay and jitter problems, severe jitter can overwhelm the jitter buffer, resulting in excessive delay. Increasing the buffer size can help, but only to a point.

Measure #2: Packet Loss

Another phenomenon to monitor is packet loss. Networks tend to either sporadically drop single packets (these periods are called “gaps” in packet loss) or large numbers of packets in a “burst.” Packet loss concealment techniques typically have no trouble handling packet loss during gap periods; it is the sustained bursts which are difficult to manage.

Measure #3: VoIP Quality

It is easier to manage what can be measured. Managing VoIP, therefore, requires hard numbers beyond subjective user assessments of quality (although these are obviously important as well). Beyond monitoring the network parameters, having an overall quality score such as a Mean Opinion Score or R-factor score can also be a useful VoIP network health index.

VoIP monitoring tools calculate the MOS and R-factor scores using a formula known as the E-model. An analyzer calculates how much the various impairment factors (such as codec, jitter, delay, and packet loss) affect the typical user’s perception of call quality.

VoIP quality may be degraded when any of the following conditions arise:

• MOS score falls under 3.5
• Factor score drops below 70
• Unidirectional delay exceeds 150 ms

Call quality may be perceived differently depending on the environment. For example, a user making a call from a factory floor may tolerate lower quality than a user making a call from a conference room.

Measure #4: Call Progress

When VoIP performance is unacceptable (such as when a user can’t get a dial tone, or there are excessive delays in ringing the other party’s phone,) examining a display that shows how the call is progressing between the parties and the call manager or gateway can reveal the problem.

A connection dynamics display highlights which party isn’t responding, or which party is responding slowly. Ultimately, this capability can dramatically cut down troubleshooting time.

Measure #5: Normalcy

Although metrics provide a good indication of current VoIP performance, understanding how VoIP normally performs over the network is helpful.

An analysis tool that tracks, stores, and analyzes long-term activity will define what is considered “normal” for a particular VoIP environment. This insight will help you intelligently configure alarms on the monitoring tool to notify you when VoIP performance deviates from the norm—allowing you to become aware of problems as they arise so you can immediately resolve them.

It is also important to maintain a database of call detail records (CDRs) from which you can generate reports for management or service providers.

Conclusion

By closely observing the network conditions that affect VoIP, you can begin to address developing infrastructure problems before they lead to user complaints or downtime. Understanding what to monitor and having a sophisticated VoIP network analyzer makes this crucial task much more manageable.