Voice is just one service of many being rolled out in enterprise networks, branches, and home offices, but voice has its own peculiarities! Specifically, voice-over-IP (VoIP) is a real-time service and is sensitive to delay and packet loss. VoIP deployment may be complicated by the need to maintain legacy equipment such as PBXs, vendor-specific phones, and so on. This article from Informit examines elements of VoIP QoS.
VoIP is a special area because it bridges the old with the new; that is, time-division multiplexing (TDM) and IP. VoIP can be deployed end-to-end using IP phones at each end, or it can be phased in over a period of time. The latter is usually referred to as a migration strategy, and is beginning to get a lot of traction with service providers—not just for voice service, but for others such as data and video. If a phased approach is used, legacy phones can be used in conjunction with special-purpose hardware. The latter can be PBX-based, if necessary, to extend the lifecycle of the PBX equipment.
An important point about VoIP deployment is that it can be configured to use WAN links as far as possible, spilling over into PSTN links only when all allocated WAN capacity has been exhausted. This arrangement allows for toll savings up to the limit of the WAN allocation.
VoIP may also be deployed for functional rather than financial reasons. Many large enterprises can easily negotiate reduced tariffs for telephone service. But VoIP can allow for useful applications to integrate with the telephony function. This is the rich area of computer-telephony integration (CTI). Examples of applications are call centers that allow customer details to be accessed by agents while they're still on the phone. All the applications use the LAN as the transport mechanism.
Let's take a look at some of the components of VoIP QoS.
We typically encounter the following types of delay components:
- Coder delay: Analog-to-digital speech conversion and PCM compression
- Packetization delay: Time to fill a packet payload
- Serialization delay: Time to push a packet payload onto the wire
- Output queuing: Scheduling a voice packet out of device queues
- WAN delay: Transmission delay across the WAN
- Dejitter delay: Smoothing the inter-arrival time of voice packets
Each of these delay components contributes to the overall budget; the design engineer must mix-and-match them appropriately. If the engineer gets it wrong, the network may exhibit variable delay, giving rise to unsatisfied customers.
Perhaps the most subtle delay component is that of output queuing, which relates to basically getting voice packets through intermediate routing devices. The complication arises because routers are used for data and voice packets; in other words, non-real time and real time.
Let's say that three packets are in transit through a router. The data packet is already being queued prior to be being pushed onto the line. Even if a voice packet has just arrived behind this data packet, the queuing already in progress will not stop. However, the next voice packet will skip the data packet ahead of it. This strategy helps to preserve the high priority of voice packets as they move through the network.
If the router becomes congested, it may start to drop packets. This is not such a disaster for data packets (which can be retransmitted) but it has a serious impact on a voice service. As we've seen, voice service is unforgiving of both delay and packet loss. The same is true of advanced surveillance projects such as planetary exploration (or possibly even in the unmanned drone planes operating over Afghanistan). Clearly, this class of application has a one-way data stream—we're looking at them and not vice versa! Data is acquired and transmitted back to the receiving station; by the time it's received, it can't be retransmitted due to speed of acquisition, limited storage, and so forth. Given the critical nature of these applications, it's commonplace to add extra redundancy to the data prior to transmission, which allows the receiver to correct any errors that occur in transit. This is the interesting area of error correction as opposed to error detection. Correction schemes allow receivers to fix errors, whereas detection schemes usually require retransmission.
The funny thing about QoS is that it tends to be known only after the fact; you use some application and you experience the QoS as you use it. The same principle applies to VoIP—poor QoS shows itself in excessive delay and possibly lost packets. In a more general sense, we can say that the major determinants of QoS are as follows:
- Throughput: Amount of available bandwidth between two network points of interest
- Availability: Fraction of time during which service is available between two points
- Delay: Length of time taken to transmit data from one point to another
- Delay variation: Difference in the delay for two packets
- Loss: Ratio of packets received to packets sent
These simple elements are sufficient to put together very complex service level agreements (SLAs). Some service providers now supply stringent SLAs with proactive tariff rebates—all part of the increased importance of service differentiation. As time goes on, we can expect to see this enlarged role of SLAs emerge increasingly within enterprise networks.
Read more about VoIP QoS at Informit.