Ask the Expert How to: Give Voice to Wi-Fi

Can voice over wireless LAN (VoWLAN) become just as ubiquitous? David Confalonieri, VP of Corporate Marketing for Extricom, outlines the key considerations for making VoWLAN a reality.

We have all seen and experienced two inexorable trends in communications: ‘cutting the cord’ and ‘convergence of voice and data’. Cutting the cord by leveraging wireless technology means productivity and user convenience, with examples spanning cellular phones, PDAs and wireless-enabled laptops. Convergence promises efficiency and lower total costs stemming from a multi-purpose infrastructure.

Now these two trends are themselves converging in the form of voice over wireless LAN (VoWLAN). With the WLAN business case already strong for data applications, this new, highly sought-after voice use promises to make the value equation even stronger. Ultimately, the ideal goal is to build a single, multi-purpose WLAN infrastructure, capable of supporting voice, data and video applications – the vaunted triple play. In such a scenario, WLAN is expected to be deployed ever more broadly and deeply across the enterprise.

So where are the potential pitfalls? To find them, you need look no further than the IEEE 802.11 specification itself. WLAN, and the 802.11 specification it’s based on, was primarily conceived for data communications, which are not real-time in nature and can tolerate instabilities in a wireless link without discernable performance impact to the user. Real-time applications such as voice and streaming data are a completely different matter, requiring far greater wireless performance. For that reason, building a WLAN to enable voice applications requires close attention to four inter-related design considerations: coverage, capacity, mobility and quality of service (QoS). And if you already have a WLAN deployed, don’t assume that what worked for bursty data will work for the new demands of voice.

• Coverage: Compared to data users, voice users will ratchet up the coverage requirement in three ways: location, strength and uplink stability. Voice users hold conversations in locations where you don’t usually transmit data – hallways, stairwells, even the bathroom. This means that coverage not only has to be broadened to cover more areas of a building, it must also be strengthened to ensure wire-like voice quality everywhere. The objective, then, will be to build out interference-free coverage that is ubiquitous, in spite of the fact that there are very few 802.11 channels to work with. Equally as important, the uplink (i.e. transmission from the phone to the VoWLAN) needs to be far more stable than for data. Interruption of a Wi-Fi connection for even a fraction of a second is a minor annoyance for a data user who is accessing a website. For a VoWLAN customer, however, it can mean dropping an important call, leaving a frustrated and unhappy end-user.
• Capacity: VoWLAN is attractive to enterprises because it allows the same WLAN system to be used for both voice and data. Such expanded use, however, requires increasing the overall capacity of the system, as measured by the number of users who can connect to a specific wireless channel at any one time. This is particularly critical in light of the fact that voice transmissions are burdened by high packet overhead, which limits the number of simultaneous voice calls that can be accommodated.
• Mobility: The great irony of Wi-Fi is that it is a specification for portability, not mobility. But once the cord is cut, VoWLAN users tend to be mobile users. And since voice is a real-time application in which packets must be sent at regular and consistent time intervals, it will not tolerate packet processing delays that arise when a phone moves from one access point to another in traditional WLAN systems. Therefore, employing technologies that bring truly seamless mobility to Wi-Fi will be critical to the success of VoWLAN.
• QoS: WLANs operate in a shared medium, in which both voice and data users must contend for access to the network. If this contention between voice and data is not managed, it winds up degrading system performance for all. When deploying a VoWLAN, some means of addressing the QoS issue is essential. The goal should therefore be to identify and employ those mechanisms that provide the highest level of predictability and guarantee of quality, not just at the logical layer, but (perhaps even more importantly) at the physical access level.

Vendors have developed a variety of architectures to address the challenges of VoWLAN. These can generally be organized into two types, commonly known as cell-based and the channel blanket approach. The cell-based architecture is the more traditional approach taken for data-centric solutions, and is geared to optimize either coverage or capacity. In this architecture, access points (APs) are assigned a specific radio channel for communications, and are then distributed to form a honeycomb coverage pattern.

The channel blanket approach is targeted at the real-time application environment of voice, delay-sensitive applications, multi-tenant/multi-device requirements, mobility demands and simplicity of design and implementation. In this type of system, a central switch governs a set of access points so as to create extended zones of coverage for every available channel, by using the channel(s) at every AP.

The potential for VoWLAN is significant, and it will have as great an impact as mobilizing computing has had on the business. The key to success will be to recognize the possible architectural roadblocks in the WLAN infrastructure, with particular focus on those fundamental wireless transport elements that have often been ignored when data-centric, hotspot-type WLANs were the primary goal.