The Killer App for Wi-Fi

The Killer App for Wi-Fi


Everyone looks for the Killer App. At 7signal, we not only found the “Killer App for Wi-Fi”, but also the “App that Kills Wi-Fi.”

It turns out that the Killer App is voice. Enterprises are starting to realize it is expensive and undesirable to make the desk phone the primary means for office communications. People want to be mobile and they want to bring and use their own device. The unified communications vendors are providing the applications for smartphones to provide the productivity features people are used to with their desktop phones.

This suits the enterprise CIOs – they want to stop paying for phones and the infrastructure required to support wired communications. In fact, the desktop phone is the last holdout to completely untether the desktop and keep the need for cabling. The enterprise can shift to one in which the WLAN is the main communications network. In addition, if the WLAN can support voice well, the need to provide mobile coverage in difficult spots goes away.

The challenge with Voice over Wi-Fi (VoWiFi) replacing the desk phone is Quality of Experience (QoE). IP phones on the desk had the same issue, but well designed and managed wired networks addressed the quality concerns. A well designed and managed WLAN can also allay any fear of migrating to a wireless-centric architecture. But an unmanaged WLAN carrying voice can exacerbate existing Wi-Fi challenges and create a poor experience for everyone. For example, voice as an application does not create a lot of bandwidth, but each conversation creates more packets and more attempts to access the shared spectrum. Many times an extra SSID is used for voice, increasing management overhead. Roaming users may also impact the processing power of WLAN access points.

None of these challenges are insurmountable when you can measure performance and understand the network. For Voice over WLAN (VoWLAN), it is imperative to have a Quality of Experience metric such as the Mean Opinion Score (MOS). MOS depends on many factors but provides a useful way to characterize the QoE:

MOS Quality Impairment
5 Excellent Imperceptible
4 Good Perceptible but not annoying
3 Fair Slightly annoying
2 Poor Annoying
1 Bad Very annoying


An accompanying blog by Martin Varella discusses MOS and the Quality of Experience further.   7signal has long been successful in helping institutions such as hospitals where the workforce is very mobile and the WLAN carries voice and data. The MOS metric has been an important Key Performance Indicator for analyzing the network.   As more enterprises move to a similar wireless-centric architecture, make sure MOS plays a role in your Wi-Fi performance management.

2 thoughts on "The Killer App for Wi-Fi"

  1. Avatar Devin Akin says:

    Consider the unbridled interference caused by VoIP desk phones that are using Bluetooth or proprietary FHSS for their headsets. It trashes the 2.4G band, and administrators rarely pull out a spectrum analyzer to see why performance in that band is so poor. Further, VoWiFi is hard on an AP because of WMM. Since the voice access category (queue) has a small contention window, the collisions increase logarithmically as more phones are added. For this reason, high-density VoWiFi is a VERY BAD IDEA.

    1. Avatar Jeff Reedy says:

      Devin – good points there. I agree with you that voice over Wi-Fi has challenges, particularly in the 2.4G band. It helps to have the benefit of spectrum analysis, whether it is built into a sensor or an AP, and it is essential to have a proactive Wi-Fi performance management approach that provides a Voice MOS measurement so you can understand the network’s ability to handle voice. And certainly VoWiFi is quite feasible in the 5G band, particularly in a general corporate BYOD environment where enterprises are trying to cut the cord. Good design by guys like you and continuous verification are the keys to successful operation.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.