Last week, Mojo introduced the C-130, the industry’s first tri-radio Wave 2 access point with a 2x2 802.11ac third radio. (In an earlier post, Hemant Chaskar explained the security benefits of a 2x2 third radio.) In this post, I will show how that third radio provides significant performance improvements, especially for enterprises running voice and video over WiFi.
The Importance of Voice and Video Applications
As enterprises expand their global footprint with employees in different locations across the world, and often telecommuting from home, the importance of collaboration tools like GoToMeeting, Skype for Business, and Google Hangouts has never been more significant. K-12 and higher education are also relying more on collaboration tools and delivery of multimedia content, as they adopt digital classroom strategies and integrate digital learning. Given that WiFi is the de facto connectivity standard for devices in these environments, WLANs need to be increasingly capable and resilient.
How Background Scanning Hurts Voice and Video
It’s well known that wireless architects must take extra care when designing for voice and video applications. These applications typically generate constant streams of data, and any small delay in the network can degrade or even ruin the end user’s experience.
Background scanning is a technique in which the radio of an access point momentarily stops serving clients in order to scan non-operating channels for RF information. During these brief periods, no data traffic is exchanged. The scanning information is then used by the wireless system to detect security threats and to optimize the operating RF parameters with radio resource management (RRM).
To measure the impact of background scanning on voice, our engineers performed a simple test. VOIP calls were set up between 16 iPad Minis, and the mean opinion score (MOS) was measured. An MOS is a numerical rating of the quality of voice. An MOS of above 4 is considered to be good, and anything below 4 is not acceptable.
|Background scanning||MOS (mean opinion score)||Acceptable?|
You can see how background scanning drops the MOS from 4.06 to 2.7, far below acceptable voice quality levels. Even though these scans are 5-10 seconds apart and the time spent on each scan is only 100 ms, the impact on VOIP clients is significant. We saw very similar results with other real-time streaming applications like web sharing, video chatting, etc.: background scanning noticeably hurts user experience.
When “Intelligent Scanning” is Dumb
This is why most vendors recommend turning off background scanning when running voice or video on a wireless network. In fact, many vendors do what they call “intelligent scanning,” which simply turns off background scanning when voice or video are detected! That may ensure that voice and video applications do not suffer, but security does.
No More Trade-Offs
In today’s WiFi world, with the proliferation of mobile devices, BYOD, collaboration tools, etc., it’s a safe bet that voice and video applications are always running in any wireless network.
The traditional approach of using the same radio for both client access and background scanning will no longer work: the constant presence of voice or video means that radios are never going to be able to perform background scanning. This creates a security hole in your network.
Also, various RF optimization functions like channel selection, power selection, load balancing, smart steering, will fail to operate properly, leading to suboptimal radio resource management (RRM) decisions. More on this in a later post.
Having a dedicated third radio for performing full-time security scanning ensures that the other two radios are dedicated to serving clients. This ensures unfaltering application experience without compromising on security, and prevents the access point from making suboptimal RRM decisions.
In short, improved performance is another reason you should check out our newest access point.