The Basics of Wireless Roaming Latency Analysis

Your new wireless network is fully deployed, and it worked well in the testing phase, including new services like voice over Wi-Fi (VoFi). But after only a few months of “real world” testing, performance issues and dropped connections are already plaguing you. As we discussed in our blog a few weeks back “The Basics of Wireless Channel Aggregation,” one way to investigate some of these issues includes the ability to have a complete view of your WLAN, not just a single channel view. When working with wireless it’s important to remember that users are mobile, and when they move around their network connection moves from on access point (AP) to another, and typically from one channel to another. This creates unique problems with both applications and VoIP calls, especially if the mobile devices are also using a VPN. Without full visibility into the WLAN, network engineers cannot determine if the problem stems from the network, the device, or the application.

“The Basics of Wireless Channel Aggregation” gives insight into why it is important to aggregate multiple channels simultaneously in a wireless environment to gain better visibility. In today’s post, we’ll take it a step further, discussing one of the major pain points within wireless systems: roaming latency, and how wireless roaming latency analysis can help network engineers determine what’s causing latency issues.

What is Roaming Latency?

Roaming Latency is the amount of time it takes for a wireless device to move from one AP to another, also known as re-association. However, there is no standard measurement for roaming, and several approaches can be taken, all equally valid, depending on the issue at hand. At the shortest end of the spectrum, one can measure just the time difference between the last packet on one AP and the first packet on the next AP. It can be argued that this is the true latency, but it’s really the application that’s currently in use that matters for many end-user problems, especially with VoFi. A more conservative, but somewhat more real-life measurement, is to look for the last data packet on the previous AP and the first data packet on the new AP. Better yet, being armed with both measurements allows you to determine your base network latency as compared with your overall application latency. As is so often the case, more information will likely lead you to the core problem more quickly.

Wireless VoIP on mobile devices is one of the more interesting applications when it comes to roaming latency, because the real-time nature of VoIP data makes this data type particularly vulnerable to network latency, regardless of the source. More than a few hundred milliseconds of delay between packets leads to unacceptable quality and can even result in dropped calls. The ITU recommends that the total latency for VoIP not exceed 150ms, so that means that wireless roaming latencies had best be well below that value.

Roaming Latency Analysis

The key to performing roaming latency analysis is having synchronized data for each of the channels the roaming device is using. To get this, you need to be performing simultaneous captures on all channels in use, so you can see the complete roaming transaction. Once you have the synchronized data, it’s a relatively simple job for software to categorize the information into individual communication streams, and then “do the math” to make the various types of latency measurements discussed above. To fully characterize the issue, you’ll want roaming data from multiple wireless clients using the same APs to be able to narrow down the problem to a specific wireless client or an overall AP. With that determined, you can now dig into the specific device to determine exactly what is going wrong.

As you begin to rely more heavily on your WLAN, or if you already do rely heavily on it, you need to make sure you have the appropriate analysis capabilities in place to troubleshoot all levels of problems, including complex issues like wireless roaming, even if there was no trace of these issues in the testing phase. For more information on getting the most out of your wireless network, check out our whitepaper.

One thought on “The Basics of Wireless Roaming Latency Analysis

  1. drbaker

    Very well written!
    Wireless VoIP is a pain in the backside. The fact is that no matter how well you design a mesh network; roaming is dependent on the CPE. 802.11s sets the stage for standardizing mesh networks and queued packet forwarding from one AP to the next. But in a mesh environment, there is not a standard for CPE roaming and every manufacturer has their own proprietary formula as to when to reassociate. There is even one popular brand of wireless VoIP that I see sitting on shelves almost all the time. The reason is that the manufacturer has never programmed the phone to roam. Every change from one AP to the next is done as if it were nomadic roaming. Even on the company website they boast designed to take advantage of a single Access Point; as if it were an enhanced feature. This is why Zibby in the second to last paragraph writes “To fully characterize the issue, you’ll want roaming data from multiple wireless clients using the same APs to be able to narrow down the problem to a specific wireless client or an overall AP”

Leave a Reply