pointer

The Basics of Wireless Channel Aggregation

You’ve bought the controllers, the access points (APs), the VoIP Phones – the entire infrastructure required for wireless – and while all of these pieces worked well together during the testing phase, when people in your company start using wireless, a variety of problems begin to pop-up. Problems such as: dropped VoFi calls, low signal strength, and connectivity issues.

So the question now is: why does everything seem fine during testing, but malfunction under real world conditions?

One of the answers may be that you do not have adequate visibility of your WLAN set up. Specifically, in today’s highly mobile environment, it is not enough to monitor each channel of a WLAN independently. As wireless users move around, or roam, they typically transition from one AP to another and from one channel to another; and, it is often these transitions which cause serious issues for wireless-based application use.

To adequately monitor, analyze, and troubleshoot your WLAN you must collect data across multiple channels simultaneously for visibility when users roam. At WildPackets we call this channel aggregation. With traditional wired network analysis, there’s only one “channel” in use, so channel aggregation is a function that is unique to WLAN analysis.

Wireless channel aggregation is relatively straightforward, if not widely available. Most WLAN analysis products scan through the channels of interest to compile overall statistics for WLAN performance. Scanning creates gaps in the data, and the more channels scanned the bigger the gaps. This technique, though adequate for generating statistics, falls far short for detailed analysis and root-cause resolution. For example, let’s say your WLAN uses 10 channels. Because the data capture is only on one channel at a time, you are only receiving 10% of the data on any single channel. This means that you’re blind to 90% of the activity on that channel, woefully inadequate for detailed analysis. This is especially true for analyses that involve critical timing, like roaming. Roaming should take place within a few hundred milliseconds, or less. If you go back to our 10 channel example, and assume that the dwell time on each channel is 0.5 seconds, then you will have a 5 second gap between times when data is collected on any given channel, and this is obviously inconsistent with the millisecond granularity needed to detailed timing analysis.

Wireless channel aggregation gets us past these problems. All that is needed is an analysis solution capable of receiving data from multiple channels simultaneously, and enough wireless adapters to cover each of the channels to be analyzed. The data is then captured into a single analysis session, and you have all the data from all the channels and can perform any level of detailed analysis that’s required.

Now that you have a better understanding of wireless channel aggregation, you can be better informed when looking for solutions capable of true root-cause wireless troubleshooting. At a minimum, we suggest that these tools be able to capture wireless packets from multiple channels simultaneously (without scanning) and measure vital statistics on each channel separately, as it provides you with a better understanding of the activities happening on each channel. Additionally, having a wireless channel aggregator that calculates latency of devices roaming between access points is also very helpful.

For more details on what is out on the market, and the hardware and software needed to perform wireless channel aggregation, check out this whitepaper by the Certified Wireless Network Professional titled The Triple Blendy.

One thought on “The Basics of Wireless Channel Aggregation

  1. network management

    Wireless channel aggregation is really an impressive and important tool in making sure that everything is fine within your network system. It’s a in depth root-cause troubleshooting that is capable of capturing and measuring vital statistics and wireless packets from multiple channels simultaneously.

Leave a Reply