RF Detection: Solving the Next Great Wi-Fi Challenge
Radio-frequency interference is the carbon monoxide of the wireless world – invisible, tasteless, odorless, arriving without warning, and with the ability to really ruin your day – unless you know the symptoms and have the right tools at your disposal. And that’s just what we have for you in this new series of articles from noted wireless expert Craig Mathias of Farpoint Group.
Let’s start this time with the key reason why interference is such a problem. And the reason here is that interference is just like network congestion in terms of its impact – what we see is a reduction in capacity.
Detecting an interference problem
Network congestion can be dealt with by adding more access points, using traffic-management techniques like prioritization, and even banning non-essential traffic or any traffic that’s outside of local usage policies.
In the case of interference, however, potential solutions can be much more challenging.
So, how do you know if you have an interference problem? As was mentioned above, significant interference most often shows up as a reduction in throughput and capacity. Users are often baffled, as they will normally see more than adequate signal strength (four or five bars), but network performance is sluggish.
These bars of signal strength are calculated based on the relationship between the AP and the user device – one can see a very strong signal, but interference prevents successful transmissions at least to some degree.
The first step, then, is to determine if the observed reduction in throughput is due to Wi-Fi congestion, capacity issues elsewhere within the network, equipment misconfiguration or even failure, or other elements under control of network operations. The network management console, network logs, third-party monitoring and diagnostic tools, and related systems and techniques are very useful here. And it’s easy to check for and (usually) remediate any discovered issues.
The spectrum analyzer
RF interference is a bit more complex to detect, though. If capacity seems OK within the network itself, the next step is to check the airwaves themselves. There’s a special tool that can be applied here, the spectrum analyzer. For many years, these devices were so expensive and complex that only engineers developing new products could afford them.
Necessity being the mother of invention, though, a number of firms got the bright idea to build PC-based but otherwise standalone spectrum analyzers that examine only the 2.4- and 5-GHz. bands used by wireless LANs.
This development was, in my humble opinion, one of the truly great leaps forward in the history of WLANs, enabling mere-mortal IT and network managers to examine the airwaves at the physical layer and determine, with no special training, if interfering signals are present.
The next great leap forward was to build spectrum-analyzer functionality into the chipsets used to implement access points, meaning than an AP could now, on a full or part-time basis, serve as an spectrum analyzer. Network-wide, facility-wide, and even multi-site spectrum analysis became possible, enhancing the productivity of operations staffs that no longer needed to grab a notebook PC with spectrum-analyzer capability and walk the halls.
Such is still often desirable in localizing an interferer, and we’ll return to this mode of operation in a subsequent article. But if you’re interested in a picture of what’s really going on in your local radio environment, and even identifying and localizing a wideband jammer, this is the way to go.
There’s still a good deal of debate over whether spectrum-analysis functionality should be integrated into APs, or whether a separate overlay Wi-Fi assurance product should be used. Both techniques work very well, so the precise choice depends upon local goals, objectives, budgets, and other operational requirements. As long as facilities have some form of interference detection, the precise implementation is a secondary consideration.
You see, with the continuing explosion of Wi-Fi systems, competing unlicensed services like cordless keyboards and mice, and potentially dramatic increases in demand as the Internet of Things gets going, interference can no longer be ignored. It is, in fact, likely to get worse.
Next week, I’ll look at sources for interference. Being able to detect the problem is one thing, but understanding where the problem is coming from is another battle.
All Posts In This Series: