A lot of companies these days are saying “hey, connect that via wireless, connect this via wireless. You don’t need a wire.” As more and more devices come on board, as more and more devices demand connectivity, it becomes increasingly difficult to manage the network to make sure that you have proper coverage, don’t have channel contention, and everybody is getting the best performance possible. All too often it’s just kind of a “throw a dart at the dartboard” and hope like hell you’ve got the right coverage.
Wireless has become just as demanding, just as ubiquitous and just as pervasive as wired networking. As such, the planning, configuration and testing for implementing a wireless network is just as critical. The challenge is that the technology is still not, what I would call, as mature as the wired network is currently.
Many considerations must be made
There are many considerations that come with configuring a wireless network and making sure that it’s fit-for-purpose. What I mean by that is understanding who the user base is, what business purposes or what functional purposes the network is going to serve. Will they be trying to make wireless VoIP calls over the network? Are they trying to stream video? Do they just need it for web and email services? Will there be any automation involved?
Wireless technology not as mature as wired
Wireless networking is a shared medium and people forget this. So as you add more clients, that available bandwidth gets divided amongst those clients — slower network. If you take advantage of the newer technologies, you get higher throughput, higher transfer rates — faster network . . . in theory. However, you also get fewer clients. Either way, it’s a balancing act of “pay me now or pay me later.”
Advertised wireless throughputs may not materialize
Another challenge is that these new technologies claim these massive throughputs — up to 1.3 gigabit throughput capable on the right client, on the right device, etc. But that’s all testing done in a laboratory. Here, in reality, the real world throughput, even under ideal conditions, might not exceed 100 meg in some instances and that’s using 802.11ac, 80 megahertz-wide channels!
Technology at the mercy of its environment
The technology itself is still at the mercy of its environment and it always will be. So, if you’ve got a smart home, you’ve got microwaves, cordless phones, baby monitors — all of those things impact the 802.11 communication protocols. If you’re in a smaller home, that is say 2,000 square feet, a single access point (AP) may work just fine — no big deal. But when you get into larger homes, more access points are better, right? — Um, no. I was talking with a client down in Highland Park. He has a 15,000 square foot home and is having coverage problems even though there are 20 access points in the house. Turns out he has walls made of concrete with steel reinforcement in them. A shining example of technology at the mercy of its environment.
2.4 gig space is congested
The 2.4 gig space is very congested. Everybody wants to live there because that was the first technology and the 5 gig technology has got its own set of unique challenges. In a smart home, you may have ZigBee or Z-Wave, both of which operate in the 2.4 gig spectrum which means that you have channel overlap and channel contention which impacts performance.
Planning is the byword
So it really comes down to proper planning. This applies both in enterprise and automation environments. You’ve got to have an understanding of what the coverage needs to be and who the user base is going to be. If you don’t properly plan, you are likely to have issues roaming from one access point to another. Your devices will disconnect from the network and then have to reconnect when it should occur automatically, transparent to the user. That’s where the wireless site surveys come into play.
Surveys are critical
All this is good information, but so what? Well, if planning is the byword then surveys are critical. Taking all these things into consideration during the planning phase to ensure you have a solid implementation does you little good without a survey. You’ve got to have some sort of plan, some sort of approach to ensure the proper placement of access points.
To determine optimal AP placement, you should do a preinstall survey where you take an access point, power it up and take transmission readings or a predictive survey in which you model the space in a computer environment. Otherwise, you’re going to have an office where connectivity is constantly dropping. This will make work productivity come to a halt, or in an automation solution, you’ll have video streaming that is pixilated and halting or audio that fades in and out. Remember the house with concrete walls? I bet he wished he had a preinstall survey done. Just kind of licking your thumb and sticking it up in the air to determine AP placement, saying to yourself “this seems right” is setting you up for failure and setting your users up for disappointment.
You can have the best planning possible done, but without performing a survey you could (probably will) be setting up an inadequate, failing network before you ever get started on the installation.
Must be able to test, validate and confirm coverage
Whether you’re installing a wireless network for a home, an office or a campus, once completed, you’ve got to be able to test, validate and confirm coverage. Hence a post install survey is necessary.
A survey of this nature ensures that there are no ‘surprises’ post install on a new build that could impact connectivity and performance. Such a survey can also detect problems that might occur even further down the road once a facility is operational and has been in use for a while. This testing gives you the ability to tweak the network by making AP adjustments to obtain optimal network connectivity and performance.
Here is an example. I’ve got a client right now that has a corporate executive hanger and they want us to come in and do a complete post install survey on both the wireless network and whatever other noise is in the environment that could impact it. One of the concerns they want us to validate is to make sure that their wireless fuel sending radios, which are sending fuel levels for the remote tanks that they use to fuel the jets, isn’t going to interfere with their main wireless network and that the wireless network isn’t going to interfere with the fuel system wireless radios so they get accurate fuel levels.
Planning, surveys and post install testing – you will not have a truly successful wireless network, operating at its fullest potential without them.