10 Common Wireless Network Mistakes [Video]
In this episode of Whiteboard Wednesday, Senior VP of Engineering, Michael McNamee, breaks down the most common Wi-Fi mistakes his teams sees on a regular basis and how you can avoid these pitfalls.
Hi, and welcome to another edition of our Whiteboard Wednesday.
Today I want to talk about some wireless network myths and fallacies that we run into in the field very often.
These are ten common wireless mistakes that people make when deploying their WLAN’s and this ranges from anything from the customers that we engage with to other engineers, other system administrators just not having the RF know-how about Wi-Fi and how it actually works.
1. Automated RF Management Fixes All Issues
The first one that we run into a lot is that the understanding that automated RF management fixes all issues. This is a false statement. A lot of vendors and manufacturers will sell their product as having this automated RF management system that can take care of all issues and that’s false. That’s not true. It all goes back to the design. If you start with a good design and your design is solid, those automated RF management systems or features work really well because they can compensate for that five to ten percent dynamic shift in the WLAN behaviors and performance, but if you’re asking that system or that feature to address twenty to thirty percent change in the network it’s just not going to work. So, what you have to do is start with a good solid design and make sure everything is tuned really well and then those features can work when you have those small dynamic changes in your WLAN.
2. More AP's is Always Better
The second myth or fallacy that we run into a lot is that more AP’s is always better, and this is definitely false. A lot of times we’ll go into an environment where two or three access points properly designed and tuned will actually do the job but we’ll see five to ten access points in that environment that are being asked to do the job and what’s happened in the past is that they put a couple of access points thinking that was enough to accommodate the capacity and coverage in that area and they run into Wi-Fi issues because they don’t have the insight or the tools to understand what the actual root cause is. They just throw up more access points hoping that it’s going to address issue and what it really does is it actually implements more issues for them and induces more challenges on the RF side for the network and causes a myriad of other problems. So, more AP’s is definitely not better.
3. More Transmit Power is Always Better
Another myth or fallacy that we run into the field very often is that the idea that more transmit power on the access points is actually better. I don’t know if any of you can reference back to the Spinal Tap documentary where they had eleven as the highest level for the volume. A lot of time we’ll see access points turned up to eleven or full power and that’s not the way that you design or configure a WLAN. The AP’s have to be tuned to have symmetrical power with the transmit and receive of the client devices whether it’s a phone, tablet, or laptop. You always want to address the least capable device which in most instances is that phone. It might be a small milliwatt transmitter like a twenty to thirty-milliwatt transmitter where as a laptop is maybe a fifty to sixty milliwatt transmitter and then the AP’s are one hundred milliwatt. If you had the AP at one hundred milliwatts the devices could see it just fine but at the edge of the cell where the coverage ends the devices are still seeing that AP transmit very loudly but they can’t transmit back that same distance so what you have is the devices appear as to have a great Wi-Fi connection but their performance is really slow and it’s confusing to them. It’s a mystery and what’s happening is the device can’t transmit back that same distance so it’s never getting an acknowledgement from the system that it’s received it’s transmission and so the device just keeps transmitting that same packet that same information over and over again and what that does is it ties up airtime for the system and impacts everybody on that WLAN so more transmit power is definitely not better.
4. I Don't Allow Wi-Fi in my Business
Another myth or fallacy statement that we see a lot or that we run into is that we’ll go to see a customer or potential customer and they tell us that they don’t allow Wi-Fi in their business. This could be for security reasons or cost or they just don’t think that they have a need for Wi-Fi devices so this statement is false. What happens a lot of times is we will go into these meetings and they will tell us that they don’t allow Wi-Fi because of security concerns or what have you and I will actually have my laptop open and the Wi-Fi scanning and I will see other Wi-Fi networks within the facility. I know they are within the facility because the signal level is at a certain level that I know it’s going to be within their four walls. So, if you don’t have Wi-Fi, you won’t know that there is another Wi-Fi there. Now you could do audits or surveys periodically to see if there is Wi-Fi but again that’s a snapshot in time and you don’t know if those things are popping up intermittently, coming and going, or it could be an intrusion attempt from a hacker trying to provide Wi-Fi connectivity into your wired network. So, we always recommend at a minimum at the base have some sort of Wi-Fi in your facility just to do Wi-Fi scanning real time twenty-four seven, three sixty-five.
5. Surveys Need to be Done Only Once
Another myth or fallacy that we run into very often is that surveys only need to be done once. This again is a false statement. Wi-Fi surveys are done for a myriad of reasons it’s either to validate a design that’s going to be deployed in that facility, the survey could be done just to prove that there is good coverage, or they could be an active survey actually testing some performance metrics but that’s a snapshot in time. WLAN’s are dynamic. They change over time whether it’s users migrating from one area to another, new applications, new devices being deployed into the environment and so surveys really can only give you an idea of what the WLAN was performing or looked like at that time when the survey was done and what we recommend is having surveys done periodically over time using the first one as a bench mark and then see how your WLAN changes over time whether it’s every six months, every year, what have you, and supplement that with some sort of measurement tool or utility that’s real time, twenty-four seven, three sixty-five measuring the performance of your WLAN throughout the area whether you’ve got a campus or a single building or you’re in an office on a single floor and then correlate those two things. So, you’ve got these snapshots in time and then correlate them with your real time metrics that are being measured on a daily basis all of the time and then see how is your WLAN changing over time and is it addressing the needs of what you’re trying to do those devices or those applications or is it failing and so those will give you some insight into how you can address the needs of your WLAN.
6. 80MHz Channels are Always Faster
This next Wi-Fi myth or fallacy that we run into is a challenge for us because what happens a lot of times is that manufacturers and the vendors actually enable this by default and people want to know why we don’t recommend that it be run and that’s the statement that eighty MHz channels are always faster. This is again false but it kind of walks the line of being true and false and it depends. Eighty MHz can work if you are in an area where you’re not surrounded by other Wi-Fi networks or you don’t have any type of Wi-Fi contention or congestion going on. If you’re a remote business and you don’t have any neighbors and there’s no RF impacting your environment, eighty MHz may work well for you if you have just got a couple of access points up to three or four would actually work but if you get anything over three or four access points and you’re in an environment where you’ve got multiple whether it’s dozens, hundreds, thousands of access points or you’re in an environment where there are other networks surrounding you whether your in a mall or a shared office tower. Eighty MHz channels not only are you going to not be a good RF neighbor, but you are also going to impact the performance for yourself. So, in those environments we only recommend twenty MHz. That gives you the most channels to choose from, reduces your co-channel and adjacent channel contention, and will just improve the overall performance of your network. So, eighty MHz channels are not always faster.
7. It's Always the Wi-Fi at Fault
So, this next myth or fallacy statement that we love is that it’s always the wireless that’s at fault. This is false. I’ve been doing this for a long time and we get the calls and the tickets, and we’ll drill in to look at what the issue is with the Wi-Fi and we find out it’s really not the Wi-Fi that’s at fault it’s something else. So, there is a myriad of things that can impact Wi-Fi performance. This could be anything from back-end network services like DHCP and DNS, network congestion, even things like authentication. So, if an authentication server is down or it’s not replying quickly enough, lag time and response times go up, devices may time out. There is a myriad of problems that affect how Wi-Fi performs. So, what we always recommend is having some sort of tool or utility that can monitor your Wi-Fi and your network over time and correlate the two so it will help you get to a root cause of what your Wi-Fi issue is whether it’s a wired network issue, a back-end network services issue or if it’s truly an RF issue something that has to do with spectrum contention or channel contention or just overall RF performance. We recommend some sort of tool or utility or solution that can measure that and give you root cause, so you can resolve the issue quickly.
8. The Wi-Fi System Controls Client Device Behavior
Another myth or fallacy that we run into and have to explain a lot is that Wi-Fi systems controls the client device behavior. This is false. In the cellular world when you’re travelling down the road with your cell phone and you’re on the interstate the cell towers actually tell your phone when to disconnect from one tower and connect to the next closest tower. In the Wi-Fi world, I wish that it worked that way, but it wasn’t written that way in the outset. The way Wi-Fi works is that the focus was put on the client devices to make the decision on when to disconnect from an access point and reconnect to another. Now historically, the devices don’t make the best decisions on how to do that. That can be improved by making sure that patches and updates to the operating system are done and that client drivers for the wireless nic are updated into the most recent, stable, best behaving version of the driver sometimes you can reference the vendor websites and they will tell you which ones have been tested and which versions are most interoperable and what they recommend. Typically, just having them up-to-date will improve the Wi-Fi system performance for those client devices so there are some things that you can do on the design side and we always address this by looking at the applications and the types of devices that were going to be using on the WLAN and that we design the WLAN to accommodate those devices the best way possible and were always looking at the least capable device whether it’s a phone or some sort of low power device and were tuning the network to make sure that those devices are accommodated so they can roam from one access point to another and it’s seamless and if we know that were tuning for something like a VOIP application or even a video application, we are addressing that with different types of quality of service or prioritization through the air and on the wire to make sure that performance works really well for those devices so that the Wi-Fi doesn’t get blamed for making improper decisions.
9. One-for-One Replacement of my Legacy WLAN is Acceptable
The next myth or fallacy that we run into a lot is that one-for-one replacement of your legacy WLAN is acceptable. This depends but most instances this is false. In the old 802.11abg legacy Wi-Fi days, when we move from the legacy solution or technology to 802.11n, we learned some lessons in that one-for-one replacement of the access points in the WLAN was not a good method for design. There were several things that were happening about that same time that caused us to rethink our design schema and our methodology about how we design WLAN’s back then. The first thing that changed was we went from the ABG diversity antenna on the access points to the MIMO antennas and technology on the 802.11 access points and this did a couple of things. It increased channel contention whether it’s adjacent or co-channel contention. We also saw the introduction of really low power devices like Ipads and Iphones which had a fractional power capability of laptops which would traditionally be the things that we would design for before and so we had to rethink how we were actually doing our design methodology and what we found was that the access points had to move closer to the clients so this meant we had to shift the AP’s out of the hallways and into the rooms whether it was a hotel room, a hospital room, a dorm room, a classroom. We had to put more access points in an auditorium due to coverage and capacity concerns. So, just taking the existing AP locations and replacing them with new AP’s is not a good methodology. Even today when we move from N deployments to AC were still going back and revisiting to make sure that were doing proper coverage for five GHz because a majority of the devices are migrated to that spectrum or that band which has the most channels and the most bandwidth and will perform the best and we want to make sure link symmetry is there and so were designing for that so again one-for-one replacement of your legacy WLAN is really not acceptable. You should always go back and address and look at the design and make sure that the current technology fits into that plan and if it doesn’t then you have got to make some changes on the design side.
10. Wi-Fi is the Source of my Headaches
So, as a solutions provider and an engineering firm we run into this next myth and fallacy quite often and it’s the statement that Wi-Fi is the source of all of my headaches and this is definitely false. What we find a lot of times is it’s the Wi-Fi devices that are more likely to blame. These are the tablets, the phones, the laptops. They all use different chipsets. The chipsets have different driver versions. Those drivers are written by different people to behave in different ways. We’ve even seen two same devices, same driver, sitting right next to each other stationary on a desktop perform completely different. One will stay connected, the other one will flip back and forth between two access points and so we know that it’s the devices that are to blame not to say that they are always to blame but they’re definitely a part of it. My mom used to say it takes two to tango, so you have to look at the devices. You have to look at the Wi-Fi. You have to look at the network and the network services as to what could be the root cause and again we always recommend some sort of tool, solution, or utility that can monitor and test the performance of your network over time to help you drill down and understand what’s that root cause for any performance issues that are degrading my network.
If you have more questions, you can make comments down below and we will respond to those and I hope you guys join us for our next Whiteboard Wednesday.
End of Transcript
Michael is the Practice Manager of Security and Mobility at SecurEdge Networks. A true Wi-Fi “Guru”, he has an incredible ability at solving the most challenging wireless mess and then helping you understand it all.