Most WLAN Engineers I know don’t have to do outdoor APoS surveys very often, however when you need to, this post might come in handy.
The last time we did an outdoor survey, we used a lightweight Cisco 1532i series access point, a small PoE+ switch, a Cisco 2504 WLAN controller and a fairly large UPS to power it all. We learned that an AC inverter plugged into a 12v power outlet in a vehicle just didn’t work for us and would not charge the UPS when driving from point A to point B. We had to look for power outlets and drag long extension cords around – which might get damaged if people drive over them.
This time, we changed it up a bit. Our task was to test a Cisco 1532e series access point with an external directional antenna. For this, we purchased the new Ventev Venvolt MK1 power supply that can supply PoE+ (802.3af & 802.3at) power to an autonomous AP that requires 802.3at power – for hours on a single charge. We loaded the autonomous code on to the access point and then configured it like you would an autonomous AP during an APoS survey.
We used the same old survey cart and telescoping pole that we always use, mounted everything like you would expect it to see on an outdoor wall or pole, and plugged it into the new Venvolt MK1. A few minutes later, the AP was online and ready to go. Here’s what the rig looked like:
We configured our Ekahau ESS site survey software to do an outdoor survey. There are a few HowTo’s floating around on how to do it. I must admit, we spent the previous day getting ESS to work with the GPS adapter. We had to download drivers, etcetera, and go through all the motions to get it working. It wasn’t simply plugging in the GPS receiver and running out the door. That said, spend the time to get all of that working first. I used a BU-353 GPS receiver, if you are wondering. Set an hour or two aside the day before (or longer) and get that working. Do not wait until you are in the parking lot with your AP up high on a pole and then decide to embark on that task. You might need access to the Internet to get the drivers, etc. Familiarize yourself with how to use ESS with a GPS outdoors – figure out how to start and stop the survey, etc. Practice with it at home if you live in a quiet neighborhood, or in a park, or somewhere else where someone won’t call the police on you.
When you are setting up your project, you need three locations in a triangle on your “floor plan” before you start surveying. We were indoors when creating the project, and we figured out that when looking at maps.google.com, it was giving us the coordinates in Decimal degrees, and ESS wanted Degrees, minutes and seconds (DMS).
We searched around and found this website to convert decimal degrees to DMS: https://www.latlong.net/lat-long-dms.html I believe if you install Google Earth on your laptop, it will give you the requirements you need in the format ESS wants. We didn’t want to go that route – just our preference.
We also used the same website where we got the “floor plan” to measure the distance between two corner parking spaces, and then used that measurement in our project. Worked beautifully.
I cannot stress enough to set everything up before you go out on-site to do your survey!
With our project ready to go in ESS this is what we looked like:
Our first “driveabout” was to see how much energy would be behind the panel antenna. We aimed the antenna to the south, put the GPS and Wi-Fi adapter on the roof of the car and started driving around. As we expected, we had some RF propagation behind the panel antenna, seen below:
Why is this important? Keep in mind all the channels that we use/don’t use, and if this antenna is on a pole, it is susceptible to interference from that direction. Since our application will be pole mounted, we mounted it on a pole to see what might happen in the installed environment. If we were going to install this antenna on a brick wall, we would have parked our survey rig up against a brick wall and walked the other side.
For our next driveabout, we moved the survey rig to an area that had a row of small trees between the rows of parking. I think it is rather obvious where the trees are:
The point of this is to see how far the 5 GHz signal will propagate outside, when impeded by a number of small trees. That distance is about 100 feet. The fewer the trees, the farther the signal goes. We have to keep in mind that these trees are going to likely grow in the future, so if we were covering this parking lot, we would have to plan for that. One thing to note – having trees is not necessarily a bad thing. Having attenuation outdoors keep your cell sizes smaller – all part of a carefully crafted RF plan.
Our third test spot was closer to the road that is more like a long hallway in a building with less attenuation. As you can see from the graphic below, the signal traveled almost twice as far in that spot:
From this testing, we learned a few things:
- We now have a good feeling of how the antenna’s RF propagates.
- Height and antenna down-tilt affects the size and shape of the cell.
- Trees attenuate RF which affects the cell size.
- We measured the 2.4 GHz cell size, which was larger, but won’t be using it for the deployment and will be turned off.
- Outdoor site surveys attract Police cars.
- *Keep in mind the transmit power and channel selections may increase/decrease the cell size.
To sum it all up – a little planning at the beginning of an outdoor deployment may save a lot of time and money in the long run, since installing outdoor Wi-Fi gear can be expensive.