Technology Blog

To Survey or Not to Survey?

by ‎07-19-2016 09:00 AM

It seems that 2016 is the year of confrontation in regards to wireless network designs and techniques. A lot of spirited debates have sprung up over many different topics, the most recent one being whether or not an AP on Stick (APoS) site survey is required. Let’s take a look at this from the standpoint of someone who maybe has their ACMA or is just starting on getting their ACMA.

 

As you enter into a new technology you need to learn not only what products are available but also how they operate and function. In order to do this, you need to do testing, physical testing. Sure you could fire up VisualRF Planner and start working on developing a predictive design but without any physical testing prior you have to trust but can’t verify what it is showing you. This is where a physical APoS survey becomes important. By doing physical testing with your Aruba 325 for example that you want to deploy you will be able to accurately depict coverage areas. But where do you start on an APoS survey? Where is your first, second, etc AP location? An APoS survey should always be validating a predictive design, so we are kind of stuck in a catch 22 here. How can I do a predictive design if I don’t know what it’s going to look like?

 

EXPERIENCE!

 

One simple word answers that question: experience! This is the importance of being a skilled wireless engineer and working towards understanding RF propagation. My personal experience has led me to believe that the best way outside of the CWNP program exams to get this experience is to do APoS surveys! By surveying a lot of different environments you get that understanding of how different walls truly impact signal propagation.  This experience I got surveying gave me the ability to yes start to not have to do APoS surveys for locations but I still prefer to for majority of environments. For example, in a hospital, if you are placing APs in clean rooms or surgery suites you need to get that right the first try. You’ll be hard pressed to go back and state we need to move an AP 5’ either direction because of the downtime required to do work in those areas as well as patch any holes created.

 

So, Survey or No?

 

Well it comes down to the typical wireless answer we always give: it depends. Am I going to go and survey a stadium? Most likely not. It just isn’t cost effective and I have a very good idea of how signal will propagate using the few different placement options I have. Will I survey an office building or hospital? Most likely I will. There might be areas that I can use previous experience or even just a simple site walk to check loss values that I build into my predictive design to ensure proper design. There are times that doing an APoS can reduce your total number of APs by 30% or more. If you do a predictive design with wall loss values of 6dB let’s say and by doing an on-site APoS you determine the wall loss value is only 3dB that drastically changes your design. You could have done a site walk that determined that was the wall loss value instead of a full APoS survey as well.

 

This is the major problem we all face when doing wireless designs: no two engineers perform wireless designs alike. A lot of people will always say surveys are not needed and we can use predictive designs. Others state you absolutely must do an APoS survey every time. This hurts the industry as a whole because we are not delivering the same message. It would be great if we could come up with a standard that at a minimum explains why there is a need for on-site APoS surveys for certain environments so everyone is playing on a level playing field.

 

Validation Required

 

The one survey that is absolutely required and always needs to be performed is the validation survey. Even if you do an APoS survey you should always do a validation survey to see what ARM is doing if you enabled it. The validation survey is the only way to ensure your network is doing what you want it to do and how you designed it. This requires you to use a tool like Ekahau or AirMagnet that are developed for the sole purpose of visualizing the RF signals being propagated by your APs.

 

Closing Thought

 

Finally, a quick statement about surveys. Many believe that surveys are the work of unskilled worker bees. I always hear statements about site survey work being beneath an ACMX/ACDX, CWNE, or CCIE. This couldn’t be further from the truth. Even as experts we still need to hone our skills and perform surveys to stay informed of the latest propagation issues RF designers are facing in new building designs. Plus doing site surveys makes sure we win all the step tracker contests!

Comments
kjohnston@securedgenetworks.com

Blake, your comment about doing a post-deployment RF verification is spot on. I always suggest to customers that if they are using ARM or RRM, that they wait at least two weeks after deployment, or occupation if it is a new facility, to perform the RF verification. This will allow the RM algorithms to settle in with the typical user/client base in the facility.

Waiting is a great point to make! It is key to let the algorithms run for a bit prior to surveying otherwise your data is skewed.

Search Airheads
Showing results for 
Search instead for 
Did you mean: 
Announcements
Read all about it! If it’s happening now, it’s in the community.

Check out the latest blogs from your community team, the community experts and other industry sources.
Labels