Understanding the Wi-Fi Engineering Process: A Step-by-Step Guide

“I’m just going to throw up some access points” is a statement I bet I’ve heard a hundred times since we started deploying large scale wireless systems almost 11 years ago. Well, for the really small offices that need 1-2 access points, that solution will probably work and this article is a waste of your time to read. However, if you have for example, a 100,000+ sq. ft. warehouse, a school or campus, or a hospital of any size, putting up a wireless network, unfortunately, it's not that simple.

The challenge is that many people think of wireless in the same way that they think of a cat6 wire running to a work-station. But wires are simple, they go from point A to point B. 

Wi-Fi Planning is inherently more complex for a number of reasons:

1) You can’t see RF Signals.

RF (Radio Frequency) signals are invisible to the naked eye (without the right software tools), and if we can’t see it, this alone makes the entire process challenging.

2) Physics.

Wireless access points have a limited number of channels they can work with. If you don’t have enough access points you have coverage or capacity issues. But if you pack too many access points into a small area you end up causing more problems than you actually resolve…it’s a little bit of science and a little bit of art to get it right.

3) Interference.

Because we have limited RF channels, other systems use the same radio frequencies we use for Wi-Fi (that physics problem again). This means that sooner or later, you’re going to interfere with yourself, or that old microwave is going to start pushing users off of your wireless network. So even if you get it right, inevitably your RF environment will change.

4) Variable Devices and Applications.

7 years ago, the devices on our networks were all windows machines, we controlled them, and we refreshed them every 3-4 years. Those good ole days are gone. Now we’ve got personal phones, tablets, laptops, and IoT devices on the network. Those devices change frequently and have different performance requirements from your Wi-Fi network.

As goofy as it sounds, a good way to think about wireless today is that it’s a living thing. It’s constantly changing and adjusting based on what’s happening inside your building, your campus, or your hospital.

Your physical environment changes and Wi-Fi coverage changes with it. The devices and applications being used change almost daily. So Wi-Fi is not like running a static wire to a work station.

Spoiler Alert! Wi-Fi is a dynamic system that has to evolve over time. So this process of planning, measuring and adjusting will be repeated over and over.

This brings up a question we get asked all the time, “what is the proper way to design a wireless network?”

The answer to that question is that it depends on what you’re designing it for and the size of the project.

Sometimes planning the system and allowing for adjustments is more cost effective for the customer. Sometimes it makes more sense to get a plan that is 100% on target before you deploy.

The answer is not the same for everyone. So to answer the question with more than, "it depends", I’m going to talk you through the entire process of what “should” be done when deploying a large scale wireless network to ensure that it is designed, implemented, and supported properly.

Keep in mind that any individual discussions about a specific environment really need to involve an RF Engineer, and some conversations around the specific project.

So here you go, in a perfect world, here is the entire WLAN process:

1) Predictive RF Design

Software tools today are pretty accurate when looking at access points and predicting coverage.

We can load in a drawing of the building with dimensions. We can load in building materials like walls and racks of equipment in a warehouse and the software will spit out an RF design with predictive coverage for the building.

We can also look at how the system will be used (i.e. hand held scanners in a warehouse) and adjust the performance requirements and RF plan.

It does require a lot of experience in order to be able to use the software properly because there are some manual adjustments (Art) to it. But this is the foundation for every other part of this process.

A word of caution here: if the software is free, it’s most likely wrong. As with anything else, RF Planning software is expensive because it takes a lot of development and updating to make it accurate.

Many times, candidly, the access point manufacturers have their own “free RF planning software” that is used as more of a marketing tool. It will over estimate the number of access points needed by as much as 40%! So it's important to watch out for that.

2) Wireless Site Survey

We have a starting point because of the Predictive RF Design, but your environment may have stuff going on that the software doesn’t know about.

In this step, we’re not throwing out the predictive design, just validating it or making adjustments where necessary.

Typically, office environments and school environments, are pretty accurate using predictive software. We've found that we’re accurate within 5-10% of the real world.

However, other environments have a lot more moving parts, and they almost always require an on-site RF site survey:

  • Warehousing/Manufacturing – Storing a 30-foot rack of furniture produces different RF issues than a rack of lighting products. The difference in inventory and inventory levels by time of year produce a different RF pattern. Large metal machinery (robotics or fork lifts) reflect RF signal. In order to adjust the plan for your specific inventory and machinery, a physical site survey is necessary.
  • Medical - Hospital wireless networks are right up there with the most complex RF environments that exist. Medical systems are typically big pieces of metal (reflective) and/or have their own wireless communication (interference). The other challenge for healthcare IT teams is that so many devices are enabled for Wi-Fi, i.e. insulin pumps, Wi-Fi temperature sensors, drug dispensers, carts on wheels, just to name a few. And all of these systems have different performance standards to plan for.

3) WLAN Engineering

The RF Planning produces a list of equipment, antennas, mounts, as well as a list of things that need to happen on the network to implement the system. Now it becomes about execution. The engineering phase will include:

  • Configuration - Setting up the system’s SSIDs, security policies and roles, and provisioning the access points.
  • Integration - Connecting the platform to your directory services and (hopefully) management software.
  • System Testing - The authentication methods and security needs to be tested thoroughly, and with the devices that will actually be used.

4) Performance Assessment

The RF Design has been done, and the system has been engineered according to the design. This is where you really find out how well the planning process worked.

This is also where more expensive software tools and some know-how are required. In order to validate the system’s performance, we’re going to do two things here, using two different suites of software.

  • Coverage and Capacity Testing - We’re going to look at the real-time coverage in the live environment. This stage requires WLAN management software that can look at access point coverage as well as look at users connected to access points to measure capacity. This will determine what/if any adjustments need to be made in the live environment.
  • Application Performance Testing - The wireless network looks good from test #1, but what’s the user experience like? This is the most important test! For this, we’re going to use a suite of software tools that measure the applications being used (streaming video for example) and we’re going to do it on the devices the customer uses. This test will measure the actual system performance and benchmark it against the industry standard for that specific application. No getting around this one, it either passes or it fails.

5) Performance Monitoring & System Management

Congratulations, you've passed all the tests! It’s time to move on! Well, not really. Devices change, the physical environments change as well. Hackers never stop attacking, so software gets out of date.

In order to keep the system from being obsolete we’ve got to monitor it, keep the software up to date, address security issues, and make adjustments as necessary.

In order to do this successfully, we use a process to manage and update the system month to month, and we monitor performance in the same way we did system testing on step 4.

  • Network Monitoring - We’re going to look at coverage, capacity, and utilization on a monthly basis, and we’re going to set up alerts on any critical system issue like access points, controllers, or other network components going down. If this is done well, many times changes will be made before critical problems happen.
  • User Health - With yet another set of software tools and sensors, we can monitor the system performance from the user’s perspective. We’re going to benchmark industry performance, and measure our user’s experience to that benchmark.

So there you have it from start to never ending.


At SecurEdge, we provide a Wi-Fi as a Service platform that makes networking more affordable and easier to implement. If you have any questions or would like to discuss an upcoming project, please contact us here.

wireless network design kit, wireless service providers,
Philip Wegner

Philip Wegner

Philip is the founder and CEO of SecurEdge Networks. He’s the consummate strategist and frequently writes for the strategy blog. You can follow him at @philipwegner

Comments

wireless network design kit, WLAN design, wifi design help