The Wireless Network Lifecycle
A wireless network lifecycle typically starts with a component of upfront requirements gathering that is then parsed by Wi-Fi professionals and translated into a design. From there, the design is implemented, validated, tuned, monitored, optimized, and monitored some more until the requirements change or go out of spec and you do it all again. Wi-Fi isn’t a “set-it-and-forget-it” technology and because so many organizations rely on wireless networks for their operations, it’s even more critical to get it right and keep it optimized.
Gathering Network and Client Requirements
Arguably the most critical thing you’ll do when building a Wi-Fi network is gathering specific requirements. It’s also often the most difficult. But the effort placed into this first phase of the wireless network lifecycle will pay dividends when you’re able to move into optimization phases instead of starting back from the beginning because requirements weren’t properly identified the first time.
Identify the Least Capable, Most Important Device
Every industry is different and while some businesses may operate a BYOD policy with a massive range of client types, others — like warehouses — may be heavily dependent on a barcode scanner that is several generations out of date. For these types of clients, breaking down the manufacturer’s posted information can range from simple criteria mapping to a more complex investigation.
Quick Tip: Client manufacturer documentation is generally specified for default values, and any changes made to registry settings will need to be included in the plan. It’s best to check with the admin to see if any customization or tuning has been made.
Get Sign-off on Requirements
Once you have all of your requirements gathered and documented, it’s important to have sign-off to ensure you’ve captured everything correctly. Oftentimes, this stage reveals a few additional requirements or requirements that have dropped in priority as clients suddenly remember the use of video streaming in the lobby or that the living atrium was added after the building was constructed.
Predictive vs. Empirical Data vs. Hybrid Design
While there is an immense amount of variety in the types of data you’re going to get gathering client requirements, by and large, there are about three different ways you can design a network. You can predicatively design a network, you can empirically measure your design or you can do a bit of both.
Wi-Fi Predictive Design
The concept of predictive designs, whether you call it guesswork design, crystal ball design, or divination-based design, can be as crude as walking into a building and just placing APs based on look and feel, or it can be more advanced, such as utilizing AI to perform calculations to optimize around known information. And while it’s possible to do a proper WLAN design that is 100% predictive, there is a higher level of risk involved in a predictive-only design. At the very least, predictive designs are invaluable for doing building materials estimation, level of effort estimation and to give you a starting place for an AP on a Stick (APoS) survey.
- Apantomancy to Numerology
- Usually requires a crystal ball or a tremendous amount of intuition or instinct
- It is possible to do a proper WLAN design using 100% predictive modeling with no M/A/C Post Deployment (weatherman modeling)
- It’s super useful for BOM and LOE approximation
- It’s a fantastic place to start your APoS Design work
- You probably want to check your work afterwards
When you build a network using a predictive model, you probably want to go back after the fact and check it—and that’s where empirical measurement gets layered in. The empirical design of a network is gathering irrefutable data in the air of the actual environment. A major benefit to empirical data is the mitigation of risk. When a design has to be right before it goes into the ceiling, you just can’t skip the empirical data.
- Known by many names, we’ll use APoS
- Heavy lifting, high cost, significant effort
- Can be hard to get right the first try – but when it’s right, it’s right
- No arguing with empirical
- Requires more than a passing understanding of things like tx power, RRM, etc – but requires little understanding of what is in the walls
- When it has to be right, before it goes on the ceiling
Hybrid Wi-Fi Design
And then there is the mixture of the two—a hybrid of predictive modeling and empirically measured modeling. This can take the form of a more simple survey, placing an AP in the environment and measuring attenuation to feed back into a predictive model, resulting in moderate risk, cost and effort. One caveat is that you can elicit a false sense of security including an element of empirical measurement without following through with a full validation survey.
- Hybrid can be done to model predictive data or empirical data
- It is still a form of predictive design, but tempered with a measure of empirical data
- Moderate risk, cost, and effort
- It is still possible to get wrong
- Can elicit a false sense of security in your design – make doubly sure you validate your design after the fact
Which is Good, Better & Best?
Once you get the requirements right, design is about balancing the level of effort with risk versus cost. It is about finding the right balance for your wireless project. Unfortunately, no amount of technology or skill can guarantee a guess. A guess is still a guess at the end of the day.
Remember, wireless design isn’t about placing APs. Wireless design is about mitigating risk and translating user requirements into a deployment that meets the company’s needs. If the needs are mission-critical Wi-Fi before it goes in the ceiling, empirical design is best. If the needs are still undefined, a hybrid design with some recommended follow up may be better. And if good enough is good enough, well, there’s your answer.
More Wi-Fi Best Practices
Be sure to watch the full on-demand session, “Good, Better, Best Practices in Wireless” for more tips from Sam including vendor best practice conflicts, the uptick in contact tracing due to COVID-19, and ways to get involved with the Wi-Fi community to take advantage of the enormous amount of shared knowledge that can save you time, effort and money.