In this third installment of my prep for the CWAP-402 exam, I am going to explain the process of chasing down knowledge. I’ll use the WLAN Power Save mechanisms as my case study. This method works well for any type of technical exam, especially anything CWNP related.
Disclaimer – while I do hope you learn something about Power Save from this article, it’s not intended to be a lesson on it. It’s a look at a method of learning, which I hope you will find useful.
In the chase for knowledge, you need to assess your knowledge against three criteria.
you don’t know something
you think you know something
you know you know something
If you read part 2, you will remember I highlighted some items in the CWAP-402 exam objectives document, identifying what I felt I needed to study. I have been teaching CWNA and CWAP courses for a number of years, yet there has always been something that didn’t sit quite right with me about power management. So right there I am assessing it as “I think I know” rather than “I know I know”. Following my proscribed method, I reached for the Sybex PW0-270 study guide and read all of chapter 8 – Power Management. The following items from the exam objectives are covered in this chapter:
With the chapter duly read, some troubling points came to mind. I understood the terminology but did I KNOW it ? When faced with this, you need to get creative with how you look for further information.
Many WLAN professionals blog about their studies and WiFi in general. This is a superb source of information and will give you new ways to think about things.
Before we go any further, let’s make sure everyone is on board with the original 802.11 power save mechanism: PS-Poll. If you aren’t sure how this works, read this excellent blog post from CWNE #153 Rasika Nayanajith for a detailed explanation. PS-Poll produced a lot of overhead, so something better was needed, and the 802.11e amendment was tasked with looking for ways to improve power management and QoS. So how did that work out ?
I was curious to see what power setting WERE in use and how power settings were advertised in the BSS, so the first thing I did was to fire up AirMagnet WiFi Analyzer to do some captures. Let’s examine what frame capture can tell us.
APSD – Automatic Power Save Delivery
APSD was defined in the 802.11e amendment. APSD support is advertised in the APSD subfield of the Capability information Field in the beacon header. Let’s take a look:
So, APSD not supported ? Nope. Why ? Because we all know that the WiFi Alliance took inspiration from 802.11e and implemented their own spin on it as Wi-Fi Multi Media (WMM). Wi-Fi CERTIFIED WMM® was introduced in 2004, and Wi-Fi CERTIFIED WMM Power Save in 2005. So, time to get some whitepapers.
Whitepapers are the food and drink of the CWNE. Whatever you are researching, there will be a whitepaper somewhere. From corporate documents to university research papers, whitepapers shine light into the dark corners of WLAN operations. You should become adept at hunting down, saving and dissecting whitepapers.
The WiFi alliance has a number of papers to choose from here . The one I selected is titled: WMM™ Power Save for Mobile and Portable Wi-Fi® CERTIFIED Devices. Page 4 mentions that WMM power save is based on U-APSD – that’s good, that term is on the list. U-APSD is part of the QoS mechanism, so we should see support in the beacons. S-APSD was never implemented so can be discounted from here on. So, back to AirMagnet:
So, why do I not see U-APSD support ? Could it be my access point is using one of the other Power Save mechanisms instead ?
SMPS and PSMP
Spatial Multiplexing Power Save and Power Save Multi Poll are, as far as I have been able to ascertain, simply not implemented by any vendor. I’m happy to be corrected, just never seen it…
So, ruling them out brings us back to U-APSD. I am now concerned as to why I am not seeing it advertised as supported. Get out more access points and do more captures !
If you are serious about a career as a WLAN professional, and passing CWAP (or going for CWNE) you simply must be able to capture and read frames. No skill is more important to being able to really see the matrix. Investing in the right tools is investing in yourself.
This is really odd, across multiple AP’s, running standalone code or managed by different controller platforms, I am still unable to see the U-APSD support advertised in a beacon. It’s time to to reach for the Batphone.
I pinged Marcus Burton, CWNE #78 – we work for the same company and he’s a very smart guy, so I shared my findings. He immediately sent back a WireShark screenshot capture showing where I would find the U-APSD support in the beacon:
So, what’s going on ? It seems that AirMagnet wasn’t interpreting the decode. Let’s compare the Hex to be sure:
So it was there, but just not displaying. It’s not uncommon for data to be displayed differently between vendors. So, now I am happy – U-APSD is in use in the network, and I know I know it ! Now I can focus on how it operates.
In frame capture, I noticed a lot of other strangeness… lots of RTS/CTS frames. These are commonly described as a function of SMPS, which we know isn’t in use, so what gives ? Back to the Batphone, and this time I cast a wider net by emailing a bunch of my peers. I am never ashamed to contact fellow professionals and ask for advice, and likewise I’ll try to help anyone who asks me. If you are studying or questioning something, don’t suffer in silence. Write up what you think and send it out for comments. Look for us at events. Come and say hello and chat to us. Follow us online and see what we tweet and blog about.
WLAN professionals love sharing knowledge. If we have a public profile then you can bet we are approachable and will help you whenever we can.
First to reply was Devin Akin, @DevinAkin and CWNE #1. He gave me some great responses to my questions and some links to whitepapers for further reading. Ronald van Kleunen, @Globeron, CWNE #108 chipped in with some questions and before long an email thread on advanced Power Save functionality was doing the rounds. Peter Mackenzie, @MackenzieWiFi, CWNE #33 and co-author of the PW0-270 study guide chipped in to say if I hadn’t got the answers I needed, to let him know and he would provide more info. I am honoured by these gentlemen, and their dedication to WLAN truth.
This article is aimed at teaching you what you can learn about learning. Here are the main points:
- Identify what you don’t know, what you think you know and what you know you know.
- Read blogs and whitepapers to find out more information.
- Challenge your ideas by running labs and testing scenarios to confirm what you know.
- Be prepared to ask for help.
- If research raises more questions, follow them up too.
- Don’t stop at “think you know“, go all the way to “know you know“
My thanks to:
- Marcus Burton for inspiring this article
- Devin Akin for selflessly sharing knowledge
- Ronald van Kleunen for additional insight
- Peter Mackenzie for offering further support
And here I am, it’s Sunday 11 September on a cold and grey Scottish afternoon, and it’s taken me a week to research and cross off 4 of the exam objectives from my list. Did I need to go so deep to pass the exam ? No. So why did I spend so much time chasing this knowledge…. ? I’m a CWNE, it’s what we do.
Some people are content to study until they get something right. Experts do it until they CAN’T get it wrong.
CWNE is not the last word for WLAN professionals. Many excellent WLAN pros are working in our industry with no desire to take CWNP exams or become a CWNE. Those that do hold CWNE are universally recognised as having spent a long time mastering their craft. I myself took 6 years to attain it. Anyone who holds CWNE has almost certainly approached learning using some or all of the methods I describe above. CWNE’s are not special people. We are simply not satisified with “think we know”. By our nature, we can only settle on “know we know”. I contend that anyone who works with WLANs and who has the commitment can gain the knowledge and experience required to become a CWNE. There are still only 188 in the world (Sep 2016), and I’m happy to see it’s growing. Come and join us.