I just want to preface this post by saying…. I’m going out on a limb here. Like way way out on a limb.

Somebody posted on Twitter the other day that they were purchasing their 4th set of noise cancelling headphones. If you don’t have a pair of these magical devices, you should stop reading this, get on your favorite purchasing website and buy some, especially if you are ever on an airplane. After you’re done buying, make sure to come back and keep reading.

Side note: Turns out that the white noise caused by airplane engines can actually increase your stress hormones and potentially cause cardiovascular issues. Who knew? http://www.onlinejacc.org/content/71/6/688

But, noise cancelling headphones got me thinking about WiFi. Signal to Noise Ratio (SNR) is something we talk about all the time. Trying to achieve that ideal SNR (ideally no less than 20 dB in a high density environment) is what we spend a lot of time doing as Wireless Engineers. So it got me thinking about how active noise cancelling headphones work and if the same idea could be applied to reducing noise in dense environments. (I told you I was out on a limb)

So essentially, active noise cancelling requires 3 things. A receiver that senses the frequency and amplitude of the noise. A CPU for making sense of the receivers input and then telling the transmitter that offers a rarefaction wave that is in line with the compression waves of the received noise. This is called “destructive interference”.

What if there was a way to have “destructive interference” that was able to detect non-WiFi RF to reduce noise in our environments? What if we had separate hardware in the environment that did only this? Would it even be possible? Would it even be able to differentiate between WiFi RF and interfering RF? Why wouldn’t this work? Anybody want to take a stab at the math?

Anybody want to join me on the limb?

I recently delivered a presentation on the upcoming 802.11ax amendment. Within this presentation I did a couple of hands on demonstrations. One of them was intended to give the audience a better understanding of SNR and why it matters to RF Modulation. In the demo the participants were asked to put on some “fog glasses”. The effect doesn’t translate well in video, but just to give you an idea, here is a picture of the fog glasses I created for them. I hope you find this explanation of modulation useful and feel free to share!

I was recently listening to one of my favorite Podcasts… The Way I Heard It with Mike Rowe. I’ve been working my way back through the archives of this show and got to Episode 1: The Million Dollar Kiss. He was telling the story of a young woman, who escaped from a life she didn’t want to take part in, only to become famous for her face in Hollywood, but was also a savvy inventor. During WWII, she patented a machine that would alter technology forever. This woman was, of course, Hedy Lamarr.

I’m obviously a little behind on learning about her, but most of the media available about her talks about how she was mostly recognized for her beauty and it took many decades for her to be recognized for her contribution to modern day communications… Notably, WiFi. The most interesting part for me is the invention itself.

Hedy Lamar sat down in 1942 with the intention of developing an idea to help out the navy. She disliked Nazis and wanted to be apart of the retaliation effort. So she started by thinking about torpedos. And how we communicate with torpedos to keep them on track. And why that wasn’t working. She discovered that the Germans were discovering the radio frequency on which we were communicating to the torpedos, and then sending out opposing messages on the same frequency in order to push the torpedo off track.

She postulated that frequencies could be changed intermittently and for incongruent amounts of time but still carry the necessary data to guide the torpedo. She began to develop this idea and enlisted the help of her composer friend, George Antheil in order to implement this idea. George was a master at synchronizing player pianos in order to compose symphonies. He and Hedy surmised that they could use a miniaturized version of a player piano scroll in order to synchronize the transmitter (“Mother Ship”) and receiver (torpedo). To be able to send and receive on the same hopping frequencies.

She invented channels, and labeled them A-H. But to add a level of complexity only allowed the receiver (torpedo) to receive on channels D, E, F, and G. She did this so that the radio operator at the transmitter could send out spoof signals on a frequency that was not actually received in order to confuse the enemy operators. The radio operator at the transmitter would get a warning indicator when an unusable channel was being used and could still send across that channel, but know that it wouldn’t be received by the torpedo.

The data being sent from transmitter to receiver was not terribly complex. Very simply it was an R Key (Right) and an L Key (Left) for rudder control. It’s easy to see how this was easily adapted into 1s and 0s in the first implementations of WiFi.

Her entire invention is purely mechanical in nature. It works using the miniaturized piano scroll along with mechanical oscillators, modulators, and amplifiers and runs on a battery.

The idea of frequency hopping and RF modulation is a fairly complicated one in modern times. I’m blown away by it’s ingenious implementation so long ago.

You can find the text and diagram of the original patent here. It’s a good read.

 

I recently listened to the awesome episode of the Wireless LAN Professionals podcast which you can check out here:

https://www.wlanpros.com

This was Episode 116 where Keith Parsons was interviewing Jim Palmer about some of his design principles at an unnamed international airport that he is employed at.

So here are the Stats:

Business Goals: To provide internet to the public at speeds of 200-300 mbps

Channel Width: 40 MHz channels in most public areas

Internet Backhaul: 2 10GB Ethernet pipes

Rate Limiting: No

Antennas: Directional

DHCP Lease time: 15 Minutes

DHCP Pool: /16 Flushed every 30 minutes

Security: None – No Captive Portal, No WPA2 Password, Nothing…

So the thinking for the business is that the customers of the airport want to be able to download content for airplane rides or to stream something while waiting on a layover. To allow customers to have a positive experience there are very few limitations on the public Wi-Fi at this particular airport.

There is so much logic and data to backup this highly unusual public Wi-Fi design implementation that it makes me wonder why more airports and other public areas aren’t more like this one.

Give this episode a listen, and then comment away. I’d like to know the community’s thoughts on this.

 

I imagine that for any of you that are actually reading this post it’s because you have, in some way, a commitment or passion for Wi-Fi. And so, as a community we have a common experience in that somewhere in our path we were introduced to this technology and it sparked in us a curiosity, an itch, that we just had to find out more about.

We caught the bug.

What was that moment for you? What was the moment where you were like “This is some seriously cool business!!!”?

Mine was modulation.

I’ll never forget the moment where I found out about the way that 256 QAM works and I leaned over to my classmate and was like “My mind is completely blow… I can’t believe this stuff actually works”

I was taking the CWNA class by the magnanimous Rick Murphy from Wireless Training Solutions and he started to explain signaling techniques. He started by talking about that all symbols need to have meaning. And that in computing we assign meaning to bits… 1s and 0s. And different combinations of these two bits that are sent by the transmitter are interpreted by the receiver when the language is commonly understood.

Then he went on to simplify and solidify the concept by offering a visual allegory.  He said that if two friends were on the top of two buildings within visual range of one another and wanted to send information they could start with one piece of cardboard. The cardboard was black on one side and white on the other. And it was commonly understood among the friends that the white side was equal to the number 1 and the black side was equal to the number 0. And so if you wanted to send your friend across the way some piece of information you could hold up the white side for the 1 and the black side for the 0.

So black, black, black, white, black, white, black, white, white is interpreted by the receiver as 0,0,0,1,0,1,0,1,1.  And depending further up the level of interpretation would get you some kind of meaningful information (At the application layer).

If the data was robust, it would take a very long time to transmit the data this way so it would be much better if you could send combinations of bits. So you’d still need only cardboard, but now you would need enough colors for bit combinations of  00, 01, 10, 11. So perhaps you now have two pieces of cardboard, one with black on one side (00), and white on the other (11) and another with red on one side (01) and blue on the other (10).

So Black, white, blue, red, red, black, blue, white, black is interpreted by the receiver is 0,0,1,1,1,0,0,1,0,1,0,0,1,0,1,1,0,0. Same number of transmissions as the previous example, but twice the data.

This can go farther if we add an additional bit. 000, 001, 010, 011, 100, 110, 101, 111. If we assigned each on of those colors they would be black, gray, blue, red, green, pink, yellow, white. The problem we run into as we add colors is that the interpretation of those colors could get mixed up. Especially if something like a fog rolls in, or if the daylight starts to fade.  Pink could be interpreted as red, or green as blue. Which then causes data corruption and requires retransmission.

The two modulation techniques he outlined in these examples are BPSK (Binary Phase Shift Keying) and QPSK (Quadrature Phase Shift Keying). This was cool, but it hadn’t quite blown my mind just yet. When he started getting into QAM is really where it knocked me off my feet.

Obviously all of this talk about cardboard and colors is simply an analogy for a number of complex algorithms and error correction decision making that allows transmitters to adjust their RF waves to be modulated in just the right phase to be able to send the right combination of bits and for it to be received correctly. As the number of bits per transmission increases the harder it is for the data to be received correctly.

When I saw the constellation diagram of 64 QAM for the first time and recognized how minutely each of the waves had to be modulated for both amplitude and phase and how precise those changes had to be made in order to be able to not have data retransmission my mind was officially blown. Then the ante was upped again, when we look at 802.11ac and 256 QAM where 8 symbols per bit are transmitted. I couldn’t even believe that it was possible.

And yet, this goes on silently beyond our seeing all the time in the ubiquitous space around us. Even after studying Wi-Fi even further, my mind is still blown at our ability to manipulate the air in this way.

Mind. Blown.

Bug. Caught.

 

Wireless networks as we know them today is celebrating it’s 20th birthday this year. Happy Birthday Wi-Fi!

Wi-Fi has come a long way from 802.11 Prime with 1mbps and 2mbps data rates supporting STAs such as Personal Device Assistants and external PCI cards for laptops. Now Wi-Fi is common place, in fact not having is highly unusual and even car manufacturers are beginning to install wireless connectivity in vehicle so that end users have continuous connectivity from home, to school, to work to the coffee shop.

And in the time between 802.11 Prime in 1997 to now looking at 802.11ac wave 2 the number one goal has been speed. More speed. More speed to support high bandwidth applications such as YouTube and Netflix. More speed to support latency sensitive devices such as VoIP phones. Speed so you don’t have to wait even 2 seconds to receive an email, or wait for a webpage to load. Speed is tangible. Speed is sexy. Speed sells devices.

It would seem though, that the Wi-Fi community is at an impasse. How much more speed can we really suck out of our given RF medium? And will the end user even notice if we change their modulation scheme from 256 QAM to 1024 QAM with the introduction of 802.11ax? Or if we change from MIMO to MU-MIMO?

I think the answer to this question is… it depends.

I think for the average end user they’ll not notice a difference if their smart phone supports MU-MIMO or if it has 3 or 4 spatial streams. Except that, they may notice some significant challenges with the battery life of their device. I think that the implementation of the latest Wi-Fi technologies are at the mercy of battery technology. Until batteries can support more spatial streams and MU-MIMO while also being able to fit into smaller and smaller form factors, end users will stop paying attention to how fast their Wi-Fi is.

I think where the importance of newer Wi-Fi technologies lies is in the ability to plan for capacity. This is especially true in schools. The education environment is perhaps the only Wi-Fi critical environment where every user on an AP could be downloading or streaming the same content at the same time. Therefore some of the less common Wi-Fi technologies are going to become really critical for classroom environments where speed is related to capacity.

MU-MIMO is one of the technologies in 802.11ac Wave 2 that could be potentially very useful in a classroom environment.  It’s most useful when devices are equally separated around an AP, all or most devices are homogeneously supporting MU-MIMO (in a district provided 1:1 this would be presumed), and all of the devices are trying to receive roughly an equal amount of data.  This sounds like an ideal technology for a classroom where students would be working on roughly the same assignment at the same time where they are required to be on a particular website or book, or looking at a particular video.

The problem with the usefulness of this is that most device manufacturers are not going to spend the extra money on a chipset that supports MU-MIMO, or extra money trying to develop a battery that will support it because it is a unique and relatively small market problem to solve.

So is it worth it to keep investing in newer Wi-Fi technologies if device manufacturers are always going to take the cheapest route that covers the majority of the marketshare? I think so. I think it’s worth it to keep pushing the boundaries, but I also think it’s time that the Wi-Fi Communities governing bodies start working to establish not just standards for the PHY but standards for the devices as well.

End users will always have a better experience when device manufacturers and technology innovators are working hand in hand.