Hacker News new | comments | show | ask | jobs | submit login
Wi-FM listens to FM signals to determine best times to send and receive data (northwestern.edu)
43 points by zw123456 677 days ago | hide | past | web | 16 comments | favorite

The article is somewhat confusingly worded (the actual paper it links is more clear: http://networks.cs.northwestern.edu/publications/wifm/icnp20...) and I initially thought it was describing something like Microsoft's White Spaces project where devices would automatically transmit over unused spectrum. In reality, the idea is that the devices all automatically select timing relative to an FM Radio Data System transmission and then can infer each other's chosen timing without actually communicating (or they "implicitly communicate" in the words of the paper's author as quoted in the article).

So a transmission time slot selector seeded by FM signals. I haven't read it yet, but is there a good summary for how it handles varying loads? As the number of users and packets go up, you have to select smaller timeslots from the same FM signal seed.

Paper http://networks.cs.northwestern.edu/publications/wifm/icnp20...

> We show that the digital signal that accompanies broadcast FM radio has sufficient structure to enable effective scheduling relative to it. It thus provides a common reference for neighboring devices to harmonize their transmissions, yet without requiring any explicit communication among them.

The link to the paper provide more detail. The basic idea is to use the FM radio RDS sub-channel as a sort of control plane or control channel. One of the things that differentiates LTE for example from 802.11 (WiFi) is that LTE has a control channel that coordinates the activities, when to transmit an on which channel via a control plane and hence achieves a higher spectral efficiency. Wifi uses CSMA/CD to sense and detect collisions. But sensing when they happen is not as good as just avoiding it all together. But to do that you need a control channel. The idea of using something as ubiquitous as the RDS sub-channel as a control channel has a lot of potential applications beyond Wifi, it could also be used for white space control to open up vast amounts of un-used spectrum potentially.

Doesn't the WiFi standard already do this for devices running in the same channel? I.e. if 3 devices are on Channel 1 they will talk to each other to avoid conflicts. Hence why you should only use one of 3 channels (the only 3 which never overlap)

That is the popular wisdom, but in my experience it's not universally true. I have had good results in an apartment building by choosing channels between 1, 6, and 11. Signal strength sometimes went up quite a bit.

For example, if there are 3 APs on each of the three non-overlapping channels, using a channel between those channels may provide better results than directly competing with 3 other APs for the entire channel.

However, it also depends on the observed signal strength for each of the competing APs. For best results, you should probably test your equipment.

But if you really want decent transfer rates between devices, just use Ethernet. I've given up on ever reaching even 50% of rated speeds on wifi. With competing APs, walls, distance, etc, it's just never going to happen.

Or if you can't run ethernet cables, I've had quite good success in my house with a powerline adapter (like this one: http://www.amazon.com/Netgear-Powerline-Extra-Outlet-PLP1200...).

Couldn't vouch for how well it'd work in an apartment setting, though.

I think this is a bad idea... to use non-standard channels. See: http://superuser.com/questions/443178/is-it-better-to-use-a-...

You should read the links you cite before you cite them. Among the answers found on that page are these:

"The 1-6-11 recommendation contained in Cisco's whitepaper about IEEE 802.11 deployment in the corporate environment certainly does not apply to all circumstances! For example, in moderately congested neighbourhoods, one stands a very good chance to benefit from not sticking to this proposed scheme."

"the situation usually gets worse when one voluntary abides to the 1-6-11 non-overlapping channel scheme. Doing so will expose your devices to the IEEE 802.11 RTS/CTS/ACK (Request to Send / Clear to Send / Acknowledge) of alien devices, effectively silencing your devices and hence forcedly lowering your bandwidth. This problem is known as the exposed node problem. In a corporate setting this problem can be solved by synchronising the nodes. In the wild, this is not readily achievable."

"In big corporate networks it's common practice to use channels 1,6 & 11 because it is fairly straightforward (at least on a diagram) to design non-overlapping cells of coverage. As a home user you don't have the same constraints so it makes sense to experiment and look for the best channel."

"I am a Ham Radio operator. I have done extensive testing. On my Actiontec or ZyXcel, channel 1 is abysmal! Channel 11 is a close second to the death of channel 1. ACTUAL power readings put 3 and 4 as the strongest signal output and throughput. Channels 6 and 9 are the standard preset. so actually avoid 1,6,9,11. I am a DSL tech also. I have walked people through changing channels from 9-10-11 to 3 or 4. They are amazed at the doubling of the wifi signal on all devices across the board."

"Hence, I call for actually measuring one's own signal-to-noise level...The Quality value takes into account noise from overlapping channels."

The bottom line is as I said: test your own equipment in your own environment and find what works best for you.

The solution to all wireless problems is... wires.

Makes me hope that "li-fi" pans out.

Basically combo powerline data with rapidly blinking (1000+ hz, iirc) leds to be able to give specific rooms wireless data. And if the light is polarized, a simple coating on the windows that match can let light in, but not data out.

Better yet, directional li-fi through transparent panels on your devices guided by compact adjustable optics.

Example of what tech such optics could use:

Solid-state variable optics: https://www.osapublishing.org/oe/abstract.cfm?uri=oe-18-13-1...

Compact non-conical optics: http://www.nasa.gov/feature/goddard/out-with-the-old-in-with...

Actually need to blink closer to 10^8 Hz :)

Is this still a problem for 5Ghz wifi?

Url changed from http://phys.org/news/2015-11-wi-fm-fm.html, which points to this.

There is a MAC you know...

Applications are open for YC Winter 2018

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact