[Make-wifi-fast] Tinfoil hats for WiFi routers

[Make-wifi-fast] Tinfoil hats for WiFi routers

[Toke Høiland-Jørgensen]

Supposed to be 3D-printed to point the signal in the right direction. Can't find the paper, just the press release: https://www.eurekalert.org/pub_releases/2017-11/dc-aw110717.php

-Toke

Re: [Make-wifi-fast] Tinfoil hats for WiFi routers

[Bruno Antunes]

The links for the papers are on the youtube video description:

https://www.youtube.com/watch?v=1KVBFtSuCm0&feature=youtu.be

http://www.cs.dartmouth.edu/~xia/papers/buildsys17-3dprint.pdf

https://homes.cs.washington.edu/~jucha/docs/3dprint-hotw.pdf

On 28 November 2017 at 17:57, Toke Høiland-Jørgensen <toke@toke.dk> wrote:
> Supposed to be 3D-printed to point the signal in the right direction. Can't find the paper, just the press release: https://www.eurekalert.org/pub_releases/2017-11/dc-aw110717.php
>
> -Toke
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast

Re: [Make-wifi-fast] Tinfoil hats for WiFi routers

[dpreed]

[-- Attachment #1: Type: text/plain, Size: 1804 bytes --]


Thanks. Reviewing the paper, all it claims is that "signal strength" is increased in desired areas. The impact on MIMO and DSSS decoding of the phase mixing is unclear, but the reflector appears to be small - about the same size as the AP..
 
There is some assertion at the beginning of the paper that it is more fine grained than MIMO in its control. But they don't demonstrate that in the paper, and don't provide any theoretical justification.
 
So it's kind of cute, but I'm skeptical of its actual utility relative to other "dynamic" approaches that can optimize for situations like door positions, furniture movement, etc.
 
A hybrid approach including a reflector and 4x4 MIMO might be good, but getting that optimized is more complex than just optimizing RSSI.
 


On Tuesday, November 28, 2017 1:29pm, "Bruno Antunes" <baantunes@gmail.com> said:



> The links for the papers are on the youtube video description:
> 
> https://www.youtube.com/watch?v=1KVBFtSuCm0&feature=youtu.be
> 
> http://www.cs.dartmouth.edu/~xia/papers/buildsys17-3dprint.pdf
> 
> https://homes.cs.washington.edu/~jucha/docs/3dprint-hotw.pdf
> 
> On 28 November 2017 at 17:57, Toke Høiland-Jørgensen
> <toke@toke.dk> wrote:
> > Supposed to be 3D-printed to point the signal in the right direction. Can't
> find the paper, just the press release:
> https://www.eurekalert.org/pub_releases/2017-11/dc-aw110717.php
> >
> > -Toke
> > _______________________________________________
> > Make-wifi-fast mailing list
> > Make-wifi-fast@lists.bufferbloat.net
> > https://lists.bufferbloat.net/listinfo/make-wifi-fast
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast

[-- Attachment #2: Type: text/html, Size: 3174 bytes --]

Re: [Make-wifi-fast] Tinfoil hats for WiFi routers

[dpreed]

[-- Attachment #1: Type: text/plain, Size: 2095 bytes --]


Though it is said to have "worked" with 802.11ac, I wonder what you do to preserve diversity in the 4 antennas, esp. if not in a straight line.
 
In principle, diffusive scattering actually helps MIMO performance, but the scattering has to be "random" to be most effective.  For example, if you put up a flat mirror behind the 4 antennas, and you picture the coverage as a "directed beam" (which is not an ideal way to think about 4x4 MIMO at all, but it's easier to picture), you'll end up with a reflected version of the backside of the beam that goes off in a different direction from the first.  That would "split" the energy so the backlobe gets wasted because it does not reach the intended receiver.
 
OFDM has an advantage in reflective environments, because the symbol duration is quite large compared to the value c/f where c is lightspeed and f is the central frequency of the band.  This means that phase mixing of the subcarriers due to reflections/scattering has much less impact on OFDM.
 
So, if you can concentrate the transmitted energy from the four antennas of the source onto the four antennas of the receiver, even with phase shifts, you can do better.
 
But 4x4 MIMO is already doing a lot of concentration of energy, so I would think that no matter what the device is, it won't help 802.11ac as much as it would help non-MIMO systems.
 
And 802.11b (DSSS) would suffer from symbol distortion due to phase mixing. But that depends on the physical dimensions of the device that is wrapped. If small compared to the symbol duration * c, then it won't matter that much.
 
 
 
 
 


On Tuesday, November 28, 2017 12:57pm, "Toke Høiland-Jørgensen" <toke@toke.dk> said:



> Supposed to be 3D-printed to point the signal in the right direction. Can't find
> the paper, just the press release:
> https://www.eurekalert.org/pub_releases/2017-11/dc-aw110717.php
> 
> -Toke
> _______________________________________________
> Make-wifi-fast mailing list
> Make-wifi-fast@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/make-wifi-fast

[-- Attachment #2: Type: text/html, Size: 4181 bytes --]