Page 1 of 1

1457:"Feedback"

Posted: Mon Dec 08, 2014 7:04 am UTC
by keithl
Image

title text: A new study finds that if you give rats a cell phone and a lever they can push to improve the signal, the rats will chew on the cell phone until it breaks and your research supervisors will start to ask some questions about your grant money.

Looking at this XKCD at a B&B in Berkeley using wireless. Riots a few blocks away - reception varies with news helicopter position.

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 7:49 am UTC
by Eternal Density
Forget wireless, I'm having enough problems getting good reception over wired broadband! Signal to noise ratio keeps dropping and disconnecting everything a few times a day :(

But there are a few spots in my house where I can get a mobile phone signal. Sometimes. So yeah on the rare occasions I try I end up standing on my bed or in the bath.

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 8:00 am UTC
by azule
You place it, however it needs to be orientated, in the window. After that, no touchie! Tethering is now your best friend.

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 8:30 am UTC
by TheEngineer
RIM suggest that holding an apple works best with their products.

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 9:24 am UTC
by da Doctah
I think the more important question is: why aren't you standing on a chair holding a pineapple?

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 10:08 am UTC
by Neil_Boekend
I was a minute ago, but I chose a folding chair. Reception was great though.

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 1:02 pm UTC
by rmsgrey
Back in the day, we used to do similar things with TV antennae.

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 1:55 pm UTC
by orthogon
rmsgrey wrote:Back in the day, we used to do similar things with TV antennae.

(Ahem) Terrestrial TV is still a thing!

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 2:20 pm UTC
by rmsgrey
orthogon wrote:
rmsgrey wrote:Back in the day, we used to do similar things with TV antennae.

(Ahem) Terrestrial TV is still a thing!


Yeah, but since the digital switchover, there's no "barely watchable" - it's either full quality or garbage, with no feedback on improved signal...

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 2:30 pm UTC
by Neil_Boekend
rmsgrey wrote:
orthogon wrote:
rmsgrey wrote:Back in the day, we used to do similar things with TV antennae.

(Ahem) Terrestrial TV is still a thing!


Yeah, but since the digital switchover, there's no "barely watchable" - it's either full quality or garbage, with no feedback on improved signal...

Dunno about DVB-T, but most DVB-S recievers can get simple signal and quality indicators on screen. Those can be used to direct the dish, assuming a near infinite amount of patience (there is usually a few seconds delay in this signal. That means that the feedback loop is severely damaged)

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 4:23 pm UTC
by suso
I was wondering if this was some reference to Hak5's wifi pineapple:

http://hakshop.myshopify.com/products/w ... t=81044992

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 10:29 pm UTC
by Flumble
suso wrote:I was wondering if this was some reference to Hak5's wifi pineapple:

http://hakshop.myshopify.com/products/w ... t=81044992

Oh wow, I didn't know there were pentesting routers on the market. Maybe I should buy one someday. Wifi access in trains* is unreliable at best, so maybe that should help.

*don't judge us Dutch! :o

Re: 1457:"Feedback"

Posted: Mon Dec 08, 2014 10:40 pm UTC
by azule
Any reason some think this is about Wi-Fi? It is clearly about cell phone signals.

Re: 1457:"Feedback"

Posted: Tue Dec 09, 2014 3:07 am UTC
by steaxauce
I think this is a good analogy for the stock market. It's basically what most traders do for a living.

Re: 1457:"Feedback"

Posted: Tue Dec 09, 2014 4:27 am UTC
by Farabor
azule wrote:You place it, however it needs to be orientated, in the window. After that, no touchie! Tethering is now your best friend.


You're assuming the router is orientable......

(Okay, I've been doing too much topology lately. Anyone want the first homotophy group of a sphere adjoined to a circle by a single point?)

Re: 1457:"Feedback"

Posted: Tue Dec 09, 2014 7:03 am UTC
by Neil_Boekend
azule wrote:Any reason some think this is about Wi-Fi? It is clearly about cell phone signals.

Wifi and cell signals encounter the same kind of problems.

Re: 1457:"Feedback"

Posted: Tue Dec 09, 2014 11:51 am UTC
by Kit.
Neil_Boekend wrote:
rmsgrey wrote:Yeah, but since the digital switchover, there's no "barely watchable" - it's either full quality or garbage, with no feedback on improved signal...

Dunno about DVB-T, but most DVB-S recievers can get simple signal and quality indicators on screen. Those can be used to direct the dish, assuming a near infinite amount of patience (there is usually a few seconds delay in this signal. That means that the feedback loop is severely damaged)

Any digital RF receiver can do it. Not every one shows it to the user, though.

The signal quality information that is (or can be) available:
1. The signal level (say, in dBmV).
2. The signal/noise ratio. Every digital sample is a discrete dot in a signal phase space. Noise makes real samples to form (probabilistic) clouds around that ideal dots. The distance between the closest dots divided by the "typical" cloud size gives the SNR. It will require some time to calculate the cloud size statistics.
3. The rates of recoverable and non-recoverable digital errors. Digital RF signal is protected by error-correcting codes, so a receiver can assemble their statistic as well. Also, some time is required to gather the statistics.

Re: 1457:"Feedback"

Posted: Tue Dec 09, 2014 12:09 pm UTC
by Mental Mouse
This is a classic example of the dire effects of random reinforcement: In B.F. Skinner's classic experiments with operant conditioning, he noted that animals (in the classic experiment, pigeons), could also be "conditioned" by an arbitrarily presented stimulus. Rereading the article (it's been a while), I see that he actually used a regular clock, from which I infer that pigeons are pretty bad at recognizing short time intervals. But randomly timed reinforcement will do the same thing, the key point being that the reward is not actually dependent on the animal's behavior. Skinner himself drew the connection to superstition and "luck" rituals....

Re: 1457:"Feedback"

Posted: Tue Dec 09, 2014 1:00 pm UTC
by Neil_Boekend
Kit. wrote:
Neil_Boekend wrote:
rmsgrey wrote:Yeah, but since the digital switchover, there's no "barely watchable" - it's either full quality or garbage, with no feedback on improved signal...

Dunno about DVB-T, but most DVB-S recievers can get simple signal and quality indicators on screen. Those can be used to direct the dish, assuming a near infinite amount of patience (there is usually a few seconds delay in this signal. That means that the feedback loop is severely damaged)

Any digital RF receiver can do it. Not every one shows it to the user, though.

All DVB-S (digital satellite TV) recievers I have seen and worked with (quite a few. I used to do tech support in retail, so I handled probably 50 different models) did show that on screen. For aligning the dish it wasn't as useful as an E15 satbeeper but it was an incredible useful diagnostic tool.
Before your post I assumed that DVB-T (digital ground based) receivers had a similar feature.

Re: 1457:"Feedback"

Posted: Tue Dec 09, 2014 1:58 pm UTC
by Kit.
Neil_Boekend wrote:
Kit. wrote:
Neil_Boekend wrote:
rmsgrey wrote:Yeah, but since the digital switchover, there's no "barely watchable" - it's either full quality or garbage, with no feedback on improved signal...

Dunno about DVB-T, but most DVB-S recievers can get simple signal and quality indicators on screen. Those can be used to direct the dish, assuming a near infinite amount of patience (there is usually a few seconds delay in this signal. That means that the feedback loop is severely damaged)

Any digital RF receiver can do it. Not every one shows it to the user, though.

All DVB-S (digital satellite TV) recievers I have seen and worked with (quite a few. I used to do tech support in retail, so I handled probably 50 different models) did show that on screen. For aligning the dish it wasn't as useful as an E15 satbeeper but it was an incredible useful diagnostic tool.
Before your post I assumed that DVB-T (digital ground based) receivers had a similar feature.

A typical DVB-T (or -T2) "receiver" is just a TV set that is designed to work with digital channels, as well as with analog ones. They may have service menus that allow qualified personnel to look at these data, but it's not how they are supposed to work for an end-user, who normally just puts an antenna cable into the TV set (and usually the antenna is already pre-installed and pre-oriented on the roof), then asks it to automatically scan for the list of the available channels. Then the digital channels are supposed to just work.

Re: 1457:"Feedback"

Posted: Tue Dec 09, 2014 8:00 pm UTC
by slinches
Kit. wrote:Then the digital channels are supposed to just work.

Yes, this works great for everyone who is within a few miles of the towers, has unobstructed line of sight and no other interference sources. For the other 90% of people in the "coverage area" (that used to receive poor, but watchable, analog reception) it's a black box that doesn't work. Luckily, my TV does have an accessible signal level indicator.

It took a whole day of wandering around on my roof to find the best spot for the antenna, but now I get 4 of the 5 major local network broadcasts and several of the others (~35 total channels). Well worth the effort considering that I'm only paying about $10 per month in replacement pineapples instead of $80 on cable TV.

Re: 1457:"Feedback"

Posted: Tue Dec 09, 2014 8:50 pm UTC
by azule
I thought I saw one channel come in. I just have of those digital receivers (a paper sized rectangle). Are you talking about also using a giant on-roof antenna?

Re: 1457:"Feedback"

Posted: Wed Dec 10, 2014 1:26 am UTC
by slinches
Yes, an on-roof antenna, but not giant at about 3x4ft as mounted. By the way, for anyone in the US who is looking into OTA broadcasts, there are decent coverage maps at the FCC and tvfool.com which gives much more detailed technical info on the available channels.

Re: 1457:"Feedback"

Posted: Tue Jan 27, 2015 3:03 pm UTC
by sfnhltb
Mental Mouse wrote:This is a classic example of the dire effects of random reinforcement: In B.F. Skinner's classic experiments with operant conditioning, he noted that animals (in the classic experiment, pigeons), could also be "conditioned" by an arbitrarily presented stimulus. Rereading the article (it's been a while), I see that he actually used a regular clock, from which I infer that pigeons are pretty bad at recognizing short time intervals. But randomly timed reinforcement will do the same thing, the key point being that the reward is not actually dependent on the animal's behavior. Skinner himself drew the connection to superstition and "luck" rituals....


Derren Brown did a fun human version of the experiment a few years back as well. Can probably find it on youtube or whatever but the ones I found were locked in my region.