Smart meters

[Roland Perry]

In message , at 19:31:12 on Tue, 10 Mar 2015, john james remarked:

Sigh - you use DSP techniques.

[Mark]

I've just changed to a cheaper tariff with OVO and was not charged anything for the privilege. Were you locked into a fixed-rate contract at the time?

[john james]

Heavy breathing isn't going to save your bacon.

Not even possible to do that in this situation.

[Roland Perry]

In message , at 20:29:24 on Tue, 10 Mar 2015, john james remarked:

Darn, those post-grad security researchers were wrong.

[john james]

Easy to claim. Have fun citing even a single example of them doing that.

[Roland Perry]

In message , at 21:09:06 on Tue, 10 Mar 2015, john james remarked:

See bottom of page 7 here:

Although I predict you'll find something to quibble about, which is the problem when rising to the sort of bait you've been dangling.

Good luck, and goodbye.

[Capitol]

As far as I can tell from a quick read, the only statement on TV monitoring was one saying it may be possible. It may be possible to catch a bus to Tokyo, but it doesn't seem likely. If a TV station is captured on a PVR, and the resulting output is sent via scart, it is not possible to determine which of two systems is being watched. With only one TV, the chance of determining which channel is being watched is extremely close to zero as the power supply will wipe out most of the signal information. You will get lots of white noise. In addition, if you have ever worked on smart meters(which I have) you will rapidly find out that they are not very smart at isolating current changes. They are in general integrating devices. The level of computing power required to DSP a household current waveform is way beyond the economics of smart meters, which are designed down to a price. There is a good reason why spectrum analysers cost so much.

[Roland Perry]

In message , at

12:36:42 on Tue, 10 Mar 2015, Capitol remarked: [snip, some points which have been have been answered several times earlier in the thread]

The analysis in question isn't done in the smart meter, but by analysing fine-grained consumption figures it could be asked to send to a central facility.

[whisky-dave]

And that's without considering other devices such as a computer that also changes currrent consumption depending on what they are doing, and all those devicers charging or not charging.

[Andy Burns]

When I was half-way through a 2 year fix with Eon, I switched to a cheaper 1 year fix, expecting to pay the exit charge but I was told (in writing) they only hit you for the exit charge if you leave them altogether.

[john james]

Which says absolutely nothing about using it to work out what the individual backlight level is on two separate TVs each watching different programs from the total power consumption of the entire house.

THAT isn't even possible.

[john james]

And even if its technically possible with a single TV, its never going to be possible with two TVs watching different programs because it isn't going to be possible to work out what the backlight level is on each of the TVs from the total house current.

[john james]

No they have not.

Still not even possible to do that with the total current in the house to work out what current a particular TV is taking when you have two TVs watching two different programs.

Just proclaiming DSP cuts no mustard.

[john james]

Those aren't such a big problem because those change the current they are taking much more slowly than TVs change what is on the screen.

The other massive insurmountable problem is with stuff that isn't watched at the time its broadcast but watched from the PVR or from the net with one of the catchup or replay systems after its been broadcast.

Yes, it may well be possible to snoop on what people are watching who watch live broadcasts and who are only watching a single TV at a time, but that's it. Or more strictly what their TV is showing, anyway.

[harryagain]

You need to get her trained up. It's not clever leaving any unused appliance plugged in. This is how fires start.

[Jethro_uk]

I don't see you'd need metering there. As long as you know energy supplied + transmission losses (known factor) then you can deduce that if the sum of electricity paid for in an area is less, there's a bypassed meter.

It never ceases to amaze me that people undertaking one criminal enterprise - particularly one like growing cannabis, which can be made undetectable from the street - then go and make themselves a magnet by stealing electricity.

Most court reports of people arrested for growing cannabis will mention they bypassed the meter.

[Jethro_uk]

Which may catch a true commercial grow - not a 250/400W singleton.

Given how easy it would be to set up a proper industrial unit with lights and fans, on a busy industrial estate, I do wonder if these spectacularly amateurish cannabis "farms" they keep busting aren't simply decoys to keep the police busy and away from the *real* action ....

[Roland Perry]

In message , at 07:09:44 on Wed, 11 Mar 2015, john james remarked:

That's because you fundamentally fail to understand the process involved.

The signal processing separates out the components from numerous different appliances - two TVs being only a trivial subset of that.

[whisky-dave]

,

I thought that these places made a profit just from one growing seession ev en if they lose the equipment and product on the second batch. evenn if the street prices quoted are a bity 'high' (pun intended) at £10

-£50K the equipment doesn,t cost that much.

[whisky-dave]

How will it do that ? I don't think it can unless teh individual appliances to it themselves. I'll could ask as we run DSP labs here I have one running at 2pm.