[Cryptography] Gaslighting ~= power droop == side channel attack

[David Richardson]

A little thumbnail electrical engineering work on the supercap idea:


On 11/29/2016 11:29 PM, Ray Dillinger wrote:
> On 11/29/2016 04:36 AM, Chris Tonkinson wrote:
>
>> I wonder if some Simple™ device could be created based on a combination
>> of battery, capacitor, and timer to smooth the "curves" of consumption
>> for a household (or on a smaller scale for specific appliances),
>> rendering this type of analysis ineffectual.
>
> Sure.  Attach between the meter and the breaker switches a high-current
> bridge rectifier, and on the other side of the bridge rectifier put the
> biggest supercapacitor you can get.
>
> It'll absorb transient spikes and droops leaving only the very tiniest
> fluctuations for the snoopy device to look at.
>
> Upside: bonus, some of your appliances may last longer.  Smoothing out
> little spikes and droops helps avoid some kinds of damage.
>
Hmmmm....  Does anyone make a supercap rated for 400V at 100A? Nominal 
calculated fully rectified 240VAC is about 340VDC IIRC, so that will be 
your supercap working voltage. Calculating for 240V/100A full load 
output with unity power factor, the "float" current flow - rectifier to 
inverter - is about 70.6A but there is always power factor loss with 
motors (heat pumps and dryer) and the inverter that converts it back to 
AC has quite a few losses too.

Another problem is finding a 240V inverter to attach to your capacitor 
to convert the DC back to 240 AC at 100A. Finally, put a 30 KVA 
240V:240V w/CT isolation transformer ahead of your fuse box to make 
everything work like you are used to.

What this basically amounts to is an always-on UPS, and a very expensive 
one at that.