Contactless bank cards

[Roland Perry]

In article 
<AANLkTi=4GEkzT1UkQ=TH-HJmAQXfikNproi1ArDnUQZz at mail.gmail.com>, 
Cybergibbons <cybergibbons at gmail.com> writes
>>> If you use USB, the latency and unpredictability causes issues with
>>> some card readers - this doesn't work to relay to TFL readers for
>>> example. However, you can use any microprocessor to perform the relay
>>
>> I don't really understand what you mean by "any microprocessor"
>
>Sorry - I meant you can use any simple microprocessor such as an AVR
>or MSP430 to perform the communication between the two PN53x.
>
>>> to avoid this, and use an RF link instead of wires.
>>
>> An RF link with more bandwidth than USB, presumably? Or is it just "less
>> latency per packet"?
>
>It's a question of latency - by the time the data leaves the reader,
>goes through some kind of USB interface, back to the PC, through an
>application on the PC, back down to another USB interface and into the
>emulator, the latency is high and unpredictable.
>
>Using something far more simple and ditching USB keeps the latency low
>and predictable.

If this is a proposal for an attack in random persons in the same shop 
as the crooks [you need one behind the till, and another out scouting 
for cards] (I think that's how it was supposed to play out) then you'd 
need something a bit more physically elegant than a laptop to be 
pressing up against the victims. So you'd suggest some sort of custom 
hardware built around the chips you mention, and with a fairly high 
bandwidth RF connection between them?
-- 
Roland Perry