Virus.Org  IT Security News and Information Portal. We offer the latest IT security news, updates, product reviews, books, and articles for all you IT security professionals out there. Enter and get the best IT security information on the Internet.

 

. Welcome to the Virus.Org Mailing List Archive  
.
.


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]


Re: Disabling RFID by mains force
.

  • To: [EMAIL PROTECTED]
  • Subject: Re: Disabling RFID by mains force
  • From: Stefek Zaba <[EMAIL PROTECTED]>
  • Date: Sat, 1 Nov 2003 22:15:09 +0000
  • In-reply-to: <[EMAIL PROTECTED]>
  • References: <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
  • Reply-to: [EMAIL PROTECTED]
  • Sender: [EMAIL PROTECTED]
.
 
On Fri, Oct 31, 2003 at 07:01:45PM -0000, Dave Howe wrote:
> > Actually there are protocols that can count the number of items being
> > scanned at any one point in time, even when these have the same ID.
> Interesting - do you have any online references?
www.rfid.org is an obvious place to start, while the RSALabs paper
on selective blocking gives a good description of the "individuation"
protocol.

> remember we are talking about a number of RFID tags, in a 3D volume that
> could well contain an assortment of metal reflective surfaces, where worst
> case only the left and right hand "panels" (possibly the floor too, if the
> unit is seated into the flooring) are available for receivers/emitters.
> Possibly this can be done in a fully enclosed volume - say a "cupboard"
> arrangement where you insert trolly, close door, and scanning takes
> place - but it seems very hard to me.
> 
The tags cooperate in a recursive-descent enumeration. Roughly, the
reader says "first bit 0?" and all tags with first-bit-0 speak up
with their second bit. If the reader hears responses implying the
presence of tags with both "00" and "01" prefixes, it recurses down
both branches of the tree; if it hears a response from only one
subbranch, it recurses down only that branch. This way, it gets the
uniqIDs of all responding tags in a number of steps bounded above by
<number-of-tags-present> times <bitlength-of-a-tag-ID>,

As I handwaved in another response, the uniq-IDs are structured, with
bits at the top being a manufacturer-of-item code, bits in the middle
being an item-made-by-that-mfr code, and bits at the end being the
item-serial-number. So apps which only care about how many items you
have don't bother indexing into any DB with the full ID (though they
still "read" it so as to know how many items of a set mfr-item-type
they've seen), while the ones which care, can.

As I understand it, manufacturers who stick RFID tags onto their
pallets/boxes/individual-items typically don't program them themselves;
for now at least, they buy a reel or sheet of tags pre-programmed by
their tag supplier, and stick 'em on. Some tags have a "read-write"
area beyond the "read-only" ID portion, for those apps which want to
carry updated info around on the tag rather than in the Gert Big
Database.

HTH, Stefek


 
.
.
 
Copyright (c) Virus.Org 1997-2006.
All Trademarks Acknowledged.
Please view our Terms and Conditions and our Privacy Policy.