From: Peter Williams (home_pwmsn.com)
Date: Sat Apr 02 2005 - 13:15:53 CST

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

I'd like a little help from any knowledgeable person on tuning the
inductance properties of a "proposed open source smartcard antenna", for an
educational toolkit board we built at ventavia.com for experimenting with
commodity smartcard technology. The board allows one to study wireless uses
and configuration of musclecards in various operational theatres - where you
can add suitable electronics for each environment, to otherwise commodity
components. I'd like to ship the board one day as an educational toolkit,
with some cutting edge experimentation cookbook that gives the buyer a head
start in open design issues.

The board can be used in various inductance loop environments - such as
those built widely into local city streets across the US. These centralized
traffic light systems already allow occupancy estimation, and per-vehicle
inductive-signature correlation between remote intersections. to allow for
surveillance of movement of persons in automobiles for a variety of
purposes: legal fact collection following intersection accidents, traffic
condition monitoring, flow rates, police incident response, and who knows
what covert signals analysis purposes! The question now is: how can and how
easily can the existing infrastructure usefully ping an antenna and
non-antenna enabled musclecard, while you are passing over any road
intersection? The complementary question then exists, what are the
appropriate design properties for musclecard cardedge security/privacy that
address this environmental opportunity/threat.

It costs about $500 to manufacture each board - which are professionally
manufactured, but hand tuned/tested for component choices. In mine, I
mounted a commodity-grade GEMPLUS combi SIM, and set jumpers to access the
board's onboard antenna. The antenna has parameters similar to those found
in commercial USB dongles supporting combi-cards, and one can mount scoping
points easily for field analysis on suitable scoping equipment. The board
also has an external antenna attachment point for use in with an alternative
Near Field antenna, and also other experimental antenna such as that to be
design in the study I'm proposing. If you can fabricate your own LQFP
packaged chips (e.g. the university EE lab) you can also play with antennas
within the chip packaging itself.

The design issues that seem pertinent to characterize are not necessarily
those of addressing id tracking and movement privacy, per se. Its not the
big brother conspiracy - that is really the design focus, here. The focus is
perhaps more about the degree to which general infrastructure can activate
or interfere with configured security features, remotely, or the correct
reporting of capabilities to discovery processes. It's the question of
whether, for example, the signals can remotely disable the bus to the crypto
co-processor, remotely undelete data in certain types of EEPROM technology,
or deposit covert signature information into the device's eeprom
technologies electrical fields. It's the issue of technology _interaction_
that are perhaps the real study topics - electronic, electrical, RF, and
memory technologies. Then we consider what software OS and applet design
supporting a feature discovery capability should do, to accommodate
commodity die with we can assume will have such properties.

Moving feature discovery from the host driver or traditional terminal (which
rarely move) to the USB dongle reader (which moves with the chip) or to the
card itself has various systemic impacts that we should understand - before
choosing how to best update the current cardedge. Beyond creating proposed
software changes, we have to look for and consider wider issues.

Peter.

_______________________________________________
Muscle mailing list
Musclelists.musclecard.com
http://lists.drizzle.com/mailman/listinfo/muscle

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]