Surpriv: RFID Surveillance and Privacy

Computer Crime Research Center and many others reports on Cisco’s offering technology for (active) RFID location and tracking:

Another new Cisco offering out of the Airespace labs is the Wireless Location Appliance 2700, which is designed to help customers track and locate 802.11 devices, such as laptops, PDAs and Wi-Fi enabled RFID tags, to within a few meters. This will be used for recovering lost property and for asset management applications.

Thus Cisco has come under fire from privacy groups as it prepares to launch a wireless RFID server that can track people and equipment using existing Wi-Fi networks.

(I’m reminded of the idea of using cell phone signal as the active part of a radar system for detecting targets… I think we’ll find “existing X,” e.g., existing Wi-Fi networks, RFID readers previously deployed for [this old application], etc., continually fitted with new applications, and, with more and more things being tagged, the product of [more readers] and [more tags] is lots of opportunities for tag readings and data collection.)

The BBC reports on research on RFID-aware robotic helpers for the blind, i.e., robots capable of identifying and reading various flavors of machine-readable tags (bar codes, RFID, etc.) to assist the blind in navigating/understanding the world.

The robots would be rather domain-specific (i.e., you’d pick one up upon entering the store, and return it when leaving); one could imagine a general-purpose assistant (think “magician’s familiar”) if it were cheap to annotate the physical world, e.g., putting tags on light poles at intersection pedestrian crossings, to allow for more informed navigation. That’s a bit reminiscent of this NSF-funded project.

Found on the HS-ARPA collaboration listings: http://www.hsarpasbir.com/currentopps.asp:

5/6/2005 Discreet Target Discrimination Eugene,OR
Marker Gene Technologies, Inc. is developing novel tagging and tracking systems that will be useful for distinguishing personnel or other tagged living organisms from non-labeled personnel in law enforcement, military or security operation zones. The tag is made up of a chemical reagent that is sensitive only to living tissues or organisms and can easily be sprayed, dropped or included into conventional cartridge ammunition or other munitions and thereby applied onto a target or target area giving it unique characteristics that generate several measurable signatures (e.g., UV absorption, UV fluorescence, and IR fluorescence). When interrogated with a UV/IR light source, the tagged individual or personnel can then be detected with an optical sensing device.

The system involves the use of optical (infrared, visible or ultraviolet) detection techniques to fully exploit the potential of this chemical tagging technology especially as a tactical tool in the surveillance of personnel movement, in the detection and recognition of sensitive personnel, and in a range of other law enforcement applications. The present proposal relates to the use of the chemical tagging agents and optical techniques to detect the absorption, transmittance, reflectance or fluorescence of the chemical taggants.
Contact: Dr. John J. Naleway jnaleway@uoregon.edu Phone: 541-342-3760

Discussion of GPS and pervasive monitoring.

[Satellite Security Systems Inc.'s] clients include school districts such as the District and Fairfax County, state and federal government agencies, police departments and companies. But there are plenty of individual customers, too — people interested in keeping tabs on new teenage drivers, Alzheimer’s patients, philandering spouses.

…

D.C. Public Schools is taking a more aggressive approach to monitoring. The information it receives on each bus and child is detailed: a driver’s route throughout the day, when the bus stops, when the doors open and close, the speed, and when the ignition is turned on or off. The system also features a database that will hold information on all the children — names, addresses, contact information, disabilities, allergies and when their school day begins and ends.

Where RFID will be largely a “point surveillance” technology, GPS provides a more continuous track, at the cost of willingness of the target to be observed, of course (apart from those situations where a covert device is used).

Silicom.com’s CIO Jury survey says, “RFID is overhyped.”

At the opposite end of the spectrum are several high-profile technology areas that CIOs believe are still over-hyped.

Almost unanimously RFID tracking tags were picked out as one of these over-hyped areas, and just one of the CIO Jury will be looking at this technology in 2005. Many said RFID will continue to be over-hyped in 2005.

Phil Young, head of IT operations at Amtrak, said: “Without standards it’s a turkey at the moment – forgive the pun at this time of year.”

AP reports on GPS used to track teens’ driving; the article was pegged to Gen. Tommy Franks (ret.) joining the company as the spokesperson for Teen Arrive Alive.

In a nutshell, phones equipped with GPS monitor their own speed, and call to alert the bearers’ parents if they exceed a pre-set limit. Presumably it wouldn’t be much of a stretch to program in “forbidden zones,” e.g., to keep one’s kid within a certain radius of the malt shop, etc.; just a tad more intelligence (and regular coordinate reporting) to implement “teen restraining orders” to keep young paramours apart, etc.

While RFID has been touted as a solution to theft/shrinkage in retail, it feels like that’s demanding enormous deployments (e.g., universal item-level tagging) that might be more efficiently addressed through other means. One obvious substitute technology would be video surveillance; here’s an SBIR (Smal Business Innovative Research grant) just awarded by the NSF for “video mining:”

SBIR Phase I: Video Mining for Customer Behavior in Retail Enterprises

This Small Business Innovation Research (SBIR) Phase I project aims at developing video mining techniques for automatically generating customer behavior statistics to help retail enterprises. These statistics can be very valuable for supporting critical decisions in merchandising, in-store marketing, and customer service. This video mining tool can be used for accurate assessment of the effectiveness of all consumer-facing elements in retail environments designed to promote or sell products. The key problems addressed will include view-independent person detection, multi-person tracking, a method for specifying behaviors, and robust behavior recognition. The approach will be to use a variety of computer vision and statistical learning techniques under the constraints of a typical retail environment. An experimental prototype will be implemented in Phase I and will use actual surveillance videos collected from retail enterprises to establish the feasibility of the approach…

This is pretty much the entire focus of the company that received the award: monitor people through video, and make inferences and take actions on the results. http://advancedinterfaces.com/index.shtml They don’t seem particularly shy about rather shpooky imagery of faces being profiled in a crowd…

(I ran across the cartoon below some time back, and have to heartily recommend The New Yorker as a great source for cleverly humorous presentation art!)

I’m pretty skeptical of the value of item-level RFID tagging in the retail environment, especially for such pipe dreams as RFID-based checkout (too fraught with problems from failed reads, and otherwise susceptible to fraud); use of RFID to deter shoplifting, or theft by employees, also seems to be a poor rationale to push for universal RFID tagging. It seems like there ought to be cheaper, more targeted alternatives, e.g., enhanced use of better systems for video surveillance. If it may take 5-6 (or more) years for item-level tagging to become more widespread, what will the machine vision researchers have achieved in that time?