Whither RFID Security

1
Whither RFID Security?
2
A few questions to bootstrap the discussion

• Are RFIDs just retarded (anemic) sensors that can
  only communicate with a reader (sink, in sensor
  jargon)?
• Put another way, are RFIDs just sensors that
  don't sense anything?
• If so, why invent new solutions? Can't we just
  re-use what was done for sensor security?
• BTW, Under what conditions can RFIDs "talk" among
  themselves? Should they?

3
A few questions to bootstrap the discussion

• What, if any, degree of tamper-resistance is
  attainable with today's (and near-future) RFIDs?
• Is there way too much hype surrounding the whole
  issue of RFID privacy?
• For example, can Joe User simply neutralize
  them, e.g., by waving around a strongly-magnetic
  "magic wand"?

4
A few questions to bootstrap the discussion

• One of my favorite perpetually open problems in
  Computer Science
• Can a computer provably forget?
• I believe the answer is NO
• If an RFID is a computer, is that still the case?
  
• E.g., if you know that an RFID has only t bits
  of storage and you ask it to store a t-bit value
  (and check that it did by asking it to echo it),
  does that mean that it has provably forgot what
  it stored earlier?

5
A few questions to bootstrap the discussion

• Are RFIDs only able to perform XORs and other
  trivial boolean ops? Are simple modular
  arithmetic operations -- e.g., add, multiply,
  square, inverse -- viable?

6
RFID Mythology

• What is the real range of RFIDs?
• Various fantastic-sounding claims have been made
  about being able to read RFIDs over 20 meters