4 ms·

Microsoft Ponders UV Password Tattoos to Protect Pacemakers

5 points by one010101 16y ago

nfnaaron 16y ago
Makes me think of mandatory ID tattoos. I realize that's a leap, but that is what first came to mind.
- contextfree 16y ago
  The paper actually brings this up. "The use of UV ink may be of less consolation to those patients who associate tattoos with their use as involuntary identifiers. In the holocaust, identification tattoos reminded prisoners no longer controlled their own bodies. Giving the patient a choice of whether or not to use micropigmentation, the type of encoding to use, and some control over the process that generates the encoding should help to address these concerns."
  - cstross 16y ago
    Also note, IIRC tattoos are forbidden in judaism (http://www.myjewishlearning.com/practices/Ethics/Our_Bodies/Adorning_the_Body/Tattoos.shtml http://www.myjewishlearning.com/practices/Ethics/Our_Bodies/...) and possibly other religions. (Although there's a halachic "out" for forbidden practices where the intention is to save lives, so this might be negotiable for medical emergency access to a pacemaker.)
teuobk 16y ago
The problems with these types of techniques is that they don't necessarily improve security, but they do add complexity. In current medical devices that support so-called "distance telemetry," the implanted device is bonded with a base device using "proximal telemetry." Proximal telemetry works only at distances of a few centimeters or less and tends to be limited by the sensitivity of the receivers within the implanted devices. Only after the authentication and key exchange has occurred in the proximal mode is the distance mode activated, which then affords the users several meters of freedom. The assumption is that a person who can stay close enough to the patient for the length of time required for the devices to bond is implicitly worthy of access. This is the same assumption that has served the industry well for the past few decades, although before the distance devices, the entire session would be conducted in proximal mode. The bigger risk, as I see it, is that an attacker could successfully impersonate an already-bonded device. In that situation, neither proximal bonding nor UV passwords would help.

5 points by one010101 16y ago

nfnaaron 16y ago

Makes me think of mandatory ID tattoos. I realize that's a leap, but that is what first came to mind.

contextfree 16y ago
The paper actually brings this up. "The use of UV ink may be of less consolation to those patients who associate tattoos with their use as involuntary identifiers. In the holocaust, identification tattoos reminded prisoners no longer controlled their own bodies. Giving the patient a choice of whether or not to use micropigmentation, the type of encoding to use, and some control over the process that generates the encoding should help to address these concerns."
- cstross 16y ago
  Also note, IIRC tattoos are forbidden in judaism (http://www.myjewishlearning.com/practices/Ethics/Our_Bodies/Adorning_the_Body/Tattoos.shtml http://www.myjewishlearning.com/practices/Ethics/Our_Bodies/...) and possibly other religions. (Although there's a halachic "out" for forbidden practices where the intention is to save lives, so this might be negotiable for medical emergency access to a pacemaker.)

teuobk 16y ago

The problems with these types of techniques is that they don't necessarily improve security, but they do add complexity. In current medical devices that support so-called "distance telemetry," the implanted device is bonded with a base device using "proximal telemetry." Proximal telemetry works only at distances of a few centimeters or less and tends to be limited by the sensitivity of the receivers within the implanted devices. Only after the authentication and key exchange has occurred in the proximal mode is the distance mode activated, which then affords the users several meters of freedom. The assumption is that a person who can stay close enough to the patient for the length of time required for the devices to bond is implicitly worthy of access. This is the same assumption that has served the industry well for the past few decades, although before the distance devices, the entire session would be conducted in proximal mode. The bigger risk, as I see it, is that an attacker could successfully impersonate an already-bonded device. In that situation, neither proximal bonding nor UV passwords would help.