0
\$\begingroup\$

I am an intern within the R&D branch of a company that designs tools that must operate with low noise at extremely high external temperatures (upwards of 175C). My boss asked that I look into finding a chip that would help prevent counterfeit tools from being connected to our power supply.

I'm looking for a chip that will check for a certain serial number or a "password" of sorts before allowing operation of the tool. In other words, the chip would expect to be given a correct response from the tool being attached to the power supply before allowing the tool to be operated.

I have found chips like the ATSHA204A (https://www.microchip.com/wwwproducts/en/ATsha204a), but I can't figure out how it should actually be connected to operate the way described above - I'm not even sure if this type of chip would be useful for what I have been tasked with.

\$\endgroup\$

closed as off-topic by brhans, Voltage Spike, Elliot Alderson, Dave Tweed May 22 at 12:30

This question appears to be off-topic. The users who voted to close gave this specific reason:

If this question can be reworded to fit the rules in the help center, please edit the question.

  • 2
    \$\begingroup\$ Why wouldn't you just use a microcontroller? \$\endgroup\$ – DKNguyen May 21 at 20:54
  • \$\begingroup\$ An off-the-shelf solution won't help you as any "counterfeit" instrument manufacturer can buy and use it. \$\endgroup\$ – Eugene Sh. May 21 at 20:55
  • \$\begingroup\$ The chip ATsha204a you linked to is only rated to 85C. I don't see it operating at 175C. \$\endgroup\$ – user4574 May 21 at 20:56
  • 3
    \$\begingroup\$ @Toor A tamper-resistant microcontroller protected from reading. And the handshake protocol with the instrument has to be encrypted or otherwise cryptographically protected. \$\endgroup\$ – Eugene Sh. May 21 at 20:56
  • 2
    \$\begingroup\$ Why do you want to prevent your power supply being used for other devices? This reminds me of the FTDI fiasco a little while back. \$\endgroup\$ – Hearth May 21 at 21:25
0
\$\begingroup\$

Use whatever $0.25 micro-controller you want. Put one controller in your power supply and another one in each device you want to attach.

The controller in both devices (supply and attached peripheral) contain a 128-bit secret key K.

The supply sends out a 128-bit random challenge word W that never repeats. The seed for the challenge word W should be stored using non-volatile memory in such a way that the seed does not reset if you loose power (to avoid play-back attacks).

The device sends back the SHA-512 hash H1 of K xor W.

The controller in the supply also computes the same hash H2. If H1 and H2 match then the attached device knows the secret key and is therefore authenticated. If H1 does not match H2 then the attached device is either broken or counterfeit.

This scheme avoids ever sending the secret key K between the devices. Therefore no-one can record it.

Any code-protection in the controller in both devices should be turned on so no-one can read out your secret key by opening the boxes.

\$\endgroup\$
  • 1
    \$\begingroup\$ Once and adversary owns one "genuine" device they can use it as an oracle to authenticate an arbitrary number of fake ones. But I guess this threat model is too advanced here. And of course the key once compromised, there is no way to revoke it. \$\endgroup\$ – Eugene Sh. May 21 at 21:14
  • \$\begingroup\$ @EugeneSh. Couldn't you still revoke it by allowing some code to order the chip to self-write some OTP fuses? \$\endgroup\$ – DKNguyen May 21 at 21:23
  • \$\begingroup\$ @Toor Sure you can if you implement such a mechanism. Also you need a way to trigger it. \$\endgroup\$ – Eugene Sh. May 21 at 21:24

Not the answer you're looking for? Browse other questions tagged or ask your own question.