Pascal Thubert (pthubert) writes:
> But then, if the JN does not know K1, how can it validate the MIC?

It does not need to validate it. MIC is security feature, it is not
meant to be used to protect against transmission erros, even when it
will also detect those.

Your comment would be about the same as saying: The IPsec/AH provides
the 96-bit ICV, and if the other end does know the key (or we do not
use the AH at all), how it can know that it received the TCP packet
properly.

> Do we have another way to protect against transmission errors on
> EBs?

Yes. Same as we normally use for non-secured frames. The level of
protection provided depends on the PHY used. It might either be
2-octet FCS, or it might be 4-octet FCS, or it might be some kind of
forward error correction code (FEC) etc.

None of these feaures are disabled when we enable the security. I.e
when we receive the the secured frames we first check the FCS, if that
is right, then we go through the filterin rules to see if is for us,
and if it is meant for us, only then we do security processing. For
the beacons received during active/passive scanning there is special
case that even if the security processing fails, we still use the
contents of the beacon, for other cases if the security processing
fails, we drop the frame (of course the 802.15.4e does not say whether
you still process the IEs included in the frame).
-- 
[email protected]

_______________________________________________
6tisch mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/6tisch

Reply via email to