The problem is that you're trying to cross layer boundaries and IEEE 802.15.4 was never intended to allow such usage.
Which brings us back to the two kinds of coexistence. In one case, we have coexistance of 6lowpan, zigbee, wireless hart. This is easily solvable with Kris' proposal. In the other, we have coexistance of YFNP and OTAP. Which 6lowpan has provided provisions for with the use of NALP.
IEEE 802.15.4 specifies a link key--a key that can be used to validate traffic between two hosts at the data link level. What you are describing is network keying, which is beyond the scope of IEEE 802.15.4. YFNP, YFNPng, and OTAP can all specify their desired security scheme and keying approach (for example, OTAP may implement a network-wide keying scheme while YFNP is a multicast protocol that uses group keying). Either way, these are concerns above the data link layer, and we should not confuse security at the network layer with security at the data link layer.
There's no confusion here. Kris' argument was based on the fact that an 802.15.4 4-byte MIC provide better integrity checks than 16-bit CRCs while also providing a convenient mechanism for segmenting a collection of nodes. With link-layer MICs in place, then a simple and now plausible solution is to have some bits in the payload define what upper-layer protocol is running. Having a 6lowpan-specific key still works since NALP protocols are necessarily aware of 6lowpan, since they conform to its protocol-identifier mechanism. Another way is, as you suggest, to provide network-level keying for segmentation, which then gets back to the "how do i determine what key to use" problem.
-- Jonathan Hui [EMAIL PROTECTED] _______________________________________________ 6lowpan mailing list [email protected] https://www1.ietf.org/mailman/listinfo/6lowpan
