In order to address the aforementioned goals and questions the discussions focused on the following topics having running discussion around the ACE charter and active drafts in mind:
The outcome of the meeting can be summarized as follows. Concerning the discussion on layer dependencies the team agreed on starting with solutions in the application layer to support high flexibility, and for the future look on layers below. Concerning the assumed architectures some parts, especially the backend, should be assumed to be a black box, but the devices are able to speak to it in a transparent way. Furthermore, the solutions must get along with sleepy devices that is common when addressing energy harvesting issues to ensure long life time of the system. DTLS approaches are great, but might have drawbacks when looking on big networks over hundreds of devices. DTLS introduces overhead to the system, but this is not a problem if devices have enough resources. Thus, the team agreed on not limiting the solutions within ACE to devices upto 256 kByte RAM, instead included also bigger devices as constraint devices but not in the size of servers. The main concept is to keep the devices smart! Furthermore, the solutions and uses-cases should be as general as possible to be applicable for complexed ones later. Key management plays a big role especially in wireless communication. The participants agreed on not limit the cryptography used to secure communication. So all solution can use either asymmetric or symmetric cryptography, but should advise where to use what. Furthermore, key establishment should not be limited as well, but for secure distribution trust within the system must be supported (e.g., trustworthy authority). Additional information about ACE activity is available under IETF. |
News >