• controlled publishing
• distributed system
• machine to machine
• peer to peer
• security

Machine To Machine (M2M) is a network paradigm, based on self organizing devices that automatically retrieve and share information from the surrounding environment. It could be used to implement several features of the Future Internet and offering several services. An efficient implementation of such a system could be achieved through Peer To Peer (P2P), that is a distributed and scalabale network communication paradigm known for its efficiency and reliability.
Handling security on these system is fundamental, since the information handled on these systems may also be sensitive, thus it is necessary to ensure data confidentiality and integrity. Furthermore, by means of access control, we wish to avoid that a malicious user harms the system behavior, filling the network with malicious data or denying the service access to the other users.
In this thesis we designed a security architecture, distributed and scalable in order to handle the security issues of such a system. We analyzed a specific implementation of the M2M paradigm, based on a P2P architecture finding its security issues and designing a method to solve them. Indeed, we focused on the P2P typical security issues, such as Sybil Attack and on the specific issues introduced by this specific implementation. Furthermore, we tried to be as less invasive as possible, in order to ensure security with a minimal impact on the system performance. We obtained these results by means of an access control system that we named “Controlled Publishing” and the design of a Distributed Certification Authority.