Full Program
Summary:
The paper discusses the challenges of publish-subscribe communication schemes, particularly the MQTT standard, where a central broker facilitates message exchange between clients. While this method offers advantages over direct communication, it poses privacy risks since the broker can monitor and analyze the messages, gaining insights into client behavior. Traditional encryption fails to protect message topics, as the broker needs to interpret them. To address this issue, the paper proposes Functional Encryption (FE), a sophisticated public-key encryption method. With FE, the broker can process encrypted inputs without learning their content, revealing only the output of the function. Additionally, this approach allows for randomized message topics for each message type, enhancing privacy. The authors plan to evaluate the concept's computational overhead and scalability, aiming to improve security in publish-subscribe systems.Author(s):
Marten Fischer
University of Applied Sciences Osnabrueck
Germany
Ralf Tönjes
University of Applied Sciences Osnabrueck
Germany