Anonymizing communication flow identifiers in the Internet of Things

Abstract

We tackle the problem of privacy breaching in IPv6 over Low-Power Wireless Personal Area Networks (6LoWPANs) where an attacker may be able to identify the communicating entities. We propose a solution based on: (i) anonymizing both IP and MAC addresses, as well as port number at the source host, using a random pseudonyming scheme, and (ii) anonymizing the IP address and port number of the destination host, using a Tor-like network. We analyzed the effect of the Tor entry node location on the performance of our solution in three different scenarios: the Tor entry node is located (i) inside the 6LoWPAN, (ii) at the IPv6 Low-power Border Router (6LBR), or (iii) completely outside the 6LoWPAN. Using Cooja simulator, we showed that our solution outperforms state-of-the-art solutions by making it more difficult to identify communication flows by improving the anonymity and unlinkability of the communicating entities without significantly affecting energy consumption, communication delay, and network bandwidth.