Practical limits of the secret key-capacity for IoT physical layer security

The confidentiality of communications in the Internet of Things (IoT) is critical, with cryptography being currently the most widely employed method to achieve it. Establishing cryptographically secure communication links between two transceivers requires the pre-agreement on some key, unknown to an external attacker. In recent years there has been growing interest for techniques that generate a shared random key through observation of the channel and its effects on the exchanged messages. The maximum length of that key is characterised by the Mutual Information (MI) between the observations of the two radios. In this work we examine the practical limits of the MI of off-the-shelf transceivers communicating through the IEEE 802.15.4 specification in an indoor office environment, and calculate the secret-key capacity, that is, the maximum length of an extracted secret-key in the presence of an eavesdropper. Furthermore, we study how using groups of observations can affect the MI and both analytically and experimentally prove that grouping observations leads to better results and an increased key-capacity.