Evaluating Secrecy Capacity for In-Body Wireless Channels

Abstract

The next generation of implanted medical devices is expected to be wireless, bringing along new security threats. Thus, it is critical to secure the communication between legitimate nodes inside the body from a possible eavesdropper. This work assesses the feasibility of securing next generation multi-nodal leadless cardiac pacemakers using physical layer security methods. The secure communication rate without leakage of information to an eavesdropper, referred to as secrecy capacity, depends on the signal-to-noise ratios (SNRs) of the eavesdropper and legitimate channels and will be used as a performance metric. Numerical electromagnetic simulations are utilized to compute the wireless channel models for the respective links. These channel models can be approximated with a log-normal distribution which can be used to evaluate the probability of positive secrecy capacity and the outage probability of this secrecy capacity. The channels are modeled for three different frequency bands and a comparison between their secrecy capacities is provided with respect to the eavesdropper distance. It has been found that the positive secrecy capacity is achievable within the personal space of the human body for all the frequency bands, with the medical implant communication systems (MICS) band outperforming others.