Secrecy Optimization for Diffusion-Based Molecular Timing Channels

Security in the context of molecular communication systems is an important design aspect that has not attracted much attention till date. This letter analyzes the information-theoretic secrecy of diffusive molecular timing channels when the distance of the eavesdropper is assumed to be random and uniformly distributed. Using an existing upper bound on the timing channel capacity, we calculate the optimal secrecy rate and optimal transmission rate for Bob which would help in achieving an improved secrecy throughput performance. Based on this optimal rate, we calculate the maximum achievable throughput. We then use this formulation to minimize the generalized secrecy outage probability (GSOP) by simultaneously maximizing the average fractional equivocation and minimizing the average information leakage rate. The numerical results show that while choosing the system parameters, there is always a trade-off between different performance metrics like GSOP, average fractional equivocation, and average information leakage rate. The proposed secrecy optimization provides a robust understanding of the physical layer secrecy at the molecular level, enabling the design of secure molecular communication systems.