Extracting Randomness from the Trend of IPI for Cryptographic Operations in Implantable Medical Devices

摘要: Achieving secure communication between an Implantable Medical Device (IMD) and a gateway or programming device outside the body has showed its criticality in recent reports of vulnerabilities cardiac devices, insulin pumps neural implants, amongst others. The use asymmetric cryptography is typically not practical solution for IMDs due to scarce computational power resources. Symmetric key preferred but security relies on agreeing using strong keys, which are difficult generate. A generate shared keys without extensive resources, extract them from physiological signals already present inside such as Inter-Pulse interval (IPI). must therefore be sources randomness that meet five conditions: Universality (available all people), Liveness at any-time), Robustness (strong random number), Permanence (independent history) Uniqueness other sources). However, these conditions (mainly last three) have been systematically examined current methods extraction IPI. In this study, we first propose methodology measure three Information secrecy measures , Santha-Vazirani Source $delta$ d e l t value dependency analysis . Then, large dataset IPI values (almost 900,000,000 IPIs), show does source. Thus, uniform number impossible. Third, trend IPI, instead value, source new method named Martingale Randomness Extraction (MRE-IPI). We evaluate MRE-IPI it satisfies condition completely some level. Finally, NIST STS Dieharder test suites able outperform IPIs achieves quality roughly half AES generator. still cannot used communications general. can one-time pad securely exchange parties. usage will thus kept minimum reduces probability breaking it. To best our knowledge, work area uses comprehensive examine signals.