Pattern Regulator: Protecting Temporal Usage Privacy for Wireless Body Area Sensor Networks

摘要

In a wireless body area sensor network (WBAN), biosensors are implanted or worn on an individual to acquire medical data for a clinical diagnosis or physical monitoring. Traditionally, authentication and cryptography techniques are used to achieve the data confidentiality and privacy of a WBAN system. However, for WBAN systems, besides data itself, the pattern of data transmission may leak critical information about the user. For example, various medical sensors deliver data in different patterns and these patterns leak many critical facts, e.g., what sensors are used (which implies diseases the individual may have), and how frequent the sensor is sampling (which implies how urgent some abnormal observation is about a patient). In this work, to address this challenge, we design a regulator, which packs real data session into a type-independent transmission model at transmission layer. All valid data packets are equably sent at the frequency defined by the regulator and at the same length to clutter the inherent pace of valid data transmission and other parameters.We also propose a strategy PAS to minimize the overhead while preventing attackers from locating the patients. Our extensive experiments validate our regulator and PAS design.