Forensic Identification of Environmental Sensors Through Challenge-Based Biometrics

As Internet of Things (IoT) sensors continue to become increasingly prevalent in everyday technology, smart homes and smart environments are becoming more feasible. Environmental gas sensors -- which include various technologies, types, and sizes -- are an important aspect of smart systems. These sensors collect crucial data that can be used to analyze and interpret the surroundings of the user while being able to alarm the user if any potentially dangerous analyte is in the environment. The differences that naturally exist in both the type of sensor and the raw, uncalibrated response to varied environments and environmental analytes give the opportunity for a level of forensic environment through securing a home with biometric principles. This paper proposes a method of using challenge-based biometrics, or biometrics that result in a response from the object or person that they are being applied to, to secure a smart home environment forensically with "abiometrics;" that is, biometrics approaches applied to a non-living item or object. The MQ sensor series (specifically, the MQ-5, MQ-7, and MQ-135) were utilized in this study to test whether an environmental change or change in exposure to gas, which are both commonly used calibration measurements in controlled environments, are simultaneously effective for challenge-based abiometrics of sensors. It was shown that a range of four to seven environmental settings would be required to achieve a one in a billion chance that the sensor was forensically validated.