Exploiting Induced skin electric potential for body-area IoT system functions

• Main Author: Yan, Zhenyu
• Other Authors: Tan Rui
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Computer science and engineering::Computer systems organization::Special-purpose and application-based systems; Engineering::Computer science and engineering::Computer systems organization::Computer-communication networks
• Online Access: https://hdl.handle.net/10356/137802
• Institution: Nanyang Technological University

Internet-of-Things (IoT) devices are increasingly deployed in indoor environments. In particular, there will be an IoT device proliferation in the household settings. For instance, it is estimated that by 2022, a typical family home could contain more than 500 smart devices. IoT devices are penetrating the industry environments to enable the realization of Industry 4.0. IoT devices are generally equipped with processing units, various sensors, communication modules, and software platforms to meet the application needs. A key characteristic of the IoT systems is the heterogeneity of hardware and software. Specifically, there are numerous IoT hardware platforms with diverse specifications and a number of embedded IoT operating systems (OSes) such as Android Things, Wear OS, Contiki, and Arduino. Such heterogeneity introduces significant challenges to the implementation of various basic system functions, such as clock synchronization and device authentication. Clock synchronization aims at keeping the clocks of the IoT devices under consideration having the same value all the time. However, the design of clock synchronization faces the trade-off between the synchronization accuracy and the universality over a variety of platforms. Specifically, high synchronization accuracy generally requires platform-dependent hardware-level network packet timestamping, which may not be available on any IoT platform. Thus, designing clock synchronization approaches to improve the Pareto frontier of accuracy versus universality is of great research interest. Device authentication verifies whether a device has a certain property (e.g., possession of a secret key). It can be used to manage the accesses to IoT devices. However, the existing device authentication approaches require cumbersome sensor hardware or user’s manual input. Thus, they are devised for specific devices and are generally inapplicable to a range of heterogeneous IoT devices. This thesis studies the clock synchronization and the device authentication problems for the IoT devices that can interact with the human bodies in the indoor environments. Such body-interacting devices include the wearable devices (e.g., smart watches, smart glasses, or medical sensors) and the indoor objects that the user can physically touch (e.g., light switches, remote controls for television, or voice assistants). This thesis designs and develops new clock synchronization and device authentication approach based on a new sensing modality called {\em induced skin electric potential} (iSEP) that provides desirable properties to solve the aforementioned challenges. Through extensive measurement studies on the electromagnetic radiation signals emitted from the electric power lines in indoor environments, we find that human body can act as an effective antenna and generate iSEP. The iSEP signal has three properties: First, it shows the same periodicity as the powerline voltages, which is 50 or 60 Hz. In particular, two iSEP signals have good synchrony across a large area, e.g., a city, due to the synchrony of the power grid voltages. Second, iSEP signals from the same location and the same body are similar, whereas those from different bodies are distinct. Third, sensing iSEP only requires a conductive wire connecting human body skin and the pin of an analog-to-digital converter (ADC). As ADC is widely available on most microcontrollers (MCUs), such a simple sensing mechanism can be universal for any IoT objects equipped with MCUs. Based on iSEP, we design TouchSync and TouchAuth systems for clock synchronization and device authentication for wearables and touchable IoT objects. TouchSync achieves millisecond synchronization accuracy while preserving universality in that it uses standard system calls only, such as reading system clock, sampling sensors, and sending/receiving network messages. It is based on the synchrony of iSEP signals collected from the same human body or different human bodies. TouchSync integrates the iSEP signal into the universal principle of Network Time Protocol and solves an integer ambiguity problem by fusing the ambiguous results in multiple synchronization rounds to conclude an accurate clock offset between two synchronizing objects. With our shared code, TouchSync can be readily integrated into any wearable applications. Example applications include delivering audio stream synchronously for wireless earphones, motion analysis of multiple on-body sensors, and gesture detection for multiple wearable gaming controllers. The design of TouchAuth is based on the electrostatics of iSEP generation and a resulting property, i.e., the iSEPs from two close locations and the same human body are similar. Extensive experiments verify the above property and show that TouchAuth achieves high-profile receiver operating characteristics in implementing the touch-to-access authentication policy. Our experiments show that a range of possible interfering sources including appliances’ electromagnetic emanations and noise injections into the power network do not affect the performance of TouchAuth. TouchAuth can be used in various applications with touchable IoT devices, e.g., family members wearing tokens can authenticate TV remote controller to access personalized programmes. In addition, the user with a wearable TouchAuth token can unlock the smart phone or authenticate a payment terminal by a simply touch. The designs of TouchSync and TouchAuth exemplify an important cyber-physical approach that leverages on certain properties of the physical processes (i.e., ambient electric field and generation of iSEP) to improve performance, usability, and security of the basic IoT system functions. Applying such a cyber-physical approach to implement a broader scope of system functions for a wider range of IoT systems beyond wearable and touchable devices will be of great research interest in the future work.