Multi-party applications are becoming popular due to the development of mobile smart devices. In this work, we explore Secure Device Pairing （SDP）, a novel pairing mechanism, which allows users to use smart watches ...Multi-party applications are becoming popular due to the development of mobile smart devices. In this work, we explore Secure Device Pairing （SDP）, a novel pairing mechanism, which allows users to use smart watches to detect the handshake between users, and use the shaking information to create security keys that are highly random. Thus, we perform device pairing without complicated operations. SDP dynamically adjusts the sensor＇s sampling frequency and uses different classifiers at varying stages to save the energy. A multi-level quantization algorithm is used to maximize the mutual information between two communicating entities without information leakage. We evaluate the main modules of SDP with 1800 sets of handshake data. Results show that the recognition accuracy of the handshake detection algorithm is 98.2%, and the power consumption is only 1/3 of that of the single sampling frequency classifier.展开更多
基金supported in part by the National Natural Science Foundation of China (Nos. 61472219 and 61672372)Shaanxi NSF (No. 2017JM6109)
文摘Multi-party applications are becoming popular due to the development of mobile smart devices. In this work, we explore Secure Device Pairing （SDP）, a novel pairing mechanism, which allows users to use smart watches to detect the handshake between users, and use the shaking information to create security keys that are highly random. Thus, we perform device pairing without complicated operations. SDP dynamically adjusts the sensor＇s sampling frequency and uses different classifiers at varying stages to save the energy. A multi-level quantization algorithm is used to maximize the mutual information between two communicating entities without information leakage. We evaluate the main modules of SDP with 1800 sets of handshake data. Results show that the recognition accuracy of the handshake detection algorithm is 98.2%, and the power consumption is only 1/3 of that of the single sampling frequency classifier.
