Keystroke Dynamics Identification Based on Triboelectric Nanogenerator for Intelligent Keyboard Using Deep Learning Method. Issue 1 (16th August 2018)

• Record Type: Journal Article
• Authors: Zhao, Guangquan; Yang, Jin; Chen, Jun; Zhu, Guang; Jiang, Zedong; Liu, Xiaoyong; Niu, Guangxing; Wang, Zhong Lin; Zhang, Bin
• Is Part Of: Advanced materials technologies. Volume 4:Issue 1(2019)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2018-08-16
• Subjects: deep belief network -- feature extraction -- intelligent keyboard -- keystroke dynamics identification; Materials science -- Periodicals; Technological innovations -- Periodicals; Materials science; Technological innovations; Periodicals; 620.1105
• DOI: 10.1002/admt.201800167
• Languages: English
• ISSNs: 2365-709X

Abstract

Abstract: Due to the heavy reliance on computers and networks, security issues have become a major concern for individuals, companies, and nations. Traditional security measures such as personal identification numbers, tokens, or passwords only provide limited protection. With the development of intelligent keyboard (IKB), this paper proposes a deep‐learning‐based keystroke dynamics identification method for increased security. The IKB is a kind of self‐powered, nonmechanical‐punching keyboard, which converts mechanical stimuli applied to the keyboard into local electronic signals. Multilayer deep belief network (DBN) is established to mine the useful information from raw electronic signals and output the keystroke dynamics identification result. The contributions include development of a novel solution that does not rely on manual feature extraction, and provides promising recognition accuracy on large amount of typing samples. One significant advantage of the proposed method is that it extracts features adaptively from the raw current signals and automatically recognizes the typing pattern, which simplifies the design of verification and identification system. The experimental results on 104 typing samples demonstrate the effectiveness of the proposed method. The proposed method has extensive applications in keyboard‐based information security. Abstract : A novel keystroke dynamics identification method is developed for intelligent keyboard using deep learning technologies. Experiment results on 104 typing datasets show that the proposed method has high identification accuracy, stable and reliable performance.