Browse Prior Art Database

Method for Robust Tap Force Augmented Keystroke Dynamics Analysis for Mobile Devices

IP.com Disclosure Number: IPCOM000249853D
Publication Date: 2017-Apr-18

Publishing Venue

The IP.com Prior Art Database

Abstract

Keystroke dynamics biometrics provides a non-intrusive and low cost authentication mechanism for computer users. It further allows for continuous authentication for improved secure access. With the ubiquity of mobile devices and people's increasing dependency on these devices to store sensitive and even critical information, keystroke dynamics has been explored to authenticate mobile device users. Studies have suggested that although keystroke dynamics serves as an effective authentication scheme for mobile devices, it comes short of meeting strong security requirements when used as the sole criterion. In this work we note that the force a user exerts on keys during typing is as idiosyncratic as the timing information. As a result, we extract a user's tapping signature using the motion captures by the accelerometer and gyroscope embedded in a mobile device during typing events to augment the performance of keystroke dynamics biometrics for mobile devices, usually computed using the key timing information. We in particular use gait dynamics image, a highly discriminative representation for motion captured by 3-axis accelerometers and gyroscopes while invariant to sensor rotation, to represent taping signatures due to forces during keying events to achieve invariance to variations in orientation of the mobile device for real-world applications. As the tapping force and time are complementary to each other, this method is expected to provide augmented keystroke dynamics based user authentication for mobile devices.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 14% of the total text.

Method for Robust Tap Force Augmented Keystroke Dynamics Analysis for Mobile Devices

                                                         Abstract

Keystroke dynamics biometrics provides a non-intrusive and low cost authentication mechanism for computer users. It further allows for continuous authentication for improved secure access. With the ubiquity of mobile devices and people’s increasing dependency on these devices to store sensitive and even critical information, keystroke dynamics has been explored to authenticate mobile device users.  Studies have suggested that although keystroke dynamics serves as an effective authentication scheme for mobile devices, it comes short of meeting strong security requirements when used as the sole criterion. In this work we note that the force a user exerts on keys during typing is as idiosyncratic as the timing information. As a result, we extract a user’s tapping signature using the motion captures by the accelerometer and gyroscope embedded in a mobile device during typing events to augment the performance of keystroke dynamics biometrics for mobile devices, usually computed using the key timing information. We in particular use gait dynamics image, a highly discriminative representation for motion captured by 3-axis accelerometers and gyroscopes while invariant to sensor rotation, to represent taping signatures due to forces during keying events to achieve invariance to variations in orientation of the mobile device for real-world applications. As the tapping force and time are complementary to each other, this method is expected to provide augmented keystroke dynamics based user authentication for mobile devices.

Keystroke dynamics is a well-investigated behavioral biometric which provides a natural choice for secure “password-free” computer access [2, 3, 4, 11, 32]. Keystroke dynamics refer to the habitual patterns or rhythms an individual exhibits while typing on a keyboard input device.  These rhythms and patterns of tapping are idiosyncratic [7, 13], in the same way as a person’s handwriting or signature, due to their similar governing neurophysiological mechanisms. In fact, as early as the 19th century, telegraph operators could recognize each other through their specific tapping styles.  This suggests that keystroke dynamics contain sufficient information to serve as a biometric identifier. Keystroke dynamics biometrics has been an active research area for the past decade.

Keystroke dynamics features are usually extracted using the timing information of the key down/hold/up events. The hold time or dwell time of individual keys, and the latency between two keys, i.e., the time interval between the release of a key and the pressing of the next key are typically exploited.  Digraphs, which are the time latencies between two successive keystrokes, are commonly used. Trigraphs, which are the time latencies between every thre...