Mobile keystroke dynamics for biometric recognition: An overview

Abstract

In the relatively recent past, the analysis of keystroke dynamics for biometric recognition purposes has intrigued researchers, since practical evidences have shown differences in the typing behaviours of distinct subjects. This area of research has become even more appealing since the emergence and evolution of mobile smartphones, given their pervasiveness and intensive use in real‐life applications. In addition, unlike hard keyboards used with computers, mobile smartphones offer the possibility of exploiting embedded sensors to augment the information acquired when typing on their soft keyboards. This study discusses the state of the art of keystroke‐dynamics‐based automatic recognition system, exclusively when dealing with mobile devices, for both verification and identification purposes. In more details, the databases employed, the features selected, the methodologies implemented, and the performance achieved through the technological advances introduced in the last years, are here overviewed.

Masterkey attacks against free-text keystroke dynamics and security implications of demographic factors

“…The original DSN dataset [25] also has a mobile equivalent, i.e., the mobile-DSN [9]. Mobile keystroke dynamics ML models are very similar to physical keyboard dynamics, with the difference that there are no key release timing data [26]. To accommodate this data loss, IMU sensor data is typically included.…”

Mobile behavioral biometrics for passive authentication

“…To date, behavioral biometrics for CA on mobile devices has been an active field of research for over a decade. Throughout the years, several studies have focused on the modalities explored within our work, employing a variety of different classification algorithms (mobile keystroke: [16,17]; touch data: [18,19,20,21]; background sensor data: [22]; multimodal biometrics: [23,24]). Regarding touch data information it is worth mentioning several studies in the literature.…”

“…In contrast, on the one hand, gesturebased related studies take into account unimodal systems [7,16,17,18,19,20,21]; on the other hand, related studies based on DL models for multimodal behavioral biometrics do not have a specific focus on human gestures [26,27]; ii) we analyze the information captured by the touchscreen in combination with simultaneous background sensor data to exploit the complementarity between taskdependent features and background sensors features (accelerometer, gravity sensor, gyroscope, linear accelerometer, magnetometer). Interaction database), a novel and public database comprising more than 5GB from a wide range of mobile sensor data acquired under unsupervised scenario for user passive authentication [30].…”

Mobile Keystroke Biometrics Using Transformers

“…Recent authentication methods propose to increase the security through an additional transparent layer based on the user's behavioural biometric information 1 , overcoming potential identity theft in a user-friendly and continuous way [2], [14]. Among the different behavioural biometric traits, keystroke dynamics is one of the most popular authentication methods in the literature [4], [10]. The information considered is the timestamp of the actions of pressing and releasing a key, together with the information of the key typed.…”

“…In addition, focusing on keystroke biometrics on mobile scenarios, many challenges must be considered to develop robust authentication systems. In this particular scenario, keystroke is typically acquired under uncontrolled circumstances, which can be affected by the user's activity, body position, emotional state, and the acquisition device [10], [17]. The performance might also be affected if the same subject is able to speak different languages [3].…”