دانلود رایگان مقاله تاثیر خصوصیات مهاجم در مکانیزم احراز هویت مستمر مبتنی بر ECG برای اینترنت چیزها

abstract

Wearable devices enable retrieving data from their porting user, among other applications. When combining them with the Internet of Things (IoT) paradigm, a plethora of services can be devised. Thanks to IoT, several approaches have been proposed to apply user data, and particularly ElectroCardioGram (ECG) signals, for biometric authentication. One step further is achieving Continuous Authentication (CA), i.e., ensuring that the user remains the same during a certain period. The hardness of this task varies with the attacker characterization, that is, the amount of information about the attacker that is available to the authentication system. In this vein, we explore different ECG-based CA mechanisms for known, blind-modelled and unknown attacker settings. Our results show that, under certain configuration, 99.5 % of true positive rate can be achieved for a blind-modelled attacker, 93.5 % for a known set of attackers and 91.8 % for unknown ones.

6. Conclusions

The use of wearable devices to extract biosignals that can be shared leveraging the Internet of Things (IoT) opens up the door to promising security applications. In this paper, we have focused on the use of ElectroCardioGram (ECG) signals for Continuous Authentication (CA). Such application is possible thanks to IoT, enabling an authenticator to process ECG data. However, a proper design of an IoT-enabled CA mechanism needs to take the attacker into account. Thus, the key difference with existing works is that we present three different mechanisms for known, unknown and blind attacker settings. In this way, we study the effect of attacker characterization. Our results exhibit promising accuracy figures, which support the use of ECG data as an identifier. Moreover, balanced practicability and reasonable easiness for the set-up are achieved in the three settings.

Future work will have three main directions. First, the use of variable ECG records (e.g., data recorded during physical activities) will be considered. Second, the use of another vital signals (e.g., EEG or GSR) will also be explored in the context of CA. Finally, the adoption of mobile-edge or fog computing schemes in this context will be assessed, taking into consideration the underlying security and privacy requirements.