Highly equipped computers (laptops or desktops) have been used as the preferred means of not only assessing online services but of staying connected with the rest of the world. However, with their enhanced capabilities and relative low cost, besides a few high-end vendors, smartphones (SPs) have replaced computers as our primary computing devices, especially for secure services. Nowadays, SPs or smart watches (SWs) are ubiquitously integrated into our work and home environment, and individuals frequently use SPs/SWs as a portal to access cloud-based services. Moreover, it is common practice for users to store on their SPs personal, sensitive, and confidential information, such as bank account information, credit card numbers, emails, passwords, photos, and videos.

In addition to malware and viruses that can spread, theft is also a very serious threat. A stolen or lost SP/SW can easily be used for identity theft and the sensitive information stored on it can then be used for several malicious purposes. This has therefore highlighted the importance of reliable identification and authentication of legitimate users and blocking imposters. It is important to develop novel and state-of-the-art methods for SP and SW-based legitimate user identification and authentication, which need to be both fast and accurate: accurate because both false positives and false negatives should be as infrequent as possible and fast because ideally, it should detect the impostor at the same moment that the SP/SW is stolen and not with the legitimate user accidentally triggering the alarm or some other mechanism.

This Special Issue aims to present novel ideas as well as systematic overviews and surveys of deep learning and artificial intelligence techniques as well as state-of-the-art strategies for securing smart devices and for legitimate user identification and authentication.

Potential topics include but are not limited to the following:

• Security implications for smartphones and smart watches
• Legitimate user identification and authentication
• Runtime impostor recognition
• User behavior analysis (physical activity recognition) and physical activity-based identification/authentication
• Machine learning and deep learning for smart solutions
• Gate-based identification/authentication
• Other methods of identification and authentication, including password-based, multi-factor, certificate-based, biometric-based, or token-based