The Internet of Things (IoT) is widely regarded as a key component of the Internet of the future and thereby has drawn significant interests in recent years. IoT consists of billions of intelligent and communicating “things”, which further extend borders of the world with physical and virtual entities. Such ubiquitous smart things produce massive data everyday, posing urgent demands on quick data analysis on various smart mobile devices. Fortunately, the recent breakthroughs in deep learning has enabled us to address the problem in an elegant way. Deep models can be exported to process massive sensor data and learn underlying features quickly and efficiently for various IoT applications on smart devices. In this article, we survey literatures on leveraging deep learning to various IoT applications. We aim to give insights on how deep learning tools can be applied from diverse perspectives to empower IoT applications in four representative domains, including smart healthcare, smart home, smart transportation, and smart industry. A main thrust is to seamlessly merge the two disciplines of deep learning and IoT, resulting in a wide-range of new designs in IoT applications, such as health monitoring, disease analysis, indoor localization, intelligent control, home robotics, traffic prediction, traffic monitoring, autonomous driving, and manufacture inspection. We also discuss a set of issues, challenges, and future research directions that leverage deep learning to empower IoT applications, which may motivate and inspire further developments in this promising field.

Exploiting the unique properties of the physical layer to enhance or complement authentication strength in wireless networks has attracted a lot of research attention recently. In this paper, we propose a novel PHYsical layer Challenge-Response Authentication Mechanism (PHY-CRAM) for wireless networks. PHY-CRAM is suitable for both one-way and mutual authentication. It fully utilizes the randomness, reciprocal, and location decorrelation features of the wireless fading channel, and is immune to various passive and active attacks. In the authentication procedure, challenge-response signals are exchanged at the physical layer, which allow two devices to verify their shared secrets while not revealing these secrets to attackers. PHY-CRAM adopts orthogonal frequency-division multiplexing (OFDM) technique which separately modulates the higher layer information and shared keys on subcarriers' phases and amplitudes respectively, in order to prevent channel probing from traffic-related information. We conduct extensive simulation study and develop a prototype using field-programmable gate array (FPGA) and discrete radio frequency (RF) components to evaluate PHY-CRAM in real-world environments. It shows that PHY-CRAM achieves both high successful authentication rate and low false acceptance rate in various channel conditions and under various attacks.