Navigation services based on Global Navigation Satellite Systems (GNSS) are essential for a wide range of global applications, including ensuring the safety of aerial vehicles. However, these GNSS signals (e. g., civilian GPS) are vulnerable to jamming and spoofing attacks. To mitigate such threats, we propose GridNet, a vision-based deep-learning technique to counter GPS jamming and spoofing attacks on aerial vehicles. GridNet uses visual devices (e.g., cameras) to determine geolocation during such attacks by extracting geolocation from aerial photos via a pre-trained neural network model. This non-invasive approach does not modify existing GPS infrastructure, merely relying on real-time visual data. Unlike other methods that focus solely on attack detection, GridNet provides a countermeasure, calculating geolocation data without GPS. We analyze the potential applications and discuss the performance in different flight environments. Experimental results show that GridNet can extract geolocation data from real aerial photos, achieving nearly 93% spoofing detection accuracy with a minimum average geolocation error of 7.33 meters within 4ms on a ground station server and 2.7ms on a typical UAV, respectively, offering a practical anti-spoofing solution.

With the rapid ubiquity of Internet of Things (IoT) technology, a growing number of devices are being connected to the Internet, thereby increasing the potential for cyberattacks. For instance, due to firmware compatibility issues and release delays, N-day vulnerabilities pose significant threats to IoT devices that run outdated firmware versions. Consequently, accurately and efficiently identifying firmware versions of devices is crucial for detecting device vulnerabilities and enhancing the security of IoT ecosystems. In this work, we present DeepFW, which utilizes a Fusion Feature Attention Network (FFAN) to extract subtle differences in embedded web interfaces within the firmware, facilitating the identification of firmware versions in online IoT devices. To address the challenge of high similarity between versions caused by firmware homogeneity in the supply chain, we propose a novel metric loss, namely the Hard Mining Cosine Triplet-Center Loss (HCTCL), to improve intraclass compactness and inter-class separability. To validate the effectiveness of our method, we collected 4,442 firmware images and obtained $\mathbf{1 3 0, 4 4 5}$ valid embedded web pages. Experimental results show that DeepFW outperforms the state-of-the-art approaches by over $25 \%$ on average in both precision and recall. Furthermore, DeepFW revealed that only $2.28 \%$ of devices in our dataset were running the latest firmware version. Our evaluation also indicates that $\mathbf{6, 6 8 4}$ devices (approximately $\mathbf{6 1. 2 6 \%}$) with outdated firmware versions remain vulnerable to known exploits.

Home and workplace networks typically safeguard against external threats but allow internal devices to communicate freely with each other. As a result, malicious code on an internal device can collect sensitive data about other devices or directly attack them. In this paper, we study mobile apps as potential sources of local network attacks, analyse their behaviour, design new defences, and evaluate and bypass existing mitigations. We first focus on Android, where apps with only the Internet permission can access all devices in the Local Area Network (LAN), meaning malicious apps can extract private LAN data, manipulate discovery protocols to obtain a Machine-in-the-Middle (MitM) position, and directly attack devices. To defend against such mobile-based attacks, we define an access model to securely differentiate between LAN and global Internet access. We implement this model on Android by creating LANShield: an app that refines Android's permission model, and can monitor and block LAN access of apps using a virtual network interface. We use LANShield to manually perform tests of 399 Android apps and find, among other observations, that 89 apps unexpectedly access the LAN, and 93 apps scan the network. In contrast to Android, iOS already separates the local and global Internet, but does so based on a proprietary LAN access model. We compare this access model to ours, and present multiple bypasses for an app to circumvent Apple's local network permission. Finally, we reported all our findings to affected vendors, and hope our work will motivate the adoption of stronger permission models on mobile devices.

Wireless reprogramming is a significant yet challenging issue in the Internet of Things (IoT). Existing methods designed for wireless sensor networks (WSN) are inadequate for IoT scenarios and have exposed severe security vulnerabilities. To address this problem, we propose Software-Defined Function (SDF), a secure and wireless reprogramming architecture for IoT named SFIOT. The key is to implement a secure communication interface between the control layer and the infrastructure layer. To this end, a set of security protocols which ensure authentication and confidentiality are designed, and their security is proved theoretically and formally verified. In addition, the constrained capabilities of IoT devices are taken into consideration in our design. We have tested the performance of SDF through a set of experiments and presented a use case on a photovoltaic energy system. The evaluation results show that the proposed protocols can be implemented in real-world IoT applications.

Multiple Object Tracking (MOT) has been a useful yet challenging task in many real-world applications such as video surveillance, intelligent retail, and smart city. The challenge is how to model long-term temporal dependencies in an efficient manner. Some recent works employ Recurrent Neural Networks (RNN) to obtain good performance, which, however, requires a large amount of training data. In this paper, we proposed a novel tracking method that integrates the auto-tuning Kalman method for prediction and the Gated Recurrent Unit (GRU), and achieves a near-optimum with a small amount of training data. Experimental results show that our new algorithm can achieve competitive performance on the challenging MOT benchmark, with higher efficiency and more robustness compared to the state-of-the-art RNN-based online MOT algorithms.

The recent surge in the prosperity of the Internet of Things (IoT) has been attracting an increasing number of researchers and experts with great attention due to its significant economic and social values. The IoT brings appealing opportunities and new challenges for both the current and future Internet. In practice, various IoT smart devices are generally pre-programmed and deployed specifically in the proper place to fulfill corresponding functions according to divergent requirements. However, lately, these pre-stored functions tend to be upgraded or reprogrammed more frequently on account of the increment of dynamic needs or urgent situations. Inspired by Software Defined Networking (SDN), the authors propose a framework in this work: Software Defined Function (SDF) for IoT, enabling IoT smart devices to be upgraded or reprogrammed securely and remotely. The authors further present a protocol named as OpenFunction stemmed from OpenFlow. Moreover, the security properties of this protocol are analyzed. Finally, the authors implement a preliminary SDF system and evaluate its performance. Experimental results indicate that OpenFunction allows a controller to update or rewrite functions in IoT devices, as well as to obtain flexibility and security. Accordingly, this work contributes to the future fusion of SDN and IoT technologies.

Unmanned Aerial Vehicles (UAVs), better known as drones, have significantly advanced fields such as aerial surveillance, military reconnaissance, cadastral surveying, disaster monitoring, and delivery services. However, UAVs rely on civilian (unauthenticated) GPS for navigation which can be trivially spoofed.

This paper studies the problem of Person Re-Identification (ReID) for large-scale applications. Recent research efforts have been devoted to building complicated part models, which introduce considerably high computational cost and memory consumption, inhibiting its practicability in large-scale applications. This paper aims to develop a novel learning strategy to find efficient feature embeddings while maintaining the balance of accuracy and model complexity. More specifically, we find by enhancing the classical triplet loss together with cross-entropy loss, our method can explore the hard examples and build a discriminant feature embedding yet compact enough for large-scale applications. Our method is carried out progressively using Bayesian optimization, and we call it the Progressive Learning Algorithm (PLA). Extensive experiments on three large-scale datasets show that our PLA is comparable or better than the-state-of-the-arts. Especially, on the challenging Market-ISOI dataset, we achieve Rank-1=94.7%/mAP=89.4% while saving at least 30 % parameters than strong part models.