Chao Yang

Collecting key-value data from IoT devices is crucial for providing users with more accurate recommendation services and other applications. However, existing collection protocols for key-value data cannot provide differential privacy protection for data with varying sensitivities, which affects data utility. In this paper, we introduce a novel utility-optimized local differential privacy definition for key-value data, UKVLDP. This definition enables key-value data of different sensitivity levels to be properly protected by appropriately relaxing the strict LDP definition, and adaptively perturbing them according to their different sensitivities, thereby improving the utility of the collected data while protecting the privacy of the key-value data. Based on UKVLDP, we propose a privacy-sensitivity-calibrated key-value data sampling algorithm for the privacy-preserving collection of key-value IoT data of different sensitivities, which solves the distortion problem of the random uniform sampling method when the key domain space is too large. In addition, we propose a key-value data collection protocol UKV-GRR that conforms to the UKVLDP definition, which achieves a better privacy-utility trade-off by adaptively perturbing the sampled key-value data. We theoretically prove that UKVLDP can provide optimized privacy guarantees. Evaluations on simulated and real datasets demonstrate that UKV-GRR outperforms the state-of-the-art key-value Generalized Randomized Response mechanism (PCKV-GRR), with an average enhancement of 31.6% in data utility. Our work offers a novel, practical solution for key-value IoT data collection mechanisms.

Simulation is a vital method to test autonomous vehicle software. It provides a low-cost and convenient virtual environment to verify the control program of autonomous vehicles. However, to deploy simulation, the vehicle software must either have built-in adaptability or undergo the source code level modification to adapt to simulators. Unfortunately, most autonomous vehicles on the market are not capable of adapting to simulators, let alone their source code are unavailable to the public. Therefore, we propose an innovative off-chip peripheral emulation-based methodology to assist drone software in adapting to simulators. The experiments demonstrate that our methodology can help simulate drones without modifying their software. This makes simulation more feasible for thirdparty analysts. Furthermore, we deployed a prototype called DVATAR and simulated a real-world drone using DVATAR for demonstration.

Autopilot is an important component of robotic vehicles. It comprises the hardware and software utilized to automatically control and drive robotic vehicles. Due to its extensive usage and significant role in robotic vehicles, its security has garnered considerable attention. Currently, there are many open-source autopilot hardware options produced by vendors and distributed through intermediaries. These open-source autopilot hardware solutions are utilized by companies and individuals to build their own robotic vehicles. However, in the supply chain, there is potential risk that the autopilot hardware might be modified. Due to the unwavering trust in open source solutions and limited familiarity with autopilot hardware, developers of robotic vehicles might inadvertently procure hardware that has been deliberately modified by attackers. In this research, we proposed a stealthy, low-cost, hardware based supply chain attack that targets autopilot hardware. This attack is independent of the autopilot software, it is malicious to any normal autopilot software that is executed by the infected hardware.

Recently, Unmanned Aerial Vehicles (UAVs) deployed with deep learning models have been widely applied both in civil and military. However, the vulnerability of the deployed model to adversarial attacks has raised security concerns. Previous studies have mainly explored adversarial attacks in the digital domain. While physical attacks have posed a more serious threat to UAVs. In this paper, we have explored a novel Robust Physical Attack against UAVs named, which directly threatens the flight safety of UAVs. Specifically, three attacks are proposed in : Hiding Attack (HA), Yaw Attack (YA), and Obstacle Attack (OA). To launch the attacks, we overcome three domain-design challenges, including continuous perturbation, digital-physical domain gap, and optimum perturbation generation. For continuous perturbation, we have introduced a nested patch that realizes attacks at any distance. Further, a series of transformations are considered to narrow the gap between the digital and physical domains. Then, we proposed a time-dependent mechanism for generating optimum perturbation. We conducted comprehensive experiments in the digital domain, simulation environment, and physical domain. The experimental results validate the robustness of the proposed framework. In the digital domain, outperforms the baseline by $54.9\%$ average attack success rate (ASR). More importantly, is still effective in both the simulation environment and the physical domain, achieving an average ASR of $100\%$ .

Inter-app communication mechanism allows app developers to improve Android apps’ usability and provide users with rich functions via interacting with exposed components or performing data sharing. However, this mechanism may be leveraged by malicious developers or unintentionally misused by inexperienced developers. For end users, this type of attack may cause privacy breaches or remote controls of personal phone, which have a great threat of the user’s data security. The key to analyzing whether there is a collusion attack is to analyze the existence of communication channels between apps. The existing static analysis tools have done some work in this area, but the comprehensiveness of the analysis is insufficient, and such research lacks a unified test standard. In this paper, we present collusion attacks using more concealed inter-app communication channels, which can bypass existing security detection mechanisms. To defend against the new attacks, we design IAFDroid, an analysis framework that combines static and taint analysis. By examining 20K real-world apps, IAFDroid found that 94.4% of the most exposed components of Android may be leveraged to perform collusion attacks. Furthermore, the evaluation showed that the feature set extracted by IAFDroid could be used to promote the accuracy of Android malware detection. We contribute a more comprehensive benchmark for IAC analysis, IACBench, which includes the new attacks we propose. To facilitate follow-up studies, we open-sourced IAFDroid and IACBench based on the GPL agreement.

With the rapid development of the IoT, smart home plays an increasingly important role in daily life. At the same time, privacy issues have gradually aroused the concern of researchers. Past research proves that there are privacy risks when people use smart home devices and some protection methods are proposed to protect people's privacy. However, attackers can still infer people's behaviour and activities through analysing the wireless traffic of smart home devices. Herein, an attack method under complicated WiFi scenarios is demonstrated, including different locations, different buildings and different networks. In addition, a new protection method, Air‐Padding, is developed, which can change the traffic patterns of smart home devices through injecting constructed packets into the link between devices and routers. The results prove that Air‐Padding can prevent people's behaviour and activities from being analysed and inferred.

The authors analyse the security threats caused by personal wireless local area network (WLAN) sharing, propose schemes under two different conditions, and evaluate the performance of their schemes. WLAN is a widely used low‐cost wireless networking technology. Most personal WLANs use the Wi‐Fi‐protected access II (WPA2)‐personal to ensure robust security. Exposing the passphrase of WLAN is the only way to share it. Passphrase exposure can cause three threats, i.e. eavesdropping, evil twin attack, and resource abuse. This study addresses these threats by proposing two schemes under different device upgrade difficulties. For devices that are difficult to upgrade, their scheme only upgrades wireless routers. All WPA2‐personal certified user devices can address these threats without any changes. For easy‐upgrade and new devices, their scheme uses the attribute‐based key exchange to address threats and provide ease of use, anonymity, and fine‐grained access control. To solve the problem practically, they propose a mutual authentication method based on trust‐on‐first‐use and a convenient attribute assignment method based on the existence of social information. The attribute authority already has numerous social information to provide services and cannot obtain more private information from participants in their scheme. The analysis shows that these proposed schemes are secure and practical.