Recent studies show that deep learning-based traffic sign recognition systems are vulnerable to backdoor attacks. These compromised models can be activated to misclassify traffic signs when exposed to specific backdoor patterns during inference. Nevertheless, existing attack methods rely on exogenous triggers (e.g., stickers or patches) that introduce external features to associate backdoor patterns with target labels, significantly increasing attack complexity. In this paper, we propose FlipBAT, a stealthy endogenous backdoor attack method that uses the image’s self-flipping as the built-in trigger, eliminating the need for external trigger patterns. Our attack supports two distinct attack modes: a multi-class backdoor attack that enables flexible target diversification via cyclic mappings, and a single-class backdoor attack that achieves higher stealthiness by minimally perturbing the source class. Extensive experiments conducted on two standard traffic sign recognition datasets (GTSRB and BelgiumTS) across three different victim models demonstrate that FlipBAT effectively establishes robust mappings between backdoor images and target classes. Notably, our method achieves efficient backdoor attacks with significantly lower poisoning rates compared to conventional approaches. Our method has also been shown to be robust against state-of-the-art backdoor defenses.

The rapid growth of the low-altitude economy has driven the widespread adoption of unmanned aerial vehicles (UAVs). This growing deployment presents new challenges for UAV trajectory planning in complex urban environments. However, existing studies often overlook key factors, such as urban airspace constraints and economic efficiency, which are essential in low-altitude economy contexts. Deep reinforcement learning (DRL) is regarded as a promising solution to these issues, while its practical adoption remains limited by low learning efficiency. To overcome this limitation, we propose a novel UAV trajectory planning algorithm that integrates DRL with the large language model (LLM) reasoning to enable safe, compliant, and economically viable trajectory planning. Specifically, we model the trajectory planning task as a partially observable Markov decision process, explicitly incorporating obstacle avoidance, regulation awareness, and energy constraints. We design a hybrid optimization algorithm based on the soft actor-critic algorithm and LLM reasoning to enable adaptive decision-making in uncertain and dynamic environments. Experimental results demonstrate that our algorithm achieves the best overall performance, with the highest data collection rate (99.50%), almost zero collision avoidance rate and regulation violation rate, a successful landing rate of nearly 100%, and the lowest energy consumption rate (76.95%). These results validate the effectiveness of our algorithm in addressing UAV trajectory planning key challenges under constraints of the low-altitude economy networking.

Availability and/or performance of PBFT (Practical Byzantine Fault Tolerance) consensus service has been widely studied. However, the existing studies overlook the situation of software aging of honest nodes, which can degrade system performance over time. Rejuvenation techniques can mitigate the negative impact of aging. This paper aims to make a quantitative joint analysis of availability and performance (a.k.a performability) of PBFT consensus service in the scenario where honest nodes are susceptible to software aging and rejuvenation techniques are adopted for recovery. We propose a Semi-Markov process (SMP) based approach for model-based evaluation. Unlike traditional models that rely on exponential distributions, our approach allows the time intervals of all events to follow general distributions, thereby enable a more nuanced analysis of PBFT dynamics. We detail the modeling process and the derivation of metric formulas. We also carry out numerical analysis for the evaluation to assess the performability of PBFT consensus service.

The rapid development of Integrated Sensing and Communication (ISAC) has driven the need for robust and adaptable indoor positioning systems. Channel State Information (CSI)-based fingerprint localization has emerged as a promising solution, but existing methods face significant challenges as their sensitivity to noise often leads to poor localization accuracy and deep neural networks suffer from computational inefficiency. In this paper, we propose KFBK, a novel framework that integrates wavelet decomposition, Kolmogorov-Arnold Network (KAN)-based feature learning, and a dynamic fusion mechanism of the Broad Learning System (BLS) and KAN. It first applies a hybrid wavelet denoising strategy with decomposition and adaptive thresholding to suppress noise while preserving critical CSI patterns. Then, a KAN-based feature extractor with splineoptimized activation functions captures complex nonlinear spatiotemporal dependencies in high-dimensional CSI data, enabling dimensionality reduction without losing essential information. Finally, a temperature-controlled dynamic weighting mechanism adaptively adjusts the contributions of BLS and KAN to improve overall localization performance under varying conditions. Extensive evaluations in two real-world environments demonstrate that KFBK outperforms state-of-the-art methods in localization accuracy and environmental adaptability while maintaining realtime responsiveness and computational efficiency.

Connected Autonomous Vehicles (CAVs) represent a transformative advancement in transportation, offering enhanced safety, improved traffic efficiency, and reduced environmental impact through intelligent driving. As CAVs operate without human intervention, they heavily rely on secure vehicle-to-vehicle (V2V) communication for cooperative perception and coordinated decision-making. These real-time inter-vehicle exchanges underpin safe coordination, dynamic decision-making, and collision avoidance. To ensure trust in such communication, robust and efficient authentication mechanisms are essential. However, existing schemes often fall short in terms of security resilience and operational practicality. In this paper, we propose a novel Certificateless Signature Scheme with Conditional Privacy-Preserving Authentication (CLSS-CPPA) tailored to CAV environments. The proposed scheme addresses three fundamental limitations in existing schemes: signature forgery vulnerabilities, single-authority dependency, and lack of dynamic revocation capability. Our approach employs distributed key generation to prevent signature forgery attacks, utilizes prefix tree structures for efficient dynamic key revocation, and implements dual-agency pseudonym management with mutual authority constraints to prevent single-entity power abuse. Lightweight cryptographic operations are also adopted to suit resource-constrained vehicular systems. Formal security analysis and extensive evaluations demonstrate that CLSS-CPPA enhances privacy preserving and reduces signing and verifying costs by 20%–90% compared to state-of-the-art schemes, making it a promising solution for real-world CAV deployments.

Existing research has demonstrated that autonomous driving policies based on deep reinforcement learning (DRL) are vulnerable to adversarial attacks, which poses challenges for the practical deployment of these policies. Designing effective adversarial attacks is a crucial prerequisite for building robust driving policies. In view of this, we propose a novel adversarial attack method, which can attack the DRL-based autonomous driving agents in a stealthy and efficient manner. This method models the attack as a mixed-integer optimization problem that aims to maximize the safety violations (e.g., collisions) of the agents while minimizing the number of attack steps. Then, a DRL-based adversary is devised in this method to solve the problem to automatically learn the optimal attack policy without domain knowledge. To further enhance the adversarys learning capability, this method incorporates attack-related information into its observations to provide more decisionmaking context and employ a trajectory clipping technique to enhance sample quality. Extensive evaluation results reveal that our method achieves a remarkable 105% enhancement in attack efficiency compared to existing methods.

Federated Learning (FL), a privacy-oriented distributed ML paradigm, is gaining great interest in the Internet of Things because of its capability to protect participants’ data privacy. Studies have been conducted to address the challenges of communication efficiency and privacy-preserving, which exist in standard FL. However, they cannot achieve the goal of making a tradeoff between communication efficiency and model accuracy while guaranteeing privacy. This paper proposes a Conditional Random Sampling (CRS) method and implements it into the standard FL (CRS-FL) to tackle the above-mentioned challenges. CRS explores a Poisson-sampling-based stochastic coefficient to achieve a higher probability of obtaining zero-gradient unbiasedly and then decreases the communication overhead effectively without model accuracy degradation. Moreover, we dig out the relaxation Local Differential Privacy (LDP) guarantee conditions of CRS theoretically. Extensive experiment results indicate that (1) in communication efficiency, CRS-FL performs better than the existing methods in metric accuracy per transmission byte without model accuracy reduction in more than 7% sampling ratio (# sampling size / # model size); (2) in privacy-preserving, CRS-FL achieves no accuracy reduction compared with LDP baselines while holding the efficiency, even exceeding them in model accuracy under more sampling ratio conditions.

Structured data is widely used in big data storage and analytics but only a few Structured Data Symmetric Searchable Encryption (SD-SSE) schemes were designed. Moreover, they at least have two security issues: lack of both forward security and keyword privacy. In addition, the existing various SSE schemes designed for unstructured data cannot be applied to structured data. The paper proposes a Secure and Efficient SSE Scheme (SES2) for structured data. SES2 can not only address the above two security issues but also is more efficient than the existing Structured Data SSE (SD-SSE) schemes. Forward security is achieved by using a new key to generate the related index when the data is updated. Keyword privacy is assured by adding noise to the query trapdoor. Efficiency is improved by generating indexes with Bloom filter in a more efficient way.

With the popularization of emerging technologies such as artificial intelligence, 5G and beyond, extended reality and blockchain, the next generation Internet is rapid expansion. “Metaverse” as an evolving paradigm of next-generation Internet, can be recognized as a fully immersive, hyper spatiotemporal and self-sustaining virtual shared space, and its concept is continuous development and evolution. It is moving from imagination to the coming reality, but it is still far from being realized. One of reasons is that distinct sub-metaverses deploying their services on heterogeneous blockchains results in major problems for interoperability, preventing the implementation of seamless integrated metaverse. Facing the challenge, this paper proposes a decentralized identifiers (DIDs) assisted secure cross-metaverse authentication scheme for MEC-enabled metaverse, which is based on a novel designed infrastructure build on MEC and blockchain. In addition, the proposed scheme adopts DIDs, which can not only achieve the secure cross-metaverse authentication, but also increase the decentralization of the metaverse. In addition, the adoption of ID-based aggregate signature can reduce the overhead of computation, communication and storage.

Vehicle-to-everything (V2X) communication is the key enabler for emerging intelligent transportation systems. Applications built on top of V2X require both authentication and privacy protection for the vehicles. The common approach to meet both requirements is to use pseudonyms which are short-term identities. However, both industrial standards and state-of-the-art research are not designed for resource-constrained environments. In addition, they make a strong assumption about the security of the vehicle's on-board computation units. In this paper, we propose a lightweight auto-refreshing pseudonym protocol (LARP) for V2X. LARP supports efficient operations for resource-constrained devices, and provides security even when parts of the vehicle are compromised. We provide formal security proof showing that the protocol is secure. We conduct experiments on a Raspberry Pi 4. The results demonstrate that LARP is feasible and practical.

Electronic health (e-health) information system relies on cloud computing technologies to provide massive medical data computing and storage services. Especially, the recently proposed Machine Learning as a Service (MLaaS) on these medical data can not only effectively improve the healthcare service quality, but also support the end users with limited computing resources. However, MLaaS on the massive medical data faces the challenge of privacy. Homomorphic encryption technology has been explored to assure the privacy of medical data owners in MLaaS but with the weaknesses of limited homomorphic operations and low efficiency. To alleviate these weaknesses, this paper proposes a novel privacy-preserving non-collusion dualcloud (NCDC) model-based e-health information system using neural network (NN) computing. The system can not only assure medical data privacy through adopting homomorphic encryption technology but also assure NN model privacy by adding fake neurons to the NN. In addition, the proposed e-health information system also has the following advantages: (i) Simple key generation. (ii) No constraint on the size of medical data to be encrypted. (iii) The less loss of prediction accuracy between encrypted and original medical data. (iv) Supporting more homomorphic operations and having better computing efficiency through experiment verification.

Vehicular fog computing (VFC) complements vehicular cloud computing as a promising solution for accommodating the surge of mobile traffic and reducing latency. This paper considers vehicular fog service (VFS) provided by a vehicular fog (VF), which is formed on-the-fly by integrating computing and storage resources of parked vehicles. VF dynamicity, due to vehicles' random arrivals and departures, poses a number of challenges for reliable and secure VFS provision to client vehicles. We propose a novel mechanism which consists of a VF construction method and a VFS access method to ensure VFS reliability and security without sacrificing performance. The reliability and security of VFS under our mechanism are discussed in detail. Moreover, we investigate the impact of the proposed mechanism on VF throughput and show that the mechanism is lightweight enough to be used in the latency-sensitive VFC.

As modern vehicles and distributed fog services advance apace, vehicular fog services (VFSs) are being expected to span across multiple geo-distributed datacenters, which inevitably leads to cross-datacenter authentication. Traditional cross-datacenter authentication models are not suitable for the scenario of high-speed moving vehicles accessing VFS, because these models either ignored user privacy or ignored the delay requirement of driving vehicles. This paper proposes a blockchain-assisted lightweight anonymous authentication (BLA) mechanism for distributed VFS, which is provisioned to driving vehicles. BLA can achieve the following advantages: 1) realizing a flexible cross-datacenter authentication, in which a vehicle can decide whether to be reauthenticated or not when it enters a new vehicular fog datacenter; 2) achieving anonymity, and granting vehicle users the responsibility of preserving their privacy; 3) it is lightweight by achieving noninteractivity between vehicles and service managers (SMs), and eliminating the communication between SMs in the authentication process, which significantly reduces the communication delay; and 4) resisting the attack that the database governed by one center is tampered with. BLA achieves these advantages by effectively combining modern cryptographical technology and blockchain technology. These security features are demonstrated by carrying out security analysis. Meanwhile, extensive simulations are conducted to validate the efficiency and practicality of BLA.

A current soft-commutation Space Vector Modulation (SVM) scheme for three-phase current source converter (CSC) is proposed in this paper. With the presented SVM, the incoming switch has a positive collector-to-emitter voltage while the outgoing switch has a negative collector-to-emitter voltage during current commutations. That is, natural soft-commutations for switching transitions are achieved with accelerated current-commutating processes and lessened durations. As a result, distortion of the line currents can be reduced. Besides, due to the lessened overlapping time, the switching frequency is expected to reach as high as possible to reduce the ac filter and improve the line current quality. Especially, since the presented SVM is based on a closed-loop strategy, soft commutations can be reached in CSC with any power factor operation. The experiment verifies the proposition.