A compact broadband four-port multiple-input multipleoutput (MIMO) antenna for UAV to assist automotive communication is proposed. By utilizing a 3D ground dielectric layer, the microstrip patch elements are vertically placed on each side of the cube to achieve omnidirectional coverage. Grounding branches are introduced on both sides of the radiating patch to realize short circuit, significantly improves impedance matching, resulting in a broadband of 4.16 GHz. 8.84 GHz and a compact size of 70 mm × 70 mm × 10 mm. Additionally, parasitic ground structures are incorporated inside the 3D ground dielectric layer, effectively enhancing port isolation. Measurements show that the port coupling coefficient is better than.18 dB in the entire frequency band, and it is even superior to.25 dB in the V2X frequency band. The envelope correlation coefficient remains below 0.0048 throughout the operating band. Furthermore, the diversity gain achieves 9.992 dB, while the channel capacity loss is maintained below 0.4 bits/s/Hz. This design provides an efficient and stable signal transmission solution for drone-assisted automotive communication systems

Federated Learning enables collaborative model training across multiple IoT devices while preserving data privacy. However, the trustworthiness of the aggregation server remains a critical security vulnerability, especially in IoT environments that rely on potentially untrusted servers. Existing secure aggregation schemes exhibit critical flaws: verification mechanisms and aggregation processes are decoupled, allowing a malicious server to generate valid verification tokens while returning incorrect aggregation results. This enables covert attacks where verification succeeds despite erroneous models, thereby seriously compromising system reliability. As a result, designing a privacy-preserving aggregation scheme that integrates verification with aggregation, while keeping both low computational and communication costs, remains a persistent challenge. To address this issue, we propose a lightweight verifiable secure aggregation based on functional encryption for federated learning on IoT devices (VFE). Our protocol cryptographically binds multi-client function encryption to identity-based aggregate signatures, thereby deeply integrating aggregation computation with verification.We further employ an efficient bilinear-pairing-based verification protocol that supports single-step verification, thereby eliminating auxiliary mechanisms and significantly reducing verification complexity. The security analysis and extensive testing demonstrate that the proposed VFE protocol achieves robust verifiable aggregation in federated learning, while simultaneously preserving client privacy and substantially reducing both computation and communication overhead, making it highly suitable for IoT deployments.

Federated Learning enables decentralized model training without exposing raw data, but remains fundamentally vulnerable to poisoning attacks from malicious clients. Existing defenses rely heavily on passive anomaly detection, honest majority assumptions, or unrealistic statistical priors, making them ineffective against adaptive and stealthy adversaries. In this paper, we propose SpecShield, a proactive defense mechanism that actively probes client models through calibrated adversarial perturbations. By leveraging the Fast Gradient Sign Method on the server side, SpecShield elicits dynamic response patterns from each client. These responses are then analyzed in the frequency domain using the Discrete Wavelet Transform. These frequency-domain features uncover distinctive response patterns between benign and malicious clients, enabling robust detection of model poisoning attacks in both non-IID environments and Byzantine majority scenarios. We further derive theoretical upper bounds on perturbation magnitudes to guarantee detection accuracy while preserving benign client performance. Through extensive experiments conducted on real-world datasets under six state-of-the-art poisoning attacks, SpecShield consistently outperforms existing defenses in both detection accuracy and model robustness. Our results demonstrate that active perturbation-induced profiling provides a new dimension for securing federated learning against sophisticated adversarial threats.

Federated learning (FL) enables cross-device collaboration by sharing local model updates without exposing raw data. However, its distributed nature introduces complex, multi-layered security threats that threaten both data privacy and model robustness. One of the most significant threats is the model poisoning attack, which exploits the server’s limited verification of client updates to inject malicious gradients, undermining aggregated model integrity and amplifying vulnerabilities in dynamic FL environments. Traditional defense mechanisms are notably vulnerable to highly adaptive, dynamic model poisoning attacks, struggling to respond effectively to attackers’ real-time adjustments in strategy. To expose these vulnerabilities and advance federated learning defense strategies, we propose a Dynamic Adaptive Model Poisoning Attack (DamPa), the first adaptive poisoning method that combines multiobjective optimization with dynamic strategy adjustments. DamPa exploits dynamic optimization to generate malicious updates that closely imitate benign patterns. It achieves significant early-stage performance degradation while maintaining both stealth and effectiveness throughout training. Our experimental evaluation on multiple real-world datasets demonstrates that DamPa outperforms existing attack methods in terms of effectiveness, particularly against robust aggregation defenses like Bulyan, DnC, FLtrust. It drastically reduces model accuracy to near-random classification levels (e.g., on the CIFAR-10 dataset, accuracy drops to 10.53%). This work reveals the limitations of existing defenses against dynamic attacks and highlights the urgent need to advance FL security. The DamPa framework offers valuable insights for designing more resilient defense mechanisms. Code is available at: https://github.com/HUFangjie/code.

In mesh-based numerical simulations, sweep is an important computation pattern. During sweep on meshes, computations on cells are strictly ordered by data dependencies in given directions. Due to this order constraint, parallelizing sweep is challenging, especially for unstructured and deforming meshes. Meanwhile, recent high-fidelity multi-physics simulations of particle transport, including nuclear reactor and inertial confinement fusion, require sweeps on large scale meshes with billions of cells and hundreds of directions. In this paper, we present JSweep, a parallel data-driven framework integrated in the JAxMIN infrastructures. The essential of JSweep is a general patch-centric data-driven abstraction, coupled with a high performance runtime system leveraging hybrid parallelism of MPI+threads and achieving dynamic communication on contemporary multi-core clusters. Built on JSweep, we implement a representative data-driven algorithm, Sn transport, featuring optimizations of vertex clustering, multi-level priority strategy and patch-angle parallelism. Experimental evaluation with two real-world applications on structured and unstructured meshes respectively, demonstrates that JSweep can scale to tens of thousands of processor cores with reasonable parallel efficiency.

Federated learning has emerged as a promising paradigm by collaboratively training a global model through sharing local gradients without exposing raw data. However, the shared gradients pose a threat to privacy leakage of local data. The central server may forge the aggregated results. Besides, it is common that resource-constrained devices drop out in federated learning. To solve these problems, the existing solutions consider either only efficiency, or privacy preservation. It is still a challenge to design a verifiable and lightweight secure aggregation with drop-out resilience for large-scale federated learning. In this article, we propose VOSA, an efficient verifiable and oblivious secure aggregation protocol for privacy-preserving federated learning. We exploit aggregator oblivious encryption to efficiently mask users’ local gradients. The central server performs aggregation on the obscured gradients without revealing the privacy of local data. Meanwhile, each user can efficiently verify the correctness of the aggregated results. Moreover, VOSA adopts a dynamic group management mechanism to tolerate users’ dropping out with no impact on their participation in future learning process. Security analysis shows that the VOSA can guarantee the security requirements of privacy-preserving federated learning. The extensive experimental evaluations conducted on real-world datasets demonstrate the practical performance of the proposed VOSA with high efficiency.

Permission delegation has become a new way for data sharing by delegating the authorized permission to other users. A flexible authorization model with strict access control policies is promising for electronic health record (EHR) sharing with security. In this paper, a blockchain-based multi-hop permission delegation scheme with controllable delegation depth for EHR sharing has been presented. We use the interplanetary file system (IPFS) for storing the original EHRs. Smart contracts and proxy re-encryption technology are implemented for permission delegation. In order to ensure data security, we use attribute-based encryption to provide fine-grained access control. Additionally, blockchain is used to achieve traceability and immutability. We deploy smart contracts so that the delegation depth can be set by delegators. Security analysis of the proposed protocol shows that our solution meets the designed goals. Finally, we evaluate the proposed algorithm and implement the scheme on the Ethereum test chain. Our scheme outperforms the competition in terms of performance, according to the results of our experiments.

With the fast boom of Internet of Medical Things (IoMT) devices and an increasing focus on personal health, personal health data are extensively collected by IoMT and stored as personal health records (PHRs). PHRs are frequently shared for accurate diagnosis, prognosis prediction, health advice consulting, etc. Since PHRs are highly private, the data-sharing process leads to wide-ranging concerns on privacy leakage and security compromise. Existing research has shown that the centralized systems, as the mainstream mode, are under the great risks. Motivated by this, we propose a consortium blockchain-based PHR management and sharing scheme, which is both security aware and privacy preserving. We adopt the interplanetary file system (IPFS) to store the PHR ciphertext of IoMT. Then, zero-knowledge proof can provide evidence for verifying keyword index authentication on blockchain. Moreover, the scheme jointly leverages modified attribute-based cryptographic primitives and tailor-made smart contracts to achieve secure search, privacy preservation, and personalized access control in IoMT scenarios. Security analysis is conducted to show that the designed protocols attain the expected design goals. This is followed by extensive evaluation results derived from real-world data sets, which demonstrate the superiority of the proposed scheme over current leading ones.

Due to its salient features, such as immutability and auditability, blockchain is becoming more integrated into the Internet of Things (IoT) for enhancing security and developing a decentralized IoT framework. However, different IoT applications require different transaction processing performance, which brings challenges to the convergence of blockchains in IoT. Moreover, the membership of a distributed IoT system may fluctuate when an IoT device joins or leaves the system. The dynamic nature of IoT systems also introduces new challenges for device management. Accordingly, we propose an application-oriented block generation (AOBG) scheme for blockchain-enabled IoT with dynamic device management and conditional traceability. Specifically, we first construct a framework for a consortium blockchain-based IoT system, including structures for application-oriented transactions and blocks, and consensus mechanism. We present different miners, respectively, for processing urgent and ordinary transactions adaptively with applications. Then, an AOBG protocol is proposed for this framework based on group signature. The group signature is used to achieve anonymity, traceability, and nonframeability. Combining time-bound keys in group signature with node accounts in blockchain, the proposed scheme can realize efficient transaction verification, dynamic device management, conditional traceability with data security, and privacy preservation. Extensive experiments demonstrate high efficiency of the proposed scheme.

With extensive promising applications of M2M (machine-to-machine) or MTC (machine type communication), while supporting multiple MTC device access networks has been considered essential for M2M communication. In a roaming environment, it has always been a great challenge to ensure safe and efficient access for MTC device groups. In this paper, in order to solve the real-time secure and efficient access problem of multiple MTCs, we proposed a group authentication protocol based on bilinear-pairing and aggregate signature. In proposed protocol, node key is generated jointly by KGC (Key Generation Center) and node simultaneously to resist camouflage attack, and the computational complexity in authentication process is significantly ameliorated as the session key is engendered by DLP (Discrete Logarithm Problem). Security analysis shows the strong security of proposed protocol, and performance evaluation proves that both transmission overhead and computational complexity decrease significantly compared with conventional schemes. In addition, it overcomes the weakness of key escrow in identity based aggregate signature protocol.

With the rapid advancement of technology, healthcare systems have been quickly transformed into a pervasive environment, where both challenges and opportunities abound. On the one hand, the proliferation of smart phones and advances in medical sensors and devices have driven the emergence of wireless body area networks for remote patient monitoring, also known as mobile-health (M-health), thereby providing a reliable and cost effective way to improving efficiency and quality of health care. On the other hand, the advances of M-health systems also generate extensive medical data, which could crowd today’s cellular networks. Device-to-device (D2D) communications have been proposed to address this challenge, but unfortunately, security threats are also emerging because of the open nature of D2D communications between medical sensors and highly privacy-sensitive nature of medical data. Even, more disconcerting is healthcare systems that have many characteristics that make them more vulnerable to privacy attacks than in other applications. In this paper, we propose a light-weight and robust security-aware D2D-assist data transmission protocol for M-health systems by using a certificateless generalized signcryption (CLGSC) technique. Specifically, we first propose a new efficient CLGSC scheme, which can adaptively work as one of the three cryptographic primitives: signcryption, signature, or encryption, but within one single algorithm. The scheme is proved to be secure, simultaneously achieving confidentiality and unforgeability. Based on the proposed CLGSC algorithm, we further design a D2D-assist data transmission protocol for M-health systems with security properties, including data confidentiality and integrity, mutual authentication, contextual privacy, anonymity, unlinkability, and forward security. Performance analysis demonstrates that the proposed protocol can achieve the design objectives and outperform existing schemes in terms of computational and communication overhead.

Due to health information sensitivity, privacy-preserving is a crucial issue in electronic health record systems. Users must provide their health information to insurance companies for their applications. This introduces potential threats to user privacy. In this paper, we propose the fairness-aware and privacy-preserving (FAPP) protocol for online health insurance systems. In the FAPP protocol, a user's health condition is encapsulated into a ciphertext with random numbers and sent to the health insurance company. The company will be unable to access the plaintext without prior user permission. However, the company will still be able to verify user integrity based on the ciphertext. In contrast to current health insurance schemes where insurance quotes are calculated by the company, the quote is calculated by the user based on the company's public policy in the proposed FAPP protocol. Additionally, the company is able to determine whether users have cheated when generating quotes. Furthermore, we propose a concept of privacy-preserving quote, which ensures that user health details cannot be derived from a generated quote. Security analysis demonstrates that the proposed FAPP protocol can achieve privacy-preservation and transparency.

Electronic health record exchanges are crucial functions of modern healthcare systems. These components are fundamental in providing quality care and enable for a larger spectrum of services. A framework which protects patient information during data exchanges is essential for healthcare systems. To achieve security and privacy-preservation for information exchange, we propose a consent-based access control (CBAC) mechanism for healthcare systems. A consent is an authorization initiated by a patient for an intended data requester via an agreement between them. After obtaining the consent from the patient, a healthcare organization can gain access to the data, which is encrypted by a healthcare provider. This is achieved by a cryptographic primitive: conditional proxy re-encryption. By doing so, patient medical data is protected against access of unauthorized parties, including public data center. Additionally, the proposed scheme achieves collusion resistance. Furthermore, mutual authentication and contextual privacy are attained. Performance evaluation demonstrates that the proposed CBAC scheme can achieve security and privacy preservation with high computational efficiency. Copyright © 2016 John Wiley & Sons, Ltd.

Abstracdt Content delivery via device‐to‐device (D2D) communications is a promising technology for offloading the heavy traffic for future mobile communication networks. As security is a critical concern for the users, we focus on improving the secrecy capacity for content dissemination in D2D communications. In this work, we explore the inherent characteristics of wireless channels to prevent eavesdropping. Firstly, we propose a power control scheme to obtain the optimal transmission powers for the D2D links without violating secrecy requirement of cellular users. Then, we formulate the problem as a stochastic optimization problem, aiming at maximizing the secrecy capacity gain of D2D communications. By solving the expected value model for the stochastic optimization problem, the optimal D2D links are selected to realize maximal ergodic secrecy capacity gain. Specifically, a weighted conflict graph is formulated according to the protocol model. Thus, the optimization problem has been transformed to the maximum weighted independent set problem, which is solved by a greedy weighted minimum degree algorithm. Simulation results demonstrate that the content dissemination scheme with power control can bring high secrecy capacity gain to the network. Copyright © 2016 John Wiley & Sons, Ltd.

Abstract Device‐to‐device (D2D) communications have been viewed as a promising data offloading solution in cellular networks because of the explosive growth of multimedia applications. Because of the nature of distributed device location, distributed caching becomes an important function of D2D communications. By taking advantage of the caching capacity of the device, in this work, we explore the device storage and file frequent reuse to realize distributed content dissemination, that is, storing contents in mobile devices (namedhelpers). Specifically, we first investigate the average and lower bound of helper amount by dividing the network into small areas where the nodes are within each other's communication radius. Then, optimal helper amount is derived based on average helper amount and network topology. Subsequently, a location‐based distributed helper selection scheme for distributed caching is proposed based on the given optimal helper amount. In particular, nodes are selected as helpers according to their locations and degrees, and contents are placed in the manner for maximizing total user utility. Extensive simulation results demonstrate the factors that affect the optimal helper amount and the total user utility. Copyright © 2015 John Wiley & Sons, Ltd.

Device-to-Device (D2D) communication has been proposed to be a promising data offloading solution in the coming big data age, with multimedia dominating the digital contents. As quality of experience (QoE) is the major determining factor in the success of new multimedia applications, we novelly propose a QoE-driven cooperative content dissemination (QeCS) scheme in the paper. Specifically, all the users predict the QoE of the potential connections characterized by mean opinion score (MOS) and send the results to the content provider (CP). Then CP formulates a weighted oriented graph based on the network topology and MOS of each potential connection. By factorizing the graph, the content dissemination fashion is established through seeking 1-factor with the maximum weight thus achieving maximum total user MOS.

This paper presents a parallel framework of grid-based numerical algorithms where data dependencies between grid zones can be modeled by a directed acyclic graph (DAG). It consists of three parts on how to partition, order and calculate the vertices of digraph. Numerical results using hundreds of processors on two parallel machines show the efficiencies and moderate scalability of this framework