Search Results (752)

Millimeter-wave (mmWave) radar point clouds offer a privacy-preserving solution for Human Activity Recognition (HAR), but their inherent sparsity and noise limit single-modal performance. While multimodal fusion mitigates this issue, existing methods often suffer from severe negative transfer during visual degradation and incur high computational costs, unsuitable for edge devices. To address these challenges, we propose Tac-Mamba, a lightweight cross-modal state space model. First, we introduce a topology-guided distillation scheme that uses a Spatial Mamba teacher to extract structural priors from visual skeletons. These priors are then explicitly distilled into a Point Transformer v3 (PTv3) radar student with a modality dropout strategy. We also developed a Trust-Aware Cross-Modal Attention (TACMA) module to prevent negative transfer. It evaluates the reliability of visual features through a SiLU-activated cross-modal bilinear interaction, smoothly degrading to a pure radar-driven fallback projection when visual inputs are corrupted. Finally, a Lightweight Temporal Mamba Block (LTMB) with a Zero-Parameter Cross-Gating (ZPCG) mechanism captures long-range kinematic dependencies with linear complexity. Experiments on the public MM-Fi dataset under strict cross-environment protocols demonstrate that Tac-Mamba achieves competitive accuracies of 95.37% (multimodal) and 87.54% (radar-only) with only 0.86M parameters and 1.89 ms inference latency. These results highlight the model’s exceptional robustness to modality missingness and its feasibility for edge deployment. Full article

Millimeter-wave radar has gradually gained popularity as a sensor mode for Human Activity Recognition (HAR) in recent years because it preserves the privacy of individuals and is resistant to environmental conditions. Nevertheless, the fast inference of high-dimensional and sparse 4D radar data is still difficult to perform on low-resource edge devices. Current models, including 3D Convolutional Neural Networks and Transformer-based models, are frequently plagued by extensive parameter overhead or quadratic computational complexity, which restricts their applicability to edge applications. The present paper attempts to resolve these issues by introducing RadarSSM as a lightweight spatiotemporal hybrid network in the context of radar-based HAR. The explicit separation of spatial feature extraction and temporal dependency modeling helps RadarSSM decrease the overall complexity of computation significantly. Specifically, a spatial encoder based on depthwise separable 3D convolutions is designed to efficiently capture fine-grained geometric and motion features from voxelized radar data. For temporal modeling, a bidirectional State Space Model is introduced to capture long-range temporal dependencies with linear time complexity $O\left(T\right)$, thereby avoiding the quadratic cost associated with self-attention mechanisms. Extensive experiments conducted on public radar HAR datasets demonstrate that RadarSSM achieves accuracy competitive with state-of-the-art methods while substantially reducing parameter count and computational cost relative to representative convolutional baselines. These results validate the effectiveness of RadarSSM and highlight its suitability for efficient radar sensing on edge hardware. Full article

The Internet of Things (IoT) plays an increasingly central role in healthcare by enabling continuous patient monitoring, remote diagnosis, and data-driven clinical decision-making through interconnected medical devices and sensing infrastructures. Despite these advances, IoT healthcare systems remain constrained by persistent challenges related to data privacy, computational efficiency, scalability, and regulatory compliance. Federated learning (FL) reduces reliance on centralised data aggregation but remains vulnerable to inference-based privacy risks, while edge-oriented approaches are limited by device heterogeneity and restricted computational and energy resources; the deployment of large language models (LLMs) further exacerbates concerns surrounding privacy exposure, communication overhead, and practical feasibility. This study introduces Trustworthy Intelligence (TI) as a guiding framework for privacy-preserving distributed intelligence in IoT healthcare, explicitly integrating predictive performance, privacy protection, and deployment-oriented system design. Within this framework, split learning (SL) is examined as a core architectural mechanism and extended to support split-aware LLM integration across heterogeneous devices, supported by a structured taxonomy spanning architectural configurations, system adaptation strategies, and evaluation considerations. The study establishes a systematic mapping between SL design choices and representative healthcare scenarios, including wearable monitoring, multi-modal data fusion, clinical text analytics, and cross-institutional collaboration, and analyses key technical challenges such as activation-level privacy leakage, early-round vulnerability, reconstruction risks, and communication–computation trade-offs. An energy- and resource-aware adaptive cut layer selection strategy is outlined to support efficient deployment across devices with varying capabilities. A proof-of-concept experimental evaluation compares the proposed SL–LLM framework with centralised learning (CL), federated learning (FL), and conventional SL in terms of training latency, communication overhead, model accuracy, and privacy exposure under realistic IoT constraints, providing system-level evidence for the applicability of the TI framework in distributed healthcare environments and outlining directions for clinically viable and regulation-aligned IoT healthcare intelligence. Full article

The Internet of Drones (IoD) has emerged as a critical extension of the Internet of Things, enabling unmanned aerial vehicles to support diverse applications, including precision agriculture, logistics, disaster monitoring, and security surveillance. Despite its rapid growth, securing IoD communications remains a significant challenge due to the open wireless environment, high drone mobility, and strict computational and energy constraints. Existing authentication mechanisms either rely on computationally expensive cryptographic operations or remain validated only at the protocol or simulation level, leaving a critical gap in practical, hardware-validated solutions suitable for resource-constrained drone platforms. This gap motivates the need for a lightweight, privacy-preserving authentication scheme that is both theoretically sound and experimentally deployable on real hardware. To address this, we propose a Physically Unclonable Functions (PUF)-assisted lightweight authentication scheme for IoD environments that binds cryptographic keys to each drone’s intrinsic hardware characteristics via PUFs. The scheme employs dynamically generated pseudo-identities to conceal permanent drone identities and prevent tracking, while authentication and key agreement are achieved using efficient symmetric cryptographic primitives, including SHA-256 for key derivation and updates, AES-256 for secure communication, and lightweight XOR operations to minimize overhead. Forward secrecy is ensured through rolling key updates, and periodic renewal of PUF challenges enhances resistance to replay and modeling attacks. To validate practicality, both software-based and hardware-based implementations were developed and evaluated. The software evaluation demonstrates a low communication overhead of 708.5 bytes and an average computation time of 18.87 ms. The hardware implementation on a Nexys A7-100T FPGA operates at 100 MHz with only 12.49% LUT utilization and low dynamic power consumption of approximately 182.5 mW. These results confirm that the proposed framework achieves an effective balance between security, privacy, and efficiency. The significance of this work lies in providing a fully hardware-validated, PUF-based authentication framework specifically tailored to the real-world constraints of IoD environments, offering a practical foundation for securing next-generation drone networks. Full article

Payment channel networks enable scalable off-chain payments, but their practical deployment remains constrained by a persistent tension among routing efficiency, liquidity visibility, transaction privacy, and settlement security. Existing multipath routing mechanisms can improve payment success under fragmented liquidity, yet they often expose sensitive balance information, leak structural features of payment routes, and enlarge the attack surface for probing, channel exhaustion, and selective forwarding. This paper presents a novel framework, Adaptive Multipath Proofs (AMPs), for privacy protection and security in payment channel networks. The core idea is to bind multipath routing decisions with lightweight zero-knowledge verifiability, allowing intermediate nodes to validate path feasibility, fragment consistency, and settlement constraints without learning exact channel balances, the complete payment amount, or the global route structure. AMP integrates three mechanisms: a hidden-liquidity feasibility proof that supports privacy-preserving route selection, an adaptive payment-splitting strategy that dynamically determines fragment allocation according to network congestion and balance uncertainty, and a proof-coupled settlement guard that enforces atomicity and timeout consistency across all payment fragments. Together, these mechanisms reduce information leakage while preserving robust payment execution under dynamic network conditions. Experimental evaluation on real Lightning Network topologies and synthetic stress scenarios demonstrates that AMP significantly lowers balance disclosure and endpoint inference risk, improves payment completion under skewed liquidity distributions, and introduces only moderate computational and communication overhead. The results indicate that adaptive proof-carrying multipath routing offers a practical and effective direction for building secure, privacy-preserving, and high-success payment channel networks. Full article

This study presents ZK-EHR, a decentralized access control framework designed to enable secure and privacy-preserving sharing of encrypted electronic health records across institutional boundaries. Unlike existing blockchain-based EHR access control systems that expose user identities on-chain or lack cryptographic privacy guarantees, ZK-EHR decouples authorization from identity disclosure by integrating zk-SNARK-based proofs with blockchain smart contracts to verify policy compliance without revealing user roles, affiliations, or credentials. The framework employs three differentiated actor roles—Patient (Data Owner), Doctor (Care Provider), and Researcher (Authorized Analyst)—with distinct policy-driven access workflows, a custom Groth16 zero-knowledge circuit for role-based constraint enforcement, and a modular architecture combining on-chain verification with off-chain encrypted storage via IPFS. Concrete design proposals for access revocation and replay attack prevention are introduced to address operational security requirements. The system was evaluated under multiple operational and adversarial scenarios. Experimental results indicate consistent on-chain verification latency (approximately 390 ms), reliable rejection of tampered submissions, and per-verification gas consumption of 216,631 gas. A comparative analysis against representative baseline systems demonstrates that ZK-EHR uniquely combines identity anonymity, on-chain cryptographic policy enforcement, and auditable encrypted record retrieval. These findings establish the feasibility of zk-SNARK-based access control for decentralized, verifiable, and privacy-aware EHR management. Full article

The integration of blockchain into parked vehicle edge computing (PVEC) has emerged as a promising approach to mitigate the inherent trust challenges in distributed and untrusted computing environments. However, during task offloading and consensus, vehicles are vulnerable to location information disclosure, leading to privacy leakage. To address this problem, we propose a location differential privacy-enabled blockchain PVEC (DBPVEC) framework to protect location information during offloading and consensus. Specifically, we design a location differential privacy mechanism based on the Laplace mechanism and theoretically prove that it satisfies ε-differential privacy. This mechanism perturbs vehicles’ locations, and a privacy-preserving offloading strategy is designed to enhance the Hotstuff consensus and protect location privacy in edge computing. Subsequently, we formulate a joint optimization problem, considering system energy consumption, latency, and privacy strength. To solve it, we design a two-layer deep reinforcement learning (DRL) algorithm, with a Deep Q-Network (DQN) as the upper layer and a Deep Deterministic Policy Gradient (DDPG) as the lower layer, to determine the optimal offloading strategy. The experimental results demonstrate that our scheme achieves significant reductions compared to the two baseline methods: the total cost decreases by 68.31% and 63.25%, energy consumption by 9.96% and 16.27%, and delay by 31.46% and 18.07%, respectively. Moreover, it effectively preserves vehicle location privacy during task offloading and consensus while maintaining favorable performance in energy consumption and latency. Full article

Federated learning (FL) enables collaborative model training across distributed devices and robotic systems while preserving data privacy, making it well-suited for swarm robotics and edge-device-powered intelligence. However, FL remains vulnerable to adversarial behaviors such as data and model poisoning, particularly in real-world deployments where detection methods must operate under strict computational and communication constraints. This paper presents a practical, real-world federated learning framework that enhances robustness to adversarial agents in a ROS2-based edge-device swarm environment. The proposed system integrates the Federated Averaging (FedAvg) algorithm with a lightweight average cosine similarity-based filtering method to detect and suppress harmful model updates during aggregation. Unlike prior work that primarily evaluates poisoning defenses in simulated environments, this framework is implemented and evaluated on physical hardware, consisting of a laptop-based aggregator and multiple Raspberry Pi worker nodes. A convolutional neural network (CNN) based on the MobileNetV3-Small architecture is trained on the MNIST dataset, with one worker executing a sign-flipping model poisoning attack. Experimental results show that FedAvg alone fails to maintain meaningful model accuracy under adversarial conditions, resulting in near-random classification performance with a final global model accuracy of 11% and a loss of 2.3. In contrast, the integration of cosine similarity filtering demonstrates effective detection of sign-flipping model poisoning in the evaluated ROS2 swarm experiment, allowing the global model to maintain model accuracy of around 90% and loss around 0.37, which is close to baseline accuracy of 93% of the FedAvg algorithm only under no attack with a very minimal increase in loss, despite the presence of an attacker. The proposed method also maintains a false positive rate (FPR) of around 0.01 and a false negative rate (FNR) of around 0.10 of the global model in the presence of an attacker, which is a minimal difference from the baseline FedAvg-only results of around 0.008 for FPR and 0.07 for FNR. Additionally, the proposed method of FedAvg + cosine similarity filtering maintains computational statistics similar to baseline FedAvg with no attacker. Baseline results show an average runtime of about 34 min, while our proposed method shows an average runtime of about 35 min. Also, the average size of the global model being shared among workers remains consistent at around 7.15 megabytes, showing little to no increase in message payload sizes between baseline results and our proposed method. These results demonstrate that computationally lightweight cosine similarity-based detection methods can be effectively deployed in real-world, resource-constrained robotic swarm environments, providing a practical path toward improving robustness in real-world federated learning deployments beyond simulation-based evaluation. Full article

Keyword Private Information Retrieval (Keyword PIR) enables private querying over keyword-based databases, which are typically sparse, as opposed to the dense arrays used in standard Index PIR. However, existing Keyword PIR schemes are limited to single-keyword queries and generally assume that keywords serve as unique identifiers, making them inadequate for practical scenarios where keywords are non-unique attributes and clients need to retrieve records matching multiple keywords simultaneously. To bridge this gap, we propose MkCwPIR, the first single-round, exact-match multi-keyword PIR protocol that supports conjunctive keyword queries while preserving strict keyword privacy against the server. Our construction employs Constant-weight codes and Newton–Girard identities to encode multi-keyword selection into a compact algebraic representation, representing a functional extension of CwPIR (Usenix Security ’22). While this functional expansion introduces additional computational overhead due to the processing of multiple keywords, we further introduce VMkCwPIR—an optimized variant leveraging BFV vectorized homomorphic encryption. Experimental results demonstrate that although the base MkCwPIR incurs higher latency due to its enhanced logical capabilities, the vectorized optimizations in VMkCwPIR effectively close this performance gap. Consequently, VMkCwPIR achieves a performance level comparable to the single-keyword CwPIR. Experimental results demonstrate that when processing a query with eight keywords, VMkCwPIR achieves a server-side execution time comparable to executing only four independent single-keyword queries in CwPIR, while maintaining constant communication overhead for up to 16 keywords. Full article

Remote photoplethysmography (rPPG) extracts physiological signals from facial videos by analyzing subtle skin color variations caused by blood flow. While this technology enables contactless health monitoring, it also raises privacy concerns because facial videos reveal both identity and sensitive biometric information. Existing privacy-preserving techniques, such as blurring or pixelation, degrade visual quality and are unsuitable for practical rPPG applications. This paper presents rePPG, a framework that inserts a desired rPPG signal into facial videos while preserving the original facial appearance. The proposed method disentangles facial appearance and physiological features, enabling replacement of the physiological signal without altering facial identity or visual quality. Skin segmentation restricts modifications to skin regions, and a cycle-consistency mechanism ensures that the injected rPPG signal can be reliably recovered from the generated video. Importantly, the extracted rPPG signals are evaluated against the injected target physiological signals rather than the subject’s original physiological state, ensuring that the evaluation measures signal rewriting accuracy. Experiments on the PURE and UBFC datasets show that rePPG successfully embeds target PPG signals, achieving 1.10 BPM MAE and 95.00% PTE6 on PURE while preserving visual quality (PSNR 24.61 dB, SSIM 0.638). Heart rate metrics are computed using a 5-second temporal window to ensure a consistent evaluation protocol. Full article

Background: Gastrointestinal (GI) cancers remain a major global health concern, where timely and accurate interpretation of endoscopic findings plays a decisive role in patient outcomes. In recent years, deep learning–based decision support systems have shown considerable potential in assisting GI diagnosis; however, their broader adoption is often limited by patient privacy regulations, uneven data availability, and the fragmented nature of clinical data across institutions. Federated learning (FL) offers a practical solution by enabling collaborative model training while keeping patient data local to each hospital. Methods: Vision Transformers (ViTs) are particularly well suited for endoscopic image analysis due to their ability to capture long-range contextual information. Nevertheless, their high computational and communication costs pose a significant challenge in federated settings, especially when data distributions vary across clients. To address this issue, we propose a privacy-preserving federated framework that combines ViTs with a Clinical-Entropy Guided Quantum Evolutionary Algorithm (CEQEA) for adaptive token pruning. The CEQEA leverages the diagnostic diversity of each client’s local dataset to guide population initialization, evolutionary updates, and mutation strength, allowing the pruning strategy to adapt naturally to different clinical profiles. Results: The proposed framework was evaluated on curated upper- and lower-GI tract subsets of the HyperKVASIR dataset under realistic non-IID federated conditions. On the final test sets, the model achieved a mean micro-averaged accuracy of 92.33% for lower-GI classification and 90.19% for upper-GI classification, while maintaining high specificity across all diagnostic classes. At the same time, the adaptive pruning strategy reduced the number of tokens processed by approximately 40% and decreased the number of required federated communication rounds by 33% compared to ViT-based federated baselines. Conclusions: Overall, these results indicate that entropy-aware, quantum-inspired evolutionary optimization can effectively balance diagnostic performance and efficiency, making transformer-based models more practical for privacy-preserving, multi-institutional gastrointestinal endoscopy. Full article

Cybersecurity operations in IoT and edge environments require fast, evidence-grounded decisions under strict resource and trust constraints. While large language models can support triage and incident analysis, their parametric knowledge may be outdated and prone to hallucination. Retrieval-augmented generation (RAG) improves grounding by conditioning responses on retrieved evidence, but also introduces new risks such as knowledge-base poisoning, indirect prompt injection, and embedding leakage. Federated learning enables collaborative adaptation without centralizing sensitive data, motivating federated RAG (FedRAG) architectures for distributed cybersecurity deployments. This study presents a deployment-oriented scoping review of FedRAG for cybersecurity. The review follows PRISMA-ScR reporting guidance and synthesizes 82 studies published between 2020 and 2026, identified through keyword search and citation snowballing over OpenAlex, arXiv, and Crossref. We develop a taxonomy that clarifies the components of federated systems, deployment locations, trust boundaries, and protected assets. We further map the combined RAG+FL attack surface, summarize practical defenses and system patterns, and distill actionable guidance for secure, privacy-preserving, and efficient FedRAG deployment in real-world IoT and edge scenarios. Our synthesis highlights recurring trade-offs among robustness, privacy, latency, communication overhead, and maintainability, and identifies open research priorities in benchmark design, governance mechanisms, and cross-silo evaluation protocols for practical deployment. Full article

Accurate liver and tumor segmentation from abdominal computed tomography (CT) scans is essential for diagnosis and treatment planning; however, centralized deep learning approaches are often constrained by privacy regulations and inter-institution data-sharing limitations. To address these challenges, we propose a skip-free feature-forward collaborative segmentation framework called Feature-Forward Residual U-Net (FF-ResUNet), in which each institution executes the encoder locally and transmits only compact bottleneck representations to a central server. High-resolution encoder features and skip connections remain strictly within institutional boundaries, reducing privacy exposure and communication overhead. The server reconstructs segmentation masks using a multi-scale dilated residual decoder with progressive upsampling and returns lightweight updates for encoder refinement. FF-ResUNet is evaluated on the Liver Tumor Segmentation (LiTS) Challenge dataset, with cross-domain testing on 3D-IRCADb and AMOS-CT to assess robustness under distribution shifts and simulated multi-institution collaboration. On LiTS, the proposed framework achieves a liver Dice score of 0.952 ± 0.015 and a tumor Dice score of 0.737 ± 0.060, with a tumor HD95 of 10.9 ± 4.1 mm. Cross-domain experiments demonstrate stable generalization to unseen datasets, while multi-client simulations show improved performance as the number of participating institutions increases before saturation. Compared with skip-based collaborative U-Net architectures, FF-ResUNet reduces communication payload by 92–98% per training iteration while maintaining competitive segmentation accuracy. These results indicate that FF-ResUNet provides an effective balance between segmentation performance, communication efficiency, and privacy preservation evaluated under simulated multi-institution collaborative settings, supporting practical multi-center clinical deployment in bandwidth- and policy-constrained healthcare environments. Full article

Fifth-generation (5G) networks are facing critical security challenges in device authentication for massive Internet of Things deployments while preserving privacy. Traditional federated learning approaches depend on the computationally expensive homomorphic encryption to protect model gradients, resulting in substantial latency and communication overhead, leading to impractical energy consumption for resource-constrained 5G devices. This paper proposes Zero-Knowledge Federated Learning (ZK-FL), eliminating homomorphic encryption by enabling devices to prove model correctness without revealing gradients. Our approach integrates zero-knowledge proofs with FL updates, where each device generates a proof ${\mathrm{Proof}}_i=\mathrm{ZK}({\mathrm{Gradient}}_i,{\mathrm{Hash}}_i)$, demonstrating computational integrity. The experimental results from 10,000 authentication attempts demonstrate ZK-FL achieves 78.4 ms average authentication latency versus 342.5 ms for homomorphic encryption-based FL (77% reduction), proof sizes of 0.128 kB versus 512 kB (99.97% reduction), and energy consumption of 284.5 mJ versus 6525 mJ (95% reduction), while maintaining 99.3% authentication success rate with formal privacy guarantees. These results demonstrate ZK-FL enables practical privacy-preserving authentication for massive-scale 5G deployment. Full article

The rapid deployment of surveillance cameras in urban, industrial, and domestic environments has intensified the need for intelligent systems capable of analyzing video streams beyond the limitations of single-camera setups. Unlike traditional single-camera approaches, multi-camera systems expand spatial coverage, reduce blind spots, and enable consistent tracking of people and objects across non-overlapping views, thereby improving robustness against occlusions and viewpoint changes. This article presents a comprehensive review of multi-camera vision systems published between 2020 and 2025, covering application domains including public security and biometrics, intelligent transportation, smart cities and IoT, healthcare monitoring, precision agriculture, industry and robotics, pan–tilt–zoom (PTZ) camera networks, and emerging areas such as retail and forensic analysis. The review synthesizes predominant technical approaches, including deep-learning-based detection, multi-target multi-camera tracking (MTMCT), re-identification (Re-ID), spatiotemporal fusion, and edge computing architectures. Persistent challenges are identified, particularly in inter-camera data association, scalability, computational efficiency, privacy preservation, and dataset availability. Emerging trends such as distributed edge AI, cooperative camera networks, and active perception are discussed to outline future research directions toward scalable, privacy-aware, and intelligent multi-camera infrastructures. Full article