With the rapid advancement of Unmanned Aerial Vehicle (UAV) technology, fleets are increasingly deployed in disaster relief, logistics, and environmental monitoring, necessitating real-time task execution and efficient UAV resource allocation. However, UAVs' constrained computational and communication capabilities, particularly in dynamic and resource-limited environments, pose significant challenges to effective resource management. To address these issues, we propose a dual-layer auction mechanism within a UAV-enhanced edge-computing architecture to achieve fair and efficient resource allocation. The mechanism integrates an improved double-Dutch auction for dynamic initial resource matching and a Vickrey auction for reallocating unmatched resources, ensuring optimal utilization and fairness. Furthermore, we incorporate the Proximal Policy Optimization reinforcement learning algorithm to adaptively refine resource pricing and auction strategies, and the Gale-Shapley algorithm to optimize UAV-edge-server matching, enhancing allocation flexibility and fairness. Extensive experiments validate the effectiveness of our approach, demonstrating up to 28.7% improvement in social welfare, enhanced resource utilization, and reduced communication costs, particularly in large-scale markets. These findings highlight the potential of the proposed mechanism in optimizing UAV-edge resource management and advancing real-time UAV applications.

Sharing Electronic Health Record (EHR) data is critical for improving medical decision-making and emergency care. However, EHRs contain sensitive patient information, making data breaches and misuse a significant concern during sharing. To tackle this challenge, we propose the Blockchain-Based Secure Access and Sharing (BCAS) scheme, a patient-centered, secure, and efficient solution for large-scale EHR access and sharing that intelligently combines blockchain with proxy re-encryption. Four smart contracts are designed for the BCAS to achieve effective user identity registration, precise virtual identity verification, strict access control, and secure data management. The adoption of hybrid on-chain and off-chain storage structures, coupled with the four smart contracts, effectively optimizes the efficiency of the blockchain in BCAS. Proxy re-encryption further enhances data confidentiality and streamlines EHR sharing among authorized users. The security analysis indicates that BCAS can resist DDoS, spoofing, and man-in-the-middle attacks, ensuring the privacy and security of patient EHR data. The performance evaluation demonstrates that the CPU usage of smart contracts does not exceed 15%. The on-chain and off-chain storage structures require less than 30s for uploading and downloading an 800MB file. Additionally, the proxy re-encryption scheme can process a 10MB file in less than 43ms. These results indicate that BCAS offers efficient performance without compromising data security and privacy. Compared with existing blockchain-based EHR sharing systems, BCAS offers superior performance, a more comprehensive solution, and increased resistance to attacks.

The advancement of face recognition technology has significantly enhanced human productivity and daily life. However, it has simultaneously triggered escalating global concerns regarding personal privacy security, primarily due to the pervasive collection and potential misuse of facial data across diverse applications. While existing methods can protect target soft attributes (e.g., gender, age, race, etc.) while preserving identity utility, they suffer from two limitations: (1) inadequate precision in attribute manipulation, which hinders fine-grained control over specific attributes without compromising the stability of non-target regions; and (2) deficiencies in generated protected images, often characterized by detail loss and structural drift. To address these challenges, we propose PFNet (PrivacyFaceNet), a multi-attribute (e.g., gender, age, and smiling) face soft biometric privacy enhancement framework. Specifically, PFNet achieves precise attribute manipulation and high-fidelity generation through a synergistic paradigm integrating two key modules: the Adaptive Spatial Attention Module (ASAM) and the Style Guidance Module (SGM). ASAM operates at the feature level to accurately localize and anchor target attribute regions. Meanwhile, SGM executes directional attribute disentanglement and regulation within the latent space. Their collaborative mechanism effectively mitigates interference with non-target attributes during manipulation. Additionally, PFNet incorporates the Dual-Frequency Compensation Module (DFCM), which simultaneously reinforces low-frequency structural integrity and high-frequency textural details. This dual-path enhancement alleviates issues such as blurring and detail degradation, yielding post-augmentation face images with superior clarity and naturalness.

Dynamic vehicular networks present significant challenges for Byzantine consensus because node reliability, network connectivity, and attack intensity can vary significantly over time. Consequently, conventional Practical Byzantine Fault Tolerance (PBFT)-style protocols are difficult to apply directly in such settings because they rely on equal-weight voting, static fault thresholds, and limited responsiveness to intermittent adversarial behaviors such as On-Off attacks. To address these limitations, this paper presents a trust-adaptive consensus framework tailored for dynamic vehicular networks, in which continuous trust assessment guides consensus influence, leader selection, and feedback-based correction within a closed-loop design. By translating trust assessment into a control signal within the consensus process, the framework dynamically adjusts voting weights and effectively mitigates the disruptive impact of unreliable nodes. Representative NS-3 simulations demonstrate encouraging performance trends over conventional baselines in the evaluated settings, suggesting that trust-adaptive control is a promising direction for resilient coordination in dynamic vehicular networks.

Data aggregation is evolving into an extremely crucial role for facilitating decision-making in Internet of Vehicles (IoV). Multi-region data is a typical attribute of IoV, containing sensitive information and driving trajectories. However, the existing privacy-preserving schemes face problems such as regional statistics, messages integrity, and collusion attacks. In order to tackle this challenge, we propose a privacy-enhanced multi-region data aggregation (PRDA) scheme for IoV. Specifically, PRDA protects both sensing data and location as masked values by multi-secret sharing. We design regional vectors to generate mask keys using symmetric bivariate polynomial without interaction. In addition, vehicles spontaneously generate verifiable and aggregatable signature to ensure messages integrity in insecure communication networks. Batch verification of bilinear pairing can improve efficiency while resisting tampering attacks by malicious adversaries. Experiments demonstrate that as the number of regions increases, comparing with existing works, PRDA has lower communication overhead, and decreases computational cost by over 32.6%.

With the rapid growth of video streaming services, content providers (CPs) face significant challenges in efficiently delivering multimedia content via heterogeneous content delivery network(CDN) nodes. Resource allocation and pricing strategies that overlook node heterogeneity and information asymmetry often result in inefficient utilization and degraded service quality. To address these issues, we propose a contract-theoretic incentive mechanism that enables the CP to design differentiated resource–reward bundles. These bundles satisfy individual rationality (IR) and incentive compatibility (IC) constraints, allowing CDN nodes to voluntarily select optimal contracts based on their private information. We derive the conditions for feasible contract design and develop an optimization framework to maximize social welfare. Extensive simulations demonstrate that the proposed mechanism significantly outperforms baseline strategies such as linear pricing and random contract assignment, yielding notable gains in social welfare, CP utility, and CDN node participation. This work provides new insights into efficient and incentive-compatible content caching for heterogeneous video streaming delivery networks.

The proliferation of the Internet of Vehicles (IoV) has spurred the development of computationally intensive and latency-sensitive applications in vehicles. However, limited onboard computing resources hinder their execution. Vehicle edge computing addresses this by providing readily accessible resources from edge servers. However, efficient resource allocation remains a challenge due to limited edge server capacity and dynamic traffic patterns. This paper proposes a novel three-layer collaborative intelligence framework that leverages edge and cloud resources to minimize system latency in IoV. We employ a deep reinforcement learning algorithm to adapt to dynamic traffic conditions and make optimal resource allocation decisions. Extensive experiments demonstrate a significant reduction in latency (26.75

The explosion of cameras embedded in IoT devices—from mobile phones to autonomous vehicles—has positioned video analytics as a transformative AI tool across healthcare, smart cities, and beyond. Yet, the substantial computing and bandwidth demands of these applications outstrip what IoT devices alone can handle, particularly when low latency is required. Computing Power Networking (CPN) is an emerging solution that unifies cloud, edge, and device resources, enabling seamless, efficient task distribution for real-time analytics. While recent advances in cloud-edge frameworks show promise, current approaches often neglect the economic incentives that drive resource availability. To address this, we present a novel, privacy-enabled dual-layer Stackelberg game model that establishes a dynamic pricing strategy for video analytics in CPN. Our model introduces a two-stage negotiation: IoT devices contract with edge servers for computational and bandwidth resources, while edge servers may offload tasks to the cloud for enhanced service. Using game theory, we derive optimal pricing and offloading strategies under both complete and incomplete information, proving a Nash equilibrium. Comprehensive simulations validate our approach, showing improvements in resource efficiency, reduced latency, and incentivized resource-sharing across all CPN tiers. Specifically, our hybrid offloading strategy significantly reduces latency compared to edge-only and cloud-only computation models. For varying IoT device quantities, the average latency reduction across all scenarios is approximately 30.5%. This work provides an economically sustainable, privacy-conscious solution to the computational challenges of video analytics in an interconnected, resource-sharing ecosystem.

With the rapid development of 6G network technology and intelligent transportation system (ITS), the edge deployment and lightweight of network applications for modern users have gradually become a possibility. In this work, we propose a mobile intelligent vehicular task offloading method efficient mobile edge computing assisted task offloading using generative adversarial network (MEGAN) based on task representation learning and federated optimization for lightweight task recognition and energy consumption optimization of mobile edge computing (MEC) in 5G/6G transportation networks. The extended dataset of tasks is constructed based on generative adversarial network (GAN) to overcome the problems of data model overfitting and sample imbalance. A task preprocessing model between the edge server and the mobile users is established by using the federated deep learning with the knowledge distillation. The task classification accuracy, energy consumption of signal transmission and data computing are the optimization objectives to realize the MEC and lightweight application deployment. Experimental results show that compared with other state-of-the-art vehicular task offloading methods, the MEGAN method has great potential to promote the efficiency and energy consumption optimization of new transportation service processing in the future.

Vehicular edge computing (VEC) tackles the escalating computational demands of intelligent transportation systems by offloading tasks to nearby roadside units (RSUs) for processing. However, in the dynamic vehicular network environment, where vehicles are constantly moving, effective VEC demands a sophisticated approach to managing computing, caching, and communication resources. This involves coordinating resource allocation and data caching across multiple vehicles and RSUs while making complex decisions about task placement. In this paper, we present VECO, a Vehicular Edge Caching and Offloading framework powered by digital twins (DTs). VECO leverages DTs for real-time monitoring of network conditions and resource states, enabling predictive analysis and intelligent decision-making. The framework incorporates a Dynamic Task Caching and Computation Offloading (DT2C) mechanism to optimize data caching and adapt task offloading based on task characteristics and dynamic resource availability. Specifically, we develop a utility-based caching algorithm for RSUs and a novel task offloading strategy using a Proximal Policy Optimization-based deep reinforcement learning algorithm. Extensive experiments demonstrate that VECO, augmented by the DT2C mechanism, significantly outperforms baseline approaches, achieving faster learning convergence and a 21% reduction in total costs, including system latency and energy consumption.

Although data sharing and fusion between connected autonomous vehicles (CAVs) can effectively enhance environment awareness and improve driving safety, it has to face severe challenges of privacy disclosure. Outsourcing encrypted data to edge servers for data analysis and model learning can alleviate this issue without imposing additional computing load on CAVs. In this article, we propose a privacy-preserving outsourcing learning (PPOL) framework based on lightweight additive secret sharing (ASS). Firstly, we propose a privacy-preserving outsourcing object detection method over secretly shared image and point cloud data. Secondly, we construct a privacy-preserving outsourcing depth estimation model over fused stereo image shares, aiming to provide a feasible solution for outsourcing data fusion and learning. Finally, we point out open issues under the PPOL framework, and give research perspectives in aspects of privacy-preserving data fusion for multi-frame, multi-modal and multi-view data, as well as privacy-preserving model optimization.

Computing Power Networking (CPN) represents a transformative paradigm in distributed computing, harnessing the collective capabilities of edge servers dispersed across diverse geographical locations. CPN’s core strengths lie in its ability to accelerate data processing, diminish latency, and scale efficiently, rendering it particularly apt for real-time applications and the Internet of Things. When coupled with blockchain technology, CPN extends its potential by facilitating secure and transparent allocation and trading of computing resources, bolstering data integrity and reliability. However, current research at the intersection of CPN and blockchain primarily focuses on framework development and technology integration, often overlooking the challenge of delivering dependable computing services, especially in the presence of potentially unreliable nodes. To tackle this issue, we introduce a reputation-enhanced resource trading framework, designed to ensure equitable and trustworthy computing power transactions. We establish a decentralized reputation model, capable of accurately assessing node behavior over extended periods. Additionally, we present three optimization mechanisms for reputation updates, accounting for transaction history, quality of service, and transaction amount. Furthermore, our work introduces a reputation-enhanced consensus mechanism within the trading system, strategically employing incentives to motivate participants to deliver high-quality services, thereby increasing their rewards. Simultaneously, it effectively mitigates wealth inequality among resource providers of varying sizes. To validate our approach, we develop a prototype system and conduct performance evaluations, which affirm the superiority of our system in enhancing reputation and delivering robust economic features.

Outsourced cloud computing can be considered as an effective way to overcome the data island among users and relieve the pressure of limited resources. However, due to the concerns about trust in cloud servers, outsourcing the users' data and model training task has considerable privacy disclosure risks. This article presents a PriKPM scheme by using additive secret sharing (ASS), so as to implement the privacy-preserving k-prototype clustering for mixed data (i.e., including numerical and categorical attributes). In PriKPM, data samples are randomly split into two shares and delivered offline to two collaborative servers. We design a secure initialization method for determining the location and number of cluster centers. Then, both servers securely calculate the mixed distance between samples and cluster centers, and execute the samples partion and cluster updating operations. An efficient and secure comparison protocol is developed to offer flexibly the "less than or equal" and "equal" functions during the entire clustering process. Furthermore, theoretical analysis proves the effectiveness and security of PriKPM. Sufficient experiments demonstrate that PriKPM is computationally more efficient than existing secure clustering works. PriKPM can achieve the approximate accuracy of the plaintext k-prototype clustering scheme.

The Metaverse, enabled by internet connectivity and immersive technologies like Virtual Reality (VR) and Augmented Reality, represents a collective virtual shared space. However, VR technology’s resource-intensive tasks, such as rendering, display, and tracking, result in substantial energy usage. Fortunately, the emergence of edge computing, a technology that places computing and storage resources closer to end-users, promotes the concept of computation offloading. In this paper, we employ an energy-efficient 360-degree VR video transmission scheme that offloads complex VR operations to edge servers. This approach minimizes latency, conserves energy, and adopts a field-of-view-based method for 360-degree VR video generation and transmission. Furthermore, we address the resource allocation challenge for multiple devices and tasks by proposing an enhanced whale optimization algorithm designed to optimize global system performance. Simulation experiments demonstrate the effectiveness of our proposed mechanism, showcasing significant reductions in energy consumption and latency for VR devices while maintaining an enhanced user experience.

The Metaverse, as the next-generation Internet, disrupts the physical world to create an immersive virtual digital realm. Its implementation relies on key enablers, including virtual reality, AI, blockchain, and 6G. However, these technologies involve computationally intensive operations with high energy consumption, posing a challenge to the sustainability of the Metaverse. A promising approach is to leverage ubiquitous and heterogeneous computing power, encompassing end-edge-cloud devices, to optimize resource allocation for computing and networking. Unfortunately, existing edge-cloud architectures fall short in meeting the diverse resource demands of Metaverse applications. In this article, we present an energy-efficient framework that delivers Metaverse services meeting quality-of-service metrics. The framework is based on computing power networking, a network-driven distributed computing environment enabling the free flow of isolated computing power. We introduce a typical Metaverse scenario involving video analytics and formulate the problem of optimal service placement, which we solve using the deep reinforcement learning algorithm, proximal policy optimization. Experimental results demonstrate the superior energy efficiency of our proposed framework. Finally, we discuss future directions for building a sustainable Metaverse, emphasizing the need for ongoing advancements in energy efficiency.

Collaborative intelligence is a new paradigm of distributed machine learning, in which a centralized server publishes an AI task and smart edge devices join together to complete the task. Applying collaborative intelligence in a wide range of applications has become a hot trend. However, it faces two intractable challenges: participant recruitment and service guarantee. In this paper, an incentive mechanism is proposed for collaborative intelligence. First, we employ a Stackelberg game model to analyze the participants' utility and design a winner-determination algorithm to select the optimal edge devices. Then, we study the long-term interactions of participants with a repeated game model and put forward a service-guarantee mechanism by setting up a provable trigger strategy. Finally, the simulation results validate our proposed strategies.

SummaryCollaborative cloud applications have become the dominant application mode in the big data era. These applications usually generate plenty of cooperative files, which share their ownerships with all collaborative participants. Data deduplication is a promising solution to improve the storage efficiency and save the user expenditure. However, it remains an open issue on how to securely prove the shared ownerships for the shared files and address the attacks on account of using data deduplication. To tackle the above issue, in this paper, we introduce a novel concept of the Proof of Shared oWnership (PoSW) and construct a secure multi‐server‐aided PoSW (ms‐PoSW) scheme for securing client‐side deduplication for the shared files, which is based on the convergent encryption, secret sharing, and bloom filter. In the ms‐PoSW scheme, we employ a sharing convergent key to avoid the single point of failure, introduce the secret sharing algorithm to implement the shared ownership, and construct a novel interaction protocol between the shared owners and the cloud server to prove the shared ownership. Furthermore, a hybrid PoSW scheme is constructed to address the secure proof of hybrid cloud architectures. Finally, security analysis and performance evaluation show the security and efficiency of the proposed schemes.

Code offloading is a promising way to accelerate mobile applications and reduce the energy consumption of mobile devices by shifting some computation to the cloud. However, existing code offloading systems suffer from a long communication delay between mobile devices and the cloud. To address this challenge, in this paper, we consider to deploy edge nodes in close proximity to mobile devices and study how they benefit code offloading. We design an edge-centric code offloading system, called Echo, over a three-layer computing hierarchy consisting of mobile devices, the edge, and the cloud. A critical problem needs to be addressed by Echo is to decide which methods should be offloaded to which computing platform (the edge or the cloud). Different from existing offloading systems that let mobile devices individually make offloading decisions, Echo implements a centralized decision engine at the edge. This edge-centric design can fully exploit limited hardware resources at the edge to provide offloading services with the quality-of-service guarantee. Furthermore, we propose some novel mechanisms, e.g., lazy object transmission and differential object update, to further improve system performance. The results of a small-scale real deployment and trace-driven simulations show that Echo significantly outperforms existing code offloading systems at both execution time and energy consumption.

We are living in a world where massive end devices perform computing everywhere and everyday. However, these devices are constrained by the battery and computational resources. With the increasing number of intelligent applications (e.g., augmented reality and face recognition) that require much more computational power, they shift to perform computation offloading to the cloud, known as mobile cloud computing (MCC). Unfortunately, the cloud is usually far away from end devices, leading to a high latency as well as the bad quality of experience (QoE) for latency-sensitive applications. In this context, the emergence of edge computing is no coincidence. Edge computing extends the cloud to the edge of the network, close to end users, bringing ultra-low latency and high bandwidth. Consequently, there is a trend of computation offloading toward edge computing. In this paper, we provide a comprehensive perspective on this trend. First, we give an insight into the architecture refactoring in edge computing. Based on that insight, this paper reviews the state-of-the-art research on computation offloading in terms of application partitioning, task allocation, resource management, and distributed execution, with highlighting features for edge computing. Then, we illustrate some disruptive application scenarios that we envision as critical drivers for the flourish of edge computing, such as real-time video analytics, smart “things” (e.g., smart city and smart home), vehicle applications, and cloud gaming. Finally, we discuss the opportunities and future research directions.