Machine unlearning in learned cardinality estimation (CE) systems presents unique challenges due to the complex distributional dependencies in multi-table relational data. Specifically, data deletion, a core component of machine unlearning, faces three critical challenges in learned CE models: attribute-level sensitivity, inter-table propagation and domain disappearance leading to severe overestimation in multi-way joins. We propose Cardinality Estimation Pruning (CEP), the first unlearning framework specifically designed for multi-table learned CE systems. CEP introduces Distribution Sensitivity Pruning, which constructs semi-join deletion results and computes sensitivity scores to guide parameter pruning, and Domain Pruning, which removes support for value domains entirely eliminated by deletion. We evaluate CEP on state-of-the-art architectures NeuroCard and FACE across IMDb and TPC-H datasets. Results demonstrate CEP consistently achieves the lowest Q-error in multi-table scenarios, particularly under high deletion ratios, often outperforming full retraining. Furthermore, CEP significantly reduces convergence iterations, incurring negligible computational overhead of 0.3%-2.5% of fine-tuning time.

Abstract Vehicular networks increasingly necessitate a robust paradigm that harmonizes high-performance multi-task optimization with stringent data transmission privacy. Nevertheless, current research often struggles to achieve an ideal equilibrium between rigorous security, efficient coordination logic, and the constrained computational capacities of on-board units. To surmount these hurdles, this paper proposes a homomorphic encryption-based privacy preservation for adaptive quantum cross-task optimization (QHE-PSO). The proposed scheme synergistically integrates localized identity authentication, homomorphic fitness appraisal, and a cross-domain strategy migration mechanism. By managing encrypted particle swarms in distinct local regions where particles represent specific strategy configurations, QHE-PSO implements a secure migration protocol to facilitate seamless multi-task coordination. Extensive experimental evaluations confirm that QHE-PSO delivers a superior balance of cryptographic resilience and optimization efficacy compared to state-of-the-art benchmarks.

ABSTRACT Objectives In rapidly evolving industrial environments, resource management in Mobile Edge Computing (MEC) has gained increasing attention, aiming to ensure Quality of Service (QoS) for Artificial Intelligence of Things (AIoT) applications. While MEC reduces end‐to‐end delay, tasks offloaded to the cloud still encounter bottlenecks when processing massive AIoT‐generated data streams. To overcome this, we introduce a Collaborative Edge‐Edge (CE2) architecture that integrates heterogeneous edge servers and devices, enabling real‐time latency‐energy trade‐offs and accelerating AI‐driven decision‐making at the network edge. Managing resources in such dynamic, multi‐task, multi‐server environments remains challenging, especially under variable task‐arrival rates. Methods To tackle this, we propose LyDRM, a hybrid dynamic resource management scheme that synergistically combines model‐based optimization with model‐free deep reinforcement learning (DRL). A Lyapunov optimization module is embedded to enforce queue‐stability constraints, ensuring bounded task backlogs over time. Result To validate its effectiveness, extensive simulations show that LyDRM reduces the average weighted system cost‐defined as a combination of latency and energy metrics‐by at least 39.89%, significantly lowers both latency and energy consumption, accelerates convergence, and maintains long‐term stability in dynamic AIoT scenarios.

Federated learning (FL) is a widely studied framework for privacy-preserving collaborative training among multiple clients. However, real-world deployments reveal a persistent challenge: free-riders, i.e., participants who benefit from the system without contributing meaningful updates. If not properly addressed, free-riders can discourage honest contributors and ultimately impair both the fairness and efficiency of FL ecosystems. Existing defenses mainly rely on post-hoc per-client, per-round evaluation, leading to limited deterrence and high resource overhead, particularly in large-scale deployments. To tackle these problems, we propose FLET, a novel federated learning framework with test tasks. FLET introduces dedicated test tasks into the training process, blending them with real FL tasks while concealing their types from participants. These test tasks, generated from the reference datasets, serve as decoys that enable accurate detection of free-riding behavior. To enforce accountability, an economic penalty mechanism is employed, achieving both proactive (ex-ante) deterrence and reactive (expost) detection.We analyze the interactions between the server and participants through a free-riding suppression game model with asymmetric information (i.e.,task type), and develop a strategic information disclosure scheme (i.e.,revealing task requirements) to mislead attackers and proactively shape participant behavior. We characterize both pure-strategy and mixed-strategy perfect Bayesian Nash equilibria, and propose a lightweight strateg-ymaking algorithm that guides players toward equilibrium strategies under different conditions with modest overhead. Extensive experiments validate that FLET effectively suppresses free-riding and enhances the utility of both the server and participants. Our findings provide insights for designing cost-effective free-riding defenses in practical FL.

In Lie group convolutional neural networks (LG-CNNs), the calculation and storage of Lie group distances have quadratic space complexity. In order to improve the memory utilization efficiency of LG-CNNs, a novel Lie group convolutional neural network called LieCConv is proposed. LieCConv utilizes an innovative sampling algorithm and a linear space complexity calculation and storage approach for Lie group distances, substantially enhancing network memory efficiency. Firstly, LieCConv employs a novel sampling algorithm called array-neighborhood sampling (ANS) in the downsampling stage. ANS only requires neighborhood information to obtain an excellent sample set with a low threshold of use. The sample set generated by ANS reflects the distribution of the original set. Then, LieCConv adopts a batch calculation and storage scheme for Lie group distances, which effectively declines the space complexity of calculating and storing Lie group distances from quadratic complexity to linear complexity, reducing the memory consumption during training. Finally, the contrast between ANS and farthest point sampling was presented, demonstrating that ANS better captures the distribution characteristics of the original dataset. The memory usage of LieCConv and LieConv was compared, revealing that LieCConv reduces the memory usage for calculating and storing Lie group distances to less than 500 MB. And the performance of LieCConv was evaluated on RotMNIST, RotFashionMNIST and TT100K, validating that LieCConv is universal and effective.

Verifying the correctness of outsourcing computation is both cumbersome and expensive, and it also requires third-party verification if auditing or arbitrating is involved. Due to its complexities, users are likely to re-outsource the verification workload to trusted third-party vendors. Multiple outsourcing tasks increase the expense of communications and expose more vulnerabilities. To address this problem, we propose an identity-based strong designated verifier dual signature scheme with constrained-delegatability. The particular innovation lies in two aspects. Firstly, in cases where there are multiple parties involved, users can autonomously designate verifiers. Second, we first present the concept of constrained-delegatability, where a signature cannot be delegated to any other than the cloud service provider. In this scheme, a user can specify a trusted verifier for the dual signature co-signed by both him and the service provider on the outsourcing result. The provider cannot designate anyone else to check the signature. The proposed scheme is provably secure based on elliptic curve bilinear pairing and the hardness assumption of computational diffie-hellman and bilinear diffie-hellman problems. Moreover, our scheme simplifies the outsourcing process and reduces the total computational costs and communication time compared to previously reported ones.

As a typical application of mobile crowdsourcing, streaming media has been attracting increasing attention since recent years. However, traditional streaming media platforms, such as Netflix, Disney+, and Hulu, may suffer some problems like inflexible billing modes, lacking sustainability in the incentive mechanisms, and management censorship. These problems may lead to a decrease in user participation rate, which will directly affect the interests of streaming media platforms. To address these issues, we propose a secure, efficient and flexible streaming media platform framework based on blockchain and well-designed smart contracts. In particular, we design a new billing model based on pay-as-you-go strategy and a new incentive mechanism with probabilistic payment technique. To improve the fairness of our incentive model, we introduce a secondary fee refund protocol where a user’s second consecutive payment could be refunded, which in turn can attract more users to participate in the platform. Since blockchain has the natural properties of decentralization and transparency, the proposed framework is resistant to censorship and enables the transactions to be publicly auditable. Based on the proposed framework, we have implemented two streaming media platform schemes. Scheme I relies primarily on smart contracts to implement the framework’s functionality, while Scheme II moves the main flow of framework to off-chain channels. As the execution of smart contracts requires transaction fees, Scheme I is more expensive but can provide much more security and accountability as well. Scheme II can execute the transaction process much faster and with only a small transaction fee. Finally, we deployed these two schemes on Ropsten and conduct a series of experiments. The results show the effectiveness and efficiency of the proposed schemes.

Mobile crowdsensing (MCS) has been applied in various fields to realize data sharing, where multiple platforms and multiple Mobile Users () have appeared recently. However, aiming at mutual selection, the existing works ignore making ’ utilities with the limited resources and platforms’ utilities while achieving the desired sensing data quality maximum as far as possible. Thus, they cannot motivate both and platforms to participate. To address this problem, standing on both sides of and platforms with conflicting interests, we propose a Progressive two-stage Auction-based Multi-platform Multi-user Mutual Selection scheme (). Specifically, in, we treat mutual selection as a two-stage auction and devise the auction models for and platform using forward and reverse auction ideas, presenting and maximizing the utilities from their respective perspectives. Then, based on the proposed progressive two-stage auction structure, we adopt 0-1 knapsack and Myerson’s price theory to construct the first stage -oriented auction and the second stage platform-oriented auction, achieving devised models. Theoretical analysis shows that is economically robust. Extensive experiments on the real dataset demonstrate that respectively promotes platforms’ and ’ utilities by 76.23% and 10.74 times, compared with the existing works.

Current 3-D object detection technology for autonomous driving usually cannot efficiently utilize local sensitive points. Meanwhile, contextual feature extracted from a object is not sufficient, which easily leads to deteriorated detection accuracy of the final object estimation. For the problems, a point–voxel-based 3-D dynamic object detection algorithm is proposed. First, local points are grouped with a camera frustum. Then, the global feature extracted by the submanifold 3-D voxel CNNs is aggregated into frustum key points. Second, a module of vector pool with feature aggregation is used to aggregate multiscale features of the point cloud. Moreover, the frustum raw feature and BEV feature are used for feature extension. Subsequently, the fine multiscale feature extracted from the point cloud is used as input to a subsequent fully convolutional network for final classification and continuous estimation of oriented 3-D boxes. The proposed method was compared with other state-of-the-art algorithms on the KITTI, Waymo, and nuScenes data sets. Experimental results showed that the proposed algorithm was better in accuracy, robustness, and generalization capabilities in 3-D dynamic object detection. Experiments on a real scenario and extensive ablation studies also demonstrated that the proposed algorithm not only effectively controls computational cost but also achieved more efficient results in 3-D object detection.

Generally, power utilities can utilize smart-meter data to extract load patterns through load-profiling technologies, such as$K$-means clustering. To improve the efficiency of load profiling, both$K$-means clustering and smart-meter data can be outsourced to powerful clouds. However, clouds are not completely trustworthy: private meter data may be used for commercial interests;$K$-means clustering may also be performed with fewer iterations to save computational costs, which violates the integrity of outsourced clustering. In this article, therefore, a secure$K$-means-clustering scheme is proposed, called privacy-preserving and integrity-preserving clustering (PIPC), which aims to protect the privacy and integrity of load profiling. To this end, two techniques are designed: 1) encrypted distance measurement, in which a public comparison matrix is constructed by securely embedding a secret key matrix and 2) integrity assurance, in which a specific Stackelberg game is designed to create economic incentives. The former, as the core of$K$-means clustering, can protect the privacy of meter data. The latter ensures that clouds can obtain the maximum utility only when clouds execute$K$-means clustering in an honest manner, thereby preserving the integrity of outsourced computing. Experimental results demonstrate that PIPC reaches high clustering accuracy and computational efficiency for load profiling while retaining smart-meter data privacy and outsourced-clustering integrity.

Mobile cloud storage (MCS) provides clients with convenient cloud storage service. In this article, we propose an efficient, secure and privacy-preserving mobile cloud storage scheme, which protects the data confidentiality and privacy simultaneously, especially the access pattern. Specifically, we propose an oblivious selection and update (OSU) protocol as the underlying primitive of the proposed mobile cloud storage scheme. OSU is based on onion additively homomorphic encryption with constant encryption layers and enables the client to obliviously retrieve an encrypted data item from the cloud and update it with a fresh value by generating a small encrypted vector, which significantly reduces the client’s computation as well as the communication overheads. Compared with previous works, our presented work has valuable properties, such as fine-grained data structure (small item size), lightweight client-side computation (a few of additively homomorphic operations) and constant communication overhead, which make it more suitable for MCS scenario. Moreover, by employing the “verification chunks” method, our scheme can be verifiable to resist malicious cloud. The comparison and evaluation indicate that our scheme is more efficient than existing oblivious storage solutions with the aspects of client and cloud workloads, respectively.

The development of Industrial Internet of Things (IIoT) provides massive abundant data resources for trading and mining. However, the existing data trading schemes achieve data usage control at the cost of high latency, thereby resulting in poor service quality as the values of IIoT data degrade over time. This article proposes a monitor-based usage control model to enforce data usage policies on the user side, which eliminates frequent interactions between owners and users. Based on that, a data trading scheme with efficient usage control for IIoT (called DTSI) is devised, which utilizes blockchain smart contract and software guard extensions (SGX) to enable owners to fully control users’ identities and operations at minimal overhead. Security analysis shows that DTSI effectively prevents data abuse and ensures the fair exchange of data. Meanwhile, extensive experiments are conducted on the DTSI prototype comparing with the state-of-the-art schemes with real-world IIoT datasets, which demonstrates the efficiency of DTSI.

Current 3D target detection methods used in the field of autonomous driving generally have low real-time performance and insufficient target context feature to detect dynamic multi-target accurately. In order to solve these problems, a dynamic multi-target detection algorithm of voxel point cloud fusion based on PointRCNN is proposed, which adopts a two-stage detection structure. The first stage directly processes the point cloud to extract key point features and divides voxel space. A novel submanifold sparse convolution is used to extract voxel features. Then key point features and voxel features of the point cloud are merged to generate pre-selection boxes. In the second stage, reference points are set based on the voxel features. The features of key points around reference points are merged for the second time to achieve optimized detection boxes. Finally, for the problem of inconsistent confidence, a mandatory consistency loss function is proposed to improve the accuracy of the detection box. The proposed algorithm was compared with other algorithms in three different datasets, and further tested on a self-made dataset from an actual vehicle platform. Results showed that the proposed algorithm had higher accuracy, better robustness, stronger generalization ability for dynamic multi-target detection.

The rapid development of machine learning technology has prompted the applications of Automatic Speech Recognition(ASR). However, studies have shown that the state-of-the-art ASR technologies are still vulnerable to various attacks, which undermines the stability of ASR destructively. In general, most of the existing attack techniques for the ASR model are based on white box scenarios, where the adversary uses adversarial samples to generate a substituted model corresponding to the target model. On the contrary, there are fewer attack schemes in the black-box scenario. Moreover, no scheme considers the problem of how to construct the architecture of the substituted models. In this paper, we point out that constructing a good substituted model architecture is crucial to the effectiveness of the attack, as it helps to generate a more sophisticated set of adversarial examples. We evaluate the performance of different substituted models by comprehensive experiments, and find that ensemble substituted models can achieve the optimal attack effect. The experiment shows that our approach performs attack over 80% success rate (2% improvement compared to the latest work) meanwhile maintaining the authenticity of the original sample well.

When authenticating a group of radio-frequency identification tags, a common method is to authenticate each tag with some challenge-response exchanges. However, sequentially authenticating individual tags one by one might not be desirable, especially when considering that a reader often has to deal with multiple tags within a limited period, since it will incur long scanning time and heavy communication costs. To address these problems, we put forward a novel efficient group authentication protocol, where a group of tags can be authenticated simultaneously with only one challenge and one response. The protocol is built on a newly designed symmetric key-based algorithm and the bit-collision pattern technique, so that authentication responses transmitted by multiple tags in a group at the same time will result in a verifiable bit-collision pattern that represents the authentication response for the entire group. The proposed approach can significantly reduce the authentication time and communication cost in the sense that the verifier can authenticate the entire group within a period that is comparable to the time taken to perform a single-tag authentication and requires only one challenge. In addition, we extend our protocol to support the privacy-preserving property, which prevents the tagged items from being tracked by illegitimate parties. A thorough security analysis shows that the proposed protocol can resist common practical attacks and experimental results show that the protocol is very efficient in terms of time and communication costs. We also discuss important practical aspects that should be considered when implementing these protocols.

This paper investigates the secure transmission in multiple access wiretap channels, where multiple legitimate users transmit private information to an intended receiver in the presence of multiple eavesdroppers. In order to improve security, we propose a novel cooperative jamming scheme, in which users do not share channel state information (CSI) but the legitimate channels will not be degraded by the artificial noise. The basic idea is to make each user exploit its own CSI in two slots to design artificial noise, so that the intended receiver can eliminate all the artificial noise but the eavesdroppers cannot. In this process, the interference between users plays a key role to achieve security, because it guarantees that the artificial noise from different users helps each other. We consider the non-collusion and collusion of eavesdroppers and analyze the secrecy performance for both scenarios. We adopt the secrecy sum-rate as the main metric, and show that positive secrecy sum-rate can be achieved by using the proposed scheme. Especially, we observe that when eavesdroppers collude and their additive white Gaussian noise (AWGN) close to zero, the number of users must not be less than twice the number of eavesdroppers to ensure positive secrecy sum-rate. Finally, simulation results are provided to corroborate our theoretical findings.

Industrial automation and control systems (IACS) are tremendously employing supervisory control and data acquisition (SCADA) network. However, their integration into IACS is vulnerable to various cyber-attacks. In this article, we first present population extremal optimization (PEO)-based deep belief network detection method (PEO-DBN) to detect the cyber-attacks of SCADA-based IACS. In PEO-DBN method, PEO algorithm is employed to determine the DBN's parameters, including number of hidden units and the size of mini-batch and learning rate, as there is no clear knowledge to set these parameters. Then, to enhance the performance of single method for cyber-attacks detection, the ensemble learning scheme is introduced for aggregation of the proposed PEO-DBN method, called EnPEO-DBN. The proposed detection methods are evaluated on gas pipeline system dataset and water storage tank system dataset from SCADA network traffic by comparing with some existing methods. Through performance analysis, simulation results show the superiority of PEO-DBN and EnPEO-DBN.

Gait recognition aims to recognize persons' identities by walking styles. Gait recognition has unique advantages due to its characteristics of non-contact and long-distance compared with face and fingerprint recognition. Cross-view gait recognition is a challenge task because view variance may produce large impact on gait silhouettes. The development of deep learning has promoted cross-view gait recognition performances to a higher level. However, performances of existing deep learning-based cross-view gait recognition methods are limited by lack of gait samples under different views. In this paper, we take a Multi-view Gait Generative Adversarial Network (MvGGAN) to generate fake gait samples to extend existing gait datasets, which provides adequate gait samples for deep learning-based cross-view gait recognition methods. The proposed MvGGAN method trains a single generator for all view pairs involved in single or multiple datasets. Moreover, we perform domain alignment based on projected maximum mean discrepancy to reduce the influence of distribution divergence caused by sample generation. The experimental results on CASIA-B and OUMVLP dataset demonstrate that fake gait samples generated by the proposed MvGGAN method can improve performances of existing state-of-the-art cross-view gait recognition methods obviously on both single-dataset and cross-dataset evaluation settings.

The widespread use of edge devices in E-Health such as smartphones and wearables means richer electronic health records (EHR) are becoming available. Training deep learning models on these data can effectively improve the quality of healthcare services. Recently, federated learning (FL) has received extensive attention in E-Health because it can train a model by only sharing gradients without disclosing the original EHR of owners. In this case, however, the adversary can still violate EHR owners' privacy based on shared gradients. To mitigate privacy threat, several privacy-preserving FL protocols have been proposed by utilizing different cryptography techniques. Unfortunately, existing privacy-preserving FL schemes do not take into account irrelevant updates, which are useless for the convergence of the global model. This may reduce the predictive accuracy and worse may lead to the uselessness of the final model. In this paper, we propose PFL-IU, an efficient and privacy-preserving FL framework that is compatible with irrelevant updates. Specifically, we first design a communication-efficient secure aggregation protocol by using a non-interactive key generation algorithm. Then we present a sign method to mitigate the negative impact incurred by irrelevant updates, which will accelerate model convergence and improve predictive accuracy. Moreover, PFL-IU is robust to EHR owners' dropout during the whole training phase. Extensive experiments using the real-world dataset demonstrate that PFL-IU can achieve better performance in terms of accuracy, convergence and efficiency.

By leveraging deep learning-based technologies, industrial artificial intelligence (IAI) has been applied to solve various industrial challenging problems in Industry 4.0. However, for privacy reasons, traditional centralized training may be unsuitable for sensitive data-driven industrial scenarios, such as healthcare and autopilot. Recently, federated learning has received widespread attention, since it enables participants to collaboratively learn a shared model without revealing their local data. However, studies have shown that, by exploiting the shared parameters adversaries can still compromise industrial applications such as auto-driving navigation systems, medical data in wearable devices, and industrial robots' decision making. In this article, to solve this problem, we propose an efficient and privacy-enhanced federated learning (PEFL) scheme for IAI. Compared with existing solutions, PEFL is noninteractive, and can prevent private data from being leaked even if multiple entities collude with each other. Moreover, extensive experiments with real-world data demonstrate the superiority of PEFL in terms of accuracy and efficiency.

Using searchable encryption (SE), users' data can be outsourced to an untrusted server while ensuring privacy of both the queries and the data. Meanwhile, to efficiently support data updating, dynamic SE (DSE) has also been proposed and applied to a variety of scenarios. However, recent work shows that even with little information leakage on updated keywords, most of existing DSE schemes are also vulnerable to adaptative attacks breaking the privacy of the queries. To address this problem, several privacy-preserving DSE have been exploited to mitigate the two major privacy issues in the data update process: i.e., Forward privacy and Backward privacy. Nevertheless, it is still an open problem to support clients multi-keyword-based searching over dynamic cloud data. In reality, as a promising query requirement, it is assurance that the cost of all participants can be fundamentally reduced by implementing multi-keyword-based querying. To combat that, in this paper, we design the first multi-keyword based search proposals ensuring forward and backward privacy over dynamic cloud data. Specifically, we utilize Symmetric Hidden Vector Encryption (SHVE) as the underlying structure to build multi-keyword search protocol. Then, Bloom filter integrating with pseudo-random function will be further adopted to enhance query efficiency. The security analysis proves the high security of our model, and extensive experiments conducted on real-world data also demonstrate the practical performance of our proposed scheme.

Let N={n0, n1, ... , nk-1} be a set of positive integers and M= {m0, m1, ... , mk-1} be a multiset of positive integers. By an (N, M, w,1; λ)-multilength optical orthogonal code (MLOOC), we mean an MLOOC of autocross correlation value and intracross correlation value one and intercross correlation value λ. The code contains micodewords of weight w and length nifor 0 ≤ i ≤ k-1. The study of MLOOCs is motivated by an application in optical networks requiring multiple signaling rates and quality-of-services. In this paper, we study (N, M, w,1; λ)-MLOOCs with λ =2 (the least value among the nontrivial intercross correlations). Some new upper bounds on code size are derived under certain restrictions and a novel encoding approach is established. A number of series of new MLOOCs are then produced. These codes are of optimal sizes with respect to the new bounds.