Existing solutions on differentially private deep learning (DPDL) either require the assumption of a trusted data server (centralized DPDL) or suffer from poor utility (local DPDL); and hence their adoptions are hampered in real-world scenarios.We present CRYPTDP, a crypto-assisted differentially private deep learning approach in the two-server model. CRYPTDP employs two non-colluding servers to collaboratively and efficiently train differentially private deep learning over the secret shares of data owners' private data while protecting the confidentiality of the data from untrusted servers. CRYPTDP is the first approach with the best of both local DPDL and centralized DPDL models, which does not resort to trusted server like local DPDL and has the utility like centralized DPDL. In particular, we also make innovations for addressing the major challenges like poor performance and security that beset CRYPTDP: We introduce a new secure computation and differential privacy friendly activation function; we propose a novel garbled-circuits-free most significant bit extraction protocol, and using the protocol we propose an efficient and secure garbled-circuits-free protocol for activation function over secret shares. Exhaustive experiments show that CRYPTDP delivers significantly better performance than the state-of-the-art local DPDL, yields higher accuracy than the state-of-the-art centralized DPDL, and can achieve two orders of magnitude faster runtime than the state-of-the-art approach.

Large-scale pre-trained vision-language models (VLMs) like CLIP show exceptional performance and zero-shot generalization. However, their reliability may be severely undermined by a critical vulnerability to subtle adversarial perturbations. Our work reveals a critical cross-modal vulnerability: visual-only perturbations induce substantial, synchronous shifts in decision attribution maps across both image and text. This phenomenon signifies a fundamental disruption of the VLM's internal logic, as it alters both the model's perceptual focus and its decision rationale. To counter this vulnerability, we introduce Cross-modal Bidirectional Attribution guided Few-shot Adversarial Prompt Tuning (CBA-FAPT), a novel method that leverages the model's internal decision rationale as a regularizer for robust learning. Our framework's core mechanism is the alignment of a novel bidirectional attribution map. This map is a unique fusion of two components. It combines forward feature attention to capture the model's perceptual focus. It also incorporates backward decision gradients to act as a proxy for the model's decision rationale, quantifying how each feature influences the final outcome. We enforce consistency on this bidirectional map between clean and adversarial examples. This approach corrects the model's internal logic on two fronts and effectively restores its adversarial robustness. Comprehensive experiments on 11 datasets demonstrate that CBA-FAPT outperforms the state-of-the-art, establishing a superior trade-off between robust and natural accuracy.

Abstract Recently, image encryption is becoming increasingly important, many chaotic models have been prevalently used to design kinds of cryptographic schemes in chaos cryptography. Among those models, coupled map lattice (CML), as a classics spatiotemporal chaotic model with good performance, is popularly used for those chaos-based cryptographic schemes. However, there exist no scientific studies in the three dimensional (3D) CML model from the view of theoretical and application perspectives besides our previous research. To further improve the complicated chaotic dynamic behavior and extend the application scenarios of CML into a3D CML one, therefore it is introduced into our paper for constructing chaos-based image encryption scheme with higher security. First of all, properties of 3D CML are comprehensively analyzed to fully verify that it possesses more complicated chaotic behavior than one dimensional and two dimensional CML. Subsequently, the chaotic sequences are produced by means of intercepting 32 bit from each node of the 3D CML model, both NIST and TestU01 testing certificate those chaotic sequences have high randomness, which are pretty suitable for designing the chaos-based cryptographic scheme. Based on those above-mentioned analyses, a new color image encryption scheme is proposed via diffusion and confusion. In our scheme, diffusion is performed based on those above-mentioned chaotic sequences via the 3D CML model, while confusion is carried out according to the Baker map. Specially, before diffusion and confusion operations, red, green and blue channel of a plain color image are combined into one pixel including 24 bit (0 or 1) for improving the efficiency of our scheme. To sum up, all corresponding simulations show our scheme has excellent encryption performance, and it would be popularly applied in the real-life cryptographic domains. Our research contributes to enriching the theoretical research on chaotic cryptography and provides new chaotic image encryption schemes.

Public-key Encryption with Keyword Search (PEKS) enables users to search encrypted data stored on an untrusted server without revealing any sensitive information. However, existing PEKS schemes are typically inefficient and lack the flexibility to support complex search policies. To address this, a novel PEKS scheme based on Arithmetic Span Programs (PEKS-ASP) is proposed in this paper. This is the first scheme to enable flexible and efficient search policies by using directed acyclic graphs. This approach enhances the efficiency of complex search queries that implement AND, OR, and NOT gates, enabling a more efficient representation of complicated search policies without redundancy in keyword usage. And our proposed PEKS-ASP scheme guarantees constant-size public parameters regardless of the number of keywords. Additionally, the proposed scheme achieves adaptive security under the matrix decisional Diffie-Hellman (MDDH) assumption, employing dual system encryption techniques. Both theoretical analysis and experimental results demonstrate that PEKS-ASP significantly improves efficiency and practicality, making it well-suited for practical applications in various cloud environments.

Large language models in deep learning have numerous parameters, requiring significant storage space and computational resources. Compression techniques are highly effective in addressing these challenges. With the development of hardware like Graphics Processing Unit (GPU), Tensor Core can accelerate low-precision matrix multiplication but achieve acceleration for sparse matrices is challenging. Due to its sparsity, the utilization of Tensor Cores is relatively low. To address this, we propose the based onTensorCoreCompressedSparseRow format (TC-CSR), which facilitates data loading on GPUs and matrix operations on Tensor Cores. Based on this format, we designed block Sparse Matrix-Matrix Multiplication (SpMM) and Sampled Dense-Dense Matrix Multiplication (SDDMM) kernels, which are common operations in deep learning. Utilizing these designs, we achieved a$\mathbf {1.41\times }$speedup on Sputnik in scenarios of moderate sparsity and a$\mathbf {1.38\times }$speedup with large-scale highly sparse matrices. Benefit from our design, we achieved a$\mathbf {1.75\times }$speedup in end-to-end inference with sparse Transformers and save memory.

Backdoor attacks in federated learning (FL) face challenges such as lower attack success rates and compromised main task accuracy (MA) compared to local training. Existing methods like distributed backdoor attack (DBA) mitigate these issues by modifying malicious clients’ updates and partitioning global triggers to enhance backdoor persistence and stealth. The recent full combination backdoor attack (FCBA) further improves backdoor efficiency with a full combination strategy. However, these methods are mainly applicable in small-scale FL. In large-scale FL, small trigger patterns weaken impact, and scaling them requires controlling exponentially more clients, which poses significant challenges, while simply reverting to DBA may decrease backdoor performance. To overcome these challenges, we propose the self-adaptive distributed backdoor attack (SADBA), which achieves similar performance to FCBA with a lower percentage of malicious clients (PMC). It also adapts more flexibly through an optimized model poisoning strategy and a self-adaptive data poisoning strategy. Experiments demonstrate SADBA outperforms state-of-the-art methods, achieving higher or comparable backdoor performance and MA across various datasets with limited PMC.

Healthcare social networks (HSNs) are pivotal in spreading healthcare knowledge, providing support to both potential patients and medical professionals, and enhancing healthcare services. However, identifying unseen data in HSN poses a significant challenge due to their intrinsic heterogeneity, dynamic characteristics, and the scarcity of labeled data. Employing semantic knowledge transfer for class-agnostic zero-shot recognition stands out as a promising and innovative solution to this problem, but the visual-semantic gap and domain shift problems considerably hinder advancements in zero-shot recognition capabilities. Previous zero-shot models often impose constraints between vision and semantics in the loss part without explicitly injecting intermodality guidance into the feature refinement process. This article yields a novel zero-shot recognition framework for HSN, named the dual tensor prototype graph network, devoted to improving the performance of recognizing unseen objects in HSN leveraging semantic knowledge. We have developed an iterative and interactive updating strategy for dual tensor prototype graphs, explicitly leveraging the distribution information from one modality to guide the prototype graph updates of another modality. We constrain the update process of the dual prototype graphs by several tailored loss functions and episodic training, alleviating the inconsistency between semantic and visual manifolds. Extensive comparative experiments conducted on two medical imaging datasets and five zero-shot benchmarks affirm the stronger generalization ability of our proposed method compared with other advanced approaches, showing the potential of addressing zero-shot problems in HSN.

Secure two-party neural network (2P-NN) inference allows the server with a neural network model and the client with inputs to perform neural network inference without revealing their private data to each other. However, the state-of-the-art 2P-NN inference still suffers from large computation and communication overhead especially when used in ImageNet-scale deep neural networks. In this work, we design and build Panther, a lightweight and efficient secure 2P-NN inference system, which has great efficiency in evaluating 2P-NN inference while safeguarding the privacy of the server and the client. At the core of Panther, we have new protocols for 2P-NN inference. Firstly, we propose a customized homomorphic encryption scheme to reduce burdensome polynomial multiplications in the homomorphic encryption arithmetic circuit of linear protocols. Secondly, we present a more efficient and communication concise design for the millionaires’ protocol, which enables non-linear protocols with less communication cost. Our evaluations over three sought-after varying-scale deep neural networks show that Panther outperforms the state-of-the-art 2P-NN inference systems in terms of end-to-end runtime and communication overhead. Panther achieves state-of-the-art performance with up to 24.95× speedup for linear protocols and 6.40× speedup for non-linear protocols in WAN when compared to prior arts.

Identifying unseen faults is a crux of the digital transformation of process manufacturing. The ever-changing manufacturing process requires preset models to cope with unseen problems. However, most current works focus on recognizing objects seen during the training phase. Conventional zero-shot recognition methods perform poorly when they are applied directly to these tasks due to the different scenarios and limited generalizability. This article yields a tensor-based zero-shot fault diagnosis framework, termed MetaEvolver, which is dedicated to improving fault diagnosis accuracy and unseen domain generalizability for practical process manufacturing scenarios. MetaEvolver learns to evolve the dual prototype distributions for each uncertain meta-domain from seen faults and then adapt to unseen faults. We first propose the concept of the uncertain meta-domain and then construct corresponding sample prototypes with the guidance of class-level attributes, which produce the sample-attribute alignment at the prototype level. MetaEvolver further collaboratively evolves the uncertain meta-domain dual prototypes by injecting the prototype distribution information of another modality, boosting the sample-attribute alignment at the distribution level. Building on the uncertain meta-domain strategy, MetaEvolver is prone to achieving knowledge transferring and unseen domain generalization with the optimization of several devised loss functions. Comprehensive experimental results on five process manufacturing data groups and five zero-shot benchmarks demonstrate that our MetaEvolver has great superiority and potential to tackle zero-shot fault diagnosis for smart process manufacturing.

Recurrent neural network (RNN), a branch of deep learning, is a powerful model for sequential data that has outstanding performance on a wide range of important Internet of Things (IoT) tasks. This unprecedented growth of RNN model has however encountered both heterogeneous IoT data and privacy issues. Existing RNN model can not deal with heterogeneous sequential data; often the larger datasets used in training of RNN model contain sensitive information. To tackle these challenges and for the first time, this research proposes a novel differentially private tensor-based RNN (DPTRNN) that can be applied in many challenging deep learning sequence tasks for IoT systems. Specifically, to process heterogeneous sequential data, we propose a tensor-based RNN model. To guarantee privacy, we develop a tensor-based back-propagation through time algorithm with perturbation to avoid exposing the sensitive information for training the tensor-based RNN model within the framework of differential privacy. Thorough security analysis shows that the differential private tensor-based RNN efficiently protects the confidentiality of sensitive user information for IoT. Our results from extensive experiments on two challenging large video datasets suggest that our proposed scheme is practical with guarantee of data privacy preservation and acceptable accuracy loss.

Federated semi-supervised learning (FSSL) has been perceived as a promising approach that leverages semi-supervised learning and federated learning (FL) to provide powerful privacy preservation while reducing the burden on human supervision. However, due to the lack of strict participant identification and the significant proportion of unlabeled samples, FSSL is more susceptible to covert backdoor attacks than traditional machine learning. To validate this speculation, a novel semantic-perturbed stealthy backdoor attack (SemSBA) scheme is proposed for FSSL-based systems. In SemSBA, we select original natural semantic features in the unlabeled training samples as backdoor triggers and then generate poisoned samples by adding adversarial perturbations that move them across the model decision boundary. With SemSBA, the adversary can trigger the hidden backdoor in the victim model during the inference stage without any deliberate modifications on testing samples. To further improve the strength and robustness of the attack, a pseudo label steering enhancement strategy is also designed to perturb the weakly-augmented version of unlabeled samples to induce target pseudo label allocations. Additionally, to improve the attack success rate, we amplify the weight of the local backdoored model during FSSL’s model aggregation process to manipulate the game between benign clients and malicious clients. Extensive experiments based on two benchmark datasets demonstrate that the proposed SemSBA scheme can achieve comparable stealthiness against existing attacks.

The cooperative works between connected smart devices in the Industrial Internet of Things (IIoT) have greatly made the growth in productivity and economics for the conventional industry. However, due to the introduction of the communication network, the budding IIoT also confronts the unprecedented cyber threats. To prevent the data from being intercepted by malicious intruders, we propose an efficient and fully secure data sharing work with a short revocation list (DS-SRL) for IIoT. The DS-SRL not only enables flexible access control to the massive data in IIoT but also provides a direct revocation approach for handling the potential issues of key disclosure and membership expiring in application scenarios. Particularly, compared with existing directly revocable ABE works, the revocation list in the DS-SRL scheme will keep constant size even with the increasing number of users. Thus, the consumption for computing and disseminating the revoke-related part of ciphertext are low. This resource-saving merit makes our DS-SRL scheme suitable for IIoT where the smart devices are weak in the ability of both processing and storage. The DS-SRL works without boundary such that the public parameters involved in the system require no predefinitions and can be dynamically adjusted after deployment. Furthermore, the proposed DS-SRL work is demonstrated to be fully secure under the decisional linear assumption. Hence, it owns high flexibility, scalability, and security, which are essential and desirable in real-life applications. Finally, the superior feasibility, efficiency, as well as effectiveness of our DS-SRL work are fairly confirmed by the detailed performance evaluation.

Various stream learning methods are emerging in an endless stream to provide a wealth of solutions for artificial intelligence in streaming data scenarios. However, when each data stream is oriented to a different target space, it forces stream learning approaches oriented to the same task to be no longer applicable. Due to inconsistent target spaces for different tasks, the previous approaches fail on the new streaming tasks or it is impracticable to be trained from scratch with few labeled samples at the beginning. To this end, we have proposed an adaptive learning scheme for few-shot streaming tasks with the contributions of tensor and meta-learning. This adaptive scheme is conducive to mitigating the domain shift when a new task has few labeled samples. We elaborate a novel tensor-empowered attention mechanism derived from nonlocal neural networks, which enables to capture long-range dependency and preserve the high-dimensional structure to refine the global features of streaming tasks. Furthermore, we develop a fine-grained similarity computing approach, which is prone to better characterize the difference across few-shot streaming tasks. To show the superiority of our method, we have carried out extensive experiments on three popular few-shot datasets to simulate streaming tasks and evaluate the performance of adaptation. The results show that our proposed method has achieved competitive performance for few-shot streaming tasks compared with the state-of-the-art (SOTA).

Contrary to the public cloud storage services that impose users to accept the security restrictions delivered by the service provider, users in the private cloud benefit from self-managed, authenticated data access services. However, this may lead to security issues. A critical challenge is the provision of secure and authenticated data storage for the data owner. Moreover, the data owner should be able to access the stored data and share it with others in a controlled manner. In this article, a secure and authenticated data storage, access, and sharing model is proposed for private cloud storage, which has three components. The data storage component provides the user with secure storage of information. The data-sharing component enables sharing the stored data under the control of the data owner. The data access component enables authenticated access to the cloud storage. The security analysis demonstrates that the model is secure against various attacks. The scheme is validated to be secure via the Scyther tool, BAN Logic, and in Random Oracle Model. The performance analysis regarding the computation and communication cost via simulation in OMNeT++ show that it obtains the required security goals and efficiency of computation and communication, compared to the related methods.

Extensive growth in developing new and efficient methods for tensor factorizations has made their intelligent applications in cyber–physical–social systems (CPSS) a hot research topic. Tensor factorizations facilitate the need for recommendations that are accurate and circumstantial, which pushes the limits of traditional collaborative filtering methods to multifaceted versions based on real intelligent environments. Nevertheless, recommenders in edge–cloud computing require information encapsulated in user models to give useful suggestions on user preferred items, which presents stern privacy trepidations. In this article, a novel edge–cloud-aided differentially private tucker decomposition scheme is proposed to avert data owner’s private data from being learned by other data owners, untrusted edge, and cloud during tucker decomposition for CPSS. Our design dissevers users’ private data computations in tucker decomposition to edges from the cloud, and the cloud is forced to perform perturbed results aggregation while preserving privacy. The scheme employs perturbation to ensure differential privacy, and the perturbation noise components are decomposed into small manageable parts that can be locally and independently resolved by edges. Our extensive experiments on two real data sets show the proposed scheme is efficient and has tolerable side effects on the results’ utility.

Nowadays, Internet of Things (IoT) is an attractive system to provide broad connectivity of a wide range of applications, and clouds are natural promoters. Cloud-assisted IoT combines the advantages of cloud computing and IoT, which is able to collect data from the real world and maximizes the value of the collected data by the means of data sharing and data analysis. Meanwhile, secure and convenient data retrieval in cloud servers becomes an important requirement for both enterprises and individual users. Public-key encryption with search functionality (shorten as PKE-SF) is a widely used cryptographic technique that allows users to retrieve encrypted data without decryption. PKE-SF mainly contains the primitives of public-key encryption with keyword search (PKE-KS), public-key encryption with equality test (PKE-ET), and plaintext-checkable encryption (PCE). In light of the overwhelming variety and multitude of PKE-SF schemes, this survey presents these schemes from different perspectives to provide better comprehension for beginners and advanced researchers. More concretely, this survey concentrates on the state of the art of PKE-SF by analyzing the design rationale, examining the framework and security model, and assessing the existing schemes in accordance with theoretic efficiency, security properties, and experimental performance. Furthermore, we discuss the extensions of traditional PKE-SF schemes which feature with the access control delegation, conjunctive keyword search, certificate-free, and offline keyword guessing attack resilience. Finally, we point out some promising directions for readers.

Industrial cyber–physical–social systems (CPSSs), a prominent data-driven paradigm, tightly couple and coordinate social space into cyber–physical systems (CPSs) within industrial environments. With the proliferation of cloud–fog computing, cloud–fog computing becomes the most prominent computing paradigm used to implement industrial data analysis. However, the open environment of cloud–fog computing and the limited control of industrial CPSSs users make industrial data analysis without compromising users’ privacy one great research challenge in practical cloud–fog-based industrial applications. High-order Bi-Lanczos (HOBI-Lanczos) approach has shown remarkable success in heterogeneous data analysis in industrial applications. In this article, a novel privacy preserving HOBI-Lanczos approach using tensor train in cloud–fog computing is proposed for industrial data applications. Specifically, a privacy preserving industrial data analysis model using cloud–fog computing and tensor train is firstly proposed. The proposed model enables fogs and clouds to securely carry out industrial data analysis for large-scale tensors given in a tensor train format. In addition, by using this model, a privacy preserving HOBI-Lanczos approach is provided. Last but not least, by using a brain-controlled robot system case study, the proposed approach is theoretically and empirically analyzed. Our proposed approach is proven to be secure. A series of experiments corroborate the superiority of the proposed approach in cloud–fog computing for industrial applications.

Cyber-physical-social big data concern heterogeneous, multiaspect, large-volume data generated in cyber-physical-social systems (CPSS). Orthogonal tensor SVD (OTSVD) has emerged as a powerful tool to reduce cyber-physical-social big data. In this work, we propose an improved secure high-order-Lanczos based OTSVD for cyber-physical-social big data reduction in clouds. Specifically, to take advantage of the parallel processing capability of cloud computing, the improved secure high-order Lanczos algorithm is derived by restructuring the original high-order Lanczos algorithm such that only one synchronization point per iteration is required. To protect data privacy, the improved secure high-order-Lanczos based OTSVD employs homomorphic encryption integrated with batching technique, and garbled circuits, and makes all computations of the OTSVD algorithm in clouds come true. To our knowledge, this is the first study to efficiently tackle big data reduction in clouds in a privacy-preserving manner. Finally, we prove that our improved approach is secure in semi-trusted model. And we evaluate the proposed improved secure OTSVD on real datasets. The results show that our proposed improved secure approach is efficient and scalable for cyber-physical-social big data reduction.

In the recent past, deep learning has received remarkable acceptance in real-world applications. Social computing expands the existing notion of cyber space and physical space to a more advance cyber-physical-social system (CPSS). Therefore, deep learning provides a propitious technique for accurate mining of information from CPSS, thus facilitates CPSS to offer services of exceptional quality efficiently. However, most of the current deep learning methods are struggling to keep up with the ever-increasing heterogeneous and highly nonlinear dissemination of data. Furthermore, the advancement of deep learning presents privacy concerns. This article proposes a deep private tensor autoencoder (dPTAE), where tensors are used for data representation, and differential privacy guarantees strong privacy. The core idea of our work is to enforce differential privacy through noise injection into the objective functions instead of the results they produce. In addition, the proposed method preserves the privacy of information shared amongst CPSS in smart environments. We applied dPTAE on three representative data sets. Rigorous experimental evaluations and theoretical analysis demonstrate that dPTAE is significantly effective and efficient.

Tucker decomposition has been widely used to extract meaningful and underlying data from heterogeneous data generated by different kinds of devices in a wide range of industrial Internet of Things (IIoT) applications. IIoT data uploaded to the cloud contain personal and sensitive information; thus, there is a growing concern about data privacy. Current existing data analysis solutions, however, assume that the data are reliably and securely collected from different IIoT data providers, an assumption that is not always true in the real world. To address the issues, in this article we propose a privacy-preserving tucker train decomposition based on gradient descent over blockchain-based encrypted IIoT data. Specifically, we use blockchain techniques to enable IIoT data providers to reliably and securely share their data by encrypting them locally before recording them in the blockchain. We use tensor train (TT) theory to build an efficient TT-based tucker decomposition based on gradient descent that tremendously reduces the number of elements to be updated during the tucker decomposition. We utilize the massive resources of fogs and clouds to implement an efficient privacy-preserving tucker train decomposition scheme. We use homomorphic encryption to build our scheme that does complete tucker train decomposition without the involvement of users. Results from a series of extensive experiments on synthetic datasets and real-world datasets demonstrate that our proposed scheme is efficient.

Deep learning techniques have shown significant success in cyber-physical-social systems (CPSS). As an instance of deep learning models, generative adversarial nets (GAN) model enables powerful and flexible image augmentation, image generation, and classification, thus can be applied to real-world CPSS settings. GAN model training needs a large collection of cyber-physical-social data originating from various CPSS devices. Numerous prevailing GAN models depend on a tacit assumption that several cyber-physical-social data providers present a reliable source to collect training data, which is seldom the case in real CPSS. The existing GAN models also fail to consider multi-dimensional latent structure. In our work, we put forward a novel blockchain-enabled tensor-based conditional deep convolutional GAN (TCDC-GAN) model for cyber-physical-social systems. The blockchain is employed to develop a decentralized and reliable cyber-physical-social data-sharing platform between numerous cyber-physical-social data providers, such that the training data and the model are documented on a ledger that is distributed. Furthermore, a tensor-based generator and a tensor-based discriminator are well designed by employing the tensor model. The results of extensive simulation experiments show the efficacy of the proposed TCDC-GAN model. Compared with the state-of-the-art models, our model gains superior estimation performance.

Cyber-physical-social systems (CPSS) are revolutionizing the relationships between humans, computers, and things. Outsourcing computation to the cloud can offer resources-constrained enterprises and consumers sustainable computing in CPSS. However, ensuring the security of data in such an outsourced environment remains a research challenge. Principal eigentensor computation has emerged as a powerful tool dealing with multidimensional cyber-physical-social systems data. In this paper, we present two novel secure principal eigentensor computation (SPEC) schemes for sustainable CPSS. To the best of our knowledge, this is the first effort to address SPEC over encrypted data in the cloud without the interaction need between multiple users and cloud. More specifically, we leverage cloud server and trusted hardware component to design a collaborative cloud model. Using the model, we propose (1) a basic SPEC scheme based on homomorphic computing and (2) an efficient SPEC scheme that combines the advantages of homomorphic computing and garbled circuits, and exploits packing technology to reduce computational cost. Finally, we theoretically and empirically analyze the security and efficiency of our SPEC schemes. Findings demonstrate that the proposed schemes provide a secure and efficient way of outsourcing computation for CPSS. In addition, from the cloud user's perspective, our proposal is lightweight.

With the advent of the 5G era, the Internet of Things (IoT) will flourish in the future. Millions of IoT devices will be connected by 5G networks, which will bring great challenges to network management. Software-defined network (SDN) is a novel solution for managing a large number of IoT devices over the network. In order to solve the problems of secure data analysis in SDN-based IoT, a differentially private tensor computing model (DPTCM) is proposed in this article. Our approach utilizes tensor to model and analyze the SDN-based IoT big data, and an algorithm named differentially private tensor train decomposition (DPTTD) is proposed to achieve secure computing in SDN-based IoT. The algorithm can make full use of the flexible computing power of edge-cloud computing so as to realize collaborative computing between edges, cloud, and the third party. By separating the calculation process of the private data and nonprivate data, the algorithm implements the localized calculation and preservation of the original data which can protect the data security from the source. Meanwhile, we use the differential privacy technology to protect the privacy of data transmitted to the cloud. Finally, we prove that the algorithm satisfies $\varepsilon $ -differential privacy. In the experiments, we verify our model on two real-world data sets. The experimental results show that differential privacy has a little side effect on prediction results, and our model has good performance in data prediction.

Tensors are popular and versatile tools which model multidimensional data. Tensor decomposition has emerged as a powerful technique dealing with multidimensional data. With the booming development of cloud computing, a large number of users are inclined to outsource big data storage and computations to the cloud. However, because of the rise of various privacy concerns, sensitive data usually need to be encrypted prior to being outsourced to a cloud. Computations over encrypted data in the cloud without compromising the privacy of data is still a challenge. This paper presents a novel privacy-preserving tensor decomposition approach over semantically secure encrypted big data. The proposed approach leverages properties of homomorphic encryption and employs a federated cloud to securely decompose an encrypted tensor for multiple users, without the clouds learning any knowledge about users' data. This is, to our knowledge, the first attempt to solve privacy-preserving tensor decomposition without requiring interaction between users and cloud service providers. In addition, in our approach, we present the first secure integer division and integer square root schemes over encrypted data (the dividend, divisor and radicand are in encrypted format). Finally, we prove the security of our approach under semi-trusted model and empirically analyze its effectiveness, which demonstrates the utility of our proposed approach in cloud deployments.

The significant growth of the Internet of Things (IoT) takes a key and active role in healthcare, smart homes, smart manufacturing, and wearable gadgets. Due to complexness and difficulty in processing multimedia data, the IoT based scheme, namely Internet of Multimedia Things (IoMT) exists that is specialized for services and applications based on multimedia data. However, IoMT generated data are facing major processing and privacy issues. Therefore, tensor-based deep computation models proved a better platform to process IoMT generated data. A differentially private deep computation method working in the tensor space can attest to its efficacy for IoMT. Nevertheless, the deep computation model comprises a multitude of parameters; thus, it requires large units of memory and expensive computing units with higher performance levels, which hinders its performance for IoMT. Motivated by this, therefore, the paper proposes a deep private tensor train autoencoder (dPTTAE) technique to deal with IoMT generated data. Notably, the compression of weight tensors to manageable tensor train format is achieved through Tensor Train (TT) network. Moreover, TT format parameters are trained through higher-order back-propagation and gradient descent. We applied dPTTAE on three representative datasets. Comprehensive experimental evaluations and theoretical analysis show that dPTTAE enhances training time efficiency, and greatly improve memory utilization efficiency, attesting its potential for IoMT.

Secure cyber-physical-social big data computations are being increasingly used to protect the users' data security in cyber-physical-social systems (CPSS). Despite the increasing popularity, how to process the tasks of the secure cyber-physical-social big data computations, while taking care of the energy consumption and meeting the users' requirements, remains challenging. To address the problem, in this work, we propose a novel tensor-based optimization model for the secure sustainable cyber-physical-social big data computations. The proposed model is a general and fine-grained model, which can jointly optimize the execution time, energy consumption, reliability, and quality of experience, and can comprehensively take into account step, task, time slot, type, node, core, cryptosystem, and security level. To our knowledge, this is the first study to holistically optimize the tasks in the secure cyber-physical-social big data computations. To illustrate the proposed model, the case study of the secure high-order Lanczos in cloud-assisted CPSS is presented. Finally, the proposed model is empirically evaluated by using multi-objective optimization, and the extensive results demonstrate that from the users' perspective our proposed tensor-based optimization model is preferable for the secure sustainable cyber-physical-social big data computations.

Singular value decomposition (SVD) has been applied to cyber security and cyber forensics since it can reduce data. However, SVD is hard to reduce high-order big data because it is designed for only matrix data initially. Reducing high-order big data is desired for cyber security applications, and is a very challenging issue. In this paper, we propose a novel orthogonal tensor SVD method using big data techniques for high-order big data (naturally represented as tensors) reduction, which can be extensively used in big data applications of cyber security and cyber forensics. More specifically, we first present a high-order lanczos-based orthogonal tensor SVD algorithm to reduce high-order data. Then, for utilizing the incomparable benefits of cloud, we develop a secure orthogonal tensor SVD method to outsource the computation task of the orthogonal tensor SVD algorithm to cloud. The secure orthogonal tensor SVD method can protect data security from untrusted cloud by applying garbled circuits to the orthogonal tensor SVD algorithm. This is, to our best knowledge, the first work to address high-order big data reduction by employing cloud computing. Finally, we analyze the security and efficiency of our proposed orthogonal tensor SVD on synthetic dataset and real network intrusion detection dataset, and the results demonstrate that our proposed method is very promising for big data reduction.

Smart environments, also referred to as cyber-physical-social systems (CPSSs), are expected to significantly benefit from the integration of edge, fog, and cloud for intelligence service flexibility, efficiency, and cost saving. High-order Bi-Lanczos method has emerged as a powerful tool serving as multi-dimensional data processing, such as prevailing feature extraction, classification, and clustering of high-order data, in CPSSs. However, integrated edge-fog-cloud architecture is open and users have very limited control; how to carry out big data processing without compromising the security and privacy is a challenging issue in edge-fog-cloud-assisted smart applications. In this work, we propose a novel and practical privacy-preserving high-order Bi-Lanczos scheme in integrated edge-fog-cloud architectural paradigm for smart environments. More precisely, we first propose a privacy-preserving big data processing model using the synergy of edge, fog, and cloud. The proposed model enables edge, fog, and cloud to cooperatively complete big data processing without compromising users’ privacy for large-scale tensor data in CPSSs. Subsequently, making use of the model, we present a privacy-preserving high-order Bi-Lanczos scheme. Finally, we theoretically and empirically analyze the security and efficiency of the proposed privacy-preserving high-order Bi-Lanczos scheme based on an intelligent surveillance system case study. And the results demonstrate that the proposed scheme provides a privacy-preserving and efficient way of computations in integrated edge-fog-cloud paradigm for smart environments.

With an objective to provide the proactive and personalized services for human beings, Cyber-Physical-Social Systems (CPSS), which combine the cyber space, physical space, and social space together, need to process the large scale heterogenous data first. Tensor, as an appropriate data representation tool, has been widely used for representation of heterogeneous Cyber-Physical-Social big data. When computationally processing such tensor, many necessary constraints have to be taken into account, e.g., the execution time, energy consumption, economic cost, security as well as reliability. However, the systematic integration of these constraints and then the modelling of general optimization for tensor processing become more challenging. In this paper, with such constraints being considered together, a general model for tensor computation that optimizes the execution time, energy consumption, and economic cost with acceptable security and reliability is proposed. From diverse perspectives of user requirements, a case study for the tree-based distributed High-Order Singular Value Decomposition (HOSVD) is measured. With the focus on multi-objective combination, the experimental results validate the applicability and generality of the proposed model.

As the rapidly growing volume of data are beyond the capabilities of many computing infrastructures, to securely process them on cloud has become a preferred solution which can both utilize the powerful capabilities provided by cloud and protect data privacy. This paper puts forward a new approach to securely decompose tensor, the mathematical model widely used in data-intensive applications, to a core tensor and some truncated orthogonal bases. The structured, semi-structured as well as unstructured data are all transformed to low-order sub-tensors which are then encrypted using the fully homomorphic encryption scheme. A unified high-order cipher tensor model is constructed by collecting all the cipher sub-tensors and embedding them to a base tensor space. The cipher tensor is decomposed through a proposed secure algorithm, in which the square root operations are eliminated during the Lanczos procedure. The paper makes an analysis of the secure algorithm in terms of time consumption, memory usage and decomposition accuracy. Experimental results reveals that this approach can securely decompose tensor models. With the advancement of fully homomorphic encryption scheme, the proposed secure tensor decomposition method is expected to be widely applied on cloud for privacy-preserving data processing.