Large Language Models (LLMs) often exhibit sycophantic behavior, agreeing with user-stated opinions even when those contradict factual knowledge. While prior work has documented this tendency, the internal mechanisms that enable such behavior remain poorly understood. In this paper, we provide a mechanistic account of how sycophancy arises within LLMs. We first systematically study how user opinions induce sycophancy across different model families. We find that simple opinion statements reliably induce sycophancy, whereas user expertise framing has a negligible impact. Through logit-lens analysis and causal activation patching, we identify a two-stage emergence of sycophancy: (1) a late-layer output preference shift and (2) deeper representational divergence. We also verify that user authority fails to influence behavior because models do not encode it internally. In addition, we examine how grammatical perspective affects sycophantic behavior, finding that first-person prompts (“I believe...”) consistently induce higher sycophancy rates than third-person framings (“They believe...”) by creating stronger representational perturbations in deeper layers. These findings highlight that sycophancy is not a surface-level artifact but emerges from a structural override of learned knowledge in deeper layers, with implications for alignment and truthful AI systems.

The most efficient asynchronous Byzantine fault tolerant (BFT) framework innagreement settings is due to Ben-Or, Kemler, and Rabin (BKR), wherenis the total number of replicas. Despite recent efforts to bring BKR closer to practical deployment, state-of-the-art designs are still hampered by the inherent mutual waiting between the broadcast and agreement phases. In response, we propose ByzTopia, a new asynchronous BFT protocol that removes this performance bottleneck. Its technical core is to decouple these two phases without introducing additional cryptographic primitives. To enable a more efficient decoupling of the broadcast and agreement phases, we introduce multi-shot reliable broadcast (MRBC), which ensures that replicas deliver messages in a well-ordered sequence across consecutive slots. We implement ByzTopia and evaluate it in various settings. Experimental results show that ByzTopia achieves up to 6.54× the throughput of PACE (for n == 31), the state-of-the-art asynchronous BFT of the same type, and 2.65× that of FIN (for n = 16), the state-of-the-art signature-free asynchronous BFT.

The widespread deployment of large language models (LLMs) in downstream applications has increased the demand for high-quality system prompts, which have become valuable intellectual assets in the commercial prompt marketplace. Recent studies have demonstrated that system prompts are vulnerable to prompt stealing attacks, where adversaries can extract system prompts from LLM applications by crafting adversarial queries, thereby compromising developers' intellectual property and undermining existing business models. However, prior attack methods suffer from critical limitations including gradient dependency and poor scalability, severely restricting their practical applicability. To address these limitations, we present AGFPS, an automated gradient-free framework that leverages evolutionary optimization to systematically steal prompts. Our approach formulates prompt stealing as a discrete optimization problem, where adversarial queries are modeled as individuals in an evolving population. These individuals are optimized through elite retention, selection, adaptive crossover, and mutation operations. To mitigate local optima convergence, we introduce a progressive fitness evaluation strategy based on adaptive sequence fragmentation that exploits LLMs' autoregressive properties. Comprehensive evaluations across multiple benchmark datasets and mainstream LLMs demonstrates that AGFPS achieves a 95.2% exact system prompt stealing success rate, significantly outperforming manual baselines and surpassing gradient-based methods in 80.6% of scenarios. The generated adversarial queries exhibit remarkable transferability across heterogeneous models and diverse datasets, while maintaining robustness against various defense mechanisms. Our work exposes critical vulnerabilities in current LLM deployment practices and underscores the urgent need for enhanced security measures in LLM applications.

It is well understood that the accuracy of machine learning models heavily depends on the amount of training data collected from individuals. However, the collection of sensitive information brings privacy risks to users. Recently, differential privacy (DP) has emerged as a rigorous privacy model for sensitive data collection. When applying DP to training data collection, a common practice to improve utility is that labels are sanitized whereas attribute values are not, a.k.a., label differential privacy (LabelDP). In this paper, we point out that LabelDP can hardly guarantee the expected privacy on labels due to the correlation between attributes and labels. To address this privacy leakage, we propose a stronger privacy model,correlation-aware label local differential privacy(CLLDP), to protect each individual user with the consideration of correlations between attributes and labels. Under CLLDP, we propose a perturbation protocol$k$heads response($k$HR) to estimate the joint probabilistic distribution of attributes and labels. This distribution can be used for a variety of machine learning tasks, such as Naïve Bayes and decision tree, both of which are illustrated in this paper. Through extensive experiments, we show the strong privacy guarantee of CLLDP and its effectiveness in real-life machine learning tasks.

Anonymous credential systems (ACs) enable users to selectively disclose attributes in an anonymous and unlinkable manner, but their misuse is hard to address without traceability. Existing traceable ACs introduce regulators to handle this issue, but they impose substantial communication and computational overhead on regulators, especially in complex scenarios involving large numbers of authentication records and multiple independent regulators. In this work, we propose Public-Key Encryption with Equality Test and Variable Public Generator (PKEET-VPG) and its verifiable version (VPV) for adapting to traceable ACs to support secure outsourceable record retrieval from regulators to service providers. The core technical idea is a session-user-level tracing key whose validity is bound to a single authentication session, thereby preventing the abuse of tracing keys.We formally prove that both schemes achieve OW-CCA2 security, and show that, when integrated into ACs, they preserve user anonymity, support non-frameability, and retain cross-session unlinkability. Theoretical analysis and comparative experiments demonstrate that our schemes can reduce the overhead of regulators at a moderate cost in the authentication phase. Furthermore, our schemes support session-level ciphertext deduplication, which may be of independent interest in some scenarios, such as anonymous voting or one-action-per-session authentication.

Non-transferable learning (NTL) has emerged as a promising method to protect the intellectual property of deep learning models by restricting cross-domain knowledge transfer. However, the robustness of its transferability constraints against potential attacks has not been explored, especially in practical deployment scenarios. In this paper, we propose a novel black-box attack framework - distribution drift learner (DDL), which effectively bypasses NTL protection mechanisms by only accessing input-output queries of protected models. The theoretical foundation of DDL is derived from the concept of data drift, which takes advantage of the variability of the statistical distribution between the source and target domains. The core innovation of DDL is the integration of distributed perception regularization into a lightweight autoencoder architecture, enabling efficient manipulation of data distribution by optimizing dual objectives (distributed perception loss and reconstruction loss). Training for DDL involves two key steps: First, DDL reconstructs a moderate amount of target domain samples and feeds the reconstructed images into the NTL model to obtain prediction labels. The DDL parameters are then updated by optimizing distributed perception loss and reconstruction loss. Through extensive experiments against standard NTL benchmarks (Digits, CIFAR10, and STL10), we demonstrate that DDL has successfully overcome the barriers of the transferable NTL model and improved the accuracy of the target domain by 81% from 10%. Our work reveals critical vulnerabilities in the NTL framework, particularly with respect to ownership verification and applicability authorization mechanisms, providing valuable insights for developing more robust model protection strategies in real-world applications.

The widespread deployment of Internet of Things (IoT) devices has driven their segmentation into distinct trust domains for the purpose of governance, creating a critical need for secure cross-domain authentication (CDA). CDA must preserve both anonymity and traceability of device identities to enable trustworthy data exchange. However, existing approaches, while exploring this trade-off, remain vulnerable to single points of failure and Sybil attacks—threats that are especially severe for unattended and resource-constrained devices. In this paper, we propose a Self-Sovereign and Supervised Cross-domain authentication scheme (SCross) to tackle these issues. The main building block we designed is a pseudonym management scheme (PMS) that allows devices to generate and use pseudonyms without relying on a trusted party. Although devices has full control of their identities, PMS still ensures traceability, Sybil resistance, and revocability. We define the formal security models of PMS, instantiate it under two different approaches, namely group signature (SCross-GS) and zero-knowledge succinct non-interactive arguments of knowledge (zkSNARKs, SCross-ZK), and present security proofs for our proposal. We implemented and evaluated SCross. The result shows that our scheme achieves an effective trade-off between security and efficiency.

Uncovering cause-and-effect mechanisms from data is fundamental to scientific progress. While large language models (LLMs) show promise for enhancing causal discovery (CD) from unstructured data, their application to the increasingly prevalent multimodal setting remains a critical challenge. Even with the advent of multimodal LLMs (MLLMs), their efficacy in multimodal CD is hindered by two primary limitations: (1) difficulty in exploring intra- and inter-modal interactions for comprehensive causal variable identification; and (2) insufficiency to handle structural ambiguities with purely observational data. To address these challenges, we propose MLLM-CD, a novel framework for multimodal causal discovery from unstructured data. It consists of three key components: (1) a novel contrastive factor discovery module to identify genuine multimodal factors based on the interactions explored from contrastive sample pairs; (2) a statistical causal structure discovery module to infer causal relationships among discovered factors; and (3) an iterative multimodal counterfactual reasoning module to refine the discovery outcomes iteratively by incorporating the world knowledge and reasoning capabilities of MLLMs. Extensive experiments on both synthetic and real-world datasets demonstrate the effectiveness of the proposed MLLM-CD in revealing genuine factors and causal relationships among them from multimodal unstructured data. The implementation code and data are available at https://github.com/JinLi-i/MLLM-CD.

Incomplete scenario is a prevalent, practical, yet challenging setting in Multimodal Recommendations (MMRec), where some item modalities are missing due to various factors. Recently, a few efforts have sought to improve the recommendation accuracy by exploring generic structures from incomplete data. However, two significant gaps persist: 1) the difficulty in accurately generating missing data due to the limited ability to capture modality distributions; and 2) the critical but overlooked visibility bias, where items with missing modalities are more likely to be disregarded due to the prioritization of items' multimodal data over user preference alignment. This bias raises serious concerns about the fair treatment of items. To bridge these two gaps, we propose a novel Modality-Diffused Counterfactual (MoDiCF) framework for incomplete multimodal recommendations. MoDiCF features two key modules: a novel modality-diffused data completion module and a new counterfactual multimodal recommendation module. The former, equipped with a particularly designed multimodal generative framework, accurately generates and iteratively refines missing data from learned modality-specific distribution spaces. The latter, grounded in the causal perspective, effectively mitigates the negative causal effects of visibility bias and thus assures fairness in recommendations. Both modules work collaboratively to address the two aforementioned significant gaps for generating more accurate and fair results. Extensive experiments on three real-world datasets demonstrate the superior performance of MoDiCF in terms of both recommendation accuracy and fairness. The code and processed datasets are released at https://github.com/JinLi-i/MoDiCF.

Privacy leakage poses a significant threat when machine learning foundation models trained on private data are released. One such threat is membership inference attacks (MIA), which determine whether a specific example was included in a model's training set. This article shifts focus from developing new MIA algorithms to measuring a model's risk under MIA. We introduce a novel metric, Relative Membership Risk (RMR), which assesses a model's MIA vulnerability from a comparative standpoint. RMR calculates the difference in prediction loss for training examples relative to a predefined reference model, enabling risk comparison across models without needing to delve into details like training strategy, architecture, or data distribution. We also explore the selection of the reference model and show that using a high-risk reference model enhances the accuracy of the RMR measure. To identify the most vulnerable reference model, we propose an efficient iterative algorithm that selects the optimal model from a set of candidates. Through extensive empirical evaluations on various datasets and network architectures, we demonstrate that RMR is an accurate and efficient tool for measuring the membership privacy risk of both individual training examples and the overall machine learning model.

Membership inference attacks (MIAs) compromise the privacy of training data through interrogating a victim machine learning model and inferring whether or not a query sample is in the training data. Existing defenses against MIAs include preprocessing the training data of the model, modifying loss functions, and perturbing the inference output. However, all these mechanisms have to change either the training or inference process, which might be out of reach of the defenders, especially when the models are deployed in a third-party cloud service. In this article, we propose preprocessing the query samples before feeding them into the models for inference. Specifically, we design aMembership Cleansermodule to remove the member information in the query sample by moving it closer to non-member area in the feature space. The membership cleanser does not modify the training or inference process of the machine learning model, so it can be applied to any machine learning system. Through extensive evaluation on four datasets against different models, our approach consistently outperforms the state-of-the-art defense mechanisms in resilience and practicality against various MIAs while retaining good inference accuracy.

Numerical vector aggregation plays a crucial role in privacy-sensitive applications, such as distributed gradient estimation in federated learning and statistical analysis of key-value data. In the context of local differential privacy, this study provides a tight minimax error bound of$O(\frac{ds}{n\epsilon ^{2}})$, where$d$represents the dimension of the numerical vector and$s$denotes the number of non-zero entries. By converting the conditional/unconditional numerical mean estimation problem into a frequency estimation problem, we develop an optimal and efficient mechanism called Collision. In contrast, existing methods exhibit sub-optimal error rates of$O(\frac{d^{2}}{n\epsilon ^{2}})$or$O(\frac{ds^{2}}{n\epsilon ^{2}})$. Specifically, for unconditional mean estimation, we leverage the negative correlation between two frequencies in each dimension and propose the CoCo mechanism, which further reduces estimation errors for mean values compared to Collision. Moreover, to surpass the error barrier in local privacy, we examine privacy amplification in the shuffle model for the proposed mechanisms and derive precisely tight amplification bounds. Our experiments validate and compare our mechanisms with existing approaches, demonstrating significant error reductions for frequency estimation and mean estimation on numerical vectors.

The shuffle model employs a shuffler to anonymize and permute user messages, thereby enhancing privacy/utility trade-offs compared to the local model. Ideally, it assumes perfect message anonymity protection against adversaries, allowing each user to hide among a large population. However, in contexts like mobile/edge networks or in scenarios where the shuffler is curious, this assumption is frequently unrealistic. In this study, we demonstrate the vulnerability of the shuffle model to communication side-channel attacks, which substantially compromise privacy amplification via shuffling. We categorize side-channel information in the shuffle model into three types: (i) in-out information, revealing the victim user’s participation and timing, (ii) message-cardinality information, indicating the victim’s message count, and (iii) message-length information, disclosing the victim’s message length(s). Numerical results indicate these attacks increase privacy loss by 200% to 4100%, revealing secret value with probability more than 90%. After theoretically analyzing the remaining privacy amplification effects, we suggest several countermeasures and principles to alleviate degradation caused by these attacks: (a) appending padding bits to each message to counter message-length attacks, (b) maximizing query parallelization to elude in-out attacks and increase the population for privacy amplification, and (c) sending dummy messages to exchange communication costs for improved privacy amplification effects. The newly proposed paradigms and principles significantly save privacy budget in comparison to current models under attack.

Deep neural network (DNN) models embody the intellectual property of a model owner, as the process of training the DNN model is a complex and resource-intensive task that requires significant investments in data preparation and computing resources. Numerous efforts have been made to protect the intellectual property of DNN models. However, existing methods often come with a critical limitation: they lack robustness, proving effective only in specific intellectual property threat scenarios or they either sacrifice the utility/accuracy of the model owner’s classifier because it interferes with the classifier’s training. To address these issues, we propose GMFIP, a novel generator-based model fingerprinting technology tailored for DNN intellectual property protection. GMFIP stands out for its robustness, extending its utility to various intellectual property threat scenarios rather than specific ones. Furthermore, GMFIP ensures that the utility/accuracy of the model is not affected by protection measures. Specifically, GMFIP begins with the training of the generator, which lays the groundwork for the model fingerprint. The generator generates fingerprints of the unique properties of the source model for verifying model ownership. To further improve the quality of these fingerprints, an extra selection phase dedicated to refining the fingerprints is integrated. Moreover, GMFIP is complemented by a binary classifier, which adapts the threshold setting to get optimal results. Our empirical evaluation includes an ablation study over four state-of-the-art technologies and three image benchmark datasets. Our results demonstrate that GMFIP outperforms other state-of-the-art technologies in effectively distinguishing pirated models from benign models.

We study online data analytics with differential privacy (DP) in decentralized settings. Specifically, online data analytics with local DP protection is widely adopted in real-world applications. Despite numerous endeavors in this field, significant gaps in utility and functionality remain when compared to its offline counterpart. We present an optimal, streamable mechanism:ExSub, for local DP sparse vector estimation. The mechanism enables a range of online analytics on streaming binary vectors, including multi-dimensional binary, categorical, or set-valued data. By leveraging the negative correlation of occurrence events in the sparse vector, we attain an optimal error rate under local privacy constraints, only requiring streamable computations. To surpass the error barrier of local privacy, we also studyExSubrandomizer in the newly emerging (single-message) shuffle model of DP, and provide nearly-tight privacy amplification bounds therein. Additionally, we leverage the online shuffle model that independently permutes users' messages at each timestamp, to design a simplified randomization strategy that can approximately reach Gaussian accuracy in central DP. Through experiments with both synthetic and real-world datasets,ExSubmechanism in the local model have been shown to reduce error by$40\%-60\%$compared to SOTA approaches. TheExSubin the shuffle model can further reduce over$85\%$error, and the online shuffle protocol reduces over$99.7\%$error.

No abstract available.

Facing enormous data generated at the network edge, edge intelligence (EI) emerges as the fusion of edge computing and AI, revolutionizing edge data processing and intelligent decision-making. Nonetheless, this emergent mode presents a complex array of security challenges, particularly prominent in image-centric applications due to the sheer volume of visual data and its direct connection to user privacy. These challenges include safeguarding model/image privacy and ensuring model integrity against various security threats, such as model poisoning. Essentially, those threats originate from data attacks, suggesting data protection as a promising solution. Although data protection measures are well-established in other domains, image-centric EI necessitates focused research. This survey examines the security issues inherent to image-centric EI and outlines the protection efforts, providing a comprehensive overview of the landscape. We begin by introducing EI, detailing its operational mechanics and associated security issues. We then explore the technologies facilitating security enhancement (e.g., differential privacy) and EI (e.g., compact networks and distributed learning frameworks). Next, we categorize security strategies by their application in data preparation, training, and inference, with a focus on image-based contexts. Despite these efforts on security, our investigation identifies research gaps. We also outline promising research directions to bridge these gaps, bolstering security frameworks in image-centric EI applications.

DRAM-based Processing in Memory (PIM) addresses the “memory wall” problem by incorporating computing units (PIM units) into main memory devices for faster and wider local data access. However, critical challenges prevent PIM units from being compatible with existing CPU hosts. Memory interleaving and virtual memory limit the size of contiguous data visible to PIM units that constrains the granularity of PIM tasks. Fine-grained PIM tasks result in significant CPU-PIM offloading overhead, offsetting the speed-up of PIM. Existing PIM systems adopt drastic measures to ensure PIM task offloading efficiency, including isolating PIM memory space and turning off global memory interleaving. These interventions, however, decrease the CPU’s memory bandwidth and introduce extra data transfer, leading to an additional “system memory wall”. This new “wall” must be eliminated before fully embracing the PIM technology. In this work, we propose UM-PIM, a PIM system with interleaved CPU pages and non-interleaved PIM pages coexisting in a Uniform and Shared Memory space. UM-PIM enables zero-copy during PIM task offloading and maintains the CPU’s memory bandwidth while ensuring PIM offloading efficiency. Firstly, we propose a dual-track memory management mechanism consisting of independent page allocation and address translation for the two kinds of pages, respectively. Second, we design UM-PIM interface hardware on the DIMM (with PIMs) side to provide a dynamic address mapping for accelerating the data re-layout. Finally, we provide APIs to reduce PIM-to-PIM communication overhead by optimizing the CPU’s access to PIM pages in different communication modes. We compare UM-PIM with a CPU system and the current PIM systems. Results show negligible performance degradation for CPU workloads ($\lt 0.1 \%$) on UM-PIM, contrasting with the $25.8 \%$ degradation on the current PIM system with memory interleaving switched off. For PIM workloads partitioned to CPU and PIM units, UM-PIM can reduce the CPU time by $4.93 \times$, resulting in an end-to-end $1.96 \times$ speedup on average.

The shuffle model of differential privacy provides promising privacy-utility balances in decentralized, privacy-preserving data analysis. However, the current analyses of privacy amplification via shuffling lack both tightness and generality. To address this issue, we propose the variation-ratio reduction as a comprehensive framework for privacy amplification in both single-message and multi-message shuffle protocols. It leverages two new parameterizations: the total variation bounds of local messages and the probability ratio bounds of blanket messages, to determine indistinguishability levels. Our theoretical results demonstrate that our framework provides tighter bounds, especially for local randomizers with extremal probability design, where our bounds are exactly tight. Additionally, variation-ratio reduction complements parallel composition in the shuffle model, yielding enhanced privacy accounting for popular sampling-based randomizers employed in statistical queries (e.g., range queries, marginal queries, and frequent itemset mining). Empirical findings demonstrate that our numerical amplification bounds surpass existing ones, conserving up to 30% of the budget for single-message protocols, 75% for multi-message ones, and a striking 75%-95% for parallel composition. Our bounds also result in a remarkably efficient Õ ( n ) algorithm that numerically amplifies privacy in less than 10 seconds for n = 10 8 users.

Off-chain transactions seek to address the low on-chain scalability and enable blockchain-based payments over unreliable on-chain networks. The key problem with existing works is that they fail to balance security and flexibility in their designs. These studies would have been more useful if they could provide a sense of security without compromising their flexibility. We hypothesize that two offline parties having loosely synchronized clocks and channels with known bounded latency can conduct off-chain transactions while maintaining a high level of security and flexibility: we introduce a novel blockchain-based offline payment protocol that supports our hypothesis. Our work leverages on-chain smart contracts and offline wallet interactions to build resilience against intermittent on-chain connectivity. Our protocol achieves flexible and trusted computations with the use of platform-agnostic Trusted Execution Environments (TEEs) and open transactions. We empirically evaluate our design over the mainstream Intel Software Guard Extensions (SGX) and compare our protocol with state-of-the-art solutions. We found that our protocol attains high efficiency and exhibits an advanced level of security and flexibility in functionality. We evaluate our construction against several real-world attacks. We prove the security and robustness of our scheme based on a practical universally composable framework with synchronous settings. This work contributes to the existing knowledge of safe and user-friendly offline payment solutions for the blockchain technology.

Recent studies have shown that deep learning-based classifiers are vulnerable to malicious inputs, i.e., adversarial examples. A practical solution is to construct a perceptible but localized perturbation called patch, making the well-trained models misclassified. However, most existing patch-based adversarial attacks focus on designing patches with localized rectangles, squares, or grids, ignoring the effect of the non-local patch. In this paper, we propose a novel cross-shaped patch attack paradigm (CSPA), a simple yet efficient and effective adversarial attack in Black-box scenarios. Specifically, the cross-shaped patch consists of two line segments intersected and perpendicular to each other at the midpoint. These two line segments are designed to be sufficiently thin and long to reach the four corners of the input image nearly. Thus, the patch has a globalized perturbation capacity while preserving its continuousness. The content and location of cross-shaped patch are then iteratively optimized by a carefully contrived random search-based algorithm to maximize this global property. Comprehensive experiments are conducted on four benchmark datasets against various victim networks. The results show that the proposed CSPA outperforms the existing patch-based attacks regarding both attack success rate and query efficiency by a large margin. Specifically, compared with the baselines, CSPA increases the success rate by up to 20% on ImageNet and reaches 100% on the CIFAR-100 and CIFAR-10 datasets. Meanwhile, CSPA reduces the average number of queries by up to 7 times. Even for the white-box attack scenario, our designed cross-shaped patch can still be applicable, achieving state-of-the-art performance.

A wide range of Arm endpoints leverage integrated and discrete GPUs to accelerate computation. However, Arm GPU security has not been explored by the community. Existing work has used Trusted Execution Environments (TEEs) to address GPU security concerns on Intel-based platforms, but there are numerous architectural differences that lead to novel technical challenges in deploying TEEs for Arm GPUs. There is a need for generalizable and efficient Arm-based GPU security mechanisms. To address these problems, we presentStrongBox, the first GPU TEE for secured general computation on Arm endpoints.StrongBoxprovides an isolated execution environment by ensuring exclusive access to GPU. Our approach is based in part on a dynamic, fine-grained memory protection policy as Arm-based GPUs typically share a unified memory with the CPU. Furthermore,StrongBoxreduces runtime overhead from the redundant security introspection operations. We also design an effective defense mechanism withinsecure worldto protect the confidential GPU computation. Our design leverages the widely-deployed Arm TrustZone and generic Arm features, without hardware modification or architectural changes. We prototypeStrongBoxusing an off-the-shelf Arm Mali GPU and perform an extensive evaluation. Results show thatStrongBoxsuccessfully ensures GPU computation security with a low (4.70%–15.26%) overhead.

Asynchronous common subset (ACS) is an essential building block for Byzantine fault-tolerance and multi-party computation. The classic ACS framework is due to Ben-Or, Kemler, and Rabin (BKR), consisting of${n}$reliable broadcast (RBC) instances and${n}$asynchronous binary agreement (ABA) instances (where${n}$is the total number of replicas). Despite recent progresses of practical BKR-ACS, the state-of-the-art designs are still trapped by a validity-latency tradeoff. In this paper, we propose Sweeper, a new ACS protocol that breaks the tradeoff, achieving optimal validity and latency. Moreover, Sweeper maintains other benefits including optimal resilience, signature-free, and information-theoretic settings. Sweeper is built on RBC and composable biased reproposable ABA (CBiased RABA). Different from the conventional RABA, CBiased RABA allows replicas to be more biased towards specific RABA instances. We provide generic strategies to transform existing ABA/RABA protocols and a new RABA protocol that we introduce, to CBiased RABA. Furthermore, Sweeper can achieve up to$2 \times $the throughput of PACE-ACS (CCS 2022), the state-of-the-art ACS protocol that follows the BKR-ACS framework.

With the rapid growth of DeepFake video techniques, it becomes increasingly challenging to identify them visually, posing a huge threat to our society. Unfortunately, existing detection schemes are limited to exploiting the artifacts left by DeepFake manipulations, so they struggle to keep pace with the ever-improving DeepFake models. In this work, we propose DeepMark, a scalable and robust framework for detecting DeepFakes. It imprints essential visual features of a video into DeepMark Meta (DMM) and uses it to detect DeepFake manipulations by comparing the extracted visual features with the ground truth in DMM. Therefore, DeepMark is future-proof, because a DeepFake video must aim to alter some visual feature, no matter how “natural” it looks. Furthermore, DMM also contains a signature for verifying the integrity of the above features. And an essential link to the features as well as their signature is attached with error correction codes and embedded in the video watermark. To improve the efficiency of DMM creation, we also present a threshold-based feature selection scheme and a deduced face detection scheme. Experimental results demonstrate the effectiveness and efficiency of DeepMark on DeepFake video detection under various datasets and parameter settings.

To counter adversarial facial accessory presentation attacks (PAs), a detection method based on local face differential is proposed in this article. It extracts the local face differential features from a suspected face image and a reference face image, and then adaptively fuses the differential features of different local face regions to detect adversarial facial accessory PAs. Meanwhile, the principle of the proposed method is explained by theoretically investigating the local facial differences between a bona fide presentation and an adversarial facial accessory PA when they are compared with a reference face image. To evaluate the proposed method, this article builds a database with different adversarial examples (AEs), presentation attack instruments (PAIs), illumination conditions, and cameras. The experimental results show that it can effectively distinguish between adversarial facial accessory PAs and bona fide presentations, and it has good generalization ability to unseen AEs, PAIs, illumination conditions, and cameras. Moreover, it outperforms the existing AE detection and presentation attack detection methods in detecting adversarial facial accessory PAs.

Deep neural networks excel at solving intuitive tasks that are hard to describe formally, such as classification, but are easily deceived by maliciously crafted samples, leading to misclassification. Recently, it has been observed that the attack-specific robustness of models obtained through adversarial training does not generalize well to novel or unseen attacks. While data augmentation through mixup in the input space has been shown to improve the generalization and robustness of models, there has been limited research progress on mixup in the latent space. Furthermore, almost no research on mixup has considered the robustness of models against emerging on-manifold adversarial attacks. In this paper, we first design a latent-space data augmentation strategy called dual-mode manifold interpolation, which allows for interpolating disentangled representations of source samples in two modes: convex mixing and binary mask mixing, to synthesize semantic samples. We then propose a resilient training framework, LatentRepresentationMixup (LarepMixup), that employs mixed examples and softlabel-based cross-entropy loss to refine the boundary. Experimental investigations on diverse datasets (CIFAR-10, SVHN, ImageNet-Mixed10) demonstrate that our approach delivers competitive performance in training models that are robust to off/on-manifold adversarial example attacks compared to leading mixup training techniques.

Knowledge distillation has emerged as a scalable and effective way for privacy-preserving machine learning. One remaining drawback is that it consumes privacy in a client-level manner. In order to attain fine-grained privacy accountant and improve utility, this work proposes a model-free reverse k-NN labeling method towards record-level private knowledge distillation, where each private record is employed for labeling at most k queries. Theoretically, we provide bounds of labeling error rate under the centralized/local model of differential privacy. Experimentally, we demonstrate that it achieves new state-of-the-art accuracy in MNIST/SVHN/CIFAR-10 dataset with one order of magnitude lower of privacy loss.

Local Differential Privacy (LDP) is now widely adopted in large-scale systems to collect and analyze sensitive data while preserving users’ privacy. However, almost all LDP protocols rely on a semi-trust model where users are curious-but-honest, which rarely holds in real-world scenarios. Recent works [6], [11], [62] show poor estimation accuracy of many LDP protocols under malicious threat models. Although a few works have proposed some countermeasures to address these attacks, they all require prior knowledge of either the attacking pattern or the poison value distribution, which is impractical as they can be easily evaded by the attackers.In this paper, we adopt a general opportunistic-and-colluding threat model and propose a multi-group Differential Aggregation Protocol (DAP) to improve the accuracy of mean estimation under LDP. Different from all existing works that detect poison values on individual basis, DAP mitigates the overall impact of poison values on the estimated mean. It relies on a new probing mechanism EMF (i.e., Expectation-Maximization Filter) to estimate features of the attackers. In addition to EMF, DAP also consists of two EMF post-processing procedures (EMF* and CEMF*), and a group-wise mean aggregation scheme to optimize the final estimated mean to achieve the smallest variance. Extensive experimental results on both synthetic and real-world datasets demonstrate the superior performance of DAP over state-of-the-art solutions.

The social trust assessment can spur extensive applications such as social recommendations, shopping, financial investment strategies, etc, but remain a challenging problem having limited exploration. Such explorations mainly limit their studies to static network topology or simplified dynamic networks, toward the social trust relationship prediction. In contrast, in this paper, we explore the social trust by taking into account the time-varying online social networks whereas the social trust relationship may vary over time. The DTrust, a dynamic graph neural network-based solution, will be proposed for accurate social trust prediction. In particular, DTrust is composed of a static aggregation unit and a dynamic unit, respectively responsible for capturing both the spatial dependence features and temporal dependence features. In the former unit, we stack multiple NNConv layers derived from the edge-conditioned convolution network for capturing the spatial dependence features correlated to the network topology and the observed social relationships. In the latter unit, a gated recurrent unit (GRU) is employed for learning the evolution law of social interaction and social trust relationships. Based on the extracted spatial and temporal features, we then employ a fully connected neural network for learning, able to predict the social trust relationships for both current and future time slots. Extensive experimental results exhibit that our DTrust can outperform the benchmark counterparts on two real-world datasets.

Federated Learning (FL), the de-facto distributed machine learning paradigm that locally trains datasets at individual devices, is vulnerable to backdoor model poisoning attacks. By compromising or impersonating those devices, an attacker can upload crafted malicious model updates to manipulate the global model with backdoor behavior upon attacker-specified triggers. However, existing backdoor attacks require more information on the victim FL system beyond a practical black-box setting. Furthermore, they are often specialized to optimize for a single objective, which becomes ineffective as modern FL systems tend to adopt in-depth defense that detects backdoor models from different perspectives. Motivated by these concerns, in this paper, we propose 3DFed, an adaptive, extensible, and multi-layered framework to launch covert FL backdoor attacks in a black-box setting. 3DFed sports three evasion modules that camouflage backdoor models: backdoor training with constrained loss, noise mask, and decoy model. By implanting indicators into a backdoor model, 3DFed can obtain the attack feedback in the previous epoch from the global model and dynamically adjust the hyper-parameters of these backdoor evasion modules. Through extensive experimental results, we show that when all its components work together, 3DFed can evade the detection of all state-of-the-art FL backdoor defenses, including Deepsight, Foolsgold, FLAME, FL-Detector, and RFLBAT. New evasion modules can also be incorporated in 3DFed in the future as it is an extensible framework.

In the era of the Internet of Things (IoT), remote sensors and endpoint appliances generate vast amounts of data. Decentralized and collaborative learning builds on these IoT data to enable classification and recognition tasks by inviting multiple data owners. Federated learning (FL), as a popular collaborative learning framework, can significantly improve the performance of models without collecting the original data. To invite data owners to participate in FL, various incentive mechanisms are designed to address this issue by researchers. However, existing solutions still face high costs and low utility due to information asymmetry, where the reputation, computation power, and data quantity of the data owners are not known in advance. Therefore, we propose a Stackelberg game-based multifactor incentive mechanism for FL (SGMFIFL). First, we design the Top-$K$cost selection algorithm based on reverse auction, which can reduce the cost of selecting data owners. Next, we devise a multifactor reward function based on reputation, accuracy, and reward rate, the data owners with high reputation and high accuracy will be of more reward. In particular, to ensure that SGMFIFL can provide reliable incentives in IoT, we use blockchain to provide a secure and trusted environment. Finally, we construct a two-stage Stackelberg game model for the task publisher and the data owners and derive an optimal Equilibrium solution for both stages of the whole game. Experiments conducted on two well-known data sets, MNIST and CIFAR10, demonstrate the significant performance of the proposed mechanism.

Aiming at the large distortion and low tampering localization accuracy of the existing semi-fragile reversible watermarking for 3D mesh models, a novel semi-fragile reversible watermarking for 3D models using spherical crown volume division is proposed. The crown volume of a sphere is divided to reduce the embedding distortion. The possible geometric and topological transformations are separately considered in the watermark generation, and the vertices of the one-ring neighbourhood are grouped to improve the tampering localization accuracy. Experimental results show that the proposed scheme can achieve better localization accuracy and lower embedding distortion than some state-of-the-art algorithms. It has good potential for the applications in integrity authentication for 3D mesh models.

Federated learning allows a large number of resource-constrained clients to train a globally-shared model together without sharing local data. These clients usually have only a few classes (categories) of data for training, where the data distribution is non-iid (not independent identically distributed). In this article, we put forward the concept ofcategory privacyfor the first time to indicatewhich classes of data a client has, which is an important but ignored privacy goal in the federated learning with non-iid data. Although secure aggregation protocols are designed for federated learning to protect the input privacy of clients, we perform the first systematic study oncategory inference attackand demonstrate that these protocols cannot fully protect category privacy. We design a differential selection strategy and two de-noising approaches to achieve the attack goal successfully. In our evaluation, we apply the attack to non-iid federated learning settings with various datasets. On MNIST, CIFAR-10, AG_news, and DBPedia dataset, our attack achieves$>90\%$accuracy measured in F1-score in most cases. We further consider a possible detection method and propose two strategies to make the attack more inconspicuous.

The success of machine learning (ML) depends on the availability of large-scale datasets. However, recent studies have shown that models trained on such datasets are vulnerable to privacy attacks, among which membership inference attack (MIA) brings serious privacy risk. MIA allows an adversary to infer whether a sample belongs to the training dataset of the target model or not. Though a variety of defenses against MIA have been proposed such as differential privacy and adversarial regularization, they also result in lower model accuracy and thus make the models less unusable. In this article, aiming at maintaining the accuracy while protecting the privacy against MIA, we propose a new defense against membership inference attacks by generative adversarial network (GAN). Specifically, sensitive data is used to train a GAN, then the GAN generate the data for training the actual model. To ensure that the model trained with GAN on small datasets can has high utility, two different GAN structures with special training techniques are utilized to deal with the image data and table data, respectively. Experiment results show that the defense is more effective on different data sets against the existing attack schemes, and is more efficient compared with most advanced MIA defenses.

While Deep Reinforcement Learning (DRL) has achieved outstanding performance in extensive applications, exploiting its vulnerability with adversarial attacks is essential towards building robust DRL systems. In this work, we aim to propose a novel Decoupled Adversarial Policy (DAP) for attacking the DRL mechanism, whereas the adversarial agent can decompose the adversarial policy into two separate sub-policies: 1) the switch policy which determines if an attacker should launch the attack, and 2) the lure policy which determines the action an attacker induces the victim to take. If the adversarial agent samples an injection action from the switch policy, the attacker can query the pre-constructed database for universal perturbation in the real-time manner, misleading the victim to take the induced action sampled from the lure policy. To train the adversarial agent to learn DAP, we utilize those samples wherein both of the sub-actions from DAP are not restricted by each other or by the external constraint, but can actually affect the attacker’s behaviors. Therefore, we propose trajectory clipping and padding in data pruning, and Decoupled Proximal Policy Optimization (DPPO) in optimizing. Extensive experiments on different Atari games demonstrate the effectiveness of our proposed method. In addition, it can simultaneously implement the real-time and few-steps attack, which outperforms the existing counterparts.

Recently, generative steganography that transforms secret information to a generated image has been a promising technique to resist steganalysis detection. However, due to the inefficiency and irreversibility of the secret-to-image transformation, it is hard to find a good trade-off between the information hiding capacity and extraction accuracy. To address this issue, we propose a secret-to-image reversible transformation (S2IRT) scheme for generative steganography. The proposed S2IRT scheme is based on a generative model, i.e., Glow model, which enables a bijective-mapping between latent space with multivariate Gaussian distribution and image space with a complex distribution. In the process of S2I transformation, guided by a given secret message, we construct a latent vector and then map it to a generated image by the Glow model, so that the secret message is finally transformed to the generated image. Owing to good efficiency and reversibility of S2IRT scheme, the proposed steganographic approach achieves both high hiding capacity and accurate extraction of secret message from generated image. Furthermore, a separate encoding-based S2IRT (SE-S2IRT) scheme is also proposed to improve the robustness to common image attacks. The experiments demonstrate the proposed steganographic approaches can achieve high hiding capacity (up to 4bpp) and accurate information extraction (almost 100% accuracy rate) simultaneously, while maintaining desirable anti-detectability and imperceptibility.

In big data era, companies and organizations are keen to collect data from users and analyse their behaviour patterns to make decisions or predictions for profits. However, it undermines users’ privacy because the collected data can be quite sensitive and easy to leak. To address privacy problems, local differential privacy (LDP) has been proposed for untrusted data collectors to obtain statistical information without compromising user privacy. Most studies on LDP assume that all users fully cooperate and contribute to the data collection process and thus the collected dataset is complete. However, in practice, especially when user population is large, such assumption seldom holds due to communication loss, user unresponsiveness or unwillingness, and incomplete user-side data. Unfortunately, state-of-the-art LDP-based data collection schemes, such as GRR, OUE and OLH, cannot handle partial data collection effectively. In this paper, we propose collaborative sampling to address partial data collection in a multi-dimensional setting. Thanks to a two-phase mechanism, we can derive the optimal sampling rate for each dimension. The optimality is shown and proved with respect to the variance of estimated frequency. Besides that, collaborative sampling is general and can be used in GRR, OUE and OLH with minimal adaption. Through experimental results, we show collaborative sampling outperforms existing mainstream data collection schemes in partial multi-dimensional data collection.

Human action recognition (HAR) is one of most important tasks in video analysis. Since video clips distributed on networks are usually untrimmed, it is required to accurately segment a given untrimmed video into a set of action segments for HAR. As an unsupervised temporal segmentation technology, subspace clustering learns the codes from each video to construct an affinity graph, and then cuts the affinity graph to cluster the video into a set of action segments. However, most of the existing subspace clustering schemes not only ignore the sequential information of frames in code learning, but also the negative effects of noises when cutting the affinity graph, which lead to inferior performance. To address these issues, we propose a sequential order-aware coding-based robust subspace clustering (SOAC-RSC) scheme for HAR. By feeding the motion features of video frames into multi-layer neural networks, two expressive code matrices are learned in a sequential order-aware manner from unconstrained and constrained videos, respectively, to construct the corresponding affinity graphs. Then, with the consideration of the existence of noise effects, a simple yet robust cutting algorithm is proposed to cut the constructed affinity graphs to accurately obtain the action segments for HAR. The extensive experiments demonstrate the proposed SOAC-RSC scheme achieves the state-of-the-art performance on the datasets of Keck Gesture and Weizmann, and provides competitive performance on the other 6 public datasets such as UCF101 and URADL for HAR task, compared to the recent related approaches.

Social trust assessment that characterizes a pairwise trustworthiness relationship can spur diversified applications. Extensive efforts have been put in exploration, but mainly focusing on applying graph convolutional network to establish a social trust evaluation model, overlooking user feature factors related to context-aware information on social trust prediction. In this article, we aim to design a new trust assessment framework GATrust which integrates multi-aspect properties of users, including user context-specific information, network topological structure information, and locally-generated social trust relationships. GATrust can assigns different attention coefficients to multi-aspect properties of users in online social networks, for improving the prediction accuracy of social trust evaluation. The framework can then learn multiple latent factors of each trustor-trustee pair to establish a social trust evaluation model, by fusing graph attention network and graph convolution network. We conduct extensive experiments on two popular real-world datasets and the results exhibit that our proposed framework can improve the precision of social trust prediction, outperforming the state-of-the-art in the literature by 4.3% and 5.5% on both two datasets, respectively.

Jointly learning from multiple datasets can help building versatile intelligent systems yet may give rise to serious concerns of data privacy and model selection. Specifically, on the one hand, these datasets can be distributed at various local clients, who may not be willing or do not ought to share data with each other. On the other hand, it is unrealistic to choose a model architecture that can well suit the disparate patterns and distributions carried by the various datasets in a priori. Whereas many works in federated learning [1] and neural architecture search [2] have been proposed to address one of the two concerns, very few have attempted the both. To close the gap, in this paper we deliver a framework, termedMulti-Granular Federated Neural Architecture Search(MGFNAS), to enable the automation of model architecture search in a federated and thus privacy-preserved setting. We argue that our MGFNAS framework is general in the sense that it does not impose any restriction on the search space or strategy, such that most existing neural architecture search techniques can be readily implemented in. The main idea of our framework is to search the optimal neural network architecture in two levels of granularity, enabling the neural-operator-basedmicro-levelsearch and the cell-basedmacro-levelsearch. The main challenge of implementing our framework lies in the fact that, due to the decentralized nature, the local architectures searched by multiple clients can differ drastically in order to fit their own datasets, while a general method to form the global model by aggregating the local architectures in both micro and macro levels is missing. To solve the issue, we propose a novel aggregation function, named Network Architecture Probabilistic Aggregation (NAPA). The key idea of our NAPA function is to treat the network architectures as graphs, of which the sub-graph structures being frequently appeared across multiple clients are modeled by probabilistic distributions. At each round, a global model is formed by sampling from those distributions in an exploration-exploitation fashion. Extensive experiments are carried out, and the results substantiate the viability and effectiveness of our proposed framework.

Mobile edge computing (MEC) is a promising edge technology to provide high bandwidth and low latency shared services and resources to mobile users. However, the MEC infrastructure raises major security concerns when the shared resources involve sensitive and private data of users. This paper proposes a novel blockchain-based key management scheme for MEC that is essential for ensuring secure group communication among the mobile devices as they dynamically move from one subnetwork to another. In the proposed scheme, when a mobile device joins a subnetwork, it first generates lightweight key pairs for digital signature and communication, and broadcasts its public key to neighbouring peer users in the subnetwork blockchain. The blockchain miner in the subnetwork packs all the public key of mobile devices into a block that will be sent to other users in the subnetwork. This enables the mobile device to communicate with its peers in the subnetwork by encrypting the data with the public key stored in the blockchain. When the mobile device moves to another subnetwork in the tree network, all the mobile devices of the new subnetwork can quickly verify its identity by checking its record in the local or higher hierarchy subnetwork blockchain. Furthermore, when the mobile device leaves the subnetwork, it does not need to do anything and its records will remain in the blockchain which is an append-only database. Theoretical security analysis shows that the proposed scheme can defend against the 51 percent attack and malicious entities in the blockchain network utilizing Proof-of-Work consensus mechanism. Moreover, the backward and forward secrecy is also preserved. Experimental results demonstrate that the proposed scheme outperforms two baselines in terms of computation, communication and storage.

A wide range of Arm endpoints leverage integrated and discrete GPUs to accelerate computation such as image processing and numerical processing applications. However, in spite of these important use cases, Arm GPU security has yet to be scrutinized by the community. By exploiting vulnerabilities in the kernel, attackers can directly access sensitive data used during GPU computing, such as personally-identifiable image data in computer vision tasks. Existing work has used Trusted Execution Environments (TEEs) to address GPU security concerns on Intel-based platforms, while there are numerous architectural differences that lead to novel technical challenges in deploying TEEs for Arm GPUs. In addition, extant Arm-based GPU defenses are intended for secure machine learning, and lack generality. There is a need for generalizable and efficient Arm-based GPU security mechanisms.

Summary Network intrusions are a big security threat to current computer networks. For protection, collaborative intrusion detection networks (CIDNs) are developed attempting to reach better detection performance than a single detector, by allowing a set of detectors to switch data or information with each other. However, there is a need to implement suitable trust management schemes, with the aim to safeguard such distributed detection networks against insider threats. In the literature, previous studies have indicated that the notion of intrusion sensitivity can be used to enhance the effectiveness of trust management, by highlighting the feedback from expert nodes. In addition, machine learning can be used to assign the value of intrusion sensitivity automatically. In this work, we evaluate the performance of typical supervised learning classifiers in allocating the value of intrusion sensitivity, and figure out some limitations under different data sets. Then we investigate the impact of intrusion sensitivity in a real network environment under adversarial conditions. The results demonstrate that a wrongly assigned sensitivity value may greatly degrade the detection effectiveness of insider attacks. There is a significant need to choose a suitable classifier in allocating the value of intrusion sensitivity in practice.

A contrastive distortion-level learning-based no-reference image-quality assessment (NR-IQA) framework is proposed in this study to further effectively model various distortion types with the same or different distortion levels. The proposed method aims to improve the prediction accuracy of NR-IQA. The proposed method consists of three parts: multiscale distortion-level representation learning, single-image NR-IQA, and a representation affinity module, which can reduce NR-IQA computational complexity while maintaining a low-distortion representation of high-distortion inputs. The proposed NR-IQA method aims to extract distributional features of samples in real distorted images and predict ambiguity based on distortion-level learning. Experimental results show that by comparing on many NR-IQA data sets the proposed method can outperform state-of-the-art methods.

Resource allocation is a fundamental research issue in IoT edge computing, and reinforcement learning is fast becoming a common solution. The majority of the current techniques involve decision makers who determine how and where resources should be distributed. In a standard cloud system, this decision maker is a central server. In an edge system, the decision maker is an edge host. Both approaches have drawbacks. Edge hosts do not always have access to enough global information to create the most optimal resource allocation strategy. Central servers do but at the cost of privacy. A solution is needed that can do both. This article, therefore, presents a novel resource allocation method called concurrent federated reinforcement learning. The scheme inherits the privacy protection of federated learning, the complex problem solving power of reinforcement learning and adds concurrency in the form of joint decision making so the resource allocation strategies work to the benefit of the global system. The experiments demonstrate that the approach provides the state-of-the-art performance in system-wide utility, speed of task completion, and resource utilization.

Although significant progress has been achieved recently in automatic learning of steganographic cost, the existing methods designed for spatial images cannot be directly applied to JPEG images which are more common media in daily life. The difficulties of migration are mainly caused by the characteristics of the$8\times 8$DCT mode structure. To address the issue, in this paper we extend an existing automatic cost learning scheme to JPEG, where the proposed scheme called JEC-RL (JPEG Embedding Cost with Reinforcement Learning) is explicitly designed to tailor the JPEG DCT structure. It works with the embedding action sampling mechanism under reinforcement learning, where a policy network learns the optimal embedding policies via maximizing the rewards provided by an environment network. Following a domain-transition design paradigm, the policy network is composed of three modules, i.e., pixel-level texture complexity evaluation module, DCT feature extraction module, and mode-wise rearrangement module. These modules operate in serial, gradually extracting useful features from a decompressed JPEG image and converting them into embedding policies for DCT elements, while considering JPEG characteristics including inter-block and intra-block correlations simultaneously. The environment network is designed in a gradient-oriented way to provide stable reward values by using a wide architecture equipped with a fixed preprocessing layer with$8\times 8$DCT basis filters. Extensive experiments and ablation studies demonstrate that the proposed method can achieve good security performance for JPEG images against both advanced feature-based and modern CNN-based steganalyzers.

Since big data becomes a main impetus to the next generation of IT industry, data privacy has received considerable attention in recent years. To deal with the privacy challenges, differential privacy has been widely discussed and related private mechanisms are proposed as privacy-enhancing techniques. However, with today’s differential privacy techniques, it is difficult to generate a sanitized dataset that can suit every machine learning task. In order to adapt to various tasks and budgets, different kinds of privacy mechanisms have to be implemented, which inevitably incur enormous costs for computation and interaction. To this end, in this article, we propose two novel schemes for outsourcing differential privacy. The first scheme efficiently achieves outsourcing differential privacy by using our preprocessing method and secure building blocks. To support the queries from multiple evaluators, we give the second scheme that employs a trusted execution environment to aggregately implement privacy mechanisms on multiple queries. During data publishing, our proposed schemes allow providers to go off-line after uploading their datasets, so that they achieve a low communication cost which is one of the critical requirements for a practical system. Finally, we report an experimental evaluation on UCI datasets, which confirms the effectiveness of our schemes.

To achieve efficiently search and update on outsourced encrypted data, dynamic searchable symmetric encryption (DSSE) was proposed by just leaking some well-defined leakages. Though small, many recent works show that an attacker can exploit these leakages to undermine the security of existing DSSE schemes. In particular, an attacker can exploit even seemingly harmless size pattern to perform severe attacks. Many exiting schemes resort to oblivious RAM (ORAM) to hide search/access pattern; however, even such powerful cryptographic primitive cannot protect size pattern leakage. In this article, we first show that size pattern can lead to more information leakages, which is not well studied or protected by existing schemes. We then extend the existing privacy notion for DSSE to capture the size pattern leakage, achieving a strong forward and backward privacy definition. Following the definition, we propose a new DSSE scheme Eurus. Eurus can eliminate search/access pattern by relying on a multi-server ORAM scheme, meanwhile reducing size pattern with reasonable efficiency. We show that Eurus can reduce leakage significantly with better efficiency, compared with state-of-the-art leakage reduction schemes.

Order-preserving encryption (OPE) is a cryptographic primitive that preserves the order of plaintexts. In the past few years, many OPE schemes were proposed to solve the problem of executing range queries in encrypted databases. However, OPE leaks some certain information (for example, the order of ciphertext), so it is vulnerable to many attacks. Subsequently, order-revealing encryption (ORE) was proposed by Bonehet al.(Eurocrypt 2015) as a generalization of order-preserving encryption. It breaks through the limitation of the numeric order of OPE plaintext. It implements ciphertext comparison for any specific form of plaintext through a publicly computable comparison function. In this article, we aim to design a new ORE scheme which reduces the leakages and preserves the practicality in terms of ciphertext length and encryption time. We first propose the hybrid model namedHybridORE. Then, we propose an improved scheme namedEncodeOREwhich achieves acceptable security and appropriate ciphertext length. They both explore the encode strategy of encoding plaintext into different parts and apply suitable ORE algorithms to each part according to its security characteristics to reduce leakages. Compared with the typical CLWW scheme (FSE 2016) and Lewi-Wu (CCS 2016) in large domain, they have fewer leakages. The experiment shows that the proposedEncodeOREis very practical.

Homomorphic encryption (HE) schemes, such as fully homomorphic encryption (FHE), support a number of useful computations on ciphertext in a broad range of applications, such as e-voting, private information retrieval, cloud security, and privacy protection. While FHE schemes do not require any interaction during computation, the key limitations are large ciphertext expansion and inefficiency. Thus, to overcome these limitations, we develop a novel cryptographic tool, MAS-Encryption (MASE), to support real-value input and secure computation on the multiply-add structure. The multiply-add structures exist in many important protocols, such as classifiers and outsourced protocols, and we will explain how MASE can be used to protect the privacy of these protocols, using two case study examples. Specifically, the first case study example is the privacy-preserving Naive Bayes classifier that can achieve minimal Bayes risk, and the other example is the privacy-preserving support vector machine. We prove that the constructed classifiers are secure and evaluate their performance using real-world datasets. Experiments show that our proposed MASE scheme and MASE based classifiers are efficient, in the sense that we achieve an optimal tradeoff between computation efficiency and communication interactions. Thus, we avoid the inefficiency of FHE based paradigm.

Networks that constantly transmit information and change structure are becoming increasingly prevalent. However, traditional privacy models are designed to protect static information, such as records in a database or a person’s profile information, which seldom changes. This conflict between static models and dynamic environments is dramatically hindering the effectiveness and efficiency of privacy preservation in today’s dynamic world. Hence, in this paper, we formally define the concept of dynamic privacy, present two novel perspectives, privacy propagation and accumulation, on the way private information can spread through dynamic cyberspace, and develop associated theories and mechanisms for preserving privacy in advanced complex networks, such as social networking sites where data are constantly being released, shared, and exchanged.

Recently, applying deep learning to no-reference image quality assessment (NR-IQA) has received significant attention. Especially in the last five years, an increasing interest has been drawn to the studies of rank learning since it can help mitigate the problem of small IQA datasets. However, on one hand, existing rank learning is not suitable for the authentically distorted images due to the lack of generated rank samples. On the other hand, the output of existing rank loss functions is uncontrollable, resulting in reduced performance. Motivated by these two limitations, we propose a novel rank learning based NR-IQA method, termed controllable list-wise ranking IQA (CLRIQA) in this paper. To be specific, we first present an imaging-heuristic approach, in which the over- and under-exposure is formulated as an inverse of the Weber-Fechner law, and fusion strategy and compression are adopted, to simulate the authentic distortion and generate the rank image samples. These samples are label-free yet associated with quality ranking information. Then we design a controllable list-wise ranking (CLR) loss function by setting an upper and lower bound of rank range and introducing an adaptive margin to tune rank interval. Finally, both the generated rank samples and proposed CLR are used to pre-train a convolutional neural network. Moreover, to obtain a more accurate prediction model, we take advantage of the IQA datasets to fine-tune the pre-trained network further. Various experiments are conducted on the IQA benchmark datasets, and experimental results demonstrate the effectiveness of the proposed CLRIQA method. The source code and network model can be downloaded at the following web address:https://github.com$/$GZHU-DVL$/$CLRIQA.

Subscription-based services, where broadcasting companies make profits by providing subscription programs to users, have become a popular application of wireless broadcast systems. Driving by business interests and data security, companies usually encrypt content of programs with session keys, then only authorized users can access the contents of their subscribed programs, respectively. It is challenging to flexibly distribute and update the session key, we are therefore motivated to propose an identity-based many-to-many subscription scheme (IMS) with efficient key management. The IMS scheme possesses three novel advantages: First, it provides a whole set of subscription flow, from the selection of programs of interest to the generation and updating of keys. Second, it supports program updating and all subscription behaviors. Third, it employs a signature and authentication mechanism to stipulate that only authorized users can access the programs’ ciphertext. The theoretical analysis demonstrates that the proposed scheme is secure against chosen-ciphertext attacks (CCA) under the$ k$-bilinear Diffie-Hellman exponent (BDHE) problem, and has a decent performance in storage and computational overhead. The experimental analysis shows the total time cost of the proposed scheme is reduced more than 75.02 percent compared against the popular peer schemes under the condition of 2500 users.

Data deduplication technique has been widely adopted by commercial cloud storage providers, which is both important and necessary in coping with the explosive growth of data. To further protect the security of users’ sensitive data in the outsourced storage mode, many secure data deduplication schemes have been designed and applied in various scenarios. Among these schemes, secure and efficient re-encryption for encrypted data deduplication attracted the attention of many scholars, and many solutions have been designed to support dynamic ownership management. In this paper, we focus on the re-encryption deduplication storage system and show that the recently designed lightweight rekeying-aware encrypted deduplication scheme (REED) is vulnerable to an attack which we call it stub-reserved attack. Furthermore, we propose a secure data deduplication scheme with efficient re-encryption based on the convergent all-or-nothing transform (CAONT) and randomly sampled bits from the Bloom filter. Due to the intrinsic property of one-way hash function, our scheme can resist the stub-reserved attack and guarantee the data privacy of data owners’ sensitive data. Moreover, instead of re-encrypting the entire package, data owners are only required to re-encrypt a small part of it through the CAONT, thereby effectively reducing the computation overhead of the system. Finally, security analysis and experimental results show that our scheme is secure and efficient in re-encryption.

Ethereum holds multiple billions of U.S. dollars in the form of Ether cryptocurrency and ERC-20 tokens, with millions of deployed smart contracts algorithmically operating these funds. Unsurprisingly, the security of Ethereum smart contracts has been under rigorous scrutiny. In recent years, numerous defense tools have been developed to detect different types of smart contract code vulnerabilities. When opportunities for exploiting code vulnerabilities diminish, the attackers start resorting to social engineering attacks, which aim to influence humans - often the weakest link in the system. The only known class of social engineering attacks in Ethereum are honeypots, which plant hidden traps for attackers attempting to exploit existing vulnerabilities, thereby targeting only a small population of potential victims.

Collaborative filtering (CF) recommender systems have been extensively developed and widely deployed in various social websites, promoting products or services to the users of interest. Meanwhile, work has been attempted at poisoning attacks to CF recommender systems for distorting the recommend results to reap commercial or personal gains stealthily. While existing poisoning attacks have demonstrated their effectiveness with the offline social datasets, they are impractical when applied to the real setting on online social websites. This paper develops a novel and practical poisoning attack solution toward the CF recommender systems without knowing involved specific algorithms nor historical social data information a priori. Instead of directly attacking the unknown recommender systems, our solution performs certain operations on the social websites to collect a set of sampling data for use in constructing a surrogate model for deeply learning the inherent recommendation patterns. This surrogate model can estimate the item proximities, learned by the recommender systems. By attacking the surrogate model, the corresponding solutions (for availability and target attacks) can be directly migrated to attack the original recommender systems. Extensive experiments validate the generated surrogate model's reproductive capability and demonstrate the effectiveness of our attack upon various CF recommender algorithms.

Numerical vector aggregation has numerous applications in privacy-sensitive scenarios, such as distributed gradient estimation in federated learning, and statistical analysis on key-value data. Within the framework of local differential privacy, this work gives tight minimax error bounds of O(d s/(n epsilon^2)), where d is the dimension of the numerical vector and s is the number of non-zero entries. An attainable mechanism is then designed to improve from existing approaches suffering error rate of O(d^2/(n epsilon^2)) or O(d s^2/(n epsilon^2)). To break the error barrier in the local privacy, this work further consider privacy amplification in the shuffle model with anonymous channels, and shows the mechanism satisfies centralized (14 ln(2/delta) (s e^epsilon+2s-1)/(n-1))^0.5, delta)-differential privacy, which is domain independent and thus scales to federated learning of large models. We experimentally validate and compare it with existing approaches, and demonstrate its significant error reduction.

Human motion prediction is an increasingly interesting topic in computer vision and robotics. In this paper, we propose a new end-to-end feedforward network, TrajectoryCNN, to predict future poses. Compared with the most existing methods, we introduce a new trajectory space and focus on modeling motion dynamics of the input sequence with coupled spatio-temporal features, dynamic local-global features, and global temporal co-occurrence features in the new space. Specifically, the coupled spatio-temporal features describe the spatial and temporal structural information hidden in a natural human motion sequence, which can be easily mined using CNN by simultaneously covering the spatial and temporal dimensions of the sequence with the convolutional filters. The dynamic local-global features encode different correlations among joint trajectories of human motion (i.e. strong correlations among joint trajectories of one part and weak correlations among joint trajectories of different parts), which can be captured by stacking multiple residual trajectory blocks and incorporating our skeletal representation. The global temporal co-occurrence features represent different importance of different input poses to mine the motion dynamics for predicting future poses, which can be obtained automatically by learning free parameters for each pose with our TrajectoryCNN. Finally, we predict future poses with the captured motion dynamic features in a non-recursive manner. Extensive experiments show that our method achieves state-of-the-art performance on five benchmarks (e.g. Human3.6M, CMU-Mocap, 3DPW, G3D, and FNTU), which demonstrates the effectiveness of our proposed method. The code is available at https://github.com/lily2lab/TrajectoryCNN.git.

Lattice code has been widely used for data hiding. It can provide security for data hiding by randomly translating its codebook with a secret dither. However, besides the secret dither, there are an infinite amount of points that are near the secret dither and can also be used for perfect decoding in the noiseless scenario. This means that lattice-based data hiding has a serious security hole, named decoding security hole (DSH) in this paper. After a theoretical analysis of DSH, we find that these points form a convex polytope and the centroid of this convex polytope is the secret dither. Based on this finding, a simple yet effective attack method is presented to estimate the secret dither of lattice-based data hiding. Extensive experimental results show that the proposed method significantly outperforms state-of-the-art attack methods, especially when the number of observations is small or the document-to-watermark ratio changes over a wide range.

Searchable symmetric encryption (SSE) has been widely applied in the encrypted database for queries in practice. Although SSE is powerful and feature-rich, it is always plagued by information leaks. Some recent attacks point out that forward privacy which disallows leakage from update operations, now becomes a basic requirement for any newly designed SSE schemes. However, the subsequent search operations can still leak a significant amount of information. To further strengthen security, we extend the definition of forward privacy and propose the notion of “forward search privacy”. Intuitively, it requires search operations over newly added documents do not leak any information about past queries. The enhanced security notion poses new challenges to the design of SSE. We address the challenges by developing the hidden pointer technique (HPT) and propose a new SSE scheme called Khons, which satisfies our security notion (with the original forward privacy notion) and is also efficient. We implemented Khons and our experiment results on large dataset (wikipedia) show that it is more efficient than existing SSE schemes with forward privacy.

In the recent decade, deep learning techniques have been widely adopted for founding artificial Intelligent applications, which led to successes in many data analysis tasks, such as risk assessment, medical predictions, and face recognition. Since the effectiveness of deep learning is directly proportional to the amount of data available, a large-scale collection of massive data is essential. Considering privacy and security concerns often prevent data owners from contributing sensitive data for training, researchers proposed several techniques to provide privacy guarantees of data in machine learning systems that contains multiple parties. However, all these works incurred frequent interactions between data owners during training, such that they came at a high communicational cost for data owners. To this end, in this article, we propose a new server-aid framework called non-interactive privacy-preserving multi-party machine learning (NPMML), which supports secure machine learning tasks without the participation of data owners. The NPMML framework significantly reduces data owners’ communicational overheads in multi-party machine learning. Moreover, we design a concrete construction for multi-layer neural networks based on NPMML. Finally, we evaluate the performance of NPMML by prototype implementation. The experimental result demonstrates that NPMML is communicational-efficient for data owners.

Voice interface has been a dominant User Interface (UI) channel in the popular smart home environment. Although Voice Control System (VCS) brings users conveniences, it is extremely vulnerable to spoofing attacks (e.g., hidden/inaudible command attack) due to its broadcast nature. In this study, to thwart spoofing attacks, we propose WSVA, a device-free voice liveness detection system based on the prevalent wireless signals generated by IoT devices without requiring user to carry any additional sensor or device. The basic insight of WSVA to distinguish the authentic voice command from a spoofed one is checking the consistency between the voice signal and its corresponding mouth motions, which can be captured by wireless signals. To achieve this goal, WSVA builds a theoretical model to describe the correlations among the wireless signal changes, the mouth motions, and the syllables in the voice command. Then, WSVA selects appropriate features from both voice and wireless signals, and calculates the consistency between these two types of signals to determine whether the VCS is suffering from the spoofing attack. To demonstrate the feasibility of WSVA, we conduct a case study on Samsung SmartThings platform and include WSVA as a new application, which is expected to significantly enhance the security of the existing VCS. We evaluate WSVA with various voice commands in different scenarios. Experimental results demonstrate that WSVA achieves the overall 99 percent true accept rate with 1 percent false accept rate with a good scalability and low latency.

Federated learning (FL) enables a large number of clients to collaboratively train a global model through sharing their gradients in each synchronized epoch of local training. However, a centralized server used to aggregate these gradients can be compromised and forge the result in order to violate privacy or launch other attacks, which incurs the need to verify the integrity of aggregation. In this work, we explore how to design communication-efficient and fast verifiable aggregation in FL. We propose VeriFL, a verifiable aggregation protocol, with O(N) (dimension-independent) communication and O(N+ d) computation for verification in each epoch, where N is the number of clients and d is the dimension of gradient vectors. Since d can be large in some real-world FL applications (e.g., 100K), our dimension-independent communication is especially desirable for clients with limited bandwidth and high-dimensional gradients. In addition, the proposed protocol can be used in the FL setting where secure aggregation is needed or there is a subset of clients dropping out of protocol execution. Experimental results indicate that our protocol is efficient in these settings.

Automatic cost learning for steganography based on deep neural networks is receiving increasing attention. Steganographic methods under such a framework have been shown to achieve better security performance than methods adopting hand-crafted costs. However, they still exhibit some limitations that prevent a full exploitation of their potentiality, including using a function-approximated neural-network-based embedding simulator and a coarse-grained optimization objective without explicitly using pixel-wise information. In this article, we propose a new embedding cost learning framework called SPAR-RL (Steganographic Pixel-wise Actions and Rewards with Reinforcement Learning) that overcomes the above limitations. In SPAR-RL, an agent utilizes a policy network which decomposes the embedding process into pixel-wise actions and aims at maximizing the total rewards from a simulated steganalytic environment, while the environment employs an environment network for pixel-wise reward assignment. A sampling process is utilized to emulate the message embedding of an optimal embedding simulator. Through the iterative interactions between the agent and the environment, the policy network learns a secure embedding policy which can be converted into pixel-wise embedding costs for practical message embedding. Experimental results demonstrate that the proposed framework achieves state-of-the-art security performance against various modern steganalyzers, and outperforms existing cost learning frameworks with regard to learning stability and efficiency.

In the feature matching of some Vehicular Ad Hoc Network (VANET) technologies, users' privacy disclosure issue has seriously threatened personal safety and caused considerable economic loss. In this paper, we proposed Oblivious Transfer (OT) protocol and Private Set Intersection (PSI) protocol to protect the users' privacy in the situation of VANET's feature matching. In our schema, an efficient k-out-of- n OT protocol is constructed. Then, this OT protocol is adopted to give a PSI protocol with equality test. Based on the PSI protocol, the two parties of VANET can obtain the intersection of their characteristic sets and any information outside such intersection is unavailable. Accordingly, internal attacker fails to get any useful information from the two parties in the feature matching, and the two parties cannot obtain extra data of each other. Moreover, to reduce the computational cost of the OT protocol, the non-communicative algebraic structure-group ring Zq[Sm] is employed. In addition, we analyzed that the proposed scheme does not use complex calculations and can resist against the current quantum attacks.

The primitive of verifiable database (VDB) can enable a resource-limited client to securely outsource an encrypted database to an untrusted cloud server and the client could efficiently retrieve and update the data at will. Meanwhile, the client can undoubtedly detect any misbehavior by the server if the database has been tampered with. We argue that most of the existing VDB schemes can only support the updating operation of replacement, rather than other common updating operations such asinsertionanddeletion. Recently, the first publicly verifiable VDB schemes that supports all updating operations was proposed based on the idea of hierarchical vector commitment. However, one disadvantage of the proposed VDB scheme is that the computation and storage complexity increases linearly when the client continually inserts data records in the same index of the database. As a result, it remains an open problem how to construct an efficient (and publicly verifiable) VDB scheme that can support all updating operations regardless of the manner of insertion. In this paper, we first introduce a new primitive called committed invertible Bloom filter (CIBF) and utilize it to propose a new publicly verifiable VDB scheme that can support all kinds of updating operations. Additionally, the proposed construction is efficient regardless of the manner of updating operations and thus provides an affirmative answer to the above open problem.

In this article, we propose a(τ, ε)-greedy reinforcement learning algorithm for anti-jamming wireless communications, which chooses previous action with probability τ and applies ε-greedy with probability 1-τ. The key idea of our algorithm is that the more valuable the previous action is, the higher probability of directly performing it at the current time slot without learning. For this purpose, the average utility of several previous actions is first calculated as a threshold for the valuable action judgment. Then, probability τ is formulated as a Gaussian-like function with respect to the difference between the threshold and the utility of the previous action, which makes the wireless devices find the optimal action at a faster speed in the early stage, and eventually ensures the convergence. As a concrete example, the proposed algorithm is implemented in a wireless communication system against multiple jammers. Simulation results show that compared with ε-greedy, the (τ, ε)-greedy obtains faster convergence rate and slightly higher signalto-interference-plus-noise ratio when being applied to Qlearning, deep Q-networks (DQN), double DQN (DDQN), and prioritized experience reply based DDQN (PDDQN). The source code is available at https://github.com/GZHUDVL/tau-epsilon-greedy-RL.

Evasion attack on discrete data is a challenging, while practically interesting research topic. It is intrinsically an NP-hard combinatorial optimization problem. Characterizing the conditions guaranteeing the solvability of an evasion attack task thus becomes the key to understand the adversarial threat. Our study is inspired by the weak submodularity theory. We characterize the attackability of a targeted classifier on discrete data in evasion attack by bridging the attackability measurement and the regularity of the targeted classifier. Based on our attackability analysis, we propose a computationally efficient orthogonal matching pursuit-guided attack method for evasion attack on discrete data. It provides provably computational efficiency and attack performances. Substantial experimental results on real-world datasets validate the proposed attackability conditions and the effectiveness of the proposed attack method.

BACKGROUND: Neoantigen-based personal vaccines and adoptive T cell immunotherapy have shown high efficacy as a cancer treatment in clinical trials. Algorithms for the accurate prediction of neoantigens have played a pivotal role in such studies. Some existing bioinformatics methods, such as MHCflurry and NetMHCpan, identify neoantigens mainly through the prediction of peptide-MHC binding affinity. However, the predictive accuracy of immunogenicity of these methods has been shown to be low. Thus, a ranking algorithm to select highly immunogenic neoantigens of patients is needed urgently in research and clinical practice. RESULTS: We develop TruNeo, an integrated computational pipeline to identify and select highly immunogenic neoantigens based on multiple biological processes. The performance of TruNeo and other algorithms were compared based on data from published literature as well as raw data from a lung cancer patient. Recall rate of immunogenic ones among the top 10-ranked neoantigens were compared based on the published combined data set. Recall rate of TruNeo was 52.63%, which was 2.5 times higher than that predicted by MHCflurry (21.05%), and 2 times higher than NetMHCpan 4 (26.32%). Furthermore, the positive rate of top 10-ranked neoantigens for the lung cancer patient were compared, showing a 50% positive rate identified by TruNeo, which was 2.5 times higher than that predicted by MHCflurry (20%). CONCLUSIONS: TruNeo, which considers multiple biological processes rather than peptide-MHC binding affinity prediction only, provides prioritization of candidate neoantigens with high immunogenicity for neoantigen-targeting personalized immunotherapies.

Text sentiment classification is of critical importance to improve the autonomous decision making and communication ability among object peers in the Social Internet of Things (SIoT). To classify sentiment polarity on a fine-grained level, aspect-level sentiment classification has become a promising direction in recent years. However, the existing solutions typically ignore the mutual information between sentences and their respective aspect terms while generally performing sentiment classification by using the simple attention mechanism. Thus, the relevant results seem to be unpromising. We aim to develop a novel neural-network-based model, by relying on the natural language processing model for rich feature extraction, called mutual attention neural networks (MANs), to conduct the aspect-level sentiment classification tasks in this article. Compared with the previous work, our proposed MAN model takes advantage of the bidirectional long short-term memory (Bi-LSTM) networks to obtain semantic dependence of sentences and their respective aspect terms, while learning the sentiment polarities of aspect terms in sentences by proposing the mutual attention mechanism. To evaluate the performance of MAN, we conduct our experiments on three real-world data sets, i.e., LAPTOP, REST, and TWITTER. The experimental results show that our proposed MAN model has significant performance improvements when compared to several existing models, in terms of aspect-level sentiment classification.

Encryption alone is not enough to protect data privacy, because access pattern leaks some sensitive information. Oblivious RAM (ORAM), the solution to this problem, is still far from practical deployment for heavy storage and communication/computation overhead. To reduce them, an insightful idea was proposed to utilize non-colluding clouds to shift client computation and client-cloud communication to the clouds. The proposed multi-cloud ORAM achieved O(1) client-cloud bandwidth cost and removed most of client computation. In this paper, we exploit “disconnected ORAMoperation” and design “two-layerencryption” to further reduce these overheads. Experiments show that our proposed scheme, NewMCOS, significantly reduces evict cache size from GB/MB to KB level with about 2-3 times lower response time and 20 percent savings in bandwidth for clouds, compared to other schemes. Theoretically speaking, we reduce evict cache size from O(√N) to O(ZK), where N is the number of real data blocks, K is the number of clouds (2 <; K <; <; √N ), and Z is the number of real blocks uploaded from the client for eviction. By employing “lazy eviction operation”, the write frequency is reduced by O(Z), the shuffling bandwidth cost is reduced by Ω(Z log Z). Meanwhile, NewMCOS is proved to be secure.

Oblivious RAM (ORAM) is important for applications that require hiding of access patterns. However, most of existing implementations of ORAM are very expensive, which are infeasible to be deployed in lightweight devices, like the terminal equipment for Internet of Things (IoT). In this article, we focus on how to apply the expensive ORAM to protect access pattern in IoT devices and propose an ORAM scheme supporting thin-client, called “ThinORAM”, under non-colluding clouds. Our proposed scheme removes complicated computations in the client side and requires only O(1) communication cost with a reasonable response time. We further show how to securely deploy ThinORAM in the fog computing environment to achieve oblivious data access with minimum client cost. Experiments show that our scheme can eliminate most of the client storage and reduce the cloud-cloud bandwidth by 2×, with 3× faster response time, when compared to the best scheme that aims at reducing client side overheads.

Summary Based on cloud servers' powerful storage and computing resources, users can store mass encrypted data in the cloud and outsource complex encryption computations to the cloud servers. Since cloud servers cannot be completely trusted, then data privacy and integrity are concerned about hot issues. We focus on the following problems in outsourced encryptions: how to protect the data privacy and how to check the integrity of data and the correctness of cloud server's outsourcing computations. In this paper, we at first propose a verifiable chaotic encryption based on Chebyshev polynomials. The scheme supports verifiable function for data integrity. To further improve the efficiency of the scheme, a corresponding outsourced encryption scheme is constructed, where the heavy overhead evaluations of Chebyshev polynomials are transferred from the user side to the cloud server. The outsourced encryption also provides the checkability for data integrity and correctness of cloud computations. The scheme is suitable for mobile users with limited computing resources. Moreover, the newly proposed scheme no longer depends upon the simple heuristic analysis. It achieves the indistinguishability under chosen‐ciphertext attacks (IND‐CCA) in the standard model based on the Chebyshev‐based Decisional Diffie‐Hellman (CDDH) assumption. Thus, we answer a long‐term open problem for building a chaotic encryption scheme with provable security in the sense of the IND‐CCA.

Efficient medical image sampling and transferring becomes one of key research areas in computer science and healthcare application industries. In particular, the technique of body area networking and personal area networking are very useful in various image-based medical monitoring systems that cover a wide range of healthcare services, such as early detecting of emergency conditions and remote online instructing of surgeries. However, medical images are highly privacy sensitive and redundant. Thus, proper protection on privacy and secure data aggregation/compression are also highly expected in medical image processing. Based on compressive sensing theory, we conceive a so-called “one-stone-three-bird” solution for medical image acquisition and transmission in this paper. The size of the original medical images can be reduced to 20%, the resulted images have very well confidentiality and supporting additively homomorphic aggregation.

As an important application of the Internet of Things technologies, mobile healthcare crowd sensing (MHCS) still has challenging issues, such as privacy protection and efficiency. Quite recently in the IEEE Internet of Things Journal (DOI: 10.1109/JIOT.2018.2828463), Liuet al.proposed a large-scale concurrent data anonymous batch verification scheme for MHCS, claiming to provide batch authentication, nonrepudiation, and anonymity. However, after a close look at the scheme, we point out that the scheme suffers two types of signature forgery attacks and hence fails to achieve the claimed security properties. In addition, a reasonable and rigorous probability analysis indicates that the security reduction from the security of the scheme to the hardness of the computational Diffie–Hellman problem is invalid. We hope that similar design flaws can be avoided in future design of anonymous batch verification schemes for MHCS.

In this paper, we study decode-and-forward relaying networks in the presence of direct links, where they are used by the eavesdropper to overhear the confidential message from the source and relay. The secure data transmission can go through from either the direct or the relaying branch, and we focus on the practical communication scenarios, where the main and eavesdropper channels are correlated. Although traditional opportunistic selection techniques can choose one better branch to ensure the secure performance, it needs to continuously know the channel state information (CSI) of both branches and may result in a high branch switching rate. To overcome these limitations, we propose a distributed secure switch-and-stay combining (DSSSC) protocol, where only one between direct and relaying branches is activated to assist the secure data transmission, and the switching occurs when the branch cannot support the secure communication any longer. The DSSSC protocol uses either the instantaneous or the statistics of the eavesdropping CSI. For both cases, we quantify the impact of correlated fading on secure communication by deriving an analytical expression for the secrecy outage probability (SOP) as well as an asymptotic expression for the high main-to-eavesdropper ratio region. From the asymptotic SOP, we can conclude that the DSSSC can achieve the optimal secure performance of opportunistic selection with less implementation complexity, and the channel correlation can further enhance the transmission security.

With the digitalization of various industries, the combination of cloud computing and the industrial Internet of Things (IIoT) has become an attractive data processing paradigm. However, the cloud-assisted IIoT still has challenging issues, including authenticity of data, untrustworthiness of third parties, and system robustness and efficiency. Recently, a lightweight certificateless signature (CLS) scheme for the cloud-assisted IIoT, that was claimed to address both authenticity of data and untrustworthiness of third parties, has been proposed by Karati et al. (2018). In this paper, we demonstrate that the CLS scheme fails to achieve the claimed security properties by presenting four types of signature forgery attacks. We also propose a robust certificateless signature (RCLS) scheme to address the aforementioned challenges. Our RCLS only needs public channels and is proven secure against both public key replacement attacks and malicious-but-passive third parties in the standard model. Performance evaluation indicates that the RCLS scheme outperforms other CLS schemes and is suitable for the IIoT.

The rich cloud services in the Internet of Things create certain needs for edge computing, in which devices should be able to handle storage tasks securely, reliably, and efficiently. When processing the storage requests from edge devices, each cloud server is supposed to eliminate duplicate copies of repeating data to reduce the amount of storage space and save on bandwidth. To protect data confidentiality while supporting deduplication, some convergent-encryption-based techniques have been proposed to encrypt the data before uploading. However, all these works cannot meet two requirements while preventing brute-force attacks: (i) power-constrained edge nodes should update encryption keys efficiently when an edge node is abandoned; and (ii) the access privacy of edge nodes should be guaranteed. In this article, we propose a novel encryption scheme for secure chunk-level deduplication. Based on this scheme, we present two constructions of the secure deduplication system that support an efficient key update protocol. The key update protocol does not involve any edge node in computational tasks, so that the deduplication system can adopt an active key update strategy. Moreover, one of our constructions, which is called advance construction, can provide access privacy assurances for edge nodes. The security analysis is given in terms of the proposed threat model. The experimental analysis demonstrates that the proposed deduplication system is practical.

In recent years, the rapid development of Internet of Things (IoT) and big data shows the huge demand of outsourcing computing in cloud computing to assist some clients with low ability devices to fulfill massive data processing. In the mean time, considering the uncertainty of the Internet devices, we should persist security and high-efficiency. Uploaded data and returned results should be protected from attack of adversary, which guarantees the security, while efficiency requires low overhead of clients to finish the whole calculation procedure. In this paper, we main focus on the public verifiable outsourcing scheme on matrix multiplication, which can be applied in many IoT scenes, such as path planning and aggregation operation in Internet of Vehicles and smart grids, respectively. Specially, this paper presents schemes for two different functions of matrix multiplication, which strengthens the applicability. Moreover, security analysis and performance evaluation in this paper properly present the superiority of this paper.

With the advent of artificial intelligence, machine learning has been well explored and extensively applied into numerous fields, such as pattern recognition, image processing and cloud computing. Very recently, machine learning hosted in a cloud service has gained more attentions due to the benefits from the outsourcing paradigm. Based on cloud-aided computation techniques, the heavy computation tasks involved in machine learning process can be off-loaded into the cloud server in a pay-per-use manner, whereas outsourcing large-scale collection of sensitive data risks privacy leakage since the cloud server is semi-honest. Therefore, privacy preservation for the client and verification for the returned results become two challenges to be dealt with. In this paper, we focus on designing a novel privacy-preserving single-layer perceptron training scheme which supports batch patterns training and verification for the training results on the client side. In addition, adopting classical secure two-party computation method, we design a novel lightweight privacy-preserving predictive algorithm. Both two participants learns nothing about other’s inputs, and the calculation result is only known by one party. Detailed security analysis shows that the proposed scheme can achieve the desired security properties. We also demonstrate the efficiency of our scheme by providing the experimental evaluation on two different real datasets.

Group data sharing in cloud environments has become a hot topic in recent decades. With the popularity of cloud computing, how to achieve secure and efficient data sharing in cloud environments is an urgent problem to be solved. In addition, how to achieve both anonymity and traceability is also a challenge in the cloud for data sharing. This paper focuses on enabling data sharing and storage for the same group in the cloud with high security and efficiency in an anonymous manner. By leveraging the key agreement and the group signature, a novel traceable group data sharing scheme is proposed to support anonymous multiple users in public clouds. On the one hand, group members can communicate anonymously with respect to the group signature, and the real identities of members can be traced if necessary. On the other hand, a common conference key is derived based on the key agreement to enable group members to share and store their data securely. Note that a symmetric balanced incomplete block design is utilized for key generation, which substantially reduces the burden on members to derive a common conference key. Both theoretical and experimental analyses demonstrate that the proposed scheme is secure and efficient for group data sharing in cloud computing.

The alarming growth rate of malicious apps has become a serious issue that sets back the prosperous mobile ecosystem. A recent report indicates that a new malicious app for Android is introduced every 10 s. To combat this serious malware campaign, we need a scalable malware detection approach that can effectively and efficiently identify malware apps. Numerous malware detection tools have been developed, including system-level and network-level approaches. However, scaling the detection for a large bundle of apps remains a challenging task. In this paper, we introduce Significant Permission IDentification (SigPID), a malware detection system based on permission usage analysis to cope with the rapid increase in the number of Android malware. Instead of extracting and analyzing all Android permissions, we develop three levels of pruning by mining the permission data to identify the most significant permissions that can be effective in distinguishing between benign and malicious apps. SigPID then utilizes machine-learning-based classification methods to classify different families of malware and benign apps. Our evaluation finds that only 22 permissions are significant. We then compare the performance of our approach, using only 22 permissions, against a baseline approach that analyzes all permissions. The results indicate that when a support vector machine is used as the classifier, we can achieve over 90% of precision, recall, accuracy, and F-measure, which are about the same as those produced by the baseline approach while incurring the analysis times that are 4-32 times less than those of using all permissions. Compared against other state-of-the-art approaches, SigPID is more effective by detecting 93.62% of malware in the dataset and 91.4% unknown/new malware samples.

During the last few years, the speaker recognition technique has been widely attractive for its extensive application in many fields, such as speech communications, domestics services, and smart terminals. As a critical method, the Gaussian mixture model (GMM) makes it possible to achieve the recognition capability that is close to the hearing ability of human in a long speech. However, the GMM is failing to recognize a short utterance speaker with a high accuracy. Aiming at solving this problem, in this paper, we propose a novel model to enhance the recognition accuracy of the short utterance speaker recognition system. Different from traditional models based on the GMM, we design a method to train a convolutional neural network to process spectrograms, which can describe speakers better. Thus, the recognition system gains the considerable accuracy as well as the reasonable convergence speed. The experiment results show that our model can help to decrease the equal error rate of the recognition from 4.9% to 2.5%.

Vehicular Ad Hoc Network (VANET) allows vehicles to exchange information about road and traffic conditions through wireless communications. Nevertheless, providing reliable and authenticated information without violating the user's privacy seems contradictory. In this paper, we propose an Anonymous Announcement System especially designed for Electric Vehicle (EV) in VANETs to achieve the aforementioned contradictory goals. We demonstrated the feasibility of the protocol with a prototype implementation on a suitable device and a network simulation with our protocol added on top of a normal VANET.

Cloud computing is growing exponentially, whereby there are now hundreds of cloud service providers (CSPs) of various sizes. While the cloud consumers may enjoy cheaper data storage and computation offered in this multi-cloud environment, they are also in face of more complicated reliability issues and privacy preservation problems of their outsourced data. Though searchable encryption allows users to encrypt their stored data while preserving some search capabilities, few efforts have sought to consider the reliability of the searchable encrypted data outsourced to the clouds. In this paper, we propose a privacy-preserving STorage and REtrieval (STRE) mechanism that not only ensures security and privacy but also provides reliability guarantees for the outsourced searchable encrypted data. The STRE mechanism enables the cloud users to distribute and search their encrypted data across multiple independent clouds managed by different CSPs, and is robust even when a certain number of CSPs crash. Besides the reliability, STRE also offers the benefit of partially hidden search pattern. We evaluate the STRE mechanism on Amazon EC2 using a real world dataset and the results demonstrate both effectiveness and efficiency of our approach.

There is an increasing trend for data owners to store their data in a third-party cloud server and buy the service from the cloud server to provide information to other users. To ensure confidentiality, the data is usually encrypted. Therefore, an encrypted data searching scheme with privacy preserving is of paramount importance. Predicate encryption (PE) is one of the attractive solutions due to its attribute-hiding merit. However, as cloud is not always trusted, verifying the searched results is also crucial. Firstly, a generic construction of Publicly Verifiable Predicate Encryption (PVPE) scheme is proposed to provide verification for PE. We reduce the security of PVPE to the security of PE. However, from practical point of view, to decrease the communication overhead and computation overhead, an improved PVPE is proposed with the trade-off of a small probability of error.

Rekeying refers to an operation of replacing an existing key with a new key for encryption. It renews security protection, so as to protect against key compromise and enable dynamic access control in cryptographic storage. However, it is non-trivial to realize efficient rekeying in encrypted deduplication storage systems, which use deterministic content-derived encryption keys to allow deduplication on ciphertexts. We design and implement REED, a rekeying-aware encrypted deduplication storage system. REED builds on a deterministic version of all-or-nothing transform (AONT), such that it enables secure and lightweight rekeying, while preserving the deduplication capability. We propose two REED encryption schemes that trade between performance and security, and extend REED for dynamic access control. We implement a REED prototype with various performance optimization techniques. Our trace-driven testbed evaluation shows that our REED prototype maintains high performance and storage efficiency.

We propose a generic construction that yields efficient identity-based encryption (IBE) schemes secure against chosen ciphertext attack (CCA) in the standard model. Our construction extends Kiltz et al.'s (Eurocrypt '09) method of constructing CCA-secure public-key encryption schemes via randomness extraction to the identity-based setting. The main idea of our construction is to transform ‘|$\epsilon _1$|-almost |$\kappa $|-entropic’ and valid/invalid ciphertext indistinguishable (VI-IND) identity-based hash proof system to the one that satisfies the stronger ‘|$\epsilon _2$|-universal’ and VI-IND property. This transformation is realized by a randomness extractor based on the 4-wise hash function. We demonstrate that our generic construction can produce CCA-secure IBE schemes whose efficiency is comparable with the most efficient but non-generic CCA-secure IBE schemes without random oracles in the literature.

Abstract As a public key cryptographic primitive, attribute‐based encryption (ABE) is promising in implementing fine‐grained access control in cloud computing. However, before ABE comes into practical applications, two challenging issues have to be addressed, that is, users' attribute privacy protection and access policy update. In this paper, we tackle the aforementioned challenge for the first time by formalizing the notion of anonymous ciphertext‐policy attribute‐based proxy re‐encryption (anonymous CP‐ABPRE) and giving out a concrete construction. We propose a novel technique called match‐then‐re‐encrypt, in which a matching phase is additionally introduced before the re‐encryption phase. This technique uses special components of the proxy re‐encryption key and ciphertext to anonymously check whether the proxy can fulfill a proxy re‐encryption or not. Theoretical analysis and simulation results demonstrate that our anonymous CP‐ABPRE scheme is secure and efficient. Copyright © 2016 John Wiley & Sons, Ltd.

In order to realize attribute-based data sharing in cloud computing, multi-authority attribute-based encryption (MA-ABE) is extremely attractive. However, most of the existing MA-ABE schemes cannot support a fully large attribute universe and are not suitable for resource-constrained mobile data owners in that the computation cost in secret key generation and encryption is extremely heavy. To tackle the earlier challenges, we propose an online/offline MA-ABE scheme, which realizes both the online/offline secret key generation and the online/offline encryption while supporting a fully large attribute universe. In the offline phase, one global-identity authority and multiple attribute authorities do the majority of the work to issue attribute secret keys before knowing users' global identity and attributes. The data owner can perform most of the encryption computation tasks before knowing the actual message and access structure. Furthermore, the online phase can rapidly assemble the final decryption key and ciphertexts when related specifications become known. Particularly, global-identity authority and attribute authorities need not to cooperate in the whole process. Our online/offline MA-ABE scheme allows the access policies encoded in linear secret sharing schemes. The formal selective security proof and extensive performance analysis indicate that our scheme is very suitable for data sharing in mobile cloud computing. Copyright © 2016 John Wiley & Sons, Ltd.

The notion of verifiable database (VDB) enables a resource-constrained client to securely outsource a very large database to an untrusted server so that it could later retrieve a database record and update a record by assigning a new value. Also, any attempt by the server to tamper with the data will be detected by the client. When the database undergoes frequent while small modifications, the client must re-compute and update the encrypted version (ciphertext) on the server at all times. For very large data, it is extremely expensive for the resources-constrained client to perform both operations from scratch. In this paper, we formalize the notion of verifiable database with incremental updates (Inc-VDB). Besides, we propose a general Inc-VDB framework by incorporating the primitive of vector commitment and the encrypt-then-incremental MAC mode of encryption. We also present a concrete Inc-VDB scheme based on the computational Diffie-Hellman (CDH) assumption. Furthermore, we prove that our construction can achieve the desired security properties.

As the cloud computing technology develops during the last decade, outsourcing data to cloud service for storage becomes an attractive trend, which benefits in sparing efforts on heavy data maintenance and management. Nevertheless, since the outsourced cloud storage is not fully trustworthy, it raises security concerns on how to realize data deduplication in cloud while achieving integrity auditing. In this work, we study the problem of integrity auditing and secure deduplication on cloud data. Specifically, aiming at achieving both data integrity and deduplication in cloud, we propose two secure systems, namely SecCloud and SecCloud$^+$. SecCloud introduces an auditing entity with a maintenance of a MapReduce cloud, which helps clients generate data tags before uploading as well as audit the integrity of data having been stored in cloud. Compared with previous work, the computation by user in SecCloud is greatly reduced during the file uploading and auditing phases. SecCloud$^+$is designed motivated by the fact that customers always want to encrypt their data before uploading, and enables integrity auditing and secure deduplication on encrypted data.

Access control ensures that only the authorized users of a system are allowed to access certain resources or tasks. Usually, according to their roles and responsibilities, users are organized in hierarchies formed by a certain number of disjoint classes. Such hierarchies are implemented by assigning a key to each class, so that the keys for descendant classes can be efficiently derived from classes higher in the hierarchy. However, pure hierarchical access may represent a limitation in many real-world cases. In fact, sometimes it is necessary to ensure access to a resource or task by considering both its directly responsible user and a group of users possessing certain credentials. In this paper, we first propose a novel model that generalizes the conventional hierarchical access control paradigm, by extending it to certain additional sets of qualified users. Afterward, we propose two constructions for hierarchical key assignment schemes in this new model, which are provably secure with respect to key indistinguishability. In particular, the former construction relies on both symmetric encryption and perfect secret sharing, whereas, the latter is based on public-key threshold broadcast encryption.

In this paper, we introduce a new fine-grained two-factor authentication (2FA) access control system for web-based cloud computing services. Specifically, in our proposed 2FA access control system, an attribute-based access control mechanism is implemented with the necessity of both a user secret key and a lightweight security device. As a user cannot access the system if they do not hold both, the mechanism can enhance the security of the system, especially in those scenarios where many users share the same computer for web-based cloud services. In addition, attribute-based control in the system also enables the cloud server to restrict the access to those users with the same set of attributes while preserving user privacy, i.e., the cloud server only knows that the user fulfills the required predicate, but has no idea on the exact identity of the user. Finally, we also carry out a simulation to demonstrate the practicability of our proposed 2FA system.

Complex network theory has attracted considerable attentions over the past few yeas. Mutual properties and evolution process of many natural and artificial networks, can be investigated by this theory. Early work have modeled un-weighted evolution, topological structure and self-organizing mechanism of wireless sensor networks. In this paper, we explore weighted evolution dynamics and establish a Strength-Constrained Weighted (SCW) evolving model. By introducing the node's strength, this model not only maintains the dynamic growth of weight, but also confines the strength of the nodes. It is clearly observed that strength and connectivity are limited in simulation experiments, which can play a role on protecting the nodes, in case the strength and connectivity are so high that the energy is consumed fast. Interesting findings are helpful to explore the optimal route pathway for the information transmission and prolong the lifetime of wireless sensor networks.

As cloud computing becomes prevalent, more and more sensitive data is being centralized into the cloud for sharing, which brings forth new challenges for outsourced data security and privacy. Attribute-based encryption (ABE) is a promising cryptographic primitive, which has been widely applied to design fine-grained access control system recently. However, ABE is criticized for its high scheme overhead as the computational cost grows with the complexity of the access formula. This disadvantage becomes more serious for mobile devices with constrained computing resources. Aiming at tackling the challenge above, we present a generic and efficient solution to implement attribute-based access control system by introducing secure outsourcing techniques into ABE. More precisely, two cloud service providers (CSPs), namely key generation-cloud service provider (KG-CSP) and decryption-cloud service provider (D-CSP) are introduced to perform the outsourced key-issuing and decryption on behalf of attribute authority and users respectively. In order to outsource heavy computation to both CSPs without private information leakage, we formalize an underlying primitive called outsourced ABE (OABE) and propose several constructions with outsourced decryption and key-issuing. Finally, extensive experiment demonstrates that with the help of KG-CSP and D-CSP, efficient key-issuing and decryption are achieved in our constructions.

Data deduplication is a technique for eliminating duplicate copies of data, and has been widely used in cloud storage to reduce storage space and upload bandwidth. However, there is only one copy for each file stored in cloud even if such a file is owned by a huge number of users. As a result, deduplication system improves storage utilization while reducing reliability. Furthermore, the challenge of privacy for sensitive data also arises when they are outsourced by users to cloud. Aiming to address the above security challenges, this paper makes the first attempt to formalize the notion of distributed reliable deduplication system. We propose new distributed deduplication systems with higher reliability in which the data chunks are distributed across multiple cloud servers. The security requirements of data confidentiality and tag consistency are also achieved by introducing a deterministic secret sharing scheme in distributed storage systems, instead of using convergent encryption as in previous deduplication systems. Security analysis demonstrates that our deduplication systems are secure in terms of the definitions specified in the proposed security model. As a proof of concept, we implement the proposed systems and demonstrate that the incurred overhead is very limited in realistic environments.

Identity-Based Encryption (IBE) which simplifies the public key and certificate management at Public Key Infrastructure (PKI) is an important alternative to public key encryption. However, one of the main efficiency drawbacks of IBE is the overhead computation at Private Key Generator (PKG) during user revocation. Efficient revocation has been well studied in traditional PKI setting, but the cumbersome management of certificates is precisely the burden that IBE strives to alleviate. In this paper, aiming at tackling the critical issue of identity revocation, we introduce outsourcing computation into IBE for the first time and propose a revocable IBE scheme in the server-aided setting. Our scheme offloads most of the key generation related operations during key-issuing and key-update processes to a Key Update Cloud Service Provider, leaving only a constant number of simple operations for PKG and users to perform locally. This goal is achieved by utilizing a novel collusion-resistant technique: we employ a hybrid private key for each user, in which an AND gate is involved to connect and bound the identity component and the time component. Furthermore, we propose another construction which is provable secure under the recently formulized Refereed Delegation of Computation model. Finally, we provide extensive experimental results to demonstrate the efficiency of our proposed construction.

Cloud computing moves the application software and databases to the centralized large data centers, where the management of the data and services may not be fully trustworthy. In this work, we study the problem of ensuring the integrity of data storage in cloud computing. To reduce the computational cost at user side during the integrity verification of their data, the notion of public verifiability has been proposed. However, the challenge is that the computational burden is too huge for the users with resource-constrained devices to compute the public authentication tags of file blocks. To tackle the challenge, we propose OPoR, a new cloud storage scheme involving a cloud storage server and a cloud audit server, where the latter is assumed to be semi-honest. In particular, we consider the task of allowing the cloud audit server, on behalf of the cloud users, to pre-process the data before uploading to the cloud storage server and later verifying the data integrity. OPoR outsources and offloads the heavy computation of the tag generation to the cloud audit server and eliminates the involvement of user in the auditing and in the pre-processing phases. Furthermore, we strengthen the proof of retrievability (PoR) model to support dynamic data operations, as well as ensure security against reset attacks launched by the cloud storage server in the upload phase.

The notion of verifiable database (VDB) enables a resource-constrained client to securely outsource a very large database to an untrusted server so that it could later retrieve a database record and update it by assigning a new value. Also, any attempt by the server to tamper with the data will be detected by the client. Very recently, Catalano and Fiore [17] proposed an elegant framework to build efficient VDB that supports public verifiability from a new primitive named vector commitment. In this paper, we point out Catalano-Fiore's VDB framework from vector commitment is vulnerable to the so-called forward automatic update (FAU) attack. Besides, we propose a new VDB framework from vector commitment based on the idea of commitment binding. The construction is not only public verifiable but also secure under the FAU attack. Furthermore, we prove that our construction can achieve the desired security properties.

With the rapid development in availability of cloud services, the techniques for securely outsourcing the prohibitively expensive computations to untrusted servers are getting more and more attentions in the scientific community. In this paper, we investigate secure outsourcing for large-scale systems of linear equations, which are the most popular problems in various engineering disciplines. For the first time, we utilize the sparse matrix to propose a new secure outsourcing algorithm of large-scale linear equations in the fully malicious model. Compared with the state-of-the-art algorithm, the proposed algorithm only requires (optimal) one round communication (while the algorithm requires $L$ rounds of interactions between the client and cloud server, where $L$ denotes the number of iteration in iterative methods). Furthermore, the client in our algorithm can detect the misbehavior of cloud server with the (optimal) probability 1. Therefore, our proposed algorithm is superior in both efficiency and checkability. We also provide the experimental evaluation that demonstrates the efficiency and effectiveness of our algorithm.

Data deduplication is one of important data compression techniques for eliminating duplicate copies of repeating data, and has been widely used in cloud storage to reduce the amount of storage space and save bandwidth. To protect the confidentiality of sensitive data while supporting deduplication, the convergent encryption technique has been proposed to encrypt the data before outsourcing. To better protect data security, this paper makes the first attempt to formally address the problem of authorized data deduplication. Different from traditional deduplication systems, the differential privileges of users are further considered in duplicate check besides the data itself. We also present several new deduplication constructions supporting authorized duplicate check in a hybrid cloud architecture. Security analysis demonstrates that our scheme is secure in terms of the definitions specified in the proposed security model. As a proof of concept, we implement a prototype of our proposed authorized duplicate check scheme and conduct testbed experiments using our prototype. We show that our proposed authorized duplicate check scheme incurs minimal overhead compared to normal operations.

Data deduplication is an attractive technology to reduce storage space and upload bandwidth for increasing vast amount of duplicated and redundant data. In a cloud storage system with data deduplication, duplicate copies of data will be eliminated and only one copy will be kept in the storage. To protect the confidentiality of sensitive data while supporting deduplication, the convergent encryption technique has been proposed to encrypt the data before outsourcing. However, the issue of keyword search over encrypted data in deduplication storage system has to be addressed for efficient data utilization. This paper firstly proposes two constructions which support secure keyword search in this scenario. In these constructions, the integrity of the data can be realized by just checking the convergent key, without other traditional integrity auditing mechanisms. Security analysis demonstrates that our keyword search schemes are secure in terms of the definitions specified in the proposed security model.

Linear programming (LP) has been well studied in the scientific community for various engineering applications such as network flow problems, packet routing, portfolio optimization, and financial data management, etc. In this paper, we first utilize the sparse matrix to investigate secure outsourcing for large-scale LP systems, which is considered as a prohibitively expensive computation for the clients with resource-constraint devices. Besides, we propose a secure and practical scheme which is suitable for any LP problem (feasible, infeasible or unbounded) even in the fully malicious model. Compared with the state-of-the-art algorithm [30], our proposed algorithm only requires O(n2) computational overhead instead of O(nρ) for 2 <; ρ ≤ 3. Furthermore, the client C can detect the misbehavior of cloud server S with the (optimal) probability 1 under the computational complexity of O(n).

SUMMARY With the rapid development of cloud computing, more and more data are being centralized into remote cloud server for sharing, which raises a challenge on how to keep them both private and accessible. Although searchable encryption provides an efficient solution to support keyword‐based search directly on encrypted data, considering its application in file sharing, existing work depends on key sharing among authorized users, which inevitably causes the risks of key exposure and abuse. In this paper, aiming at enabling efficient and secure data sharing in cloud computing, we provide a generic construction for this purpose. The proposed construction is full‐featured: (i) It enables authorized users to perform keyword‐based search directly on encrypted data without sharing the unique secret key; and (ii) it provides two‐layered access control to limit unauthorized user's access to the shared data. On the basis of the proposed generic construction, we utilize the existing techniques on identity‐based broadcast encryption and public key searchable encryption to instantiate a concrete construction. Copyright © 2013 John Wiley & Sons, Ltd.

With the rapid development of cloud services, the techniques for securely outsourcing the prohibitively expensive computations to untrusted servers are getting more and more attention in the scientific community. Exponentiations modulo a large prime have been considered the most expensive operations in discrete-logarithm-based cryptographic protocols, and they may be burdensome for the resource-limited devices such as RFID tags or smartcards. Therefore, it is important to present an efficient method to securely outsource such operations to (untrusted) cloud servers. In this paper, we propose a new secure outsourcing algorithm for (variable-exponent, variable-base) exponentiation modulo a prime in the two untrusted program model. Compared with the state-of-the-art algorithm, the proposed algorithm is superior in both efficiency and checkability. Based on this algorithm, we show how to achieve outsource-secure Cramer-Shoup encryptions and Schnorr signatures. We then propose the first efficient outsource-secure algorithm for simultaneous modular exponentiations. Finally, we provide the experimental evaluation that demonstrates the efficiency and effectiveness of the proposed outsourcing algorithms and schemes.

Data deduplication is a technique for eliminating duplicate copies of data, and has been widely used in cloud storage to reduce storage space and upload bandwidth. Promising as it is, an arising challenge is to perform secure deduplication in cloud storage. Although convergent encryption has been extensively adopted for secure deduplication, a critical issue of making convergent encryption practical is to efficiently and reliably manage a huge number of convergent keys. This paper makes the first attempt to formally address the problem of achieving efficient and reliable key management in secure deduplication. We first introduce a baseline approach in which each user holds an independent master key for encrypting the convergent keys and outsourcing them to the cloud. However, such a baseline key management scheme generates an enormous number of keys with the increasing number of users and requires users to dedicatedly protect the master keys. To this end, we propose Dekey , a new construction in which users do not need to manage any keys on their own but instead securely distribute the convergent key shares across multiple servers. Security analysis demonstrates that Dekey is secure in terms of the definitions specified in the proposed security model. As a proof of concept, we implement Dekey using the Ramp secret sharing scheme and demonstrate that Dekey incurs limited overhead in realistic environments.

Attribute-based signature (ABS) enables users to sign messages over attributes without revealing any information other than the fact that they have attested to the messages. However, heavy computational cost is required during signing in existing work of ABS, which grows linearly with the size of the predicate formula. As a result, this presents a significant challenge for resource-constrained devices (such as mobile devices or RFID tags) to perform such heavy computations independently. Aiming at tackling the challenge above, we first propose and formalize a new paradigm called Outsourced ABS, i.e., OABS, in which the computational overhead at user side is greatly reduced through outsourcing intensive computations to an untrusted signing-cloud service provider (S-CSP). Furthermore, we apply this novel paradigm to existing ABS schemes to reduce the complexity. As a result, we present two concrete OABS schemes: i) in the first OABS scheme, the number of exponentiations involving in signing is reduced from O(d) to O(1) (nearly three), where d is the upper bound of threshold value defined in the predicate; ii) our second scheme is built on Herranz et al.'s construction with constant-size signatures. The number of exponentiations in signing is reduced from O(d2) to O(d) and the communication overhead is O(1). Security analysis demonstrates that both OABS schemes are secure in terms of the unforgeability and attribute-signer privacy definitions specified in the proposed security model. Finally, to allow for high efficiency and flexibility, we discuss extensions of OABS and show how to achieve accountability as well.

This paper initiates the study of two specific security threats on smart-card-based password authentication in distributed systems. Smart-card-based password authentication is one of the most commonly used security mechanisms to determine the identity of a remote client, who must hold a valid smart card and the corresponding password to carry out a successful authentication with the server. The authentication is usually integrated with a key establishment protocol and yields smart-card-based password-authenticated key agreement. Using two recently proposed protocols as case studies, we demonstrate two new types of adversaries with smart card: 1) adversaries with pre-computed data stored in the smart card, and 2) adversaries with different data (with respect to different time slots) stored in the smart card. These threats, though realistic in distributed systems, have never been studied in the literature. In addition to point out the vulnerabilities, we propose the countermeasures to thwart the security threats and secure the protocols.

Attribute-Based Encryption (ABE) is a promising cryptographic primitive which significantly enhances the versatility of access control mechanisms. Due to the high expressiveness of ABE policies, the computational complexities of ABE key-issuing and decryption are getting prohibitively high. Despite that the existing Outsourced ABE solutions are able to offload some intensive computing tasks to a third party, the verifiability of results returned from the third party has yet to be addressed. Aiming at tackling the challenge above, we propose a new Secure Outsourced ABE system, which supports both secure outsourced key-issuing and decryption. Our new method offloads all access policy and attribute related operations in the key-issuing process or decryption to a Key Generation Service Provider (KGSP) and a Decryption Service Provider (DSP), respectively, leaving only a constant number of simple operations for the attribute authority and eligible users to perform locally. In addition, for the first time, we propose an outsourced ABE construction which provides checkability of the outsourced computation results in an efficient way. Extensive security and performance analysis show that the proposed schemes are proven secure and practical.

Attribute-based encryption (ABE) has been widely studied recently to support fine-grained access control of shared data. Anonymous ABE, which is a relevant notion to ABE, further hides the receivers' attribute information in ciphertexts because many attributes are sensitive and related to the identity of eligible users. However, in existing anonymous ABE work, a user knows whether the attributes and the policy match or not only after repeating decryption attempts. And, the computation overhead of each decryption is high as the computational cost grows with the complexity of the access formula, which usually requires many pairings in most of the existing ABE schemes. As a result, this direct decryption method in anonymous ABE will suffer a severe efficiency drawback.

The outsourcing computations in distributed environments suffer from the trust problems between the outsourcer and the workers. All existing solutions only assume the rational lazy-but-honest workers. In this paper, we first introduce the rational lazy-and-partially-dishonest workers in the outsourcing computation model. In addition, we propose a new fair conditional payment scheme for outsourcing computation that is only based on traditional electronic cash systems. The proposed construction uses a semitrusted third party$T$to achieve the fairness and efficiency. However,$T$is only involved in the protocol in the exceptional case, namely in the case of disputes. Moreover, since neither the secret sharing/splitting scheme nor the cut-and-choose protocol is used for the generation or verification of the payment token, our solution clearly outperforms the existing schemes in terms of efficiency.

Attribute-based encryption (ABE) is a promising tool for implementing fine-grained cryptographic access control. Very recently, motivated by reducing the trust assumption on the authority, and enhancing the privacy of users, a multiple-authority key-policy ABE system, together with a semi-generic anonymous key-issuing protocol, have been proposed by Chase and Chow in CCS 2009. Since ABE allows encryption for multiple users with attributes satisfying the same policy, it may not be always possible to associate a decryption key to a particular individual. A misbehaving user could abuse the anonymity by leaking the key to someone else, without worrying of being traced. In this paper, we propose a multi-authority ciphertext-policy (AND gates with wildcard) ABE scheme with accountability, which allows tracing the identity of a misbehaving user who leaked the decryption key to others, and thus reduces the trust assumptions not only on the authorities but also the users. The tracing process is efficient and its computational overhead is only proportional to the length of the identity.

Cloud Computing has been envisioned as the next-generation architecture of IT Enterprise. It moves the application software and databases to the centralized large data centers, where the management of the data and services may not be fully trustworthy. This unique paradigm brings about many new security challenges, which have not been well understood. This work studies the problem of ensuring the integrity of data storage in Cloud Computing. In particular, we consider the task of allowing a third party auditor (TPA), on behalf of the cloud client, to verify the integrity of the dynamic data stored in the cloud. The introduction of TPA eliminates the involvement of the client through the auditing of whether his data stored in the cloud are indeed intact, which can be important in achieving economies of scale for Cloud Computing. The support for data dynamics via the most general forms of data operation, such as block modification, insertion, and deletion, is also a significant step toward practicality, since services in Cloud Computing are not limited to archive or backup data only. While prior works on ensuring remote data integrity often lacks the support of either public auditability or dynamic data operations, this paper achieves both. We first identify the difficulties and potential security problems of direct extensions with fully dynamic data updates from prior works and then show how to construct an elegant verification scheme for the seamless integration of these two salient features in our protocol design. In particular, to achieve efficient data dynamics, we improve the existing proof of storage models by manipulating the classic Merkle Hash Tree construction for block tag authentication. To support efficient handling of multiple auditing tasks, we further explore the technique of bilinear aggregate signature to extend our main result into a multiuser setting, where TPA can perform multiple auditing tasks simultaneously. Extensive security and performance analysis show that the proposed schemes are highly efficient and provably secure.

In an attribute-based signature (ABS), users sign messages with any predicate of their attributes issued from an attribute authority. Under this notion, a signature attests not to the identity of the individual who signed a message, but a claim regarding the attributes the underlying signer possesses. In ABS, users cannot forge signatures with attributes they do not possess even through colluding. On the other hand, a legitimate signer remains anonymous without the fear of revocation and is indistinguishable among all the users whose attributes satisfying the predicate specified in the signature. ABS is useful in many important applications such as anonymous authentication and attribute-based messaging systems.

Cloud computing is an emerging computing paradigm in which IT resources and capacities are provided as services over the Internet. Promising as it is, this paradigm also brings forth new challenges for data security and access control when users outsource sensitive data for sharing on cloud servers, which are likely outside of the same trust domain of data owners. To maintain the confidentiality of, sensitive user data against untrusted servers, existing work usually apply cryptographic methods by disclosing data decryption keys only to authorized users. However, in doing so, these solutions inevitably introduce heavy computation overhead on the data owner for key distribution and data management when fine-grained data access control is desired, and thus do not scale well. In this paper, we present a way to implement, scalable and fine-grained access control systems based on attribute-based encryption (ABE). For the purpose of secure access control in cloud computing, the prevention of illegal key sharing among colluding users is missing from the existing access control systems based on ABE. This paper addresses this challenging open issue by defining and enforcing access policies based on data attributes and implementing user accountability by using traitor tracing. Furthermore, both the user grant and revocation are efficiently supported by using the broadcast encryption technique. Extensive analysis shows that the proposed scheme is highly efficient and provably secure under existing security models.

Cloud computing has been acknowledged as one of the prevaling models for providing IT capacities. The off-premises computing paradigm that comes with cloud computing has incurred great concerns on the security of data, especially the integrity and confidentiality of data, as cloud service providers may have complete control on the computing infrastructure that underpins the services. This makes it difficult to share data via cloud providers where data should be confidential to the providers and only authorized users should be allowed to access the data. This work aims to construct a system for trusted data sharing through untrusted cloud providers, to address the above mentioned issue. The constructed system can imperatively impose the access control policies of data owners, preventing the cloud storage providers from unauthorized access and making illegal authorization to access the data.

As Cloud Computing becomes prevalent, sensitive information are being increasingly centralized into the cloud. For the protection of data privacy, sensitive data has to be encrypted before outsourcing, which makes effective data utilization a very challenging task. Although traditional searchable encryption schemes allow users to securely search over encrypted data through keywords, these techniques support only boolean search, without capturing any relevance of data files. This approach suffers from two main drawbacks when directly applied in the context of Cloud Computing. On the one hand, users, who do not necessarily have pre-knowledge of the encrypted cloud data, have to post process every retrieved file in order to find ones most matching their interest, On the other hand, invariably retrieving all files containing the queried keyword further incurs unnecessary network traffic, which is absolutely undesirable in today's pay-as-you-use cloud paradigm. In this paper, for the first time we define and solve the problem of effective yet secure ranked keyword search over encrypted cloud data. Ranked search greatly enhances system usability by returning the matching files in a ranked order regarding to certain relevance criteria (e.g., keyword frequency), thus making one step closer towards practical deployment of privacy-preserving data hosting services in Cloud Computing. We first give a straightforward yet ideal construction of ranked keyword search under the state-of-the-art searchable symmetric encryption (SSE) security definition, and demonstrate its inefficiency. To achieve more practical performance, we then propose a definition for ranked searchable symmetric encryption, and give an efficient design by properly utilizing the existing cryptographic primitive, order-preserving symmetric encryption (OPSE). Thorough analysis shows that our proposed solution enjoys ``as-strong-as-possible" security guarantee compared to previous SSE schemes, while correctly realizing the goal of ranked keyword search. Extensive experimental results demonstrate the efficiency of the proposed solution.

As Cloud Computing becomes prevalent, more and more sensitive information are being centralized into the cloud. For the protection of data privacy, sensitive data usually have to be encrypted before outsourcing, which makes effective data utilization a very challenging task. Although traditional searchable encryption schemes allow a user to securely search over encrypted data through keywords and selectively retrieve files of interest, these techniques support only exact keyword search. That is, there is no tolerance of minor typos and format inconsistencies which, on the other hand, are typical user searching behavior and happen very frequently. This significant drawback makes existing techniques unsuitable in Cloud Computing as it greatly affects system usability, rendering user searching experiences very frustrating and system efficacy very low. In this paper, for the first time we formalize and solve the problem of effective fuzzy keyword search over encrypted cloud data while maintaining keyword privacy. Fuzzy keyword search greatly enhances system usability by returning the matching files when users' searching inputs exactly match the predefined keywords or the closest possible matching files based on keyword similarity semantics, when exact match fails. In our solution, we exploit edit distance to quantify keywords similarity and develop an advanced technique on constructing fuzzy keyword sets, which greatly reduces the storage and representation overheads. Through rigorous security analysis, we show that our proposed solution is secure and privacy-preserving, while correctly realizing the goal of fuzzy keyword search.