Recent studies reveal that adversaries can manipulate the internal knowledge of large language models (LLMs) on selected topics through model editing, causing attacker-specified harmful or biased outputs when queried about the edited content. Once such tampered LLMs are distributed, they can mislead users on the targeted topics, thereby potentially propagating misinformation or reinforcing stereotypes. However, existing knowledge manipulation attacks rely on the ability to redistribute compromised models, which is infeasible in constrained settings like Federated Instruction Tuning (FedIT), where a central server controls LLM's training and distribution. In this work, we introduce ShadeEdit, the first attack framework that leverages strengthened model editing to enable knowledge manipulation in FedIT scenarios. ShadeEdit introduces two key components to address two challenges posed by the training process of FedIT: (1) a paraphrase-based editing dataset selection strategy to mitigate the dilution from benign updates on malicious ones by constructing a high-quality editing dataset, and (2) an adaptive manipulation mechanism to evade aggregation-based defenses via an adaptive clipping strategy. ShadeEdit achieves an average 99.5% attack success rate over eight robust aggregation algorithms while preserving instruction-following accuracy, demonstrating its strong attack effectiveness and model-utility preservation.

Website Fingerprinting (WF) is an emerging traffic analysis technique that enables a passive adversary to infer which websites a user visits. However, most existing studies, whether in single-tab or multi-tab settings, rely on the unrealistic assumption that users only access website homepages, diverging significantly from real-world browsing behavior. Even recent works extending WF to subpages primarily focus on website-level identification, without distinguishing which specific subpages are visited, thereby limiting the attack's granularity and scope. In this paper, we propose HiFi-WF (Hierarchical Fine-grained Website Fingerprinting), a novel framework that breaks the homepage-only assumption and extends WF to multi-tab recognition and fine-grained subpage identification. We formulate the task as a hierarchical multi-label classification problem, jointly modeling the distinctions and correlations between homepages and subpages. To this end, HiFi-WF integrates a unified CNN-based extractor and layered encoder with a Feature Interaction Module based on multi-head cross-attention to capture inter-level dependencies. An Enhanced SubHead enforces hierarchical constraints to suppress invalid subpage predictions, while a cascaded channel–spatial attention mechanism refines discriminative features for precise hierarchical identification. Experimental results demonstrate that HiFi-WF achieves state-of-the-art performance at both hierarchical levels, attaining F1-scores of 92.1% (homepage) and 81.9% (subpage), thereby validating its effectiveness in advancing WF attacks toward realistic, fine-grained, and multi-tab browsing scenarios. Related codes and datasets can be found in https://github.com/wusongyang02-blip/HiFi-WF.

Abstract Electricity theft detection seeks to thwart the illegal use of electricity, thereby safeguarding the safety and stability of the power system. Traditional methods, which typically rely on aggregated household consumption data to identify theft, often overlook the fact that household consumption is vulnerable to fluctuations in normal user behavior. This results in high false positive and false negative rates. To refine the accuracy, we propose a novel electricity theft detection method based on Non-Intrusive Load Monitoring (NILM) and multi-source data fusion. Our approach employs advanced NILM algorithms to cost-effectively extract individual appliance consumption data from aggregated power signals. We then integrate this data with household aggregate consumption data through a multi-source data fusion architecture. By analyzing the unique consumption patterns of different types of appliances, our approach identifies theft behaviors that cannot be detected by aggregate consumption data alone. Experimental results across three real-world datasets demonstrate that our method significantly outperforms single-source data-based benchmarks, achieving up to a 7.92% gain in F1-score and a 12.6% gain in Precision. Moreover, our method exhibits strong generalization ability across a series of typical machine learning models.

Blockchain-based steganography enables data hiding via encoding the covert data into a specific blockchain transaction field. However, previous works focus on the specific field-embedding methods while lacking a consideration on required field-generation embedding. In this paper, we propose a generic blockchain-based steganography framework (GBSF). The sender generates the required fields such as amount and fees, where the additional covert data is embedded to enhance the channel capacity. Based on GBSF, we design a reversible generative adversarial network (R-GAN) that utilizes the generative adversarial network with a reversible generator to generate the required fields and encode additional covert data into the input noise of the reversible generator. We then explore the performance flaw of R-GAN. To further improve the performance, we propose R-GAN withCounter-intuitive data preprocessing andCustom activation functions, namelyCCR-GAN. The counter-intuitive data preprocessing (CIDP) mechanism is used to reduce decoding errors in covert data, while it incurs gradient explosion for model convergence. The custom activation function named ClipSigmoid is devised to overcome the problem. Theoretical justification for CIDP and ClipSigmoid is also provided. We also develop a mechanism named T2C, which balances capacity and concealment. We conduct experiments using the transaction amount of the Bitcoin mainnet as the required field to verify the feasibility. We then apply the proposed schemes to other transaction fields and blockchains to demonstrate the scalability. Finally, we evaluate capacity and concealment for various blockchains and transaction fields and explore the trade-off between capacity and concealment. Experimental results demonstrate that R-GAN and CCR-GAN are able to enhance the channel capacity effectively and outperform state-of-the-art works.

Wireless medical sensor networks (WMSNs) have been widely adopted in healthcare for collecting users' physiological data, providing crucial references for medical diagnosis and prevention. However, transmitting sensitive data over public networks faces security risks, potentially leading to privacy breaches and financial losses. Moreover, large-scale data transmission increases energy consumption, hindering continuous monitoring. Therefore, achieving energy efficiency alongside data security is critical for WMSNs. This paper proposes a lightweight slope-based piecewise linear approximation algorithm for online data compression, utilizing slope intervals under a user defined error bound, to reduce energy consumption. Concurrently, we introduce a pairing-free certificateless aggregate signature scheme, proven secure under the random oracle model against different type adversaries, to enhance data privacy and integrity. Experimental results demonstrate that the compression algorithm achieves efficient compression while preserving trends, and the aggregate signature scheme reduces computational overhead by 20% without increasing communication costs.

Existing privacy-preserving multi-keyword retrieval schemes often suffer from reduced retrieval efficiency, lack robust verification mechanisms in dynamic environments, and are prone to symmetric key leakage issues. To address these shortcomings, we propose a dynamic and secure multi-keyword search scheme with a two-factor verification mechanism, named Mimi. Specifically, Mimi first constructs a dynamic verification tree structure to accelerate the verification of the correctness of returned results. Second, it builds an encrypted searchable index that supports sub-linear search time complexity. Third, Mimi incorporates a secure symmetric key exchange protocol to protect the confidentiality of the symmetric key. Furthermore, Mimi supports multi-user search operations without increasing the index construction costs and accommodates dynamic updates to both user roles and data. Through comprehensive security analysis, we demonstrate that Mimi ensures the security of the encrypted searchable inverted index and maintains query indistinguishability for users. Empirical evaluations show that the Mimi scheme is efficient and effective.

Natural Language Generation (NLG) applications have gained immense popularity due to the utilization of powerful deep learning techniques and large training corpora. However, the increasing prevalence of NLG models also poses a significant risk of unauthorized access or theft of intellectual property (IP). To safeguard NLG models, watermarking has emerged as a promising tool, but existing watermarking techniques based on pre-processing are prone to attacker detection and can potentially harm NLG applications. This paper proposes a novel, semantic, and stealthy watermarking scheme for IP protection of NLG models. Our approach embeds a semantic combination water mark, which is generated through a multi-stage process designed to be semantic and stealthy. This scheme endows an NLG model with a verifiable preference for specific semantic combinations, which are initiated by a foundational pattern but holistically constructed to preserve model functionality. To enhance the robustness, data embedding is systematically performed through a masked location injection. Consequently, the watermark is seamlessly integrated into NLG models without misleading their original attention mechanism. Comprehensive experiments are conducted to demonstrate that the proposed scheme is highly effective and robust in protecting the IP of NLG models while remaining stealthy to potential attackers.

Covert communication over public blockchains has emerged as a promising approach for secret data transmission. However, existing methods often suffer from high communication costs, low communication efficiency, and the risk of permanent data exposure. To overcome the above challenges, we propose FastBOC, a hybrid covert communication framework that integrates blockchain and onion networks. In FastBOC, the blockchain is employed as a covert signal channel to transmit lightweight signals, while the onion network handles high-capacity secret data transmission. This decoupling significantly reduces communication costs and avoids permanent data exposure on the blockchain. We further design an address-based encoding scheme and a dynamic port activation mechanism to enhance concealment. We implement FastBOC on the Ethereum testnet and conduct experiments to evaluate its concealment, efficiency, and cost. The results demonstrate that FastBOC (1) achieves strong concealment and (2) can transmit 1-Megabyte (MB) data within 20.11 seconds and reduce communication cost by 5–7 orders of magnitude compared to prior blockchain-based covert communication schemes.

Existing CNN inference frameworks based on FHE often suffer from reduced efficiency and accuracy due to the polynomial approximation of activation functions, and they lack effective mechanisms to prevent sensitive information leakage during the final classification stage. To address these limitations, we propose FSAT, a fast and secure inference framework enhanced with adversarial training. Specifically, FSAT employs a private CNN model architecture, where linear layers are computed through an optimized homomorphic ciphertext convolution operation, while non-linear layer operations are efficiently realized using a secure searchable index and an encrypted look-up table, which replace polynomial activation approximations and significantly improve inference accuracy and latency performance. To further mitigate information leakage, we introduce a dual-constraint adversarial training scheme that makes it substantially more difficult for an adversary to infer sensitive attributes of the input data. Experimental results demonstrate that FSAT achieves high inference accuracy and efficiency while substantially reducing the risk of sensitive data leakage.

The increasing saturation of terrestrial resources has driven the exploration of low-altitude applications such as air taxis. Low altitude wireless networks (LAWNs) serve as the foundation for these applications, and integrated sensing and communication (ISAC) constitutes one of the core technologies within LAWNs. However, the open nature of low-altitude airspace makes LAWNs vulnerable to malicious channel access attacks, which degrade the ISAC performance. Therefore, this paper develops a game-based framework to mitigate the influence of the attacks on LAWNs. Concretely, we first derive expressions of communication data’s signal-to-interference-plus-noise ratio and the age of information of sensing data under attack conditions, which serve as quality of service metrics. Then, we formulate the ISAC performance optimization problem as a Stackelberg game, where the attacker acts as the leader, and the legitimate drone and the ground ISAC base station act as second and first followers, respectively. On this basis, we design a backward induction algorithm that achieves the Stackelberg equilibrium while maximizing the utilities of all participants, thereby mitigating the attack-induced degradation of ISAC performance in LAWNs. We further prove the existence of the equilibrium. Simulation results show that the proposed algorithm outperforms existing baselines and a static Nash equilibrium benchmark, ensuring that LAWNs can provide reliable service for low-altitude applications.

The widespread deployment of deep learning models across various applications has raised significant concerns regarding data privacy. Membership inference attacks (MIAs), a major privacy threat, aim to determine whether a specific sample is used during model training, thereby posing significant risks to sensitive information. Most existing MIA methods rely on the model’s final state output, overlooking the process by which the model memorizes training samples. To better exploit model memorization for MIAs, we propose a novel attack method called machine unlearning-based membership inference attack (MU-MIA). The proposed method introduces machine unlearning to incrementally reduce the model’s memorization of specific samples, generating a forgetting trajectory for each sample. The forgetting trajectory is composed of temporal variations in different metrics of the sample during machine unlearning. To distinguish member from non-member samples, we design a BiLSTM-based binary classifier with attention, which captures discriminative temporal patterns within each forgetting trajectory. Moreover, the machine unlearning phase of our attack is conducted under a zero-shot setting, which eliminates the need for any real data during the unlearning process, thereby improving the practicality and generalizability of the attack. We evaluate the proposed MIA method across different datasets and model architectures, and the comparative experimental results show that our method outperforms existing baseline attack methods.

Federated Learning (FL) offers collaborative model training across multiple decentralized devices without the need to share data directly, enhancing privacy and data security. However, FL systems are susceptible to backdoor attacks, where malicious clients inject poisoned weights during training. Existing defenses, primarily based on anomaly detection, are prone to erroneous rejections of normal weights while accepting poisoned ones, largely due to shortcomings in quantifying similarities among client models. Furthermore, other defenses demonstrate effectiveness only when dealing with a limited number of malicious clients, typically fewer than 10%. To alleviate these vulnerabilities, we present G2uardFL, a protective framework that translates the detection of malicious clients into an attributed graph clustering problem, thus safeguarding FL systems. Specifically, this framework employs a client graph clustering approach to identify malicious clients and integrates an adaptive mechanism to amplify the discrepancy between the aggregated model and the poisoned ones, effectively eliminating embedded backdoors. Through empirical evaluation, comparing G2uardFL with cutting-edge defenses, such as FLAME (USENIX Security 2022) [37] and DeepSight (NDSS 2022) [43], against various backdoor attacks, including 3DFed (SP 2023) [26], our results demonstrate its significant effectiveness in mitigating backdoor attacks while having a negligible impact on the aggregated model’s performance on benign samples (i.e., the primary task performance). For instance, in an FL system with 25% malicious clients, G2uardFL reduces the attack success rate to 10.61%, while maintaining a primary task performance of 80.98% on the CIFAR-10 dataset. This surpasses the performance of the best-performing baseline, which merely achieves the attack success rate of 19.54%.

Subgraph Federated Learning (FL) has emerged as a promising paradigm for node classification tasks wherein subgraphs derived from a global graph are distributed across multiple devices to mitigate data leakage risks. Similar to other FL systems, subgraph FL faces significant security challenges, particularly from backdoor attacks, an area that remains extensively underexplored. Existing attacks typically follow a two-phase strategy to implant backdoors. However, in subgraph FL, such attacks often lead toDivergence Amplification, a phenomenon characterized by significant parameter discrepancies between normal and backdoored models, thereby compromising attack stealthiness. To tackle this challenge, we propose BEEF, a Backdoor attack with an End-to-End Framework designed for effectiveness, stealth, and durability. Unlike conventional methods, BEEF incorporates a dedicated trigger generator, which is jointly trained with a backdoored model. To increase its stealthiness, BEEF crafts adversarial perturbations as triggers that provoke misclassification while leaving the model’s parameters entirely untouched. Furthermore, by calibrating a subset of low-salience parameters associated with backdoor activation, BEEF ensures stable performance and sustained effectiveness across FL rounds. Comprehensive evaluations across eight datasets, four models, five state-of-the-art attacks, and six aggregation methods demonstrate BEEF’s effectiveness in deceiving GNNs while maintaining minimal impact on normal data performance. Additionally, we adapt BEEF to federated graph classification tasks, broadening its applicability and practicality.

As terrestrial resources become increasingly saturated, the developing attention is gradually shifting from the ground to the low-altitude airspace, which supports many emerging applications such as urban air taxis and aerial inspection. For these applications, low-altitude wireless networks (LAWNs) are the foundation, with integrated sensing, communications, and computing (ISCC) being one of the core parts. However, the openness of low-altitude airspace poses a serious threat to communications, degrading ISCC performance and ultimately compromising the reliability of applications supported by LAWNs. To address these challenges, this paper studies joint performance optimization of ISCC while considering security of the communications. Specifically, we derive beampattern error, secrecy rate, and age of information (AoI) as performance metrics for sensing, secure communication, and computing. Building on these metrics, we formulate a multi-objective optimization problem, which aims to balance sensing and computing performance while enhancing the secrecy rate of communications. We then propose a deep Q-network (DQN)-based multi-objective evolutionary algorithm, which adaptively selects evolutionary operators according to the evolving optimization objectives, thereby leading to more effective solutions. Extensive simulations show that the proposed method brings an average performance gain of about 14% compared to existing methods, thereby ensuring ISCC performance for applications supported by LAWNs.

With the rapid growth of cloud services, the storage of images in cloud environments requires secure and effective data encryption methods. Many thumbnail-preserving encryption (TPE) methods have thus been proposed to balance privacy and usability of image data. However, the exposure of thumbnail information in TPE methods may introduce privacy leakage risk, and a systematic evaluation of their security has not yet been conducted. In this paper, we propose a new Mamba-Transformer cooperation Network (MTNet) to recover the original images from the limited exposed thumbnail information, highlighting the information disclosure problem in TPE. Specifically, the core model component integrates a Mamba block and a Transformer block, which employ the powerful capabilities of the Mamba for wide field dependency modeling and the Transformer for effective channel interaction. Besides, the cascade architecture incorporates an intermediate output that provides supplementary information and achieves multilevel supervision, thereby improving the quality of the final output. Finally, to better utilize the subtle details in different levels, we propose a multi-scale fusion module that adaptively integrates features from various stages of the encoding process. The experimental results achieved by our proposed MTNet reveal that the privacy risk associated with TPE is significantly underestimated and more robust defense mechanisms are required. Source code is available athttps://github.com/HITLiXincodes/MTNet.

The security of collaboration among drone swarms necessitates the creation of inter-swarm/group keys. However, current solutions lack an inter-group key establishment mechanism that supports uniformity, flexibility, trustworthiness, efficiency and scalability to enable secure and efficient inter-swarm communications. In this paper, we propose DeGKG, an efficient decentralized inter-group key generation scheme that offers the construction of inter-swarm encryption keys for various drone swarms. It leverages the regional similarity of satellite cluster signals to construct drone swarm public/private key pairs, and employs the Chinese remainder theorem to integrate the swarm public keys for creating inter-group encryption keys, significantly reducing the number of complex cryptographic operations and the burden of inter-swarm key creation, and ensuring the scalability. In addition, the inter-swarm key creation allows the division of drones without the regional similarity of satellite cluster signals into various swarms, each composed of drones with the signal similarity, thus supporting the key establishment among all the drones and efficacy. An efficient blockchain consensus mechanism is implemented to uniformly generate inter-group encryption keys for various swarm combinations without relying on a trusted third party, thus ensuring the efficiency, flexibility, and trustworthiness of the generation. We prove the security of DeGKG, and demonstrate its efficacy and efficiency through simulations and comparisons.

Based on analyzing the character of cascaded decoder architecture commonly adopted in existing DETR-like models, this paper proposes a new decoder architecture. The cascaded decoder architecture constrains object queries to update in the cascaded direction, only enabling object queries to learn relatively-limited information from image features. However, the challenges for object detection in natural scenes (e.g., extremely-small, heavily-occluded, and confusingly mixed with the background) require an object detection model to fully utilize image features, which motivates us to propose a new decoder architecture with the parallel Multi-time Inquiries (MI) mechanism. MI mechanism is very simple, enabling object queries to parallelly perform multi-time inquiries to learn more comprehensive information from image features. Our MI based model, MI-DETR, outperforms all existing DETR-like models on COCO benchmark under different backbones and training epochs, achieving +2.3 AP and +0.6 AP improvements compared to the most representative model DINO and SOTA model Relation-DETR under ResNet-50 backbone.

Visual reprogramming (VR) leverages well-developed pre-trained models (e.g., a pre-trained classifier on ImageNet) to tackle target tasks (e.g., a traffic sign recognition task), without the need for training from scratch. Despite the effectiveness of previous VR methods, all of them did not consider the adversarial robustness of reprogrammed models against adversarial attacks, which could lead to unpredictable problems in safety-crucial target tasks. In this paper, we empirically find that reprogramming pre-trained models with adversarial robustness and incorporating adversarial samples from the target task during reprogramming can both improve the adversarial robustness of reprogrammed models. Furthermore, we propose a theoretically guaranteed adversarial robustness risk upper bound for VR, which validates our empirical findings and could provide a theoretical foundation for future research. Extensive experiments demonstrate that by adopting the strategies revealed in our empirical findings, the adversarial robustness of reprogrammed models can be enhanced.

Embodied agents exhibit immense potential across a multitude of domains, making the assurance of their behavioral safety a fundamental prerequisite for their widespread deployment. However, existing research predominantly concentrates on the security of general large language models, lacking specialized methodologies for establishing safety benchmarks and input moderation tailored to embodied agents. To bridge this gap, this paper introduces a novel input moderation framework, meticulously designed to safeguard embodied agents. This framework encompasses the entire pipeline, including taxonomy definition, dataset curation, moderator architecture, model training, and rigorous evaluation. Notably, we introduce EAsafetyBench, a meticulously crafted safety benchmark engineered to facilitate both the training and stringent assessment of moderators specifically designed for embodied agents. Furthermore, we propose Pinpoint, an innovative prompt-decoupled input moderation scheme that harnesses a masked attention mechanism to effectively isolate and mitigate the influence of functional prompts on moderation tasks. Extensive experiments conducted on diverse benchmark datasets and models validate the feasibility and efficacy of the proposed approach. The results demonstrate that our methodologies achieve an impressive average detection accuracy of 94.58%, surpassing the performance of existing state-of-the-art techniques, alongside an exceptional moderation processing time of merely 0.002 seconds per instance. The source code and datasets can be found at https://github.com/ZihanYan-CQU/EAsafetyBench.

Modern Text-to-Image (T2I) models deploy multi-layered defenses to block Not-Safe-For-Work (NSFW) content generation. These defenses typically include sequential layers such as prompt filters, concept erasers and image filters. While existing adversarial attacks have demonstrated vulnerabilities in isolated defense layers, they prove largely ineffective against multi-layered defenses deployed in real-world T2I systems. In this paper, we demonstrate that exploiting overlapping vulnerabilities across these distinct defense layers enables adversaries to systematically bypass the entire safeguard of T2I systems. We propose Transstratal Adversarial Attack (TAA), a novel black-box framework to compromise T2I models with multi-layered protection. It generates transstratal adversarial prompts to evade all defense layers simultaneously. This is accomplished through transstratal adversarial candidate generation using LLMs to fulfill implicit and subjective adversarial requirements against different defense layers, combined with adversarial genetic optimization for efficient black-box search to maximize the bypass rates and generated image harmfulness. Evaluated across 14 T2I models (e.g., Stable Diffusion, DALL·E, and Midjourney) and 17 safety modules, our attack achieves an average attack success rate of 85.6\%, surpassing state-of-the-art methods by 73.5\%. Our findings challenge the isolated design of safety mechanisms and establish the first benchmark for holistic robustness evaluation in multi-layered safeguarded T2I models. The code can be found in https://github.com/Bluedask/TAA-T2I.

Blockchain-based covert communication (BCC) provides high reliability and anonymity by embedding secret data into blockchain transactions. However, existing BCC approaches still face three fundamental limitations: (i) data mining risk, since transactions containing the secret data are permanently recorded on-chain and may be detected perpetually; (ii) limited efficiency, as only small payloads (e.g., 256 bits) can be carried per transaction; and (iii) private key leakage, where receivers often need access to the sender’s private key and may incur private key exposure. To address these issues, we propose a novel covert communication model with transaction withdrawal (BCC-TW) and a multisig-based data embedding scheme (MUL-DE). BCC-TW prevents covert transactions from being confirmed by constructing higher-fee double-spend transactions, thereby ensuring that secret data only exists temporarily in the mempool. MUL-DE encodes data into redundant public keys of Bitcoin multisig addresses, thus enabling higher efficiency and avoiding private key exposure. We implement a prototype on Bitcoin testnet and evaluate its concealment and efficiency. Experimental results demonstrate that the proposed approach achieves strong indistinguishability against statistical and deep-learning-based detectors, improves communication efficiency up to 251 bits per public key, and significantly reduces cost compared with state-of-the-art baselines.

Website Fingerprinting (WF) attacks allow passive eavesdroppers to deduce the websites a user visits by analyzing encrypted traffic, threatening user privacy. While current WF attacks achieve high accuracy, they typically assume single-tab browsing, which is unrealistic as users often open multiple tabs, creating mixed traffic. Existing multi-tab WF approaches require large datasets and frequent retraining due to evolving website content, limiting their practicality. In this paper, we introduce Few-shot Multi-tab Website Fingerprinting (FMWF), a novel approach designed to address the limitations of existing multi-tab WF attacks. FMWF directly tackles the challenges of mixed, overlapping traffic traces generated from multi-tab browsing, leveraging two key innovations: (1) an advanced data augmentation technique that synthesizes realistic multi-tab traffic sequences from easily collected single-tab traces, thereby dramatically reducing the need for large-scale real-world traffic data; and (2) a powerful fine-tuning algorithm based on transfer learning that adapts pre-trained models to new, multi-tab environments with minimal additional data. This two-stage framework enables FMWF to capture the complex effectively, overlapping traffic patterns inherent in multi-tab browsing while maintaining a high level of flexibility and significantly lowering computational and data collection burdens. Our experiments, conducted using real traffic traces collected from three widely-used browsers-Microsoft Edge, Google Chrome, and Tor Browser-highlight the superior performance of FMWF in both closed-world and open-world scenarios. Notably, FMWF achieves a minimum 12.3% improvement in accuracy compared to ARES (SP'23) [7], TMWF (CCS'23) [13], and BAPM (ACSAC'21) [10] in the open-world scenario. The code with related datasets is available at https://github.com/WW-Meng/FMWF.

Pre-trained models (PTMs) are widely adopted across various downstream tasks in the machine learning supply chain. Adopting untrustworthy PTMs introduces significant security risks, where adversaries can poison the model supply chain by embedding hidden malicious behaviors (backdoors) into PTMs. However, existing backdoor attacks to PTMs can only achieve partially task-agnostic and the embedded backdoors are easily erased during the fine-tuning process. This makes it challenging for the backdoors to persist and propagate through the supply chain. In this paper, we propose a novel and severer backdoor attack, TransTroj, which enables the backdoors embedded in PTMs to efficiently transfer in the model supply chain. In particular, we first formalize this attack as an indistinguishability problem between poisoned and clean samples in the embedding space. We decompose embedding indistinguishability into pre- and post-indistinguishability, representing the similarity of the poisoned and reference embeddings before and after the attack. Then, we propose a two-stage optimization that separately optimizes triggers and victim PTMs to achieve embedding indistinguishability. We evaluate TransTroj on four PTMs and six downstream tasks. Experimental results show that our method significantly outperforms SOTA task-agnostic backdoor attacks -- achieving nearly 100% attack success rate on most downstream tasks -- and demonstrates robustness under various system settings. Our findings underscore the urgent need to secure the model supply chain against such transferable backdoor attacks. The code is available at https://github.com/haowang-cqu/TransTroj

The nonintrusive load monitoring (NILM) technique, through its use of various deep neural networks (DNNs), is capable of learning residential appliances’ usage patterns from networked smart meters. However, such learned information may pose a serious privacy risk to users. In response to this privacy concern, in this article, we introduce an innovative adversarial attack. This attack can effectively restrict the NILM models’ ability to dissect power signals while maintaining accurate electricity charges for users. Given that previous adversarial attacks—which are designed for image classifiers and regressors with one-time output—cannot adequately handle NILM models and regressors with time-series output, we formally present the attack objective by leveraging the unique characteristics of regression and time-series data. Our proposed solution algorithms for this attack objective can generate imperceptible perturbations, effectively misleading the prediction of NILM models. To further ensure accurate billing calculation, we refine the attack objective to a practical version and propose a post-process that can iteratively remove the added perturbation in a certain period without compromising attack effectiveness. Experimental results on two real-world datasets, REDD and UK-DALE, demonstrate the effectiveness, transferability, and practicality of our proposed adversarial attack scheme.

The rapid evolution of the Industrial Internet of Things (IIoT) has driven unprecedented growth in industrial data volumes. To enhance cost efficiency and data-sharing capabilities, massive amounts of this data are stored in the cloud. Public Key Encryption with Keyword Search (PEKS) technology enables efficient encrypted data retrieval without key management and distribution issues and has been extensively studied for this purpose. However, due to inherent IIoT characteristics—such as heterogeneous data formats, resource-constrained devices, and heightened vulnerability to attacks, existing PEKS schemes face significant limitations: 1) typically restricted to single-keyword searches; 2) prohibitive computational overhead for resource-limited IIoT devices; 3) heightened risks of exploitation by attackers. To address these issues, we propose a lightweight certificateless authenticated encryption with multi-keyword search scheme, named CLAEMKS. It enables efficient multi-keyword search while substantially enhancing computational efficiency by eliminating the intensive bilinear pairing operations. Meanwhile, by leveraging certificateless cryptography, CLAEMKS solves certificate management problems while avoiding key escrow issues. Furthermore, formal security proofs and efficiency analyses are conducted to validate the effectiveness of our proposed scheme. The results demonstrate that CLAEMKS delivers substantial performance and security improvements.

With the rapid development of natural language processing (NLP), backdoor attacks have emerged as a significant security threat. These attacks inject malicious triggers into NLP models, causing them to produce adversarial output while remaining functional under normal input. To eliminate backdoors, existing data-driven defense methods typically transform backdoored samples into normal samples. However, these defenses lack the scalability to adapt effectively to various backdoor attacks. To address this challenge, we propose LLMBD, a novel data-driven backdoor defense method that leverages large language models (LLMs) for paraphrasing. Specifically, LLMBD uses large language models with optimized prompts to paraphrase the input text, eliminating potential backdoors while maintaining semantic integrity and textual fluency. During the training and inference phase, we apply grouping and major voting mechanisms to bypass residual backdoors in the paraphrased dataset. Finally, we validate the robustness and defense effectiveness of LLMBD through comprehensive model evaluations. Experimental results on datasets including SST-2, IMDB, and HSOL under various backdoor attack types (BadNets, AddSent, Synbkd, Stylebkd) show that LLMBD significantly outperforms existing methods such as RAP, STRIP, ParaFuzz, and TextGuard. On the SST-2, HSOL, and IMDb datasets, LLMBD achieves an average ASR drop of 0.278, with the average CACC maintained at 0.897. LLMBD exhibits superior robustness, generalization, and performance preservation without modifications to the backdoored model, providing an efficient and model-agnostic defense strategy against diverse backdoor threats.

The pursuit of refined data analysis and the preservation of privacy in Big Data pose significant concerns. Among the paramount paradigms for addressing these challenges, differential privacy stands out as a vital area of research. However, traditional differential privacy tends to be excessively restrictive when it comes to individuals’ control over their own data. It often treats all data as inherently sensitive, whereas in reality, not all information related to individuals is sensitive and requires an identical level of protection. In this paper, we define secret specification-based differential privacy (SSDP), where the term “secret specification” implies enabling users to decide what aspects of their information are sensitive and what are not, prior to data generation or processing. By allowing individuals to independently define their secret specifications, the SSDP achieves personalized privacy protection and facilitates effective data analysis. To enable the targeted application of SSDP, we further present task-specific mechanisms designed for database and graph data scenarios. Finally, we assess the trade-offs between privacy and utility inherent in the proposed mechanisms through comparative experiments conducted on real datasets, demonstrating the utility enhancements offered by SSDP mechanisms in practical applications.

Transferability and imperceptibility of adversarial examples are pivotal for assessing the efficacy of black-box attacks. While diffusion models have been employed to generate adversarial examples, leveraging their advanced image generation capability to enhance transferability and imperceptibility, these methods typically focus only on perturbing the image or latent space. They often ignore the critical role of semantic information in the denoising process, thereby impeding the improvement of the transferability of adversarial examples. Furthermore, the modification of high-level semantics inevitably introduces image blurring. This degradation in visual quality makes the adversarial examples more susceptible to detection. To overcome the above limitations, we are the first to utilize image latent encoding and semantic embedding perturbations to enhance the performance of adversarial attacks. Then, the LESEP method is proposed. In the LESEP framework, we first apply image latent encoding attack to achieve deception of the target model. Second, the semantic embedding attack enhances the transferability of adversarial examples. Additionally, we utilize the image restoration technique to guarantee the high imperceptibility of the crafted adversarial examples. Through comprehensive experiments on diverse datasets, different network architectures and defense methods, we have demonstrated that the LESEP method achieves outstanding transferability and imperceptibility while displaying strong robustness.

Perceptual image encryption serves as a pivotal mechanism for delegating processing while ensuring the visual security of image data. The robustness of such encryption schemes is traditionally evaluated through cryptanalysis techniques, yet these approaches heavily rely on manual labor and prerequisite knowledge of the encryption algorithms. Recently, some works attempt to reveal visual content from perceptually encrypted images Based on CNN architectures. However, it is still tricky for these increasingly complex methods to reveal informative visual details. In this study, we focus on the extraction and utilization of inherent hierarchical features within the input image itself to significantly advances the field. To achieve it, we present a novel Progressive Fusion Attack Network (PFAN) to fully explore the hierarchical features. PFAN incorporates multiple subbranches, forming a progressive fusion structure that facilitates informative hierarchical feature representations and offers robust model fault tolerance. To enhance the reconstruction of encryption-induced distortions, we incorporate a Multiscale Feature Extraction Module (MFEM) that captures robust hierarchical features across various scales. Meanwhile, a Hierarchical Feature Fusion Module (HFFM) is designed to adaptively integrate and highlight the optimal feature representations, further optimizing the visual content reconstruction process. Extensive experimental evaluation demonstrates that PFAN exhibits remarkable agnosticism towards different perceptual encryption schemes and encryption strengths, achieving superior performance. Furthermore, PFAN outperforms state-of-the-art CNN-based image restoration methods in terms of effectiveness and generalizability.

The sixth-generation (6G) mobile communication networks are perceived as large-scale heterogeneous networks. With their increased heterogenization and densification, it is crucial to guarantee the security and efficiency of user equipment's handovers between networks. However, existing cross-domain handover authentication schemes cannot ensure handover authentication efficiency and cannot balance privacy and system efficiency, which thus cannot be directly applied in heterogeneous networks. In this paper, we present PECHA, a privacy-preserving and efficient cross-domain handover authentication scheme for heterogeneous networks, which enables anonymous authentication on user equipment (UE) through the collision property of chameleon hash functions. PECHA ensures authentication efficiency by employing the interplanetary file system and blockchain to synchronize UE's authentication information to target networks in advance. The privacy and system efficiency are balanced by modeling the unlinkability of UE's new and old chameleon hash values and determining the update frequency of UE chameleon hash value. PECHA also achieves correctness, mutual authentication and key agreement, anonymity, unlinkability, conditional privacy, forward/backward secrecy, robustness, known randomness secrecy, key escrow freeness and rapid response, and resists against spoofing attacks, replay attacks and man-in-the-middle attacks. Comprehensive performance analysis, evaluation and comparisons show that PECHA is efficient with respect to both computation and communication.

With the prevalence of online social networks (OSNs), much personal information is collected and maintained by trusted service providers for third-party queries and analyses. Existing works regarding differentially private social network data publication overlook the fact that different users exhibit distinct privacy preferences or sensitivity inclinations. Neglecting these individual nuances may lead to privacy mechanisms that are overly conservative or inadequately protective. Furthermore, the injection of excessive noise into OSN data perceived by users as non-personal or less sensitive can incur additional privacy costs, resulting in lower service quality. This paper introduces a fine-grained, sensitivity-aware personalized edge differential privacy model (SPEDP) for OSNs. Specifically, SPEDP enables each OSN user to individually define the sensitivity level of their social connections, facilitating user-friendly personalized privacy settings. We design a privacy-aware mechanism that operates within a trusted service provider, capable of establishing privacy protection levels based on user-perceived sensitivity settings. Additionally, we propose a sensitivity-aware sampling mechanism to implement SPEDP. To further optimize the privacy mechanism, we explore a privacy threshold optimization strategy aimed at minimizing privacy budget waste. Finally, the personalized privacy protections and utility improvements achieved by the SPEDP mechanism are rigorously validated through theoretical analysis and comprehensive comparative experiments on benchmark datasets.

Reconfigurable Intelligent Surfaces (RIS) can optimize spectrum and energy efficiency in the sixth-generation (6G) wireless communication system through dynamic electromagnetic wave manipulation. The programmable control of spatial electromagnetic signals by RIS presents a double-edged sword, and it can also be exploited by malicious attackers. However, few studies have focused on the detection and identification of such malicious RIS. To fill this gap, we propose a novel spoofing detection framework combining dynamic key-embedded phase codebooks with a dual-channel feature extraction mechanism. This approach jointly decodes wireless channel fingerprints and cryptographic signatures from received signals. A hybrid discriminator, integrating autoencoder-based signal reconstruction fidelity and key-matching validation, enables robust legitimacy verification. The prototype experiments using USRP SDR and RIS hardware show that the verification accuracy of the scheme can reach 100%, when the signal-to-noise ratio (SNR) is above 10dB, the number of training sample points is more than 128, and the codebook dimension is near 32.

Assessing the stealthiness of adversarial perturbations is challenging due to the lack of appropriate evaluation metrics. Existing evaluation metrics, e.g.,$L_{p}$norms or Image Quality Assessment (IQA), fall short of assessing the pixel-level stealthiness of subtle adversarial perturbations since these metrics are primarily designed for traditional distortions. To bridge this gap, we present the first comprehensive study on the subjective and objective assessment of the stealthiness of adversarial perturbations from a visual perspective at a pixel level. Specifically, we propose new subjective assessment criteria for human observers to score adversarial stealthiness in a fine-grained manner. Then, we create a large-scale adversarial example dataset comprising 10586 pairs of clean and adversarial samples encompassing twelve state-of-the-art adversarial attacks. To obtain the subjective scores according to the proposed criterion, we recruit 60 human observers, and each adversarial example is evaluated by at least 15 observers. The mean opinion score of each adversarial example is utilized for labeling. Finally, we develop a three-stage objective scoring model that mimics human scoring habits to predict adversarial perturbation’s stealthiness. Experimental results demonstrate that our objective model exhibits superior consistency with the human visual system, surpassing commonly employed metrics like PSNR and SSIM.

Backdoor attacks pose a serious security threat to Natural Language Processing (NLP) models, allowing adversaries to manipulate model outputs through hidden triggers. Although backdoor detection methods have been developed to address this issue, existing approaches based on trigger inversion are effective only for simple, visible triggers. These methods struggle to handle semantically enhanced, invisible triggers and often fail to provide accurate backdoor determinations due to reliance on unreliable heuristics, making it difficult to reliably distinguish backdoored models from benign ones. This presents a critical gap in current detection techniques. To address these challenges, we propose a novel trigger inversionSemInvthat consists of two key contributions: consistent semantics inversion and identifiable condition inspection. Consistent semantics inversion introduces a new regularization technique into the trigger optimization process, enabling more effective inversion of semantically constrained triggers. Identifiable condition inspection assesses the attack performance margin across different identifiable conditions, providing robust evidence for distinguishing backdoored models from benign ones. We evaluateSemInvusing the TrojAI round 6–8 datasets and demonstrate that it significantly outperforms state-of-the-art approaches in both backdoor detection accuracy and trigger inversion performance. Our method also proves effective against models with stealthy triggers, advancing the field of NLP security by offering a more comprehensive solution for identifying backdoor attacks. The code repository is in https://github.com/Bluedask/SemInv.

AI-synthesized speech, also known as deepfake speech, has recently raised significant concerns due to the rapid advancement of speech synthesis and speech conversion techniques. Previous works often rely on distinguishing synthesizer artifacts to identify deepfake speech. However, excessive reliance on these specific synthesizer artifacts may result in unsatisfactory performance when addressing speech signals created by unseen synthesizers. In this paper, we propose a robust deepfake speech detection method that employs feature decomposition to learn synthesizer-independent content features as complementary for detection. Specifically, we propose a dual-stream feature decomposition learning strategy that decomposes the learned speech representation using a synthesizer stream and a content stream. The synthesizer stream specializes in learning synthesizer features through supervised training with synthesizer labels. Meanwhile, the content stream focuses on learning synthesizer-independent content features, enabled by a pseudo-labeling-based supervised learning method. This method randomly transforms speech to generate speed and compression labels for training. Additionally, we employ an adversarial learning technique to reduce the synthesizer-related components in the content stream. The final classification is determined by concatenating the synthesizer and content features. To enhance the model’s robustness to different synthesizer characteristics, we further propose a synthesizer feature augmentation strategy that randomly blends the characteristic styles within real and fake audio features and randomly shuffles the synthesizer features with the content features. This strategy effectively enhances the feature diversity and simulates more feature combinations. Experimental results on four deepfake speech benchmark datasets demonstrate that our model achieves state-of-the-art robust detection performance across various evaluation scenarios, including cross-method, cross-dataset, and cross-language evaluations.

Federated Learning (FL) is a collaborative machine learning paradigm allowing participants to train a global model collaboratively without sharing training data. The distributed nature makes FL vulnerable to the untargeted or backdoor model poisoning attacks (MPAs). Hence, lots of defense methods are proposed to secure FL. However, existing defenses are ineffective in defending against the emerging stealthy layer-space MPA, since the defenses either focus on the model space or ignore the disparities between the layers. In this paper, we propose a novel layer-space defense method called FLGuardian that can protect the global model from the state-of-the-art MPAs. FLGuardian first employs a new layer-wise detection to find out the benign clients for each layer through pairwise cosine distances and pairwise Euclidean distances combined with a clustering algorithm. Then, FLGuardian assigns a trust score for each client according to the detection results of all the layers, where a deeper layer in the model brings a higher weight in the scoring. Finally, we select several clients with the highest scores for updating the global model. Experimental results show that FLGuardian excels nine typical defense methods against seven state-of-the-art MPAs in most cases. Particularly, under LPattack, the emerging layer-space backdoor MPA, FLGuardian secures Backdoor Success Rate (BSR) below 3% while other defenses have over 93% BSRs on CIFAR-10. Moreover, FLGuardian remains robust against adaptive attacks tailored to FLGuardian.

Collaborative learning has gained significant traction for training deep learning models without sharing the original data of participants, particularly when dealing with sensitive data such as facial images. However, current gradient inversion attacks are employed to progressively reconstruct private data from gradients, and they have shown successful in extracting private training data. Nonetheless, our observations reveal that these methods exhibit suboptimal performance in face reconstruction and result in the loss of numerous facial details. In this paper, we propose DFLeak, an effective approach to boost face leakage from gradients using residual optimization and thwart the privacy of facial applications in collaborative learning. In particular, we first introduce a superior initialization method to stabilize the inversion process. Second, we propose to integrate prior-free face restoration (PFFR) results into the gradient inversion optimization process in a residual manner, which enriches facial details. We further design a pixel update schedule to mitigate the adverse effects of image regularization terms and preserve fine facial details. Comprehensive experimentation demonstrates the effectiveness of our approach in achieving more realistic and higher-quality facial image reconstructions, surpassing the performance of state-of-the-art gradient inversion attacks.

This is a corrigendum for the article “An Unbiased Risk Estimator for Partial Label Learning with Augmented Classes” published in ACM Trans. Intell. Syst. Technol. 15(6): 131:1-131:22 (2024).

Authenticated key agreement (AKA) between vehicles and road side units (RSUs) is crucial in vehicular ad-hoc networks (VANETs). However, existing solutions still suffer from high overheads of AKA and lack a mechanism to balance privacy strength and system efficiency. In this paper, we present a lightweight anonymous authenticated key agreement (LWAKA) scheme for VANETs, supporting lightweight anonymous authentication and key agreement between vehicles and RSUs simultaneously. In particular, vehicles’ authentication information is synchronized to target RSUs in advance for accelerating authentication, and lightweight cryptographic operations (i.e., hash function, hash-based message authentication, physical unclonable function, fuzzy extractor and symmetric encryption) are employed to ensure the high efficiency of AKA in terms of computation and communication overheads. The system efficiency and privacy are balanced through modeling the relationship between the frequency of pseudonym updates and the unlinkability of the vehicles’ new and old pseudonyms. Security analysis shows that LWAKA not only achieves anonymity, conditional privacy, pseudonym unlinkability, key escrow freeness, and physical security, but also resists against most known attacks. Comparative experimental results demonstrate that LWAKA outperforms existing schemes in terms of lightweight design.

Smart grid, through networked smart meters employing the non-intrusive load monitoring (NILM) technique, can considerably discern the usage patterns of residential appliances. However, this technique also incurs privacy leakage. To address this issue, we propose an innovative scheme based on adversarial attack in this paper. The scheme effectively prevents NILM models from violating appliance-level privacy, while also ensuring accurate billing calculation for users. To achieve this objective, we overcome two primary challenges. First, as NILM models fall under the category of time-series regression models, direct application of traditional adversarial attacks designed for classification tasks is not feasible. To tackle this issue, we formulate a novel adversarial attack problem tailored specifically for NILM and providing a theoretical foundation for utilizing the Jacobian of the NILM model to generate imperceptible perturbations. Leveraging the Jacobian, our scheme can produce perturbations, which effectively misleads the signal prediction of NILM models to safeguard users' appliance-level privacy. The second challenge pertains to fundamental utility requirements, where existing adversarial attack schemes struggle to achieve accurate billing calculation for users. To handle this problem, we introduce an additional constraint, mandating that the sum of added perturbations within a billing period must be precisely zero. Experimental validation on real-world power datasets REDD and U.K.-DALE demonstrates the efficacy of our proposed solutions, which can significantly amplify the discrepancy between the output of the targeted NILM model and the actual power signal of appliances, and enable accurate billing at the same time. Additionally, our solutions exhibit transferability, making the generated perturbation signal from one target model applicable to other diverse NILM models.

In this paper, we present a novel task of source-free cross-modal adversarial example generation, which generates adversarial examples based on textual descriptions of attackers. This task has two challenges as follows. First, how to generate adversarial examples when the clean examples are missing or inaccessible. Second, how to achieve fine-grained custom adversarial example generation according to the semantic descriptions of the attackers. Existing adversarial example generation methods can not effectively deal with these two challenges. To address these challenges, we propose a Source-Free Cross-Modal Adversarial Example Generation framework, abbreviated as SFCM-AEG. Within the SFCM-AEG model, we firstly leverage a pre-trained GPT as a simulator to construct textual descriptions of attackers by labels. Following this, we employ a diffusion model to synthesize an image that aligns with the generated textual description. Finally, the generated images are converted into adversarial examples using an adversarial example generation method. Experimental results demonstrate that our proposed SFCM-AEG method can generate adversarial examples with customized semantic descriptions, without relying on clean examples, while achieving strong attack performance in a white-box setting.

Fully connected (FC) layers play a significant role in deep neural networks (DNNs) models. Owing to the complexity of its parameters, an FC layer has sufficient capacity to manage high-dimensional tasks, so a large amount of memory and powerful computing capabilities become essential requirements. However, the large number of parameters in an FC layer greatly limits the practical application of this model. To address this problem, we apply matrix optimization to an FC layer. First, an added penalty term properly maintains the sparsity of the imposed weights. Second, a rank constraint is applied to the two components of the factorized weight matrix. Our compression algorithm can effectively reduce the number of required network parameters, which not only reduces the computational complexity of the network but also results in better generalizability on a test dataset. Finally, the effectiveness of the proposed method is verified in two different computer vision task domains. Experiments show that our sparse reduced-rank method achieves a better compression ratio with a lower accuracy loss relative to the competing approaches. The code is available at https://github.com/cheer79/Compress_FC .

Personalized federated learning (PFL) combines client needs and data characteristics to train personalized models for local clients. However, the most of previous PFL schemes encountered challenges such as low model prediction accuracy and privacy leakage when applied to practical datasets. Besides, the existing privacy protection methods fail to achieve satisfactory results in terms of model prediction accuracy and security simultaneously. In this paper, we propose a Privacy-preserving Personalized Federated Learning under Secure Multi-party Computation (SMC-PPFL), which can preserve privacy while obtaining a local personalized model with high prediction accuracy. In SMC-PPFL, noise perturbation is utilized to protect similarity computation, and secure multi-party computation is employed for model sub-aggregations. This combination ensures that clients' privacy is preserved, and the computed values remain unbiased without compromising security. Then, we propose a weighted sub-aggregation strategy based on the similarity of clients and introduce a regularization term in the local training to improve prediction accuracy. Finally, we evaluate the performance of SMC-PPFL on three common datasets. The experimental results show that SMC-PPFL achieves 2% ∼ 15% higher prediction accuracy compared to the previous PFL schemes. Besides, the security analysis also verifies that SMC-PPFL can resist model inversion attacks and membership inference attacks

The centralized server in traditional federated learning (FL) is authoritative (i.e. decisive control), which may cause immeasurable damage to the system's security in the event of decision failure or attack. To weaken the authority of the central server, existing studies have proposed blockchain-based federated learning (BFL) approaches. However, existing BFL still suffers from high resource overhead and difficulty in resisting high malicious ratio (more than 50%) attacks. To address the above challenge, this paper proposes an efficient and secure non-authoritative (i.e. highly decentralized) anarchic (i.e. distributed self-governance) federated learning framework which is named NAAFL. During the local process of NAAFL, an area credit-based screening mechanism for participating devices is proposed to ensure that participating devices are always highly trusted devices with higher total credit values. Then, to effectively exclude a high percentage of malicious training gradients, a multi-device validation voting mechanism based on historical information is designed to construct the global gradient. Subsequently, to weaken the central server authority and reduce the resource overhead while guaranteeing security, a secure and low-consumption consensus mechanism based on the federation chain is proposed, and the overhead is further reduced by a momentum acceleration algorithm. Finally, the theoretical analysis and experimental simulation are conducted on the proposed NAAFL. The results further show that the proposed NAAFL outperforms existing studies and can defend against attacks with up to 80% malicious ratio, which exceeds the common threshold (50%) of existing studies. Meanwhile, the overhead of NAAFL is reduced by about 77.51% compared to BFL.

Privacy-preserving Machine Learning as a Service (MLaaS) enables the powerful cloud server to run its well-trained neural model upon the input from resource-limited client, with both of server's model parameters and client's input data protected. While computation efficiency is critical for the practical implementation of privacy-preserving MLaaS and it is inspiring to witness recent advances towards efficiency improvement, there still exists a significant performance gap to real-world applications. In general, state-of-the-art frameworks perform function-wise efficiency optimization based on specific cryptographic primitives. Although it is logical, such independent optimization for each function makes noticeable amount of expensive operations unremovable and misses the opportunity to further accelerate the performance by jointly considering privacy-preserving computation among adjacent functions. As such, we propose COIN: Conjunctive Optimization with Interleaved Nexus, which remodels mainstream computation for each function to conjunctive counterpart for composite function, with a series of united optimization strategies. Specifically, COIN jointly computes a pair of consecutive nonlinear-linear functions in the neural model by reconstructing the intermediates throughout the whole procedure, which not only eliminates the most expensive crypto operations without invoking extra encryption enabler, but also makes the online crypto complexity independent of filter size. Experimentally, COIN demonstrates 11.2x to 29.6x speedup over various function dimensions from modern networks, and 6.4x to 12x speedup on the total computation time when applied in networks with model input from small-scale CIFAR10 to large-scale ImageNet.

With the advent of various mobile IoT devices, a large amount of e-health record (EHR) data has been generated. This data has great potential to improve medical research. However, there are many challenges regarding the sharing of medical data. Firstly, users are more inclined to interact anonymously. Secondly, verifying the validity of certificates in the case of anonymous interactions is challenging. In addition, it is necessary to uncover the identities of the actual interacting parties in the event of malicious behavior. Therefore, we address the above challenges and propose group signatures with time constraints to support anonymous and traceable EHR data sharing. First, we propose a group signature scheme that supports traceability. Second, to address the issue of validating anonymous certificates, we propose group signatures with time constraints that enable dynamic updates to validity. Through this, we can dynamically revoke group members. Lastly, security proofs and efficiency analyses demonstrate that our scheme is both secure and efficient.

Adversarial robustness is an important standard for measuring the quality of learned models, and adversarial training is an effective strategy for improving the adversarial robustness of models. In this paper, we disclose that adversarially trained models are vulnerable to two-faced attacks, where slight perturbations in input features are crafted to make the model exhibit a false sense of robustness in the verification phase. Such a threat is significantly important as it can mislead our evaluation of the adversarial robustness of models, which could cause unpredictable security issues when deploying substandard models in reality. More seriously, this threat seems to be pervasive and tricky: we find that many types of models suffer from this threat, and models with higher adversarial robustness tend to be more vulnerable. Furthermore, we provide the first attempt to formulate this threat, disclose its relationships with adversarial risk, and try to circumvent it via a simple countermeasure. These findings serve as a crucial reminder for practitioners to exercise caution in the verification phase, urging them to refrain from blindly trusting the exhibited adversarial robustness of models.

Watching micro-videos is becoming a part of public daily life. Usually, user watching behaviors are thought to be rooted in their multiple different interests. In the paper, we propose a model named OPAL for micro-video matching, which elicits a user’s multiple heterogeneous interests by disentangling multiple soft and hard interest embeddings from user interactions. Moreover, OPAL employs a two-stage training strategy, in which the pre-train is to generate soft interests from historical interactions under the guidance of orthogonal hyper-categories of micro-videos and the fine-tune is to reinforce the degree of disentanglement among the interests and learn the temporal evolution of each interest of each user. We conduct extensive experiments on two real-world datasets. The results show that OPAL not only returns diversified micro-videos but also outperforms six state-of-the-art models in terms of recall and hit rate.

Text-to-image (T2I) diffusion models enjoy great popularity and many individuals and companies build their applications based on publicly released T2I diffusion models. Previous studies have demonstrated that backdoor attacks can elicit T2I diffusion models to generate unsafe target images through textual triggers. However, existing backdoor attacks typically demand substantial tuning data for poisoning, limiting their practicality and potentially degrading the overall performance of T2I diffusion models. To address these issues, we propose EvilEdit, a training-free and data-free backdoor attack against T2I diffusion models. EvilEdit directly edits the projection matrices in the cross-attention layers to achieve projection alignment between a trigger and the corresponding backdoor target. We preserve the functionality of the backdoored model using a protected whitelist to ensure the semantic of non-trigger words is not accidentally altered by the backdoor. We also propose a visual target attack EvilEdit VTA, enabling adversaries to use specific images as backdoor targets. We conduct empirical experiments on Stable Diffusion and the results demonstrate that the EvilEdit can backdoor T2I diffusion models within one second with up to 100% success rate. Furthermore, our EvilEdit modifies only 2.2% of the parameters and maintains the model's performance on benign prompts. Our code is available at https://github.com/haowang-cqu/EvilEdit.

Monocular Depth Estimation (MDE) enables the prediction of scene depths from a single RGB image, having been widely integrated into production-grade autonomous driving systems, e.g., Tesla Autopilot. Current adversarial attacks to MDE models focus on attaching an optimized adversarial patch to a designated obstacle. Although effective, this approach presents two inherent limitations: its reliance on specific obstacles and its limited malicious impact. In contrast, we propose a pioneering attack to MDE models that \textit{decouples obstacles from patches physically and deploys optimized patches on roads}, thereby extending the attack scope to arbitrary traffic participants. This approach is inspired by our groundbreaking discovery: \textit{various MDE models with different architectures, trained for autonomous driving, heavily rely on road regions} when predicting depths for different obstacles. Based on this discovery, we design the Adversarial Road Marking (AdvRM) attack, which camouflages patches as ordinary road markings and deploys them on roads, thereby posing a continuous threat within the environment. Experimental results from both dataset simulations and real-world scenarios demonstrate that AdvRM is effective, stealthy, and robust against various MDE models, achieving about 1.507 of Mean Relative Shift Ratio (MRSR) over 8 MDE models. The code is available at \url{https://github.com/a-c-a-c/AdvRM.git}

This paper addresses the problem of on-road object importance estimation, which utilizes video sequences captured from the driver's perspective as the input. Although this problem is significant for safer and smarter driving systems, the exploration of this problem remains limited. On one hand, publicly-available large-scale datasets are scarce in the community. To address this dilemma, this paper contributes a new large-scale dataset named Traffic Object Importance (TOI). On the other hand, existing methods often only consider either bottom-up feature or single-fold guidance, leading to limitations in handling highly dynamic and diverse traffic scenarios. Different from existing methods, this paper proposes a model that integrates multi-fold top-down guidance with the bottom-up feature. Specifically, three kinds of top-down guidance factors (i.e., driver intention, semantic context, and traffic rule) are integrated into our model. These factors are important for object importance estimation, but none of the existing methods simultaneously consider them. To our knowledge, this paper proposes the first on-road object importance estimation model that fuses multi-fold top-down guidance factors with bottom-up feature. Extensive experiments demonstrate that our model outperforms state-of-the-art methods by large margins, achieving 23.1% Average Precision (AP) improvement compared with the recently proposed model (i.e., Goal).

This paper is motivated by an interesting phenomenon: the performance of object detection lags behind that of instance segmentation (i.e., performance imbalance) when investigating the intermediate results from the beginning transformer decoder layer of MaskDINO (i.e., the SOTA model for joint detection and segmentation). This phenomenon inspires us to think about a question: will the performance imbalance at the beginning layer of transformer decoder constrain the upper bound of the final performance? With this question in mind, we further conduct qualitative and quantitative pre-experiments, which validate the negative impact of detection-segmentation imbalance issue on the model performance. To address this issue, this paper proposes DI-MaskDINO model, the core idea of which is to improve the final performance by alleviating the detection-segmentation imbalance. DI-MaskDINO is implemented by configuring our proposed De-Imbalance (DI) module and Balance-Aware Tokens Optimization (BATO) module to MaskDINO. DI is responsible for generating balance-aware query, and BATO uses the balance-aware query to guide the optimization of the initial feature tokens. The balance-aware query and optimized feature tokens are respectively taken as the Query and Key&Value of transformer decoder to perform joint object detection and instance segmentation. DI-MaskDINO outperforms existing joint object detection and instance segmentation models on COCO and BDD100K benchmarks, achieving +1.2 $AP^{box}$ and +0.9 $AP^{mask}$ improvements compared to SOTA joint detection and segmentation model MaskDINO. In addition, DI-MaskDINO also obtains +1.0 $AP^{box}$ improvement compared to SOTA object detection model DINO and +3.0 $AP^{mask}$ improvement compared to SOTA segmentation model Mask2Former.

As an advanced one-to-many public key encryption system, attribute-based encryption (ABE) is widely believed to be a promising technology for achieving flexible and fine-grained access control of encrypted data on untrusted storage servers (e.g., public cloud servers). However, user revocation in ABE is a critical but challenging problem, and designing efficient revocable ABE has been an active research topic in the past decade. Almost all the existing revocable ABE schemes incorporate a timestamp in the encryption algorithm such that revoked users cannot decrypt ciphertexts generated in future time intervals. To prevent revoked users from decrypting past ciphertexts, the storage server needs to perform a process called ciphertext delegation (Sahai et al., CRYPTO’12) that periodically updates the timestamp for all ciphertexts. As the number of ciphertexts could be huge in a storage system, ciphertext delegation could pose a huge computation overhead to the server.Motivated by the popularity of commodity Trusted Execution Environment (TEE) technologies, this paper initiates the study on hardware-based revocable ABE (HR-ABE) to eliminate the (unscalable) ciphertext delegation and prevent collusion attacks between an untrusted storage server and revoked users. We formalize this new notion and present an efficient HR-ABE construction that also supports outsourced decryption for resource-constrained data users. Furthermore, HR-ABE is also designed to address the potential secret leakage problem suffered by TEE (e.g., due to side-channel attacks) so that the leakage of secrets possessed by TEE does not lead to leakage of user data. We prove HR-ABE’s security formally and benchmark its performance experimentally.

Privacy-preserving Machine Learning as a Service (MLaaS) enables the resource-limited client to cost-efficiently obtain inference output of a well-trained neural model that is possessed by the cloud server, with both client’s input and server’s model parameters protected. While efficiency plays a core role for practical implementation of privacy-preserving MLaaS and it is encouraging to witness recent advances towards efficiency improvement, there still exists a significant performance gap to real-world applications. The basic logic in state-of-the-art frameworks involves an individual computation for each function of the neural model, based on specific cryptographic primitives. While it is definitely logical, we look back to the necessity of this function-wise methodology and initiate the comprehensive exploration towards allied optimization for efficient privacy-preserving MLaaS. Under such fresh perspective, we remodel the computation process that is always from input to output of the same function in mainstream works, to the allied counterpart that is from one function’s input associated with the start of expensive overhead to another function’s output enabling effective circumvention of unnecessary cost within the procedure. As such we propose FIT (Function Input Tuning) which features by a computation module for composite function with a series of joint optimization strategies. Theoretically, FIT not only eliminates the most expensive crypto operations without invoking extra encryption enabler, but also makes the running-time crypto complexity independent of filter size. Experimentally, FIT demonstrates tens of times speedup over various function dimensions from modern models, and 4.5× to 35.5× speedup on the total computation time when plugged in neural networks with data from small-scale MNIST to large-scale ImageNet.

Currently, power data trading typically relies on Web pages as the conventional mode of mediation. Nevertheless, dishonest trading Web may secretly resell the data sets of grid companies or have no way of knowing what the buyer has done with the power data, thereby compromising the privacy of power user. This article proposes a privacy-preserving supply-demand consistency matching service scheme (PPMSE) to address the problem of whether the power data provided by the seller aligns with the requirements of the buyer in power data trading. The scheme utilizes enhanced public-key searchable encryption (PKSE) to establish a matching environment that fulfills privacy protection needs, thereby facilitating consistency matching between supply and demand, all while preserving user privacy. Then, the PPMSE ensures that matching service can only occur within the designated platform by equipping the power data trading cloud platform with public and private keys. Additionally, by applying ciphertext policy attribute-based encryption (CP-ABE) to the data processing tasks of the buyer, the scheme enables the seller to decrypt and obtain what the buyer has done with the data after successful matching and meeting specific attributes. Ultimately, a comprehensive analysis and performance evaluation are provided, validating the feasibility and superiority of the proposed scheme.

A long-standing topic in artificial intelligence is the effective recognition of patterns from noisy images. In this regard, the recent data-driven paradigm considers 1) improving the representation robustness by adding noisy samples in training phase (i.e., data augmentation) or 2) pre-processing the noisy image by learning to solve the inverse problem (i.e., image denoising). However, such methods generally exhibit inefficient process and unstable result, limiting their practical applications. In this paper, we explore a non-learning paradigm that aims to derive robust representation directly from noisy images, without the denoising as pre-processing. Here, the noise-robust representation is designed as Fractional-order Moments in Radon space (FMR), with also beneficial properties of orthogonality and rotation invariance. Unlike earlier integer-order methods, our work is a more generic design taking such classical methods as special cases, and the introduced fractional-order parameter offers time-frequency analysis capability that is not available in classical methods. Formally, both implicit and explicit paths for constructing the FMR are discussed in detail. Extensive simulation experiments and robust visual applications are provided to demonstrate the uniqueness and usefulness of our FMR, especially for noise robustness, rotation invariance, and time-frequency discriminability.

Federated learning is a technique that enables multiple parties to collaboratively train a model without sharing raw private data, and it is ideal for smart healthcare. However, it raises new privacy concerns due to the risk of privacy-sensitive medical data leakage. It is not until recently that the privacy-preserving FL (PPFL) has been introduced as a solution to ensure the privacy of training processes. Unfortunately, most existing PPFL schemes are highly dependent on complex cryptographic mechanisms or fail to guarantee the accuracy of training models. Besides, there has been little research on the fairness of the payment procedure in the PPFL with incentive mechanisms. To address the above concerns, we first construct an efficient non-interactive designated decryptor function encryption (NDD-FE) scheme to protect the privacy of training data while maintaining high communication performance. We then propose a blockchain-based PPFL framework with fair payment for medical image detection, namely ESB-FL, by combining the NDD-FE and an elaborately designed blockchain. ESB-FL not only inherits the characteristics of the NDD-FE scheme, but it also ensures the interests of each participant. We finally conduct extensive security analysis and experiments to show that our new framework has enhanced security, good accuracy, and high efficiency.

With the development of big data technology, the power industry has also entered the data-driven intelligence era. Cloud computing-based smart grids give the power industry stronger capabilities in data analytics. Electricity load forecasting in the cloud helps smart grids allocate resources appropriately. However, the users' privacy is easily compromised in the load forecasting process with cloud computing. The electricity usage data collected by the system may contain sensitive information about the users, which could lead to serious privacy leakage. In order to solve the issues, we propose a novel privacy-preserving cloud-aided load forecasting scheme for the cloud computing-based smart grid. It contains a secure online training algorithm and an efficient real-time forecasting algorithm. Meanwhile, the two-party interaction security scheme is more suitable for real-world applications. Before being sent to the cloud server, the control center of the smart grids encrypts the data using homomorphic encryption. During the process of model training and forecasting, the data remains securely encrypted at all times to avoid the risk of data privacy breaches. Finally, security and experimental analyses show that our scheme effectively avoids privacy leakage while reducing resource consumption.

First-person view attention has been widely studied in computer science domain since 1990s while third-person view attention in natural scenarios begins to gain the intensive interest until 2015. This paper focuses on the problem of third-person view attention prediction in natural scenarios where a human freely performs daily activities without constraints. To handle the two insuffiencies of existing methods: (i) assuming some extra information (except for input images) are given in advance and (ii) ignoring the importance of human-scene interaction, this paper proposes a model with weak information dependency, which helps to alleviate annotation costs. In addition, a transformer-based human-scene interaction mechanism is proposed to explore the global and long-dependency contexts between the human and scene. The pipeline of the proposed model is firstly extracting human and scene features, then inferring human attention probability map by fusing human and scene features via a transformer-based network, and finally predicting human attention object based on human attention probability map and object detection. The experiments on two public datasets validate the effectiveness of our model.

Perceptual image encryption degrades image quality by selectively encrypting some key information of the plain images. The encrypted images are partially perceptible according to the security or quality requirements. Although several types of attacks have tried to infer privacy information from the encrypted images, they can only either extract statistical information or enhance image sketch. In this paper, we take one step further and fully recover the plain images from perceptually encrypted counterparts by designing a non-local attack network (NL-ANet). NL-ANet is composed of densely cascaded multiscale non-local modules (MSNL) and a hierarchical attention fusion module (HAFM). In particular, to better reconstruct encryption distortion, we introduce MSNL to capture powerful hierarchical features from different scales, and propose HAFM to adaptively aggregate and enhance informative hierarchical features for reconstruction. We also propose a new instantiation of the multi-head non-local block with channel attention (MHCA) to explore the long-range dependencies of global contextual information. Extensive experiments show that NL-ANet is encryption-agnostic and superior on different perceptual encryption schemes under different encryption strengths. NL-ANet also achieves better performance than state-of-the-art image restoration methods.

Data aggregation plays a crucial role in smart grid communication as it enables the collection of data in an energy-efficient manner. However, the widespread deployment of smart meters has raised significant concerns regarding the privacy of users' personal data. Therefore, in this paper, we present an efficient and privacy-preserving data aggregation and trust management scheme (PATM) for an IoT-enabled smart grid based on smart contract. Firstly, we propose a five-layer architecture for smart grid communication to support secure and efficient data aggregation and management. Under the architecture, the Boneh-Goh-Nissim cryptosystem with blind factor is improved to facilitate privacy protection. In addition, the tamper-evident nature of blockchain is utilized for effective data management. Our designs also enhance the resistance to differential attack and prevent privacy breaches during the aggregation process. Detailed security proof and theoretical analysis confirm that our PATM can satisfies the necessary security and privacy requirements while maintaining the required efficiency for smart grid operations. Furthermore, comparative experiments demonstrate that PATM outperforms other proposed work in terms of storage cost, computational complexity, and utility of differential privacy.

The increasing privacy concerns associated with cloud-assisted image retrieval have captured the attention of researchers. However, a significant number of current research endeavors encounter limitations, including suboptimal accuracy, inefficient retrieval, and a lack of effective result verification mechanisms. To address these limitations, we propose an adaptive verifiable privacy-preserving medical image retrieval (AVPMIR) scheme in the outsourced cloud. Specifically, we utilize the convolutional neural network (CNN) ResNet50 model to extract the feature of each medical image within the dataset of the medical institution, aiming to enhance retrieval accuracy. To enhance retrieval efficiency, we build an encryption searchable index based on a mini-batch$k$-means clustering algorithm. Furthermore, we present an index merging method in which multi-data owners build a different index tree according to different standards. To check the correctness of the returned results from the cloud server, we construct an adaptive verification framework for the obtained results based on chameleon hash and BLS signature. To provide strong security for the medical image datasets, we design an improved logistic chaotic mapping algorithm. The security analysis demonstrates that AVPMIR can defend various threat models. The experiment analysis further indicates that the AVPMIR can improve retrieval efficiency and demonstrate its practicability.

Group-based content push can be widely applied in integrated networks, where group key management is crucial for the push's security. Existing group key management methods mainly include symmetric group key agreement, broadcast encryption, asymmetric group key agreement, and attribute-based encryption. However, most of them do not consider user equipment (UE) identity privacy and unlinkability, cannot support flexibility and efficiency due to each UE maintaining group keys, and lack the trustworthiness of UE and group key management, which hinders the widespread adoption of group-based content push in trustless environments like integrated networks. In this paper, we investigate a novel decentralized group key management (DeGKM) scheme for group-based content push in integrated networks, where different operators manage pseudonyms and group keys across domains in a decentralized manner. In particular, our scheme adopts verifiable shuffling to establish a unified and trustworthy inter-domain pseudonym management approach that can preserve UE identity privacy and pseudonym unlinkability without relying on a trusted third party, and introduces a unified inter-domain group key management method based on Chinese remainder theorem and blockchain that significantly guarantees the flexibility, efficiency and trustworthiness. We formally prove the security of DeGKM and show its efficiency through simulations and comparisons with related works.

In cloud storage systems, secure deduplication plays a critical role in saving storage costs for the cloud server and ensuring data confidentiality for cloud users. Traditional secure deduplication schemes require users to encrypt their outsourced files using specific encryption algorithms that cannot provide semantic security. However, users are unable to directly benefit from the storage savings, as the relation between the actual storage cost and the offered prices remains not transparent. As a result, users may be unwilling to cooperate with the cloud by encrypting their data using semantically secure algorithms. Moreover, data integrity is a significant concern for cloud storage users. To address these issues, this paper proposes a novel transparent and secure deduplication scheme that supports integrity auditing. Compared to previous works, our design can verify the number of file owners and the integrity through one-time proof verification. It also protects the private contents of files and the privacy of file ownership from malicious users. Moreover, our scheme includes a batch auditing method to simultaneously verify the numbers of file owners and the integrity of multiple files. Theoretical analysis confirms the correctness and security of our scheme. Comparison results demonstrate its competing performance over previous solutions

Blockchain is extensively applied to many fields as an immutable distributed ledger. However, the immutability contradicts regulations such as the GDPR ruling “the right to be forgotten” of data. Besides, numerous emerging blockchain-based applications call for elastic data management. To erase some data, redactable blockchains are proposed for breaking the immutability in a controlled way. Unfortunately, the prior solutions may suffer from poor security and centralized control of the redaction privilege. They cannot support dynamic nodes, where the departure of participators will result in a single point of failure. This paper proposes a noveldynamic and decentralizedattribute-basedchameleonhash (DACH) to make blockchain history mutable, achieving asecurely anddynamicallyredactable blockchain (SDR-chain) in a decentralized setting. We first propose the formal definition, security models, and concrete construction of our DACH. Meanwhile, we design a delegation algorithm of DACH to support a dynamically changing committee, where participators can freely and securely leave and join the network. Then, the transactions of the SDR-chain are redacted by computing DACH collisions. The security is analyzed in the random oracle model. Finally, theoretical analysis and experimental evaluation demonstrate that our SDR-chain is superior to the prior solutions in terms of security and functionality.

Recent research findings suggest that machine learning models are highly susceptible to backdoor poisoning attacks. Backdoor poisoning attacks can be easily executed and achieve high success rates, as the model exhibits anomalous behavior even if a small quantity of malicious data is incorporated into the training dataset. In conventional backdoor defense technologies, fine-tuning is employed as an invasive method that involves adjusting the parameters of model neurons to eliminate backdoors in the attacked model. Nevertheless, this method poses a challenge as the same neurons are responsible for both the original and backdoor tasks, resulting in a decline in the accuracy of the original task during the fine-tuning process. In order to address this issue, we propose a non-invasive approach known as Anti-Backdoor Model (ABM), which does not involve modifying the parameters of the attacked model. ABM employs an external model to counteract the influence of the backdoor task on the attacked model, thereby achieving a balance between eliminating backdoors and preserving the accuracy of the original task. Specifically, our approach involves initially embedding a controllable backdoor in the dataset and leveraging the strong and weak relationships between backdoors to identify a highly concentrated poisoned dataset. Subsequently, we employ the standard training method to train the attacked model (the teacher model). Finally, we utilize this dataset with low volume to train an external model (the student model) that exclusively focuses on backdoors by means of knowledge distillation to counteract the backdoor task in the attacked model (the teacher model). In the experimental part, we assess the effectiveness of ABM by testing eight mainstream attacks on three standard public datasets. Experimental results reveal that ABM exhibits promising efficacy in eliminating the backdoor task while preserving the accuracy of the original task. Our source codes are open athttps://gitee.com/dugu1076/ABM.git.

Time series data analytics technology significantly benefits modern scientific research, especially in fields such as medical health, financial investment, and transportation. Unfortunately, privacy issues hinder people from handing over the data to a third party for various analytical tasks; because the data may reveal much more individual sensitive information, e.g., disease information from medical data, investment tendency from financial data, or the daily trajectory from transportation data. To break down this barrier, secure computation approaches have shown their importance in processing sensitive data, and have attracted much attention from the industry and research communities. However, when considering the case of secure time-series data analytics (e.g., DTW similarity), we are still far from achieving high efficiency due to high round complexity in communication or expensive computational complexity. We observe that DTW involves a lot of comparison operations and existing approaches in dealing with the comparison require higher communication costs. To this end, this paper studies secure DTW-based analytics with practical efficiency over time series data. Specifically, we propose the framework of secure time series data analytics (STDA) and formulate the problem of top-$k$query for outsourced time series data. Based on threshold Paillier encryption, we present a top-$k$query protocol utilizing the DTW distance as a metric and its security analysis, optimizations, and performance evaluation. The experimental results demonstrate that in a wide-area network with a 10 ms latency, our top-$k$approach outperforms the state-of-the-art by 3x times, while DTW calculation outperforms by 9x times. Correspondingly, the optimized$\mathcal {F}_{\text {DTW}}$achieves 17x times better, and optimized top-$k$achieves 4-10x times better.

With the growing popularity of cloud computing, sharing media data through the cloud has become a common practice. Due to high information redundancy, media data take up a significant amount of storage space. Moreover, similar media data may have the same visual effect, resulting in unnecessary duplication. Thus, it can greatly improve the cloud storage efficiency by performing deduplication to the similar media data stored on the cloud. However, data privacy is a growing concern in cloud-based service. In this paper, we present SimLESS, a secure deduplication system for similar data in cloud media sharing. SimLESS allows the cloud to perform deduplication over the encrypted similar media data of different distributors while protecting the confidentiality and ownership of the data. When uploading a media file, SimLESS allows the distributor to set a distance threshold, and the cloud performs deduplication only when there is a file on the cloud whose distance from the file being uploaded is smaller than the threshold. Additionally, we provide fine-grained access control for distributors to ensure that only authorized media consumers can access the data. Furthermore, our system prevents any distributor from claiming ownership of a media file using only the tag of a similar file. We formally analyze the security of SimLESS and implement a system prototype to evaluate its performance. Our experimental results demonstrate that the computation and communication costs of SimLESS are practically affordable.

In the context of drone swarms, achieving efficient group secure communication is a challenging problem, due to the inherent limitations imposed by the drones’ limited energy and constrained resources. Physical layer group key generation (PLGK) is a promising technology to enable efficient group security communication. However, most existing PLGK schemes struggle to adapt to the dynamic nature of drone swarms. To address this gap, this paper proposes a novel satellite cluster state information (SCSI)-based PLGK, which leverages signal status information from all visible navigation satellites to establish the group key. The presented method utilizes the regional similarity of SCSI as a random information source to generate group keys between different drones, and employs a novel updating framework based on a fuzzy generator and a hash chain to enhance key update and alignment robustness. The proposed scheme not only significantly reduces the overhead of group key generation also mitigates the issues of key loss and reconstruction. The security of the proposed scheme is validated through formal protocol security proof and security analysis against possible attacks. Finally, experiments with real-world drones demonstrate the efficiency and effectiveness of the SCSI-based PLGK.

There is an increasing tendency to fine-tune large-scale pre-trained language models (LMs) using small private datasets to improve their capability for downstream applications. In this paper, we systematically analyze the pre-train and then fine-tune the process of generative LMs and show that the fine-tuned LMs would leak sensitive keywords of the private datasets even without any prior knowledge of the downstream tasks. Specifically, we propose a novel and efficient keyword inference attack framework to accurately and maximally recover sensitive keywords. Owing to the fine-tuning process, pre-trained and fine-tuned models might respond differently to identical input prefixes. To identify potential sensitive sentences for training the fine-tuend LM, we introduce a contrast difference score that assesses the response variations between a pre-trained LM and its corresponding fine-tuned LM. Following this, we iteratively fine-tune the pre-trained model using these sensitive sentences to minimize the disparity between the target model and the pre-trained model, thereby maximizing the number of inferred sensitive keywords. We implement two types of keyword inference attacks (i.e., domain and private) according to our framework and conduct comprehensive experiments on three downstream applications to evaluate the performance. The experimental results demonstrate that our domain keyword inference attack achieves a precision of 85%, while our private keyword inference attack can extract highly sensitive personal information for a significant number of individuals (approximately 0.3% of all customers in the private fine-tuning dataset, which contains 40,000 pieces of personal information).

Partial Label Learning (PLL) is a typical weakly supervised learning task, which assumes each training instance is annotated with a set of candidate labels containing the ground-truth label. Recent PLL methods adopt identification-based disambiguation to alleviate the influence of false positive labels and achieve promising performance. However, they require all classes in the test set to have appeared in the training set, ignoring the fact that new classes will keep emerging in real applications. To address this issue, in this article, we focus on the problem of Partial Label Learning with Augmented Class (PLLAC), where one or more augmented classes are not visible in the training stage but appear in the inference stage. Specifically, we propose an unbiased risk estimator with theoretical guarantees for PLLAC, which estimates the distribution of augmented classes by differentiating the distribution of known classes from unlabeled data and can be equipped with arbitrary PLL loss functions. Besides, we provide a theoretical analysis of the estimation error bound of the estimator, which guarantees the convergence of the empirical risk minimizer to the true risk minimizer as the number of training data tends to infinity. Furthermore, we add a risk-penalty regularization term in the optimization objective to alleviate the influence of the over-fitting issue caused by negative empirical risk. Extensive experiments on benchmark, UCI, and real-world datasets demonstrate the effectiveness of the proposed approach.

Multiple-instance learning (MIL) is a significant weakly supervised learning problem, where the training data consist of bags containing multiple instances and bag-level labels. Most previous MIL research required fully labeled bags. However, collecting such data is challenging due to the labeling costs or privacy concerns. Fortunately, we can easily collect pairwise comparison information, indicating one bag is more likely to be positive than the other. Therefore, we investigate a novel MIL problem about learning a bag-level binary classifier only from pairwise comparison bags. To solve this problem, we display the data generation process and provide a baseline method to train an instance-level classifier based on unlabeled-unlabeled learning. To achieve better performance, we propose a convex formulation to train a bag-level classifier and give a generalization error bound. Comprehensive experiments show that both the baseline method and the convex formulation achieve satisfactory performance, while the convex formulation performs better. 1

Diffusion-weighted imaging (DWI) has been extensively explored in guiding the clinic management of patients with breast cancer. However, due to the limited resolution, accurately characterizing tumors using DWI and the corresponding apparent diffusion coefficient (ADC) is still a challenging problem. In this paper, we aim to address the issue of super-resolution (SR) of ADC images and evaluate the clinical utility of SR-ADC images through radiomics analysis. To this end, we propose a novel double transformer-based network (DTformer) to enhance the resolution of ADC images. More specifically, we propose a symmetric U-shaped encoder-decoder network with two different types of transformer blocks, named as UTNet, to extract deep features for super-resolution. The basic backbone of UTNet is composed of a locally-enhanced Swin transformer block (LeSwin-T) and a convolutional transformer block (Conv-T), which are responsible for capturing long-range dependencies and local spatial information, respectively. Additionally, we introduce a residual upsampling network (RUpNet) to expand image resolution by leveraging initial residual information from the original low-resolution (LR) images. Extensive experiments show that DTformer achieves superior SR performance. Moreover, radiomics analysis reveals that improving the resolution of ADC images is beneficial for tumor characteristic prediction, such as histological grade and human epidermal growth factor receptor 2 (HER2) status.

Images are composed of “things” (i.e., structured objects) and “stuff” (i.e., textured surfaces), which have completely different effects on the human visual system (HVS). A good image quality assessment (IQA) method should fully consider the visual salience effects of image structures and the masking effects of image textures. In this article, we propose a perceptual quality analysis model using structure separation and high-order moments (SSHMPQA) in the deep domain. First, we use a total variation (TV) model to separate the perceptual structures in images from their deep feature maps, thereby maintaining meaningful object shapes with texture suppression and defining perceptual structure-aware distances in the deep domain. Then, we use the first- to fourth-order moments to calculate the mean, skewness and kurtosis of the probability distributions of the deep features. On this basis, we define a perceptual texture-aware distance in the deep domain. We then formulate the final model by solving a well-defined perceptual optimization problem. The proposed SSHMPQA model has good interpretability and is data-driven; moreover, the model does not require a complex and long training process because the optimization problem is convex and has an exact analytical solution. To verify the effectiveness of our model, comprehensive experiments are conducted. The experimental results show that the proposed model is superior to other state-of-the-art traditional and deep learning-based full-reference (FR) IQA methods.

Recognizing violence in videos is significant for the automatic identification and assessment of violence content to restrict the access to violence for specific audiences such as children. Existing methods focus on violence detection, which is only able to recognize whether there exists violence or not. Differently, this paper handles the problem of video violence rating, which provides a more granular classification of violence levels. However, there is no publicly available database for video violence rating since it asks for fine-grained violence level annotations. Therefore, this paper introduces a large-scale violence rating database, which will be publicly released. Furthermore, we propose a multimodal violence rating model. Different from existing models, our model makes use of the token-based interaction and contrastive learning techniques. The token-based interaction is able to strengthen the feature representations and make full use of multimodal features. The contrastive learning can improve the performance of the model. To evaluate our model, a wide range of experiments are conducted, and experiment results show that our model outperforms existing methods. We have made the VioShot database publicly available for downloading athttps://sites.google.com/site/xiangtaooo/.

Enhancing low-light image visibility is a critical task in computer vision since it helps to improve input for high-level algorithms. High-quality images typically have clear structural information. In previous studies, due to the lack of proper structural guidance, restored images had some problems, such as unclear structural areas and overexposed or underexposed local areas. To address the above problems, in this paper, we introduce a coefficient of variation (COV) with excellent performance in maintaining structural information, and then we propose a low-light image enhancement method that utilizes the COV to extract structural information from images. First, we apply a traditional retinex model to estimate both reflectance and illumination. Second, we use the COV to indicate the degree of dispersion of the input sample, which enables us to obtain a robust structure-distinguishing weight map for low-light images. The weight map is adaptively divided to obtain a structural weight map, which is then used to enhance the gradient image. This process is applied before the reflectance layer of the retinex model. Finally, the result is obtained by using the block coordinate descent method. According to extensive experiments, outstanding results can be achieved by our proposed method in terms of both subjective and objective evaluation metrics in comparison with other state-of-the-art methods. The source code is available at our website.

High-performance traffic flow prediction models provide reliable future road information and optimize traffic navigation systems. However, the traffic data used for model learning contains lots of private information, and the existing privacy-preserving strategies always reduce the accuracy of prediction models. Besides, an effective traffic flow prediction model needs to be continuously and rapidly updated to adapt to dynamic changes in the traffic environment. Thus, we propose a Secure and Efficient Continuous Learning Model (SE-CLM) based on broad learning, spatial correlation, and adaptive sampling processing techniques to realize accurate and efficient traffic flow prediction under strong privacy protection. Specifically, SE-CLM is constructed on the broad network architecture to enable fast and continuous model training. This model is trained on a cloud server by combining the spatial correlation of traffic flows, to achieve accurate traffic flow prediction. Besides, an adaptive sampling strategy is designed to further improve the prediction accuracy of the model under the protection with differential privacy (DP), where the budget allocation for DP is optimized by adaptively sampling traffic flows with different timestamps for noise perturbation processing. Furthermore, the experimental simulations are conducted in real vehicular mobility datasets. The experimental results show that the designed spatial-based SE-CLM achieve more accurate and efficient traffic flow prediction than those of the other existing schemes. The adaptive sampling strategy not only significantly reduces the DP-noise added in traffic flows but also a 20% reduction in communication volume compared to other strategies. Finally, the security analysis also verifies that SE-CLM satisfies w-event ε-DP.

RGBT tracking is gaining popularity due to its ability to provide effective tracking results in a variety of weather conditions. However, feature specificity and complementarity have not been fully used in existing models that directly fuse the correlation filtering response, which leads to poor tracker performance. In this article, we propose correlation filters with adaptive modality weight and cross-modality learning (AWCM) ability to solve multimodality tracking tasks. First, we use weighted activation to fuse thermal infrared and visible modalities, and the fusion modality is used as an auxiliary modality to suppress noise and increase the learning ability of shared modal features. Second, we design modal weights through average peak-to-correlation energy coefficients to improve model reliability. Third, we propose consistency in using the fusion modality as an intermediate variable for joint learning consistency, thereby increasing tracker robustness via interactive cross-modal learning. Finally, we use the alternating direction method of multipliers algorithm to produce a closed solution and conduct extensive experiments on the RGBT234, VOT-TIR2019, and GTOT tracking benchmark datasets to demonstrate the superior performance of the proposed AWCM against compared to existing tracking algorithms. The code developed in this study is available at the following website. 1

While smart grid significantly facilitates energy efficiency by using users’ power consumption data, it poses privacy leakage risk for user personal behaviors. Differential privacy (DP) has emerged as a promising solution to address this issue. However, existing approaches suffer from severe data utility degradation due to the intensive noise introduced by DP. Additionally, some of these methods are vulnerable to security attacks. To bridge this gap, in this paper, we propose ABDP (accuratebilling-enableddifferentiallyprivate), a mechanism that achieves high-strength DP while ensuring accurate aggregation and billing operations without compromising security. In particular, we propose aggregated and individual noise cancellation algorithms to counteract the negative effects of noise on data utility. Specifically, our ABDP ensures precise aggregation and accurate billing calculations for the power grid and individual users, respectively Furthermore, we present a Blockchain smart contract exploiting the pseudo random function to enforce a fair and secure data reporting process. Theoretical analysis is provided to evaluate the privacy and security guarantees of ABDP. Experimental results on real-world datasets, namely NERL-DATA and REDD, demonstrate that ABDP achieves error-free aggregation and billing calculation, offers arbitrary intensity privacy protection against non-intrusive load monitoring and filtering attacks, and outperforms existing state-of-the-art approaches.

With the widespread use of cloud storage technology by individuals and organizations, data providers usually send their data to cloud for storage to reduce memory pressure, and allow the users to retrieve these data, which has become the trend of rapid data retrieval. To guarantee the data confidentiality, several research works have been developed on encrypted cloud data for ranked multi-keyword retrieval. Nevertheless, most of these schemes are disabled since they cannot resist keyword guessing attacks. Moreover, the ranked top-$K$search results obtained by the subscriber from the encrypted cloud data are inaccurate. To overcome these drawbacks, we design a novel and efficient privacy-preserving ranked multi-keyword retrieval scheme (named as PRMKR) in this paper. With PRMKR, the data and the inverted indexes which belong to the data provider can be securely transferred to the cloud server. In addition, a registered subscriber can request accurate retrieval services without compromising his/her trapdoor information to the cloud server. Specifically, we design an encryption searchable plugin-in server and lower dimensional inverted indexesvector for data owners, which can further guarantee data confidentiality of the data owner and improve search efficiency, respectively. Our rigorous security proof demonstrates that PRMKR can withstand keyword guessing attacks. Finally, experimental evaluations confirm that PRMKR has decent computational and communication efficiency.

Point cloud registration is a basic task in computer vision and computer graphics. Recently, deep learning-based end-to-end methods have made great progress in this field. One of the challenges of these methods is to deal with partial-to-partial registration tasks. In this work, we propose a novel end-to-end framework called MCLNet that makes full use of multi-level consistency for point cloud registration. First, the point-level consistency is exploited to prune points located outside overlapping regions. Second, we propose a multi-scale attention module to perform consistency learning at the correspondence-level for obtaining reliable correspondences. To further improve the accuracy of our method, we propose a novel scheme to estimate the transformation based on geometric consistency between correspondences. Compared to baseline methods, experimental results show that our method performs well on smaller-scale data, especially with exact matches. The reference time and memory footprint of our method are relatively balanced, which is more beneficial for practical applications.

In contrast to the standard learning paradigm where all classes can be observed in training data, learning with augmented classes (LAC) tackles the problem where augmented classes unobserved in the training data may emerge in the test phase. Previous research showed that given unlabeled data, an unbiased risk estimator (URE) can be derived, which can be minimized for LAC with theoretical guarantees. However, this URE is only restricted to the specific type of one-versus-rest loss functions for multi-class classification, making it not flexible enough when the loss needs to be changed with the dataset in practice. In this paper, we propose a generalized URE that can be equipped with arbitrary loss functions while maintaining the theoretical guarantees, given unlabeled data for LAC. To alleviate the issue of negative empirical risk commonly encountered by previous studies, we further propose a novel risk-penalty regularization term. Experiments demonstrate the effectiveness of our proposed method.

The problem of predicting driver attention from the driving perspective is gaining increasing research focus due to its remarkable significance for autonomous driving and assisted driving systems. The driving experience is extremely important for safe driving, a skilled driver is able to effortlessly predict oncoming danger (before it becomes salient) based on the driving experience and quickly pay attention to the corresponding zones. However, the nonobjective driving experience is difficult to model, so a mechanism simulating the driver experience accumulation procedure is absent in existing methods, and the current methods usually follow the technique line of saliency prediction methods to predict driver attention. In this paper, we propose a FeedBack Loop Network (FBLNet), which attempts to model the driving experience accumulation procedure. By over-and-over iterations, FBLNet generates the incremental knowledge that carries rich historically-accumulative and long-term temporal information. The incremental knowledge in our model is like the driving experience of humans. Under the guidance of the incremental knowledge, our model fuses the CNN feature and Transformer feature that are extracted from the input image to predict driver attention. Our model exhibits a solid advantage over existing methods, achieving an outstanding performance improvement on two driver attention benchmark datasets.

Visual Question Answering (VQA) is the vision-language task of answering text-based questions presented in an image and has been advanced by the remarkable success of multimodal deep networks. Similar to unimodal networks, multimodal VQA models are also vulnerable to adversarial examples, which raises severe threats to the corresponding applications. Although several adversarial training methods have been proposed, most of them focus on improving the generalization ability of VQA models on clean samples instead of mitigating the adversarial attacks. In this paper, we systemically analyze the core structure of multimodal VQA networks and propose a novel adversarial training algorithm to mitigate adversarial attacks on VQA models. Specifically, our key component is a regularization term with our carefully designed Contrastive Fusion Representation (CFR), which can reduce the sensitivity of VQA models to adversarial perturbations of both the vision and language inputs. We further enhance the adversarial training with augmented CFRs. Comprehensive experimental results show that our method can mitigate adversarial attacks as well as preserve the generalization ability on clean samples under various system settings and outperforms other defense methods.

Collaborative learning has gained great popularity due to its benefit of data privacy protection: participants can jointly train a Deep Learning model without sharing their training sets. However, recent works discovered that an adversary can fully recover the sensitive training samples from the shared gradients. Such reconstruction attacks pose severe threats to collaborative learning. Hence, effective mitigation solutions are urgently desired. In this paper, we systematically analyze existing reconstruction attacks and propose to leverage data augmentation to defeat these attacks: by preprocessing sensitive images with carefully-selected transformation policies, it becomes infeasible for the adversary to extract training samples from the corresponding gradients. We first design two new metrics to quantify the impacts of transformations on data privacy and model usability. With the two metrics, we design a novel search method to automatically discover qualified policies from a given data augmentation library. Our defense method can be further combined with existing collaborative training systems without modifying the training protocols. We conduct comprehensive experiments on various system settings. Evaluation results demonstrate that the policies discovered by our method can defeat state-of-the-art reconstruction attacks in collaborative learning, with high efficiency and negligible impact on the model performance.

With the development of wireless communications and the pervasiveness of location-aware mobile electronic devices, location-based services (LBS) which can provide a convenient lifestyle for people have attracted considerable interest recently. However, there still exists the privacy disclosure problem of LBS today. To solve this problem, in this article, we present a novel privacy-preserving LBS search scheme in outsourced cloud. In the proposed LBS search scheme, the LBS providers data are first outsourced to the cloud server in an encrypted method. Then, a registered user constructs a query model to obtain accurate LBS query results without divulging his/her location information and query attribute to the LBS provider and the cloud server. Specifically, based on the designed matrix encryption technology, the LBS search scheme can achieve privacy preservation of users query and confidentiality of LBS data in the outsourced cloud server. Through security analysis, we show that our scheme can resist various known security threats. The experimental results further show that our LBS search scheme greatly reduces the communication overhead and provides convenient search experience to the users.

Following the success of cloud computing, it has been shown its importance to realize various access control models in the cloud storage setting. Chinese Wall is a traditional access control model in business for solving the conflict of interest (CoI) problem, and it would be very interesting to achieve conflict-resistant in cloud storage system. However, the access control model does not ensure the privacy of users, and it may reveal the user's interest, investment tendency, etc. Therefore, it raises a big challenge to implement the Chinese Wall without compromising the user's privacy. In this paper, we focus on the Chinese Wall model and apply it to the cloud storage while protecting the access patterns of users. Specifically, we first formulate the tree-based Chinese Wall access control and then propose the Chinese Wall Protocol (called C-Wall). We prove that our C-Wall not only realizes the conflict-resistant but also protects the user's privacy with universally composable security. Besides, we also apply C-Wall to privacy-preserving cloud storage and propose the C2-Wall, which not only maintains C-Wall's features, but also ensures the sensitive files from being touched by "honest-but-curious" cloud servers. Furthermore, we evaluate our C2-Wall by theoretical analysis and experimental validation. Experimental results show its effectiveness and efficiency for practical deployment.

As one of the key technologies to enable the internet of things (IoT), cloud computing plays a significant role in providing huge computing and storage facilities for large-scale data. Though cloud computing brings great advantages, new issues emerge, such as data security breach and privacy disclosure. In this paper, we introduce a novel secure and privacy-preserving outsourcing computing scheme (hereafter referred to as SPOCS) to tackle this issue. In SPOCS, the effective use of Intel SGX, one of the trusted execution environment (TEE), ensures the confidence and integrity of sensitive data in cloud computing and prevents data loss from causing privacy disclosure. In order to keep malicious cloud service providers (CSPs) from illegally tampering with the outsourcing results, blockchain is employed to ensure the data immutability. Significantly, our proposed scheme achieves anonymity and traceability. In the outsourcing process, smart contracts are applied to make the whole process fully automated without any human involvement. Finally, the security of the proposed scheme is analyzed in terms of its resistance to different attacks. The experiments indicate that our scheme is effective and efficient.

Vision-based practical applications, such as consumer photography and automated driving systems, greatly rely on enhancing the visibility of images captured in night-time environments. For this reason, various image enhancement algorithms (EHAs) have been proposed. However, little attention has been given to the quality evaluation of enhanced night-time images. In this paper, we conduct the first dedicated exploration of the subjective and objective quality evaluation of enhanced night-time images. First, we build an enhanced night-time image quality (EHNQ) database, which is the largest of its kind so far. It includes 1,500 enhanced images generated from 100 real night-time images using 15 different EHAs. Subsequently, we perform a subjective quality evaluation and obtain subjective quality scores on the EHNQ database. Thereafter, we present an objective blind quality index for enhanced night-time images (BEHN). Enhanced night-time images usually suffer from inappropriate brightness and contrast, deformed structure, and unnatural colorfulness. In BEHN, we capture perceptual features that are highly relevant to these three types of corruptions, and we design an ensemble training strategy to map the extracted features into the quality score. Finally, we conduct extensive experiments on EHNQ and EAQA databases. The experimental and analysis results validate the performance of the proposed BEHN compared with the state-of-the-art approaches. Our EHNQ database is publicly available for download athttps://sites.google.com/site/xiangtaooo/.

The rapid development of IoT (Internet of Things) brings great convenience to people through the utilization of IoT applications, but also brings huge security challenges. Existing IoT security breaches show that many IoT devices have authentication flaws. Although many IoT authentication schemes were proposed, they are not applicable to recent smart IoT applications covering IoT device, back-end sever, and user-end mobile applications. To build the first line of defense for trending IoT systems, this paper proposes a new authentication scheme. The proposed scheme first models the entire life cycle of the IoT device for real-world scenarios of smart IoT systems, which contains factory manufacturing, daily usage, and system resetting. For each stage in the life cycle, the proposed scheme employs efficient symmetric key mechanisms to achieve the authentication between IoT device, back-end server, and mobile application. The proposed scheme supports both server-free local area network communication and sever-involved remote public area communication. Formal security verification shows that the proposed scheme resists existing attacks. The open-source experimental evaluations also show that the proposed scheme is efficient and promising for practical usage.

Cloud-assisted electronic health records (EHRs) provide convenient medical services for patients by storing and analyzing medical data records in the cloud, but searching for sensitive data (e.g., identity, medical history) in the cloud conflicts with privacy protection requirements. Searchable encryption (SE) is a good cryptographic primitive for solving this conflict, which allows the user to store their encrypted data in the cloud and search them later in encrypted domain. However, the direct applications of most existing SE schemes in cloud-assisted EHRs may result in challenges, for example in terms of functionality, security and efficiency. In this paper, we propose BPVSE, a new verifiable and dynamic SE scheme for cloud-assisted EHR. BPVSE has the following advantages over existing approaches. First, leveraging blockchain and hash-proof chain, BPVSE allows the user to publicly verify the search result returned by the cloud without a trusted authority. Second, BPVSE supports dynamic datasets with forward and backward security, using our newly designed new hidden data structure. Third, BPVSE enables the user to launch parallel search with efficient encryption. We formally prove the security of the proposed BPVSE, and also conduct theoretical comparison and experimental evaluation to show its superiority of functionality, security, and efficiency.

Industrial Internet of Things (IIoT) is a new paradigm for building intelligent industrial control systems, and how to establish a secure channel in IIoT for machine-to-machine (M2M) communication is a critical problem because the devices in IIoT suffer from various attacks and may leak confidential information. Traditional authenticated and confidential channel establishment (ACCE) protocols neither apply for resource-constrained IIoT devices nor satisfy leakage resilience. In this paper, we introduce a new security notion: historical data based multi-factor ACCE (HMACCE) to address this issue and propose two HMACCE protocols. Our HMACCE protocols use three authentication factors, i.e., a symmetric secret key, historical data, and a set of secret tags associated with the historical data, to establish a secure communication channel between the client and the server. The key idea is to use the secret key managed by an IIoT edge device to quickly verify the relationship between the historical data and its associated tags stored on the server. Our HMACCE has the following remarkable features. First, it is lightweight and tailored for resource-constrained IIoT devices. Second, it is bounded historical tag leakage resilience, which means that if a small portion of the secret tags is leaked to an adversary, it will not affect its security with an overwhelming probability. Moreover, as a security enhancement service, our HMACCE can be easily integrated with legacy IIoT devices by running simple authenticated key exchange protocols.

Deep hiding, embedding images with others using deep neural networks, has demonstrated impressive efficacy in increasing the message capacity and robustness of secret sharing. In this paper, we challenge the robustness of existing deep hiding schemes by preventing the recovery of secret images, building on our in-depth study of state-of-the-art deep hiding schemes and their vulnerabilities. Leveraging our analysis, we first propose a simple box-free removal attack on deep hiding that does not require any prior knowledge of the deep hiding schemes. To improve the removal performance on the deep hiding schemes that may be enhanced by adversarial training, we further design a more powerful removal attack, efficient box-free removal attack (EBRA), which employs image inpainting techniques to remove secret images from container images. In addition, to ensure the effectiveness of our attack and preserve the fidelity of the processed container images, we design an erasing phase based on the locality of deep hiding to remove secret information and then make full use of the visual information of container images to repair the erased visual content. Extensive evaluations show our method can completely remove secret images from container images with negligible impact on the quality of container images.

The number of images produced by people everyday is rapidly increasing in recent years and their local storage space may be not big enough for storing all these images. As a result, people are currently accustomed to uploading images to cloud platforms, which raises privacy concerns. Traditional image encryption is a way to protect image privacy without preserving visual usability, so that image owners fail to conveniently browse and manage their images stored in the cloud. Hence some visual privacy protection schemes were proposed to balance image privacy and usability, while many of them are irreversible. Recently, a novel reversible technology, called Thumbnail-Preserving Encryption (TPE), has been a hot topic. However, existing TPE schemes are either inefficient, or cannot perfectly restore the original image and meanwhile achieve Nonce-Respecting (NR) security. In view of this, we propose a reversible framework for efficient and secure visual privacy protection, which tunably preserves image visual usability for image owners with the idea of data hiding. In the framework, the original image is firstly divided into several regions by our proposed region division methods and one of the regions is vacated by data hiding. Then, the vacated region is utilized to preserve the original thumbnail by pixel adjustment after image encryption. Finally, pixels in each sub block are permuted for security. According to our theoretical analysis, the above processes are completely reversible and the processed image achieves NR security. Furthermore, we conduct extensive experiments, including recognition by various application programming interfaces, user surveys, and efficiency comparison, to demonstrate that our framework is efficient and strikes a good balance between privacy and usability.

Adversarial attacks have threatened the application of deep neural networks in security-sensitive scenarios. Most existing black-box attacks fool the target model by interacting with it many times and producing global perturbations. However, all pixels are not equally crucial to the target model; thus, indiscriminately treating all pixels will increase query overhead inevitably. In addition, existing black-box attacks take clean samples as start points, which also limits query efficiency. In this article, we propose a novel black-box attack framework, constructed on a strategy of dual transferability (DT), to perturb the discriminative areas of clean examples within limited queries. The first kind of transferability is the transferability of model interpretations. Based on this property, we identify the discriminative areas of clean samples for generating local perturbations. The second is the transferability of adversarial examples, which helps us to produce local pre-perturbations for further improving query efficiency. We achieve the two kinds of transferability through an independent auxiliary model and do not incur extra query overhead. After identifying discriminative areas and generating pre-perturbations, we use the pre-perturbed samples as better start points and further perturb them locally in a black-box manner to search the corresponding adversarial examples. The DT strategy is general; thus, the proposed framework can be applied to different types of black-box attacks. We conduct extensive experiments to show that, under various system settings, our framework can significantly improve the query efficiency of existing black-box attacks and attack success rates.

Inmultiple-instance learning(MIL), each training example is represented by a bag of instances. A training bag is either negative if it contains no positive instances or positive if it has at least one positive instance. Previous MIL methods generally assume that training bags are fully labeled. However, the exact labels of training examples may not be accessible, due to security, confidentiality, and privacy concerns. Fortunately, it could be easier for us to access the pairwise similarity between two bags (indicating whether two bags share the same label or not) and unlabeled bags, as we do not need to know the underlying label of each bag. In this paper, we provide the first attempt to investigate MIL from only similar-dissimilar-unlabeled bags. To solve this new MIL problem, we first propose a strong baseline method that trains an instance-level classifier by employing an unlabeled-unlabeled learning strategy. Then, we also propose to train a bag-level classifier based on a convex formulation and theoretically derive a generalization error bound for this method. Comprehensive experimental results show that our instance-level classifier works well, while our bag-level classifier even has better performance.

Various image enhancement algorithms are adopted to improve underwater images that often suffer from visual distortions. It is critical to assess the output quality of underwater images undergoing enhancement algorithms, and use the results to optimise underwater imaging systems. In our previous study, we created a benchmark for quality assessment of underwater image enhancement via subjective experiments. Building on the benchmark, this paper proposes a new objective metric that can automatically assess the output quality of image enhancement, namely UWEQM. By characterising specific underwater physics and relevant properties of the human visual system, image quality attributes are computed and combined to yield an overall metric. Experimental results show that the proposed UWEQM metric yields good performance in predicting image quality as perceived by human subjects.

Research on image dehazing has made the need for a suitable dehazed image quality assessment (DIQA) method even more urgent. The performance of existing DIQA methods heavily relies on handcrafted haze-related features. Since hazy images with uneven haze density distributions will result in uneven quality distributions after dehazing, the manually extracted feature expression is neither accurate nor robust. In this paper, we design a deep CNN-based DIQA method without a handcrafted feature requirement. Specifically, we propose a blind dehazed image quality assessment model (BDQM), which consists of three components: image preprocessing, a haze-related feature extraction network (HFNet), and an improved regression network (IRNet). In HFNet, we design a perceptual information enhancement (PIE) module to learn powerful feature representations and enhance network capability according to channel attention, multiscale convolution and residual concatenation. IRNet aims to aggregate all patch information for the quality prediction of the whole image, where the effect of inhomogeneous distortion from the dehazing procedure is attenuated via a specifically designed patch attention (PA) mechanism. Experimental results on benchmark datasets demonstrate the effectiveness and superiority of the proposed network architecture over state-of-the-art methods.

Medical image fusion generates a fused image containing multiple features extracted from different source images, and it is of great help in clinical analysis and diagnosis. However, training a deep learning model for image fusion usually requires enormous computing power, especially for large volumes of medical data. Meanwhile, the privacy of images is also a critical issue. In this article, we propose a privacy-preserving blockchain-based medical image fusion (BMIF) framework. First, to ensure fusion performance, we design a new medical image fusion model based on convolutional neural network and Inception network and integrate the proposed model into the consensus process of blockchain. Next, to save computing power of blockchain, we design a consensus mechanism by requesting consensus nodes to train the fusion model instead of calculating useless hash values in traditional blockchain. Then, to protect data privacy, we further present an efficient homomorphic encryption to realize the training of fusion model on encrypted medical data. Finally, we conduct theoretical analysis and extensive experiments on public datasets to evaluate the feasibility and the performance of our proposed BMIF. The results exhibit that BMIF is efficient and secure, and our medical image fusion network performs better than state-of-the-art approaches.

This article proposes a frequency-based structure and texture decomposition model in a Retinex-based framework for low-light image enhancement and noise suppression. First, we utilize the total variation-based noise estimation to decompose the observed image into low-frequency and high-frequency components. Second, we use a Gaussian kernel for noise suppression in the high-frequency layer. Third, we propose a frequency-based structure and texture decomposition method to achieve low-light enhancement. We extract texture and structure priors by using the high-frequency layer and a low-frequency layer, respectively. We present an optimization problem and solve it with the augmented Lagrange multiplier to generate a balance between structure and texture in the reflectance map. Our experimental results reveal that the proposed method can achieve superior performance in naturalness preservation and detail retention compared with state-of-the-art algorithms for low-light image enhancement. Our code is available on the following website. 1

Distributed fog computing has received increasing attention recently and fog-assisted cloud storage can provide a real-time service to collect and manage large-scale data for the applications of Internet of Things. Encrypted data deduplication over cloud storage can significantly save storage space of the cloud server while protecting the confidentiality of the outsourced data. Previous encrypted data deduplication schemes are mostly designed for traditional cloud storage with a two-layer architecture and cannot be applied to the emerging fog-assisted cloud storage that has a more complex three-layer architecture (i.e., cloud server, fog node and endpoint device). In this paper, we design, analyze and implement FCDedup, a new encrypted data deduplication scheme for fog-assisted cloud storage. FCDedup is a two-level deduplication system that enables each fog node to detect duplicated encrypted data uploaded by different endpoint devices, as well as enables cloud server to detect duplicated encrypted data from different fog nodes. By doing so, FCDedup can achieve both intra-deduplication within a single data owner and inter-deduplication across different data owners. FCDedup is also designed to prevent cloud server and fog nodes launching the brute-force attacks, and to guarantee the reliability of files downloaded from the cloud. Formal analysis is provided to justify its deduplication correctness and security. Besides, we implement a prototype of FCDedup using Alibaba Cloud as backend storage. Our evaluations demonstrate that FCDedup is completely compatible with existing cloud storage systems and achieves modest performance overhead.

Next-generation wireless technology and machine-to-machine technology can provide the ability to connect and share data at any time among IoT smart devices. However, the traditional centralized data sharing/trading mechanism lacks trust guarantee and cannot satisfy the real-time requirement. Distributed systems, especially blockchain, provide us with promising solutions. In this paper, we propose a blockchain based non-repudiation scheme for IoT data trading to resolve the credibility and real-time limits. The proposed scheme has two parts, i.e., a trading scheme and an arbitration scheme. The trading scheme employs a divide-and-conquer method and two commitment methods to support efficient IoT data trading, which runs in a two-round manner. The arbitration scheme first leverages a smart contract to solve disputes on-chain in real time. In case of on-chain arbitration dissatisfaction, the arbitration scheme also employs an off-line arbitration to make a final resolution. Short-term and long-term analysis show that the proposed scheme enforces non-repudiation among the data trading parties and runs efficiently for rational data owners and buyers. We implemented the proposed scheme. Experimental results confirm that the proposed scheme has an orders-of-magnitude performance speedup than the state-of-the-art scheme.

Deep neural networks (DNNs) have shown their powerful capability in scene text editing (STE). With carefully designed DNNs, one can alter texts in a source image with other ones while maintaining their realistic look. However, such editing tools provide a great convenience for criminals to falsify documents or modify texts without authorization. In this paper, we propose to actively defeat text editing attacks by designing invisible "armors" for texts in the scene. We turn the adversarial vulnerability of DNN-based STE into strength and design local perturbations (i.e., "armors") specifically for texts using an optimized normalization strategy. Such local perturbations can effectively mislead STE attacks without affecting the perceptibility of scene background. To strengthen our defense capabilities, we systemically analyze and model STE attacks and provide a precise defense method to defeat attacks on different editing stages. We conduct both subjective and objective experiments to show the superior of our optimized local adversarial perturbation against state-of-the-art STE attacks. We also evaluate the portrait and landscape transferability of our perturbations.

Modern deep neural networks are highly vulnerable to adversarial examples, which attracts more and more researchers' attention to craft powerful adversarial examples. Most of these generation algorithms create global perturbations that would affect the visual quality of adversarial examples. To mitigate such drawbacks, some attacks attempt to generate local perturbations. However, existing local adversarial attacks are time-consuming and the generated adversarial examples are still distinguishable from clean images. In this paper, we propose a novel efficient local adversarial attack (ELAA) using model interpreters to generate severe local perturbations and improve the imperceptibly of the generated adversarial examples. Specifically, we take advantage of model interpretation methods to search the discriminative regions of clean images. Then, we generate local adversarial examples by adding masks to original clean images. We also propose a new optimization method to reduce the redundancy of local perturbations. Through extensive experiments, we show our ELAA can maintain a high attack ability while preserving the visual quality of clean images. Experimental results also demonstrate our local attack outperforms state-of-the-art local attack methods under various system settings.

Physical-layer fingerprinting is a promising technique to identify Internet of Things (IoT) devices. In this article, we investigate a new radio-frequency (RF) fingerprinting based on the distinctive signal-to-noise-ratio (SNR) trace in the sector-level sweep (SLS) procedure of 5G IEEE 802.11ad devices. This SLS SNR trace-based fingerprinting can directly apply to off-the-shelf devices without any extra hardware requirements and be independent of the wireless channel and environment. To tackle the impact of orientation on the RF fingerprinting, we propose a novel fingerprinting framework, involving correlation analysis, surface fitting, curve pursuing, and binary classification, named the CSCB framework. Using this framework, the proposed SLS SNR trace-based fingerprinting can achieve device authentication at any orientation with one receiver under line-of-sight (LOS) or non-LOS (NLOS) scenarios. We conduct proof-of-concept experiments using off-the-shelf IEEE 802.11ad devices (Talon AD7200 and MG360 WiGig) to evaluate the performance of the proposed fingerprinting schemes. Experimental results show the effectiveness of the proposed fingerprinting schemes where the verification accuracy of the proposed scheme can reach 99% with only 200 training samples.

Image classification is a fundamental artificial intelligence task that labels images into one of some predefined classes. However, training complex image classification models requires a large amount of computation resources and data in order to reach state-of-the-art performance. This demand drives the growth of distributed deep learning, where multiple agents cooperatively train global models with their individual datasets. Among such learning systems, decentralized learning is particularly attractive, as it can improve the efficiency and fault tolerance by eliminating the centralized parameter server, which could be the single point of failure or performance bottleneck. Although the agents do not need to disclose their training image samples, they exchange parameters with each other at each iteration, which can put them at the risk of data privacy leakage. Past works demonstrated the possibility of recovering training images from the exchanged parameters. One common defense direction is to adopt Differential Privacy (DP) to secure the optimization algorithms such as Stochastic Gradient Descent (SGD). Those DP-based methods mainly focus on standalone systems, or centralized distributed learning. How to enforce and optimize DP protection in decentralized learning systems is unknown and challenging, due to their complex communication topologies and distinct learning characteristics. In this paper, we design TOP- DP, a novel solution to optimize the differential privacy protection of decentralized image classification systems. The key insight of our solution is to leverage the unique features of decentralized communication topologies to reduce the noise scale and improve the model usability. (1) We enhance the DP-SGD algorithm with thistopology-awarenoise reduction strategy, and integrate the time-aware noise decay technique. (2) We design two novel learning protocols (synchronous and asynchronous) to protect systems with different network connectivities and topologies. We formally analyze and prove the DP requirement of our proposed solutions. Experimental evaluations demonstrate that our solution achieves a better trade-off between usability and privacy than prior works. To the best of our knowledge, this is the first DP optimization work from the perspective of network topologies.

Decentralized learning has gained great popularity to improve learning efficiency and preserve data privacy. Each computing node makes equal contribution to collaboratively learn a Deep Learning model. The elimination of centralized Parameter Servers (PS) can effectively address many issues such as privacy, performance bottleneck and single-point-failure. However, how to achieve Byzantine Fault Tolerance in decentralized learning systems is rarely explored, although this problem has been extensively studied in centralized systems. In this paper, we present an in-depth study towards the Byzantine resilience of decentralized learning systems with two contributions. First, from the adversarial perspective, we theoretically illustrate that Byzantine attacks are more dangerous and feasible in decentralized learning systems: even one malicious participant can arbitrarily alter the models of other participants by sending carefully crafted updates to its neighbors. Second, from the defense perspective, we propose Ubar, a novel algorithm to enhance decentralized learning with Byzantine Fault Tolerance. Specifically, Ubar provides aUniformByzantine-resilientAggregationRule for benign nodes to select the useful parameter updates and filter out the malicious ones in each training iteration. It guarantees that each benign node in a decentralized system can train a correct model under very strong Byzantine attacks with an arbitrary number of faulty nodes. We conduct extensive experiments on standard image classification tasks and the results indicate that Ubar can effectively defeat both simple and sophisticated Byzantine attacks with higher performance efficiency than existing solutions.

Online diagnosis is one of the data services, which can use the machine learning model placed on the cloud and collected physical data from internet of medical things (IoMT) for better medical services. However, the collected user data, diagnosis results and the deployed machine learning model contain sensitive information of users and the healthcare provider, which may lead to serious privacy leakage. To achieve a secure outsourced diagnosis, both high security and low burden for users should be considered. However, the existing works can not solve these problems simultaneously. In this article, based on two non-colluding servers, a privacy-preserving cloud-aided diagnosis scheme for IoMT is proposed. Concretely, a hybrid data encryption method based on homomorphic encryption and AES is used to generate user requests in an efficient way. Besides, we propose a class of secure two-party protocols using homomorphic encryption, such as secure kernel function computation, secure multiplication, and secure comparison, and a privacy-preserving diagnosis scheme based on multi-class SVM with these building blocks is constructed, which can keep users offline in the diagnosis process. Finally, the security analysis and evaluation further illustrate that our scheme is superior to the prior works in terms of security and user-friendliness.

Currently, the conventional mode of power data transaction is mediated by Web pages. Nevertheless, there are challenging issues such as privacy protection, transaction security and data reliability in power data trading. In this paper, we present a novel secure power data trading scheme (SPDTS). Firstly, the zero-knowledge proof is employed to achieve data availability and consistency without revealing the data. Then, SPDTS takes full advantage of the dispersibility and immutability of blockchain to ensure the reliability of data transactions. To keep the transaction process efficient, the processing tasks for power data are performed under smart contract. Meanwhile, a trusted execution environment (TEE) is adopted to guarantee the security of power data. Finally, we present a differential privacy scheme to safeguard the privacy information in the power data. Our study indicates that the proposed scheme can achieve privacy protection, transaction security and data reliability. Also, we conduct security analysis and verify the privacy protection property of the scheme in real cases.

Unrestricted adversarial examples allow the attacker to start attacks without given clean samples, which are quite aggressive and threatening. However, existing works for generating unrestricted adversary examples are quite inefficient and cannot achieve a high success rate. In this article, we explore an end-to-end and effective solution for unrestricted adversary example generation. To stabilize the training process and make our generative model converge to satisfactory results, we design a novel decoupled two-step efficient generative model (EGM), which contains a conditional reference generator and a conditional adversarial transformer. The former is responsible for generating reference samples from noises and source classes. The latter is responsible for converting the reference sample into adversarial examples corresponding to target classes. To improve the success rate, we design a new strategy, augmentation of adversarial labels to produce dynamic target labels and enhance the exploration ability of EGM. Such a strategy can be also applied to existing attacks to improve their attack success rates, which is of independent interest. We conduct extensive experiments to evaluate our proposed model and demonstrate the necessity of decoupling the generation process in EGM. Experimental results show our EGM is much faster and achieves a higher success rate than the state-of-the-art attacks.

Cloud storage synchronization among different computing terminals has attracted large-scale uses among enterprise and individual users. It enables users to maintain the same copy of data in real time, which eases users the tedious yet error-prone data management burden. However, existing cloud storage synchronization systems are in a closed form. Users are fixed to a certain cloud service provider, which makes it hard to transfer from one provider to another when balancing factors such as performance, cost, security, etc. To bridge this gap, this article proposes a new synchronization system based on standard cloudobjectstorage. Specifically, we first formulate the cloud object storage synchronization problem by defining some useful concepts. We then use the idea of state encoding and a push-pull paradigm to propose a cloud object storage synchronization system. The proposed system supports real-time, multiple-terminal, and cloud-independent storage synchronization. We also prototyped the proposed system. The experimental results show that the proposed system is promising for practical usages.

With the advent of cloud computing, it has become more and more popular to outsource various services to the cloud for releasing the burden of local data storage and maintenance. However, it may cause serious privacy problems because the cloud may be untrusted. In this article, we study the privacy-preserving reverse nearest neighbor (PPRNN) query over encrypted spatial data. First, we introduce the concept of reference-locked order-preserving encryption (RL-OPE) with its construction and security proof, which reveals less information than traditional order-preserving encryption (OPE). Then, we present a novel PPRNN scheme in static setting based on structured encryption (SE) and the proposed RL-OPE, called sPPRNN. After that, we design a generic method that extends a PPRNN scheme in static setting to the counterpart in dynamic setting, called dPPRNN. Furthermore, we present a thorough privacy analysis of our proposal. Finally, we demonstrate its efficiency and effectiveness for practical deployment through extensive experiments.

A joint reinforcement learning (RL) and game theory method is presented for segment-level continuous bitrate selection and tile-level bitrate allocation in tile-based 360-degree streaming to increase users’ quality of experience (QoE). First, a viewpoint prediction method based on single-user (SU) viewpoint traces and the saliency map (SM) model is presented to model viewing behaviours. Second, an RL method is proposed to predict segment bitrate and a cooperative bargaining game theory is proposed for bitrate allocation optimization to choose a suitable bitrate for every tile with the help of the viewpoint prediction map. Performance evaluation results indicate that the proposed method can outperform the state-of-the-art methods in terms of different QoE objectives.

Although generative adversarial networks (GANs) show impressive effects on image generation, existing GANs suffer an unstable training process, and thus result in poor image quality sometimes. To solve this problem, we first introduce a supervision mechanism into GANs and propose a teacher-supervised GAN (GAN-T) model. Specifically, we design a teacher supervision mechanism to inspect whether the features of generated images are as real as those of real images. If not, we add action into the generator. The action takes the encoding of the real image as prior knowledge to guide the generation of samples. We then apply our proposed method to existing GANs to show its compatibility with them. Finally, we conduct extensive experiments on the tasks of noise-to-image generation and image translation, and experimental results show that our proposed method can significantly stabilize the training process of the generator and improve the quality of generated images.

In this paper, a quality of experience (QoE)-based neural live streaming method with dynamic adaptive video quality control is developed to improve streaming performance. First, the dynamic adaptive streaming issue is formulated as a Markov decision process (MDP) problem. Second, an reinforcement learning (RL)-based approach is proposed as an appropriate solution, where the client functions as an RL agent and the environment is made up of various networks. User QoE is the reward by mutual consideration of video quality and play-back state. Finally, to optimize the total reward, the RL algorithm chooses the required video quality for each video segment. Experimental results show that the proposed RL-based streaming algorithm outperforms state-of-the-art schemes in terms of both temporal and visual QoE metrics by a noticeable margin while guaranteeing application-level fairness when multiple clients share a bottlenecked network. The code is available on the following website: https://github.com/OpenCode007/ICME2021.

Watermarking has become the tendency in protecting the intellectual property of DNN models. Recent works, from the adversary's perspective, attempted to subvert watermarking mechanisms by designing watermark removal attacks. However, these attacks mainly adopted sophisticated fine-tuning techniques, which have certain fatal drawbacks or unrealistic assumptions. In this paper, we propose a novel watermark removal attack from a different perspective. Instead of just fine-tuning the watermarked models, we design a simple yet powerful transformation algorithm by combining imperceptible pattern embedding and spatial-level transformations, which can effectively and blindly destroy the memorization of watermarked models to the watermark samples. We also introduce a lightweight fine-tuning strategy to preserve the model performance. Our solution requires much less resource or knowledge about the watermarking scheme than prior works. Extensive experimental results indicate that our attack can bypass state-of-the-art watermarking solutions with very high success rates. Based on our attack, we propose watermark augmentation techniques to enhance the robustness of existing watermarks.

Multiple-instance learning (MIL) is an important weakly supervised binary classification problem, where training instances are arranged in bags, and each bag is assigned a positive or negative label. Most of the previous studies for MIL assume that training bags are fully labeled. However, in some real-world scenarios, it could be difficult to collect fully labeled bags, due to the expensive time and labor consumption of the labeling task. Fortunately, it could be much easier for us to collect similar and dissimilar bags (indicating whether two bags share the same label or not), because we do not need to figure out the underlying label of each bag in this case. Therefore, in this paper, we for the first time investigate MIL from only similar and dissimilar bags. To solve this new MIL problem, we propose a convex formulation to train a bag-level classifier based on empirical risk minimization and theoretically derive a generalization error bound. In addition, we also propose a strong baseline for this new MIL problem, which aims to train an instance-level classifier by minimizing the instance-level empirical risk. Extensive experimental results clearly demonstrate that our proposed baseline works well, while our proposed convex formulation is even better.

Perceptual encryption is an efficient way of protecting image content by only selectively encrypting a portion of significant data in plain images. Existing security analysis of perceptual encryption usually resorts to traditional cryptanalysis techniques, which require heavy manual work and strict prior knowledge of encryption schemes. In this paper, we introduce a new end-to-end method of analyzing the visual security of perceptually encrypted images, without any manual work or knowing any prior knowledge of the encryption scheme. Specifically, by leveraging convolutional neural networks (CNNs), we propose a progressive recovery network (PRNet) to recover visual content from perceptually encrypted images. Our PRNet is stacked with several dense attention recovery blocks (DARBs), where each DARB contains two branches: feature extraction branch and image recovery branch. These two branches cooperate to rehabilitate more detailed visual information and generate efficient feature representation via densely connected structure and dual-saliency mechanism. We conduct extensive experiments to demonstrate that PRNet works on different perceptual encryption schemes with different settings, and the results show that PRNet significantly outperforms the state-of-the-art CNN-based image restoration methods.

Most existing cluster-based cross-domain person re-identification (re-id) methods only pre-train the re-id model on the source domain. Unfortunately, the pre-trained model may not perform well on the target domain due to the large domain gap between source and target domains, which is harmful to the following optimization. In this paper, we propose a novel Self-supervised Pre-training method on the Target Domain (SPTD), which pre-trains the model on both the source and target domains in a self-supervised manner. Specifically, SPTD uses different kinds of data augmentation manners to simulate different intra-class changes and constraints the consistency between the augmented data distribution and the original data distribution. As a result, the pre-trained model involves some specific discriminative knowledge on the target domain and is beneficial to the following optimization. It is easy to combine the proposed SPTD with other cluster-based cross-domain re-id methods just by replacing the original pre-trained model with our pre-trained model. Comprehensive experiments on three widely used datasets, i.e. Market1501, DukeMTMC-ReID and MSMT17, demonstrate the effectiveness of SPTD. Especially, the final results surpass previous state-of-the-art methods by a large margin.

Smart home is a good exemplification of the Internet of Things (IoT). Many researchers study how to make smart home systems (SHS) smarter. However, the security of SHS is also worth studying as user's information is collected from the devices and security-sensitive data are communicated through an open network. Therefore, how to guarantee the security of the data transmission in SHS is an important problem. In this paper, we propose a new secure data transmission scheme with access control to protect the data transmission in SHS. In our scheme, data transmission is secured by a new cryptographic primitive called access control encryption (ACE). Different from other existing solutions in SHS, our scheme controls not only which messages smart devices can receive, but also which messages they can send. Our experimental results demonstrate the effectiveness and efficiency of our proposed mechanism.

For the text-to-image synthesis task, most discriminators in existing generative adversarial networks based methods tend to fall into a local suboptimal state too early in the training process, resulting in the poor quality of generated images. To address the above problems, a hybrid augmented loss discriminator is designed. In this designed discriminator, to reduce the sensitivity of the discriminator classification recognition, make it pay attention to the semantic and structural changes, we add the loss value of the fake sample to the loss value of the real sample. Moreover, to indirectly guide the generator to generate samples, the loss value of the real sample is added to the fake sample. The loss value mixed with real and fake samples actually augments signal transmission. It perturbs parameter update of the discriminator during optimization and prevents the discriminator from falling into the local suboptimal state prematurely. Whereafter, we apply the proposed discriminator to two kinds of text-to-image synthesis tasks. Experimental results show that the proposed method can help the baseline models to improve performance.

Blockchain technology has attracted tremendous interest from both industry and academia. It is typically used to record a public history of transactions (e.g., payment/smart contract data), but storing nonpayment/contract data in transactions has been common. The ability to store data unrelated to payment/contract such as illicit data on blockchain may be abused for malicious purposes. For example, one may use blockchain to store the data related to child pornography and copyright violations, which are publicly visible and immutable. Moreover, an immutable blockchain is not suitable for all blockchain-based applications. So far, numerous redaction mechanisms for the mutable blockchain have been developed. In this paper, we aim at conducting a comprehensive survey that reviews and analyzes the state-of-the-art redaction mechanisms. We start by giving a general presentation of blockchain and summarize the typical methods of inserting data in blockchain. Next, we discuss the challenges of designing the redaction mechanism and propose a list of evaluation criteria. Then, redaction mechanisms of the existing mutable blockchains are systemically reviewed and analyzed based on our evaluation criteria. The analyses include algorithmic overviews, performance limitations, and security vulnerabilities. Finally, the comparisons and analyses provide new insights into these mechanisms. This survey will provide developers and researchers a comprehensive view and facilitate the design of future mutable blockchains.

The Internet of Things (IoT) is gradually stepping out of its infancy into maturity. Its widespread applications cover from tiny wearable devices to large industrial systems. Although many security solutions have been introduced to address data security and privacy problems caused by the unique characteristics of IoT, how to simultaneously achieve data confidentiality, protect the privacy of access control policy, and provide reasonable data source identification has been a challenging problem. Moreover, lacking one of the above properties may result in serious issues (e.g., leakage information and forging identity), and the situation grows steadily worse with the expansion of “things” scale. To address the above issues, we propose a new cryptographic primitive named certificateless matchmaking encryption (CL-ME), which inherits the security properties of certificateless cryptosystem and matchmaking encryption. Meanwhile, we also present two effective concrete constructions with formal security proofs based on the standard hard assumptions. The basic construction is the first instance of CL-ME based on bilinear pairing, and the enhanced construction is a pairing-free lightweight solution. Finally, we implement our proposed schemes using popular cryptography library and compare their performance with existing works. Theoretical analysis and experimental evaluations demonstrate that our proposed schemes are more suitable for IoT environment.

The visual security index (VSI) is a quantized indicator for objective visual security evaluation of selectively encrypted images. One challenging problem in current research is that the performance of VSIs is highly sensitive to the extracted features and the method of similarity measurement, and it is hard to choose appropriate handcrafted features from encrypted images, as well as to find an effective similarity measurement. In this paper, we make the first attempt to present a novel convolutional neural network-based visual security index (CNNVSI). Our proposed CNNVSI is purely data-driven and trained end-to-end. We propose three specialized designs to make the approach work for encrypted low-quality images without any handcrafted features or prior knowledge about the human vision system (HVS). First, we present a patch labeling algorithm to assign each encrypted patch a visual security score. Second, we design a multiscale attention residual network (MARNet) for feature learning. Last, we propose to fuse the learned features from plain images, encrypted images and their discrepancy images. Extensive and systematic experiments are conducted on five publicly available image databases to analyze the performance of our proposed CNNVSI, and the experimental results and their analysis demonstrate that our proposed CNNVSI significantly outperforms the existing state-of-the-art methods in terms of accuracy and stability.

This article investigates the problem of distributed online optimization for a group of units communicating on time-varying unbalanced directed networks. The main target of the set of units is to cooperatively minimize the sum of all locally known convex cost functions (global cost function) while pursuing the privacy of their local cost functions being well masked. To address such optimization problems in a collaborative and distributed fashion, a differentially private-distributed stochastic subgradient-push algorithm, called DP-DSSP, is proposed, which ensures that units interact with in-neighbors and collectively optimize the global cost function. Unlike most of the existing distributed algorithms which do not consider privacy issues, DP-DSSP via differential privacy strategy successfully masks the privacy of participating units, which is more practical in applications involving sensitive messages, such as military affairs or medical treatment. An important feature of DP-DSSP is tackling distributed online optimization problems under the circumstance of time-varying unbalanced directed networks. Theoretical analysis indicates that DP-DSSP can effectively mask differential privacy as well as can achieve sublinear regrets. A compromise between the privacy levels and the accuracy of DP-DSSP is also revealed. Furthermore, DP-DSSP is capable of handling arbitrarily large but uniformly bounded delays in the communication links. Finally, simulation experiments confirm the practicability of DP-DSSP and the findings in this article.

Due to the abundant marine resources, smart ocean has attracted much attention of the government, industry, and academy. The Internet-of-Things (IoT) architectures for smart ocean have been proposed to collect various of data from the ocean, thereby assisting environmental protection, military reconnaissance, and so on. However, few researchers have paid attention to the security and privacy issues of data collection and transmission. In this article, for the unreliable underwater environment, we present a secure, efficient, and complete data collection, and transmission and storage scheme for IoT in smart ocean. Especially, to prolong the lifetime of the underwater node, two novel data compression algorithms [lossy data compression algorithm (LCA) and lossless data compression algorithm (NLCA)] are also proposed. Moreover, due to the vulnerability of underwater nodes, we also propose a corresponding IoT framework and data collection pattern to resist the single point failure attack. Besides, to guarantee the confidentiality, reliability, and integrity of transmitting data, Elliptic Curve-ElGamal (EC-ElGamal) and elliptic curve digital signature algorithm (ECDSA) are employed. The consensus algorithm and blacklisting mechanism are also employed to detect and address failure or malicious nodes. Finally, the security analysis demonstrates that our scheme is able to resist many typical attacks for underwater nodes, such as manipulation attacks, Distributed Denial-of-Service (DDoS) attacks, malicious node injection attacks, and so on. Additionally, relevant experimental results show that the scheme is feasibility and efficiency.

Fog-cloud computing promises many new vertical service areas beyond simple data communication, storing, and processing. Among them, distributed deep learning (DDL) across fog-cloud computing environment is one of the most popular applications due to its high efficiency and scalability. Compared with the centralized deep learning, DDL can provide better privacy protection with training only on sharing parameters. Nevertheless, when DDL meets fog-cloud computing, it still faces two major security challenges: 1) how to protect users' privacy from being leaked to other internal participants in the training process and 2) how to guarantee users' identities from being forged by external adversaries. To combat them, several approaches have been proposed via various technologies. Nevertheless, those approaches suffer from drawbacks in terms of security, efficiency, and functionality, and cannot guarantee the legitimacy of participants' identities during training. In this article, we propose a secure and privacy-preserving DDL (SPDDL) for fog-cloud computing. Compared with the state-of-the-art works, our proposal achieves a better tradeoff between security, efficiency, and functionality. In addition, our SPDDL can guarantee the unforgeability of users' identities against external adversaries. Extensive experimental results indicate the practical feasibility and high efficiency of our SPDDL.

Perceptual/selective encryption has been gaining widespread attention as an emerging technology for image privacy protection. However, few studies focus on the visual security evaluation of perceptually encrypted images, which has a significant impact on measuring the effectiveness and practicality of these encryption methods. In this paper, we propose an image importance-based visual security index (IIBVSI) by leveraging spatial contrast and texture features. Based on the characteristics of perceptually encrypted images, we present an averaged high-order gradient magnitude map to describe the spatial contrast feature and introduce a combined local amplitude map of multiple log-Gabor filters to represent the texture feature. Specifically, the multiresolution representation of an image is first created by downsampling to simulate the hierarchical property of the human visual system. Next, for each scale of image resolution, the spatial contrast and the texture feature maps are extracted from both plain and encrypted images. Similarity measurements are then conducted on these feature maps to generate the contrast and the texture similarity maps. An image importance-based pooling strategy is subsequently proposed to combine these measurements and generate a visual security score. The final IIBVSI score is computed by averaging the visual security scores of all scales of image resolution. Extensive experiments are conducted on several publicly available databases, and the results demonstrate the superiority and robustness of our proposed IIBVSI compared with existing state-of-the-art work in the low and moderate image quality ranges.

Perceptual image encryption provides an efficient and effective way to preserve the confidentiality of visual information, and the measurement of content leakage is of fundamental importance for perceptually encrypted images. Numerous visual security indexes (VSIs) have been proposed to evaluate visual content leakage. Due to the lack of perceptually encrypted image databases, image quality assessment (IQA) databases are widely adopted to evaluate the performance of existing VSIs. However, there are huge differences between VSIs and IQAs. The misuse of databases may lead to an inaccurate evaluation. In this paper, we propose a perceptually encrypted image database (PEID) which contains 1080 encrypted images from 20 plain images with 10 well-known perceptual encryption techniques. Both visual quality and content leakage scores of the encrypted images are obtained through a comprehensive subjective evaluation. We also propose a systemic methodology to accurately evaluate the monotonicity, fitness, and accuracy of VSIs. We conduct extensive experiments on the proposed PEID to evaluate the performance of existing state-of-the-art VSIs. We have made the database publicly available for download and hope that the proposed PEID can facilitate the research of visual security evaluation and beyond.

Blind image quality assessment (BIQA) aims to develop quantitative measures to automatically and accurately estimate the visual quality of an image without any prior information about its reference image. This issue has been attracting a great deal of attention for a long time; however, little work has been done on night-time images, which are crucially important for consumer photography and practical applications such as automated driving systems. In this paper, to the best of our knowledge, we conduct the first exploration on subjective and objective quality assessment of night-time images. First, we build a large-scale natural night-time image database (NNID) containing 2240 images with 448 different image contents captured by different photographic equipment in real-world scenarios. Subsequently, we carry out a subjective experiment to evaluate the perceptual quality of all the images in the NNID database. Thereafter, we perform objective assessment of night-time images by proposing a blind night-time image quality assessment metric using brightness and texture features (BNBT). Finally, extensive experiments are conducted to evaluate the performance and efficiency of the proposed BNBT metric on the NNID database. The experimental results demonstrate that this metric outperforms existing state-of-the-art BIQA methods in terms of all evaluation criteria and has an acceptable computational cost at the same time. We have made the NNID database publicly available for downloading at https://sites.google.com/site/xiangtaooo/.

Cloud storage security has gained considerable research efforts with the wide adoption of cloud computing. As a security mechanism, researchers have been investigating cloud storage auditing schemes that enable a user to verify whether the cloud keeps the user's outsourced data undamaged. However, existing schemes have usability issues in compatibility with existing real world cloud storage applications, error-tolerance, and efficiency. To mitigate this usability gap, this article proposes a new general cloud storage auditing scheme that is more usable. The proposed scheme uses the idea of integrating linear error correcting codes and linear homomorphic authentication schemes together. This integration uses only one additional block to achieve error tolerance and authentication simultaneously. To demonstrate the power of the general construction, we also propose one detailed scheme based on the proposed general construction using the Reed Solomon code and the universal hash based MAC authentication scheme, both of which are implemented over the computation-efficient Galois field GF(28). We also show that the proposed scheme is secure under the standard definition. Moreover, we implemented and open-sourced the proposed scheme. Experimental results show that the proposed scheme is orders of magnitude more efficient than the state-of-the-art scheme.

Differential evolution (DE) is a simple yet powerful evolutionary algorithm for the solving of continuous optimization problems. In the last decades, a plethora of DE variants have been proposed in the literature for enhanced optimization performance. However, most of these DE variants are designed to solve a single problem in a single run. Recently, a multifactorial DE (MFDE) has been proposed to conduct evolutionary search on multiple tasks simultaneously. Benefitting from the implicit knowledge transfer among different tasks, MFDE has demonstrated a superior performance against the single-task DE in terms of convergence speed and solution quality. In MFDE, the knowledge transfer is realized via the mutation operation conducted on solutions with different skill factors. However, despite a lot of mutation strategies suggested in the literature, the current MFDE takes DE/rand/1 as the only strategy for knowledge transfer. The impacts of different mutation strategies on the performance of MFDE is still unexplored. Taking this cue, in this paper, we embark a study to investigate how different mutation strategies for knowledge transfer affect the performance of MFDE. In particular, besides DE/rand/1, another four commonly-used mutation strategies are adapted for the purpose of multitask optimization. Further, towards effective mutation for knowledge transfer in MFDE, a new mutation strategy called DE/best/1+ρ, which is able to adjust its behavior along the search process is proposed. Lastly, comprehensive empirical studies are conducted to investigate the performance of existing and the new proposed mutation strategies on the 9 single-objective multitasking benchmarks.

With the explosive growth of online images and the popularity of search engines, a great demand has arisen for small and medium-sized enterprises to build and outsource large-scale image retrieval systems to cloud platforms. While reducing storage and retrieval burdens, enterprises are at risk of facing untrusted cloud service providers. In this paper, we take the first step in studying the problem of query authentication for large-scale image retrieval. Due to the large size of image files, the main challenges are to (i) design efficient authenticated data structures (ADSs) and (ii) balance search, communication, and verification complexities. To address these challenges, we propose two novel ADSs, the Merkle randomized k-d tree and the Merkle inverted index with cuckoo filters, to ensure the integrity of query results in each step of image retrieval. For each ADS, we develop corresponding search and verification algorithms on the basis of a series of systemic design strategies. Furthermore, we put together the ADSs and algorithms to design the final authentication scheme for image retrieval, which we name ImageProof. We also propose several optimization techniques to improve the performance of the proposed ImageProof scheme. Security analysis and extensive experiments are performed to show the robustness and efficiency of ImageProof.

Summary In the presence of scale, dynamism, uncertainty, and elasticity, cloud engineers face several challenges when allocating resources for cloud‐based software services. They should allocate appropriate resources in order to guarantee good quality of services as well as low cost of resources. Self‐adaptive ability is needed in this process because engineers' intervention is difficult. Traditional self‐adaptive resource allocation methods are policy‐driven. Thus, cloud engineers usually have to develop separate sets of rules for each systems in order to allocate resources effectively, which leads to high administrative cost and implementation complexity. Machine learning has made great achievements in many fields, and it can be also applied to resource allocation. In this paper, we present a self‐learning and self‐adaptive approach to resource allocation for cloud‐based software services. For a given cloud‐based software service, its QoS model is firstly trained on history data, which is capable to predict the QoS value as output by using the information on workload and allocated resources as inputs. Then, on‐line decision‐making on resource allocation can be carried out automatically based on genetic algorithm, which is aimed to search reasonable resource allocation plan by using the QoS model. We evaluate our approach on RUBiS benchmark, demonstrating the accuracy of the QoS model over 90% and the improvement of resource utilization by 10%‐30%.

As the next generation of information and communication infrastructure, Internet of Things (IoT) enables many advanced applications such as smart healthcare, smart grid, smart home, and so on, which provide the most flexibility and convenience in our daily life. However, pervasive security and privacy issues are also increasing in IoT. For instance, an attacker can get health condition of a patient via analyzing real-time records in a smart healthcare application. Therefore, it is very important for users to protect their private data. In this paper, we present two efficient data aggregation schemes to preserve private data of customers. In the first scheme, each IoT device slices its actual data randomly, keeps one piece to itself, and sends the remaining pieces to other devices which are in the same group via symmetric encryption. Then, each IoT device adds the received pieces and the held piece together to get an immediate result, which is sent to the aggregator after the computation. Moreover, homomorphic encryption and AES encryption are employed to guarantee secure communication. In the second scheme, the slicing strategy is also employed. Noise data are introduced to prevent the exchanged actual data of devices from disclosure when the devices blend data each other. AES encryption is also employed to guarantee secure communication between devices and aggregator, compared to homomorphic encryption, which has significantly less computational cost. Analysis shows that integrity and confidentiality of IoT devices’ data can be guaranteed in our schemes. Both schemes can resist external attack, internal attack, colluding attack, and so on.

In cloud computing environments, data centers can provide high-performance computing resources and distributed storage space. Scientific workflows often need to be implemented across multiple data centers, where copious amounts of application data are stored. Moving data across geographically distributed data centers leads to intolerable delays and hinders the efficient execution of scientific workflows, which are large-scale data-intensive. Reasonable data placement can reduce data scheduling between the data centers effectively. In this paper, an adaptive discrete particle swarm optimization (PSO) algorithm based on genetic algorithm has been proposed to decrease the number of data transmissions across data centers. The algorithm overcame the premature convergence defect of PSO by introducing the mutation and crossover of genetic algorithm. Moreover, it effectively improved the diversity in the process of population evolution. Compared with the previous work, the simulation results showed that the proposed strategy greatly reduced the volume of data transfer while reducing the number of data movement across data centers.

Cloud storage security has been gaining research interest in recent years. Although considerable work has been conducted on verifying the integrity of the outsourced data in the cloud, how to efficiently verify the file search results returned from the cloud is still a challenge to be resolved. Towards this direction, we tackle the verifiable file search problem in this paper. We formulate and solve this problem by proposing two protocols. The first protocol enables verifying the correctness of the file search result when all users have the same security privilege in accessing the outsourced data. The second protocol, which builds on the first protocol, further enables user differentiation, i.e., different users can only access files that fit their security privileges. In our protocols, we employ two key strategies in enabling file search verifiability. One is to separate all possible filenames into two finite sets and the other is to embed some secret information in the outsourced data. Further, we leverage the key chaining and recursion mechanisms to enable user differentiation. We have conducted experiments to validate the effectiveness of our proposed protocols. Our results show that both protocols are efficient in terms of computation, storage, and communication cost.

Ensemble learning plays an important role in big data analysis. A great limitation is that multiple parties cannot share their knowledge extracted from ensemble learning model with privacy guarantee, therefore it is a great demand to develop privacy-preserving collaborative ensemble learning. This paper proposes a privacy-preserving collaborative ensemble learning framework under differential privacy. In the framework, multiple parties can independently build their local ensemble models with personalized privacy budgets, and collaboratively share their knowledge to obtain a stronger classifier with the help of central agent in a privacy-preserving way. Under this framework, this paper presents the differentially private versions of two widely-used ensemble learning algorithms: collaborative random forests under differential privacy (CRFsDP) and collaborative adaptive boosting under differential privacy (CAdaBoostDP). Theoretical analysis and extensive experimental results show that our proposed framework achieves a good balance between privacy and utility in an efficient way.

We present a new high-payload joint reversible data-hiding scheme for encrypted images. Instead of embedding data in the encrypted image directly, the content owner first uses an interpolation technique to estimate whether the location can be used for embedding and generates a location map before encryption. Next, the data hider embeds the additional data through flipping the most significant bits (MSBs) of the encrypted image according to the location map. At the receiver side, before extracting the additional data and reconstructing the image, the receiver decrypts the image first. Experimental results demonstrate that the proposed method can achieve real reversibility, which means data extraction and image recovery are free of error. Moreover, our scheme can embed more payloads than most existing reversible data hiding schemes in encrypted images.

In practice, the greatest threat against the security of a digital signature scheme is the exposure of signing key, since the forward security of past signatures and the backward security of future signatures could be compromised. There are some attempts in the literature, addressing forward-secure signature for preventing forgeries of signatures in the past time; however, few studies addressed the backward-security of signatures, which prevents forgeries in the future time. In this paper, we introduce the concept of key-evolving signature with bilateral security, i.e., both forward security and backward security. We first define the bilateral security formally for preventing the adversaries from forging a valid signature of the past and the future time periods in the case of key exposure. We then provide a novel construction based on hub-and-spoke updating structure and the random oracle model, and show that the construction achieves bilateral security and unbounded number of time periods. Finally, we compare our scheme with the existing work by rigorous analysis and experimental evaluation, and demonstrate that our construction is more secure and efficient for practical applications.

This paper reveals an intrinsic relationship between secure cloud storage and secure network coding for the first time. Secure cloud storage was proposed only recently while secure network coding has been studied for more than ten years. Although the two areas are quite different in their nature and are studied independently, we show how to construct a secure cloud storage protocol given any secure network coding protocol. This gives rise to a systematic way to construct secure cloud storage protocols. Our construction is secure under a definition which captures the real world usage of the cloud storage. Furthermore, we propose two specific secure cloud storage protocols based on two recent secure network coding protocols. In particular, we obtain the first publicly verifiable secure cloud storage protocol in the standard model. We also enhance the proposed generic construction to support user anonymity and third-party public auditing, which both have received considerable attention recently. Finally, we prototype the newly proposed protocol and evaluate its performance. Experimental results validate the effectiveness of the protocol.

With the development in recent decades of various efficient image encryption algorithms, such as selective encryption, a great demand has arisen for methods of evaluating the visual security of encrypted images. Existing solutions usually adopt well-known metrics of visual quality assessment to measure the quality of encrypted images, but they often exhibit undesired behavior on perceptually encrypted images of low quality. In this paper, we propose a novel visual security index (VSI) based on the human visual system. The proposed VSI evaluates two aspects of the content similarity between plain and encrypted images: the edge similarity extracted via multi-threshold edge detection and the texture similarity measured by means of the co-occurrence matrix. These two components are further integrated to obtain the proposed VSI through adaptive similarity weighting. Extensive experiments were performed on two publicly available image databases. Our experimental results demonstrate that compared with many existing state-of-the-art visual security metrics, the proposed VSI exhibits a better performance and stability on low-quality images.

Cloud storage has gained a remarkable success in recent years with an increasing number of consumers and enterprises outsourcing their data to the cloud. To assure the availability and integrity of the outsourced data, several protocols have been proposed to audit cloud storage. Despite the formally guaranteed security, the constructions employed heavy cryptographic operations as well as advanced concepts (e.g., bilinear maps over elliptic curves and digital signatures), and thus are inefficient to admit wide applicability in practice. In this paper, we design a novel secure cloud storage protocol, which is conceptually and technically simpler and significantly more efficient than previous constructions. Inspired by a classic string equality checking protocol in distributed computing, our protocol uses only basic integer arithmetic (without advanced techniques and concepts). As simple as the protocol is, it supports both randomized and deterministic auditing to fit different applications. We further extend the proposed protocol to support data dynamics, i.e., adding, deleting and modifying data, using a novel technique. As a further contribution, we find a systematic way to design secure cloud storage protocols based on verifiable computation protocols. Theoretical and experimental analyses validate the efficacy of our protocol.

In this paper, we study the verification problem of particle swarm optimization (PSO) when it is outsourced to the cloud, i.e. making sure that the cloud executes PSO algorithm as requested. A verifiable PSO algorithm and its verification algorithm are proposed. The proposed scheme does not involve expensive cryptography, and it is efficient and effective to verify the honesty of the cloud.

This paper investigates the intrinsic relationship between secure cloud storage and secure network coding for the first time. Secure cloud storage was proposed only recently while secure network coding has been studied for more than ten years. We show in general how to construct a secure cloud storage protocol given any secure network coding protocol. Our construction suggests a systematic way to construct various secure cloud storage protocols. We also show that it is secure under a definition which captures the real world uses of the cloud storage. From our general construction, we propose a secure cloud storage protocol based on a recent secure network coding protocol. The protocol is the first publicly verifiable secure cloud storage protocol in the standard model, while the previous work is either not publicly verifiable, or security argument is only argued heuristically in the random oracle model. We also enhance the proposed protocol to support third-party public auditing, which has received considerable attention recently. Finally, we prototype the proposed protocol and evaluate its performance. Experimental results validate the effectiveness of the protocol.

Attribute based encryption (ABE) has been applied to many applications nowadays [1][2] and it effectively achieves a fine grained access control. Even the encryptor needs only one encryption operation, and all the decryption operations are distributed to the receiver's end, the computational cost of encryption is still impractical when there is a large amount of encryption with different access structures. In this paper, we examine existing techniques that optimize the decentralized attribute based encryption scheme [3], such as a better construction of e Linear Secret-Sharing Scheme (LSSS) matrix, pre-processing of scalar multiplication and pairing, multi-pairing and so on. We proposed the deployment of offline pools to improve the real-time operations. We proposed the method to construct offline pools and designed algorithms to achieve better hitting rate of offline pool topples. We evaluated the optimization techniques, the result shows that there is a 45 times of performance improvement in the encryption (6-8 times for the decryption) after we applied all the real-time optimization techniques mentioned in our paper. With deployment of the offline pools, the optimization can be improved at least 100-200 times than without offline pools.

With cloud computing and mobile computing becoming more and more popular, there are a lot potential applications for computation outsourcing to the cloud. This paper investigates the linear regression outsourcing problem, which is a quite common engineering task and employed in various applications, as a case study to find out the possible problems that need to be solved. We propose two protocols which can enable secure and efficient outsourcing of linear regression problems to the cloud. The protocols can protect the client’s data privacy well and at the same time have good efficiency. We show all subtleties and the techniques in designing such protocols. The main idea to protect the privacy is employing some transformations to the original linear regression problem to get a new problem which is sent to the cloud; and then transforming the answer returned back from the cloud to get the true solution to the original problem. Experimental results validate the practical usability of our protocols.

The Arnold cat map is employed in various applications where chaos is utilized, especially chaos-based cryptography and watermarking. In this paper, we study the problem of period distribution of the generalized discrete Arnold cat map over the Galois ring \BBZ2e. Full knowledge of the period distribution is obtained analytically by adopting the Hensel lift approach. Our results have impact on both chaos theory and its applications as they not only provide design strategy in applications where special periods are required, but also help to identify unstable periodic orbits of the original chaotic cat map. The method in our paper also shows some ideas how to handle problems over the Galois ring \BBZ2e.

In this paper, we analyze the period distribution of the generalized discrete cat map over the Galois ring where is a prime. The sequences generated by this map are modeled as 2-dimensional LFSR sequences. Employing the generation function and the Hensel lifting approaches, full knowledge of the detail period distribution is obtained analytically. Our results not only characterize the period distribution of the cat map, which gives insights to various applications, but also demonstrate some approaches to deal with the period of a polynomial in the Galois ring.

In this paper, an improved particle swarm optimization algorithm with momentum (mPSO) is proposed based on inspiration from the back propagation (BP) learning algorithm with momentum in neural networks. The momentum acts as a lowpass filter to relieve excessive oscillation and also extends the PSO velocity updating equation to a second-order difference equation. Experimental results are shown to verify its superiority both in robustness and efficiency.