Ensuring data confidentiality while achieving efficient deduplication is critical for cloud storage services. Existing works on encrypted data deduplication operate under a deduplication-before-storage (DbS) model, identifying duplicates only during the data uploading phase. They require the owners of duplicate data or additional key servers to assist in the data uploading phase for enabling encrypted deduplication. This makes it hard to balance data security and deduplication effectiveness. This paper introduces a new paradigm named deduplication-while-storage (DwS), which differs from the existing DbS model by enabling dynamic deduplication throughout the entire storage lifecycle, activated whenever owners of potential duplicate data are available. This paradigm aims to significantly boost deduplication ratio without compromising the security of outsourced data. To instantiate the DwS paradigm, we develop SGX-DwS, a shielded storage system that utilizes Intel SGX to support encrypted data deduplication. SGX-DwS builds on an ownership-based encryption mechanism that enables secure equality checks on ciphertexts with the cooperation of owners of duplicate data. It adopts a two-phase deduplication strategy. The first phase processes duplicates during uploading if a previous uploader of duplicate data is online and willing to assist (with incentives provided for participation), while the second phase addresses remaining duplicates during the storage phase when owners of the potential duplicates are simultaneously online and cooperative. Our experimental results demonstrate that SGX-DwS achieves high deduplication ratio with modest overhead, even in large-scale cloud storage environments. Specifically, the improvement in deduplication ratio achieved by SGX-DwS is up to 92.49% as compared to that in the DbS model.

Cloud services have been commonly leveraged to store and manage the exponential growth of images, yet this also comes with critical data privacy concerns. Reversible data hiding over encrypted images (RDH-EI) techniques can embed data into encrypted images and support lossless recovery, which can provide an effective solution for securely managing private images in the cloud. However, existing schemes generally suffer from low embedding capacity. Moreover, most of them rely on a single cloud server, which introduces a single point of failure. In this paper, we first propose a pixel correlation recovery (PCR) technique for restoring the pixel correlation excessively disrupted during encryption. Using the PCR technique, we develop a secure (r, n)-threshold RDH-EI scheme with large embedding capacity and avoidance of single point of failure. In our scheme, a content owner encrypts a confidential image into n shares and distributes them across n independent cloud servers. We design a new encoding method enabling each cloud server to efficiently encode the share, preserving capacity for data embedding. An authorized receiver can later extract the embedded data and reconstruct the confidential image from r shares. Experiments demonstrate that our scheme achieves significantly larger embedding capacity over state-of-the-art schemes.

Vertical federated learning (VFL) enables collaborative learning across different entities with disjoint data features for the same instances. In VFL, passive parties with partial data features extract embeddings from their data and forward them to the active party, which holds disjoint data features and labels, for aggregation and subsequent prediction. However, some passive parties may act as free-riders and submit valueless embeddings to deceive rewards, which undermines the fairness of collaborative training and increases communication overhead. Compared to horizontal federated learning (HFL), detecting free-riders in VFL is more challenging due to the distinct data features and heterogeneous embeddings each party produces. This makes it difficult to identify disguised embeddings of free-riders using anomaly detection methods typically employed in HFL. This paper proposes the first free-rider detection strategy in VFL using an unsupervised auxiliary task based on maximum mean discrepancy (MMD). It helps benign parties capture shared information from the active party, resulting in smaller MMD distances for benign embeddings compared to those of free-riders. Additionally, considering that embeddings may be exploited to infer local data, we introduce PVF-FD, a ciphertext-domain verifiable embedding learning scheme that enables the main and auxiliary tasks to be performed simultaneously in a privacy-preserving manner. We formally analyze the security of PVF-FD. Experimental results demonstrate that PVF-FD can effectively detect free-riders, reduce communication overhead, and maintain the performance of the main task.

As cloud storage is extensively utilized in the contemporary digital age, assuring data integrity and conserving cloud storage space has become a priority for all. However, existing cross-user deduplication audit schemes conflict with the pay-as-you-go model, causing unnecessary costs and violating data isolation. Moreover, retaining a single copy of identical data across multiple users introduces maintenance challenges during data operations. To address these issues, we propose a new blockchain-enabled efficient deduplication and mixed auditing scheme which intricately integrates Message-Locked Encryption (MLE) to construct Homomorphic Verifiable Tags (HVTs), enabling deduplication without exposing confidential data. Our scheme supports single-user deduplication at both block and file levels, as well as plaintext-ciphertext mixed auditing, thereby preventing redundant payments while preserving data isolation to simplify maintenance during data operations and ownership transfers. By employing Elliptic Curve Cryptography (ECC) to encrypt keys and storing the encrypted keys on the blockchain, we ensure data confidentiality while reducing the burden of local key management. Leveraging blockchain-based smart contracts, we further design a self-auditing mechanism that eliminates reliance on trusted third-party auditors. Moreover, our scheme embraces dynamic data operations through an optimized Merkle Hash Tree (MHT) and enables secure cloud data ownership transfer via identity verification. Finally, we prove the correctness and security of our scheme and evaluate its performance through experiments and comparisons with state-of-the-art works, demonstrating its efficiency, particularly in the data upload phase.

Federated learning (FL) is a compelling privacy-friendly paradigm that allows multiple clients to jointly train a model by sharing only gradient updates instead of their local datasets. Since gradient updates may still expose sensitive information, a line of research has explored the use of local differential privacy (LDP) mechanisms to formally safeguard these updates. Under LDP, each client perturbs its gradients locally prior to sharing. However, LDP often leads to a significant degradation in model utility due to the addition of large noises. To enable a better balance between privacy and utility, an increasing trend is to leverage the shuffle model of differential privacy (DP) in FL, which introduces an intermediate shuffling operation on the perturbed gradients, enabling privacy amplification. Following this trend, we present${\sf Camel}$, a communication-efficient and maliciously secure FL framework operating under the shuffle model of DP. A key difference of${\sf Camel}$from existing works is its new support for integrity checks on the shuffle computation, providing security against a malicious adversary. To achieve this,${\sf Camel}$builds on a trending cryptographic technique called secret-shared shuffle, and augments it by our custom methods for system-wide communication optimization and lightweight server-side integrity verification. Furthermore, we provide a formal analysis of privacy loss by employing Rényi differential privacy (RDP) for the entire FL process, which allows a tighter privacy bound. Our comprehensive experimental results show that${\sf Camel}$outperforms current state-of-the-art approaches in achieving better privacy-utility trade-offs, while maintaining promising performance.

Image encryption techniques generally encrypt a secret image into a meaningless noise-like format, which could easily attract attention from attackers who then may try to crack it. On the other hand, image steganography typically embeds secret image data within a cover image, but it inevitably incurs a lot of distortion to the cover image. This makes the secret image data vulnerable to attacks by steganalysis tools. In light of the above, in this paper, we propose a Visually Meaningful Image Encryption (VMIE) scheme via image-to-image reversible transformation based on the Glow model. In this scheme, a secret image is encoded and compressed as a latent vector by the deep compression autoencoder. Then, the latent vector is scrambled and inputted into the Glow model to generate a visually meaningful encrypted image. Extensive experiments demonstrate that the proposed VMIE scheme not only provides desirable security against attacks, but also enables the reconstruction of the original images with negligible quality loss. Codes are available athttps://github.com/AIMS-Group-ZhiliZhou/VMEI.

Federated Learning (FL) enables collaborative machine learning training while preserving data privacy. However, reliance on a central server of the typical FL confronts the risk of single server failure. Decentralized Federated Learning (DFL) emerges as a promising distributed framework, allowing clients to directly share models without server intervention, thereby addressing this challenge. Nevertheless, due to its decentralized nature, DFL is highly susceptible to Byzantine attacks orchestrated by malicious clients. Existing Byzantine-resilient DFL algorithms, though few, remain vulnerable to adaptive attacks due to their heavy reliance ongradient checksof local models, which can be adaptively manipulated by intelligent adversaries. To tackle this issue, we propose a DFL aggregation scheme called FORCE (Byzantine-Resilient Decentralized Federated Learning via Game-Theoretic Contribution Aggregation). Drawing inspiration from the Shapley value in game theory, FORCE shifts from gradient-checking approaches to employ a universal metric, the loss of the local model—independent of specific gradients, to identify potentially malicious clients. Moreover, to reduce the computational overhead of FORCE as the number of neighboring clients scales up, we propose a computationlightweight variant, FORCE−, which is optimized through approximating Shapley value computation. This variant becomes more scalable for resource-restricted DFL clients that are also aggregators. Experimental results on three diverse datasets (two image modalities and one textual modality datasets) under three existing attacks demonstrate that FORCE outperforms existing state-of-the-art Byzantine-resilient DFL aggregation methods, effectively defending against Byzantine attacks.

Streaming graphs widely exist in various application domains due to their excellent capability to capture temporal relationships between different entities. In recent years, outsourcing streaming graphs to the cloud for storage and analytics has become increasingly popular. Among others, pattern detection on streaming graphs, which aims to continuously detect subgraphs matching a given query pattern, benefits practical applications like credit card fraud detection and cyber-attack detection. However, conducting such streaming graph analytics in the cloud also raises critical privacy concerns. This paper introduces GraphGuard, the first system aimed at privacy-preserving pattern detection on outsourced streaming graphs. GraphGuard is designed through a tailored synergy of insights from graph modeling, lightweight secret sharing, edge differential privacy, and data encoding/padding. It conceals edge and vertex labels, as well as the relationship between vertices, for both the outsourced streaming graph and query pattern. We implement GraphGuard and perform comprehensive performance evaluations. The results show that GraphGuard is able to securely perform one detection on a streaming graph's snapshot (with a sliding time window of size 50,000) in just a few seconds. In comparison to a baseline utilizing general secure multiparty computation techniques, GraphGuard is up to 60× faster in query latency and achieves up to 98% savings in communication.

Reversible data hiding over encrypted images (RDH-EI) is an important technique for secure cloud image management but existing schemes often exhibit high computational complexity, low embedding rates, and excessive data expansion. This article addresses these issues by analyzing block-based secret sharing, revealing significant intra-block data redundancy. Based on this observation, we propose two space-preserving methods: the direct space-vacating method and the image-shrinking-based space-vacating method. Using these techniques, we design two novel RDH-EI schemes: a high-capacity RDH-EI scheme and a size-reduced RDH-EI scheme. The high-capacity RDH-EI scheme directly creates embedding space in encrypted images, eliminating the need for complex space-vacating operations and achieving higher and more stable embedding rates. In contrast, the size-reduced RDH-EI scheme minimizes data expansion by discarding unnecessary shares, resulting in smaller encrypted images. Experimental results show that the high-capacity RDH-EI scheme outperforms existing methods in terms of embedding capacity, while the size-reduced RDH-EI scheme achieves strong performance in minimizing data expansion. Both schemes offer effective solutions for RDH-EI challenges.

The field of privacy-preserving computation has recently seen a surge in specialized methods for Private Set Intersection (PSI) and Private Set Union (PSU). The primary focus of existing research lies in the multi-party setting, where set owners collaboratively execute PSI/PSU protocols on their sets. Limited research has investigated the more scalable outsourced service setting, where set owners secretly share their sets among a set of servers that collaboratively provide PSI/PSU query services over the secret-shared data. In this paper, we present FastPSC, a new system design supporting maliciously secure PSI/PSU in the outsourced service setting. FastPSC delicately bridges lightweight secure computation techniques and differential privacy mechanisms. The key insight is to leverage differentially private leakage to achieve a significant efficiency boost in secure and accurate intersection and union query services. Experiments show that with differentially private leakage allowed, FastPSC can achieve a significant performance advantage over the-state-of-the-art prior works without differentially private leakage. Specifically, compared with the work by Mohasselet al.(CCS'20) with semi-honest security, FastPSC achieves a$1.5\times$–$52.2\times$speedup and reduces server-side communication cost by 78%–98%. Compared with the work by Asharovet al.(CCS'23) with malicious security, FastPSC achieves a$4.2\times$–$7.4\times$speedup and reduces server-side communication cost by 99%.

Federated learning (FL) enables multiple clients to jointly train a model by sharing only gradient updates for aggregation instead of raw data. Due to the transmission of very high-dimensional gradient updates from many clients, FL is known to suffer from a communication bottleneck. Meanwhile, the gradients shared by clients as well as the trained model may also be exploited for inferring private local datasets, making privacy still a critical concern in FL. We present Clover, a novel system framework for communication-efficient, secure, and differentially private FL. To tackle the communication bottleneck in FL, Clover follows a standard and commonly used approach---top-k gradient sparsification, where each client sparsifies its gradient update such that only k largest gradients (measured by magnitude) are preserved for aggregation. Clover provides a tailored mechanism built out of a trending distributed trust setting involving three servers, which allows to efficiently aggregate multiple sparse vectors (top-k sparsified gradient updates) into a dense vector while hiding the values and indices of non-zero elements in each sparse vector. This mechanism outperforms a baseline built on the general distributed ORAM technique by several orders of magnitude in server-side communication and runtime, with also smaller client communication cost. We further integrate this mechanism with a lightweight distributed noise generation mechanism to offer differential privacy (DP) guarantees on the trained model. To harden Clover with security against a malicious server, we devise a series of lightweight mechanisms for integrity checks on the server-side computation. Extensive experiments show that Clover can achieve utility comparable to vanilla FL with central DP and no use of top-k sparsification. Meanwhile, achieving malicious security introduces negligible overhead in client-server communication, and only modest overhead in server-side communication and runtime, compared to the semi-honest security counterpart.

Graph learning has garnered increasing attention in recent years, which aims to train machine learning models over graph data to support various graph analytic tasks. Coming with the popularity of graph learning are critical privacy concerns regarding the information-rich graphs in many application domains (e.g., finance, social networks, and healthcare). There is thus an urgent call for privacy-preserving graph learning. In this paper, we target an emerging decentralized graph scenario, where a graph is fully decentralized among a set of nodes in such a way that each node only has a limited local view about the global graph. We propose PDGL, a new system framework that can effectively support privacy-assured model training over a decentralized graph, with privacy protection for the links among the nodes as well as the nodes’ private feature data and labels. In contrast to PDGL, prior work does not provide protection for the nodes’ links, feature data, and labels simultaneously. Extensive experiments demonstrate that while providing strong privacy protection for decentralized graph data, PDGL can achieve model utility comparable to the baseline setting of centralized graph learning.

Decentralized social graphs, where no single entity possesses the information of the entire graph, and each user maintains only a limited view of the graph, contain great value for different applications. However, simply collecting local views for analytics raises privacy concerns due to the sensitive information of social relationships they capture. To address this, a canonical approach involves privately fitting a generative graph model to the decentralized social graph, generating a differentially private synthetic graph that serves as a proxy for analytics. Existing solutions, however, often fail to capture the inherent directionality of edges and attribute-edge correlations when dealing with decentralized directed social graphs, leading to synthetic graphs with poor utility. To bridge this gap, we present PrivAGM, a new solution that harnesses the synergies among differential privacy, secure multiparty computation, and generative graph models, enabling the secure construction of differentially private directed attributed graph models on decentralized social graphs while ensuring the privacy preservation of individuals. We evaluate PrivAGM on three real-world directed social graph datasets. The results show that PrivAGM outperforms the state-of-the-art methods, generating synthetic graphs with significantly higher utility.

Due to the properties of autonomy and scalability, decentralized storage networks (DSNs) leveraging blockchain technology have attracted growing attention. Integrity auditing and verifiable searchable encryption are two essential functions for DSNs. The former ensures reliable and fair storage services, while the latter enables users to conduct keyword searches over encrypted data and guarantees the public verifiability of search results. However, all existing research in DSN has focused either on integrity auditing or on verifiable searchable encryption separately. In this paper, we propose a novel scheme for encrypted decentralized storage that simultaneously supports verifiable search and integrity auditing. It employs a unified proof and supports one-time proof verification to validate both the correctness of the returned file identifiers and the integrity of the files associated with these identifiers. As a result, compared to previous schemes supporting only integrity auditing, our scheme maintains a similar proof size and the support for search result verification does not significantly increase the on-chain storage overhead. Additionally, our scheme allows users to dynamically update their outsourced files while ensuring forward security during the file insertion process. We formally analyze the correctness and security of our scheme, and implement a system prototype to evaluate its performance. The experimental results demonstrate that it achieves verifiable searchable encryption and integrity auditing with practically affordable overhead.

As deepfake technology poses severe threats to information security, significant efforts have been devoted to deepfake detection. To enable model generalization for detecting new types of deepfakes, it is required that the existing models should learn knowledge about new types of deepfakes without losing prior knowledge, a challenge known as catastrophic forgetting (CF). Existing methods mainly utilize domain adaptation to learn about the new deepfakes for addressing this issue. However, these methods are constrained to utilizing a small portion of data samples from the new deepfakes, and they suffer from CF when the size of the data samples used for domain adaptation increases. This resulted in poor average performance in source and target domains. In this paper, we introduce a novel approach to boost the generalizability of deepfake detection. Our approach follows a two-stage training process: training in the source domain (prior deepfakes that have been used for training) and domain adaptation to the target domain (new types of deepfakes). In the first stage, we employ expansive learning to train our expanded model from a well-trained teacher model. In the second stage, we transfer the expanded model to the target domain while removing assistant components. For model architecture, we propose the frequency extraction module to extract frequency features as complementary to spatial features and introduce spatial-frequency contrastive loss to enhance feature learning ability. Moreover, we develop a confidence judgement module to eliminate conflicts between new and prior knowledge. Experimental results demonstrate that our method can achieve better average accuracy in source and target domains even when using large-scale data samples of the target domain, and it exhibits superior generalizability compared to state-of-the-art methods.

Multi-replica storage is an advanced extension of traditional cloud storage that allows data owners to customize the number of backups for file blocks based on their relative importance. In such settings, remote auditing mechanisms are essential for verifying data integrity and ensuring that the cloud service provider (CSP) maintains the pre-negotiated number of replicas. However, existing schemes often expose block positions and backup quantities to the CSP, making users' data vulnerable to template attacks. Meanwhile, secure deduplication significantly reduces storage overhead and user costs while preserving data confidentiality. In this paper, we propose a novel multi-replica cloud storage scheme that, for the first time, simultaneously supports cross-user deduplication and integrity auditing in the ciphertext domain. The proposed scheme can not only protect data privacy from template attacks but also enable the elimination of redundant ciphertext replicas and audit authentication tags across users at the block level. Formal analysis validates the correctness and security guarantees of our scheme. Experimental results demonstrate its effectiveness with modest overhead.

Protocol Reverse Engineering (PRE) serves as the foundation for numerous security analysis techniques, such as vulnerability mining and intrusion detection, etc. The PRE analysis precision can directly affect the accuracy of these downstream techniques. The network-trace-based PRE technique has become the mainstream PRE technique attributed to its ease of implementation. However, without the prerequisite of additional dedicated devices or knowledge of information, the analysis precision of existing network-trace-based PRE methods is often achievable at only byte or half-byte level but not the fine-grained bit-level, which makes it increasingly challenging to meet the precision requirements of those downstream security applications. In this work, we propose a fine-grained PRE scheme, named FineBID, which makes the identification capability in a fine-grained manner for existing network-trace-based PRE methods into bit-level fields. FineBID follows the global characteristics of protocol fields and constructively models the bit-level field identification problem as a multi-objective decision model, which thus effectively overcomes the insufficient representativeness of bit-level fields’ local characteristics. Then, the multi-objective decision model is solved to obtain the Pareto solution set for different field segmentation levels, and the utility value per bit is further computed. The utility value can be used as the immediate indicator to determine whether each bit is a field boundary or not. Meanwhile, we propose an Actual Ground Truth that is more in line with the actual usage of each bit. With extensive experiments on the Internet, wireless, and industrial protocols, we affirm that FineBID can not only significantly reduce the search space for Ground Truth or Actual Ground Truth with a space reduction of 95.3% compared to exhaustive search, but also identify Ground Truth or Actual Ground Truth more accurately than other similar methods.

Organizations these days capitalize on crowdsourcing to learn collective wisdom from a population of individuals. Vast amounts of data have been gathered, making the crowdsourcing platforms a lucrative target to steal data from and thus raising severe privacy concerns. Data contributed by workers may carry sensitive individual information. Meanwhile, organizations deem the aggregate statistics as intellectual property. In this paper, we propose, design, and evaluate GoCrowd, a system framework for obliviously aggregating wisdom with quality assurance in crowdsourcing. At its core, we propose constructions for two procedures. The starting point is a gold-standard based private worker quality control procedure that provides privacy-friendly worker quality assurance under the widely popular gold-standard mechanism. The subsequent procedure is an oblivious wisdom aggregation procedure that obliviously learns aggregate statistics over workers’ data while considering their quality. We securely realize these procedures with only lightweight secret sharing techniques. Our system is utterly oblivious to the service provider, and ensures that only the requester can learn the aggregate quality-aware statistics but nothing more. Extensive evaluations show that GoCrowd can produce quality statistics over data from 500 workers for 200 16-choice questions within 1 s.

Passwords are widely used for authentication in Internet applications. Recently, users tend to adopt graphical passwords instead of traditional alphanumeric passwords, since it is much easier for humans to remember images than verbal representations. However, the existing graphical password authentication systems generally suffer from three main issues. 1) It is required to remember and perform complicated operations during the registration/login phases, which significantly limits the systems’ usability; 2) The users’ passwords are simply stored as plaintexts in servers, and thus the security is compromised; 3) The users need to register/login to each server separately when they are applied in multi-server environment. To address the above issues, we propose a user-friendly and secure Google map-based graphical password (FS-GMGP) system using tolerant distance and homomorphic encryption. By using a homomorphic encryption scheme, each user encrypts his password point and response point selected on Google map, while the servers compute and decrypt the distance between the two encrypted points and then compare the resulting value with a tolerant distance for authentication. Moreover, the FS-GMGP system is extended for multi-server environment. The evaluation results and security analysis show that the FS-GMGP and its extended version achieve desirable usability and security in single-server environment and multi-server environment, respectively.

Model watermarking is a widely adopted mechanism for protecting deep learning (DL) model intellectual property (IP). Black-box verifiable watermarking typically involves injecting backdoors that cause the model to produce predetermined outputs for specific inputs. In contrast, white-box verifiable watermarking uses steganographic techniques to embed watermarks into weight parameters or activation values. However, the former poses new security risks, while the latter often lacks robustness against removal techniques. In this paper, we propose a latent model watermarking, constructing upon the model Division and Union operating concept, dubbed as DUO, leveraging the strengths of two watermarking methods above while eliminating each shortcoming. Once the model owner or provider embeds a watermark into the model using watermark data, the watermarked model is divided into two parts: the main model, which corresponds to the primary task and is made publicly available, and a small sub-network privately reserved by the owner. The watermark resides latently within the main model and can only be activated through the private sub-network (the reserved parameters) when they are united. Consequently, DUO does not adversely affect the performance of the main model on its primary task and does not induce any security risks, even in the presence of watermark data. We extensively validate DUO on four benchmark datasets (CIFAR-10, ImageNette, CIFAR-100, and Tiny-ImageNet) using various model architectures, including standardized ResNet and VGG. The results affirm its capability to accurately verify model ownership without compromising model accuracy. It exhibits a 100% detection accuracy on pirated/positive testing models (96 models are tested) with a 0% false positive rate on normal/negative testing models (64 models are tested). Due to its latent nature, DUO is both effective and robust, capable of withstanding a wide range of state-of-the-art watermark laundering including severe model fine-tuning and pruning. We further evaluate and demonstrate that DUO remains robust against adaptive attacks, even when both the watermark data and the reserved parameters are known to the adversary.

In contrast to prevalent Federated Learning (FL) privacy inference techniques such as generative adversarial networks attacks, membership inference attacks, property inference attacks, and model inversion attacks, we devise an innovative privacy threat: the Data Distribution Decompose Attack on FL, termedDecaf. This attack enables an honest-but-curious FL server to meticulously profile the proportion of each class owned by the victim FL user, divulging sensitive information like local market item distribution and business competitiveness. The crux ofDecaflies in the profound observation that the magnitude of local model gradient changes closely mirrors the underlying data distribution, including the proportion of each class.Decafaddresses two crucial challenges: accurately identify the missing/null class(es) given by any victim user as a premise and then quantify the precise relationship between gradient changes and each remaining non-null class. Notably,Decafoperates stealthily, rendering it entirely passive and undetectable to victim users regarding the infringement of their data distribution privacy. Experimental validation on five benchmark datasets (MNIST, FASHION-MNIST, CIFAR-10, FER-2013, and SkinCancer) employing diverse model architectures, including customized convolutional networks, standardized VGG16, and ResNet18, demonstratesDecaf’s efficacy. Results indicate its ability to accurately decompose local user data distribution, regardless of whether it is IID or non-IID distributed. Specifically, the dissimilarity measured using$L_{\infty }$distance between the distribution decomposed byDecafand ground truth is consistently below 5% when no null classes exist. Moreover,Decafachieves 100% accuracy in determining any victim user’s null classes, validated through formal proof.

End-to-end encrypted messaging services (EEMSs) empower private communication through encrypting messages, yet also make content moderation for combating the spread of abusive messages challenging. There is an urgent call for supporting content moderation in EEMSs while ensuring user privacy. In this paper, we present a new system design for privacy-assured content moderation in EEMSs. At a high level, users in our system can privately report abusive messages, and the EEMS traces the source if a message has an aggregated report count exceeding a predefined threshold and is audited to be abusive. Our system mainly departs from prior works in that it allows flexible and adaptable thresholds, offers robustness against dishonest reporters providing malformed reports, and better ensures the privacy of all users during the moderation process. We also take a step further and propose a privacy-aware detection mechanism that relies on a blocklist built with transparency to mitigate the further spread of identified abusive messages from forwarders. Formal security analysis is provided and extensive experiments demonstrate the practical efficiency of our system.

Time-series data is prevalent in many applications like smart homes, smart grids, and healthcare. And it is now increasingly common to store and query time-series data in the cloud. Despite the benefits, data privacy concerns in such outsourced services are pressing, making it imperative to embed privacy assurance mechanisms from the outset. Most existing related works have been focused on querying for different types of aggregate statistics. In this article, we instead focus on the secure support for advanced interval skyline queries, which allow to identify time series that are not dominated by any other time series within a query time interval. This is valuable for time-series data analytics in applications like remote health monitoring (e.g., identifying patients with high heart rates in a certain week). We present OblivTime, a new system framework for oblivious and efficient interval skyline query processing over encrypted time-series data. OblivTime is built from a synergy of time-series data analytics, lightweight cryptography, and GPU parallel computing, achieving stronger security guarantees and lower online query latency over the state-of-the-art prior work. Extensive experiments demonstrate that OblivTime can achieve up to$666\times$speedup in online query latency over the state-of-the-art prior work.

The proliferation of cloud computing has spurred the popularity of storing and querying databases in the cloud. Among others, skyline queries play an important role in the database field due to its usefulness in multi-criteria decision support systems. To accommodate the tailored needs of users, user-defined skyline query has recently emerged, allowing users to define custom preferences in their skyline queries. However, user-defined skyline query services, if deployed in the cloud, may raise critical privacy concerns as the outsourced databases and skyline queries may contain proprietary/privacy-sensitive information. In light of the above, this paper presents ObliuSky, a new solution enabling oblivious user-defined skyline query processing in the cloud. ObliuSky departs from prior work by not only providing confidentiality protection for the content of the outsourced database, the user-defined skyline queries, and the query results, but also hiding the data patterns (e.g., user-defined dominance relations among database points and search access patterns) which may indirectly cause data leakages. We formally analyze the security guarantees and conduct extensive performance evaluations. The results show that while achieving much stronger security guarantees than the state-of-the-art prior work, ObliuSky is superior in database and query encryption efficiency, and scalable in oblivious query processing.

Spatial crowdsourcing (SC) has recently emerged as a new crowdsourcing service paradigm, where workers move physically to designated locations to perform tasks. Most SC systems perform task assignment based on the spatial proximity between task locations and worker locations. Under such a strategy, workers can only perform tasks near them, which may result in low social welfare (i.e., the total profit of the platform and workers). In contrast, the newly emerging strategy of competitive task assignment (CTA) stimulates workers to compete for their preferred tasks, allowing optimization of the overall profit of SC systems. Among others, one novel CTA setting is competitive detour tasking, which allows workers to compete for tasks that need them to make detours from their original travel paths. However, it requires collecting each worker’s bidding profile which may expose private information. In light of this, in this article, we design, implement, and evaluate PrivCO, a new system framework enabling privacy-preserving competitive detour tasking services in SC. PrivCO delicately bridges state-of-the-art competitive detour tasking algorithms with lightweight cryptography, providing strong protections for workers’ bidding profiles. Extensive experiments over real-world datasets demonstrate that while offering strong security guarantees, PrivCO achieves social welfare comparable to the plaintext domain.

Federated learning (FL) has rapidly become a compelling paradigm that enables multiple clients to jointly train a model by sharing only gradient updates for aggregation, without revealing their local private data. In order to protect the gradient updates which could also be privacy-sensitive, there has been a line of work studying local differential privacy (LDP) mechanisms to provide a formal privacy guarantee. With LDP mechanisms, clients locally perturb their gradient updates before sharing them out for aggregation. However, such approaches are known for greatly degrading the model utility, due to heavy noise addition. To enable a better privacy-utility trade-off, a recently emerging trend is to apply the shuffle model of DP in FL, which relies on an intermediate shuffling operation on the perturbed gradient updates to achieve privacy amplification. Following this trend, in this paper, we present Camel, a new communication-efficient and maliciously secure FL framework in the shuffle model of DP. Camel first departs from existing works by ambitiously supporting integrity check for the shuffle computation, achieving security against malicious adversary. Specifically, Camel builds on the trending cryptographic primitive of secret-shared shuffle, with custom techniques we develop for optimizing system-wide communication efficiency, and for lightweight integrity checks to harden the security of server-side computation. In addition, we also derive a significantly tighter bound on the privacy loss through analyzing the Rényi differential privacy (RDP) of the overall FL process. Extensive experiments demonstrate that Camel achieves better privacy-utility trade-offs than the state-of-the-art work, with promising performance.

To implement encrypted data deduplication in a cloud storage system, users must encrypt files using special encryption algorithms (e.g., convergent encryption (CE)), which cannot provide strong protection. The confidential level of an outsourced file is determined by the user himself/herself subjectively or by the owner number of the file objectively. These files owned by a few users are considered strictly confidential and require strong protection. In this paper, we design, analyze and implement LSDedup, which attains a high storage efficiency while providing strictly confidential files (SCFiles) with strong protection. LSDedup allows cloud users to securely interact with cloud servers to check the confidential level of an outsourced file. Users encrypt the SCFiles using standard symmetric encryption algorithms to achieve a high security level, whereas encrypting the less confidential files (LSFiles) using CE such that cloud servers can perform deduplication. LSDedup is designed to prevent cloud servers reporting fake confidential level and a fake file user claiming the ownership of the file. Formal analysis is provided to justify its security. Besides, we implement an LSDedup prototype using Alibaba Cloud as backend storage. Our evaluations demonstrate that LSDedup can work with existing cloud service providers’ APIs and achieves modest performance overhead.

Deep neural networks (DNNs) are vulnerable to backdoor attacks, where a backdoored model behaves normally with clean inputs but exhibits attacker-specified behaviors upon the inputs containing triggers. Most previous backdoor attacks mainly focus on either the all-to-one or all-to-all paradigm, allowing attackers to manipulate an input to attack a single target class. Besides, the two paradigms rely on a single trigger for backdoor activation, rendering attacks ineffective if the trigger is destroyed. In light of the above, we propose a new M-to-N attack paradigm that allows an attacker to manipulate any input to attack N target classes, and each backdoor of the N target classes can be activated by any one of its M triggers. Our attack selects M clean images from each target class as triggers and leverages our proposed poisoned image generation framework to inject the triggers into clean images invisibly. By using triggers with the same distribution as clean training images, the targeted DNN models can generalize to the triggers during training, thereby enhancing the effectiveness of our attack on multiple target classes. Extensive experimental results demonstrate that our new backdoor attack is highly effective in attacking multiple target classes and robust against pre-processing operations and existing defenses.

Cloud service is a natural choice to store and manage the exponentially produced images. Data privacy is one of the most concerned points in cloud-based image services. Reversible data hiding over encrypted images (RDH-EI) is an effective technique to securely store and manage confidential images in the cloud. However, existing RDH-EI schemes have obvious weaknesses such as reliable key management system dependence and single point of failure. To securely store and manage confidential images in the cloud, in this study, we propose a new reversible data hiding strategy via image secret sharing. We first design a secure (r,n)-threshold preprocessing-free matrix secret sharing (PFMSS) technique. It can directly sharem-bit data by matrix multiplication without preprocessing. Using the PFMSS, we further design a secure (r,n)-threshold reversible data hiding scheme over encrypted images. The content owner divides a confidential image intonshares without accessing to a secret encryption key, and then sends thenshares toncloud-based image servers from competing providers. For each share, some additional data, e.g., integrity and identification of the image, can be embedded into it and these data can also be losslessly extracted. An authorized receiver can recover the confidential image fromrshares. By designing, the content owner doesn’t need to access a secret key when encrypting the image and the scheme can withstandn-rpoints of failure. Simulation results show that our scheme can ensure image content confidentiality and has a much larger embedding capacity compared to state-of-the-art schemes.

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

Mobile crowdsensing (MCS) has rapidly emerged as a popular paradigm for sensory data collection and benefited various location-based services and applications like road monitoring, smart transportation, and environmental monitoring. In practice, there often exist data-missing regions in the target sensing area, due to factors like limited budget, large area size, and scarcity of participants. This poses a demand for data recovery, which is commonly done based on the compressive sensing (CS) technique. However, CS-based data recovery requires access to sensory data tagged with locations, raising critical concerns on participants’ location privacy. While a plethora of location privacy techniques exist, most of them breach the data correlation inherently required by CS-based data recovery. Meanwhile, existing works mostly focus on protecting locations and overlook sensory data which may also indirectly lead to location leakages. In this paper, we propose SecDR, a new system design supporting secure, efficient, and accurate data recovery for location-based MCS applications. SecDR protects both locations and sensory data, and is built from a delicate synergy of CS-based data recovery and lightweight cryptography techniques. Extensive evaluations demonstrate that SecDR achieves promising performance and, even with stronger security guarantees, outperforms the state-of-the-art, with accuracy close to the plaintext domain.

Words are treated as atomic units in natural language processing tasks and it is a fundamental step to represent them as vectors for supporting subsequent computations. GloVe is a widely used machine learning model to train word vectors. Generally, a large corpus and high computation resources are required to train high-quality word vectors using GloVe, making it difficult for users to train their own word vectors by themselves. A natural choice nowadays is to outsource the training process to the cloud. However, coming with such cloud-based training services are serious privacy concerns, which should be well addressed. In this paper, we design, implement, and evaluate PPGloVe, the first system framework that supports privacy-preserving word vectors training using GloVe over encrypted data of multiple participants. We first decompose the training task and show that previous privacy-preserving machine learning techniques are not practical for this task. We then construct a new secure training strategy to delicately bridge lightweight cryptographic techniques with GloVe in depth to support privacy-preserving GloVe training on the cloud. By design, the corpora of the participants and the trained word vectors are kept private along the whole training process. Extensive experiments over three datasets of different scales demonstrate that PPGloVe produces word vectors with promising quality comparable to plaintext training, with practically affordable overhead.

To mitigate recent insidious backdoor attacks on deep learning models, advances have been made by the research community. Nonetheless, state-of-the-art defenses are either limited to specific backdoor attacks (i.e., source-agnostic attacks) or non-user-friendly in that machine learning expertise and/or expensive computing resources are required. This work observes that all existing backdoor attacks have an inadvertent and inevitable intrinsic weakness, termed as non-transferability —that is, a trigger input hijacks a backdoored model but is not effective in another model that has not been implanted with the same backdoor. With this key observation, we propose non-transferability enabled backdoor detection to identify trigger inputs for a model-under-test during run-time. Specifically, our detection allows a potentially backdoored model-under-test to predict a label for an input. Moreover, our detection leverages a feature extractor to extract feature vectors for the input and a group of samples randomly picked from its predicted class label, and then compares the similarity between the input and the samples in the feature extractor’s latent space to determine whether the input is a trigger input or a benign one. The feature extractor can be provided by a reputable party or is a free pre-trained model privately reserved from any open platform (e.g., ModelZoo, GitHub, Kaggle) by a user and thus our detection does not require the user to have any machine learning expertise or perform costly computations. Extensive experimental evaluations on four common tasks affirm that our detection scheme has high effectiveness (low false acceptance rate) and usability (low false rejection rate) with low detection latency against different types of backdoor attacks.

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.

It is increasingly common for enterprises/organizations to outsource graph analytics services to the cloud. For example, enterprises may leverage the cloud to store and query large attributed graphs. Among others, subgraph matching over a large attributed graph is a common and fundamental query functionality for graph analytics. It aims to retrieve all isomorphic subgraphs for a small query graph and greatly benefits various application domains like cheminformatics, social network analysis, and anti-money laundering. Deploying subgraph matching service in the cloud, however, poses a threat to the privacy of the information-rich graph data as the cloud gains access to the attributed graph, query graph, and query result. Given this, several works have been proposed for supporting privacy-aware subgraph matching. However, prior works only consider a weak semi-honest threat model and cannot provide integrity guarantees for the subgraph matching results in case of malicious adversary. In light of this, we design, implement, and evaluate eGrass, a new system enabling maliciously secure attributed subgraph matching service outsourced to the cloud. In addition to offer protection for graph data confidentiality, eGrass is also designed to hide search access patterns as well as defend against malicious cloud server that attempt to compromise the result integrity. We conduct extensive experiments on a real-world dataset. The results demonstrate that compared to the state-of-the-art previous protocol with semi-honest security, eGrass is only 3 × −4.7× slower in query latency, uses 3 × −3.5× more communication, and does not require extra cloud-side storage.

Secure deduplication over encrypted data can greatly improve cloud storage efficiency and protect data privacy. Recently, there have been some research efforts aiming at designing secure deduplication schemes with the assistance of key servers (KSs). However, prior works are unsatisfactory in that they suffer from some limitations such as security degradation (the leakage at partial KSs will lead to all the ciphertexts being subject to offline brute-force attacks) or lack of scalability for handling the change of KSs. In this paper, we propose a new secure deduplication scheme for large-scale cloud storage service, which, to our best knowledge, is the first server-aided scheme that supports both tolerance of partial KSs leakage and dynamic change of KSs. Our scheme divides all the KSs into multiple groups and each KS group keeps a randomly generated secret key using threshold cryptography. We design a file-related KS group selection mechanism for assisting encryption key generation, which guarantees that the identical files of different users can be encrypted using the same keys. Our scheme is designed to update the KS groups regularly for supporting the joining and leaving of the KSs as well as maintaining long-term security. We leverage the blockchain to help divide KSs into groups in a fair way and securely migrate group secret keys during KS group updating. Formal analysis is provided to verify the correctness of our scheme and justify its security, and both theoretical and experimental results demonstrate that it has modest performance overhead.

Enforcing end-to-end data encryption is vital for protecting the interests of requesters in crowdsourcing services, who initiate crowdsourcing tasks and need to pay the service provider and reward workers for the crowdsourced data. It ensures that the data encrypted by workers can be decrypted by the requester only. This yet makes it challenging to protect workers in getting rewards as the data is now only accessible to the requester, who may falsely report workers’ data quality. There is thus an urgent call for enforcing end-to-end data encryption while achieving robustness against such false-reporting. However, this is not yet sufficient for worker protection because most platforms with quality awareness involve a screening process for worker selection in advance, which requires collecting personal worker profiles for assessment against task requirements and raises privacy concerns. In light of the above, we propose PrivRo, a new system framework for privacy-preserving crowdsourcing service with robust quality awareness. PrivRo supports private profile matching for secure screening as well as secure data collection with verifiable quality reporting through custom secure protocols. To our best knowledge, no prior work has simultaneously and adequately supported the secure functionalities compared to PrivRo. Extensive experiments demonstrate the practical efficiency of PrivRo.

Currently, a high demand for on-device deep neural network (DNN) model deployment is limited by the large model size, computing-intensive floating-point operations (FLOPS), and intellectual property (IP) infringements (i.e., easy access to model duplication for the avoidance of license payments). One appealing solution to addressing the first two concerns is model quantization, which reduces the model size and uses integer operations commonly supported by microcontrollers (MCUs usually do not support FLOPS). To this end, a 1-bit quantized DNN model or deep binary neural network (BNN) significantly improves the memory efficiency, where each parameter in a BNN model has only 1 bit. However, BNN cannot directly provide IP protection (in particular, the functionality of the model is locked unless there is a license payment). In this article, we propose a reconfigurable BNN (RBNN) to further amplify the memory efficiency for resource-constrained Internet of Things (IoT) devices while naturally protecting the model IP. Generally, RBNN can be reconfigured on demand to achieve any one of$M$($M>1$) distinct tasks with the same parameter set, thus only a single task determines the memory requirements. In other words, the memory utilization is improved by a factor of$M$. Our extensive experiments corroborate that up to seven commonly used tasks ($M=7$, six of these tasks are image related and the last one is audio) can co-exist (the value of$M$can be larger). These tasks with a varying number of classes have no or negligible accuracy drop-off (i.e., within 1%) on three binarized popular DNN architectures, including VGG, ResNet, and ReActNet. The tasks span across different domains, e.g., computer vision and audio domains validated herein, with the prerequisite that the model architecture can serve those cross-domain tasks. To fulfill the IP protection of an RBNN model, the reconfiguration can be controlled by both a user key and a device-unique root key generated by the intrinsic hardware fingerprint (e.g., SRAM memory power-up pattern). By doing so, an RBNN model can only be used per paid user per authorized device, thus benefiting both the user and the model provider. The source code is released athttps://github.com/LearningMaker/RBNN.

It is increasingly popular for cloud providers to offer middlebox service that supports content-based similarity detection for enterprises. However, redirecting network traffic to the cloud for such service raises security concerns. While trusted execution environments such as Intel SGX have emerged as a pragmatic solution for designing secure in-the-cloud middleboxes, it remains challenging to practically support content-based similarity detection. In this paper, we design a secure in-the-cloud middlebox system that can detect content-based similar flows in encrypted traffic dynamically. To cope with the constrained enclave memory, we adopt the caching technique and devise a compact index to increase the cache hit rate for effective similarity detection inside the enclave. We also present a parallel algorithm for performance speedup, with an efficient enclave thread management mechanism. Extensive evaluations demonstrate that the overhead of our system compared to native processing (without SGX) is limited to 2.1×. Meanwhile, our tailored design can achieve up to$14.4\times$better computational efficiency compared to simply moving the target functionality to the SGX enclave via existing popular library operating systems like Graphene-SGX and Occlum. Our secure system can achieve a normalized similarity detection precision of about 90%.

Cloud computing offers advantages in handling the exponential growth of images but also entails privacy concerns on outsourced private images. Reversible data hiding (RDH) over encrypted images has emerged as an effective technique for securely storing and managing confidential images in the cloud. Most existing schemes only work on uncompressed images. However, almost all images are transmitted and stored in compressed formats such as JPEG. Recently, some RDH schemes over encrypted JPEG bitstreams have been developed, but these works have some disadvantages such as a small embedding capacity (particularly for low quality factors), damage to the JPEG format, and file size expansion. In this study, we propose a permutation-based embedding technique that allows the embedding of significantly more data than existing techniques. Using the proposed embedding technique, we further design a large-capacity RDH scheme over encrypted JPEG bitstreams, in which a grouping method is designed to boost the number of embeddable blocks. The designed RDH scheme allows a content owner to encrypt a JPEG bitstream before uploading it to a cloud server. The cloud server can embed additional data (e.g., copyright and identification information) into the encrypted JPEG bitstream for storage, management, or other processing purpose. A receiver can losslessly recover the original JPEG bitstream using a decryption key. Comprehensive evaluation results demonstrate that our proposed design can achieve approximately twice the average embedding capacity compared to the best prior scheme while preserving the file format without file size expansion.

Traditional schemes for reversible data hiding in encrypted images (RDH-EI) focus on one data hider and cannot resist the single point of failure. Besides, the image security is determined by one party, rather than multiple parties. Thus, it is valuable to design RDH-EI schemes with multiple data hiders for stronger security. In this article, we propose a multiple data hiders-based RDH-EI scheme using a new secret sharing technique. First, we devise an$(r,n)$-threshold$(r\leq n)$matrix-based secret sharing (MSS) using matrix theory, and theoretically verify its efficacy and security properties. Then, using the MSS, we propose an$(r,n)$-threshold RDH-EI scheme called MSS-RDHEI. The content owner encrypts an image to be$n$encrypted images using the MSS with an encryption key, and outsources these encrypted images to$n$data hiders. Each data hider can embed some data, e.g., copyright and identification information, into the encrypted image for the purposes of storage, management, or other processing, and these data can also be losslessly extracted. An authorized receiver can recover the confidential image from$r$encrypted images. By designing, our MSS-RDHEI scheme can withstand$n-r$points of failure. Experimental results show that it ensures the image content confidentiality and achieves a much larger embedding capacity than state-of-the-art schemes.

Neural network (NN) inference services enrich many applications, like image classification, object recognition, facial verification, and more. These NN inference services are increasingly becoming an essential offering from cloud computing providers, where end-users’ data are offloaded to the cloud for inference under a customized model. However, current cloud-based inference services operate on clear inputs and NN models, raising paramount privacy concerns. Individual user data may contain private information that should always remain confidential. Meanwhile, the NN model is deemed proprietary to the model owner as model training requires substantial resources. In this article, we present, tailor, and evaluateSonic, a lightweight secure NN inference service delegated in the cloud.Sonicleverages the cloud computing paradigm to fully outsource the secure inference, freeing end devices and model owners from being actively online for assistance.Sonicguards both user input and model privacy along the whole service flow. We design a series of secure and efficient NN layer functions purely using lightweight cryptographic primitives. Extensive evaluations demonstrate thatSonicachieves up to$60\times$bandwidth saving in online inference compared to prior art.

Cloud computing promises great advantages in handling the exponential data growth. Secure deduplication can greatly improve cloud storage efficiency while protecting data confidentiality. In the meantime, when data are outsourced to the remote cloud, there is an imperative need to audit the integrity. Most existing works only consider the support for either secure deduplication or integrity auditing. Recently, there have been some research efforts aiming to integrate secure deduplication with integrity auditing. However, prior works are unsatisfactory in that they suffer from the leakage of ownership privacy and forgeability of auditing results for low-entropy data. In this paper, we propose a new scheme that delicately bridges secure deduplication and integrity auditing in encrypted cloud storage. In contrast with prior works, our scheme protects the ownership privacy and prevents the cloud service provider from forging the auditing results for low-entropy data. Furthermore, we propose a blockchain-based mechanism that helps to ensure key recoverability and reduce local storage cost of keys. Formal analysis is provided to justify the security guarantees. Experiment results demonstrate the modest performance overhead of our scheme.

Federated learning has recently emerged as a paradigm promising the benefits of harnessing rich data from diverse sources to train high quality models, with the salient features that training datasets never leave local devices. Only model updates are locally computed and shared for aggregation to produce a global model. While federated learning greatly alleviates the privacy concerns as opposed to learning with centralized data, sharing model updates still poses privacy risks. In this paper, we present a system design which offers efficient protection of individual model updates throughout the learning procedure, allowing clients to only provide obscured model updates while a cloud server can still perform the aggregation. Our federated learning system first departs from prior works by supporting lightweight encryption and aggregation, and resilience against drop-out clients with no impact on their participation in future rounds. Meanwhile, prior work largely overlooks bandwidth efficiency optimization in the ciphertext domain and the support of security against an actively adversarial cloud server, which we also fully explore in this paper and provide effective and efficient mechanisms. Extensive experiments over several benchmark datasets (MNIST, CIFAR-10, and CelebA) show our system achieves accuracy comparable to the plaintext baseline, with practical performance.

Outsourcing decision tree inference services to the cloud is highly beneficial, yet raises critical privacy concerns on the proprietary decision tree of the model provider and the private input data of the client. In this paper, we design, implement, and evaluate a new system that allows highly efficient outsourcing of decision tree inference. Our system significantly improves upon prior art in the overall online end-to-end secure inference service latency at the cloud as well as the local-side performance of the model provider. We first present a new scheme which securely shifts most of the processing of the model provider to the cloud, resulting in a substantial reduction on the model provider's performance complexities. We further devise a scheme which substantially optimizes the performance for secure decision tree inference at the cloud, particularly the communication round complexities. The synergy of these techniques allows our new system to achieve up to$8 \times$better overall online end-to-end secure inference latency at the cloud side over realistic WAN environment, as well as bring the model provider up to$19 \times$savings in communication and$18 \times$savings in computation.

The rapid growth of the Internet of Vehicles (IoV) paradigm sparks the generation of large volumes of distributed data at vehicles, which can be harnessed to build models for intelligent applications. Federated learning has recently received wide attentions, which allows model training over distributed datasets without requiring raw datasets to be shared out. However, federated learning is known to be vulnerable to poisoning attacks, where malicious clients may manipulate the local datasets or model updates to corrupt the global model. Such attacks have to be countered when federated learning is adopted in IoV systems, given that the training process is distributed among a large number of vehicles in an open environment. In addition, IoV systems present a hierarchical architecture in practice where other types of nodes sit between the cloud server and vehicles, allowing intermediate aggregation for reducing overall training latency. Yet the intermediate aggregation nodes may also pose threats. In this paper, we propose a robust hierarchical federated learning framework named RoHFL, which allows hierarchical federated learning to be suitably applied in the IoV with robustness against poisoning attacks. We develop a robust model aggregation scheme that contains a logarithm-based normalization mechanism to cope with scaled gradients from malicious vehicles. We integrate the notion of reputation into the aggregation process and develop a scheme for reputation updating. We provide a formal analysis of RoHFL’s convergence guarantees. Experiment results over several popular datasets demonstrate the promising performance of RoHFL, which is superior to prior work in the robustness against poisoning attacks.

Graphs are widely used to model complex structured data in many applications. With the proliferation of cloud computing, it is popular to store and query graphs in the cloud. Among others, graph similarity search, which aims to retrieve from a graph database graphs similar to a query graph, has received wide attentions and benefited various domains such as cheminformatics, computer vision, and more. Deploying graph similarity search services on the cloud, however, raises critical privacy concerns on the information-rich graphs. In this article, we initiate the first study on privacy-preserving graph similarity search in cloud computing. We design, implement, and evaluate PrigSim, a novel system allowing the cloud to host an outsourced encrypted graph database and support secure graph similarity search, where the graph similarity is measured by the well-known metric called graph edit distance. PrigSim is built from a customized and delicate synergy of insights on graph modelling, lightweight cryptography, and data encoding and padding, providing protections for the confidentiality of data content associated with graphs, as well as hiding the connections among vertices. Extensive experiments demonstrate that the security design of PrigSim is accuracy-preserving, and presents modest performance overheads (with$9\times$-$15\times$higher query latency than the plaintext baseline).

Analytics over social graphs allows to extract valuable knowledge and insights for many fields like community detection, fraud detection, and interest mining. In practice, decentralized social graphs frequently arise, where the social graph is not available to a single entity and is decentralized among a large number of users, each holding only a limited local view about the whole graph. Collecting the local views for analytics of decentralized social graphs raises critical privacy concerns, as they encode private information about the social interactions among individuals. In this paper, we design, implement, and evaluate PrivGED, a new system aimed at privacy-preserving analytics over decentralized social graphs. PrivGED focuses on the support for eigendecomposition, one popular and fundamental graph analytics task producing eigenvalues/eigenvectors over the adjacency matrix of a social graph and benefits various practical applications. PrivGED is built from a delicate synergy of insights on graph analytics, lightweight cryptography, and differential privacy, allowing users to securely contribute their local views on a decentralized social graph for a cloud-based eigendecomposition analytics service while gaining strong privacy protection. Extensive experiments over real-world social graph datasets demonstrate that PrivGED achieves accuracy comparable to the plaintext domain, with practically affordable performance superior to prior art.

The well-known benefits of cloud computing have spurred the popularity of database service outsourcing, where one can resort to the cloud to conveniently store and query databases. Coming with such popular trend is the threat to data privacy, as the cloud gains access to the databases and queries which may contain sensitive information, like medical or financial data. A large body of work has been presented for querying encrypted databases, which has been mostly focused on secure keyword search. In this paper, we instead focus on the support for secure skyline query processing over encrypted outsourced databases, where little work has been done. Skyline query is an advanced kind of database query which is important for multi-criteria decision-making systems and applications. We propose SecSkyline, a new system framework building on lightweight cryptography for fast privacy-preserving skyline queries. SecSkyline ambitiously provides strong protection for not only the content confidentiality of the outsourced database, the query, and the result, but also for data patterns that may incur indirect data leakages, such as dominance relationships among data points and search access patterns. Extensive experiments demonstrate that SecSkyline is substantially superior to the state-of-the-art in query latency, with up to$813\times$improvement.

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.

Graphs are widely used to model the complex relationships among entities. As a powerful tool for graph analytics, graph neural networks (GNNs) have recently gained wide attention due to its end-to-end processing capabilities. With the proliferation of cloud computing, it is increasingly popular to deploy the services of complex and resource-intensive model training and inference in the cloud due to its prominent benefits. However, GNN training and inference services, if deployed in the cloud, will raise critical privacy concerns about the information-rich and proprietary graph data (and the resulting model). While there has been some work on secure neural network training and inference, they all focus on convolutional neural networks handling images and text rather than complex graph data with rich structural information. In this article, we design, implement, and evaluate SecGNN, the first system supporting privacy-preserving GNN training and inference services in the cloud. SecGNN is built from a synergy of insights on lightweight cryptography and machine learning techniques. We deeply examine the procedure of GNN training and inference, and devise a series of corresponding secure customized protocols to support the holistic computation. Extensive experiments demonstrate that SecGNN achieves comparable plaintext training and inference accuracy, with promising performance.

Vertical federated learning (VFL) has recently emerged as an appealing distributed paradigm empowering multi-party collaboration for training high-quality models over vertically partitioned datasets. Gradient boosting has been popularly adopted in VFL, which builds an ensemble of weak learners (typically decision trees) to achieve promising prediction performance. Recently there have been growing interests in using decision table as an intriguing alternative weak learner in gradient boosting, due to its simpler structure, good interpretability, and promising performance. In the literature, there have been works on privacy-preserving VFL for gradient boosted decision trees, but no prior work has been devoted to the emerging case of decision tables. Training and inference on decision tables are different from that in the case of generic decision trees, not to mention gradient boosting with decision tables in VFL. In light of this, we design, implement, and evaluate Privet, the first system framework enabling privacy-preserving VFL service for gradient boosted decision tables. Privet delicately builds on lightweight cryptography and allows an arbitrary number of participants holding vertically partitioned datasets to securely train gradient boosted decision tables. Extensive experiments over several real-world datasets and synthetic datasets demonstrate that Privet achieves promising performance, with utility comparable to plaintext centralized learning.

Due to its powerful feature learning capability and high efficiency, deep hashing has achieved great success in large-scale image retrieval. Meanwhile, extensive works have demonstrated that deep neural networks (DNNs) are susceptible to adversarial examples, and exploring adversarial attack against deep hashing has attracted many research efforts. Nevertheless, backdoor attack, another famous threat to DNNs, has not been studied for deep hashing yet. Although various backdoor attacks have been proposed in the field of image classification, existing approaches failed to realize a truly imperceptive backdoor attack that enjoys invisible triggers and clean label setting simultaneously, and they cannot meet the intrinsic demand of image retrieval backdoor.

Federated learning is a newly emerging distributed learning framework that facilitates the collaborative training of a shared global model among distributed participants with their privacy preserved. However, federated learning systems are vulnerable to Byzantine attacks from malicious participants, who can upload carefully crafted local model updates to degrade the quality of the global model and even leave a backdoor. While this problem has received significant attention recently, current defensive schemes heavily rely on various assumptions, such as a fixed Byzantine model, availability of participants' local data, minority attackers, IID data distribution, etc. To relax those constraints, this paper presents Robust-FL, the first prediction-based Byzantine-robust federated learning scheme where none of the assumptions is leveraged. The core idea of the Robust-FL is exploiting historical global model to construct an estimator based on which the local models will be filtered through similarity detection. We then cluster local models to adaptively adjust the acceptable differences between the local models and the estimator such that Byzantine users can be identified. Extensive experiments over different datasets show that our approach achieves the following advantages simultaneously: (i) independence of participants' local data, (ii) tolerance of majority attackers, (iii) generalization to variable Byzantine model.

In this paper, we propose CryptMed, a system framework that enables medical service providers to offer secure, lightweight, and accurate medical diagnostic service to their customers via an execution of neural network inference in the ciphertext domain. CryptMed ensures the privacy of both parties with cryptographic guarantees. Our technical contributions include: 1) presenting a secret sharing based inference protocol that can well cope with the commonly-used linear and non-linear NN layers; 2) devising optimized secure comparison function that can efficiently support comparison-based activation functions in NN architectures; 3) constructing a suite of secure smooth functions built on precise approximation approaches for accurate medical diagnoses. We evaluate CryptMed on 6 neural network architectures across a wide range of non-linear activation functions over two benchmark and four real-world medical datasets. We comprehensively compare our system with prior art in terms of end-to-end service workload and prediction accuracy. Our empirical results demonstrate that CryptMed achieves up to respectively 413 ×, 19 ×, and 43 × bandwidth savings for MNIST, CIFAR-10, and medical applications compared with prior art. For the smooth activation based inference, the best choice of our proposed approximations preserve the precision of original functions, with less than 1.2% accuracy loss and could enhance the precision due to the newly introduced activation function family.

Cloud computing has been a dominant paradigm for a variety of information processing platforms, particularly for enabling various popular applications of sharing economy. However, there is a major concern regarding data privacy on these cloud-based platforms. This work presents novel cloud-based privacy-preserving solutions to support collaborative consumption applications for sharing economy. In typical collaborative consumption, information processing platforms need to enable fair cost-sharing among multiple users for utilizing certain shared facilities and communal services. Our cloud-based privacy-preserving protocols, based on homomorphic Paillier cryptosystems, can ensure that the cloud-based operator can only obtain an aggregate schedule of all users in facility sharing, or a service schedule conforming to service provision rule in communal service sharing, but is unable to track the personal schedules or demands of individual users. More importantly, the participating users are still able to settle cost-sharing among themselves in a fair manner for the incurred costs, without knowing each other’s private schedules or demands. Our privacy-preserving protocols involve no other third party who may compromise privacy. We also provide an extensive evaluation study and a proof-of-concept system prototype of our protocols.

Medical time series data analytics based on dynamic time warping (DTW) greatly benefits modern medical research. Driven by the distributed nature of medical data, the collaboration of multiple healthcare institutions is usually necessary for a sound medical conclusion. Among others, a typical use case is disease screening for public health, where multiple healthcare institutions wish to collaboratively detect over their joint datasets the patients whose medical records have similar features to the given query samples. However, sharing the medical data faces critical privacy obstacles with the increasingly strict legal regulations on data privacy. In this article, we present the design of a novel system enabling privacy-preserving DTW-based analytics on distributed medical time series datasets. Our system is built from a delicate synergy of techniques from both cryptography and data mining domains, where the key idea is to leverage observations on the advancements in plaintext DTW analytics (e.g., clustering and pruning) to facilitate the scalable computation in the ciphertext domain, through our tailored security design. Extensive experiments over real medical time series datasets demonstrate the promising performance of our system, e.g., our system is able to process a secure DTW query computation over 15K time series sequences in 34 minutes.

Outsourcing machine learning inference services to the cloud is getting increasingly popular. However, this also entails privacy risks to the provider's proprietary model and the client's sensitive data. Focusing on inference with decision trees, this article proposes a framework for securely and efficiently outsourcing decision tree inference. Targeting both privacy and efficiency, we propose a customized protocol using only lightweight cryptography in the online execution of secure inference. We resort to additive secret sharing and tackle the problems in various components including secure input feature selection, decision node evaluation, and inference result generation. Our protocol requires no interaction from the provider and client during online secure inference, a distinct advantage over prior works for practical deployment as they all operate under the client-provider setting where synchronous and continuous interaction is required. Performance evaluation demonstrates our security design's efficiency, as well as substantial performance benefits for the client (up to four orders of magnitude in computation and 163 times in communication), as opposed to prior art in the non-outsourcing setting. To facilitate the practical usage for meeting more service demands, we also investigate the extensions for secure outsourced inference of random forests and categorical feature-based decision trees.

In recent years there has been growing popularity of leveraging cloud computing for storing and querying attributed graphs, which have been widely used to model complex structured data in various applications. Such trend of outsourced graph analytics, however, is accompanied with critical privacy concerns regarding the information-rich and proprietary attributed graph data. In light of this, we design, implement, and evaluate OblivGM, a new system aimed at oblivious graph analytics services outsourced to the cloud. OblivGM focuses on the support for attributed subgraph matching, one popular and fundamental graph query functionality aiming to retrieve from a large attributed graph subgraphs isomorphic to a small query graph. Built from a delicate synergy of insights from attributed graph modelling and advanced lightweight cryptography, OblivGM protects the confidentiality of data content associated with attributed graphs and queries, conceals the connections among vertices in attributed graphs, and hides search access patterns. Meanwhile, OblivGM flexibly supports oblivious evaluation of varying subgraph queries, which may contain equality and/or range predicates. Extensive experiments over a real-world attributed graph dataset demonstrate that while providing strong security guarantees, OblivGM achieves practically affordable performance (with query latency on the order of a few seconds).

With the widespread adoption of cloud computing, it is increasingly popular for online social network (OSN) service providers to leverage the public cloud as a back-end to manage their services for the cloud’s well-understood benefits. However, the cloud is also notoriously subject to a wide attack surface, making it an imperative need to embed security in the cloud-backed OSN service from the very beginning. In light of this, in this paper, we design, implement, and evaluate PeGraph, the first system simultaneously allowing private, efficient, and rich queries over encrypted social graphs. PeGraph is aimed at safeguarding the confidentiality of the social graph at the cloud, while preserving the functionality of social search, a key enabler for quality OSN services like friend discovery and user targeting. PeGraph is built from a delicate synergy of insights from social graph modelling and lightweight cryptography such as searchable encryption and additive secret sharing, supporting rich social search queries like exact queries, fuzzy queries, and mixed queries. PeGraph also allows the cloud to obliviously render the encrypted social search results in a ranked order according to their importance, as per users’ preferences. Extensive experiments demonstrate that PeGraph can securely process a wide range of practical social search queries within 1 second, over a real-world social graph consisting of millions of entities.

Image-scaling is a typical operation that processes the input image before feeding it into convolutional neural network models. However, it is vulnerable to the newly revealed image-scaling attack. This work presents an image-scaling attack detection framework, Decamouflage, consisting of three independent detection methods: scaling, filtering, and steganalysis, to detect the attack through examining distinct image characteristics. Decamouflage has a pre-determined detection threshold that is generic. More precisely, as we have validated, the threshold determined from one dataset is also applicable to other different datasets. Extensive experiments show that Decamouflage achieves detection accuracy of 99.9% and 98.5% in the white-box and the black-box settings, respectively. We also measured its running time overhead on a PC with an Intel i5 CPU and 8GB RAM. The experimental results show that image-scaling attacks can be detected in milliseconds. Moreover, Decamouflage is highly robust against adaptive image-scaling attacks (e.g., attack image size variances).

Public blockchains have emerged as a promising direction in revolutionizing existing data-driven systems relying on centralized service providers. Among others, one kind of such systems is the popular crowdsensing systems which promise convenient data collection and aggregation. Although promising, leveraging public blockchains to build crowdsensing systems is non-trivial and has to overcome several barriers. First, public blockchains are transparent and lack support for data privacy. Second, participants from the open blockchain environment may misbehave in serving crowdsensing applications, like providing invalid data or doing aggregation incorrectly. Further, on-chain processing incurs monetary cost, so simply putting all workload on-chain is highly uneconomical and a delicate joint on-chain and off-chain design is required. In this paper, we take the first research attempt and explore a new design point to bridge public blockchains with crowdsensing systems. We propose a framework for building private, robust, and verifiable blockchain-empowered crowdsensing systems. It features an open service paradigm where blockchain nodes can rent out their computing resources to serve crowdsensing applications, with custom and full-fledged mechanisms to foster a healthy and economical ecosystem and to simultaneously tackle the challenges of data privacy, robustness against misbehaving participants, and service correctness assurance. Extensive experiments demonstrate our designs practicality.

It is increasingly popular to leverage the wisdom of crowd for knowledge discovery and monetization. Among others, crowdsensing with truth discovery has emerged as a promising way for leveraging the crowd wisdom, which can mine reliable knowledge from the generally unreliable sensory data contributed collected from diverse sources. Building a knowledge marketplace based on crowdsensing with truth discovery for knowledge discovery and monetization, however, is non-trivial and has to overcome several challenges. First, the sensory data should be protected as they may carry sensitive information. Second, many real crowdsensing applications usually yield sensory data in a streaming fashion, posing the demand that truth discovery should be conducted over data streams to continuously mine reliable knowledge in each data collection epoch. Third, knowledge monetization should be well treated, fully addressing the practical needs of parties in the monetization ecosystem. In this article, we take the first research attempt and propose a new full-fledged framework for building a secure knowledge marketplace over crowdsensed data streams. Our marketplace supports secure monetization of reliable knowledge mined privately from data streams in crowdsensing applications. Our framework leverages lightweight cryptographic techniques like additive secret sharing to enable privacy-preserving streaming truth discovery, continuously producing reliable knowledge over data streams. For monetization of the learned truth, i.e., knowledge, we resort to the emerging blockchain technology and deliver a tailored and full-fledged design, which promises monetization fairness, knowledge confidentiality, and streamlined processing. Extensive experiments on Amazon cloud and Ethereum blockchain demonstrate the practically affordable performance of our design.

Large volumes of images are being exponentially generated today, which poses high demands on the services of storage, processing, and management. To handle the explosive image growth, a natural choice nowadays is cloud computing. However, coming with the cloud-based image services is acute data privacy concerns, which has to be well addressed. In this paper, we present a secure cloud-based image service framework, which allows privacy-preserving and effective image denoising on the cloud side to produce high-quality image content, a key for assuring the quality of various image-centric applications. We resort to state-of-the-art image denoising techniques based on deep neural networks (DNNs), and show how to uniquely bridge cryptographic techniques (like lightweight secret sharing and garbled circuits) and image denoising in depth to support privacy-preserving DNN based image denoising services on the cloud. By design, the image content and the DNN model are all kept private along the whole cloud-based service flow. Our extensive empirical evaluation shows that our security design is able to achieve denoising quality comparable to that in plaintext, with high cost efficiency on the local side and practically affordable cost on the cloud side.

In-network storage is recognized as a vital component of many emerging network architectures, which facilitates high-quality and efficient content-centric services. In this trend, providing content-based near-duplicate detection (NDD) services among in-network storage becomes naturally necessary for network traffic alleviation and resource optimization. However, due to the increasing attacking surfaces, storing data in the networked environment inevitably raises new concerns about user privacy exposure and unauthorized data access. Therefore, we aim to design a secure NDD service in the context of encrypted in-network storage. For efficiency, we first leverage the fingerprint techniques and locality-sensitive hashing to convert the problem of NDD into the keyword search. We then adopt an efficient multi-key searchable encryption scheme, which requires only one encrypted query from the user even the data are from multiple content providers encrypted with different keys. As simply combining the above methods does not appear to directly locate accurate results, we then devise a secure result refining scheme via Yao's garbled circuits to avoid user-side post-processing. Furthermore, we enhance our design to address the potential malicious behavior of in-network servers. Extensive evaluations of real-world image dataset demonstrate that our design can achieve comparable accuracy to the plaintext with modest security overhead.

The wide adoption of cloud greatly facilitates the sharing of explosively generated media content today, yet deprives content providers' direct control over the outsourced media content. Thus, it is pivotal to build an encrypted cloud media center where only authorized access is allowed. Enforcing access control alone, however, cannot fully protect content providers' interests, as authorized users may later become traitors that illegally redistribute media content to the public. Such realistic threat should have been seriously treated yet is largely overlooked in the literature. In this paper, we initiate the first study on secure media sharing with fair traitor tracing in the encrypted cloud media center, through a new marriage of proxy re-encryption (for secure media sharing) and fair watermarking (for fair traitor tracing). Our key insight is to fully leverage the homomorphic properties residing in proxy re-encryption to embrace operations in fair watermarking. Two protocols are proposed for different application scenarios. We also provide complexity analysis for performance, showing that our work can also be treated as secure outsourcing of fair watermarking, and thus kills two birds with one stone. We thoroughly analyze the security strengths and conduct extensive experiments to validate the effectiveness of our design.

Truth discovery in mobile crowdsensing has recently received wide attention. It refers to the procedure for estimating the unknown user reliability from collected sensory data and inferring truthful information via reliability-aware data aggregation. Though widely studied in the plaintext domain, truth discovery remains largely under-explored in privacy-aware mobile crowdsensing. Existing works either do not consider user reliability issue or fall short of achieving practical cost efficiency, due to iterative transmission and computation over large ciphertexts from homomorphic cryptosystem. In this paper, we propose two new privacy-aware crowdsensing designs with truth discovery that significantly improve the bandwidth and computation performance on individual users. Our insight is to identify the core atomic operation in the iterative truth discovery procedure, and carefully craft security designs accordingly to enable efficient truth discovery in the ciphertext domain. Our first design is highly customized for the single-server setting, while our second design under the two-server model further shifts most of user workloads to the cloud server side. Both our designs protect individual sensory data and reliability degrees throughout the truth discovery procedure. Experiments show that compared with the prior result, our designs gain at least 30x and 10x savings on user communication and computation, respectively.

With the prosperity of mobile application markets, mobile advertising is becoming an increasingly important economic force. In order to maximize revenue, ads are recommended to be delivered to potentially interested users, which requires user targeting, i.e., analyzing users' profiles and exploring users' interests. However, collecting user personal information for targeted mobile advertising services raises critical privacy concerns. Although some solutions like anonymization and obfuscation have been proposed for privacy-aware targeted advertising, they undesirably face the issues of security, efficiency, and/or ad relevance. In this paper, we propose a practical system enabling secure and efficient targeted mobile advertising services. It allows the ad network to perform accurate user targeting, while ensuring strong privacy protection for mobile users. Specifically, we show how to properly leverage a cryptographic primitive called private stream searching to support secure, accurate, and practical targeted mobile ad delivery. Moreover, we propose secure billing schemes to enable the ad network to charge advertisers in a privacy-preserving manner. The security strength of our system is thoroughly analyzed. Through extensive experiments, we show that our system achieves practical efficiency on mobile devices.

Crowdsensing, driven by the proliferation of sensor-rich mobile devices, has emerged as a promising data sensing and aggregation paradigm. Despite useful, traditional crowdsensing systems typically rely on a centralized third-party platform for data collection and processing, which leads to concerns like single point of failure and lack of operation transparency. Such centralization hinders the wide adoption of crowdsensing by wary participants. We therefore explore an alternative design space of building crowdsensing systems atop the emerging decentralized blockchain technology. While enjoying the benefits brought by the public blockchain, we endeavor to achieve a consolidated set of desirable security properties with a proper choreography of latest techniques and our customized designs. We allow data providers to safely contribute data to the transparent blockchain with the confidentiality guarantee on individual data and differential privacy on the aggregation result. Meanwhile, we ensure the service correctness of data aggregation and sanitization by delicately employing hardware-assisted transparent enclave. Furthermore, we maintain the robustness of our system against faulty data providers that submit invalid data, with a customized zero-knowledge range proof scheme. The experiment results demonstrate the high efficiency of our designs on both mobile client and SGX-enabled server, as well as reasonable on-chain monetary cost of running our task contract on Ethereum.

Fog computing has emerged as a promising paradigm in overcoming the growing challenges (e.g., low latency, location awareness, and geographic distribution) arising from many real-world Internet of Things (IoT) applications, by extending the cloud to the network edge. With the widespread deployment of fog-assisted IoT applications, unprecedentedly huge volumes of network traffic from massive IoT devices would continuously arrive at the fog nodes. Archiving the network traffic can be highly beneficial to fog computing, which forms the basis of forensic, monitoring, troubleshooting, and many other critical tasks. Such high value, however, constantly renders traffic archives the first-order target to experienced attackers. This mandates the traffic archives to be built in a trustworthy way and stayed encrypted at rest. Security aside, it is yet highly desirable to retain the utility of the encrypted traffic archives, in particular by making them privately queryable. In this paper, we take the first research attempt and explore a new design point to delicately bridge trusted hardware and searchable encryption for building trustworthy, encrypted, yet queryable network traffic archives for fog-assisted IoT applications. We take a systematic approach to address several key challenges, which are unsolvable by synthesizing out-of-box techniques, from ground up. Extensive evaluations show that our system can achieve stable archiving throughput of 350 Mb/s with one core, and saturate a 1 Gb/s link with four cores; for a real trace, it outperforms a baseline system without any of our designs by over 110× .

Image denoising via machine learning techniques, particularly neural networks, has been shown to achieve state-of-the-art performance. However, in practice security and privacy issues undesirably arise in applying a trained machine learning model to image denoising. In this paper, we propose a system framework that enables the owner of a trained machine learning model to provide secure image denoising service to an authorized user, via the aid of cloud computing. Our framework ensures that the cloud server learns nothing about the model and the user's images, while the user learns nothing about the model except denoised images. Experiments are conducted for performance evaluation, and the results show that our design can achieve denoising quality close to that in the plaintext domain. For future work, we plan to explore various directions for optimizing the runtime performance.

Leveraging the wisdom of crowd for knowledge discovery and monetization is increasingly popular nowadays. Among others, one popular way of leveraging the crowd wisdom is crowdsensing with truth discovery, which is able to discover truthful knowledge from the unreliable sensory data harvested from mobile clients. In order to become truly successful, however, a number of challenges are yet to be addressed. First, safeguarding clients' sensory data is demanded for privacy protection. Second, in many real crowdsensing applications, data are usually collected in a streaming manner, so truth discovery is naturally required to be efficiently conducted in a streaming fashion. Thirdly, knowledge monetization should be made full-fledged, endowed with features of transparency and streamlined processing while fully addressing the practical needs of parties in the monetization ecosystem. In this paper, we present our initial effort on a crowdsensing framework that enables privacy-preserving knowledge discovery and full-fledged blockchain-based knowledge monetization. Our framework enables privacy-preserving and efficient truth discovery over encrypted crowdsensed data streams for truthful knowledge discovery. Meanwhile, with careful integration of the newly emerging blockchain-based smart contract technology, our framework allows full-fledged knowledge monetization. Tackling the challenges of monetization fairness and (on-chain) knowledge confidentiality, our customized knowledge monetization design well respects the interests of knowledge seller and requester, with full support of transparency, streamlined processing, and automatic quality-aware rewards for clients. Extensive experiments on Microsoft Azure cloud and Ethereum blockchain demonstrate the practically affordable performance of our design.

Mobile crowdsensing enables convenient sensory data collection from a large number of mobile devices and has found various applications. In the real practice, however, the sensory data collected from various mobile devices are usually unreliable. To extract truthful information from the unreliable sensory data in mobile crowdsensing, the topic of truth discovery has received wide attention recently, which essentially operates by estimating user reliability degrees and performing reliability-aware truthful aggregation. Despite the effectiveness, applying truth discovery in mobile crowdsensing faces several privacy and security challenges. First, the sensory data and reliability degrees of users may reveal privacy-sensitive information and, thus, demand strong protection. Second, the requester that initiates a crowdsensing application usually needs to have monetary investment, so the inferred truths can be the requester's proprietary information and should be protected as well. In this paper, we propose a new system architecture enabling encrypted truth discovery in mobile crowdsensing. We focus on general and realistic mobile crowdsensing scenarios with varying levels of user participation, and our security design is built on the confidence-aware truth discovery (CATD) approach for its state-of-the-art accuracy in such scenarios. In our system architecture, users send encrypted sensory data to the cloud, where CATD is then conducted in the encrypted domain. The final encrypted inferred truths are sent to the requester for decryption. Along the whole workflow, the sensory data and reliability degrees of users, as well as the inferred truths of the requester, are kept private. Extensive experiments over real-world mobile crowdsensing dataset show that our design achieves practical performance on mobile devices.

Along with the rapid advancement of digital image processing technology, image denoising remains a fundamental task, which aims to recover the original image from its noisy observation. With the explosive growth of images on the Internet, one recent trend is to seek high quality similar patches at cloud image databases and harness rich redundancy therein for promising denoising performance. Despite the well-understood benefits, such a cloud-based denoising paradigm would undesirably raise security and privacy issues, especially for privacy-sensitive image data sets. In this paper, we initiate the first endeavor toward privacy-preserving image denoising from external cloud databases. Our design enables the cloud hosting encrypted databases to provide secure query-based image denoising services. Considering that image denoising intrinsically demands high quality similar image patches, our design builds upon recent advancements on secure similarity search, Yao's garbled circuits, and image denoising operations, where each is used at a different phase of the design for the best performance. We formally analyze the security strengths. Extensive experiments over real-world data sets demonstrate that our design achieves the denoising quality close to the optimal performance in plaintext.

The explosive growth of multimedia contents, especially videos, is pushing forward the paradigm of cloud-based media hosting today. However, the wide attacking surface of the public cloud and the growing security awareness from the society are both calling for data encryption before outsourcing to cloud. Under the circumstance of encrypted videos, how to still preserve all the service benefits of cloud media center remains to be fully explored. In this paper, we present a secure system architecture design as our initial effort toward this direction, which bridges together the advancements of video coding techniques and secure deduplication. Our design enables the cloud with the crucial deduplication functionality to completely eliminate the extra storage and bandwidth cost, which would have been incurred by hosting encrypted videos from different entities. The design is also carefully tailored to the scalable video coding (SVC) techniques to support heterogeneous networks and devices for high-quality adaptive video dissemination. We show fully functional system implementations with structure-aware encryption design and structure-aware deduplication strategies that are both completely compliant with the video format in SVC. Extensive security analysis and experiments via our prototype deployed on Azure cloud platform show the practicality of the design. Our work can also be easily extended to support other media applications that employ media files with scalable structures.

Near-duplicate detection (NDD) plays an essential role for effective resource utilization and possible traffic alleviation in many emerging network architectures, leveraging in-network storage for various content-centric services. As innetwork storage grows, data security has become one major concern. Though encryption is viable for in-network data protection, current techniques are still lacking for effectively locating encrypted near-duplicate data, making the benefits of NDD practically invalidated. Besides, adopting encrypted innetwork storage further complicates the user authorization when locating near-duplicate data from multiple content providers under different keys. In this paper, we propose a secure and effective NDD system over encrypted in-network storage supporting multiple content providers. Our design bridges locality-sensitive hashing (LSH) with a newly developed cryptographic primitive, multi-key searchable encryption, which allows the user to send only one encrypted query to access near-duplicate data encrypted under different keys. It relieves the users from multiple rounds of interactions or sending multiple different queries respectively. As simply applying LSH does not ensure the detection quality, we then leverage Yao's garbled circuits to build a secure protocol to obtain highly accurate results, without user-side post-processing. We formally analyze the security strength. Experiments demonstrate our system achieves practical performance with comparable accuracy to plaintext.

Multimedia contents, especially videos, are being exponentially generated today. Due to the limited local storage, people are willing to store the videos at the remote cloud media center for its low cost and scalable storage. However, videos may have to be encrypted before outsourcing for privacy concerns. For practical purposes, the cloud media center should also provide the deduplication functionality to eliminate the storage and bandwidth redundancy, and adaptively disseminate videos to heterogeneous networks and different devices to ensure the quality of service. In light of the observations, we present a secure architecture enabling the encrypted cloud media center. It builds on top of latest advancements on secure deduplication and video coding techniques, with fully functional system implementations on encrypted video deduplication and adaptive video dissemination services. Specifically, to support efficient adaptive dissemination, we utilize the scalable video coding (SVC) techniques and propose a tailored layer-level secure deduplication strategy to be compatible with the internal structure of SVC. Accordingly, we adopt a structure-compatible encryption mechanism and optimize the way how encrypted SVC videos are stored for fast retrieval and efficient dissemination. We thoroughly analyze the security strength of our system design with strong video protection. Furthermore, we give a prototype implementation with encrypted end-to-end deployment on Amazon cloud platform. Extensive experiments demonstrate the practicality of our system.