The proliferation of model-sharing platforms has intensified the need for data-free backdoor detection, as deployed models are often accessed without accompanying clean validation data. This constraint renders traditional, data-dependent detection methods ineffective. However, existing strategies to evade data-free detection are inadequate; they frequently fail to circumvent the multi-faceted discriminative criteria of modern detectors that analyze model output behavior, and they often compromise the backdoor model's primary task performance, thus failing to balance attack stealth with functionality. To evade these detections, this paper introduces a Stealthy Backdoor Attack (SBdA) based on label smoothing. Our method dynamically adjusts the training labels for backdoor samples by leveraging the feature similarity between each class and the attacker's target class. This optimization shapes the backdoored model's output distributions to closely mimic those of a benign model, thereby evading detection mechanisms that rely on outlier characterization and posterior distribution analysis. Extensive experiments demonstrate that SBdA maintains a high attack success rate (exceeding 91% under all tested conditions, with 75% of models surpassing 96%) while significantly reducing its detectability. On CIFAR-10, the average outlier score for our models was 0.554-merely 0.050 higher than benign models-compared to a 25.517 deviation for conventional attacks. On GTSRB, SBdA reduced the outlier score gap by 82.86% compared to the average-label technique. Furthermore, by carefully calibrating the posterior distribution, SBdA effectively avoids detection by posterior matrix-based methods across all four tested datasets.

Electronic medical records (EMRs) are the essential sensitive personal data that is shared between patients and doctors through the semi-trusted E-health cloud. In the real application, multiauthority ciphertext-policy attribute-based searchable encryption (MA-CP-ABSE) is suitable to protect the security of the EMR for it possesses fine-grained access permission, efficient key management, and retrieval function over the encrypted data. However, previously proposed MA-CP-ABSE schemes are commonly restricted by the central authority, which is required by all attribute authorities in some operations like generating users’ secret keys. To deal with this issue, we proposed a decentralized MA-CP-ABSE (DMA-CP-ABSE) scheme. In our scheme, any attribute authority can become an independent authority to generate a secret key, which is no longer controlled by the central authority. Furthermore, a single-keyword search will produce many disrelated search results. For this, we enhance the DMA-CP-ABMSE scheme by implementing a multikeyword search function to improve the search accuracy. Besides, to handle the dynamic changing of access permission, we have designed the attribute revocation methods. Finally, we process the formal security analysis to demonstrate our scheme is secure under the chosen-keyword attack (CKA) and implement experiments to show its efficiency and feasibility.

Allowing users to assure that their files are reliably stored into multiple replicas is critically important but challenging for secure cloud storage. Recently, Damgård et al. [1] designed the first publicly verifiable proof of replicated storage (abbreviated as PRI-POREP) in the private client setup without the fine-grained timing assumption. However, it relies on an “ideal” invertible random permutations (IRPs), whose construction with the structured domain/range remains open even in the random oracle model. Also, it is computationally inefficient in terms of both replicas generation and file update. To address challenges regarding both practicality and efficiency while guaranteeing the security of PRI-POREP, this paper aims at constructing a new proof of replicated storage scheme without timing assumption, named as μPRI-POREP. μPRI-POREP is secure against server-side deletion of replica blocks and it works efficiently, saving computation cost by orders of magnitude, compared to PRI-POREP. Moreover, we demonstrate that μPRI-POREP can also support efficient dynamic update and can be further applied to secure the RSA-Hourglass schemes. Finally, we evaluate μPRI-POREP with a prototype implementation and exhibit that it can achieve comparable performance compared to PRI-POREP and support efficient file update operation.

Deep neural networks (DNNs) greatly enable the task of automatic modulation classification (AMC) by virtue of their powerful feature extraction capability. However, extensive research has shown that DNNs are highly vulnerable to adversarial attacks, which can lead them to confidently output incorrect results with high confidence scores. Existing adversarial attack methods often focus solely on temporal characteristics of signals while neglecting frequency domain information, resulting in adversarial examples with poor transferability and inadequate performance in the closed-box scenario. An adversarial attack method based on frequency domain feature enhanced and integral gradient (FEIG) for AMC task is proposed in this paper. The approach utilizes techniques such as translation interpolation and Inverse Fast Fourier Transform to enhance the frequency domain information of original examples, thereby constructing enhanced baseline examples. Subsequently, these generated enhanced baseline examples are used as new inputs for gradient integration to obtain adversarial examples. Compared to traditional methods, the generated adversarial examples exhibit stronger transferability. Furthermore, in order to improve the defense performance of the model, an enhanced hybrid adversarial training (EH-AT) framework is proposed in this paper. The original clean example and the adversarial example generated by the proposed attack method are trained with joint loss constraints, which greatly enhances the robustness of the model. Experimental results demonstrate the effectiveness of the FEIG attack method and the EH-AT framework.

Spectrum is the fundamental medium for transmitting information services, including communication, navigation, and detection. Spectrum anomalies can lead to substantial economic losses and even endanger life safety. Anomaly detection constitutes a critical component of spectrum risk management. Through spectrum anomaly detection (SAD), anomalous spectrum usage behaviors, such as malicious user activities, can be identified. Given the significant limitations of current SAD algorithms in terms of accuracy and localization capabilities, this article proposes an approach for detecting spectral anomalies that utilizes knowledge distillation and memory-enhanced autoencoders (AEs). First, the pretrained network with robust feature extraction capabilities is distilled into the teacher network. Subsequently, both an AE and a memory-enhanced AE with an identical structure are trained to predict the teacher network’s normalized outputs on a spectrum devoid of anomalies. Finally, in the case of an anomalous spectrum, difference exist between the normalized outputs of the teacher network and the outputs of different student networks, as well as among the outputs of different student networks, which facilitates the process of anomaly detection. The outcomes of experiments reveal that the proposed algorithm is more effective on both synthetic spectral data sets and real IQ signals, demonstrating its proficiency in accurately detecting and locating anomalies.

Price discrimination has been empirically exposed where e-commercial platforms aim to gain additional profits by charging customers with different prices for the same product/service. This situation becomes even worse in nowadays’ Big Data era, giving the chance for service providers to leverage artificial intelligence technologies to have the deep analysis of personalized patterns, urgently calling for solutions to prevent such discriminated behaviors to protect customers’ rights. This article aims to defend against price discrimination by developing a secure and privacy-preserving solution, provable for e-commerce fairness. Using a newly designed cryptographic accumulator and public bulletin board, our system, called FairECom, allows an auditor (i.e., a customer or third-party auditor) to verify if customers are experiencing price discrimination. In particular, FairECom enables a customer to check if his payment to a product/service is identical to other customers through a privacy-preserving challenge-response protocol, for implementing the price transparency against discrimination. We implement a prototype using an Ethereum-based public bulletin board to conduct the system evaluation. Our evaluation indicates that FairECom can integrate with existing APIs provided by Ethereum and incur acceptable costs when deploying to the e-commercial systems.

Data deduplication is of critical importance to reduce the storage cost for clients and to relieve the unnecessary storage pressure for cloud servers. While various techniques have been proposed for secure deduplication of identical files/blocks, the effective and secure deduplication solutions on fuzzy similar data (image, video, and others) which occupy a large portion in the real world across wide applications, remain open. In this article, we propose a novel deduplication system, named Fuzzy Deduplication (FuzzyDedup), to implement the secure deduplication of similar data (i.e., similar files, chunks, or blocks). In particular, we leverage the similarity-preserving hash, a fuzzy extractor based on error-correcting codes, and the encryption with customized design to construct a fuzzy-style deduplication encryption scheme (FuzzyMLE), achieving the ciphertext-based deduplication for similar data. Besides, to defend against data ownership cheating attack and duplicate-faking attack, a fuzzy-style proof of ownership scheme (FuzzyPoW) is designed for the cloud server to securely verify a client in possession of the similar data. To further enhance security and efficiency, we also propose both server-aided and random-tag FuzzyMLE to make FuzzyDedup robust against off-line brute-force attack and to support tag randomization, respectively. Then, we design Hamming distance reduction and tag cutting optimization algorithms to improve the tag query efficiency of FuzzyDedup. In the end, we formally prove the security of our solution and conduct experiments on real-world datasets for performance evaluation. Experimental results exhibit the efficiency of FuzzyDedup in terms of computation cost and communication overhead.

Cloud computing, which provides a brand-new service model, has become an important infrastructure in the information age, and has been widely used in numerous fields. The Key-Policy Attribute-Based Encryption (KP-ABE) scheme allows the encrypted data with fine-grained access control in the cloud environment. However, achieving large-scale user revocation in the application scenario of KP-ABE becomes one of the thorny problems. Furthermore, the computation and communication costs of the previous user revocation schemes were generally high, especially when a large number of users were revoked. To address these problems, an enhanced high-performance user-revocable KP-ABE scheme combined with the puncture method was proposed. In this article, the user could be revoked by the fine-grained restriction policy. When revoking the user, the cloud would run the puncture algorithm to embed the restriction policy defined by the data owner into the ciphertext. This method could effectively omit the re-encryption and key updating processes, by which the computation and communication overhead of the user revocation are efficiently reduced, and the user revocation becomes more flexible and efficient. Moreover, the Chosen-Plaintext Attack (CPA) security proof and extensive simulation results demonstrate the reliability and efficiency of the proposed scheme for user revocation in a cloud environment.

The maturity in information technology and new energy technologies enables participants to generate, buy, and sell energy in energy trading systems. Although applying blockchain technology to energy trading has solved some drawbacks in traditional centralized energy systems, the openness and transparency characteristics make the trading records stored on the blockchain vulnerable to data-mining attacks that may cause indispensable privacy leakage. Due to high efficiency and low overhead, noise-addition is an appropriate solution for privacy preservation. Nonetheless, recent research on noise-addition needs to generate massive accounts, which brings a certain amount of waste and inconvenience for later regulation and management. To avoid the aforementioned issues, this paper proposes a consortium blockchain-enabled scheme to ensure the privacy of data stored on the blockchain and resist linking attacks initiated by data mining algorithms. Our scheme utilizes a dynamic partition algorithm to leverage an account mapping algorithm and a virtual token algorithm. Specifically, the account mapping algorithm utilizes a dynamic account allocation method to hide the trading distribution of active users. Furthermore, the virtual token algorithm applies Laplace noise to hide the actual energy consumption of inactive users and curb excessive accounts generation. Finally, we formally demonstrate the privacy and effectiveness of our proposed scheme in security analysis and experiment evaluations.

Advances in adversarial attack and defense technologies will enhance the reliability of deep learning (DL) systems spirally. Most existing adversarial attack methods make overly ideal assumptions, which creates the illusion that the DL system can be attacked simply and has restricted the further improvement on DL systems. To perform practical adversarial attacks, a detection tolerant black-box adversarial-attack (DTBA) method against DL-based automatic modulation classification (AMC) is presented in this article. In the DTBA method, the local DL model as a substitution of the remote target DL model is trained first. The training dataset is generated by an attacker, labeled by the target model, and augmented by Jacobian transformation. Then, the conventional gradient attack method is utilized to generate adversarial attack examples toward the local DL model. Moreover, before launching attack to the target model, the local model estimates the misclassification probability of the perturbed examples in advance and deletes those invalid adversarial examples. Compared with related attack methods of different criteria on public datasets, the DTBA method can reduce the attack cost while increasing the rate of successful attack. Adversarial attack transferability of the proposed method on the target model has increased by more than 20%. The DTBA method will be suitable for launching flexible and effective black-box adversarial attacks against DL-based AMC systems.

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

While the prevalent cloud storage platforms are offering convenient services in support of diverse data-driven applications for clients, various security concerns raise in terms of data confidentiality, availability, and retrievability. Among them, servers' dishonesty on the location-specific data backup becomes a serious concern when the data stands out clients' control, considering the strict regulations imposed by many governments and organizations on data storage location. This article studies location-aware data backup verification for the data stored in clouds and aims to design a secure framework, named as ReliableBox, enabling the clients to verify if their data have been backed up on the remote servers with specific geolocation. In the design of ReliableBox, we leverage the prominent proof-of-storage techniques for data possession proof, and take advantage of multilateration geolocation and Intel SGX for the precise communication delay measurement and trust computing delay measurement, respectively. In ReliableBox, a client first computes integrity tags for the files and then outsources both the files and tags to the cloud storage server. In the later attestation, with the precise network delay and distance measurement from location-known verifiers, the client verifies that the outsourced files are intact and backed-up to hosts at the specific geolocation. With the customized design, ReliableBox can support the security needs in terms of both data integrity and backup location verification for clients, even when there exists potential dishonest cloud service providers who may manipulate the network delays or forge verification proofs. We provide security analysis to show the security property of ReliableBox in terms of data access, confidentiality, and verifications. In the end, we implement the system prototype and deploy it into several prevalent and commercial cloud platforms for performance evaluation. The experimental results demonstrate that ReliableBox is secure in support of data integrity checking and location-aware backup auditing, while it is robust to the data possession and location spoofing attacks.

Data deduplication has been widely used in cloud storage to reduce storage space and communication overhead by eliminating redundant data and storing only one copy for them. In order to achieve secure data deduplication, the convergent encryption scheme and many of its variants are proposed. However, most of these schemes do not consider or cannot address the efficiently dynamic ownership changes and the secure Proof-of-Ownership (PoW), simultaneously. In this paper, we propose a secure data deduplication scheme with efficient PoW process for dynamic ownership management. Specially, our scheme supports both cross-user file-level and inside-user block-level data deduplication. During the file-level deduplication, we construct a new PoW scheme to ensure the tag consistency and achieve the mutual ownership verification. Moreover, we design a lazy update strategy to achieve efficient ownership management. For inside-user block-level deduplication, the user-aided key is used to realize convergent key management and reduce the key storage space. Finally, the security and performance analysis demonstrate that our scheme can ensure data confidentiality and tag consistency, and it is efficient in data ownership management.

Cross-client data deduplication has been widely used to eliminate redundant storage overhead in cloud storage system. Recently, Abadi et al. introduced the primitive of MLE2 with nice security properties for secure and efficient data deduplication. However, besides the computationally expensive noninteractive zero-knowledge proofs, their fully randomized scheme (R-MLE2) requires the inefficient equality-testing algorithm to identify all duplicate ciphertexts. Thus, an interesting challenging problem is how to reduce the overhead of R-MLE2 and propose an efficient construction for R-MLE2. In this paper, we introduce a new primitive called μR-MLE2, which gives a partial positive answer for this challenging problem. We propose two schemes: static scheme and dynamic scheme, where the latter one allows tree adjustment by increasing some computation cost. Our main trick is to use the interactive protocol based on static or dynamic decision trees. The advantage gained from it is, by interacting with clients, the server will reduce the time complexity of deduplication equality test from linear time to efficient logarithmic time over the whole data items in the database. The security analysis and the performance evaluation show that our schemes are Path-PRV-CDA2 secure and achieve several orders of magnitude higher performance for data equality test than R-MLE2 scheme when the number of data items is relatively large.

The advent of the cloud computing makes storage outsourcing become a rising trend, which promotes the secure remote data auditing a hot topic that appeared in the research literature. Recently some research consider the problem of secure and efficient public data integrity auditing for shared dynamic data. However, these schemes are still not secure against the collusion of cloud storage server and revoked group users during user revocation in practical cloud storage system. In this paper, we figure out the collusion attack in the exiting scheme and provide an efficient public integrity auditing scheme with secure group user revocation based on vector commitment and verifier-local revocation group signature. We design a concrete scheme based on the our scheme definition. Our scheme supports the public checking and efficient user revocation and also some nice properties, such as confidently, efficiency, countability and traceability of secure group user revocation. Finally, the security and experimental analysis show that, compared with its relevant schemes our scheme is also secure and efficient.

An Unattended Wireless Sensor Network (UWSN) collects the sensing data by using mobile sinks (MSs). It differs from the traditional multi-hop wireless sensor networks in which unbalanced traffic makes the sensors close to the base station deplete their power earlier than others. An UWSN can save the battery power and prolong the network lifetime. Unfortunately, MSs would be given too much privilege when acting as the collecting base station, which will cause security concern if compromised. Besides, UWSNs are usually deployed in unreachable and hostile environments, where sensors can be easily compromised. Thus, their security issues should be carefully addressed to deal with node compromise. In this paper, we present a novel key management scheme to secure UWSNs. We employ the Blundo symmetric polynomial mechanism to guard against the newly compromised nodes in a period while utilizing the periodic key updating based on the reverse hash chain to block the compromised nodes and revoke the compromised MSs if failing the authentication. We show that our scheme is robust against node compromised attacks and carry out comparison analysis on the intrusion-tolerance ratio, communication and computing overhead.