With the deep integration of smart IoT devices with social networks, platforms use user-interest similarity for social recommendations to connect like-minded people. Identity-based encryption with equality test (IBEET) enables equivalence testing of interest ciphertexts encrypted under different public keys and avoids costly certificate management. It is beneficial for privacy-preserving social recommendations on smart IoT devices. Nevertheless, existing IBEET schemes suffer from two critical flaws: the lack of forward security allows historical trapdoor holders to test newly generated ciphertexts, and the absence of ciphertext origin authentication impedes the legality verification of the ciphertext source. In this paper, taking into account IoT-device features like limited resources, we introduce a forward secure identity-based signcryption scheme with equality test enabling public verification (FS-IBSCET-PV) for social recommendation. It enables the platforms to leverage users’ trapdoors to perform equality tests on interest ciphertexts, accurately matching users who share common interests. Meanwhile, users can submit time-limited authorized trapdoors when needing recommendation services, thus effectively reducing the risk of trapdoor misuse. Furthermore, upon receiving recommendation results, users can publicly verify their origin without extra decryption to ensure the authenticity of the recommendations, consequently mitigating the hardware overhead. Security analysis and simulation experiments show that FS-IBSCET-PV has enhanced functionality and better testing efficiency than related schemes. It provides a secure and practical solution for privacy-preserving social recommendations on resource-limited smart IoT devices.

In the Internet of Medical Things (IoMT), ensuring the confidentiality, integrity, and authenticity of sensitive medical data poses a significant challenge. Signcryption, merging digital signature and public-key encryption into one logical step, offers a promising solution for such resource-constrained networks. Most existing signcryption schemes inherently lack public verification, which prevents third parties (such as routers) from early identifying illegal messages without decryption. In this paper, we propose a practical certificateless aggregate signcryption scheme with public verification for IoMT. Our design uses elliptic curves and follows the certificateless cryptosystem framework. Thus, it not only avoids time-consuming pairing operations but also addresses the issues of certificate management and key escrow. Meanwhile, individual signcryptions can be compressed into a single aggregate signcryption, optimizing computational and communication overheads. Furthermore, both the pre-aggregated and aggregated signcryptions support public verification, effectively enabling the early elimination of illegal messages. We further present formal security proofs for our construction and reduce its security to the hardness of the one-sided gap Diffie-Hellman (OGDH) and the elliptic curve discrete logarithm (ECDL) problems. Performance analysis reveals that, compared with prior studies, this scheme preemptively blocks illegal message spread, reduces receiving-end load, and better suits resource-constrained IoMT.

In the Federated Learning (FL) paradigm, authentication mechanisms serve as a critical safeguard by preventing unauthorized entities from joining the training process, thereby ensuring the security of both models and data. However, many existing FL authentication schemes have issues regarding security or efficiency. In this paper, we propose a novel pairing-free certificate-based signature scheme and deploy it in the context of privacy-preserving FL. At its core, the proposed scheme is rooted in an elliptic curve certificate-based cryptosystem, which ensures gradient unforgeability without relying on secure channels for key distribution. Additionally, it incorporates a pseudonym-based framework that guarantees participant anonymity during model training while enabling publicly verifiable traceability. Meanwhile, it integrates a dynamic participant management strategy that combines similarity-based quality assessment with a dynamic reputation evaluation to optimize collaborative effectiveness. Performance evaluations show that our proposed scheme reduces signature-related overhead and achieves superior model performance, outperforming existing privacy-preserving FL authentication schemes in communication and computational efficiency.

With the rapid development of Industrial Internet of Things (IIoT), data sharing as a cornerstone function of IIoT, has attracted considerable attention. Ciphertext-policy attribute-based encryption is extensively used to ensure confidentiality and fine-grained access control in such scenarios. Nevertheless, most existing schemes face critical challenges, including high decryption overhead, inadequate attribute privacy protection, and the absence of verifiable revocation mechanisms, which significantly impede their applications in industrial manufacturing systems. To address these challenges, this paper introduces a ciphertext-policy anonymous attribute-based encryption (CP-A$^{2}$BE) scheme with outsourced decryption and verifiable revocation. It encompasses three key innovations: Firstly, a distributed edge computing architecture is established, leveraging pre-deployed edge nodes within factories to offer outsourced decryption services. Secondly, while safeguarding the privacy of industrial data, the attribute privacy of devices and personnel is also taken into consideration. Thirdly, a verifiable revocation mechanism employing commitment-based techniques is presented to enable real-time access control updates while ensuring the data integrity of revocation operations performed by cloud servers. Experimental evaluation shows that our proposed scheme is practical for data sharing in resource-limited IIoT, especially as the decryption time consistently remains at 26 milliseconds, regardless of the number of attributes involved.

Public key authenticated encryption with keyword search (PAEKS) is an advanced asymmetric searchable encryption technique that provides strong resistance against inside keyword guessing attacks (KGAs). However, most PAEKS schemes are limited to one-to-one search, resulting in a linear increase in the number of trapdoors as the number of senders grows. Some solutions have been introduced, but they are still deficient in security or privacy. In this paper, we propose a novel dual-server public key authenticated searchable encryption scheme with constant trapdoor. The proposed scheme achieves robust security against inside KGAs even if a malicious server is allowed to collude with certain data senders. Meanwhile, in contrast to existing schemes, it prevents the leakage of keyword equivalence among unretrieved ciphertexts. Moreover, the data receiver in our system only needs to create one trapdoor per keyword to search for their ciphertexts produced by different data senders, which is not achievable in most prior works. The experimental results demonstrate that our proposed construction maintains comparable performance to existing solutions, especially in multi-sender scenarios.

On social platforms, the friendship recommendation models deployed suggest potential friends based on the similarities among users. As privacy concerns rise, more users choose to encrypt their attributes before requesting recommendation services. Identity-based encryption with equality test (IBEET) supports encrypted data comparison under different users' identities (public keys) and avoids the need for costly certificate management. This makes it highly suitable for implementing privacy-preserving friendship recommendation on social platforms. Nevertheless, most existing IBEET schemes lack the ability to control the validity period of trapdoors used for testing. Historical trapdoors may be maliciously utilized to test encrypted data generated in the future. To address this issue, we propose a forward secure identity-based encryption with equality test for friendship recommendation. Our construction enables an authorized tester to provide privacy-preserving recommendations based on encrypted data among different users. Furthermore, to control the tester's testing capability, time slots are encoded by the 0/1-encoding mechanism and embedded into both ciphertexts and trapdoors. This ensures that only the authorized tester possessing valid trapdoors can perform tests on specific ciphertexts. Under the bilinear Diffie-Hellman assumption in the random oracle model, we prove that the proposed IBEET achieves OW-ID-CCA security when the attacker has access to the trapdoor and IND-ID-CCA security when the attacker does not. Experiment simulation demonstrates that our construction outperforms existing ones in overall performance, particularly in testing that requires frequent execution for recommendation.

Cloud auditing enables remote data integrity verification without downloading the entire dataset, significantly advancing cloud storage services. In long-term usage scenarios, key exposure poses a critical challenge, as a compromised private key enables the cloud service provider (CSP) to forge tags for incorrect data. Although key updatable auditing protocols have been proposed to mitigate key exposure risks, most do not support tag updates triggered by key changes, allowing the CSP to still exploit the leaked historical key for tag forgery. In this article, we propose a block tag updatable certificate-based integrity auditing protocol for long-term cloud storage. By leveraging certificate-based cryptography, our protocol eliminates the need for costly secure channels. It supports simultaneous key and tag updates, ensuring that even with previous private key leaks, the CSP cannot forge valid tags under newly updated key-pairs. The protocol achieves high efficiency through: 1) 2-D data partitioning that reduces computational and storage overhead; 2) optimized tag generation requiring only one map-to-point hash function operation for multiple data blocks; 3) strategic delegation of computationally intensive tag update operations to the CSP; and 4) constant-time proof verification regardless of the number of challenged blocks. Security proofs and performance evaluations demonstrate that the protocol offers desirable security and efficiency, making it well-suited for long-term cloud storage.

In vehicular ad hoc networks (VANETs), the core feature lies in the secure and prompt exchange of information between vehicles. To further enhance this feature, many signcryption schemes adopted to VANETs are proposed. However, most existing schemes still suffer from security or efficiency drawbacks. In this article, we propose a pairing-free certificate-based online/offline signcryption scheme constructed by elliptic curve, aiming to make the communication system of VANETs more lightweight and more secure. It not only ensures data confidentiality and unforgeability within one logic step but also eliminates the need of secure channels for key distribution. When deployed in VANETs, it can precompute many intricate operations in the offline phase and support batch verification of multiple messages. Moreover, our deployment provides pseudonym-based conditional privacy preserving for vehicles and its traceability results about malicious behaviors can be publicly verified, guaranteeing accountability. It also considers common attacks involved in VANETs, including but not limited to impersonation attacks, collusion attacks, and man-in-the-middle attacks. These advantages of our construction make it highly practical for VANETs, which have been demonstrated in our simulation experiments.

In healthcare wireless sensor networks (HWSNs), a process of medical diagnosis heavily relies on the medical data collected by lightweight sensors, as any malicious modification may result in severe consequences. Furthermore, large-scale data transmission in HWSNs would impose significant communication overhead. Therefore, efficiently guaranteeing the availability of these data while reducing the communication cost is crucial in HWSNs. Aggregate signatures suit the resource-limited environments, but existing ones still face several challenges, including vulnerability to coalition attacks, privacy preservation, and revocation of misbehaving signers. In this paper, we propose an efficient revocable identity-based aggregate signature scheme with designated verifiers (R-IBAS-DV) for HWSNs. In our proposed scheme, numerous individual signatures on the collected medical data can be aggregated into a succinct aggregate signature and their validity is equivalent to that of the aggregate signature. The equivalence property of our proposed scheme is sound even under coalition attacks and exclusively verifiable by designated healthcare professionals. Meanwhile, our proposed scheme is rooted in Hess’s practical identity-based signature, thereby circumventing costly certificate management. Additionally, it incorporates RSA accumulators to facilitate the efficient revocation of malicious signers. Security analysis and performance comparisons demonstrate that our R-IBAS-DV scheme offers enhanced security and lower computation overhead, making it particularly suitable for resource-constrained HWSNs.

Due to the introduction of cloud computing in healthcare services, personal health records (PHRs) have being uploaded to cloud servers in increasing numbers. Since data confidentiality requirements exist, data owners should encrypt their PHRs in advance of transmitting them to a cloud server. Attribute-based encryption with keyword search (ABKS) technique ensures that the encrypted PHRs are able to retrieved by other data users whose attributes match access polices granted by data owners. However, access polices are public in most existing ABKS schemes, which can reveal sensitive information contained in PHRs. In this article, we provide an efficient ABKS scheme with policy hiding for PHRs that implements the following features. (1) The fine-grained access control is achieved where data owners can authorize which data users can retrieve encrypted PHRs. (2) The access policy is hidden to safeguard sensitive information from being leaked. (3) The costs of storage and computation do not grow linearly as the number of attributes increases. The security of the presented ABKS scheme is reduced to the truncated$q$-DABDHE assumption and the DDH assumption. Its performance is also demonstrated by our extensive simulation experiments.

Signcryption can efficiently ensure data confidentiality and authenticity simultaneously, which makes it ideal for use in secure communication of vehicular ad hoc networks (VANETs). Recently, Ali et al. constructed an efficient offline/online heterogeneous signcryption protocol for edge computing-based internet of vehicles (IEEE Trans. Intell. Transp. Syst., vol. 24, no. 12, pp. 14506-14519, Dec. 2023). In this paper, we demonstrate that their signcryption protocol lacks robustness against public key replacement attacks, and then introduce an enhanced signcryption protocol secure in the random oracles. Our certificate-based and identity-based heterogeneous signcryption protocol enables communication between entities in different cryptosystems, while ensuring pseudonym-based conditional privacy for vehicles. Additionally, it supports public verification for tracing malicious behavior from anonymous vehicles. Unlike other protocols, our protocol avoids costly pairings or map-to-point hashes, and incorporates online/offline construction and batch verification, significantly reducing communication delays and computational overhead. Performance analysis shows that our protocol outperforms existing approaches in VANETs.

Abstract Federated learning (FL) is a focus of research in the area of privacy protection since it does not have the privacy issues that arise from data concentration. Although its emergence has attracted widespread attention from academia and industry, existing works on FL still face security challenges. FL can be considered as a cooperative-based task to achieve global model sharing. However, the model raises issues of cooperative security, such as free-riding and poisoning attacks. Therefore, we focus on the behavior of participants with strong cooperative relationships and build a Cooperative Security-aware Federated Learning model using blockchain. In addition, we propose a credit-based economic model including profit and punishment mechanisms to ensure fairness and security among participants. Furthermore, for data privacy, we develop a participation permission strategy to protect the privacy of participants through proxy re-encryption and homomorphic encryption. Finally, the simulation results of the real datasets show that the proposed scheme achieves a good performance in security and accuracy.

In healthcare wireless sensor networks (HWSNs), a lot of personal medical data are collected by wireless sensors. Given that a medical diagnostic analysis heavily relies on these medical data, any tampering with medical data may potentially result in severe consequences. So how to efficiently authenticate the collected medical data is a challenging work in such resource-constrained application scenarios. In this article, we propose an efficient aggregate signature (AS) scheme with designated verifier based on the Schnorr signature over an elliptic curve group for HWSNs. In our signature scheme, the set of many individual signatures on very large amounts of collected medical data can be compressed into a concise AS and their availability can be reduced to the validity of the AS only believed by a designated verifier. Unauthorized entities cannot trace the origin of these medical data authenticated by the AS. Meanwhile, our proposed scheme can also stop compressing any invalid individual signature into a valid AS even if all signers collude. Malicious signers are unable to deny the authenticated data by forging a valid AS based on several invalid individual signatures. Experimental simulation shows that the computation cost of our provably secure AS scheme is notably lower than that of existing works.

Recently, Zhu et al. introduced a conditional privacy-preserving authentication scheme based on certificateless aggregate signature (CLAS) for vehicular ad-hoc networks (VANETs). In this paper, we put forward a re-cryptanalysis to their CLAS and provide four concrete forgeries to disprove its security for VANETs. In addition, we also propose a provably secure CLAS scheme with a new aggregate algorithm for VANETs. This algorithm guarantees that the validities of all single signatures are publicly deduced to the validity of a concise aggregate signature. Performance analysis shows that our scheme not only supports public verification but also has short signature size and/or low computation cost compared with existing related works, which is very practical for VANETs.

Recently, Gayathri et al. introduced an efficient certificateless aggregate signature (CLAS) for e-healthcare system and claimed that it can achieve efficient aggregation to different signatures on different messages from different medical sensors. In this article, we analyze it and find that the CLAS scheme and the underlying certificateless signature (CLS) scheme in it are not secure. Anyone can forge a valid aggregate/individual signature replacing the legitimate medical sensor only based on the public parameters set in Gayathri et al.'s CLAS scheme. Moreover, the malicious medical sensors can deny a valid aggregate signature (AS) by offering false individual signatures aggregated into this AS. We also present a secure CLAS scheme by aggregating an existing lightweight CLS scheme. The aggregation algorithm adopted in our new scheme is proven sound against inside attacks.

With the increasing development of cloud computing, a number of users choose to outsource their data to a remote database service provider and enjoy flexible data sharing with multiple parties. However, the integrity and confidentiality of outsourced data in a remote cloud server are the main threats users are concerned about. The tree structure is one of the most widely used data organization structures. It is crucial to guarantee the integrity and privacy not only for the content but also for the structure if sensitive information is organized in this structure. At present, despite some solutions have been put forward, none of these considers the additional redaction attack on tree-structured data from attackers while sharing data with others. In this article, we provide a construction of secure redactable signature scheme for tree-structured data which features a multi-party control on the redaction of vertexes in a tree while the authenticity and privacy are assured. Then, we prove that our scheme is unforgeable, private, and transparent. Furthermore, extensive theoretical and experimental analyses are conducted to assess the efficiency of our scheme. The results demonstrate our scheme achieves multi-party redaction control without sacrificing sensible resources especially when a tree has a larger number of vertexes and siblings per vertex.

Abstract In certificateless proxy signature (CLPS), the key generation center is responsible for initializing the system parameters and can obtain the opportunity to adaptively set some trapdoors in them when wanting to launch some attacks. Until now, how to withstand the malicious-but-passive key generation center (MKGC) attacks in CLPS is still an interesting problem. In this paper, we focus on the challenging issue and introduce a CLPS scheme provably secure in the standard model. To the best of our knowledge, we are the first to demonstrate its security under MKGC attacks by adopting the technology of embedding the classic difficulty problems into the target entity public key rather than the system parameters during the security proof process.

In digital signature, strong unforgeability requires that an attacker cannot forge a new signature on any previously signed/new messages, which is attractive in both theory and practice. Recently, a strongly unforgeable certificateless signature (CLS) scheme without random oracles was presented. In this paper, we firstly show that the scheme fails to achieve strong unforgeability by forging a new signature on a previously signed message under its adversarial model. Then, we point out that the scheme is also vulnerable to the malicious-but-passive key generation center (MKGC) attacks. Finally, we propose an improved strongly unforgeable CLS scheme in the standard model. The improved scheme not only meets the requirement of strong unforgeability but also withstands the MKGC attacks. To the best of our knowledge, we are the first to prove a CLS scheme to be strongly unforgeable against the MKGC attacks without using random oracles.

ABSTRACT As the key generation center (KGC) in a certificateless cryptosystem is no longer fully trusted, for practical applications, a secure certificateless encryption (CLE) scheme should withstand malicious KGC attacks. In this paper, we investigate practical CLE schemes withstanding attacks from a malicious KGC without resorting to random oracles. At first, we show the insecurity of the CLE scheme (ZW scheme) presented by Zhang and Wang. Although ZW scheme was proved to be chosen ciphertext secure without using random oracles, we find that it cannot resist attacks from a malicious KGC. We show an attack that demonstrates a malicious KGC can easily decrypt a target ciphertext using chosen ciphertext attack. After that, we introduce a new efficient CLE scheme. We prove its CCA2 security without using random oracles. Our new scheme not only gets rid of the security drawbacks of ZW scheme but also keeps its most distinctive feature of a short public key length. The formal security proofs of the new scheme are presented without using random oracles assuming the hardness of the decisional truncated q‐augmented bilinear Diffie‐Hellman exponent (q‐ABDHE) problem and the hardness of the decisional bilinear Diffie–Hellman problem. Copyright © 2013 John Wiley & Sons, Ltd.