Secure Computation Research Articles

Computation and communication efficient approach for federated learning based urban sensing applications against inference attacks

Federated learning based participatory sensing has gained much attention lately for the vital task of urban sensing due to privacy and security issues in conventional machine learning. However, inference attacks by the honest-but-curious application server or a malicious adversary can leak the personal attributes of the participants, such as their home and workplace locations, routines, and habits. Approaches proposed in the literature to prevent such information leakage, such as secure multi-party computation and homomorphic encryption, are infeasible for urban sensing applications owing to high communication and computation costs due to multiple rounds of communication between the user and the server. Moreover, for effective modeling of urban sensing phenomenon, the application model needs to be updated frequently — every few minutes or hours, resulting in periodic data-intensive updates by the participants, which severely strains the already limited resources of their mobile devices. This paper proposes a novel low-cost privacy-preserving framework for enhanced protection against the inference of participants’ personal and private attributes from the data leaked through inference attacks. We propose a novel approach of strategically leaking selected location traces by providing computation and communication-light direct (local) model updates, whereas the rest of the model updates (when the user is at sensitive locations) are provided using secure multi-party computation. We propose two new methods based on spatiotemporal entropy and Kullback–Leibler divergence for automatically deciding which model updates need to be sent through secure multi-party computation and which can be sent directly. The proposed approach significantly reduces the computation and communication overhead for participants compared to the fully secure multi-party computation protocols. It provides enhanced protection against the deduction of personal attributes from inferred location traces compared to the direct model updates by confusing the application server or malicious adversary while inferring personal attributes from location traces. Numerical experiments on the popular Geolife GPS trajectories dataset validate our proposed approach by reducing the computation and communication requirements by the participants significantly and, at the same time, enhancing privacy by decreasing the number of inferred sensitive and private locations of participants.

Semi‐Quantum Secure Multiparty Summation and its Applications to Anonymous Auction and Ranking

AbstractSemi‐quantum secure summation allows multiple parties with limited quantum capabilities to perform collective calculations under the premise of protecting privacy. However, current approaches are confined to low‐dimensional quantum states as information carriers and are restricted to scenarios involving two or three parties. This limitation hinders their universality and ability to fulfill the requirements of practical secure multiparty computations. To address these challenges, a novel semi‐quantum secure multiparty summation (SQSMS) protocol is proposed based on high‐dimensional entangled states and single‐particle states. The protocol leverages the entanglement correlation to guarantee the security of the transmission, enabling resistance against various potential attacks. Then, by utilizing the IBM Qiskit platform, corresponding quantum circuits and simulation results are presented to validate the feasibility and correctness of the protocol. Furthermore, building upon the foundation of the SQSMS protocol, semi‐quantum anonymous auction and anonymous ranking protocols are further constructed. Compared with previous approaches, the designed protocols cater to the requirements of their respective counterparts. Consequently, the approach may have broad applications in semi‐quantum secure multiparty computation scenarios.

Lattice-Based Threshold Secret Sharing Scheme and Its Applications: A Survey

As the most popular cryptographic scheme in the post-quantum field, lattices have received extensive attention and research. Not only do they provide quantum-resistant security, they also enable the construction of complex applications. Currently, lattice cryptography schemes based on different difficult problems have been applied in different fields. The threshold secret sharing (TSS) scheme is an important field of cryptography and has important application value and development prospects in key protection, secure multi-party computation, privacy protection, etc. However, with the rapid development of quantum computing, many existing cryptography-underlying technologies are facing huge difficulties and challenges. Therefore, post-quantum TSS has important research significance and value for the future development of cryptography. In this paper, we summarize the existing secret sharing schemes based on lattice-hard problems and the relevant applications of these schemes in the post-quantum realm. We classify existing lattice-based TSS according to different functions and introduce typical solutions. To the best of our knowledge, this is the first review paper on lattice-based TSS schemes.

Bidirectional Privacy-Preserving Network- Constrained Peer-to-Peer Energy Trading Based on Secure Multiparty Computation and Blockchain

The increasing integration of distributed energy resources has led to a network-constrained peer-to-peer (NCP2P) energy trading. Compared with conventional peer-to-peer trading without network constraints, NCP2P trading encounters challenges in protecting bidirectional privacy of both the prosumers and the grid operator when implementing market clearing and dispute resolution mechanisms. To deal with them, we apply a novel cryptographic technology called secure multiparty computation (SMPC). Firstly, a secure quadratic programming (QP) algorithm based on SMPC is designed to compute the NCP2P market clearing results with bidirectional privacy. Secondly, we propose dispute resolution mechanisms that employ blockchain to verify the correctness of claimed trading results of the two trading prosumers, and reconstruct the true trading bill based on Karush-Kuhn-Tucker conditions using SMPC when both prosumers claim wrong trading results. Finally, we design a bidirectional privacy-preserving NCP2P energy trading framework based on SMPC and blockchain. The case analysis shows that our proposed trading framework using the secure QP algorithm achieves optimal trading results, bidirectional privacy and low communication requirements for prosumers. The accuracy of the proposed dispute resolution mechanisms is also proven.

Secure multi-party computation in deep learning : Enhancing privacy in distributed neural networks

Ensuring data privacy while applying Deep Learning (DL) on distributed datasets represents an essential task in the current period of critical data security. Data privacy and accuracy of models are typically impacted by traditional methods. Data privacy is of the most tremendous significance in distributed data settings, and the current research presents a novel model for DL termed Secure Multi-Party Computation (SMPC). The accuracy of the mathematical models and confidentiality of the data are frequently compelled to coexist in conventional methods. In order to enable collaborative DL without compromising private information, the recommended system uses the Paillier Homomorphic Encryption Scheme (PHES). By using innovative cryptographic methods, this decentralized method secures the confidentiality of data without utilizing a Trusted Authority (TA). By performing a thorough assessment of the CIFAR-10 and IMDB datasets, the present study demonstrates that the system the author uses is simple and scalable and that it offers accuracy on par with conventional methods. By presenting an approach for achieving a balance between the two competing demands of data security and computational performance, this method signifies a vast advance forward with a confidentiality DL.

VeriTrain: Validating MLaaS Training Efforts via Anomaly Detection

Machine learning as a service (MLaaS) offers users the benefit of training state-of-the-art neural network models on fast hardware with low costs. However, it also brings security concerns since the user does not fully trust the cloud. To prove to the user that the ML training results are legitimate, existing approaches mainly adopt cryptographic techniques such as secure multi-party computation, which incur large overheads. In this paper, we model the problem of verifying ML training efforts as an anomaly detection problem. We design a verification system, dubbed <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VeriTrain</small> , which combines unsupervised anomaly detection approaches and hypothesis testing techniques to verify the legitimacy of training efforts on the MLaaS cloud. <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VeriTrain</small> is run inside trusted execution environments (TEEs) on the same cloud machine to ensure the integrity of its execution. We consider a threat model where the cloud model trainer is a lazy attacker and tries to fool <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VeriTrain</small> with minimum training effort. We perform extensive evaluations on multiple neural network models and datasets, which shows that <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VeriTrain</small> performs well in detecting parameter updates crafted by the attacker. We also implement <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VeriTrain</small> with Intel SGX and show that it only incurs moderate overheads.

SGXonerate:Finding (and Partially Fixing) Privacy Flaws in TEE-based Smart Contract Platforms Without Breaking the TEE

TEE-based smart contracts are an emerging blockchain architecture, offering fully programmable privacy with better performance than alternatives like secure multiparty computation. They can also support compatibility with existing smart contract languages, such that existing (plaintext) applications can be readily ported, picking up privacy enhancements automatically. While previous analysis of TEE-based smart contracts have focused on failures of TEE itself, we asked whether other aspects might be understudied. We focused on state consistency, a concern area highlighted by Li et al., as well as new concerns including access pattern leakage and software upgrade mechanisms. We carried out a code review of a cohort of four TEE-based smart contract platforms. These include Secret Network, the first to market with in-use applications, as well as Oasis, Phala, and Obscuro, which have at least released public test networks. The first and most broadly applicable result is that access pattern leakage occurs when handling persistent contract storage. On Secret Network, its fine-grained access pattern is catastrophic for the transaction privacy of SNIP-20 tokens. If ERC-20 tokens were naively ported to Oasis they would be similarly vulnerable; the others in the cohort leak coarse-grained information at approximately the page level (4 kilobytes). Improving and characterizing this will require adopting techniques from ORAMs or encrypted databases. Second, the importance of state consistency has been underappreciated, in part because exploiting such vulnerabilities is thought to be impractical. We show they are fully practical by building a proof-of-concept tool that breaks all advertised privacy properties of SNIP-20 tokens, able to query the balance of individual accounts and the token amount of each transfer. We additionally demonstrate MEV attacks against the Sienna Swap application. As a final consequence of lacking state consistency, the developers have inadvertently introduced a decryption backdoor through their software upgrade process. We have helped the Secret developers mitigate this through a coordinated vulnerability disclosure, after which their state consistency should be roughly on par with the rest.

Federated Cross-Incremental Self-Supervised Learning for Medical Image Segmentation.

Federated cross learning has shown impressive performance in medical image segmentation. However, it encounters the catastrophic forgetting issue caused by data heterogeneity across different clients and is particularly pronounced when simultaneously facing pixelwise label deficiency problem. In this article, we propose a novel federated cross-incremental self-supervised learning method, coined FedCSL, which not only can enable any client in the federation incrementally yet effectively learn from others without inducing knowledge forgetting or requiring massive labeled samples, but also preserve maximum data privacy. Specifically, to overcome the catastrophic forgetting issue, a novel cross-incremental collaborative distillation (CCD) mechanism is proposed, which distills explicit knowledge learned from previous clients to subsequent clients based on secure multiparty computation (MPC). Besides, an effective retrospect mechanism is designed to rearrange the training sequence of clients per round, further releasing the power of CCD by enforcing interclient knowledge propagation. In addition, to alleviate the need of large-scale densely annotated pretraining medical datasets, we also propose a two-stage training framework, in which federated cross-incremental self-supervised pretraining paradigm first extracts robust yet general image-level patterns across multi-institutional data silos via a novel round-robin distributed masked image modeling (MIM) pipeline; then, the resulting visual concepts, e.g., semantics, are transferred to the federated cross-incremental supervised fine-tuning paradigm, favoring various cross-silo medical image segmentation tasks. The experimental results on public datasets demonstrate the effectiveness of the proposed method as well as the consistently superior performance of our method over most state-of-the-art methods quantitatively and qualitatively.

Hybrid Fusion of Lightweight Security Frameworks Using Data Mining Approach in IoT

The rapid adoption of the Internet of Things throughout healthcare and smart city construction has led to a rise in networked devices and security issues. This work suggests new techniques to improve IoT safety and maximise computing resources. We develop a complete security architecture integrating lightweight cryptography, blockchain, machine learning anomaly detection, and federated learning. We did so because we know that traditional security measures are inadequate for the Internet of Things. The lightweight cryptographic algorithm (LCA) provides efficient encryption and decryption, making it ideal for low-resource Internet of Things devices. Twenty processes comprise the LCA design. These operations include key generation, data encryption, digital signatures, and integrity checking. These procedures secure IoT data transfers. ADML detects anomalies in encrypted Internet of Things data using machine learning. This approach may identify security issues better. To keep up with data trends, this method extracts features, trains models, and updates them. Blockchain-based data integrity (BDI) is the third element. Blockchain ensures that Internet of Things data is reliable and full. BDI developed an immutable ledger solution to increase IoT data security and dependability. This data integrity system generates blocks, hashes, confirms blocks, and updates the blockchain. Fourth, FLIoT (Federated Learning for the Internet of Things) emphasises data privacy and collaborative model training across IoT devices. Foundation for the Internet of Things (FIoT) protocols and standards aim to increase IoT devices' collective intelligence while safeguarding users' privacy. It includes local model training, model aggregation, and the latest global model distribution. Our work also uses Secure Multi-party Computation (SMC) to analyse data more thoroughly and continuously, addressing online transaction cybersecurity issues. The framework outperforms the current state of the art in memory use, energy consumption, anomaly detection accuracy and precision, and encryption and decryption time. The Hybrid Fusion Framework combines lightweight cryptographic algorithms with federated learning, machine learning, blockchain technology, and other similar technologies to provide an effective, adaptable, and affordable IoT security solution.

МЕТОД БАГАТОСТОРОННІХ ОБЧИСЛЕНЬ КЕРУВАННЯ КРИПТОАКТИВАМИ НА ОСНОВІ ТЕХНОЛОГІЇ MULTI-PARTY COMPUTATION

The task of building an efficient and secure crypto-currency wallet based on multi-party computing technology, which can provide reliable and secure use in various industries working with cryptocurrencies, is considered. Modern algorithms of multilateral calculations, such as those of Gennaro and Goldfeder, Lindella, Doerner, offer significant advantages in data protection, but they have certain limitations. The main ones are insufficient efficiency in the number of signature rounds and lack of cold storage support. This highlights the need for further research and development in the field of Multi-Party Computation technology to achieve higher levels of operational efficiency and flexibility. The method of ensuring the confidentiality and security of cryptoassets on the dream technology of multilateral computing for the Ethereum network consists in: ensuring a high level of security for the storage of private keys and the process of signing the transaction, using advanced cryptographic protection methods; guaranteeing the confidentiality and privacy of users, protecting their personal information from unauthorized access or data leakage; using advanced technologies, including Multi-Party Computation technology, to increase the level of security and privacy of asset storage. On the basis of the conducted research and analysis of the characteristics of multiparty computing protocol standards, a protocol is proposed that takes into account the existing improvements and shortcomings of previous protocols. To improve security, the protocol implements the principle of interval update of the shared secret. A cryptoasset management system based on multilateral computing includes several important stages: choosing cloud providers, creating and configuring instances, containerizing applications, configuring secure communication and routing. The system provides the possibility of reliable storage of cryptoassets, in particular, private keys to wallets of the Ethereum network, using the Multi-Party Computation technology, which can be used in various cryptoprojects for the safe transfer of cryptoassets. The architecture of the system involves the interaction of the main components to ensure reliable and safe operation. A scheme for establishing a secure connection of nodes is proposed and a local storage model is described, which provides secure storage of key shares, signatures, and system logs.

Integrating Improved Mobile Net and Homomorphic Encryption in Hybrid IoT Security Frameworks for Enhanced Resilience Against Advanced Persistent Threats

IoT devices have transformed smart cities and healthcare. The expanding usage of IoT devices creates major security threats, leaving critical systems vulnerable to sophisticated and persistent assaults. Our hybrid IoT security approach employs homomorphic encryption and improved MobileNet to protect data and simplify feature extraction. Our extensive testing and assessment prove that the proposed structure makes IoT settings more resistant to sophisticated persistent attacks. We discovered superior methodologies for F1 score, accuracy, precision, and memory performance measurement. To ensure data privacy and security during analysis and transmission, homomorphic encryption is incorporated. Our ablation research lays out each framework component's contributions. To increase system speed, it emphasizes safe data processing, real-time analytical optimization, lightweight feature extraction, and privacy-preserving computing. The scalability study indicates that the framework can scale with IoT installations while maintaining peak performance and resource efficiency. Finally, the hybrid IoT security architecture improves IoT security. It provides a full and effective security solution for IoT infrastructure. Lawmakers, business experts, and students in the sector may learn from this research regarding genuine IoT security systems.

Efficiently Compiling Secure Computation Protocols From Passive to Active Security: Beyond Arithmetic Circuits

This work studies compilation of honest-majority semi-honest secure multi-party protocols secure up to additive attacks to maliciously secure computation with abort. Prior work concentrated on arithmetic circuits composed of addition and multiplication gates, while many practical protocols rely on additional types of elementary operations or gates to achieve good performance. In this work we revisit the notion of security up to additive attacks in the presence of additional gates such as random element generation and opening. This requires re-evaluation of functions that can be securely evaluated, extending the notion of protocols secure up to additive attacks, and re-visiting the notion of delayed verification that points to weaknesses in its prior use and designing a mitigation strategy. We transform the computation using dual execution to achieve security in the malicious model with abort and experimentally evaluate the difference in performance of semi-honest and malicious protocols to demonstrate the low cost.

OPTIMIZED CLOUD SECURITY ECC-ENHANCED HOMOMORPHIC PAILLIER RE-ENCRYPTION

In the dynamic domain of cloud computing, ensuring data security is of utmost importance. Conventional encryption techniques, while providing a high level of security, introduce substantial computational burdens, rendering them impractical for environments with limited resources. In response to this predicament, our study introduces a novel lightweight encryption framework that amalgamates Elliptic Curve Cryptography (ECC) with Homomorphic Paillier Re-Encryption, thereby reinforcing the security of data within cloud infrastructures.Our methodology exploits the inherent advantages of ECC, notably its ability to maintain stringent security measures with relatively smaller key dimensions, thereby optimizing efficiency without sacrificing the level of security. The integration of ECC with Homomorphic Paillier Encryption facilitates the execution of secure computations on ciphered data, maintaining user privacy while permitting the cloud to perform meaningful data operations. The re-encryption feature of our scheme ensures the secure mobility and modification of data sans decryption, thus augmenting security measures and operational flexibility.The proposed encryption paradigm has been validated through rigorous theoretical scrutiny and empirical implementation, revealing marked enhancements in both computational efficiency and security measures when juxtaposed with established encryption techniques. The empirical evidence suggests that the streamlined nature of our encryption scheme renders it exceptionally compatible with real-world cloud applications, particularly in scenarios where the optimization of resources is imperative.This work contributes to the field of cloud data security by providing a scalable, efficient, and secure encryption solution, paving the way for more secure and practical cloud computing applications.

A Review of AIoT-based Human Activity Recognition: From Application to Technique.

This scoping review paper redefines the Artificial Intelligence-based Internet of Things (AIoT) driven Human Activity Recognition (HAR) field by systematically extrapolating from various application domains to deduce potential techniques and algorithms. We distill a general model with adaptive learning and optimization mechanisms by conducting a detailed analysis of human activity types and utilizing contact or non-contact devices. It presents various system integration mathematical paradigms driven by multimodal data fusion, covering predictions of complex behaviors and redefining valuable methods, devices, and systems for HAR. Additionally, this paper establishes benchmarks for behavior recognition across different application requirements, from simple localized actions to group activities. It summarizes open research directions, including data diversity and volume, computational limitations, interoperability, real-time recognition, data security, and privacy concerns. Finally, we aim to serve as a comprehensive and foundational resource for researchers delving into the complex and burgeoning realm of AIoT-enhanced HAR, providing insights and guidance for future innovations and developments.

Computational genetic epidemiology: Leveraging HPC for large-scale AI models based on Cyber Security

To better understand disease susceptibility and prevention, computational genetic epidemiology is leading research. This paper introduces GenomeMinds, a breakthrough method for scaling large-scale AI models for disease risk prediction. HPC was used to develop the method. GenomeMinds is compared to six standard methods to demonstrate its benefits. GenomeMinds' incredible potential is shown by real-world performance assessments. These measures evaluate data processing speed, forecast accuracy, scalability, computer efficiency, privacy, and ethics. GenomeMinds benefits are shown via scatter plots, which visually compare data. According to the data, GenomeMinds may revolutionize computational genetic epidemiology by doing well across all criteria. GenomeMinds has faster data processing, better prediction accuracy, stronger scalability, higher computational efficiency, enhanced privacy and security, and a comprehensive ethical awareness.

Privacy-preserving cryptographic solutions for medical image protection : The secure force algorithm and intelligent encryption systems

Privacy-preserving must comply with regulatory requirements, safeguard patient privacy, secure data integrity, enable evolving technologies, and guarantee moral and reliable healthcare practices for healthcare applications. By adopting these measures, the healthcare industry may effectively harness the advantages offered by contemporary technology while simultaneously ensuring the protection of patient confidentiality. In this paper, a Secure Force (SF) Encryption Algorithm is modified and improved to guarantee data confidentiality, integrity, and privacy for healthcare applications. Furthermore, SF is also integrated into Intelligent Encryption Systems to augment the protection and confidentiality of confidential patient information. A thorough examination of the algorithm’s efficacy is performed using MATLAB regarding computational efficiency, security, and influence on data entropy. Results demonstrate the algorithm’s ability to preserve patient anonymity, protect sensitive data, and augment data security. Addressing these vital aspects, the research contributes to the ongoing effort to secure medical data in a rapidly evolving healthcare landscape.

Libras: A Fair, Secure, Verifiable, and Scalable Outsourcing Computation Scheme Based on Blockchain

Libras: A Fair, Secure, Verifiable, and Scalable Outsourcing Computation Scheme Based on Blockchain

Blockchain-Empowered Edge Intelligence for TACS Obstacle Detection: System Design and Performance Optimization

With the significant advantages of system complexity and operating costs, Train Autonomous Circumambulate System (TACS) is gradually replacing the traditional Communication Based Train Control (CBTC) system as the next-generation train operation control system development direction. As train operation and control become more decentralized and autonomous, real-time and accurate obstacle detection, apart from route-level protection, is quite desirable in TACS. Most of the existing researches about obstacle detection focus on detection algorithm optimization based on the once-deployed lifelong use principle, while model re-optimization based on the actual operating environment under unexpected situations and model sharing among multi-users are largely ignored. In this paper, we design a novel obstacle detection system in TACS based on blockchain-empowered Edge Intelligence (EI). To make full use of the massive raw unannotated data collected online, we first propose an SSL-based TACS obstacle detection model. Considering the resource-hungry model training, we introduce EI into TACS and propose a MARL-based task offloading algorithm for secure and efficient computation offloading coordination. Furthermore, we propose a blockchain-based model sharing scheme to facilitate the multi-model parameter exchange and improve the obstacle detection accuracy. Extensive simulation results show that the designed obstacle detection system can effectively improve the TACS obstacle detection performance.

Blockchain-Based Secure Content Caching and Computation for Edge Computing

Blockchain-Based Secure Content Caching and Computation for Edge Computing

Generalized Lagrange Coded Computing: A Flexible Computation-Communication Tradeoff for Resilient, Secure, and Private Computation

Generalized Lagrange Coded Computing: A Flexible Computation-Communication Tradeoff for Resilient, Secure, and Private Computation