Privacy Requirements Research Articles

Federated In-Network Machine Learning for Privacy-Preserving IoT Traffic Analysis

The expanding use of Internet-of-Things (IoT) has driven machine learning (ML)-based traffic analysis. 5G networks’ standards, requiring low-latency communications for time-critical services, pose new challenges to traffic analysis. They necessitate fast analysis and response, preventing service disruption or security impact on network infrastructure. Distributed intelligence on IoT edge has been studied to analyze traffic, but introduces delays and raises privacy concerns. Federated learning can address privacy concerns, but does not meet latency requirements. In this article, we propose FLIP4: an efficient federated learning-based framework for in-network traffic analysis. Our solution introduces a lightweight federated tree-based model, offloaded and running within network devices. FLIP4 consumes less resources than previous solutions and reduces communication overheads, making it well-suited for IoT edge traffic analysis. It ensures prompt mitigation and minimal impact on services in the presence of false alerts using two approaches (metering and dropping), thereby balancing learning accuracy and privacy requirements.

Cloud-Based Machine Learning : Opportunities and Challenges

This comprehensive article explores the transformative impact of cloud-based machine learning (ML) on modern enterprises, examining both opportunities and challenges in implementation. The article investigates the rapidly growing cloud computing market and its ML segment, revolutionizing how organizations approach data analytics and business intelligence. Through detailed analysis of enterprise implementations, the article demonstrates how cloud ML solutions have democratized access to advanced analytics, significantly reducing operational costs while improving data processing efficiency. The article examines key aspects, including scalability advantages, cost efficiencies, and technical complexities, while providing evidence-based best practices for successful implementation. Drawing from multiple industry studies and real-world deployments, the article presents a framework for organizations to navigate challenges in data privacy, vendor dependencies, and skill requirements while maximizing the benefits of cloud-based ML solutions.

FL-APB: Balancing Privacy Protection and Performance Optimization for Adversarial Training in Federated Learning

Federated Learning (FL), as a distributed machine learning method, is particularly suitable for training models that require large amounts of data while meeting increasingly strict data privacy and security requirements. Although FL effectively protects the privacy of participants by avoiding the sharing of raw data, balancing the risks of privacy leakage with model performance remains a significant challenge. To address this, this paper proposes a new algorithm—FL-APB (Federated Learning with Adversarial Privacy–Performance Balancing). This algorithm combines adversarial training with privacy-protection mechanisms to dynamically adjust privacy and performance budgets, optimizing the balance between the two while enhancing and ensuring performance. The experimental results demonstrate that the FL-APB algorithm significantly improves model performance across various adversarial training scenarios, while effectively protecting the privacy of participants through adversarial training of privacy data.

Large language model-based optical network log analysis using LLaMA2 with instruction tuning

The optical network encompasses numerous devices and links, generating a significant volume of logs. Analyzing these logs is significant for network optimization, failure diagnosis, and health monitoring. However, the large-scale and diverse formats of optical network logs present several challenges, including the high cost and difficulty of manual processing, insufficient semantic understanding in existing analysis methods, and the strict requirements for data security and privacy. Generative artificial intelligence (GAI) with powerful language understanding and generation capabilities has the potential to address these challenges. Large language models (LLMs) as a concrete realization of GAI are well-suited for analyzing DCI logs, replacing human experts and enhancing accuracy. Additionally, LLMs enable intelligent interactions with network administrators, automating tasks and improving operational efficiency. Moreover, fine-tuning with open-source LLMs protects data privacy and enhances log analysis accuracy. Therefore, we introduce LLMs and propose a log analysis method with instruction tuning using LLaMA2 for log parsing, anomaly detection and classification, anomaly analysis, and report generation. Real log data extracted from the field-deployed network was used to design and construct instruction tuning datasets. We utilized the dataset for instruction tuning and demonstrated and evaluated the effectiveness of the proposed scheme. The results indicate that this scheme improves the performance of log analysis tasks, especially a 14% improvement in exact match rate for log parsing, a 13% improvement in F1-score for anomaly detection and classification, and a 23% improvement in usability for anomaly analysis, compared with the best baselines.

A zero-knowledge proof federated learning on DLT for healthcare data

With the increasingly widespread adoption of Healthcare 4.0 practices, new challenges have arisen for the utilization of collected sensitive data. On the one hand, these data have immense potential to unlock valuable insights for personalized medicine, early disease detection, and predictive analysis thanks to the use of Artificial Intelligence. On the other hand, ensuring the protection of patient privacy is of paramount importance to maintain trust and uphold ethical practices within the healthcare system. Classical centralized learning approaches do not fit well with the privacy and security requirements imposed by the law and the sensitivity of treated data, which is why decentralized learning approaches are gaining ground. Among these, Federated Learning (FL) stands out as a viable alternative, providing greater security and performance comparable to classic centralized learning approaches. However, there are still various attacks targeting the local parameters or gradients updated by the participants. Therefore, we present our architecture based on the conjunction of Zero-Knowledge Proof, FL, and blockchain that implements also the decentralized identifier standard. The adoption of this architecture can grant the execution, management, supervision, and updating of the FL process, guaranteeing the resilience of the system and the reliability and traceability of exchanged data. In order to test the performance, robustness, and implementation costs of the proposed architecture, we develop a case study on the prediction of blood glucose levels in people with Type-1-diabetes. The results of our analysis show an improved system in terms of balance between performance privacy and security, guaranteeing high levels of verifiability, therefore proving the proposed architecture suitable for most of the FL processes needed in the healthcare field.

TCα-PIA: A Personalized Social Network Anonymity Scheme via Tree Clustering and α-Partial Isomorphism

Social networks have become integral to daily life, allowing users to connect and share information. The efficient analysis of social networks benefits fields such as epidemiology, information dissemination, marketing, and sentiment analysis. However, the direct publishing of social networks is vulnerable to privacy attacks such as typical 1-neighborhood attacks. This attack can infer the sensitive information of private users using users’ relationships and identities. To defend against these attacks, the k-anonymity scheme is a widely used method for protecting user privacy by ensuring that each user is indistinguishable from at least k−1 other users. However, this approach requires extensive modifications that compromise the utility of the anonymized graph. In addition, it applies uniform privacy protection, ignoring users’ different privacy preferences. To address the above challenges, this paper proposes an anonymity scheme called TCα-PIA (Tree Clustering and α-Partial Isomorphism Anonymization). Specifically, TCα-PIA first constructs a similarity tree to capture subgraph feature information at different levels using a novel clustering method. Then, it extracts the different privacy requirements of each user based on the node cluster. Using the privacy requirements, it employs an α-partial isomorphism-based graph structure anonymization method to achieve personalized privacy requirements for each user. Extensive experiments on four public datasets show that TCα-PIA outperforms other alternatives in balancing graph privacy and utility.

A Novel Model for Protecting the Privacy of Digital Images in Cloud Using Permutated Fragmentation and Encryption Algorithms

Maintaining privacy is becoming increasingly challenging due to growing reliance on cloud services and software, respectively. Our data is stored in a virtual environment on unreliable cloud machines, making it susceptible to privacy breaches if not handled properly. Encrypting data before uploading it can be a solution to this problem, but it can be time-consuming. However, all of the encryption methods used to safeguard digital data so far did not fulfillment privacy and integration requirements. This is because encryption cannot function independently. Data that is encrypted and stored on a single cloud server can still be accessed by attackers, compromising the privacy of the data. In this paper, we propose a new model based on the user's classification of privacy level. The proposed model divides the digital file into multiple fragments and separately encrypts each fragment; each fragment is encrypted as separated blocks. Additionally, permutation is implemented on encrypted fragments because they are stored in the cloud with replication fragments on another cloud service. This approach ensures that even if the attacker’s gains access to one fragment, they would not be able to access the entire file, thereby safeguarding the privacy of the data.

Growth-adaptive distillation compressed fusion model for network traffic identification based on IoT cloud-edge collaboration

The development of the Internet of Things (IoT) has led to the rapid growth of the types and number of connected devices and has generated large amounts of complex and diverse traffic data. Traffic identification on edge servers solves the real-time and privacy requirements of IoT management and has attracted much attention, but still faces several problems: (1) traditional machine learning (ML) models rely on artificially constructed features, and the existing deep learning (DL) traffic identification models have reached their performance limit; and (2) insufficient computing resources of edge servers limit the possible improvement in the performance of deep learning models by increasing the number of parameters and structural complexity. To address these issues, we propose a lightweight fusion model. First, the Network-in-Network (NiN) model and Random Forest (RF) model are used on the cloud server to construct a traffic identification fusion model. The excellent representation extraction capability of the NiN compensates for the RF’s dependence on manual feature extraction, and its modular structure is suitable for the subsequent model compression operations. Then, the NiN was distilled. We propose Growth-Adaptive Distillation to lightweight the NiN model, which can reduce the operation of manually adjusting the structure of the student model and ensure the efficiency and low power consumption of the fusion model deployment. In addition, both the RF in the cloud and the distilled NiN are deployed on the edge server. Comparisons with multiple algorithms on two network traffic datasets show that the proposed model achieves state-of-the-art performance while ensuring the use of minimal computational resources.

Anti-forgetting source-free domain adaptation method for machine fault diagnosis

Unsupervised domain adaptation methods have made significant progress in the field of machine fault diagnosis. However, these methods generally assume the accessibility of source domain data during the cross-domain transfer phase. In practice, this assumption is often impractical due to data privacy requirements and the burden of data transmission and storage. Additionally, a key challenge lies in preventing the forgetting of source domain knowledge while performing cross-domain diagnosis in a source-free scenario. To address these issues, this paper proposes an anti-forgetting source-free domain adaptation method for machine fault diagnosis. Specifically, a prototype-based pseudo-label generation strategy, combined with a nearest-neighbor constraint, is employed to deeply explore diagnosis information in the unlabeled target domain data. Then, a fast nuclear-norm maximization constraint is introduced to reduce the density of target domain samples near the classification boundary, thereby decreasing the probability of misdiagnosis. Furthermore, a weighted Fisher regularization is designed to retain the diagnosis information of healthy states inherent in the source domain, thus mitigating information forgetting. Finally, comprehensive experiments are conducted to validate the superiority of our method in source-free cross-domain diagnosis from multiple perspectives, while also demonstrating its anti-forgetting properties.

Spatial interpolation of global DEM using federated deep learning

The digital elevation model (DEM) provides important data support for 3D terrain modeling. However, due to the complex and changeable terrain in the real world and the high cost of field measurement, it is extremely difficult to obtain continuous and high-density elevation data directly. Therefore, it is necessary to rely on spatial interpolation technology to restore the DEM overall picture in the original sampling area. The traditional spatial interpolation method usually has the characteristics of low model complexity and high computational cost, which leads to low real-time performance and low precision of the interpolation process. The interpolation operation based on DEM data can be considered as a special image generation process where the input is a DEM image with missing values and the output is a complete DEM image. At present, a large number of studies have proved that deep learning methods are very effective in image generation tasks. However, the training of deep learning models requires the support of a large number of high-quality data sets. DEM data in various countries, especially in key regions, are usually restricted by privacy protection regulations and cannot be disclosed. The emergence of Federated Learning (FL) provides a new solution, which supports local training on multiple end nodes, without sending local data to a remote center server for centralized training, effectively protecting data privacy. In this study, we propose a DEM interpolation model based on FL and multiScale U-Net. The experimental results show that compared with the traditional method, this model has faster processing speed and lower interpolation precision. At the same time, this research result provides a new way for efficient and secure use of terrain information, especially in those application scenarios that have strict requirements for DEM data privacy and security.

Optimum noise mechanism for differentially private queries in discrete finite sets

The differential privacy (DP) literature often centers on meeting privacy constraints by introducing noise to the query, typically using a pre-specified parametric distribution model with one or two degrees of freedom. However, this emphasis tends to neglect the crucial considerations of response accuracy and utility, especially in the context of categorical or discrete numerical database queries, where the parameters defining the noise distribution are finite and could be chosen optimally. This paper addresses this gap by introducing a novel framework for designing an optimal noise probability mass function (PMF) tailored to discrete and finite query sets. Our approach considers the modulo summation of random noise as the DP mechanism, aiming to present a tractable solution that not only satisfies privacy constraints but also minimizes query distortion. Unlike existing approaches focused solely on meeting privacy constraints, our framework seeks to optimize the noise distribution under an arbitrary (ϵ,δ)\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$(\\epsilon , \\delta )$$\\end{document} constraint, thereby enhancing the accuracy and utility of the response. We demonstrate that the optimal PMF can be obtained through solving a mixed-integer linear program. Additionally, closed-form solutions for the optimal PMF are provided, minimizing the probability of error for two specific cases. Numerical experiments highlight the superior performance of our proposed optimal mechanisms compared to state-of-the-art methods. This paper contributes to the DP literature by presenting a clear and systematic approach to designing noise mechanisms that not only satisfy privacy requirements but also optimize query distortion. The framework introduced here opens avenues for improved privacy-preserving database queries, offering significant enhancements in response accuracy and utility.

Regional Load Forecasting Scheme for Security Outsourcing Computation

Smart grids generate an immense volume of load data. When analyzed using intelligent technologies, these data can significantly improve power load management, optimize energy distribution, and support green energy conservation and emissions reduction goals. However, in the process of data utilization, a pertinent issue arises regarding potential privacy leakage concerning both regional and individual user power load data. This paper addresses the scenario of outsourcing computational tasks for regional power load forecasting in smart grids, proposing a regional-level load forecasting solution based on secure outsourcing computation. Initially, the scheme designs a secure outsourcing training protocol to carry out model training tasks while ensuring data security. This protocol guarantees that sensitive information, including but not limited to individual power consumption data, remains comprehensively safeguarded throughout the entirety of the training process, effectively mitigating any potential risks of privacy infringements. Subsequently, a secure outsourcing online prediction protocol is devised, enabling efficient execution of prediction tasks while safeguarding data privacy. This protocol ensures that predictions can be made without compromising the privacy of individual or regional power load data. Ultimately, experimental analysis demonstrates that the proposed scheme meets the requirements of privacy, accuracy, and timeliness for outsourcing computational tasks of load forecasting in smart grids.

Anonymous group structure algorithm based on community structure.

A social network is a platform that users can share data through the internet. With the ever-increasing intertwining of social networks and daily existence, the accumulation of personal privacy information is steadily mounting. However, the exposure of such data could lead to disastrous consequences. To mitigate this problem, an anonymous group structure algorithm based on community structure is proposed in this article. At first, a privacy protection scheme model is designed, which can be adjusted dynamically according to the network size and user demand. Secondly, based on the community characteristics, the concept of fuzzy subordinate degree is introduced, then three kinds of community structure mining algorithms are designed: the fuzzy subordinate degree-based algorithm, the improved Kernighan-Lin algorithm, and the enhanced label propagation algorithm. At last, according to the level of privacy, different anonymous graph construction algorithms based on community structure are designed. Furthermore, the simulation experiments show that the three methods of community division can divide the network community effectively. They can be utilized at different privacy levels. In addition, the scheme can satisfy the privacy requirement with minor changes.

Federated learning assisted distributed energy optimization

AbstractThe increased penetration of distributed energy resources and the adoption of sensing and control technologies are driving the transition from our current centralized electric grid to a distributed system controlled by multiple entities (agents). The transactive energy community serves as an established example of this transition. Distributed energy management approaches can effectively address the evolving grid's scalability, resilience, and privacy requirements. In this context, the accuracy of agents' estimations becomes crucial for the performance of distributed and multi‐agent decision‐making paradigms. This paper specifically focuses on integrating federated learning (FL) with the multi‐agent energy management procedure. FL is utilized to forecast agents' local energy generation and demand, aiming to accelerate the convergence of the distributed decision‐making process. To enhance energy aggregation in transactive energy communities, we propose an FL‐assisted distributed consensus + innovations approach. The results demonstrate that employing FL significantly reduces errors in predicting net power demand. The improved forecast accuracy, in turn, introduces less error in the distributed optimization process, thereby enhancing its convergence behaviour.

Roadmap for Linking Registry Data with Health Services Data to Support Evidence-Informed Decision-Making.

ApproachAs part of a learning period to optimize the use of RWE for decision-making for drugs for rare diseases, the [organization name removed to allow for blind review] conducted an environmental scan to map real-world data in patient registries. Over 400 patient registries were identified, signaling the potential wealth of untapped information to support decision-making by linking registry data with health services data. To better understand the challenges faced by registry holders hoping to link registry data with health services data sources available in Canada, a series of interviews were conducted with several Canadian rare disease registries. In addition, a literature review was completed, and Canadian experts in epidemiology, privacy, record linkage, registry science, and health services research were consulted to inform the development of a roadmap to meet various stakeholder needs. ResultsThe resulting roadmap consists of 8 specific steps covering topics related to registry purpose, informed consent, ethical approval, participant privacy, governance, data linkability, participant identifiability and jurisdictional requirements. ConclusionThe roadmap is currently undergoing pilot testing by a pan-Canadian rare disease registry. A final [organization name] report and an accompanying roadmap in a checklist format to facilitate implementation will be finalized, disseminated across key stakeholders, and made publicly available. ImplicationsWhile developed for registries, the roadmap applies to the linkage of clinical trial or cohort study data, or other systematically gathered patient-level data to health services data.

Applications of blockchain technology in privacy preserving and data security for real time (data) applications

SummaryBlockchain (BC) technology has been incorporated into the infrastructure of different kinds of applications that require transparency, reliability, security, and traceability. However, the BC still has privacy issues because of the possibility of privacy leaks when using publicly accessible transaction information, even with the security features offered by BCs. Specifically, certain BCs are implementing security mechanisms to address data privacy to prevent privacy issues, facilitates attack‐resistant digital data sharing and storage platforms. Hence, this proposed review aims to give a comprehensive overview of BC technology, to shed light on security issues related to BC, and to emphasize the privacy requirements for existing applications. Many proposed BC applications in asset distribution, data security, the financial industry, the Internet of Things, the healthcare sector, and AI have been explored in this article. It presents necessary background knowledge about BC and privacy strategies for obtaining these security features as part of the evaluation. This survey is expected to assist readers in acquiring a complete understanding of BC security and privacy in terms of approaches, ideas, attributes, and systems. Subsequently, the review presents the findings of different BC works, illustrating several efforts that tackled privacy and security issues. Further, the review offers a positive strategy for the previously described integration of BC for security applications, emphasizing its possible significant gaps and potential future development to promote BC research in the future.

Providing data analytic services for national knowledge users in a federated system: learnings from two Canadian use cases

Objective and ApproachCanada’s federated health data system along with access pathways geared toward academic research, pose challenges for knowledge Users (KUs) requiring timely, pan-Canadian evidence to inform decisions. To understand challenges and explore solutions, we conducted two use cases for a pan-Canadian health technology assessment organization and a federal agency, using health administrative data at one federal and six provincial data centres. The population-based cohort studies described socio-demographics, comorbidities, treatment patterns, service utilization and costs. One focused on spinal muscular atrophy, a rare disease and the other on dementia, a complex chronic disease. ResultsAdministrative challenges included aligning varied ethical review and data access policies/procedures including requiring local and/or academic principal investigators and differing definitions of “research” vs. planning, evaluation, and monitoring. Data-related challenges included differences in structure, timeliness, and completeness across regions resulting in difficulty aligning constructs such as incident cases and episodes of care. Privacy requirements prohibited pooling jurisdictional estimates resulting in “small cells” that couldn’t be shared with KUs. ConclusionsWe provided analytic outputs from multiple regions, albeit with some differences and gaps, increasing knowledge around both diseases while developing capacity for combined analyses and gaining insight into national possibilities for data access. ImplicationsAs decision makers must rely on best available information, some data is better than none. However, to improve data-analytic services for Pan-Canadian KUs, next steps will include improving data harmonization, expanding data assets and filling data gaps, implementing common data models, and exploring options for federated and/or pooled analyses.

Building access to linked data for program evaluation: lessons and opportunities from the evaluation of a pan-Canadian skills training initiative

In recent years, Canada’s statistical agency (Statistics Canada) has developed sophisticated linkage infrastructure through their Social Data Linkage Environment (SDLE), linking internal data sets held by Statistics Canada and external data held by researchers. While the SDLE presents an opportunity for better measurement of the effectiveness of policies and programs, to date it has not been used for this purpose outside of government, instead largely supporting academic research projects. This session presents a novel approach to program evaluation in the Canadian context, using the SDLE Statistics Canada administrative data alongside independently collected program data from multiple service providers to evaluate the outcomes and effectiveness of a portfolio of skills training programs. Our approach aims to both rigorously evaluate the long-term outcomes of participants in this portfolio of public skills training programs, and develop a replicable proof-of-concept for using linked data infrastructure in Canada to support program evaluations. Through this initiative, we identify the key design features needed to support this use case for linked data, including study design parameters, necessary datasets and linkage processes, and privacy and data governance policies. Finally, we identify opportunities for future replication of this approach in Canada, including strategies for expedited linkage and analysis of evaluation results that maintain privacy requirements. We find that using the SDLE infrastructure for program evaluation is both feasible and desirable, however barriers exist to replicating this approach, including sectoral capacity for collecting linkable data, and transposing linkage and confidentiality rules from an academic research to a program evaluation context.

A Novel Authenticated Quantum Anonymous Secret Sharing for Classical and Quantum Information

AbstractAnonymous secret sharing (ASS) is an essential cryptographic concept that facilitates the sharing and reconstruction of secret information while safeguarding the identity of the involved secret receivers, which has broad applications in key management, data backup, and distributed systems. In this study, a novel authenticated quantum anonymous secret sharing (QASS) protocol that emphasizes information privacy and identity anonymity protection is proposed. Employing ‐level multipartite GHZ states as a quantum resource, one‐sided anonymous entanglement (AE) is innovatively established between the dealer and anonymous receivers, enabling the dealer to distribute a random share of secret information. Additionally, by establishing a one‐sided AE between anonymous receivers and restorer, the restorer can securely collect and reconstruct the secret information using quantum teleportation (QT). Rigorous security analysis demonstrates that protocol can resist attacks from active adversaries and potentially dishonest users. Quantum experiments on IBM Qiskit validate the correctness and feasibility of the proposed QASS protocol. This work contributes to the advancement of quantum anonymous communication, addressing the requirements for information privacy and identity anonymity in practical application environments.

Reconstruction Attacks on Aggressive Relaxations of Differential Privacy

Differential privacy is a widely accepted formal privacy definition that allows aggregate information about a dataset to be released while controlling privacy leakage for individuals whose records appear in the data. Due to the unavoidable tension between privacy and utility, there have been many works trying to relax the requirements of differential privacy to achieve greater utility.One class of relaxation, which is gaining support outside the privacy community is embodied by the definitions of individual differential privacy (IDP) and bootstrap differential privacy (BDP). Classical differential privacy defines a set of neighboring database pairs and achieves its privacy guarantees by requiring that each pair of neighbors should be nearly indistinguishable to an attacker. The privacy definitions we study, however, aggressively reduce the set of neighboring pairs that are protected.To a non-expert, IDP and BDP can seem very appealing as they echo the same types of privacy explanations that are associated with differential privacy, and also experimentally achieve dramatically better utility. However, we show that they allow a significant portion of the dataset to be reconstructed using algorithms that have arbitrarily low privacy loss under their privacy accounting rules.With the non-expert in mind, we demonstrate these attacks using the preferred mechanisms of these privacy definitions. In particular, we design a set of queries that, when protected by these mechanisms with high noise settings (i.e., with claims of very low privacy loss), yield more precise information about the dataset than if they were not protected at all. The specific attacks here can be defeated and we give examples of countermeasures. However, the defenses are either equivalent to using differential privacy or to ad-hoc methods tailored specifically to the attack (with no guarantee that they protect against other attacks). Thus, the defenses emphasize the deficiencies of these privacy definitions.