Privacy-preserving Approach Research Articles

Differentially Private Decentralized Generator Maintenance and Operations for Power Networks

Increasing sensor instrumentation has been critical to the improvement of power network operations and maintenance while challenges linked to data ownership and privacy remain. To this end, decentralized methods have been popular for their computational scalability while enabling data ownership by utility stakeholders. However, decentralization also requires sharing network flow estimates over public channels raising privacy concerns. In this paper, we propose a differential privacy driven, ADMM approach for decentralized, mixed integer optimization of operations and maintenance problems which preserves privacy of network flow estimates. We prove strong privacy guarantees by leveraging the linear relationship between the phase angles and flow. To address challenges associated with the mixed integer and dynamic nature of the problem, we introduce an exponential moving average based consensus mechanism to enhance convergence, coupled with a control chart based convergence criteria to improve stability. Our results demonstrate that our privacy preserving approach is robust to a range of noise levels and operational scenarios, delivering solutions on par with benchmark methods.

ExTRUST: Reducing Exploit Stockpiles With a Privacy-Preserving Depletion System for Inter-State Relationships

Cyberspace is a fragile construct threatened by malicious cyber operations of different actors, with vulnerabilities in IT hardware and software forming the basis for such activities, thus also posing a threat to global IT security. Advancements in the field of artificial intelligence accelerate this development, either with artificial intelligence enabled cyber weapons, automated cyber defense measures, or artificial intelligence-based threat and vulnerability detection. Especially state actors, with their long-term strategic security interests, often stockpile such knowledge of vulnerabilities and exploits to enable their military or intelligence service cyberspace operations. While treaties and regulations to limit these developments and to enhance global IT security by disclosing vulnerabilities are currently being discussed on the international level, these efforts are hindered by state concerns about the disclosure of unique knowledge and about giving up tactical advantages. This leads to a situation where multiple states are likely to stockpile at least some identical exploits, with technical measures to enable a depletion process for these stockpiles that preserve state secrecy interests and consider the special constraints of interacting states as well as the requirements within such environments being non-existent. This paper proposes such a privacy-preserving approach that allows multiple state parties to privately compare their stock of vulnerabilities and exploits to check for items that occur in multiple stockpiles without revealing them so that their disclosure can be considered. We call our system EXTRUST and show that it is scalable and can withstand several attack scenarios. Beyond the intergovernmental setting, EXTRUST can also be used for other zero-trust use cases, such as bug-bounty programs.

Federated Kalman Filter for Secure IoT-Based Device Monitoring Services

Device monitoring services have increased in popularity with the evolution of recent technology and the continuously increased number of Internet of Things (IoT) devices. Among the popular services are the ones that use device location information. However, these services run into privacy issues due to the nature of data collection and transmission. In this work, we introduce a platform incorporating Federated Kalman Filter (FKF) with a federated learning approach and private blockchain technology for privacy preservation. We analyze the accuracy of the proposed design against a standard Kalman Filter (KF) implementation of localization based on the Received Signal Strength Indicator (RSSI). The experimental results reveal significant potential for improved data estimation for RSSI-based localization in device monitoring.

Compressed Sensing-Based Privacy Preserving in Labeled Dynamic Social Networks

With the rapid development of social networks, privacy- preserving is an important issue. Nowadays, privacy-preserving mainly involves static social networks. In fact, social networks are dynamic. The sequential release of social networks will lead to privacy leakage. Many studies have employed <inline-formula><tex-math notation="LaTeX">$K$</tex-math></inline-formula>-anonymity to protect users&#x0027; privacy, but because <inline-formula><tex-math notation="LaTeX">$K$</tex-math></inline-formula> affects data utility in social networks, a smaller <inline-formula><tex-math notation="LaTeX">$K$</tex-math></inline-formula> cannot provide enough privacy preserving, whereas a larger <inline-formula><tex-math notation="LaTeX">$K$</tex-math></inline-formula> will lead to a significant loss of data utility. This article proposes a new privacy-preserving approach based on compressed sensing (CS) to protect privacy in the labeled dynamic social networks. The scheme first uses the CS to compress the updated node information at each time point. Then, in the form of label grouping, the node link relationship is randomly deleted/changed to blur the node degrees to protect privacy. Finally, the reconstruction algorithm and Gaussian measurement matrix are combined to ensure operating efficiency while balancing data privacy and data utility. The simulation results show that our scheme retains data utility better than <inline-formula><tex-math notation="LaTeX">$l$</tex-math></inline-formula>-diversity and <inline-formula><tex-math notation="LaTeX">$K$</tex-math></inline-formula>-anonymity. Furthermore, according to privacy analysis, the scheme can also protect link relationship privacy and label privacy at the same time, and prevent background knowledge attacks.

UtilityAware: A framework for data privacy protection in e-health

Data privacy in e-health deals with the protection of sensitive medical information that is collected, stored, and analyzed in electronic health systems. Several organizations publish sensitive person-specific data for research purposes. E-health data and related domains are the loci of research. First, in publishing sensitive person-specific datasets, ensuring the privacy of user sensitive information is an issue. Secondly, ascertaining both privacy-preservation and data utility simultaneously are contradictory to each other. In addition, all transactions have the same prior belief that may result in erroneous modeling and privacy breaches. To refrain from the belief of an adversary and to solve the above discussed issues, a semantic privacy guarantee must be ensured before publishing data by any organization. This paper proposes a solution to the former issue, a framework for privacy preservation of structured datasets in ascertaining that an adversary has low confidence in extrapolation. The latter problem is also tackled by the proposed framework that combines stratified sampling with generalization to achieve representative semantic privacy-preservation with high data utility. Moreover, this study presents a mathematical proof that the proposed framework achieves differential privacy. Our experimental results show that our algorithm provides better data utility and privacy simultaneously. The proposed framework achieves 3% and 0.04% higher classification accuracy and low relative error, respectively, compared to state-of-the-art existing privacy-preservation approaches.

Applications of Federated Learning in Mobile Health: Scoping Review.

The proliferation of mobile health (mHealth) applications is partly driven by the advancements in sensing and communication technologies, as well as the integration of artificial intelligence techniques. Data collected from mHealth applications, for example, on sensor devices carried by patients, can be mined and analyzed using artificial intelligence-based solutions to facilitate remote and (near) real-time decision-making in health care settings. However, such data often sit in data silos, and patients are often concerned about the privacy implications of sharing their raw data. Federated learning (FL) is a potential solution, as it allows multiple data owners to collaboratively train a machine learning model without requiring access to each other's raw data. The goal of this scoping review is to gain an understanding of FL and its potential in dealing with sensitive and heterogeneous data in mHealth applications. Through this review, various stakeholders, such as health care providers, practitioners, and policy makers, can gain insight into the limitations and challenges associated with using FL in mHealth and make informed decisions when considering implementing FL-based solutions. We conducted a scoping review following the guidelines of PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews). We searched 7 commonly used databases. The included studies were analyzed and summarized to identify the possible real-world applications and associated challenges of using FL in mHealth settings. A total of 1095 articles were retrieved during the database search, and 26 articles that met the inclusion criteria were included in the review. The analysis of these articles revealed 2 main application areas for FL in mHealth, that is, remote monitoring and diagnostic and treatment support. More specifically, FL was found to be commonly used for monitoring self-care ability, health status, and disease progression, as well as in diagnosis and treatment support of diseases. The review also identified several challenges (eg, expensive communication, statistical heterogeneity, and system heterogeneity) and potential solutions (eg, compression schemes, model personalization, and active sampling). This scoping review has highlighted the potential of FL as a privacy-preserving approach in mHealth applications and identified the technical limitations associated with its use. The challenges and opportunities outlined in this review can inform the research agenda for future studies in this field, to overcome these limitations and further advance the use of FL in mHealth.

A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection

Federated learning has been a hot research topic in enabling the collaborative training of machine learning models among different organizations under the privacy restrictions. As researchers try to support more machine learning models with different privacy-preserving approaches, there is a requirement in developing systems and infrastructures to ease the development of various federated learning algorithms. Similar to deep learning systems such as PyTorch and TensorFlow that boost the development of deep learning, federated learning systems (FLSs) are equivalently important, and face challenges from various aspects such as effectiveness, efficiency, and privacy. In this survey, we conduct a comprehensive review on federated learning systems. To achieve smooth flow and guide future research, we introduce the definition of federated learning systems and analyze the system components. Moreover, we provide a thorough categorization for federated learning systems according to six different aspects, including data distribution, machine learning model, privacy mechanism, communication architecture, scale of federation and motivation of federation. The categorization can help the design of federated learning systems as shown in our case studies. By systematically summarizing the existing federated learning systems, we present the design factors, case studies, and future research opportunities.

Collaborative Privacy-preserving Approaches for Distributed Deep Learning Using Multi-Institutional Data.

Deep learning (DL) algorithms have shown remarkable potential in automating various tasks in medical imaging and radiologic reporting. However, models trained on low quantities of data or only using data from a single institution often are not generalizable to other institutions, which may have different patient demographics or data acquisition characteristics. Therefore, training DL algorithms using data from multiple institutions is crucial to improving the robustness and generalizability of clinically useful DL models. In the context of medical data, simply pooling data from each institution to a central location to train a model poses several issues such as increased risk to patient privacy, increased costs for data storage and transfer, and regulatory challenges. These challenges of centrally hosting data have motivated the development of distributed machine learning techniques and frameworks for collaborative learning that facilitate the training of DL models without the need to explicitly share private medical data. The authors describe several popular methods for collaborative training and review the main considerations for deploying these models. They also highlight publicly available software frameworks for federated learning and showcase several real-world examples of collaborative learning. The authors conclude by discussing some key challenges and future research directions for distributed DL. They aim to introduce clinicians to the benefits, limitations, and risks of using distributed DL for the development of medical artificial intelligence algorithms. ©RSNA, 2023 Quiz questions for this article are available in the supplemental material.

A Novel Approach for Location Privacy Preservation of Clusters

Location privacy is one of the burning issues in respect of the LBS and other services, since there are several risks of data leakage happening all over the world as well as the location data leakage is also happening. Withthe increment of these data breaches several advertisements and bad cyber crime happens with different person. To solve these issues and to secure the location privacy as well as location data hiding we have applied k- anonymity algorithm to anonymise the data sets so that the location information of users will hide themselves. The aim of our paper is to demonstrate the novel approach for location privacy preservation using machine learning and deep learning algorithms applicable on various parameters as well as on various data sets. We have applied L diversity and T closeness for the same so that we make sure about the protection of the individual stored data dets as well at the same time.

Non-Intrusive Privacy-Preserving Approach for Presence Monitoring Based on WiFi Probe Requests

Monitoring the presence and movements of individuals or crowds in a given area can provide valuable insight into actual behavior patterns and hidden trends. Therefore, it is crucial in areas such as public safety, transportation, urban planning, disaster and crisis management, and mass events organization, both for the adoption of appropriate policies and measures and for the development of advanced services and applications. In this paper, we propose a non-intrusive privacy-preserving detection of people's presence and movement patterns by tracking their carried WiFi-enabled personal devices, using the network management messages transmitted by these devices for their association with the available networks. However, due to privacy regulations, various randomization schemes have been implemented in network management messages to prevent easy discrimination between devices based on their addresses, sequence numbers of messages, data fields, and the amount of data contained in the messages. To this end, we proposed a novel de-randomization method that detects individual devices by grouping similar network management messages and corresponding radio channel characteristics using a novel clustering and matching procedure. The proposed method was first calibrated using a labeled publicly available dataset, which was validated by measurements in a controlled rural and a semi-controlled indoor environment, and finally tested in terms of scalability and accuracy in an uncontrolled crowded urban environment. The results show that the proposed de-randomization method is able to correctly detect more than 96% of the devices from the rural and indoor datasets when validated separately for each device. When the devices are grouped, the accuracy of the method decreases but is still above 70% for rural environments and 80% for indoor environments. The final verification of the non-intrusive, low-cost solution for analyzing the presence and movement patterns of people, which also provides information on clustered data that can be used to analyze the movements of individuals, in an urban environment confirmed the accuracy, scalability and robustness of the method. However, it also revealed some drawbacks in terms of exponential computational complexity and determination and fine-tuning of method parameters, which require further optimization and automation.

A Privacy-Preserving Approach to Identify COVID-19 Infection Origins via Volunteered Share of Health Data Records by Mobile Users

Human-beings are suffering from the rapid spread of COVID-19 throughout the world. In order to quickly identify, quarantine and cure the infected people, and to stop further infections, it is crucial to expose those origins who have been infected but are asymptomatic. However, this task is not easy, especially when the rigid security and privacy constraints on health records are taken into consideration. In this paper, we develop a new method to solve this problem. In the outbreak of a disease like COVID-19, the proposed method can find hidden infected people (or communities) through volunteered share of health data by some mobile users. Such volunteers only reveal whether they are healthy or infected e.g. through they mobile apps. This approach minimises health data disclosure and preserves privacy for the others. There are three steps in the proposed method. First, we borrow the idea from traditional epidemiology and design a novel algorithm to estimate the number of infection origins based on a Susceptible-Infected model. Second, we introduce the concept of ‘heavy centre’ to locate those origins. The probability of each node being infected will then be derived by building a spreading model based on the origins. To evaluate our method, we conduct a series of experiments on various networks with different structures. We examine the performance in estimating the number of origins as well as their origins. The results show that the proposed method yields higher accuracies than the existing methods, even when the fraction of volunteers is small.

Physical Internet towards sustainable supply chain: The significance of digital interoperability

The purpose of this study is to discover physical internet attributes for satisfying economic, social, and environmental benefits, and frame the hierarchical interrelationships of these attributes. The fuzzy Delphi method is used to define necessary criteria; fuzzy Decision-Making Trial and Evaluation Laboratory is adopted to identify the cause-and-effect relationships based on the experts’ viewpoints. Finding - The result also presents the most important criteria having causal effects on others including privacy preservation approaches, system of logistics network interconnectivity, distributed data storage, real-time locating system, and distributed multi-segment intermodal transport, from which recommend practical implications for the decision-making process. This study suggests both theoretical and managerial implications for improving the physical internet towards sustainable supply chain as physical internet has been perceived as a novel logistics model towards sustainability and a key solution for the current paradigms to overcome sustainable issues. This study contributes to the literature by pointing out the causal aspects for physical internet implementation, which are digital interoperability of supply chain, and shared warehouse and transportation networks.

Protecting Sensitive Attributes by Adversarial Training Through Class-Overlapping Techniques

In recent years, machine learning as a service (MLaaS) has brought considerable convenience to our daily lives. However, these services raise the issue of leaking users’ sensitive attributes, such as race, when provided through the cloud. The present work overcomes this issue by proposing an innovative privacy-preserving approach called privacy-preserving class overlap (PPCO), which incorporates both a Wasserstein generative adversarial network and the idea of class overlapping to obfuscate data for better resilience against the leakage of attribute-inference attacks(i.e., malicious inference on users’ sensitive attributes). Experiments show that the proposed method can be employed to enhance current state-of-the-art works and achieve superior privacy–utility trade-off. Furthermore, the proposed method is shown to be less susceptible to the influence of imbalanced classes in training data. Finally, we provide a theoretical analysis of the performance of our proposed method to give a flavour of the gap between theoretical and empirical performances.

Study and Analysis of Various Cloud Security, Authentication, and Data Storage Models

In recent days, widespread acceptance of cloud data storage applications increases various privacy problems and security problems. Outsourced data security is considered the main confrontation for cloud clients because of data control loss. This review presents a detailed survey of 50 research papers presenting privacy preservation approaches, namely authentication-based, cloud security-based, data storage-based, data security-based, and encryption-based techniques. The analysis is considered based on the categorization of approaches, dataset employed, utilized software tools, published year, and the performance metrics are discussed. Furthermore, problems raised in existing privacy preservation techniques are elucidated in the research gaps and problems section. The future work of this study is based on the research gaps and problems recognized from present research schemes. Additionally, JAVA software language is widely utilized for implementing privacy preservation models, and the Amazon access sample database is a commonly employed dataset for the privacy preservation approach.

Tokens Shuffling Approach for Privacy, Security, and Reliability in IoHT under a Pandemic

Privacy and security are unavoidable challenges in the future of smart health services and systems. Several approaches for preserving privacy have been provided in the Internet of Health Things (IoHT) applications. However, with the emergence of COVID-19, the healthcare centers needed to track, collect, and share more critical data such as the location of those infected and monitor social distancing. Unfortunately, the traditional privacy-preserving approaches failed to deal effectively with emergency circumstances. In the proposed research, we introduce a Tokens Shuffling Approach (TSA) to preserve collected data’s privacy, security, and reliability during the pandemic without the need to trust a third party or service providers. TSA depends on a smartphone application and the proposed protocol to collect and share data reliably and safely. TSA depends on a proposed algorithm for swapping the identities temporarily between cooperated users and then hiding the identities by employing fog nodes. The fog node manages the cooperation process between users in a specific area to improve the system’s performance. Finally, TSA uses blockchain to save data reliability, ensure data integrity, and facilitate access. The results prove that TSA performed better than traditional approaches regarding data privacy and the performance level. Further, we noticed that it adapted better during emergency circumstances. Moreover, TSA did not affect the accuracy of the collected data or its related statistics. On the contrary, TSA will not affect the quality of primary healthcare services.

A homomorphic approach for security and privacy preservation of Smart Airports

The concept of Smart Airport is emerging in the Aviation sector because of the industry 4.0 revolution. To provide the best facility to the air travelers, most of the airports have enhanced their infrastructure using smart automation and Cloud Services (CS), etc. which play a significant role in increasing the revenue of the aviation industry. As the use of digital technology at airports is increasing, they are becoming progressively vulnerable to cyber-attacks and data breaches. The data collected by the airports is stored in specialized cloud databases, which ensures the availability of the data at any time. However, cloud servers have been compromised several times which leads to a lack of trust by the consumers. To meet the privacy and security objectives of the data collected at the airports and stored in the cloud, this research proposes a Searchable Encryption (SE) scheme based on Partial Homomorphic Encryption (PHE). The proposed scheme allows users to search on the encrypted data stored in the cloud hence enhancing the usability of the encrypted data. In this paper, a Single Keyword Searchable Encryption (SKSE) scheme is presented which is implemented using Paillier Cryptosystem. The proposed scheme comprises of two variants (a) Secure SKSE (b) Efficient SKSE. The Secure SKSE scheme exploits the probabilistic properties of the Paillier cryptosystem, whereas, the Efficient SKSE scheme is lightweight as it is based on the deterministic properties of the Paillier cryptosystem. Both variants of SKSE scheme are deployed on a public cloud platform “Contabo” and tested on the real-world aviation data to assess the performance of both approaches. The security and performance analysis illustrates that the proposed Secure SKSE maintains a high-security level by providing probabilistic encryption and trapdoors. On the other hand, the Efficient SKSE is 84 times faster in terms of performance.

Privacy-preserving push-sum distributed cubature information filter for nonlinear target tracking with switching directed topologies

To address the data security issue of distributed state estimation, this paper proposes a novel consensus-based cubature information filtering algorithm for sensitive nonlinear target tracking under restricted communication. Through a privacy-preserving approach via state decomposition, the algorithm can protect the privacy of local information from adversaries without sacrificing global estimation accuracy. Based on the push-sum consensus, the distributed approach is further extended to switching directed topologies, which is more feasible for tracking with communication constraints. Our method’s average convergence, privacy preservation, and stability are theoretically proved. Finally, simulations are conducted to evaluate the algorithm’s performance on security, accuracy, and robustness.

Fuzz-classification (p, l)-Angel: An enhanced hybrid artificial intelligence based fuzzy logic for multiple sensitive attributes against privacy breaches

The inability of traditional privacy-preserving models to protect multiple datasets based on sensitive attributes has prompted researchers to propose models such as SLOMS, SLAMSA, (p, k)-Angelization, and (p, l)-Angelization, but these were found to be insufficient in terms of robust privacy and performance. (p, l)-Angelization was successful against different privacy disclosures, but it was not efficient. To the best of our knowledge, no robust privacy model based on fuzzy logic has been proposed to protect the privacy of sensitive attributes with multiple records. In this paper, we suggest an improved version of (p, l)-Angelization based on a hybrid AI approach and privacy-preserving approach like Generalization. Fuzz-classification (p, l)-Angel uses artificial intelligence based fuzzy logic for classification, a high-dimensional segmentation technique for segmenting quasi-identifiers and multiple sensitive attributes. We demonstrate the feasibility of the proposed solution by modelling and analyzing privacy violations using High-Level Petri Nets. The results of the experiment demonstrate that the proposed approach produces better results in terms of efficiency and utility.

An Optimized Approach for Privacy Preserving of Big Data using GDTM and Random Number Generators with GNN

In this paper we propose a Privacy preserving mechanism of big data using GDTM along with Random Number generators. Given the rapid explosion of data being used across Enterprises, Individuals and Sensors, billions of data is being streamed and exchanged across the network. There is a high possibility of sensitive data being exchanged and stored, it's important to preserve sensitive data of Individuals and Enterprise data.Most of the current techniques of privacy preserving in particular in the areas of data perturbation has been done on Static data. Given the dynamic nature of the applications and the huge data that is being generated it's important to evaluate the privacy preserving on big data without losing the accuracy.Our research contribution is on Privacy preserving of big data using Geometric data transformation, random number generator and GNN techniques [5].We would like to extend our research further on improving the accuracy of big data.

High-accuracy low-cost privacy-preserving federated learning in IoT systems via adaptive perturbation

With the rapid development of the Internet of Things (IoT), federated learning (FL) has been widely used to obtain insights from collected data while preserving data privacy. Differential privacy (DP) is an additive noise scheme that has been widely studied as a privacy-preserving approach on FL. However, privacy protection under DP usually comes at the cost of model accuracy for the underlying FL process. In this paper, we propose a novel low-cost (for both communication and computational overhead) adaptive noise perturbation/masking scheme to protect FL clients’ privacy without degrading the global model accuracy. In particular, we set the magnitude of the additive noise to adaptively change with the magnitude of the local model updates. Then, a direction-based filtering scheme is used to accelerate the convergence of the FL model. A maximum tolerable noise bound for local clients is derived using the central limit theorem (CLT). The designed noise maximizes privacy protection for clients while preserving the accuracy and convergence rate of the FL model , as a result of the noise cancelling out and forming a more concentrated distribution after the aggregation operation on the server. We theoretically prove that FL with the proposed noise perturbation scheme retains the same accuracy and convergence rate (O(1/T) for convex loss functions and O(1/T) for non-convex loss functions) as that of non-private FL with SGD. We also evaluate the performance of the proposed scheme in terms of convergence behavior, computational efficiency, and privacy protection against state-of-the-art privacy inference attacks on real-world datasets. Experimental results show that FL with our proposed perturbation scheme outperforms DP in the accuracy and convergence rate of the FL model in both client dropout and non-client dropout scenarios. Compared with DP, our proposed scheme does not incur additional computational and communication overhead. Our approach provides DP-comparable or better effectiveness in defending against privacy attacks under the same global model accuracy.