Privacy-preserving Approach Research Articles

MITIGATING POISONING ATTACKS TO FEDERATED LEARNING IN IOTs ANOMALY DETECTION WITH ATTENTION AGGREGATION

Federated Learning (FL) is a privacy-preserving approach for training deep neural networks across decentralized devices without sharing raw data. Thus, FL has been popularly applied in domains like anomaly detection in Internet of Things (IoTs). However, IoT networks/ devices have limited protection capabilities, resulting in the vulnerability of FL to data poisoning attacks. In order to address this challenge, we propose a new robust FL system designed to counter data poisoning attacks. Our approach, named as Federated Learning with Attention Aggregation (FedAA), leverages AutoEncoder (AE) models for local anomaly detection in IoT networks. In FedAA, we design to aggregate the global model from local models using a novel aggregation method, named as Attention Aggregation (AA). This method is specifically designed to mitigate the impact of data poisoning attacks, which often lead to high values of the loss functions in the local models. More precisely, the local models with high loss values are assigned lower attention weights when contributing to the global model aggregation, and vice versa. As a result, the proposed AA method enhances the robustness of FedAA against data poisoning attacks. We have conducted extensive experiments on three datasets, i.e., N-BaIoT, NSL-KDD, and UNSW, of IoT anomaly detection. The results show that FedAA is more robust than other FL systems in mitigating data poisoning attacks.

Distributed event-triggered-based encrypted control for nonlinear multi-agent systems via privacy-preserving approach

Distributed event-triggered-based encrypted control for nonlinear multi-agent systems via privacy-preserving approach

Decentralized big data mining: federated learning for clustering youth tobacco use in India

This study examines the smoking patterns of youth across various states and union territories of India using the Global Youth Tobacco Survey (GYTS) dataset. The analysis employs three clustering algorithms K-Means, DBSCAN, and Hierarchical Clustering within a federated learning framework, which ensures that sensitive public health data remains decentralized and private. Federated learning enables collaborative analysis across different regions by sharing only model parameters rather than raw data, thus enhancing privacy. Furthermore, the integration of differential privacy ensures additional protection by adding controlled noise to the model parameters, safeguarding individual-level data from exposure during the learning process. The study highlights the varying performances of the clustering algorithms, revealing valuable insights into regional smoking behaviors and the effectiveness of government anti-tobacco campaigns. These insights offer important guidance for public health authorities, allowing for the design and implementation of more targeted and effective campaigns tailored to the needs of specific regions. By leveraging federated learning and differential privacy, this study demonstrates a privacy-preserving approach to analyzing large-scale public health data, providing a blueprint for future health interventions and tobacco control strategies in India and beyond.

An Enhanced Lightweight Secure Authentication and Privacy-Preserving Approach for VANETs

Aims and Background: Vehicular-Adhoc-Networks (VANETs) have gained a lot of attention in the past ten years. Used extensively in intelligent transport systems (ITS), it facilitates the sharing of traffic data between vehicles and their immediate surroundings, resulting in a more pleasant driving experience. When it comes to VANET security, privacy and security are the two biggest obstacles. To improve security in VANET, authentication and privacy-preserving methods are required because any specific disclosure of vehicle specifics, including route data, might have devastating consequences. Objectives & Methodology: In light of this need, this research introduces a novel framework for VANETs called enhanced timed efficient stream loss-tolerant authentication (ETESLTA), which enables a new kind of lightweight authentication while simultaneously protecting users' privacy. Initialisation, registration, mutual authentication, broadcasting and verification, and vehicle revocation are all parts of the suggested model. Furthermore, the ETESLTA method requires very little memory while yet providing excellent broadcast authentication, just like TESLTA. In addition, the ETESLTA method incorporates a cuckoo filter to record the real details of cars inside the RSU's detection range. Results: The suggested approach provides strong anonymity to achieve privacy and resists common assaults, and it features lightweight mutual authentication between the parties. A wide scale of experiments are conducted and the outcomes are evaluated in terms of numerous metrics to ensure the ETESLTA technique performs well. Conclusion: The experimental results demonstrated that the ETESLTA method was superior to the most current state-of-the-art approaches.

Robustness Against Data Integrity Attacks in Decentralized Federated Load Forecasting

This study examines the impact of data integrity attacks on Federated Learning (FL) for load forecasting in smart grid systems, where privacy-sensitive data require robust management. While FL provides a privacy-preserving approach to distributed model training, it remains susceptible to attacks like data poisoning, which can impair model performance. We compare Centralized Federated Learning (CFL) and Decentralized Federated Learning (DFL), using line, ring and bus topologies, under adversarial conditions. Employing a three-layer Artificial Neural Network (ANN) with substation-level datasets (APEhourly,PJMEhourly, and COMEDhourly), we evaluate the system’s resilience in the absence of anomaly detection. Results indicate that DFL significantly outperforms CFL in attack resistance, achieving Mean Absolute Percentage Errors (MAPEs) of 0.48%, 4.29% and 0.702% across datasets, compared to the CFL MAPEs of 6.07%, 18.49% and 10.19%. This demonstrates the potential of DFL as a resilient, secure solution for load forecasting in smart grids, minimizing dependence on anomaly detection to maintain data integrity.

A novel location and time privacy-preserving approach for mobile cloud based on blockchain

A novel location and time privacy-preserving approach for mobile cloud based on blockchain

Federated Learning With Deep Neural Networks: A Privacy-Preserving Approach to Enhanced ECG Classification.

In response to increasing data privacy regulations, this work examines the use of federated learning for deep residual networks to diagnose cardiac abnormalities from electrocardiogram (ECG) data. This approach allows medical institutions to collaborate without exchanging raw patient data. We utilize the publicly available data from the PhysioNet/Computing in Cardiology Challenge 2021, featuring diverse ECG databases, to compare the classification performance of three federated learning methods against both central training with data sharing and isolated training scenarios. We show that federated learning outperforms ECG classifiers trained in isolation. In particular, our findings demonstrate that a globally trained model fine-tuned to specific local datasets surpasses non-collaborative approaches. This shows that models trained in federation learn general features that can be tailored to specific tasks. Furthermore, federated learning almost matches the performance of central training with data sharing on out-of-distribution data from non-participating institutions. These results highlight the ability of federated learning in developing models that generalize well across diverse patient data, without the need to share data among institutions, thus addressing data privacy concerns.

Federated Learning for Telecom Fraud Detection: A Privacy-Preserving Approach to Overcoming Data Fragmentation and Enhancing Security

Federated Learning for Telecom Fraud Detection: A Privacy-Preserving Approach to Overcoming Data Fragmentation and Enhancing Security

Autonomous intrusion detection for IoT: a decentralized and privacy preserving approach

The Internet of Things (IoT) has been increasingly adopted in domains such as smart infrastructure, healthcare, supply chain, transportation, and many others. However, the constrained computational resources of these devices make conventional security approaches against security threats not applicable. This limitation emphasizes the need to explore new approaches, specifically tailored to these kinds of devices. To increase protection against cyberattacks in IoT devices, Intrusion Detection Systems (IDSs) are considered an effective approach. Machine Learning (ML) techniques can be combined with Federated Learning to enhance the privacy and scalability of these systems. Many IDSs have been introduced, but there is a research gap concerning Host Intrusion Detection System (HIDS), which is the primary focus of our current work. Additionally, existing research predominantly focuses on the application of ML techniques and their evaluation, with limited attention to real-world implementation. We propose a lightweight HIDS that relies on the analysis of system call traces to detect malicious activities. The proposed HIDS achieved an accuracy rate of approximately 98%. Finally, by using eXplainable Artificial Intelligence methods, we sought to provide explanations for these these results.

AI security in different industries: A comprehensive review of vulnerabilities and mitigation strategies

Artificial intelligence has quickly and gradually spread across all sectors and fields, achieving exceptional performance in operations and decision-making. Yet, as industries process their tasks with the help of AI systems, new threats appear that require proper solutions. In the following section of this paper, the use of security solutions involving AI in various industries, which are finance, health, manufacturing, and government, as well as a discussion of risks involving artificial intelligence in particular fields and the corresponding measures to be taken against them will be discussed. AI is often applied to detect fraud and credit scoring in the financial sector but has potential risks, such as adversarial attacks and data manipulation. Similarly, the healthcare industry uses AI for predictive healthcare diagnostics and patient information protection; threats are data theft and adversarial examples attacking diagnostic models. In manufacturing, AI individuals contain predictive servicing and excellent control, though they are constantly exposed to industrial espionage and the sabotage of AI models. In retail, however, AI is used to improve marketing and predict consumer behavior. While these benefits exist, questions about using Algorithms that could be biased and privacy infringements continue to raise their ugly head. Defensive and governmental organizations that apply AI for surveillance or automatic operating systems are most vulnerable to adversarial intervention with key control systems and innate security systems. To counter such threats, this paper provides details of industry-specific measures, including adversarial training data sanitization, AI model audit, and privacy-preserving approaches. We illustrate examples linked to every one of these methods to show how they can be deployed in a variety of industries and to guard AI systems. To this end, the following review of the relationship between AI and cybersecurity demonstrates that security measures that rely on artificial intelligence must be regularly checked and updated. This reveals the call for proper, protective, and industry-specific measures to manage the risks and protect AI applications to ensure that the world’s modern, interconnected world is safe for AI.

A parameter privacy-preserving strategy for mixed-autonomy platoon control

It has been demonstrated that leading cruise control (LCC) can improve the operation of mixed-autonomy platoons by allowing connected and automated vehicles (CAVs) to make longitudinal control decisions based on the information provided by surrounding vehicles. However, LCC generally requires surrounding human-driven vehicles (HDVs) to share their real-time states, which can be used by adversaries to infer drivers’ car-following behavior, potentially leading to financial losses or safety concerns. This paper aims to address such privacy concerns and protect the behavioral characteristics of HDVs by devising a parameter privacy-preserving approach for mixed-autonomy platoon control. First, we integrate a parameter privacy filter into LCC to protect sensitive car-following parameters. The privacy filter allows each vehicle to generate seemingly realistic pseudo states by distorting the true parameters to pseudo parameters, which can protect drivers’ privacy in behavioral parameters without significantly influencing the control performance. Second, to enhance the reliability and practicality of the privacy filter within LCC, we first introduce an individual-level parameter privacy preservation constraint to the privacy filter, focusing on the privacy level of each individual parameter pair. Subsequently, we extend the current approach to accommodate continuous parameter spaces through a neural network estimator. Third, analysis of head-to-tail string stability reveals the potential impact of privacy filters in degrading mixed traffic flow performance. Simulation shows that this approach can effectively trade off privacy and control performance in LCC. We further demonstrate the benefit of such an approach in networked systems, i.e., by applying the privacy filter to a preceding vehicle, one can also achieve a certain level of privacy for the following vehicle.

Privacy-Preserving Handover Optimization Using Federated Learning and LSTM Networks.

The rapid evolution of wireless communication systems necessitates advanced handover mechanisms for seamless connectivity and optimal network performance. Traditional algorithms, like 3GPP Event A3, often struggle with fluctuating signal strengths and dynamic user mobility, leading to frequent handovers and suboptimal resource utilization. This study proposes a novel approach combining Federated Learning (FL) and Long Short-Term Memory (LSTM) networks to predict Reference Signal Received Power (RSRP) and the strongest nearby Reference Signal Received Power (RSRP) signals. Our method leverages FL to ensure data privacy and LSTM to capture temporal dependencies in signal data, enhancing prediction accuracy. We develop a dynamic handover algorithm that adapts to real-time conditions, adjusting thresholds based on predicted signal strengths and historical performance. Extensive experiments with real-world data show our dynamic algorithm significantly outperforms the 3GPP Event A3 algorithm, achieving higher prediction accuracy, reducing unnecessary handovers, and improving overall network performance. In conclusion, this study introduces a data-driven, privacy-preserving approach that leverages advanced machine learning techniques, providing a more efficient and reliable handover mechanism for future wireless networks.

Representation Magnitude Has a Liability to Privacy Vulnerability

The privacy-preserving approaches to machine learning (ML) models have made substantial progress in recent years. However, it is still opaque in which circumstances and conditions the model becomes privacy-vulnerable, leading to a challenge for ML models to maintain both performance and privacy. In this paper, we first explore the disparity between member and non-member data in the representation of models under common training frameworks.We identify how the representation magnitude disparity correlates with privacy vulnerability and address how this correlation impacts privacy vulnerability. Based on the observations, we propose Saturn Ring Classifier Module (SRCM), a plug-in model-level solution to mitigate membership privacy leakage. Through a confined yet effective representation space, our approach ameliorates models’ privacy vulnerability while maintaining generalizability. The code of this work can be found here: https://github.com/JEKimLab/AIES2024SRCM

Securing IoT Data: FDUP-RDIC—A Fully Decentralized Approach for Privacy-Preserving and Efficient Data Integrity

Securing IoT Data: FDUP-RDIC—A Fully Decentralized Approach for Privacy-Preserving and Efficient Data Integrity

DPKnob: A visual analysis approach to risk-aware formulation of differential privacy schemes for data query scenarios

Differential privacy is an essential approach for privacy preservation in data queries. However, users face a significant challenge in selecting an appropriate privacy scheme, as they struggle to balance the utility of query results with the preservation of diverse individual privacy. Customizing a privacy scheme becomes even more complex in dealing with queries that involve multiple data attributes. When adversaries attempt to breach privacy firewalls by conducting multiple regular data queries with various attribute values, data owners must arduously discern unpredictable disclosure risks and construct suitable privacy schemes. In this paper, we propose a visual analysis approach for formulating privacy schemes of differential privacy. Our approach supports the identification and simulation of potential privacy attacks in querying statistical results of multi-dimensional databases. We also developed a prototype system, called DPKnob, which integrates multiple coordinated views. DPKnob not only allows users to interactively assess and explore privacy exposure risks by browsing high-risk attacks, but also facilitates an iterative process for formulating and optimizing privacy schemes based on differential privacy. This iterative process allows users to compare different schemes, refine their expectations of privacy and utility, and ultimately establish a well-balanced privacy scheme. The effectiveness of this study is verified by a user study and two case studies with real-world datasets.

Privacy-preserving distributed frequency control for AC microgrids with constrained communication

Distributed control of AC microgrids is one of the most popular methods in the islanded operation mode. Whereas, most of the existing studies either do not consider the potential threat of privacy security or rely on the assumption of ideal communication networks. To this end, this paper presents a novel privacy-preserving distributed secondary frequency control strategy for the privacy protection problem of an islanded AC microgrid with constrained communication. The key contributions of this paper are threefold. (1) Different from the existing privacy-preserving approaches used in AC microgrids, a time-varying function is introduced to mask interactive information such that the frequency cannot be reconstructed by malicious attackers. (2) An event-triggered communication scheme is employed to cope with the constrained communication environment. (3) A privacy-preserving distributed event-triggered control strategy with communication delay is developed such that the frequency restoration and active power sharing of the microgrid are guaranteed. Moreover, the maximum communication delay that the proposed control can withstand is analyzed. Simulation results show the properties of the privacy preservation, the decrease of communication load, and the bounded communication delay allowed in the proposed control strategy.

Secure similar patients query with homomorphically evaluated thresholds

Patient-centric precision medicine requires the analysis of large volumes of genomic data to tailor treatments and medications based on individual-level characteristics. Because the amount of data held by a single institution is limited, researchers may want access to genomic data held by other institutions. Owing to the inherent privacy implications of genomic data, performing comparisons on encrypted data is preferable in certain settings. The Similar patient query (SPQ) is an application that enables a secure search across genomic databases for patients with similar genetic makeup. Query results can be used to draw meaningful conclusions regarding suitable therapies.However, existing protocols either reveal intermediate computations, such as similarity scores, which can lead to membership-inference attacks, or they realize the ideal Boolean output (similar/not similar) through multiple protocol rounds, requiring the database owners to stay online throughout.This paper introduces a two-party privacy-preserving approach to perform SPQs across encrypted genomic databases based on secure function extensions of additively homomorphic encryption. In contrast to related works, our scheme enables secure computation of genomic data similarity without an external party in a single round. This is achieved for more than 1000 positions of a genome in a single public key operation of 256-bit security level in the integer factorization setting.

A privacy-preserving approach for detecting smishing attacks using federated deep learning

A privacy-preserving approach for detecting smishing attacks using federated deep learning

Blockchain-assisted Verifiable Secure Multi-Party Data Computing

Secure multi-party computation (SMPC) is a crucial technology that supports privacy preservation, enabling multiple users to perform computations on any function without disclosing their private inputs and outputs in a distrustful environment. Existing secure multi-party computation models typically rely on obfuscation circuits and cryptographic protocols to facilitate collaborative computation of tasks. However, the efficiency and privacy leakage of users have not been paid much attention during the computation process. To address these problems, this article proposes a privacy-preserving approach Blockchain-assisted Verifiable Secure Multi-Party Data Computing (BVS-MPDC). Specifically, to prevent privacy leakage when users and multiple participants share data, BVS-MPDC uses additive homomorphic encryption to encrypt data shares; and verifies the generated Pedersen commitment of all the data. BVS-MPDC utilizes an improved Schnorr aggregation signature to improve computation efficiency between computing nodes and smart contracts by submitting an aggregation signature to the blockchain. Moreover, we design and implement a smart contract for verifying aggregation signature results on Ethereum. The security proof is presented under the UC framework. Finally, simulation experiments of performance evaluations demonstrate that our scheme outperforms existing schemes in computation overhead and verification.

Fusion of machine learning and blockchain-based privacy-preserving approach for healthcare data in the Internet of Things

Fusion of machine learning and blockchain-based privacy-preserving approach for healthcare data in the Internet of Things