Privacy-sensitive Information Research Articles

Safeguarding User-Centric Privacy in Smart Homes

Internet of Things (IoT) devices have been increasingly deployed in smart homes to automatically monitor and control their environments. Unfortunately, extensive recent research has shown that on-path external adversaries can infer and further fingerprint people’s sensitive private information by analyzing IoT network traffic traces. In addition, most recent approaches that aim to defend against these malicious IoT traffic analytics cannot adequately protect user privacy with reasonable traffic overhead. In particular, these approaches often did not consider practical traffic reshaping limitations, user daily routine permitting, and user privacy protection preference in their design. To address these issues, we design a new low-cost, open source user-centric defense system—PrivacyGuard—that enables people to regain the privacy leakage control of their IoT devices while still permitting sophisticated IoT data analytics that is necessary for smart home automation. In essence, our approach employs intelligent deep convolutional generative adversarial network assisted IoT device traffic signature learning, long short-term memory based artificial traffic signature injection, and partial traffic reshaping to obfuscate private information that can be observed in IoT device traffic traces. We evaluate PrivacyGuard using IoT network traffic traces of 31 IoT devices from five smart homes and buildings. We find that PrivacyGuard can effectively prevent a wide range of state-of-the-art adversarial machine learning and deep learning based user in-home activity inference and fingerprinting attacks and help users achieve the balance between their IoT data utility and privacy preserving.

A privacy-preserving federated learning approach for airline upgrade optimization

PurposeA key issue of making upgrade decisions is to match the most relevant upgrade offers to the right customers at the right time. To optimize upgrade strategies and profitability, companies seek to break “data silos” between themselves and other business partners for a more holistic view of customers' consumption experiences. However, multi-source data fusion may lead to potential privacy leakage. To overcome these two challenges in data silos and privacy protection, this study introduced a privacy-preserving federated learning (FL) approach and explained the process of using FL in modeling airline passengers’ willingness to pay for upgrade offers. Design/methodology/approachFederated learning is a new confidential computing technique that allows companies to train a model cooperatively by exchanging model parameters instead of the actual raw data, which might include customers' privacy sensitive information. Using a case study of a Chinese airline company, this study demonstrated how FL-based upgrade models using multi-source data can be developed to improve the accuracy of predicting customers' willingness to pay for upgrades while preserving customers’ personal data privacy. FindingsComparing with traditional unilateral model using single-source data, the federated logistic regression and SecureBoost models demonstrate better model performance. This indicates that the proposed FL approach can enhance the accuracy of modeling airline passengers' willingness to pay for upgrade offers while preserving passengers’ data privacy. The findings also show that the FL-based models generally took longer running time than the traditional unilateral model due to the design of FL approach in ensuring data privacy. OriginalityThis study contributes to the literature of upgrade optimization by introducing the new FL approach for developing machining learning models to predict customers’ reaction to upgrade offers. Although we focus on the airline industry in our case study, the proposed FL approach can be applied to other industries with a similar issue of upgrade optimization such as hotels or cruise lines, and car rental.

Post-quantum XML and SAML Single Sign-On

Extensible Markup Language (XML) is one of the most popular serialization languages. Since many security protocols are built using XML, it also provides cryptographic functionality. A central framework in this area is the Security Assertion Markup Language (SAML). This standard is one of the most widely used options for implementing Single Sign-On (SSO), which allows users to authenticate to different service providers using the credentials from a single identity provider. Like all other security protocols currently in use, the security and privacy of XML-based frameworks such as SAML is threatened by the development of increasingly powerful quantum computers. In fact, future attackers with access to scalable quantum computers will be able to break the currently used cryptographic building blocks and thus undermine the security of the SAML SSO to illegally access sensitive private information. Post-quantum cryptography algorithms have been developed to protect against such quantum attackers. While many security protocols have been migrated into the quantum age by using post-quantum cryptography, no such solutions for XML and the security protocols based on it have been developed, let alone tested. We make the following contributions to fill this gap. We have designed post-quantum solutions for the cryptographic building blocks in XML and integrated them into the SAML SSO protocol. We implemented our solutions in the OpenSAML, Apache Santuario, and BouncyCastle libraries and extensively tested their performance for various post-quantum instantiations. As a result, we have created a comprehensive and solid foundation for post-quantum XML and post-quantum SAML SSO migration.

Achieving High Accuracy in Android Malware Detection through Genetic Programming Symbolic Classifier

The detection of Android malware is of paramount importance for safeguarding users’ personal and financial data from theft and misuse. It plays a critical role in ensuring the security and privacy of sensitive information on mobile devices, thereby preventing unauthorized access and potential damage. Moreover, effective malware detection is essential for maintaining device performance and reliability by mitigating the risks posed by malicious software. This paper introduces a novel approach to Android malware detection, leveraging a publicly available dataset in conjunction with a Genetic Programming Symbolic Classifier (GPSC). The primary objective is to generate symbolic expressions (SEs) that can accurately identify malware with high precision. To address the challenge of imbalanced class distribution within the dataset, various oversampling techniques are employed. Optimal hyperparameter configurations for GPSC are determined through a random hyperparameter values search (RHVS) method developed in this research. The GPSC model is trained using a 10-fold cross-validation (10FCV) technique, producing a set of 10 SEs for each dataset variation. Subsequently, the most effective SEs are integrated into a threshold-based voting ensemble (TBVE) system, which is then evaluated on the original dataset. The proposed methodology achieves a maximum accuracy of 0.956, thereby demonstrating its effectiveness for Android malware detection.

Robust Estimation Method against Poisoning Attacks for Key-Value Data with Local Differential Privacy

Local differential privacy (LDP) protects user information from potential threats by randomizing data on individual devices before transmission to untrusted collectors. This method enables collectors to derive user statistics by analyzing randomized data, thereby presenting a promising avenue for privacy-preserving data collection. In the context of key–value data, in which discrete and continuous values coexist, PrivKV has been introduced as an LDP protocol to ensure secure collection. However, this framework is susceptible to poisoning attacks. To address this vulnerability, we propose an expectation maximization (EM)-based algorithm combined with a cryptographic protocol to facilitate secure random sampling. Our LDP protocol, known as emPrivKV, exhibits two key advantages: it improves the accuracy of statistical information estimation from randomized data, and enhances resilience against the manipulation of statistics, that is, poisoning attacks. These attacks involve malicious users manipulating the analysis results without detection. This study presents the empirical results of applying the emPrivKV protocol to both synthetic and open datasets, highlighting a notable improvement in the precision of statistical value estimation and robustness against poisoning attacks. As a result, emPrivKV improved the frequency and the mean gains by 17.1% and 25.9%, respectively, compared to PrivKV, with the number of fake users being 0.1 of the genuine users. Our findings contribute to the ongoing discourse on refining LDP protocols for key–value data in scenarios involving privacy-sensitive information.

Physical Side-Channel Attacks against Intermittent Devices

Intermittent (batteryless) devices operate solely using energy harvested from their environment. These devices turn on when they have energy and turn off during energy scarcity. Intermittent devices have recently become increasingly popular in smart buildings, manufacturing plants, and medical implantables as they eliminate the need for battery replacement and enable green computing. Despite their growing adoption in critical applications, the privacy implications of intermittent devices remain largely unexplored. In this paper, we introduce a novel remote side-channel attack. Our observation is that the network packet frequency of an intermittent device can be exploited to learn its turn-on/off patterns. From these patterns, we can infer the energy availability of a device, which reveals privacy-sensitive information about its operating environment, e.g., the presence or absence of individuals. To realize our attack, we develop a three-stage hierarchical inference framework that leverages the timestamped network packet sequence of intermittent devices. Our framework automatically extracts a set of temporal features from inter-packet-arrival timings. It then employs a series of models to uncover (1) whether a target intermittent device is present in the environment, (2) its energy harvester type (e.g., vibration or water flow), and (3) its energy availability conditions (e.g., high-vibration or no-vibration). To validate our attack effectiveness, we conduct experiments in two environments: a smart home and a miniature manufacturing plant equipped with three intermittent devices powered by solar energy, vibration, and temperature. By analyzing their energy availability patterns, we are able to infer user activities and presence in the smart home and the robot’s movement patterns in the manufacturing plant with an average accuracy of 85%. This sensitive information enables an adversary to launch domain-specific attacks, such as burglarizing a smart home when the user is asleep or timely tampering with plant sensors to cause maximum damage.

A Deep Learning-Based Method for Preventing Data Leakage in Electric Power Industrial Internet of Things Business Data Interactions.

In the development of the Power Industry Internet of Things, the security of data interaction has always been an important challenge. In the power-based blockchain Industrial Internet of Things, node data interaction involves a large amount of sensitive data. In the current anti-leakage strategy for power business data interaction, regular expressions are used to identify sensitive data for matching. This approach is only suitable for simple structured data. For the processing of unstructured data, there is a lack of practical matching strategies. Therefore, this paper proposes a deep learning-based anti-leakage method for power business data interaction, aiming to ensure the security of power business data interaction between the State Grid business platform and third-party platforms. This method combines named entity recognition technologies and comprehensively uses regular expressions and the DeBERTa (Decoding-enhanced BERT with disentangled attention)-BiLSTM (Bidirectional Long Short-Term Memory)-CRF (Conditional Random Field) model. This method is based on the DeBERTa (Decoding-enhanced BERT with disentangled attention) model for pre-training feature extraction. It extracts sequence context semantic features through the BiLSTM, and finally obtains the global optimal through the CRF layer tag sequence. Sensitive data matching is performed on interactive structured and unstructured data to identify privacy-sensitive information in the power business. The experimental results show that the F1 score of the proposed method in this paper for identifying sensitive data entities using the CLUENER 2020 dataset reaches 81.26%, which can effectively prevent the risk of power business data leakage and provide innovative solutions for the power industry to ensure data security.

Horizontal Federated Recommender System: A Survey

Due to underlying privacy-sensitive information in user-item interaction data, the risk of privacy leakage exists in the centralized-training recommender system (RecSys). To this issue, federated learning, a privacy-oriented distributed computing paradigm, is introduced and promotes the crossing field “Federated Recommender System (FedRec).” Regarding data distribution characteristics, there are horizontal, vertical, and transfer variants, where horizontal FedRec (HFedRec) occupies a dominant position. User devices can personally participate in the horizontal federated architecture, making user-level privacy feasible. Therefore, we target the horizontal point and summarize existing works more elaborately than existing FedRec surveys. First, from the model perspective, we group them into different learning paradigms (e.g., deep learning and meta learning). Second, from the privacy perspective, privacy-preserving techniques are systematically organized (e.g., homomorphic encryption and differential privacy). Third, from the federated perspective, fundamental issues (e.g., communication and fairness) are discussed. Fourth, each perspective has detailed subcategories, and we specifically state their unique challenges with the observation of current progress. Finally, we figure out potential issues and promising directions for future research.

PP-CSA: Practical Privacy-Preserving Software Call Stack Analysis

Software call stack is a sequence of function calls that are executed during the runtime of a software program. Software call stack analysis (CSA) is widely used in software engineering to analyze the runtime behavior of software, which can be used to optimize the software performance, identify bugs, and profile the software. Despite the benefits of CSA, it has recently come under scrutiny due to concerns about privacy. To date, software is often deployed at user-side devices like mobile phones and smart watches. The collected call stacks may thus contain privacy-sensitive information, such as healthy information or locations, depending on the software functionality. Leaking such information to third parties may cause serious privacy concerns such as discrimination and targeted advertisement. This paper presents PP-CSA, a practical and privacy-preserving CSA framework that can be deployed in real-world scenarios. Our framework leverages local differential privacy (LDP) as a principled privacy guarantee, to mutate the collected call stacks and protect the privacy of individual users. Furthermore, we propose several key design principles and optimizations in the technical pipeline of PP-CSA, including an encoder-decoder scheme to properly enforce LDP over software call stacks, and several client/server-side optimizations to largely improve the efficiency of PP-CSA. Our evaluation over real-world Java and Android programs shows that our privacy-preserving CSA pipeline can achieve high utility and privacy guarantees while maintaining high efficiency. We have released our implementation of PP-CSA as an open-source project at https://github.com/wangzhaoyu07/PP-CSA for results reproducibility. We will provide more detailed documents to support and the usage and extension of the community.

The Noise Blowing-Up Strategy Creates High Quality High Resolution Adversarial Images against Convolutional Neural Networks

Convolutional neural networks (CNNs) serve as powerful tools in computer vision tasks with extensive applications in daily life. However, they are susceptible to adversarial attacks. Still, attacks can be positive for at least two reasons. Firstly, revealing CNNs vulnerabilities prompts efforts to enhance their robustness. Secondly, adversarial images can also be employed to preserve privacy-sensitive information from CNN-based threat models aiming to extract such data from images. For such applications, the construction of high-resolution adversarial images is mandatory in practice. This paper firstly quantifies the speed, adversity, and visual quality challenges involved in the effective construction of high-resolution adversarial images, secondly provides the operational design of a new strategy, called here the noise blowing-up strategy, working for any attack, any scenario, any CNN, any clean image, thirdly validates the strategy via an extensive series of experiments. We performed experiments with 100 high-resolution clean images, exposing them to seven different attacks against 10 CNNs. Our method achieved an overall average success rate of 75% in the targeted scenario and 64% in the untargeted scenario. We revisited the failed cases: a slight modification of our method led to success rates larger than 98.9%. As of today, the noise blowing-up strategy is the first generic approach that successfully solves all three speed, adversity, and visual quality challenges, and therefore effectively constructs high-resolution adversarial images with high-quality requirements.

Exploring One-Shot Semi-supervised Federated Learning with Pre-trained Diffusion Models

Recently, semi-supervised federated learning (semi-FL) has been proposed to handle the commonly seen real-world scenarios with labeled data on the server and unlabeled data on the clients. However, existing methods face several challenges such as communication costs, data heterogeneity, and training pressure on client devices. To address these challenges, we introduce the powerful diffusion models (DM) into semi-FL and propose FedDISC, a Federated Diffusion-Inspired Semi-supervised Co-training method. Specifically, we first extract prototypes of the labeled server data and use these prototypes to predict pseudo-labels of the client data. For each category, we compute the cluster centroids and domain-specific representations to signify the semantic and stylistic information of their distributions. After adding noise, these representations are sent back to the server, which uses the pre-trained DM to generate synthetic datasets complying with the client distributions and train a global model on it. With the assistance of vast knowledge within DM, the synthetic datasets have comparable quality and diversity to the client images, subsequently enabling the training of global models that achieve performance equivalent to or even surpassing the ceiling of supervised centralized training. FedDISC works within one communication round, does not require any local training, and involves very minimal information uploading, greatly enhancing its practicality. Extensive experiments on three large-scale datasets demonstrate that FedDISC effectively addresses the semi-FL problem on non-IID clients and outperforms the compared SOTA methods. Sufficient visualization experiments also illustrate that the synthetic dataset generated by FedDISC exhibits comparable diversity and quality to the original client dataset, with a neglectable possibility of leaking privacy-sensitive information of the clients.

A multi-level privacy-preserving scheme for extracting traffic images

Traffic images are constantly used as a stick to assess traffic conditions. Traffic flow statistical analysis, road condition safety monitoring, vehicle violation detection, accident surveillance, and the driving environment perception of autonomous vehicles are all functionalities that depend on the processing of traffic images. However, traffic images contain privacy-sensitive information related to users, such as license plate numbers, drivers and passengers. Extracting, analyzing and sharing such privacy-sensitive information without any security measures may raise concerns among users about potential privacy violations. This paper proposes a multi-level privacy protection scheme for traffic image extraction based on compressive sensing, which has the advantages of data undersampling, privacy protection and data access control. Detailed compression rate analysis demonstrates that the proposed solution can effectively reduce the transmission load of image information. Security analysis demonstrates that the proposed approach achieves multi-level reconstruction qualities and high-security strength for users with different authorities. Thus, it has a good application prospect in an image-based intelligent traffic management system.

On Mitigating the Utility-Loss in Differentially Private Learning: A New Perspective by a Geometrically Inspired Kernel Approach

Privacy-utility tradeoff remains as one of the fundamental issues of differentially private machine learning. This paper introduces a geometrically inspired kernel-based approach to mitigate the accuracy-loss issue in classification. In this approach, a representation of the affine hull of given data points is learned in Reproducing Kernel Hilbert Spaces (RKHS). This leads to a novel distance measure that hides privacy-sensitive information about individual data points and improves the privacy-utility tradeoff via significantly reducing the risk of membership inference attacks. The effectiveness of the approach is demonstrated through experiments on MNIST dataset, Freiburg groceries dataset, and a real biomedical dataset. It is verified that the approach remains computationally practical. The application of the approach to federated learning is considered and it is observed that the accuracy-loss due to data being distributed is either marginal or not significantly high.

Do you want to know a secret? Strategic alliances and competition in product markets

Strategic alliances make firms’ boundaries permeable to information leakages that may benefit the competitive position of partnering firms. In this paper, we examine the implications of information leakage on the incentives to join a strategic alliance and the nature of competition in the product market. We show that information leakage can trigger opportunistic behaviours in which firms engage in unprofitable alliances simply because the possibility of learning sensitive information about their competitors increases the expected private rents that firms earn when competing in the product market. Thus, our findings uncover a purely informational mechanism through which information leakage affects the incentives to join a strategic alliance that does not rely on the firm’s ability to absorb spillovers from other firms. We also show that the incentives to devise alliances to gain access to the partner’s sensitive private information may remain even if the negatively affected firm can pursue compensation for the damage that this deceptive business practice may cause.

Securing Secrets in Cyber-Physical Systems: A Cutting-Edge Privacy Approach with Consortium Blockchain.

In the era of interconnected and intelligent cyber-physical systems, preserving privacy has become a paramount concern. This paper aims a groundbreaking proof-of-concept (PoC) design that leverages consortium blockchain technology to address privacy challenges in cyber-physical systems (CPSs). The proposed design introduces a novel approach to safeguarding sensitive information and ensuring data integrity while maintaining a high level of trust among stakeholders. By harnessing the power of consortium blockchain, the design establishes a decentralized and tamper-resistant framework for privacy preservation. However, ensuring the security and privacy of sensitive information within CPSs poses significant challenges. This paper proposes a cutting-edge privacy approach that leverages consortium blockchain technology to secure secrets in CPSs. Consortium blockchain, with its permissioned nature, provides a trusted framework for governing the network and validating transactions. By employing consortium blockchain, secrets in CPSs can be securely stored, shared, and accessed by authorized entities only, mitigating the risks of unauthorized access and data breaches. The proposed approach offers enhanced security, privacy preservation, increased trust and accountability, as well as interoperability and scalability. This paper aims to address the limitations of traditional security mechanisms in CPSs and harness the potential of consortium blockchain to revolutionize the management of secrets, contributing to the advancement of CPS security and privacy. The effectiveness of the design is demonstrated through extensive simulations and performance evaluations. The results indicate that the proposed approach offers significant advancements in privacy protection, paving the way for secure and trustworthy cyber-physical systems in various domains.

Security and Privacy in Vehicular Digital Twin Networks: Challenges and Solutions

With the flourishing and advancement of autonomous vehicles, the vehicular digital twin (VDT) has become an emerging paradigm, consisting of inter-twin communication and intra-twin communication. VDT can effectively address the current shortcomings of the autonomous vehicle and provide enhanced and scalable services to the users. Nevertheless, strict security and privacy concerns arise, as VDT collects a wide range of privacy-sensitive information of the users in real-time and is exposed to an open network environment in both physical and virtual horizons. In this article, we investigate the security and privacy of VDT. We first describe the vehicular digital twin network architecture in detail and then investigate some VDT-related applications. After that, we discuss the security and privacy issues in VDT. Finally, we put forward several potential countermeasures and open research issues for VDT from the perspective of security and privacy. We expect this work will bring more attention to further investigation on security and privacy solutions for vehicular digital twin networks.

Evaluation of Open-Source Tools for Differential Privacy

Differential privacy (DP) defines privacy protection by promising quantified indistinguishability between individuals who consent to share their privacy-sensitive information and those who do not. DP aims to deliver this promise by including well-crafted elements of random noise in the published data, and thus there is an inherent tradeoff between the degree of privacy protection and the ability to utilize the protected data. Currently, several open-source tools have been proposed for DP provision. To the best of our knowledge, there is no comprehensive study for comparing these open-source tools with respect to their ability to balance DP’s inherent tradeoff as well as the use of system resources. This work proposes an open-source evaluation framework for privacy protection solutions and offers evaluation for OpenDP Smartnoise, Google DP, PyTorch Opacus, Tensorflow Privacy, and Diffprivlib. In addition to studying their ability to balance the above tradeoff, we consider discrete and continuous attributes by quantifying their performance under different data sizes. Our results reveal several patterns that developers should have in mind when selecting tools under different application needs and criteria. This evaluation survey can be the basis for an improved selection of open-source DP tools and quicker adaptation of DP.

Preserving privacy in surgical video analysis using a deep learning classifier to identify out-of-body scenes in endoscopic videos

Surgical video analysis facilitates education and research. However, video recordings of endoscopic surgeries can contain privacy-sensitive information, especially if the endoscopic camera is moved out of the body of patients and out-of-body scenes are recorded. Therefore, identification of out-of-body scenes in endoscopic videos is of major importance to preserve the privacy of patients and operating room staff. This study developed and validated a deep learning model for the identification of out-of-body images in endoscopic videos. The model was trained and evaluated on an internal dataset of 12 different types of laparoscopic and robotic surgeries and was externally validated on two independent multicentric test datasets of laparoscopic gastric bypass and cholecystectomy surgeries. Model performance was evaluated compared to human ground truth annotations measuring the receiver operating characteristic area under the curve (ROC AUC). The internal dataset consisting of 356,267 images from 48 videos and the two multicentric test datasets consisting of 54,385 and 58,349 images from 10 and 20 videos, respectively, were annotated. The model identified out-of-body images with 99.97% ROC AUC on the internal test dataset. Mean ± standard deviation ROC AUC on the multicentric gastric bypass dataset was 99.94 ± 0.07% and 99.71 ± 0.40% on the multicentric cholecystectomy dataset, respectively. The model can reliably identify out-of-body images in endoscopic videos and is publicly shared. This facilitates privacy preservation in surgical video analysis.

A novel GPU based Geo-Location Inference Attack on WebGL framework

In the past few years, graphics processing units (GPUs) have become an indispensable part of modern computer systems, not only for graphics rendering but also for intensive parallel computing. Given that many tasks running on GPUs contain sensitive information, security concerns have been raised, especially about potential GPU information leakage. Previous works have shown such concerns by showing that attackers can use GPU memory allocations or performance counters to measure victim side effects. However, such an attack has a critical drawback that it requires a victim to install desktop applications or mobile apps yielding it uneasy to be deployed in the real world. In this paper, we solve this drawback by proposing a novel GPU-based side-channel Geo-Privacy inference attack on the WebGL framework, namely, GLINT (stands for Geo-Location Inference Attack). GLINT merely utilizes a lightweight browser extension to measure the time elapsed to render a sequence of frames on well-known map websites, e.g., Google Maps, or Baidu Maps. The measured stream of time series is then employed to infer geologically privacy-sensitive information, such as a search on a specific location. Upon retrieving the stream, we propose a novel online segmentation algorithm for streaming data to determine the start and end points of privacy-sensitive time series. We then combine the DTW algorithm and KNN algorithm on these series to conclude the final inference on a user’s geo-location privacy.We conducted real-world experiments to testify our attack. The experiments show that GeoInfer can correctly infer more than 83% of user searches regardless of the locations and map websites, meaning that our Geo-Privacy inference attack is accurate, practical, and robust. To counter this attack, we implemented a defense strategy based on Differential Privacy to hinder obtaining accurate rendering data. We found that this defense mechanism managed to reduce the average accuracy of the attack model by more than 70%, indicating that the attack was no longer effective. We have fully implemented GLINT and open-sourced it for future follow-up research.

Personal information disclosure under competition for benefits: Is sharing caring?

Personal information is shared extensively every day, particularly when competing for others' attention on online platforms. In this paper, we experimentally investigate the interaction of peer comparison and incentives as drivers to disclose potentially privacy-sensitive information. We find that information sharing is higher under incentives, and further increases under peer comparison. This effect is driven by those initially disclosing less, who additionally report feeling more compelled to reveal information. Our results shed light on additional drivers for the current information-sharing trend, while pointing to neglected social pressure to disclose personal information in competitive environments.