Leakage Of Sensitive Information Research Articles

Anonymous federated learning framework in the internet of things

AbstractWith the continuous development of the internet of things (IoT), federated learning is being widely applied. This technology keeps data locally to protect data security. However, during the process of uploading gradients from local clients, there is a possibility of leaking sensitive information such as identities. To address this issue, this paper presents a secure and efficient anonymous federated learning framework. In our proposed, we first design a lightweight key‐sharing protocol based on elliptic curve cryptography (ECC) to generate shared keys and ensure secure communication. We then describe the improved process of federated learning, where clients communicate with the cloud server using pseudonyms to achieve anonymity. The security of our protocol is analyzed from both formal and informal perspectives, demonstrating that our proposed protocol satisfies “session key security.” We also employ the formal verification tool ProVerif to validate the security of the protocol in terms of mutual authentication and key configuration in the Dolve–Yao threat model. Finally, the computational and communication costs of the proposed anonymous federated learning framework are evaluated, showing that both the computational and communication expenses are relatively low. A comparison was made between the proposed federated learning process and three other federated learning processes, demonstrating the clear advantages of our proposed.

Blockchain assisted encryption scheme for intellectual share estimation using medical research data

SummaryRecently, blockchain utilization with advanced encryption and decryption has gained improved attention over diverse applications. Especially in the medical field, big data storage requires an effective medium for managing user data with high confidentiality. The existing research presents diverse encryption algorithms, but certain drawbacks include sensitive information leakage, ineffective data integrity verification, and increased computational cost and time. The proposed work develops an effective blockchain mechanism for secured data storage to conquer these limitations. This article secures the medical research data from different team members for intellectual share estimation. Initially, the registration process involves user details such as name, contact number, age and so on. Next, a digital signature is generated for every registered user using the SHA‐256 hash algorithm. During data uploading, the data is encrypted using elliptic curve Diffie Hellman encryption scheme to promote security. The audio and video files are stored in inter planetary file system, and texts are stored in the blockchain. The users can retrieve the stored data from the blockchain using the encryption key. The identities and associated public keys are verified using a smart contract to validate the user certificate. Next, a decision form is made to show the polling mechanism using item preference ranking for other users viewing the data. The proposed mechanism ensures intellectual shares for all the users associated with a specified project. The performances are analyzed using PYTHON, and the encryption time obtained in the proposed model is 0.85 s at 140 security bit level.

Federated Learning for Clients’ Data Privacy Assurance in Food Service Industry

The food service industry must ensure that service facilities are free of foodborne pathogens hosted by organic residues and biofilms. Foodborne diseases put customers at risk and compromise the reputations of service providers. Fluorescence imaging, empowered by state-of-the-art artificial intelligence (AI) algorithms, can detect invisible residues. However, using AI requires large datasets that are most effective when collected from actual users, raising concerns about data privacy and possible leakage of sensitive information. In this study, we employed a decentralized privacy-preserving technology to address client data privacy issues. When federated learning (FL) is used, there is no need for data sharing across clients or data centralization on a server. We used FL and a new fluorescence imaging technology and applied two deep learning models, MobileNetv3 and DeepLabv3+, to identify and segment invisible residues on food preparation equipment and surfaces. We used FedML as our FL framework and Fedavg as the aggregation algorithm. The model achieved training and testing accuracies of 95.83% and 94.94% for classification between clean and contamination frames, respectively, and resulted in intersection over union (IoU) scores of 91.23% and 89.45% for training and testing, respectively, of segmentation of the contaminated areas. The results demonstrated that using federated learning combined with fluorescence imaging and deep learning algorithms can improve the performance of cleanliness auditing systems while assuring client data privacy.

Microservice‐driven privacy‐aware cross‐platform social relationship prediction based on sequential information

AbstractCurrently, the accurate prediction of social relationships can effectively reduce the decision‐making burden of users in various service platforms. However, in the big data environment, the users' data information used for the relationship prediction is highly fragmented distribution, so it is a non‐trivial challenge to integrate the users' sequence data information from different platforms while preventing sensitive information leakage. To this end, based on the microservice environment, we devise a cross‐platform social relationship prediction approach (CPSRP) to address the above problems. Briefly, the improved Simhash method aggregates similar users into the common bucket. Then the embedding technique converts the users' sparse data information into the low‐dimensional dense continuous feature vectors; the redefined Gated Recurrent Unit (r‐GRU) network and the Multilayer Perceptron (MLP) network are employed to extract the overall temporal sequence features of users. The relationship prediction is finally executed according to the users' sequential features. Extensive experiments are conducted on Epinions, and the experimental results further prove the benefits of our proposal in terms of relationship prediction while protecting users' sensitive information.

Multimodal Fusion-Based Image Hiding Algorithm for Secure Healthcare System

The development of artificial intelligence (AI) plays a significant role of multimedia applications, especially in the healthcare domain. However, it has brought about the problem of sensitive information leakage. To address these challenges, an interesting multimodal fusion-based robust image hiding algorithm is proposed in this paper. Firstly, fused image is considered as mark image generated from MRI and CT images using non-subsampled shearlet transform (NSST). Secondly, we employed principal component analysis (PCA) to compute the appropriate coefficients of cover image for embedding purpose. Thirdly, fused mark image is Arnold cat map encoded to address the security issue hidden mark media. Finally, the fusion of fractional dual-tree complex wavelet transform (Fr-DTCWT) and randomized singular value decomposition (RSVD) is utilized to conceal encrypted fused mark media inside host image. Our findings show that the proposed algorithm outperforms some of the recent techniques in terms of high robustness and invisibility.

Assessing the Potential of Using Artificial Intellect for Governing the Organization (illustrated by the Transportation and Logistics Holding Company)

The article provides results of the experiment in the form of an interview with chat-bot based on artificial intellect (AI) ChatGPT dealt with governing the transportation and logistics holding company. The research showed that AI systems had reached such a level of development, which allows us to use them for solving problems of organizational and strategic governance of the company. The system can give adequate answers to questions connected with various aspects of management, formulate suggestions and identify promising trends for strategic development. Chat-bot that deals with preliminary preparation of the material can be used to resolve analytical problems in order to save time of profile specialists. However, use of such systems can cause risks of sensitive information leakage and making incorrect decisions on the basis of wrong data. In spite of high assessment of such system potential, transportation and logistics companies and other organizations should use relevant technology with care.

Federated learning‐based private medical knowledge graph for epidemic surveillance in internet of things

AbstractWith the explosive development of the Internet of Things (IoT), it is convenient and important to collect health data from medical sensors and smart devices and construct medical knowledge graph. The knowledge graph contributes to investigating the connection between patient and disease, especially for epidemic surveillance. However, it is possible to cause the leakage of sensitive health information due to the untrusted data collector or various malicious attackers. In this paper, we attempt to utilise federated learning to construct a special knowledge graph, that is, individual‐symptom relationship diagram with local differential privacy (LDP‐ISRD), for epidemic risk surveillance, which presents the underlying infectious relationship among individuals. At first, we propose a federated learning‐based framework of LDP‐ISRD by utilising individuals' smart devices in IoT. Then, we leverage locations to determine the connection among individuals in terms of physical contact. Next, we propose a randomised algorithm PrivISRD to implement federated learning‐based LDP‐ISRD, which consists of symptom perturbation and aggregation. Finally, extensive experiments evaluate the impact of various parameters and results demonstrate that LDP‐ISRD has good performance.

Locally private estimation of conditional probability distribution for random forest in multimedia applications

The application of artificial intelligence models to raw multimedia data is susceptible to various data inference attacks, posing a significant risk in terms of sensitive input information leakage. Most of the existing studies on privacy-preserving multimedia applications based on artificial intelligence, focus on a single intelligent model and thus have various limitations. In this paper, we attempt to directly perturb the core component of many multimedia intelligent models in Bayesian networks and deep learning. That is, we apply conditional probability distribution estimation to guarantee the privacy of the models. At first, we present the formal problem formulation of private conditional probability distribution estimation and apply it to random forest for task classification in multimedia applications. Then, we design a simple perturbation approach called NaivePrivDistEst, to add noise to all the elements of probability estimation in random forest. Next, we present an improved approach called FastLRG, that utilizes the taxonomy tree to discretize the continuous attributes, thereby combining the attribute features to improve the prediction accuracy of random forest. Finally, we perform extensive experiments to evaluate the performance of random forest based on the proposed estimation algorithms. The experimental results indicate that the proposed models have better performance compared with existing private decision trees.

DP-Authentication: A novel deep learning based drone pilot authentication scheme

Unmanned Aerial Vehicles (UAVs), also known as drones, have recently been proposed as flying base stations for providing reliable service to IoT devices. However, due to the lack of effective authentication schemes, UAVs are often hijacked by adversaries, which raises a high potential for sensitive information leakage. Therefore, designing a real-time authentication scheme is essential to enhance UAV safety. Up to the present, several works exist about pilot authentication by classifying radio-control signals. As propagating through the open environment, radio-control signals can be sniffed, analyzed, and simulated, posing significant threats to UAV security. For this reason, we propose a novel deep learning-based drone pilot authentication scheme, DP-Authentication, to protect UAVs from malicious radio-manipulated attacks. Specifically, we collect UAV flight data from the onboard PX4 flight stack and feed them into the authentication scheme to validate pilot legal status dynamically. As verified by comprehensive experiments, the proposed authentication scheme can authenticate pilots with an accuracy of 95.24% and detect malicious hijacking with an accuracy of 96.82%. Thanks to the low system overhead, it holds great promise for deployment on the UAV side to monitor pilot legal status in real-time.

Privacy-Preserving Dynamic Average Consensus via Random Number Perturbation

This brief focuses on the study of privacy preservation of dynamic average consensus (DAC) in multi-agent networks. A privacy-preserving DAC (PP-DAC) algorithm is proposed based on a carefully designed random number perturbation mechanism. The PP-DAC algorithm is able to protect agents from the leakage of sensitive information without compromising their tracking accuracy. Furthermore, the privacy analysis for different scenarios is given to show that the PP-DAC algorithm works well unless all neighbors of the target agent collude with each other to attack this agent. Also, some numerical simulations are given to illustrate the validity of the proposed algorithm.

Utility Analysis about Log Data Anomaly Detection Based on Federated Learning

Logs that record system information are managed in anomaly detection, and more efficient anomaly detection methods have been proposed due to their increase in complexity and scale. Accordingly, deep learning models that automatically detect system anomalies through log data learning have been proposed. However, in existing log anomaly detection models, user logs are collected from the central server system, exposing the data collection process to the risk of leaking sensitive information. A distributed learning method, federated learning, is a trend proposed for artificial intelligence learning regarding sensitive information because it guarantees the anonymity of the collected user data and collects only weights learned from each local server in the central server. In this paper, we executed an experiment regarding system log anomaly detection using federated learning. The results demonstrate the feasibility of applying federated learning in deep-learning-based system-log anomaly detection compared to the existing centralized learning method. Moreover, we present an efficient deep-learning model based on federated learning for system log anomaly detection.

Enhancing the Performance of SQL Injection Attack Detection through Probabilistic Neural Networks

SQL injection attack is considered one of the most dangerous vulnerabilities exploited to leak sensitive information, gain unauthorized access, and cause financial loss to individuals and organizations. Conventional defense approaches use static and heuristic methods to detect previously known SQL injection attacks. Existing research uses machine learning techniques that have the capability of detecting previously unknown and novel attack types. Taking advantage of deep learning to improve detection accuracy, we propose using a probabilistic neural network (PNN) to detect SQL injection attacks. To achieve the best value in selecting a smoothing parament, we employed the BAT algorithm, a metaheuristic algorithm for optimization. In this study, a dataset consisting of 6000 SQL injections and 3500 normal queries was used. Features were extracted based on tokenizing and a regular expression and were selected using Chi-Square testing. The features used in this study were collected from the network traffic and SQL queries. The experiment results show that our proposed PNN achieved an accuracy of 99.19% with a precision of 0.995%, a recall of 0.981%, and an F-Measure of 0.928% when employing a 10-fold cross-validation compared to other classifiers in different scenarios.

From Attack to Identification: MEMS Sensor Fingerprinting Using Acoustic Signals

A device pretending to be behaving normally can carry out malicious acts that result in the leakage of sensitive information, invasion of privacy, damage property, or even loss of life. The first step to detect these malicious acts is to identify whether a device is authorized or not. There are existing methods for identifying devices based on cryptographic schemes or physical unclonable function (PUF), but each of these methods has disadvantages in that key management costs are incurred or additional chips or circuits are required. In this article, we propose a device identification method that overcomes these shortcomings. Based on the fact that MEMS sensors are built-in to most devices due to their various applications and the property that MEMS sensors are sensitive to acoustic signals with resonant frequencies, the proposed method uses MEMS sensor readings to acoustic signals. Many attack methods have been studied using this property to disable the normal function of the sensor or eavesdrop on speech around a device. On the other hand, our method is the first to fingerprint MEMS sensors using raw sensor readings to acoustic signals from a security point of view rather than from this conventional attack point of view. Furthermore, our method is based on a challenge-response structure to be secure from replay attacks; here, a challenge is a random acoustic signal, and a response is the unique MEMS sensor reading to the challenge. In our evaluation, we use commercial MEMS sensors and a low-cost speaker that costs less than <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\$ $ </tex-math></inline-formula> 1. The results show that under various conditions that may affect MEMS sensor readings, the macro F1-scores when identifying authorized devices is 1.0, and the macro accuracy of detecting attacks using devices other than authorized ones is 0.994. As a result, our method can identify devices well at a low cost.

Bounding information leakage in machine learning

Recently, it has been shown that Machine Learning models can leak sensitive information about their training data. This information leakage is exposed through membership and attribute inference attacks. Although many attack strategies have been proposed, little effort has been made to formalize these problems. We present a novel formalism, generalizing membership and attribute inference attack setups previously studied in the literature and connecting them to memorization and generalization. First, we derive a universal bound on the success rate of inference attacks and connect it to the generalization gap of the target model. Second, we study the question of how much sensitive information is stored by the algorithm about its training set and we derive bounds on the mutual information between the sensitive attributes and model parameters. Experimentally, we illustrate the potential of our approach by applying it to both synthetic data and classification tasks on natural images. Finally, we apply our formalism to different attribute inference strategies, with which an adversary is able to recover the identity of writers in the PenDigits dataset.

Learning Malicious Circuits in FPGA Bitstreams

Computing platforms are integrating field-programmable gate arrays (FPGAs) to support domain-specific customization. Multiple tenants can share these FPGAs by configuring them at runtime. However, attackers can abuse this capability by programming the FPGAs with malicious functions. A malicious configuration bitstream can launch denial of service, overheat the FPGA, leak sensitive information via side channels, enable remote monitoring, and launch voltage and timing attacks. We consider time-based multitenancy, where multiple tenants use the FPGA at different time intervals and not at the same time. We propose a defense based on machine learning (ML) algorithms to detect bitstreams of malicious circuits and malicious circuits mixed with legitimate circuits by analyzing the static features extracted from FPGA bitstreams. The proposed approach can help detect malicious bitstreams without the need for reverse engineering of the bitstream or having access to the design netlist. Our results on Xilinx FPGAs indicate that supervised classifiers may identify malicious bitstreams representing ring-oscillator circuits with a true-positive rate (TPR) of 100% and a false-positive rate (FPR) of only 4%. In addition, for the extremely difficult problem of detecting malicious bitstreams embedded in legitimate bitstreams, a pipeline of a random forest and a support vector machine classifiers trained on subarrays of bitstreams can help detect bitstreams of malicious circuits embedded in legitimate designs with TPR of 95.5% and FPR of 30.4%.

Destruction of stego images formed by adaptive embedding methods with dictionary learning methods

Counteraction to sensitive information leakage that processed by state and private organizations is topical task today. Of special interest are methods for prevention data leakage by usage of hidden (steganographic) communication channels by attackers. Despite wide range of proposed steganalysis methods for detection of embedded messages, theirs performance highly depends on prior information about used embedding methods. As an example, we may mention modern stegdetectors for digital images, which are based on cover rich models and deep convolutional neural networks. Therefore, the stego image destruction methods are widely applied as preventive action.Modern methods for stego image destruction are based on widespread image denoising methods, like median filter and lossy compression. The limitation of such methods is significant changes of image's statistical features that may disclosure the steganalysis process to attacker. Therefore, development of stego images processing methods that provide reliable destruction of embedded data, and preserving cover image statistical features is needed. The paper is aimed at performance evaluation of applying the novel methods of spectral analysis, namely dictionary learning, for solving this tasks. The obtained results showed limitation of state-of-the-art methods for destruction of stego image formed by adaptive embedding methods, namely considerable changes of image's statistical parameters. The proposed method allows preserving both minimal changes of a \ac{ci} parameters, and ratio of survived bits of embedded message (less than $7\%$). This makes proposed solution an attractive candidate for reliable destruction of stego images formed by novel embedding methods. However, practical usage of proposed solution requires further improvement of dictionary learning methods, namely decreasing of computation complexity of dictionary forming procedure.

Characterization and Detection of Cross-Router Covert Channels

In covert channel attacks, an adversary seeks various means to influence a tangible characteristic of a system, and then makes the systems leak information by measuring this characteristic. Covert channels are, by nature, very elusive. This makes it very difficult to identify them and defend against attacks that use these channels to leak sensitive information. Thus, they are a serious threat to the security of many systems.In this paper, we present two network timing covert channel attacks, and a defense mechanism against them. The purpose of the proposed attacks is to leak sensitive information between two logically separated (or isolated) networks that are hosted by a single router – one that is connected to the Internet, and another that is isolated and contains sensitive information. The attacks build on the fact that the response time of the router for a specific type of packet sent from a device that is connected to it is usually predictable, given the network topology. By interacting with the single shared router in a specific manner, an attacker can increase the router’s packet response time. The interaction is determined based on the information to be leaked, so the receiver (a computer located on the Internet-connected network) can measure the delayed packet response times and decode the sender’s signals (which operates from the isolated network) to receive classified information. The two classes of attacks presented in this work differ in the way that the delay is caused. In the cross-router covert channel (CRCC) attack, the sender overloads the router with control-plane packets; in the Wi-Fi micro-jamming attack, the sender uses a pre-installed implant to transmit single-tone signals in the 2.4GHz frequency range, disturbing the router’s packet transmissions. We showed that both attacks can influence a wide range of router brands and Wi-Fi capable devices and evaluated the optimal settings for both attacks when using different types of packets, transmission power, and data transmission rates.Our proposed defense mechanism is based on semi-supervised machine learning and deep learning algorithms, which are both used for novelty detection in network traffic. By detecting unusual traffic, we can identify the disturbances needed to leak the information. The system can then respond by blocking the suspicious devices that are involved in the attack. We evaluated the attacks in noisy and noise-free environments, successfully detecting both attacks in both environments. All the data and code, which includes the implementation of the attacks and the defense mechanism, is published as a benefit to the research community.

Cross-collection latent Beta-Liouville allocation model training with privacy protection and applications.

Cross-collection topic models extend previous single-collection topic models, such as Latent Dirichlet Allocation (LDA), to multiple collections. The purpose of cross-collection topic modeling is to model document-topic representations and reveal similarities between each topic and differences among groups. However, the restriction of Dirichlet prior and the significant privacy risk have hampered those models' performance and utility. Training those cross-collection topic models may, in particular, leak sensitive information from the training dataset. To address the two issues mentioned above, we propose a novel model, cross-collection latent Beta-Liouville allocation (ccLBLA), which operates a more powerful prior, Beta-Liouville distribution with a more general covariance structure that enhances topic correlation analysis. To provide privacy protection for the ccLBLA model, we leverage the inherent differential privacy guarantee of the Collapsed Gibbs Sampling (CGS) inference scheme and then propose a hybrid privacy protection algorithm for the ccLBLA model (HPP-ccLBLA) that prevents inferring data from intermediate statistics during the CGS training process without sacrificing its utility. More crucially, our technique is the first attempt to use the cross-collection topic model in image classification applications and investigate the cross-collection topic model's capabilities beyond text analysis. The experimental results for comparative text mining and image classification will show the merits of our proposed approach.

A methodology to identify identical single-board computers based on hardware behavior fingerprinting

The connectivity and resource-constrained nature of single-board devices open the door to cybersecurity concerns affecting Internet of Things (IoT) scenarios. One of the most important issues is the presence of unauthorized IoT devices that want to impersonate legitimate ones by using identical hardware and software specifications. This situation can provoke sensitive information leakages, data poisoning, or privilege escalation in IoT scenarios. Combining behavioral fingerprinting and Machine/Deep Learning (ML/DL) techniques is a promising approach to identify these malicious spoofing devices by detecting minor performance differences generated by imperfections in manufacturing. However, existing solutions are not suitable for single-board devices since they do not consider their hardware and software limitations, underestimate critical aspects such as fingerprint stability or context changes, and do not explore the potential of ML/DL techniques. To improve it, this work first identifies the essential properties for single-board device identification: uniqueness, stability, diversity, scalability, efficiency, robustness, and security. Then, a novel methodology relies on behavioral fingerprinting to identify identical single-board devices and meet the previous properties. The methodology leverages the different built-in components of the system and ML/DL techniques, comparing the device internal behavior with each other to detect variations that occurred in manufacturing processes. The methodology validation has been performed in a real environment composed of 15 identical Raspberry Pi 4 Model B and 10 Raspberry Pi 3 Model B＋ devices, obtaining a 91.9% average TPR with an XGBoost model and achieving the identification for all devices by setting a 50% threshold in the evaluation process. Finally, a discussion compares the proposed solution with related work, highlighting the fingerprint properties not met, and provides important lessons learned and limitations.

Sensitive information leakage analysis of database code by abstract interpretation

An information system stores outside data in the backend database to process them efficiently and protects sensitive data from illegitimate flow or unauthorised users. However, most information systems are made in such a way that the sensitive information stored in a database may be leaked explicitly or implicitly during data processing along with the control structure of the program to the output channels. Therefore, sensitive data leakage is one of the crucial security threat. In this paper, the main objective is to detect the illegitimate flow of confidential information in an information system. We propose a framework to detect sensitive information leakage through the data-flow paths of an information system. In particular, to compute the precise set of data-flow paths, we use the non-relational abstract property of the interval domain and the relational abstract property of the polyhedra domain that enables the framework to produce efficient security analysis results.