Provide Data Privacy Research Articles

Enhanced Secure Leach with Advanced Encryption Standard And Attack Detection Scheme

Wireless Sensor Networks (WSNs) have become widely used in a variety of applications, ranging from military surveillance to environmental monitoring, as a result of their quick development. However, WSNs are susceptible to security attacks due to their open communication channels and limited resources, which jeopardizes network dependability and data integrity. The network employs Abundant Secure LEACH protocol to create an energy-efficient system that is vulnerable to a wide range of assaults, including the HELLO flood. The proposed scheme uses Enhanced Secure-LEACH with Advanced Encryption Standard and Attack Detection (SLEAD) mechanism to protect cluster heads from sinkhole, sybil and Hello flood attacks and provide data privacy. SLEAD makes use of a unique ID for each sensor node and Advanced Encryption Standard Mechanism for the purpose of authenticating a sensor node as CH and ensure the security of data. SLEAD algorithm was implemented using Python within a Jupyter Notebook environment. The simulation result shows that SLEAD outperformed the traditional scheme in-terms of efficient energy utilization and data privacy.

Centralized vs. Decentralized Cloud Computing in Healthcare

Healthcare is one of the industries that seeks to deliver medical services to patients on time. One of the issues it currently grapples with is real-time patient data exchange between various healthcare organizations. This challenge was solved by both centralized and decentralized cloud computing architecture solutions. In this paper, we review the current state of these two cloud computing architectures in the health sector with regard to the effect on the efficiency of Health Information Exchange (HIE) systems. Our study seeks to determine the relevance of these cloud computing approaches in assisting healthcare facilities in the decision-making process to adopt HIE systems. This paper considers the system performance, patient data privacy, and cost and identifies research directions in each of the architectures. This study shows that there are some benefits in both cloud architectures, but there are also some drawbacks. The prominent characteristic of centralized cloud computing is that all data and information are stored together at one location, known as a single data center. This offers many services, such as integration, effectiveness, simplicity, and rapid information access. However, it entails providing data privacy and confidentiality aspects because it will face the hazard of a single point of failure. On the other hand, decentralized cloud computing is built to safeguard data privacy and security whereby data are distributed to several nodes as a way of forming mini-data centers. This increases the system’s ability to cope with a node failure. Thus, continuity and less latency are achieved. Nevertheless, it poses integration issues because managing data from several sites could be a problem, and the costs of operating several data centers are higher and complex. This paper also pays attention to the differences in aspects like efficiency, capacity, and cost. This paper assists healthcare organizations in determining the most suitable cloud architecture strategy for deploying secure and effective HIE systems.

Compact: Approximating Complex Activation Functions for Secure Computation

Secure multi-party computation (MPC) techniques can be used to provide data privacy when users query deep neural network (DNN) models hosted on a public cloud. State-of-the-art MPC techniques can be directly leveraged for DNN models that use simple activation functions (AFs) such as ReLU. However, these techniques are ineffective and/or inefficient for the complex and highly non-linear AFs used in cutting-edge DNN models. We present Compact, which produces piece-wise polynomial approximations of complex AFs to enable their efficient use with state-of-the-art MPC techniques. Compact neither requires nor imposes any restriction on model training and results in near-identical model accuracy. To achieve this, we design Compact with input density awareness, and use an application specific simulated annealing type optimization to generate computationally more efficient approximations of complex AFs. We extensively evaluate Compact on four different machine-learning tasks with DNN architectures that use popular complex AFs silu, gelu, and mish. Our experimental results show that Compact incurs negligible accuracy loss while being 2x-5x computationally more efficient than state-of-the-art approaches for DNN models with large number of hidden layers. Our work accelerates easy adoption of MPC techniques to provide user data privacy even when the queried DNN models consist of a number of hidden layers, and trained over complex AFs.

Detection of Covid‐19 disease by using privacy‐aware artificial intelligence system

AbstractCovid‐19 is a highly infectious respiratory disease that spreads quickly between individuals and has been recognized as a pandemic by the World Health Organization (WHO). Chest x‐ray images, lung computed tomography images, and polymerase chain reaction tests are generally used to diagnose this disease by the doctors. Nevertheless, manual diagnosis of Covid‐19 disease is laborious and requires highly experienced professionals. Therefore, automated systems are always needed to assist doctors in their diagnostic decisions. In the field of medicine and healthcare, artificial intelligence and deep learning currently offer incredibly effective and rapid automatic decision‐support systems. Since sensitive data is used to diagnose Covid‐19, privacy has become a major concern in research that uses artificial intelligence and deep learning. In order to eliminate these issues, this paper proposes a novel deep learning model that privately detects Covid‐19 disease. The proposed model utilizes differential privacy technique to provide data privacy and convolutional neural network to diagnose Covid‐19 disease. The performance of the proposed model was evaluated through experiments conducted on five different datasets, resulting a maximum accuracy rate of 97%.

Defence against Side-Channel Attacks for Encrypted Network Communication Using Multiple Paths

As more network communication is encrypted to provide data privacy for users, attackers are focusing their attention on traffic analysis methods for side-channel attacks on user privacy. These attacks exploit patterns in particular features of communication flows such as interpacket timings and packet sizes. Unsupervised machine learning approaches, such as Hidden Markov Models (HMMs), can be trained on unlabelled data to estimate these flow attributes from an exposed packet flow, even one that is encrypted, so it is highly feasible for an eavesdropper to perform this attack. Traditional defences try to protect specific side channels by modifying the packet transmission for the flow, e.g., by adding redundant information (padding of packets or use of junk packets) and perturbing packet timings (e.g., artificially delaying packet transmission at the sender). Such defences incur significant overhead and impact application-level performance metrics, such as latency, throughput, end-to-end delay, and jitter. Furthermore, these mechanisms can be complex, often ineffective, and are not general solutions—a new profile must be created for every application, which is an infeasible expectation to place on software developers. We show that an approach exploiting multipath communication can be effective against HMM-based traffic analysis. After presenting the core analytical background, we demonstrate the efficacy of this approach with a number of diverse, simulated traffic flows. Based on the results, we define some simple design rules for software developers to adopt in order to exploit the mechanism we describe, including a critical examination of existing communication protocol behavior.

Analysis on Privacy Preserving Clustering Methods for Big Datasets Using Random Number Generators

Due to the growing requirement for powerful processing capabilities, safeguarding and processing sensitive data in big data systems has become an increasingly prevalent issue. Additional effort is required to secure sensitive data on a system with numerous connections, each of which has its own privacy policy, and to exchange cryptographic keys in accordance with the appropriate protocols for each of these parties. Data encrypted using homomorphic encryption algorithms can undergo the same kinds of arithmetic operations as data encrypted using plaintext. This allows for the outsourcing of computations to cloud services while also providing data privacy by processing in the encrypted domain. Also, key exchange sessions become easier for everyone involved because of this. This study presents new privacy-preserving clustering algorithms and homomorphic encryption systems that may together operate on a shared HPC infrastructure, like the cloud. Therefore, it is not necessary for the participants in this system to have a lot of processing power, as the system would distribute the most power-intensive jobs to any provider of cloud computing. Multiple clustering techniques can have their distance matrices computed in a way that does not compromise user privacy using our technology.

Analysis on Privacy Preserving Clustering Methods for Big Datasets Using Random Number Generators

Due to the growing requirement for powerful processing capabilities, safeguarding and processing sensitive data in big data systems has become an increasingly prevalent issue. Additional effort is required to secure sensitive data on a system with numerous connections, each of which has its own privacy policy, and to exchange cryptographic keys in accordance with the appropriate protocols for each of these parties. Data encrypted using homomorphic encryption algorithms can undergo the same kinds of arithmetic operations as data encrypted using plaintext. This allows for the outsourcing of computations to cloud services while also providing data privacy by processing in the encrypted domain. Also, key exchange sessions become easier for everyone involved because of this. This study presents new privacy-preserving clustering algorithms and homomorphic encryption systems that may together operate on a shared HPC infrastructure, like the cloud. Therefore, it is not necessary for the participants in this system to have a lot of processing power, as the system would distribute the most power-intensive jobs to any provider of cloud computing. Multiple clustering techniques can have their distance matrices computed in a way that does not compromise user privacy using our technology.

A Novel Optimized Perturbation-Based Machine Learning for Preserving Privacy in Medical Data.

In recent times, providing privacy to the medical dataset has been the biggest issue in medical applications. Since, in hospitals, the patient's data are stored in files, the files must be secured properly. Thus, different machine learning models were developed to overcome data privacy issues. But, those models faced some problems in providing privacy to medical data. Therefore, a novel model named Honey pot-based Modular Neural System (HbMNS) was designed in this paper. Here, the performance of the proposed design is validated with disease classification. Also, the perturbation function and the verification module are incorporated into the designed HbMNS model to provide data privacy. The presented model is implemented in a python environment. Moreover, the system outcomes are estimated before and after fixing the perturbation function. A DoS attack is launched in the system to validate the method. At last, a comparative assessment is made between executed models with other models. From the comparison, it is verified that the presented model achieved better outcomes than others.

Privacy-preserving data aggregation scheme against deletion and tampering attacks from aggregators

Because privacy-preserving data aggregation protocols provide data privacy and data compression, they have been extensively studied in smart grids. However, most of the existing data aggregation protocols are based on honest-but-curious aggregators, or adopt computationally intensive asymmetric homomorphic encryption, making these protocols only provably secure in weak security models, or resulting in high computational complexity for terminal users. To address the aforementioned issues, in this work, we propose a novel lightweight privacy-preserving data aggregation scheme against malicious aggregators based on our symmetric homomorphic encryption. The proposed scheme can reduce the computational cost of smart meters while also resisting tampering and deletion attacks from malicious aggregators. One of the highlights of the proposed scheme is its ability to withstand deletion attacks from malicious aggregators, which makes it unnecessary for the malicious aggregator to aggregate the readings of all smart meters instead of only a part of them. Therefore, our scheme is more secure than other schemes. It is the first privacy-preserving data aggregation scheme based on a stronger security model without multiple rounds of interaction. Finally, the detailed security analysis shows that the proposed scheme satisfies desirable security properties. Moreover, experimental results demonstrate that the proposed scheme is superior to the other four schemes in terms of computational complexity and communication overhead.

VoIPChain: A decentralized identity authentication in Voice over IP using Blockchain

Confidentiality, availability, integrity, authentication, and non-repudiation when together became the top priority for secure communication. However, authentication is the first line of defense in all these security parameters. Unfortunately, several authentication mechanisms in the traditional methods cannot provide secure communication due to various vulnerabilities such as single point of failure and privacy issues. Furthermore, key distribution mechanisms such as Certificate Authority (CA) and Trusted Third Party (TTP) distribute keys over the central architecture fail, thus secure communication cannot occur. As a solution to these drawbacks, this paper proposes a new decentralized blockchain-based identity authentication mechanism for VoIP networks, VoIPChain. The proposed scheme utilizes the main features of the blockchain platform, such as immutability, transparency, and fault tolerance, to provide data privacy and secure communication in VoIP applications. The proposed scheme is evaluated as an actual implementation using the virtual Ethereum platform and Python language. The experimental results show that the proposed scheme is an efficient and cost-effective solution for the call process as a decentralized identity authentication system in VoIP. In addition, end-to-end secure call performance outperforms the existing blockchain-based authentication mechanisms by 30%–70% in terms of average time delay. The proposed scheme is almost ten times faster than the TLS process to authenticate between parties. Moreover, compared to fast but less secure basic methods over the SIP authentication the proposed scheme has an acceptable time delay in a call process.

Blockchain-Based Distributed Federated Learning in Smart Grid

The participation of prosumers in demand-response programs is essential for the success of demand-side management in renewable-powered energy grids. Unfortunately, the engagement is still low due to concerns related to the privacy of their energy data used in the prediction processes. In this paper, we propose a blockchain-based distributed federated learning (FL) technique for energy-demand prediction that combines FL with blockchain to provide data privacy and trust features for energy prosumers. The privacy-sensitive energy data are stored locally at edge prosumer nodes without revealing it to third parties, with only the learned local model weights being shared using a blockchain network. The global federated model is not centralized but distributed and replicated over the blockchain overlay, ensuring the model immutability and provenance of parameter updates. We had proposed smart contracts to deal with the integration of local machine-learning prediction models with the blockchain, defining functions for the model parameters’ scaling and reduction of blockchain overhead. The centralized, local-edge, and blockchain-integrated models are comparatively evaluated for prediction of energy demand 24 h ahead using a multi-layer perceptron model and the monitored energy data of several prosumers. The results show only a slight decrease in prediction accuracy in the case of blockchain-based distributed FL with reliable data privacy support compared with the centralized learning solution.

Disease Prediction in Edge Computing: A Privacy-Preserving Technique for PHI Collection and Analysis

Edge computing has garnered significant attention in recent years, as it enables the extension of cloud resources to the network edge. This enables the user to utilize virtually enhanced resources in terms of storage and computation at a lower cost. The edge-computing-assisted wireless wearable communication (EWWC) technology is a prime candidate for e-health edge applications to collect personal health information, which leads to disease learning and prediction. Ensuring privacy and efficiency of such a system in EWWC is extremely important. In this article, we introduce an efficient and privacy-preserving disease prediction scheme. We use the randomizable signature and matrices encryption technique to achieve identity protection and data privacy. The experimental analysis shows that our solution outperforms the existing solution in terms of computational costs and communication overhead. At the same time, it is able to provide data privacy, prediction model security, user identity protection, mendacious data traceability, and model verifiability. We also analyze potential future research directions related to this emerging area.

Multibiosignal chaotic encryption scheme based on spread spectrum and global diffusion process for e-health

E-health provides several remote health services highly convenient for patients. Nevertheless, inherent security risks are present in e-health such as data privacy since insecure communication channels are used to transmit data between patients and physicians or to cloud medical data storage. Nowadays, medical body area network provides a solution to continuous health monitoring with real-time updates of multibiosignal records in multiuser communication schemes for monitoring and early detection of medical conditions or diseases. In this paper, we propose to use chaos-based sequence direct spread spectrum technology and chaos-based cryptography to provide data privacy, high security and efficiency in monitoring multibiosignals in multiuser networks for e-health applications. First, each plain biosignal is modulated with sequence direct spread spectrum by using the chaotic Hénon map. Then, all modulated biosignals are summed to produce the modulated cryptogram. The modulated cryptogram is finally encrypted by using global diffusion process with the chaotic Hénon map to increase security by reducing the high autocorrelation of the modulated cryptogram. The experimental results are presented in a simulated multiuser scenario with 4 users (or patients) and 2 specialists (neurologist and cardiologist), with 12 biosignals acquired from database on the Internet, such as electroencephalograms, electromyograms, blood pressure signals, and electrocardiograms. A comprehensive security study is presented to show the high security, the high speed encryption, and the high noise interference robustness. According with the results, the proposed method can be used for secure multibiosignal monitoring in multiuser networks for e-health applications with high efficiency and security.

Multi-Level Security Model Developed to Provide Data Privacy in Distributed Database Systems

Multi-Level Security Model Developed to Provide Data Privacy in Distributed Database Systems

Generative Adversarial Networks in Security and Data Privacy Sector: A Survey

In the modern era, most of the data exists in the digital form, however existing security measures are insufficient to protect this data since they do not have enough ability toanticipate malicious attacks. With an upsurge in the number of security challenges, data protection systems along with security defense and attack mechanisms must keep up with the rapidly changing threats. In this survey, we review the Generative Adversarial Networks (GANs) and its different variants that have been employed to provide data privacy protection to systems, security defense, along with security attack capability to systems. The main goal of this survey is to investigate the recent developments in the field of security along with data protection based upon GANs architectures. The GANs applications, challenges along with potential solutions from recent research developments have also been discussed in brief.

Compositional Defence of Application Privacy in Resistant to Physical and Software Attacks in Untrusted Cloud Environment

Cloud computing facilities have become an increasing rise in demand for presenting computing sources. Various programs like monetary transactions, fitness care systems, video streaming, IoT programs want on-call for provisioning of cloud sources to guarantee timeliness and excessive availability. A shared culture for data interchange has evolved as a result of the fast progress of networking technology. The data security has become a difficult issue as a result of the vast volume of data that is exchanged via the internet. Thus there is a requirement for security to protect data that is transmitted across an insecure connection. In our proposed work, privacy, confidentiality, availability, integrity, and accountability that contribute to the security of information are ensured by cryptographic technique. This paper also describes an exhaustive study on the usage of hybrid algorithms i.e. the combination of Advanced Encryption Standards (AES) and Elliptical Curve Cryptography (ECC) to protect data leakage and protect the privacy of end-users and SHA256 algorithm to ensure data integrity. This type of hybrid encryption will help not only in concentrating software as well as physical attacks but also provide the model to deal with and prevent the application from these types of attacks; thus making the cloud application system more secure. The implemented web application can be used as a defence system for privacy and security in organizations as well as institutions where members of the organization can share the file or confidential data to other team members or colleagues without data leakage thus providing data privacy and security.

CP-ABSEL: Ciphertext-policy attribute-based searchable encryption from lattice in cloud storage

Ciphertext-policy attribute-based searchable encryption (CP-ABSE) is widely used in the cloud environment to provide data privacy and fine-grained access control over encrypted data. The existing CP-ABSE schemes are designed based on bilinear pairing hardness assumptions to prove their security. However, these schemes are vulnerable to quantum attacks, i.e., adversaries can break the security of these schemes with the use of quantum computers. To address this issue, in this paper, we propose a novel ciphertext-policy attribute-based searchable encryption from lattice (CP-ABSEL) in cloud storage, since lattice-based cryptography is quantum attacks free. In CP-ABSEL, we adopted learning with errors (LWE) hardness assumption to resist from quantum attacks. Further, CP-ABSEL is indistinguishable against the chosen keyword attack and indistinguishable against chosen plaintext attack. Moreover, CP-ABSEL allows only legitimate users to perform a keyword search over an encrypted index, and unauthorized users cannot get even the ciphertext form of documents. The performance analysis proves that CP-ABSEL is efficient and practical.

Information Flow Control with Decentralized Labeling Model in Information Security

Data security aims to prevent the use, modification, and spread of data by unauthorized people. In this study, our purpose was to provide data privacy and confidentiality with information flow control in distributed databases. In particular, a decentralized label model was developed that maintained confidentiality, including privacy, with data flow control. This model consists of an actor, an object, and a label. The owners of the objects are actors, and they need to share their data objects with others. Actors label the data objects and then send them out. A label contains the policy statements of data security issued by each of the owners. Each owner sets its own security and privacy policy independently of the other owners. The confidentiality of data in unsecured transport channels is ensured for all the actors in the system by means of labels while the data are in flow. Data objects are spread and shared securely among actors within unsecured environments. In addition, with the path compression, the long node chain that is formed while the data objects are passing between the source node and the destination is broken, so that the objects are retrieved fast, and the cost of access is reduced. This result was shown experimentally by modeling the distributed environment.

ContAuth

User authentication is key in user authorization on smart and personal devices. Over the years, several authentication mechanisms have been proposed: these also include behavioral-based biometrics. However, behavioral-based biometrics suffer from two issues: they are prone to degradation in performance (accuracy) over time (e.g., due to data distribution changes arising from user behavior) and the need to learn the machine learning model from scratch, when adding new users. In this paper, we propose ContAuth, a system that can enhance the robustness of behavioral-based authentication. ContAuth continuously adapts to new incoming data (data incremental learning) and is able to add new users without retraining (class incremental learning). Specifically, ContAuth combines deep learning models with online learning models to achieve learning on the fly, thereby preventing a severe drop in the accuracy between sessions (over time). To add new users, ContAuth employs class incremental learning methods. We evaluate ContAuth on multiple behavior-based user authentication modalities: breathing, gait. and EMG. Our results show that our framework can help True Positive Rate (TPR) to remain high (>85 %) compared to other methods for all the modalities except EMG (>70%) across the sessions while keeping False Positive Rates (FPR) at a minimum (0-10%). It can achieve up to 35% improvement in TPR over a traditional deep learning model. Additionally, iCaRL (an incremental learning method) enables ContAuth to allow the addition of new users by alleviating catastrophic forgetting, to a large extent. Finally, we also show that ContAuth can be deployed efficiently and effectively on device, further providing data privacy.

Privacy preserving classification over differentially private data

AbstractPrivacy preserving data classification is an important research area in data mining field. The goal of a privacy preserving classification algorithm is to protect the sensitive information as much as possible, while providing satisfactory classification accuracy. Differential privacy is a strong privacy guarantee that enables privacy of sensitive data stored in a database by determining the ratio of sensitive information leakage with respect to an ɛ parameter. In this study, our aim is to investigate the classification performance of the state‐of‐the‐art classification algorithms such as C4.5, Naïve Bayes, One Rule, Bayesian Networks, PART, Ripper, K*, IBk, and Random tree for performing privacy preserving classification. To preserve privacy of the data to be classified, we applied input perturbation technique coming from differential privacy, and observed the relationship between the ɛ parameter values and accuracy of the classifiers. To our best knowledge, this article is the first study that analyzes the performances of the well‐known classification algorithms over differentially private data, and discovers which datasets are more suitable for privacy preserving classification when input perturbation is applied to provide data privacy. The classification algorithms are compared by using the differentially private versions of the well‐known datasets from the UCI repository. According to the experimental results, we observed that, as ɛ parameter value increases, better classification accuracies are achieved with lower privacy levels. When the classifiers are compared, Naïve Bayes classifier is the most successful method. The ɛ parameter should be greater than or equal to 2 (i.e., ɛ ≥2) to achieve cloud server is malicious and untrusted, sensitive data will satisfactory classification accuracies.This article is categorized under: Commercial, Legal, and Ethical Issues > Security and Privacy Technologies > Classification