Biblio

Deep learning has been used in a wide range of applications like computer vision, natural language processing and image detection. The advancement in deep learning algorithms in image detection and manipulation has led to the creation of deepfakes, deepfakes use deep learning algorithms to create fake images that are at times very hard to distinguish from real images. With the rising concern around personal privacy and security, Many methods to detect deepfake images have emerged, in this paper the use of deep learning for creating as well as detecting deepfakes is explored, this paper also propose the use of deep learning image enhancement method to improve the quality of deepfakes created.

In recent years, the advent of deep learning-based techniques and the significant reduction in the cost of computation resulted in the feasibility of creating realistic videos of human faces, commonly known as DeepFakes. The availability of open-source tools to create DeepFakes poses as a threat to the trustworthiness of the online media. In this work, we develop an open-source online platform, known as DeepFake-o-meter, that integrates state-of-the-art DeepFake detection methods and provide a convenient interface for the users. We describe the design and function of DeepFake-o-meter in this work.

Nowadays deepfake videos are concern with social ethics, privacy and security. Deepfake videos are synthetically generated videos that are generated by modifying the facial features and audio features to impose one person’s facial data and audio to other videos. These videos can be used for defaming and fraud. So, counter these types of manipulations and threats, detection of deepfake video is needed. This paper proposes multilayer hybrid recurrent deep learning models for deepfake video detection. Proposed models exploit the noise-based temporal facial convolutional features and temporal learning of hybrid recurrent deep learning models. Experiment results of these models demonstrate its performance over stacked recurrent deep learning models.

Having a clear insight on the protocols carrying traffic is crucial for network applications. Deep Packet Inspection (DPI) has been a key technique to provide visibility into traffic. DPI has proven effective in various scenarios, and indeed several open source DPI solutions are maintained by the community. Yet, these solutions provide different classifications, and it is hard to establish a common ground truth. Independent works approaching the question of the quality of DPI are already aged and rely on limited datasets. Here, we test if open source DPI solutions can provide useful information in practical scenarios, e.g., supporting security applications. We provide an evaluation of the performance of four open-source DPI solutions, namely nDPI, Libprotoident, Tstat and Zeek. We use datasets covering various traffic scenarios, including operational networks, IoT scenarios and malware. As no ground truth is available, we study the consistency of classification across the solutions, investigating rootcauses of conflicts. Important for on-line security applications, we check whether DPI solutions provide reliable classification with a limited number of packets per flow. All in all, we confirm that DPI solutions still perform satisfactorily for well-known protocols. They however struggle with some P2P traffic and security scenarios (e.g., with malware traffic). All tested solutions reach a final classification after observing few packets with payload, showing adequacy for on-line applications.

Cyber-physical systems (CPS) data privacy protection during sharing, aggregating, and publishing is a challenging problem. Several privacy protection mechanisms have been developed in the literature to protect sensitive data from adversarial analysis and eliminate the risk of re-identifying the original properties of shared data. However, most of the existing solutions have drawbacks, such as (i) lack of a proper vulnerability characterization model to accurately identify where privacy is needed, (ii) ignoring data providers privacy preference, (iii) using uniform privacy protection which may create inadequate privacy for some provider while over-protecting others, and (iv) lack of a comprehensive privacy quantification model assuring data privacy-preservation. To address these issues, we propose a personalized privacy preference framework by characterizing and quantifying the CPS vulnerabilities as well as ensuring privacy. First, we introduce a Standard Vulnerability Profiling Library (SVPL) by arranging the nodes of an energy-CPS from maximum to minimum vulnerable based on their privacy loss. Based on this model, we present our personalized privacy framework (PDP) in which Laplace noise is added based on the individual node's selected privacy preferences. Finally, combining these two proposed methods, we demonstrate that our privacy characterization and quantification model can attain better privacy preservation by eliminating the trade-off between privacy, utility, and risk of losing information.

Enabled by the industrial Internet, intelligent transportation has made remarkable achievements such as autonomous vehicles by carnegie mellon university (CMU) Navlab, Google Cars, Tesla, etc. Autonomous vehicles benefit, in various aspects, from the cooperation of the industrial Internet and cyber-physical systems. In this process, users in autonomous vehicles submit query contents, such as service interests or user locations, to service providers. However, privacy concerns arise since the query contents are exposed when the users are enjoying the services queried. Existing works on privacy preservation of query contents rely on location perturbation or k-anonymity, and they suffer from insufficient protection of privacy or low query utility incurred by processing multiple queries for a single query content. To achieve sufficient privacy preservation and satisfactory query utility for autonomous vehicles querying services in cyber-physical systems, this article proposes a novel privacy notion of client-based personalized k-anonymity (CPkA). To measure the performance of CPkA, we present a privacy metric and a utility metric, based on which, we formulate two problems to achieve the optimal CPkA in term of privacy and utility. An approach, including two modules, to establish mechanisms which achieve the optimal CPkA is presented. The first module is to build in-group mechanisms for achieving the optimal privacy within each content group. The second module includes linear programming-based methods to compute the optimal grouping strategies. The in-group mechanisms and the grouping strategies are combined to establish optimal CPkA mechanisms, which achieve the optimal privacy or the optimal utility. We employ real-life datasets and synthetic prior distributions to evaluate the CPkA mechanisms established by our approach. The evaluation results illustrate the effectiveness and efficiency of the established mechanisms.

Modern power systems depend on cyber-physical systems to link physical devices and control technologies. A major concern in the implementation of smart power networks is to minimize the risk of data privacy violation (e.g., by adversaries using data poisoning and inference attacks). In this article, we propose a privacy-preserving framework to achieve both privacy and security in smart power networks. The framework includes two main modules: a two-level privacy module and an anomaly detection module. In the two-level privacy module, an enhanced-proof-of-work-technique-based blockchain is designed to verify data integrity and mitigate data poisoning attacks, and a variational autoencoder is simultaneously applied for transforming data into an encoded format for preventing inference attacks. In the anomaly detection module, a long short-term memory deep learning technique is used for training and validating the outputs of the two-level privacy module using two public datasets. The results highlight that the proposed framework can efficiently protect data of smart power networks and discover abnormal behaviors, in comparison to several state-of-the-art techniques.

Cyber-Physical Systems (CPS) underpin global critical infrastructure, including power, water, gas systems and smart grids. CPS, as a technology platform, is unique as a target for Advanced Persistent Threats (APTs), given the potentially high impact of a successful breach. Additionally, CPSs are targets as they produce significant amounts of heterogeneous data from the multitude of devices and networks included in their architecture. It is, therefore, essential to develop efficient privacy-preserving techniques for safeguarding system data from cyber attacks. This paper introduces a comprehensive review of the current privacy-preserving techniques for protecting CPS systems and their data from cyber attacks. Concepts of Privacy preservation and CPSs are discussed, demonstrating CPSs' components and the way these systems could be exploited by either cyber and physical hacking scenarios. Then, classification of privacy preservation according to the way they would be protected, including perturbation, authentication, machine learning (ML), cryptography and blockchain, are explained to illustrate how they would be employed for data privacy preservation. Finally, we show existing challenges, solutions and future research directions of privacy preservation in CPSs.

The integration of advanced information, communication and data analytic technologies has transformed the traditional grid into an intelligent bidirectional system that can automatically adapt its services for utilities or consumers' needs. However, this change raises new privacy-related challenges. Privacy leakage has become a severe issue in the grid paradigm as adversaries run malicious analytics to identify the system's internal insight or use it to interrupt grids' operation by identifying real-time demand-based supply patterns. As a result, current grid authorities require an integrated mechanism to improve the system's sensitive data's privacy preservation. To this end, we present a multilayered smart grid architecture by characterizing the privacy issues that occur during data sharing, aggregation, and publishing by individual grid end nodes. Based on it, we quantify the nodes preferred privacy requirements. We further introduce personalized differential privacy (PDP) scheme based on trust distance in our proposed framework to provide the system with the added benefit of a user-specific privacy guarantee to eliminate differential privacy's limitation that allows the same level of privacy for all data providers. Lastly, we conduct extensive experimental analysis on a real-world grid dataset to illustrate that our proposed method is efficient enough to provide privacy preservation on sensitive smart grid data.

Protecting Cyber-physical Systems (CPSs) is highly important for preserving sensitive information and detecting cyber threats. Developing a robust privacy-preserving anomaly detection method requires physical and network data about the systems, such as Supervisory Control and Data Acquisition (SCADA), for protecting original data and recognising cyber-attacks. In this paper, a new privacy-preserving anomaly detection framework, so-called PPAD-CPS, is proposed for protecting confidential information and discovering malicious observations in power systems and their network traffic. The framework involves two main modules. First, a data pre-processing module is suggested for filtering and transforming original data into a new format that achieves the target of privacy preservation. Second, an anomaly detection module is suggested using a Gaussian Mixture Model (GMM) and Kalman Filter (KF) for precisely estimating the posterior probabilities of legitimate and anomalous events. The performance of the PPAD-CPS framework is assessed using two public datasets, namely the Power System and UNSW-NB15 dataset. The experimental results show that the framework is more effective than four recent techniques for obtaining high privacy levels. Moreover, the framework outperforms seven peer anomaly detection techniques in terms of detection rate, false positive rate, and computational time.

Cyber-physical systems are subject to natural uncertainties and sensor noise that can be amplified/attenuated due to feedback. In this work, we want to leverage these properties in order to define the inherent differential privacy of feedback-control systems without the addition of an external differential privacy noise. If larger levels of privacy are required, we introduce a methodology to add an external differential privacy mechanism that injects the minimum amount of noise that is needed. On the other hand, we show how the combination of inherent and external noise affects system security in terms of the impact that integrity attacks can impose over the system while remaining undetected. We formulate a bilevel optimization problem to redesign the control parameters in order to minimize the attack impact for a desired level of inherent privacy.

The privacy leakage resulting from location-based service (LBS) has become a critical issue. To preserve user privacy, many previous studies have investigated to prevent LBS servers from user privacy theft. However, they only consider whether the peers are innocent or malicious but ignore the relationship between the peers, whereas such a relationship between each pairwise of users affects the privacy leakage tremendously. For instance, a user has less concern of privacy leakage from a social friend than a stranger. In this paper, we study cyber-physical-social (CPS) aware method to address the privacy preserving in the case that not only LBS servers but also every other participant in the network has the probability to be malicious. Furthermore, by exploring the physical coupling and social ties among users, we construct CPS-aware privacy utility maximization (CPUM) game. We then study the potential Nash equilibrium of the game and show the existence of Nash equilibrium of CPUM game. Finally, we design a CPS-aware algorithm to find the Nash equilibrium for the maximization of privacy utility. Extensive evaluation results show that the proposed approach reduces privacy leakage by 50% in the case that malicious servers and users exist in the network.

As the public cloud becomes one of the leading ways in data sharing nowadays, data confidentiality and user privacy are increasingly critical. Partially policy-hidden ciphertext policy attribute-based encryption (CP-ABE) can effectively protect data confidentiality while reducing privacy leakage by hiding part of the access structure. However, it cannot satisfy the need of data sharing in the public cloud with complex users and large amounts of data, both in terms of less expressive access structures and limited granularity of policy hiding. Moreover, the verification of access right to shared data and correctness of decryption are ignored or conducted by an untrusted third party, and the prime-order groups are seldom considered in the expressive policy-hidden schemes. This paper proposes a fully policy-hidden CP-ABE scheme constructed on LSSS access structure and prime-order groups for public cloud data sharing. To help users decrypt, HVE with a ``convert step'' is applied, which is more compatible with CP-ABE. Meanwhile, decentralized credible verification of access right to shared data and correctness of decryption based on blockchain are also provided. We prove the security of our scheme rigorously and compare the scheme with others comprehensively. The results show that our scheme performs better.

This paper proposes a method of encrypting data stored in the blockchain network by applying ciphertext-policy attribute-based encryption (CP-ABE) and symmetric key algorithm. This method protects the confidentiality and privacy of data that is not protected in blockchain networks, and stores data in a more efficient way than before. The proposed model has the same characteristics of CP-ABE and has a faster processing speed than when only CP-ABE is used.

Malwares are the key means leveraged by threat actors in the cyber space for their attacks. There is a large array of commercial solutions in the market and significant scientific research to tackle the challenge of the detection and defense against malwares. At the same time, attackers also advance their capabilities in creating polymorphic and metamorphic malwares to make it increasingly challenging for existing solutions. To tackle this issue, we propose a methodology to perform malware detection and family attribution. The proposed methodology first performs the extraction of opcodes from malwares in each family and constructs their respective opcode graphs. We explore the use of clustering algorithms on the opcode graphs to detect clusters of malwares within the same malware family. Such clusters can be seen as belonging to different sub-family groups. Opcode graph signatures are built from each detected cluster. Hence, for each malware family, a group of signatures is generated to represent the family. These signatures are used to classify an unknown sample as benign or belonging to one the malware families. We evaluate our methodology by performing experiments on a dataset consisting of both benign files and malware samples belonging to a number of different malware families and comparing the results to existing approach.

Authorship attribution, an NLP problem where anonymous text is matched to its author, has important, cross-disciplinary applications, particularly those concerning cyber-defense. Our research examines the degree of sensitivity that attention-based models have to adversarial perturbations. We ask, what is the minimal amount of change necessary to maximally confuse a transformer model? In our investigation we examine a balanced subset of emails from the Enron email dataset, calculating the performance of our model before and after email signatures have been perturbed. Results show that the model's performance changed significantly in the absence of a signature, indicating the importance of email signatures in email authorship detection. Furthermore, we show that these models rely on signatures for shorter emails much more than for longer emails. We also indicate that additional research is necessary to investigate stylometric features and adversarial training to further improve classification model robustness.

Content blocking is an important part of a per-formant, user-serving, privacy respecting web. Current content blockers work by building trust labels over URLs. While useful, this approach has many well understood shortcomings. Attackers may avoid detection by changing URLs or domains, bundling unwanted code with benign code, or inlining code in pages.The common flaw in existing approaches is that they evaluate code based on its delivery mechanism, not its behavior. In this work we address this problem by building a system for generating signatures of the privacy-and-security relevant behavior of executed JavaScript. Our system uses as the unit of analysis each script's behavior during each turn on the JavaScript event loop. Focusing on event loop turns allows us to build highly identifying signatures for JavaScript code that are robust against code obfuscation, code bundling, URL modification, and other common evasions, as well as handle unique aspects of web applications.This work makes the following contributions to the problem of measuring and improving content blocking on the web: First, we design and implement a novel system to build per-event-loop-turn signatures of JavaScript behavior through deep instrumentation of the Blink and V8 runtimes. Second, we apply these signatures to measure how much privacy-and-security harming code is missed by current content blockers, by using EasyList and EasyPrivacy as ground truth and finding scripts that have the same privacy and security harming patterns. We build 1,995,444 signatures of privacy-and-security relevant behaviors from 11,212 unique scripts blocked by filter lists, and find 3,589 unique scripts hosting known harmful code, but missed by filter lists, affecting 12.48% of websites measured. Third, we provide a taxonomy of ways scripts avoid detection and quantify the occurrence of each. Finally, we present defenses against these evasions, in the form of filter list additions where possible, and through a proposed, signature based system in other cases.As part of this work, we share the implementation of our signature-generation system, the data gathered by applying that system to the Alexa 100K, and 586 AdBlock Plus compatible filter list rules to block instances of currently blocked code being moved to new URLs.

In this study, a novel decentralized authentication model is proposed for establishing a secure communications structure in VoIP applications. The proposed scheme considers a distributed architecture called the blockchain. With this scheme, we highlight the multimedia data is more resistant to some of the potential attacks according to the centralized architecture. Our scheme presents the overall system authentication architecture, and it is suitable for mutual authentication in terms of privacy and anonymity. We construct an ECC-based model in the encryption infrastructure because our structure is time-constrained during communications. This study differs from prior work in that blockchain platforms with ECC-Based Biometric Signature. We generate a biometric key for creating a unique ID value with ECC to verify the caller and device authentication together in blockchain. We validated the proposed model by comparing with the existing method in VoIP application used centralized architecture.

Security plays a major role in data transmission and reception. Providing high security is indispensable in communication systems. The RSA (Rivest-Shamir-Adleman) cryptosystem is used widely in cryptographic applications as it offers highly secured transmission. RSA cryptosystem uses Montgomery multipliers and it involves modular exponentiation process which is attained by performing repeated modular-multiplications. This leads to high latency and owing to improve the speed of multiplier, highly efficient modular multiplication methodology needs to be applied. In the conventional methodology, Carry Save Adder (CSA) is used in the multiplication and it consumes more area and it has larger delay, but in the suggested methodology, the Reverse Carry Propagate (RCP) adder is used in the place of CSA adder and the obtained output shows promising results in terms of area and latency. The simulation is done with Xilinx ISE design suite. The proposed multiplier can be used effectively in signal processing, image processing and security based applications.

Distributed acoustic sensing (DAS) systems based on fiber brag grating (FBG) have been widely used for distributed temperature and strain sensing over the past years, and function well in perimeter security monitoring and structural health monitoring. However, with relevant algorithms functioning with low accuracy, the DAS system presently has trouble in signal recognition, which puts forward a higher requirement on a high-precision identification method. In this paper, we propose an improved recognition method based on relative fundamental signal processing methods and convolutional neural network (CNN) to construct a mathematical model of disturbance FBG signal recognition. Firstly, we apply short-time energy (STE) to extract original disturbance signals. Secondly, we adopt short-time Fourier transform (STFT) to divide a longer time signal into short segments. Finally, we employ a CNN model, which has already been trained to recognize disturbance signals. Experimental results conducted in the real environments show that our proposed algorithm can obtain accuracy over 96.5%.

Recent developments in deepfake technology are increasing new security threats. To solve these issues, various detection methods have been proposed including the methods utilizing biological signals captured by R-PPG. However, existing methods have limitations in terms of detection accuracy and generalized performance. In this paper, we present our approach for R-PPG-based BPM (Beats Per Minute) analysis for effective deepfake detection. With the selected deepfake datasets, we performed (a) comparison and analysis of conditions for BPM processing, and (b) BPM extraction by dividing the face into 16 regions and comparison of BPM in each region. The results showed that our proposed BPM-related properties are effective in deepfake detection.

In the current network security situation, the types of network threats are complex and changeable. With the development of the Internet and the application of information technology, the general trend is opener. Important data and important business applications will face more serious security threats. However, with the development of cloud computing technology, the trend of large-scale deployment of important business applications in cloud centers has greatly increased. The development and use of software-defined networks in cloud data centers have greatly reduced the effect of traditional network security boundary protection. How to find an effective way to protect important applications in open multi-step large-scale cloud data centers is a problem we need to solve. Threat intelligence has become an important means to solve complex network attacks, realize real-time threat early warning and attack tracking because of its ability to analyze the threat intelligence data of various network attacks. Based on the research of threat intelligence, machine learning, cloud central network, SDN and other technologies, this paper proposes an active defense method of network security based on threat intelligence for super-large cloud data centers.

Wireless communication systems are inherently vulnerable to adversarial attacks since malevolent jammers might jam and disrupt the legitimate transmission intentionally. Of particular interest are so- called denial-of-service (DoS) attacks in which the jammer is able to completely disrupt the communication. Accordingly, it is of crucial interest for the legitimate users to detect such DoS attacks. Turing machines provide the fundamental limits of today's digital computers and therewith of the traditional signal processing. It has been shown that these are incapable of detecting DoS attacks. This stimulates the question of how powerful the signal processing must be to enable the detection of DoS attacks. This paper investigates the general computation framework of Blum-Shub-Smale machines which allows the processing and storage of arbitrary reals. It is shown that such real number signal processing then enables the detection of DoS attacks.

The latest security methods are based on biometric features. The electrocardiogram is increasingly used in such systems because it provides biometric features that are difficult to falsify. This paper aims to study the use of the electrocardiogram together with the Convolutional Neural Networks, in order to identify the subjects based on the ECG signal and to improve the security. In this study, we used the Fantasia database, available on the PhysioNet platform, which contains 40 ECG recordings. The ECG signal is pre-processed, and then spectrograms are generated for each ECG signal. Spectrograms are applied to the input of several architectures of Convolutional Neural Networks like Inception-v3, Xception, MobileNet and NasNetLarge. An analysis of performance metrics reveals that the subject identification method based on ECG signal and CNNs provides remarkable results. The best accuracy value is 99.5% and is obtained for Inception-v3.

A conventional visible light communication system consists of a transmitter, a jammer that includes a few light emitting diodes, a legal listener and an eavesdropper. In this work, a similar system is designed with a collimating lens in order to create an extra layer of practical physical security measure. The use of a collimating lens makes it available to spatially limiting data transmission to an area under the lensed transmitter. Also focused data transmission through the optical lens, increases the secrecy rate. To investigate the applicability of the proposed design we designed a sample experimental setup using USRP and implemented in a laboratory environment. In the proposed set up, the receiver is in a fixed position. However, it is possible to implement an easy, practical and cheap hardware solution with respect to a beamforming type VLC that uses directional beam forming method to establish transmission to a dynamic target. In addition, it is achievable to control the size of the area where a receiver can access data by manipulating the distance between the optical lens and transmitter.