Biblio

Cloud computing has been widely used because of its low price, high reliability, and generality of services. However, considering that cloud computing transactions between users and service providers are usually asynchronous, data privacy involving users and service providers may lead to a crisis of trust, which in turn hinders the expansion of cloud computing applications. In this paper, we propose DPP, a data privacy-preserving cloud computing scheme based on homomorphic encryption, which achieves correctness, compatibility, and security. DPP implements data privacy-preserving by introducing homomorphic encryption. To verify the security of DPP, we instantiate DPP based on the Paillier homomorphic encryption scheme and evaluate the performance. The experiment results show that the time-consuming of the key steps in the DPP scheme is reasonable and acceptable.

Homomorphic encryption (HE) facilitates computing over encrypted data without using the secret keys. It is currently inefficient for practical implementation on the Internet of Things (IoT). However, the performance of these HE schemes may increase with optimized libraries and hardware capabilities. Thus, implementing and analyzing HE schemes and protocols on resource-constrained devices is essential to deriving optimized and secure schemes. This paper develops an energy profiling framework for homomorphic encryption on IoT devices. In particular, we analyze energy consumption and performance such as CPU and Memory utilization and execution time of numerous HE schemes using SEAL and HElib libraries on the Raspberry Pi 4 hardware platform and study energy-performance-security trade-offs. Our analysis reveals that HE schemes can incur a maximum of 70.07% in terms of energy consumption among the libraries. Finally, we provide guidelines for optimization of Homomorphic Encryption by leveraging multi-threading and edge computing capabilities for IoT applications. The insights obtained from this study can be used to develop secure and resource-constrained implementation of Homomorphic encryption depending on the needs of IoT applications.

With the development of sensor technology, people put forward a higher level, more diversified demand for portable rangefinders. However, its data storage method has not been developed in a large scale and breakthrough. This paper studies the design of portable sensor data storage system based on homomorphic encryption algorithm, which aims to maintain the security of sensor data storage through homomorphic encryption algorithm. This paper analyzes the functional requirements of the sensor data storage system, puts forward the overall design scheme of the system, and explains in detail the requirements and indicators for the specific realization of each part of the function. Analyze the different technical resources currently used in the storage system field, and dig deep into the key technologies that match the portable sensor data storage system. This paper has changed the problem of cumbersome operation steps and inconvenient data recovery in the sensor data storage system. This paper mainly uses the method of control variables and data comparison to carry out the experiment. The experimental results show that the success rate of the sensor data storage system under the homomorphic encryption algorithm is infinitely close to 100% as the number of data blocks increases.

This paper presents topological sorting methods to minimize the multiplicative depth of encrypted arithmetic expressions. The research aims to increase compatibility between nonlinear dynamic control schemes and homomorphic encryption methods, which are known to be limited by the quantity of multiplicative operations. The proposed method adapts rewrite rules originally developed for encrypted binary circuits to depth manipulation of arithmetic circuits. The paper further introduces methods to normalize circuit paths that have incompatible depth. Finally, the paper provides benchmarks demonstrating the improved depth in encrypted computed torque control of a dynamic manipulator and discusses how achieved improvements translate to increased cybersecurity.

Cloud computing plays major role in the development of accessing clouduser’s document and sensitive information stored. It has variety of content and representation. Cyber security and attacks in the cloud is a challenging aspect. Information security attains a vital part in Cyber Security management. It involves actions intended to reduce the adverse impacts of such incidents. To access the documents stored in cloud safely and securely, access control will be introduced based on cloud users to access the user’s document in the cloud. To achieve this, it is highly required to combine security components (e.g., Access Control, Usage Control) in the security document to get automatic information. This research work has proposed a Role Key Homomorphic Encryption Algorithm (RKHEA) to monitor the cloud users, who access the services continuously. This method provides access creation of session-based key to store the singularized encryption to reduce the key size from random methods to occupy memory space. It has some terms and conditions to be followed by the cloud users and also has encryption method to secure the document content. Hence the documents are encrypted with the RKHEA algorithm based on Service Key Access (SKA). Then, the encrypted key will be created based on access control conditions. The proposed analytics result shows an enhanced control over the documents in cloud and improved security performance.

Fully homomorphic encryption technologies allow you to operate on encrypted data without disclosing it, therefore they have a lot of potential for solving personal data storage and processing issues. Because of the increased interest in these technologies, various software tools and libraries that allow completely homomorphic encryption have emerged. However, because this subject of cryptography is still in its early stages, standards and recommendations for the usage of completely homomorphic encryption algorithms are still being developed. The paper presents the main areas of application of homomorphic encryption. The analysis of existing developments in the field of homomorphic encryption is carried out. The analysis showed that existing library implementations do not support the division and subtraction operation. The analysis revealed the need to develop a library of fully homomorphic encryption, which allows performing all mathematical operations on them (addition, difference, multiplication and division), as well as the relevance of developing its own implementation of a library of homomorphic encryption on integers. Then, implement the development of a fully homomorphic encryption library in C++ and on an ESP 32 microcontroller. The ability to perform four operations (addition, difference, multiplication and division) on encrypted data will expand the scope of application of homomorphic encryption. A method of homomorphic division and subtraction is proposed that allows performing the division and subtraction operation on homomorphically encrypted data. The level of security, the types of operations executed, the maximum length of operands, and the algorithm's running time are all described as a consequence of numerical experimentation with parameters.

Homomorphic encryption is suitable for a machine learning in the cloud such as a privacy-preserving machine learning. However, ordinary homomorphic public key encryption has a problem that public key holders can generate ciphertexts and anyone can execute homomorphic operations. In this paper, we will propose a solution based on the Keyed Homomorphic-Public Key Encryption proposed by Emura et al.

Currently, mobile ad hoc networks (MANETs) are widely used due to its self-configuring feature. However, it is vulnerable to the malicious jammers in practice. Traditional anti-jamming approaches, such as channel hopping based on deterministic sequences, may not be the reliable solution against intelligent jammers due to its fixed patterns. To address this problem, we propose a distributed game theory-based multi-agent anti-jamming (DMAA) algorithm in this paper. It enables each user to exploit all information from its neighboring users before the network attacks, and derive dynamic local policy knowledge to overcome intelligent jamming attacks efficiently as well as guide the users to cooperatively hop to the same channel with high probability. Simulation results demonstrate that the proposed algorithm can learn an optimal policy to guide the users to avoid malicious jamming more efficiently and rapidly than the random and independent Q-learning baseline algorithms,

This paper considers network protection games for a heterogeneous network system with N nodes against cyber-attackers of two different types of intentions. The first type tries to maximize damage based on the value of each net-worked node, while the second type only aims at successful infiltration. A defender, by applying defensive resources to networked nodes, can decrease those nodes' vulnerabilities. Meanwhile, the defender needs to balance the cost of using defensive resources and potential security benefits. Existing literature shows that, in a Nash equilibrium, the defender should adopt different resource allocation strategies against different types of attackers. However, it could be difficult for the defender to know the type of incoming cyber-attackers. A Bayesian game is investigated considering the case that the defender is uncertain about the attacker's type. We demonstrate that the Bayesian equilibrium defensive resource allocation strategy is a mixture of the Nash equilibrium strategies from the games against the two types of attackers separately.

Cyberspace is vulnerable to continuous malicious attacks. Traceability of network attacks is an effective defense means to curb and counter network attacks. In this paper, the evolutionary game model is used to analyze the network attack and defense behavior. On the basis of the quantification of attack and defense benefits, the replication dynamic learning mechanism is used to describe the change process of the selection probability of attack and defense strategies, and finally the evolutionary stability strategies and their solution curves of both sides are obtained. On this basis, the attack behavior is analyzed, and the probability curve of attack strategy and the optimal attack strategy are obtained, so as to realize the effective traceability of attack behavior.

At present, the research on the network security problem of underwater wireless sensors is still few, and since the underwater environment is exposed, passive security defense technology is not enough to deal with unknown security threats. Aiming at this problem, this paper proposes an offensive and defensive game model from the finite rationality of the network attack and defense sides, combined with evolutionary game theory. The replicated dynamic equation is introduced to analyze the evolution trend of strategies under different circumstances, and the selection algorithm of optimal strategy is designed, which verifies the effectiveness of this model through simulation and provides guidance for active defense technology.

In this short paper, we survey some work at the intersection of Artificial Intelligence (AI) and security that are based on game theoretic considerations, and particularly focus on the author's (our) contribution in these areas. One half of this paper focuses on applications of game theoretic and learning reasoning for addressing security applications such as in public safety and wildlife conservation. In the second half, we present recent work that attacks the learning components of these works, leading to sub-optimal defense allocation. We finally end by pointing to issues and potential research problems that can arise due to data quality in the real world.

Cyber physical system (CPS) Critical infrastructures (CIs) like the power and energy systems are increasingly becoming vulnerable to cyber attacks. Mitigating cyber risks in CIs is one of the key objectives of the design and maintenance of these systems. These CPS CIs commonly use legacy devices for remote monitoring and control where complete upgrades are uneconomical and infeasible. Therefore, risk assessment plays an important role in systematically enumerating and selectively securing vulnerable or high-risk assets through optimal investments in the cybersecurity of the CPS CIs. In this paper, we propose a CPS CI security framework and software tool, CySec Game, to be used by the CI industry and academic researchers to assess cyber risks and to optimally allocate cybersecurity investments to mitigate the risks. This framework uses attack tree, attack-defense tree, and game theory algorithms to identify high-risk targets and suggest optimal investments to mitigate the identified risks. We evaluate the efficacy of the framework using the tool by implementing a smart grid case study that shows accurate analysis and feasible implementation of the framework and the tool in this CPS CI environment.

Rational and smart decision making by means of strategic interaction and mathematical modelling is the key aspect of Game theory. Security games based on game theory are used extensively in cyberspace for various levels of security. The contemporary security issues can be modelled and analyzed using game theory as a robust mathematical framework. The attackers, defenders and the adversarial as well as defensive interactions can be captured using game theory. The security games equilibrium evaluation can help understand the attackers' strategies and potential threats at a deeper level for efficient defense. Wireless sensor network (WSN) designs are greatly benefitted by game theory. A deep learning adversarial network algorithm is used in combination with game theory enabling energy efficiency, optimal data delivery and security in a WSN. The trade-off between energy resource utilization and security is balanced using this technique.

The success of deep learning based steganography has shifted focus of researchers from traditional steganography approaches to deep learning based steganography. Various deep steganographic models have been developed for improved security, capacity and invisibility. In this work a multi-data deep learning steganography model has been developed using a well known deep learning model called Generative Adversarial Networks (GAN) more specifically using deep convolutional Generative Adversarial Networks (DCGAN). The model is capable of hiding two different messages, meant for two different receivers, inside a single cover image. The proposed model consists of four networks namely Generator, Steganalyzer Extractor1 and Extractor2 network. The Generator hides two secret messages inside one cover image which are extracted using two different extractors. The Steganalyzer network differentiates between the cover and stego images generated by the generator network. The experiment has been carried out on CelebA dataset. Two commonly used distortion metrics Peak signal-to-Noise ratio (PSNR) and Structural Similarity Index Metric (SSIM) are used for measuring the distortion in the stego image The results of experimentation show that the stego images generated have good imperceptibility and high extraction rates.

Existing anomaly detection models show success in detecting abnormal images with generative adversarial networks on the insufficient annotation of anomalous samples. However, existing models cannot accurately identify the anomaly samples which are close to the normal samples. We assume that the main reason is that these methods ignore the diversity of patterns in normal samples. To alleviate the above issue, this paper proposes a novel anomaly detection framework based on generative adversarial network, called ADe-GAN. More concretely, we construct a self-supervised learning task to fully explore the pattern information and latent representations of input images. In model inferring stage, we design a new abnormality score approach by jointly considering the pattern information and reconstruction errors to improve the performance of anomaly detection. Extensive experiments show that the ADe-GAN outperforms the state-of-the-art methods over several real-world datasets.

Generative adversarial networks (GANs) have been increasingly popular among deep learning methods. With many GANs-based models developed since its emergence, among which are conditional generative adversarial networks, progressive growing of generative adversarial networks, Wasserstein generative adversarial networks and so on. These frameworks are currently widely applied in areas such as remote sensing cybersecurity, medical, and architecture. Especially, they have solved problems of cloud removal, semantic segmentation, image-to-image translation and data argumentation in remote sensing. For example, WGANs and ProGANs can be applied in data argumentation, and cGANs can be applied in semantic argumentation and image-to-image translation. This article provides an overview of structures of multiple GANs-based models and what areas they can be applied in remote sensing.

Fuzzing is a popular technique for finding soft-ware vulnerabilities. Despite their success, the state-of-art fuzzers will inevitably produce a large number of low-quality inputs. In recent years, Machine Learning (ML) based selection strategies have reported promising results. However, the existing ML-based fuzzers are limited by the lack of training data. Because the mutation strategy of fuzzing can not effectively generate useful input, it is prohibitively expensive to collect enough inputs to train models. In this paper, propose a generative adversarial networks based solution to generate a large number of inputs to solve the problem of insufficient data. We implement the proposal in the American Fuzzy Lop (AFL), and the experimental results show that it can find more crashes at the same time compared with the original AFL.

Deep learning has been widely applied for jobs involving face inpainting, however, there are usually some problems, such as incoherent inpainting edges, lack of diversity of generated images and other problems. In order to get more feature information and improve the inpainting effect, we therefore propose a Generative Adversarial Network of Skip Connection (SCGAN), which connects the encoder layers and the decoder layers by skip connection in the generator. The coherence and consistency of the image inpainting edges are improved, and the finer features of the image inpainting are refined, simultaneously using the discriminator's local and global double discriminators model. We also employ WGAN-GP loss to enhance model stability during training, prevent model collapse, and increase the variety of inpainting face images. Finally, experiments on the CelebA dataset and the LFW dataset are performed, and the model's performance is assessed using the PSNR and SSIM indices. Our model's face image inpainting is more realistic and coherent than that of other models, and the model training is more reliable.

Speech enhancement is often applied for speech-based systems due to the proneness of speech signals to additive background noise. While speech processing-based methods are traditionally used for speech enhancement, with advancements in deep learning technologies, many efforts have been made to implement them for speech enhancement. Using deep learning, the networks learn mapping functions from noisy data to clean ones and then learn to reconstruct the clean speech signals. As a consequence, deep learning methods can reduce what is so-called musical noise that is often found in traditional speech enhancement methods. Currently, one popular deep learning architecture for speech enhancement is generative adversarial networks (GAN). However, the cross-entropy loss that is employed in GAN often causes the training to be unstable. So, in many implementations of GAN, the cross-entropy loss is replaced with the least-square loss. In this paper, to improve the training stability of GAN using cross-entropy loss, we propose to use deep regret analytic generative adversarial networks (Dragan) for speech enhancements. It is based on applying a gradient penalty on cross-entropy loss. We also employ relativistic rules to stabilize the training of GAN. Then, we applied it to the least square and Dragan losses. Our experiments suggest that the proposed method improve the quality of speech better than the least-square loss on several objective quality metrics.

In recent years, generative adversarial networks (GAN) have become a research hotspot in the field of deep learning. Researchers apply them to the field of anomaly detection and are committed to effectively and accurately identifying abnormal images in practical applications. In anomaly detection, traditional supervised learning algorithms have limitations in training with a large number of known labeled samples. Therefore, the anomaly detection model of unsupervised learning GAN is the research object for discussion and research. Firstly, the basic principles of GAN are introduced. Secondly, several typical GAN-based anomaly detection models are sorted out in detail. Then by comparing the similarities and differences of each derivative model, discuss and summarize their respective advantages, limitations and application scenarios. Finally, the problems and challenges faced by GAN in anomaly detection are discussed, and future research directions are prospected.

As one of the defensive solutions against cyberattacks, an Intrusion Detection System (IDS) plays an important role in observing the network state and alerting suspicious actions that can break down the system. There are many attempts of adopting Machine Learning (ML) in IDS to achieve high performance in intrusion detection. However, all of them necessitate a large amount of labeled data. In addition, labeling attack data is a time-consuming and expensive human-labor operation, it makes existing ML methods difficult to deploy in a new system or yields lower results due to a lack of labels on pre-trained data. To address these issues, we propose a semi-supervised IDS model that leverages Generative Adversarial Networks (GANs) and Adversarial AutoEncoder (AAE), called a semi-supervised adversarial autoencoder (SAAE). Our SAAE experimental results on two public datasets for benchmarking ML-based IDS, including NF-CSE-CIC-IDS2018 and NF-UNSW-NB15, demonstrate the effectiveness of AAE and GAN in case of using only a small number of labeled data. In particular, our approach outperforms other ML methods with the highest detection rates in spite of the scarcity of labeled data for model training, even with only 1% labeled data.

Imbalanced class distribution can cause information loss and missed/false alarms for deep learning and machine-learning algorithms. The detection performance of traditional intrusion detection systems tend to degenerate due to skewed class distribution caused by the uneven allocation of observations in different kinds of attacks. To combat class imbalance and improve network intrusion detection performance, we adopt the conditional generative adversarial network (CTGAN) that enables the generation of samples of specific classes of interest. CTGAN builds on the generative adversarial networks (GAN) architecture to model tabular data and generate high quality synthetic data by conditionally sampling rows from the generated model. Oversampling using CTGAN adds instances to the minority class such that both data in the majority and the minority class are of equal distribution. The generated security alerts are used for training classifiers that realize critical alert detection. The proposed scheme is evaluated on a real-world dataset collected from security operation center of a large enterprise. The experiment results show that detection accuracy can be substantially improved when CTGAN is adopted to produce a balanced security-alert dataset. We believe the proposed CTGAN-based approach can cast new light on building effective systems for critical alert detection with reduced missed/false alarms.

With the progress of cryptography computer science, designing cryptographic algorithms using deep learning is a very innovative research direction. Google Brain designed a communication model using generation adversarial network and explored the encrypted communication algorithm based on machine learning. However, the encrypted communication model it designed lacks quantitative evaluation. When some plaintexts and keys are leaked at the same time, the security of communication cannot be guaranteed. This model is optimized to enhance the security by adjusting the optimizer, modifying the activation function, and increasing batch normalization to improve communication speed of optimization. Experiments were performed on 16 bits and 64 bits plaintexts communication. With plaintext and key leak rate of 0.75, the decryption error rate of the decryptor is 0.01 and the attacker can't guess any valid information about the communication.

The NIST standardization process for post-quantum cryptography has been drawing the attention of researchers to the submitted candidates. One direction of research consists in implementing those candidates on embedded systems and that exposes them to physical attacks in return. The Classic McEliece cryptosystem, which is among the four finalists of round 3 in the Key Encapsulation Mechanism category, builds its security on the hardness of the syndrome decoding problem, which is a classic hard problem in code-based cryptography. This cryptosystem was recently targeted by a laser fault injection attack leading to message recovery. Regrettably, the attack setting is very restrictive and it does not tolerate any error in the faulty syndrome. Moreover, it depends on the very strong attacker model of laser fault injection, and does not apply to optimised implementations of the algorithm that make optimal usage of the machine words capacity. In this article, we propose a to change the angle and perform a message-recovery attack that relies on side-channel information only. We improve on the previously published work in several key aspects. First, we show that side-channel information, obtained with power consumption analysis, is sufficient to obtain an integer syndrome, as required by the attack framework. This is done by leveraging classic machine learning techniques that recover the Hamming weight information very accurately. Second, we put forward a computationally-efficient method, based on a simple dot product and information-set decoding algorithms, to recover the message from the, possibly inaccurate, recovered integer syndrome. Finally, we present a masking countermeasure against the proposed attack.