Biblio

At present, our life is almost inseparable from the network, the network provides a lot of convenience for our life. However, a variety of network security incidents occur very frequently. In recent years, with the continuous development of neural network technology, more and more researchers have applied neural network to intrusion detection, which has developed into a new research direction in intrusion detection. As long as the neural network is provided with input data including network data packets, through the process of self-learning, the neural network can separate abnormal data features and effectively detect abnormal data. Therefore, the article innovatively proposes an intrusion detection method based on deep convolutional neural networks (CNN), which is used to test on public data sets. The results show that the model has a higher accuracy rate and a lower false negative rate than traditional intrusion detection methods.

The Internet of Things (IoT) is rapidly evolving, while introducing several new challenges regarding security, resilience and operational assurance. In the face of an increasing attack landscape, it is necessary to cater for the provision of efficient mechanisms to collectively detect sophisticated malware resulting in undesirable (run-time) device and network modifications. This is not an easy task considering the dynamic and heterogeneous nature of IoT environments; i.e., different operating systems, varied connected networks and a wide gamut of underlying protocols and devices. Malicious IoT nodes or gateways can potentially lead to the compromise of the whole IoT network infrastructure. On the other hand, the SDN control plane has the capability to be orchestrated towards providing enhanced security services to all layers of the IoT networking stack. In this paper, we propose an SDN-enabled control plane based orchestration that leverages emerging Long Short-Term Memory (LSTM) classification models; a Deep Learning (DL) based architecture to combat malicious IoT nodes. It is a first step towards a new line of security mechanisms that enables the provision of scalable AI-based intrusion detection focusing on the operational assurance of only those specific, critical infrastructure components,thus, allowing for a much more efficient security solution. The proposed mechanism has been evaluated with current state of the art datasets (i.e., N\_BaIoT 2018) using standard performance evaluation metrics. Our preliminary results show an outstanding detection accuracy (i.e., 99.9%) which significantly outperforms state-of-the-art approaches. Based on our findings, we posit open issues and challenges, and discuss possible ways to address them, so that security does not hinder the deployment of intelligent IoT-based computing systems.

As neural networks make their way into safety-critical systems, where misbehavior can lead to catastrophes, there is a growing interest in certifying the equivalence of two structurally similar neural networks - a problem known as differential verification. For example, compression techniques are often used in practice for deploying trained neural networks on computationally- and energy-constrained devices, which raises the question of how faithfully the compressed network mimics the original network. Unfortunately, existing methods either focus on verifying a single network or rely on loose approximations to prove the equivalence of two networks. Due to overly conservative approximation, differential verification lacks scalability in terms of both accuracy and computational cost. To overcome these problems, we propose NEURODIFF, a symbolic and fine-grained approximation technique that drastically increases the accuracy of differential verification on feed-forward ReLU networks while achieving many orders-of-magnitude speedup. NEURODIFF has two key contributions. The first one is new convex approximations that more accurately bound the difference of two networks under all possible inputs. The second one is judicious use of symbolic variables to represent neurons whose difference bounds have accumulated significant error. We find that these two techniques are complementary, i.e., when combined, the benefit is greater than the sum of their individual benefits. We have evaluated NEURODIFF on a variety of differential verification tasks. Our results show that NEURODIFF is up to 1000X faster and 5X more accurate than the state-of-the-art tool.

The deep learning-based compressive Sensing (CS) has shown substantial improved performance and in run-time reduction with signal sampling and reconstruction. In most cases, moreover, these techniques suffer from disrupting artefacts or high-frequency contents at low sampling ratios. Similarly, this occurs in the multi-resolution sampling method, which further collects more components with lower frequencies. A promising innovation combining CS with convolutionary neural network has eliminated the sparsity constraint yet recovery persists slow. We propose a Deep wavelet based compressive sensing with multi-resolution framework provides better improvement in reconstruction as well as run time. The proposed model demonstrates outstanding quality on test functions over previous approaches.

We propose an Artificial Intelligence (AI)/Deep Learning (DL)-based image analysis framework for hardware assurance of digital integrated circuits (ICs). Our aim is to examine and verify various hardware information from analyzing the Scanning Electron Microscope (SEM) images of an IC. In our proposed framework, we apply DL-based methods at all essential steps of the analysis. To the best of our knowledge, this is the first such framework that makes heavy use of DL-based methods at all essential analysis steps. Further, to reduce time and effort required in model re-training, we propose and demonstrate various automated or semi-automated training data preparation methods and demonstrate the effectiveness of using synthetic data to train a model. By applying our proposed framework to analyzing a set of SEM images of a large digital IC, we prove its efficacy. Our DL-based methods are fast, accurate, robust against noise, and can automate tasks that were previously performed mainly manually. Overall, we show that DL-based methods can largely increase the level of automation in hardware assurance of digital ICs and improve its accuracy.

This letter presents a novel median filtering forensics approach, based on a convolutional neural network (CNN) with an adaptive filtering layer (AFL), which is built in the discrete cosine transform (DCT) domain. Using the proposed AFL, the CNN can determine the main frequency range closely related with the operational traces. Then, to automatically learn the multi-scale manipulation features, a multi-scale convolutional block is developed, exploring a new multi-scale feature fusion strategy based on the maxout function. The resultant features are further processed by a convolutional stream with pooling and batch normalization operations, and finally fed into the classification layer with the Softmax function. Experimental results show that our proposed approach is able to accurately detect the median filtering manipulation and outperforms the state-of-the-art schemes, especially in the scenarios of low image resolution and serious compression loss.

In this paper, we introduce a new digital image forensics approach called forensic similarity, which determines whether two image patches contain the same forensic trace or different forensic traces. One benefit of this approach is that prior knowledge, e.g., training samples, of a forensic trace is not required to make a forensic similarity decision on it in the future. To do this, we propose a two-part deep-learning system composed of a convolutional neural network-based feature extractor and a three-layer neural network, called the similarity network. This system maps the pairs of image patches to a score indicating whether they contain the same or different forensic traces. We evaluated the system accuracy of determining whether two image patches were captured by the same or different camera model and manipulated by the same or a different editing operation and the same or a different manipulation parameter, given a particular editing operation. Experiments demonstrate applicability to a variety of forensic traces and importantly show efficacy on “unknown” forensic traces that were not used to train the system. Experiments also show that the proposed system significantly improves upon prior art, reducing error rates by more than half. Furthermore, we demonstrated the utility of the forensic similarity approach in two practical applications: forgery detection and localization, and database consistency verification.

In image forensics, to determine whether the image is impurely transformed, it extracts and examines the features included in the suspicious image. In general, the features extracted for the detection of forgery images are based on numerical values, so it is somewhat unreasonable to use in the CNN structure for image classification. In this paper, the extraction method of a feature vector is using a least-squares solution. Treat a suspicious image like a matrix and its solution to be coefficients as the feature vector. Get two solutions from two images of the original and its median filter residual (MFR). Subsequently, the two features were formed into a visualized pattern and then fed into CNN deep learning to classify the various transformed images. A new structure of the CNN net layer was also designed by hybrid with the inception module and the residual block to classify visualized feature vector patterns. The performance of the proposed image forensics detection (IFD) scheme was measured with the seven transformed types of image: average filtered (window size: 3 × 3), gaussian filtered (window size: 3 × 3), JPEG compressed (quality factor: 90, 70), median filtered (window size: 3 × 3, 5 × 5), and unaltered. The visualized patterns are fed into the image input layer of the designed CNN hybrid model. Throughout the experiment, the accuracy of median filtering detection was 98% over. Also, the area under the curve (AUC) by sensitivity (TP: true positive rate) and 1-specificity (FP: false positive rate) results of the proposed IFD scheme approached to `1' on the designed CNN hybrid model. Experimental results show high efficiency and performance to classify the various transformed images. Therefore, the grade evaluation of the proposed scheme is “Excellent (A)”.

With the rapid progress in recent years, techniques that generate and manipulate multimedia content can now provide a very advanced level of realism. The boundary between real and synthetic media has become very thin. On the one hand, this opens the door to a series of exciting applications in different fields such as creative arts, advertising, film production, and video games. On the other hand, it poses enormous security threats. Software packages freely available on the web allow any individual, without special skills, to create very realistic fake images and videos. These can be used to manipulate public opinion during elections, commit fraud, discredit or blackmail people. Therefore, there is an urgent need for automated tools capable of detecting false multimedia content and avoiding the spread of dangerous false information. This review paper aims to present an analysis of the methods for visual media integrity verification, that is, the detection of manipulated images and videos. Special emphasis will be placed on the emerging phenomenon of deepfakes, fake media created through deep learning tools, and on modern data-driven forensic methods to fight them. The analysis will help highlight the limits of current forensic tools, the most relevant issues, the upcoming challenges, and suggest future directions for research.

The fast growth of encrypted traffic puts forward burning requirements on the efficiency of traffic classification. Although deep learning models perform well in the classification, they sacrifice the efficiency to obtain high-precision results. To reduce the resource and time consumption, a novel and lightweight model is proposed in this paper. Our design principle is to “maximize the reuse of thin modules”. A thin module adopts the multi-head attention and the 1D convolutional network. Attributed to the one-step interaction of all packets and the parallelized computation of the multi-head attention mechanism, a key advantage of our model is that the number of parameters and running time are significantly reduced. In addition, the effectiveness and efficiency of 1D convolutional networks are proved in traffic classification. Besides, the proposed model can work well in a real time manner, since only three consecutive packets of a flow are needed. To improve the stability of the model, the designed network is trained with the aid of ResNet, layer normalization and learning rate warmup. The proposed model outperforms the state-of-the-art works based on deep learning on two public datasets. The results show that our model has higher accuracy and running efficiency, while the number of parameters used is 1.8% of the 1D convolutional network and the training time halves.

This paper examines multiple machine learning models to find the model that best indicates anomalous activity in an industrial control system that is under a software-based attack. The researched machine learning models are Random Forest, Gradient Boosting Machine, Artificial Neural Network, and Recurrent Neural Network classifiers built-in Python and tested against the HIL-based Augmented ICS dataset. Although the results showed that Random Forest, Gradient Boosting Machine, Artificial Neural Network, and Long Short-Term Memory classification models have great potential for anomaly detection in industrial control systems, we found that Random Forest with tuned hyperparameters slightly outperformed the other models.

Deep learning methods are increasingly becoming solutions to complex problems, including the search for anomalies. While fully-connected and convolutional neural networks have already found their application in classification problems, their applicability to the problem of detecting anomalies is limited. In this regard, it is proposed to use autoencoders, previously used only in problems of reducing the dimension and removing noise, as a method for detecting anomalies in the industrial control system. A new method based on autoencoders is proposed for detecting anomalies in the operation of industrial control systems (ICS). Several neural networks based on auto-encoders with different architectures were trained, and the effectiveness of each of them in the problem of detecting anomalies in the work of process control systems was evaluated. Auto-encoders can detect the most complex and non-linear dependencies in the data, and as a result, can show the best quality for detecting anomalies. In some cases, auto-encoders require fewer machine resources.

This paper proposes AERFAD, an anomaly detection method based on the autoencoder and the random forest, for solving the credit card fraud detection problem. The proposed AERFAD first utilizes the autoencoder to reduce the dimensionality of data and then uses the random forest to classify data as anomalous or normal. Large numbers of credit card transaction data of European cardholders are applied to AEFRAD to detect possible frauds for the sake of performance evaluation. When compared with related methods, AERFAD has relatively excellent performance in terms of the accuracy, true positive rate, true negative rate, and Matthews correlation coefficient.

Over the last few years, there has been an increasing number of studies about facial emotion recognition because of the importance and the impact that it has in the interaction of humans with computers. With the growing number of challenging datasets, the application of deep learning techniques have all become necessary. In this paper, we study the challenges of Emotion Recognition Datasets and we also try different parameters and architectures of the Conventional Neural Networks (CNNs) in order to detect the seven emotions in human faces, such as: anger, fear, disgust, contempt, happiness, sadness and surprise. We have chosen iCV MEFED (Multi-Emotion Facial Expression Dataset) as the main dataset for our study, which is relatively new, interesting and very challenging.

In this study, it was aimed to recognize the emotional state from facial images using the deep learning method. In the study, which was approved by the ethics committee, a custom data set was created using videos taken from 20 male and 20 female participants while simulating 7 different facial expressions (happy, sad, surprised, angry, disgusted, scared, and neutral). Firstly, obtained videos were divided into image frames, and then face images were segmented using the Haar library from image frames. The size of the custom data set obtained after the image preprocessing is more than 25 thousand images. The proposed convolutional neural network (CNN) architecture which is mimics of LeNet architecture has been trained with this custom dataset. According to the proposed CNN architecture experiment results, the training loss was found as 0.0115, the training accuracy was found as 99.62%, the validation loss was 0.0109, and the validation accuracy was 99.71%.

Image Retrieval (IR) has become one of the main problems facing computer society recently. To increase computing similarities between images, hashing approaches have become the focus of many programmers. Indeed, in the past few years, Deep Learning (DL) has been considered as a backbone for image analysis using Convolutional Neural Networks (CNNs). This paper aims to design and implement a high-performance image classifier that can be used in several applications such as intelligent vehicles, face recognition, marketing, and many others. This work considers experimentation to find the sequential model's best configuration for classifying images. The best performance has been obtained from two layers' architecture; the first layer consists of 128 nodes, and the second layer is composed of 32 nodes, where the accuracy reached up to 0.9012. The proposed classifier has been achieved using CNN and the data extracted from the CIFAR-10 dataset by the inception model, which are called the Transfer Values (TRVs). Indeed, the Particle Swarm Optimization (PSO) algorithm is used to reduce the TRVs. In this respect, the work focus is to reduce the TRVs to obtain high-performance image classifier models. Indeed, the PSO algorithm has been enhanced by using the crossover technique from genetic algorithms. This led to a reduction of the complexity of models in terms of the number of parameters used and the execution time.

Nowadays, with the increasing use of computers and the Internet, more people are exposed to cyber-security dangers. According to antivirus companies, malware is one of the most common threats of using the Internet. Therefore, providing a practical solution is critical. Current methods use machine learning approaches to classify malware samples automatically. Despite the success of these approaches, the accuracy and efficiency of these techniques are still inadequate, especially for multiple class classification problems and imbalanced training data sets. To mitigate this problem, we use deep learning-based algorithms for classification and generation of new malware samples. Our model is based on the opcode sequences, which are given to the model without any pre-processing. Besides, we use a novel generative adversarial network to generate new opcode sequences for oversampling minority classes. Also, we propose the model that is a combination of Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM) to classify malware samples. CNN is used to consider short-term dependency between features; while, LSTM is used to consider longer-term dependence. The experiment results show our method could classify malware to their corresponding family effectively. Our model achieves 98.99% validation accuracy.

Generative Adversarial Network (GAN) and its variants have shown promising results in generating synthetic data. However, the issues with GANs are: (i) the learning happens around the training samples and the model often ends up remembering them, consequently, compromising the privacy of individual samples - this becomes a major concern when GANs are applied to training data including personally identifiable information, (ii) the randomness in generated data - there is no control over the specificity of generated samples. To address these issues, we propose imdpGAN-an information maximizing differentially private Generative Adversarial Network. It is an end-to-end framework that simultaneously achieves privacy protection and learns latent representations. With experiments on MNIST dataset, we show that imdpGAN preserves the privacy of the individual data point, and learns latent codes to control the specificity of the generated samples. We perform binary classification on digit pairs to show the utility versus privacy trade-off. The classification accuracy decreases as we increase privacy levels in the framework. We also experimentally show that the training process of imdpGAN is stable but experience a 10-fold time increase as compared with other GAN frameworks. Finally, we extend imdpGAN framework to CelebA dataset to show how the privacy and learned representations can be used to control the specificity of the output.

Modern industrial control systems (ICS) act as victims of cyber attacks more often in last years. These cyber attacks often can not be detected by classical information security methods. Moreover, the consequences of cyber attack's impact can be catastrophic. Since cyber attacks leads to appearance of anomalies in the ICS and technological equipment controlled by it, the task of intrusion detection for ICS can be reformulated as the task of industrial process anomaly detection. This paper considers the applicability of generative adversarial networks (GANs) in the field of industrial processes anomaly detection. Existing approaches for GANs usage in the field of information security (such as anomaly detection in network traffic) were described. It is proposed to use the BiGAN architecture in order to detect anomalies in the industrial processes. The proposed approach has been tested on Secure Water Treatment Dataset (SWaT). The obtained results indicate the prospects of using the examined method in practice.

Millimeter Wave (mmWave) band can be a solution to serve the vast number of Internet of Things (IoT) and Vehicle to Everything (V2X) devices. In this context, Cognitive Radio (CR) is capable of managing the mmWave spectrum sharing efficiently. However, Cognitive mmWave Radios are vulnerable to malicious users due to the complex dynamic radio environment and the shared access medium. This indicates the necessity to implement techniques able to detect precisely any anomalous behaviour in the spectrum to build secure and efficient radios. In this work, we propose a comparison framework between deep generative models: Conditional Generative Adversarial Network (C-GAN), Auxiliary Classifier Generative Adversarial Network (AC-GAN), and Variational Auto Encoder (VAE) used to detect anomalies inside the dynamic radio spectrum. For the sake of the evaluation, a real mmWave dataset is used, and results show that all of the models achieve high probability in detecting spectrum anomalies. Especially, AC-GAN that outperforms C-GAN and VAE in terms of accuracy and probability of detection.

Deep Learning (DL), in spite of its huge success in many new fields, is extremely vulnerable to adversarial attacks. We demonstrate how an attacker applies physical white-box and black-box adversarial attacks to Channel decoding systems based on DL. We show that these attacks can affect the systems and decrease performance. We uncover that these attacks are more effective than conventional jamming attacks. Additionally, we show that classical decoding schemes are more robust than the deep learning channel decoding systems in the presence of both adversarial and jamming attacks.

The Software Defined Network (SDN) provides higher programmable functionality for network configuration and management dynamically. Moreover, SDN introduces a centralized management approach by dividing the network into control and data planes. In this paper, we introduce a deep learning enabled intrusion detection and prevention system (DL-IDPS) to prevent secure shell (SSH) brute-force attacks and distributed denial-of-service (DDoS) attacks in SDN. The packet length in SDN switch has been collected as a sequence for deep learning models to identify anomalous and malicious packets. Four deep learning models, including Multilayer Perceptron (MLP), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM) and Stacked Auto-encoder (SAE), are implemented and compared for the proposed DL-IDPS. The experimental results show that the proposed MLP based DL-IDPS has the highest accuracy which can achieve nearly 99% and 100% accuracy to prevent SSH Brute-force and DDoS attacks, respectively.

Network traffic anomaly detection is of critical importance in cybersecurity due to the massive and rapid growth of sophisticated computer network attacks. Indeed, the more new Internet-related technologies are created, the more elaborate the attacks become. Among all the contemporary high-level attacks, dictionary-based brute-force attacks (BFA) present one of the most unsurmountable challenges. We need to develop effective methods to detect and mitigate such brute-force attacks in realtime. In this paper, we investigate SSH and FTP brute-force attack detection by using the Long Short-Term Memory (LSTM) deep learning approach. Additionally, we made use of machine learning (ML) classifiers: J48, naive Bayes (NB), decision table (DT), random forest (RF) and k-nearest-neighbor (k-NN), for additional detection purposes. We used the well-known labelled dataset CICIDS2017. We evaluated the effectiveness of the LSTM and ML algorithms, and compared their performance. Our results show that the LSTM model outperforms the ML algorithms, with an accuracy of 99.88%.

Today our world benefits from Internet of Things (IoT) technology; however, new security problems arise when these IoT devices are introduced into our homes. Because many of these IoT devices have access to the Internet and they have little to no security, they make our smart homes highly vulnerable to compromise. Some of the threats include IoT botnets and generic confidentiality, integrity, and availability (CIA) attacks. Our research explores botnet detection by experimenting with supervised machine learning and deep-learning classifiers. Further, our approach assesses classifier performance on unbalanced datasets that contain benign data, mixed in with small amounts of malicious data. We demonstrate that the classifiers can separate malicious activity from benign activity within a small IoT network dataset. The classifiers can also separate malicious activity from benign activity in increasingly larger datasets. Our experiments have demonstrated incremental improvement in results for (1) accuracy, (2) probability of detection, and (3) probability of false alarm. The best performance results include 99.9% accuracy, 99.8% probability of detection, and 0% probability of false alarm. This paper also demonstrates how the performance of these classifiers increases, as IoT training datasets become larger and larger.

Android, being the most widespread mobile operating systems is increasingly becoming a target for malware. Malicious apps designed to turn mobile devices into bots that may form part of a larger botnet have become quite common, thus posing a serious threat. This calls for more effective methods to detect botnets on the Android platform. Hence, in this paper, we present a deep learning approach for Android botnet detection based on Convolutional Neural Networks (CNN). Our proposed botnet detection system is implemented as a CNN-based model that is trained on 342 static app features to distinguish between botnet apps and normal apps. The trained botnet detection model was evaluated on a set of 6,802 real applications containing 1,929 botnets from the publicly available ISCX botnet dataset. The results show that our CNN-based approach had the highest overall prediction accuracy compared to other popular machine learning classifiers. Furthermore, the performance results observed from our model were better than those reported in previous studies on machine learning based Android botnet detection.