Biblio

Risk management is an essential part of software security. Assemblyline is a software security tool developed by the Canadian Centre for Cyber Security (CCCS) for malware detection and analysis. In this paper, we examined the performance of Assemblyline for assessing the risk of executable files. We developed and examined use-cases where Assemblyline is included as part of a security safety net assessing vulnerabilities that would lead to risk. Finally, we considered Assemblyline’s utility in a continuous integration / delivery context using our test results.

SELinux policies used in practice are generally large and complex. As a result, it is difficult for the policy writers to completely understand the policy and ensure that the policy meets the intended security goals. To remedy this, we have developed a tool called SEFlowViz that helps in visualizing the information flows of a policy and thereby helps in creating flow-secure policies. The tool uses the graph database Neo4j to visualize the policy. Along with visualization, the tool also supports extracting various information regarding the policy and its components through queries. Furthermore, the tool also supports the addition and deletion of rules which is useful in converting inconsistent policies into consistent policies.

The revolution caused by the advanced analysis features of Internet of Things and big data have made a big turnaround in the digital world. Data analysis is not only limited to collect useful data but also useful in analyzing information quickly. Therefore, most of the variants of the shared system based on the parallel structural model are explored simultaneously as the appropriate big data storage library stimulates researchers’ interest in the distributed system. Due to the emerging digital technologies, different groups such as healthcare facilities, financial institutions, e-commerce, food service and supply chain management generate a surprising amount of information. Although the process of statistical analysis is essential, it can cause significant security and privacy issues. Therefore, the analysis of data privacy protection is very important. Using the platform, technology should focus on providing Advanced Cryptography Policy (ACP). This research explores different security risks, evolutionary mechanisms and risks of privacy protection. It further recommends the post-statistical modern privacy protection act to manage data privacy protection in binary format, because it is kept confidential by the user. The user authentication program has already filed access restrictions. To maintain this purpose, everyone’s attitude is to achieve a changing identity. This article is designed to protect the privacy of users and propose a new system of restoration of controls.

In the era of Big Data, opportunities and challenges are mixed. The data transfer is increasingly frequent and speedy, and the data lifecycle is also extended, bringing more challenges to security and privacy risk governance. Currently, the common measures of risk governance covering the entire data life cycle are the data-related staff management, equipment security management, data encryption codes, data content identification and de-identification processing, etc. With the trend of data globalization, regulations fragmentation and governance technologization, “International standards”, a measure of governance combining technology and regulation, has the potential to become the best practice. However, “voluntary compliance” of international standards derogates the effectiveness of risk governance through this measure. In order to strengthen the enforcement of the international standards, the paper proposes a governance approach which is “the framework regulated by international standards, and regulations and technologies specifically implemented by national legislation.” It aims to implement the security and privacy risk governance of Big Data effectively.

Distributed training is a novel approach to accelerating training of Deep Neural Networks (DNN), but common training libraries fall short of addressing the distributed nature of heterogeneous processors or interruption by other workloads on the shared processing nodes. This paper describes distributed training of DNN on computational storage devices (CSD), which are NAND flash-based, high-capacity data storage with internal processing engines. A CSD-based distributed architecture incorporates the advantages of federated learning in terms of performance scalability, resiliency, and data privacy by eliminating the unnecessary data movement between the storage device and the host processor. The paper also describes Stannis, a DNN training framework that improves on the shortcomings of existing distributed training frameworks by dynamically tuning the training hyperparameters in heterogeneous systems to maintain the maximum overall processing speed in term of processed images per second and energy efficiency. Experimental results on image classification training benchmarks show up to 3.1x improvement in performance and 2.45x reduction in energy consumption when using Stannis plus CSD compare to the generic systems.

Quantized deep neural networks (QDNNs) are necessary for low-power, high throughput, and embedded applications. Previous studies mostly focused on developing optimization methods for the quantization of given models. However, quantization sensitivity depends on the model architecture. Also, the characteristics of weight and activation quantization are quite different. This study proposes a holistic approach for the optimization of QDNNs, which contains QDNN training methods as well as quantization-friendly architecture design. Synthesized data is used to visualize the effects of weight and activation quantization. The results indicate that deeper models are more prone to activation quantization, while wider models improve the resiliency to both weight and activation quantization.

3D reconstruction has piqued the interest of many disciplines, and many researchers have spent the last decade striving to improve on latest automated three-dimensional reconstruction systems. Three Dimensional models can be utilized to tackle a wide range of visualization problems as well as other activities. In this paper, we have implemented a method of Digital Surface Map (DSM) generation from Aerial images using Conditional Generative Adversarial Networks (c-GAN). We have used Seg-net architecture of Convolutional Neural Network (CNN) to segment the aerial images and then the U-net generator of c-GAN generates final DSM. The dataset we used is ISPRS Potsdam-Vaihingen dataset. We also review different stages if 3D reconstruction and how Deep learning is now being widely used to enhance the process of 3D data generation. We provide binary cross entropy loss function graph to demonstrate stability of GAN and CNN. The purpose of our approach is to solve problem of DSM generation using Deep learning techniques. We put forth our method against other latest methods of DSM generation such as Semi-global Matching (SGM) and infer the pros and cons of our approach. Finally, we suggest improvements in our methods that might be useful in increasing the accuracy.

We are interested in the design of generative networks. The training of these mathematical structures is mostly performed with the help of adversarial (min-max) optimization problems. We propose a simple methodology for constructing such problems assuring, at the same time, consistency of the corresponding solution. We give characteristic examples developed by our method, some of which can be recognized from other applications, and some are introduced here for the first time. We present a new metric, the likelihood ratio, that can be employed online to examine the convergence and stability during the training of different Generative Adversarial Networks (GANs). Finally, we compare various possibilities by applying them to well-known datasets using neural networks of different configurations and sizes.

Many portable imaging devices use the operation of "trunc" (rounding towards zero) instead of rounding as the final quantizer for computing DCT coefficients during JPEG compression. We show that this has rather profound consequences for steganography and its detection. In particular, side-informed steganography needs to be redesigned due to the different nature of the rounding error. The steganographic algorithm J-UNIWARD becomes vulnerable to steganalysis with the JPEG rich model and needs to be adjusted for this source. Steganalysis detectors need to be retrained since a steganalyst unaware of the existence of the trunc quantizer will experience 100% false alarm.

Data exfiltration is of increasing concern throughout the world. The number of incidents and capabilities of data exfiltration attacks are growing at an unprecedented rate. However, such attack vectors have not been deeply explored in the literature. This paper aims to address this gap by implementing a data exfiltration methodology, detailing some data exfiltration methods. Groups of exfiltration methods are incorporated into a program that can act as a testbed for owners of any network that stores sensitive data. The implemented methods are tested against the well-known network intrusion detection system Snort, where all of them have been successfully evaded detection by its community rule sets. Thus, in this paper, we have developed new countermeasures to prevent and detect data exfiltration attempts using these methods.

SUMMARY & CONCLUSIONSA Multi-access Edge Computing (MEC) micro data center (MEDC) consists of multiple MEC hosts close to endpoint devices. MEC service is delivered by instantiating a virtualization system (e.g., Virtual Machines or Containers) on a MEC host. MEDC faces more new security risks due to various device connections in an open environment. When more and more IoT/CPS systems are connected to MEDC, it is necessary for MEC service providers to quantitatively analyze any security loss and then make defense-related decision. This paper develops a CTMC model for quantitatively analyzing the security and dependability of a vulnerable MEDC system under lateral movement attacks, from the adversary’s initial successful access until the MEDC becomes resistant to the attack. The proposed model captures the behavior of the system in a scenario where (i) the rate of vulnerable MEC servers being infected increases with the increasing number of infected MEC servers, (ii) each infected MEC server can perform its compromising activity independently and randomly, and (iii) any infected MEC may fail and then cannot provide service. We also introduce the formulas for computing metrics. The proposed model and formula are verified to be approximately accurate by comparing numerical results and simulation results.

Intrusion detection is a traditional practice of security experts, however, there are several issues which still need to be tackled. Therefore, in this paper, after highlighting these issues, we present an architecture for a hybrid Intrusion Detection System (IDS) for an adaptive and incremental detection of both known and unknown attacks. The IDS is composed of supervised and unsupervised modules, namely, a Deep Neural Network (DNN) and the K-Nearest Neighbors (KNN) algorithm, respectively. The proposed system is near-autonomous since the intervention of the expert is minimized through the active learning (AL) approach. A query strategy for the labeling process is presented, it aims at teaching the supervised module to detect unknown attacks and improve the detection of the already-known attacks. This teaching is achieved through sliding windows (SW) in an incremental fashion where the DNN is retrained when the data is available over time, thus rendering the IDS adaptive to cope with the evolutionary aspect of the network traffic. A set of experiments was conducted on the CICIDS2017 dataset in order to evaluate the performance of the IDS, promising results were obtained.

As malware detection algorithms and methods become more sophisticated, malware authors adopt equally sophisticated evasion mechanisms to defeat them. Anecdotal evidence claims Living-Off-The-Land (LotL) techniques are one of the major evasion techniques used in many malware attacks. These techniques leverage binaries already present in the system to conduct malicious actions. We present the first large-scale systematic investigation of the use of these techniques by malware on Windows systems.In this paper, we analyse how common the use of these native system binaries is across several malware datasets, containing a total of 31,805,549 samples. We identify an average 9.41% prevalence. Our results show that the use of LotL techniques is prolific, particularly in Advanced Persistent Threat (APT) malware samples where the prevalence is 26.26%, over twice that of commodity malware.To illustrate the evasive potential of LotL techniques, we test the usage of LotL techniques against several fully patched Windows systems in a local sandboxed environment and show that there is a generalised detection gap in 10 of the most popular anti-virus products.

Windows Communication Foundation (WCF) is a communication framework for building connected, service-oriented applications, initially released by Microsoft as part of.NET Framework, but now open source. The WCF message-based communication is a very popular solution used for sending asynchronous messages from one service endpoint to another. Because WCF provides many functionalities it has a large-consuming development model and often the security measures implemented in applications are not proper. In this study we propose a methodology for offensive security analysis of an WCF endpoint or service, from red team perspective. A step by step approach, empirical information, and detailed analysis report of WCF vulnerabilities are presented. We conclude by proposing recommendations for mitigating attacks and securing endpoints.

With the rapid development of power Internet of Things (IoT), the ubiquitous edge agents are frequently exposed in a risky environment, where the white-box attacker could steal all the internal information by full observation of dynamic execution of the cryptographic software. In this situation, a new table-based white-box cryptography implementation of SM4 algorithm is proposed to prevent the attacker from extracting the secret key, which hides the encryption and decryption process in obfuscated lookup tables. Aiming to improve the diversity and ambiguity of the lookup tables as well as resist different types of white-box attacks, the random bijective nonlinear mappings are applied as scrambling encodings of the lookup tables. Moreover, in order to make our implementation more practical in the resource-constrained edge IoT agent, elaborate design is proposed to make some tables reusability, leading to less memory occupation while guaranteeing the security. The validity and security of the proposed implementation will be illustrated through several evaluation indicators.

This paper explores a practical approach to securing large wireless networks by applying Physical Layer Security (PLS). To date, PLS has mostly been seen as an information theory concept with few practical implementations. We present an Access Point (AP) selection algorithm that uses PLS to find an AP that offers the highest secrecy capacity to a legitimate user. We then propose an implementation of this algorithm using the novel concept of spectrum programming which extends Software-Defined Networking to the physical and data-link layers and makes wireless network management and control more flexible and scalable than traditional platforms. Our Wi-Fi network evaluation results show that our approach outperforms conventional solutions in terms of security, but at the expense of communication capacity, thus identifying a trade-off between security and performance. These results encourage implementation and extension to further wireless technologies.

Considered is a network of parallel wireless channels in which individual parties are engaged in secret communication under the protection of cooperative jamming. A strategic eavesdropper selects the most vulnerable channels to attack. Existing works usually suggest the defender allocate limited cooperative jamming power to various channels. However, it usually requires some strong assumptions and complex computation to find such an optimal power control policy. This paper proposes a probabilistic cooperative jamming scheme such that the defender focuses on protecting randomly selected channels. Two different cases regarding each channel’s eavesdropping capacity are discussed. The first case studies the general scenario where each channel has different eavesdropping capacity. The second case analyzes an extreme scenario where all channels have the same eavesdropping capacity. Two non-zero-sum Nash games model the competition between the network defender and an eavesdropper in each case. Furthermore, considering the case that the defender does not know the eavesdropper’s channel state information (CSI) leads to a Bayesian game. For all three games, we derive conditions for the existence of a unique Nash equilibrium (NE), and obtain the equilibria and the value functions in closed form.

Protecting and preventing sensitive data from being used inappropriately has become a challenging task. Even a small mistake in securing data can be exploited by phishing attacks to release private information such as passwords or financial information to a malicious actor. Phishing has now proven so successful, it is the number one attack vector. Many approaches have been proposed to protect against this type of cyber-attack, from additional staff training, enriched spam filters to large collaborative databases of known threats such as PhishTank and OpenPhish. However, they mostly rely upon a user falling victim to an attack and manually adding this new threat to the shared pool, which presents a constant disadvantage in the fight back against phishing. In this paper, we propose a novel approach to protect against phishing attacks using binary visualisation and machine learning. Unlike previous work in this field, our approach uses an automated detection process and requires no further user interaction, which allows faster and more accurate detection process. The experiment results show that our approach has high detection rate.

Currently and particularly with remote working scenarios during COVID-19, phishing attack has become one of the most significant threats faced by internet users, organizations, and service providers. In a phishing attack, the attacker tries to steal client sensitive data (such as login, passwords, and credit card details) using spoofed emails and fake websites. Cybercriminals, hacktivists, and nation-state spy agencies have now got a fertilized ground to deploy their latest innovative phishing attacks. Timely detection of phishing attacks has become most crucial than ever. Machine learning algorithms can be used to accurately detect phishing attacks before a user is harmed. This paper presents a novel ensemble model to detect phishing attacks on the website. We select three machine learning classifiers: Artificial Neural Network (ANN), K-Nearest Neighbors (KNN), and Decision Tree (C4.5) to use in an ensemble method with Random Forest Classifier (RFC). This ensemble method effectively detects website phishing attacks with better accuracy than existing studies. Experimental results demonstrate that the ensemble of KNN and RFC detects phishing attacks with 97.33% accuracy.

Phishing URL is a type of cyberattack, based on falsified URLs. The number of phishing URL attacks continues to increase despite cybersecurity efforts. According to the Anti-Phishing Working Group (APWG), the number of phishing websites observed in 2020 is 1 520 832, doubling over the course of a year. Various algorithms, techniques and methods can be used to build models for phishing URL detection and classification. From our reading, we observed that Machine Learning (ML) is one of the recent approaches used to detect and classify phishing URL in an efficient and proactive way. In this paper, we evaluate eleven of the most adopted ML algorithms such as Decision Tree (DT), Nearest Neighbours (KNN), Gradient Boosting (GB), Logistic Regression (LR), Naïve Bayes (NB), Random Forest (RF), Support Vector Machines (SVM), Neural Network (NN), Ex-tra\_Tree (ET), Ada\_Boost (AB) and Bagging (B). To do that, we compute detection accuracy metric for each algorithm and we use lexical analysis to extract the URL features.

Phishing is a frequent assault in which unsuspecting people’s unique, private, and sensitive information is stolen through fake websites. The primary objective of phishing websites’consistent resource allocators isto steal unique, private, and sensitive information such as user login passwords and online financial transactions. Phishers construct phony websites that look and sound just like genuine things. With the advent of technology, there are protecting users significantly increased in phishing methods. It necessitates the development of an anti-phishing technology to identify phishing and protect users. Machine learning is a useful technique for combating phishing attempts. These articles were utilized to examine Machine learning for detection strategies and characteristics.

Website phishing continues to persist as one of the most important security threats of the modern Internet era. A major concern has been that machine learning based approaches, which have been the cornerstones of deployed phishing detection solutions, have not been able to adapt to the evolving nature of the phishing attacks. To create updated machine learning models, the collection of a sufficient corpus of real-time phishing data has always been a challenging problem as most phishing websites are short-lived. In this work, for the first time, we address these important concerns and describe an adaptive phishing detection solution that is able to adapt to changes in phishing attacks. Our solution has two major contributions. First, our solution allows for multiple organizations to collaborate in a privacy preserving manner and generate a robust machine learning model for phishing detection. Second, our solution is designed to be flexible in order to adapt to the novel phishing features introduced by attackers. Our solution not only allows for incorporating novel features into the existing machine learning model, but also can help, to a certain extent, the “unlearning” of existing features that have become obsolete in current phishing attacks. We evaluated our approach on a large real-world data collected over a period of six months. Our results achieve a high true positive rate of 97 %, which is on par with existing state-of-the art centralized solutions. Importantly, our results demonstrate that, a machine learning model can incorporate new features while selectively “unlearning” the older obsolete features.

The use of AlNiCo alloys as the low coercive force (LCF) magnet in Variable Flux Memory Machines has been largely discussed in the literature, but similar magnetic materials as FeCrCo are still little explored. This paper proposes the study of a standstill magnetization strategy of a Variable Flux Memory Machine composed by a FeCrCo-based cylindrical rotor. An inverter in DC/DC mode is proposed for injecting short-time currents along the magnetization axis aiming the regulation of the magnetization state of the FeCrCo. A methodology for validating results obtained is defined from the estimation of the remanence and the excitation field characterizing the behavior of the internal recoil lines of the magnet used in the rotor. A study of the armature reaction affecting the machine when q-axis currents supply the machine is proposed by simulation.

Many techniques are available nowadays, for Intellectual Property(IP) protection of Digital circuits. Out of these techniques, the popular one is watermarking. Similar to the watermarking used in case of text, image and video, watermarking of digital circuits also modifies a digital circuit design in such a way, that only the IP owner of design is able to extract the watermark form the design. In this paper, Multi – Feedback configuration of Linear Feedback Shift Register(LFSR) is used to watermark a FSM based design. This watermarking technique improves the watermark strength of already existing LFSR based watermarking technique. In terms of hardware utilization, it is significantly efficient than some popular watermarking techniques. The proposed technique has been implemented using Verilog HDL in Xilinx ISE and the simulation is done using ModelSim.

Most of the existing identity systems are built on top of centralized storage systems. Storing identity data on these types of centralized storage platforms(e.g cloud storage, central servers) becomes a major privacy concern since various types of attacks and data breaches can happen. With this research, we are proposing blockchain and self-sovereign identity based digital identity (KYC - Know Your Customer) platform “Casper” to address the issues on centralized identity systems. “Casper ” is an Android/iOS based mobile identity wallet application that combines the integration of blockchain and a self-sovereign identity-based approach. Unlike centralized identity systems, the actual identities of the customer/users are stored in the customers’ mobile wallet application. The proof of these identities is stored in the blockchain-based decentralized storage as a self-sovereign identity proof. Casper platforms’ Self-Sovereign Identity(SSI)-based system provides a Zero Knowledge Proof(ZKP) mechanism to verify the identity information. Casper platform can be adopted in various domains such as healthcare, banking, government organization etc. As a use case, we have discussed building a digital identity wallet for banking customers with the Casper platform. Casper provides a secure, decentralized and ZKP verifiable identity by using blockchain and SSI based approach. It addresses the common issues in centralized/cloud-based identity systems platforms such as the lack of data immutability, lack of traceability, centralized control etc.