Biblio

This paper presents the details of the roving proxy framework for SMS spam and SMS phishing (SMishing) detection. The framework aims to protect organizations and enterprises from the danger of SMishing attacks. Feasibility and functionality studies of the framework are presented along with an update process study to define the minimum requirements for the system to adapt with the latest spam and SMishing trends.

This article presents results and overview of conducted testing of active optical network devices. The base for the testing is originating in Kali Linux and penetration testing generally. The goal of tests is to either confirm or disprove a vulnerability of devices used in the tested polygon. The first part deals with general overview and topology of testing devices, the next part is dedicated to active and passive exploration and exploits. The last part provides a summary of the results.

CPS is generally complex to study, analyze, and design, as an important means to ensure the correctness of design and implementation of CPS system, simulation test is difficult to fully test, verify and evaluate the components or subsystems in the CPS system due to the inconsistent development progress of the com-ponents or subsystems in the CPS system. To address this prob-lem, we designed a hybrid P2P based collaborative simulation test framework composed of full physical nodes, hardware in the loop(HIL) nodes and full digital nodes to simulate the compo-nents or subsystems in the CPS system of different development progress, based on the framework, we then proposed collabora-tive simulation control strategy comprising sliding window based clock synchronization, dynamic adaptive time advancement and multi-priority task scheduling with preemptive time threshold. Experiments showed that the hybrid collaborative simulation testing method proposed in this paper can fully test CPS more effectively.

The use of risk information can help software engineers identify software components that are likely vulnerable or require extra attention when testing. Some studies have shown that the requirements risk-based approaches can be effective in improving the effectiveness of regression testing techniques. However, the risk estimation processes used in such approaches can be subjective, time-consuming, and costly. In this research, we introduce a fuzzy expert system that emulates human thinking to address the subjectivity related issues in the risk estimation process in a systematic and an efficient way and thus further improve the effectiveness of test case prioritization. Further, the required data for our approach was gathered by employing a semi-automated process that made the risk estimation process less subjective. The empirical results indicate that the new prioritization approach can improve the rate of fault detection over several existing test case prioritization techniques, while reducing threats to subjective risk estimation.

Current testing for Deep Neural Networks (DNNs) focuses on quantity of test cases but ignores diversity. To the best of our knowledge, DeepXplore is the first white-box framework for Deep Learning testing by triggering differential behaviors between multiple DNNs and increasing neuron coverage to improve diversity. Since it is based on multiple DNNs facing problems that (1) the framework is not friendly to a single DNN, (2) if incorrect predictions made by all DNNs simultaneously, DeepXplore cannot generate test cases. This paper presents Test4Deep, a white-box testing framework based on a single DNN. Test4Deep avoids mistakes of multiple DNNs by inducing inconsistencies between predicted labels of original inputs and that of generated test inputs. Meanwhile, Test4Deep improves neuron coverage to capture more diversity by attempting to activate more inactivated neurons. The proposed method was evaluated on three popular datasets with nine DNNs. Compared to DeepXplore, Test4Deep produced average 4.59% (maximum 10.49%) more test cases that all found errors and faults of DNNs. These test cases got 19.57% more diversity increment and 25.88% increment of neuron coverage. Test4Deep can further be used to improve the accuracy of DNNs by average up to 5.72% (maximum 7.0%).

Software vulnerabilities often remain hidden until an attacker exploits the weak/insecure code. Therefore, testing the software from a vulnerability discovery perspective becomes challenging for developers if they do not inspect their code thoroughly (which is time-consuming). We propose that vulnerability prediction using certain software metrics can support the testing process by identifying vulnerable code-components (e.g., functions, classes, etc.). Once a code-component is predicted as vulnerable, the developers can focus their testing efforts on it, thereby avoiding the time/effort required for testing the entire application. The current paper presents a study that compares how software metrics perform as vulnerability predictors for software projects developed in two different languages (Java vs Python). The goal of this research is to analyze the vulnerability prediction performance of software metrics for different programming languages. We designed and conducted experiments on security vulnerabilities reported for three Java projects (Apache Tomcat 6, Tomcat 7, Apache CXF) and two Python projects (Django and Keystone). In this paper, we focus on a specific type of code component: Functions. We apply Machine Learning models for predicting vulnerable functions. Overall results show that software metrics-based vulnerability prediction is more useful for Java projects than Python projects (i.e., software metrics when used as features were able to predict Java vulnerable functions with a higher recall and precision compared to Python vulnerable functions prediction).

The rapid growth and globalization of the integrated circuit (IC) industry put the threat of hardware Trojans (HTs) front and center among all security concerns in the IC supply chain. Current Trojan detection approaches always assume HTs are composed of digital circuits. However, recent demonstrations of analog attacks, such as A2 and Rowhammer, invalidate the digital assumption in previous HT detection or testing methods. At the system level, attackers can utilize the analog properties of the underlying circuits such as charge-sharing and capacitive coupling effects to create information leakage paths. These new capacitor-based vulnerabilities are rarely covered in digital testings. To address these stealthy yet harmful threats, we identify a large class of such capacitor-enabled attacks and define them as charge-domain Trojans. We are able to abstract the detailed charge-domain models for these Trojans and expose the circuit-level properties that critically contribute to their information leakage paths. Aided by the abstract models, an information flow tracking (IFT) based solution is developed to detect charge-domain leakage paths and then identify the charge-domain Trojans/vulnerabilities. Our proposed method is validated on an experimental RISC microcontroller design injected with different variants of charge-domain Trojans. We demonstrate that successful detection can be accomplished with an automatic tool which realizes the IFT-based solution.

Machine Learning (ML) algorithms, as the core technology in Artificial Intelligence (AI) applications, such as self-driving vehicles, make important decisions by performing a variety of data classification or prediction tasks. Attacks on data or algorithms in AI applications can lead to misclassification or misprediction, which can fail the applications. For each dataset separately, the parameters of ML algorithms should be tuned to reach a desirable classification or prediction accuracy. Typically, ML experts tune the parameters empirically, which can be time consuming and does not guarantee the optimal result. To this end, some research suggests an analytical approach to tune the ML parameters for maximum accuracy. However, none of the works consider the ML performance under attack in their tuning process. This paper proposes an analytical framework for tuning the ML parameters to be secure against attacks, while keeping its accuracy high. The framework finds the optimal set of parameters by defining a novel objective function, which takes into account the test results of both ML accuracy and its security against attacks. For validating the framework, an AI application is implemented to recognize whether a subject's eyes are open or closed, by applying k-Nearest Neighbors (kNN) algorithm on her Electroencephalogram (EEG) signals. In this application, the number of neighbors (k) and the distance metric type, as the two main parameters of kNN, are chosen for tuning. The input data perturbation attack, as one of the most common attacks on ML algorithms, is used for testing the security of the application. Exhaustive search approach is used to solve the optimization problem. The experiment results show k = 43 and cosine distance metric is the optimal configuration of kNN for the EEG dataset, which leads to 83.75% classification accuracy and reduces the attack success rate to 5.21%.

Part of our team proposed a new steganalytic method based on NIST tests at MMM-ACNS 2017 [1], and it was encouraged to investigate some cipher modifications to prevent such types of steganalysis. In the current paper, we propose one cipher modification based on decompression by arithmetic source compression coding. The experiment shows that the current proposed method allows to protect stegosystems against steganalysis based on NIST tests, while security of the encrypted embedded messages is kept. Protection of contemporary image steganography based on edge detection and modified LSB against NIST tests steganalysis is also presented.

In recent years, data published and shared with third parties to develop artificial intelligence (AI) tools and services has significantly increased. When there are regulatory or internal requirements regarding privacy of data, anonymization techniques are used to maintain privacy by transforming the data. The side-effect is that the anonymization may lead to useless data to train and test the AI because it is highly dependent on the quality of the data. To overcome this problem, we propose a test-driven anonymization approach for artificial intelligence tools. The approach tests different anonymization efforts to achieve a trade-off in terms of privacy (non-functional quality) and functional suitability of the artificial intelligence technique (functional quality). The approach has been validated by means of two real-life datasets in the domains of healthcare and health insurance. Each of these datasets is anonymized with several privacy protections and then used to train classification AIs. The results show how we can anonymize the data to achieve an adequate functional suitability in the AI context while maintaining the privacy of the anonymized data as high as possible.

Point of Sale (POS) systems has become the technology of choice for most businesses and offering number of advantages over traditional cash registers. They manage staffs, customers, transaction, inventory, sale and labor reporting, price adjustment, as well as keeping track of cash flow, expense management, reducing human errors and more. Whether traditional on-premise POS, or Cloud-Bases POS, they help businesses to run more efficiently. However, despite all these advantages, POS systems are becoming targets of a number of cyber-attacks. Security of a POS system is a key requirement of the Payment Card Industry Data Security Standard (PCI DSS). This paper undertakes research into the PCI DSS and its accompanying standards, in an attempt to break or bypass security measures using varying degrees of vulnerability and penetration attacks in a methodological format. The resounding goal of this experimentation is to achieve a basis from which attacks can be made against a realistic networking environment from whence an intruder can bypass security measures thus exposing a vulnerability in the PCI DSS and potentially exposing confidential customer payment information.

Twitter being the most popular content sharing platform is giving rise to automated accounts called “bots”. Majority of the users on Twitter are bots. Various machine learning (ML) algorithms are designed to detect bots avoiding the vulnerability constraints of ML-based models. This paper contributes to exploit vulnerabilities of machine learning (ML) algorithms through black-box attack. An adversarial text sequence misclassifies the results of deep learning (DL) classifiers for bot detection. Literature shows that ML models are vulnerable to attacks. The aim of this paper is to compromise the accuracy of ML-based bot detection algorithms by replacing original words in tweets with their synonyms. Our results show 7.2% decrease in the accuracy for bot tweets, therefore classifying bot tweets as legitimate tweets.

Advance persistent threat is a primary security concerns to the big organizations and its technical infrastructure, from cyber criminals seeking personal and financial information to state sponsored attacks designed to disrupt, compromising infrastructure, sidestepping security efforts thus causing serious damage to organizations. A skilled cybercriminal using multiple attack vectors and entry points navigates around the defenses, evading IDS/Firewall detection and breaching the network in no time. To understand the big picture, this paper analyses an approach to advanced persistent threat by doing the same things the bad guys do on a network setup. We will walk through various steps from foot-printing and reconnaissance, scanning networks, gaining access, maintaining access to finally clearing tracks, as in a real world attack. We will walk through different attack tools and exploits used in each phase and comparative study on their effectiveness, along with explaining their attack vectors and its countermeasures. We will conclude the paper by explaining the factors which actually qualify to be an Advance Persistent Threat.

Malware analysis and detection based on it is very important factor in the computer security. Despite of the enormous effort of companies making anti-malware solutions, it is usually not possible to respond to new malware in time and some computers will get infected. This shortcoming could be partially mitigated through using behavioral malware analysis. This work is aimed towards machine learning algorithms comparison for the behavioral malware analysis purposes.

Recent years have seen an increase in the data usage in cars, particularly as they become more autonomous and connected. With the rise in data use have come concerns about automotive cyber-security. An in-vehicle network shown to be particularly vulnerable is the Controller Area Network (CAN), which is the communication bus used by the car's safety critical and performance critical components. Cyber attacks on the CAN have been demonstrated, leading to research to develop attack detection and attack prevention systems. Such research requires representative attack demonstrations and data for testing. Obtaining this data is problematical due to the expense, danger and impracticality of using real cars on roads or tracks for example attacks. Whilst CAN simulators are available, these tend to be configured for testing conformance and functionality, rather than analysing security and cyber vulnerability. We therefore adapt a leading, industry-standard, CAN simulator to incorporate a core set of cyber attacks that are representative of those proposed by other researchers. Our adaptation allows the user to configure the attacks, and can be added easily to the free version of the simulator. Here we describe the simulator and, after reviewing the attacks that have been demonstrated and discussing their commonalities, we outline the attacks that we have incorporated into the simulator.

This paper begins with an introduction to security metrics, describing the need for security metrics, followed by a discussion of the nature of security metrics, including the challenges found with some security metrics used in the past. The paper then discusses what makes a good security metric and proposes a rigorous step-by-step method that can be applied to design good security metrics, and to test existing security metrics to see if they are good metrics. Application examples are included to illustrate the method.

Recent years, more and more testing criteria for deep learning systems has been proposed to ensure system robustness and reliability. These criteria were defined based on different perspectives of diversity. However, there lacks comprehensive investigation on what are the most essential diversities that should be considered by a testing criteria for deep learning systems. Therefore, in this paper, we conduct an empirical study to investigate the relation between test diversities and erroneous behaviors of deep learning models. We define five metrics to reflect diversities in neuron activities, and leverage metamorphic testing to detect erroneous behaviors. We investigate the correlation between metrics and erroneous behaviors. We also go further step to measure the quality of test suites under the guidance of defined metrics. Our results provided comprehensive insights on the essential diversities for testing criteria to exhibit good fault detection ability.

Classifying malware programs is a research area attracting great interest for Anti-Malware industry. In this research, we propose a system that visualizes malware programs as images and distinguishes those using Convolutional Neural Networks (CNNs). We study the performance of several well-established CNN based algorithms such as AlexNet, ResNet and VGG16 using transfer learning approaches. We also propose a computationally efficient CNN-based architecture for classification of malware programs. In addition, we study the performance of these CNNs as feature extractors by using Support Vector Machine (SVM) and K-nearest Neighbors (kNN) for classification purposes. We also propose fusion methods to boost the performance further. We make use of the publicly available database provided by Microsoft Malware Classification Challenge (BIG 2015) for this study. Our overall performance is 99.4% for a set of 2174 test samples comprising 9 different classes thereby setting a new benchmark.

Machine learning has been adopted widely to perform prediction and classification. Implementing machine learning increases security risks when computation process involves sensitive data on training and testing computations. We present a proposed system to protect machine learning engines in IoT environment without modifying internal machine learning architecture. Our proposed system is designed for passwordless and eliminated the third-party in executing machine learning transactions. To evaluate our a proposed system, we conduct experimental with machine learning transactions on IoT board and measure computation time each transaction. The experimental results show that our proposed system can address security issues on machine learning computation with low time consumption.

In this work, we applied a deep Convolutional Neural Network (CNN) with Xception model to perform malware image classification. The Xception model is a recently developed special CNN architecture that is more powerful with less over- fitting problems than the current popular CNN models such as VGG16. However only a few use cases of the Xception model can be found in literature, and it has never been used to solve the malware classification problem. The performance of our approach was compared with other methods including KNN, SVM, VGG16 etc. The experiments on two datasets (Malimg and Microsoft Malware Dataset) demonstrated that the Xception model can achieve the highest training accuracy than all other approaches including the champion approach, and highest validation accuracy than all other approaches including VGG16 model which are using image-based malware classification (except the champion solution as this information was not provided). Additionally, we proposed a novel ensemble model to combine the predictions from .bytes files and .asm files, showing that a lower logloss can be achieved. Although the champion on the Microsoft Malware Dataset achieved a bit lower logloss, our approach does not require any features engineering, making it more effective to adapt to any future evolution in malware, and very much less time consuming than the champion's solution.

Controller area networks (CANs) are vulnerable to spoofing attacks such as frame falsifying attacks, as electronic control units (ECUs) send and receive messages without any authentication and encryption. In this paper, we propose a physical authentication scheme that exploits the voltage features of the ECU signals on the CAN bus and applies reinforcement learning to choose the authentication mode such as the protection level and test threshold. This scheme enables a monitor node to optimize the authentication mode via trial-and-error without knowing the CAN bus signal model and spoofing model. Experimental results show that the proposed authentication scheme can significantly improve the authentication accuracy and response compared with a benchmark scheme.

In order to protect individuals' privacy, data have to be "well-sanitized" before sharing them, i.e. one has to remove any personal information before sharing data. However, it is not always clear when data shall be deemed well-sanitized. In this paper, we argue that the evaluation of sanitized data should be based on whether the data allows the inference of sensitive information that is specific to an individual, instead of being centered around the concept of re-identification. We propose a framework to evaluate the effectiveness of different sanitization techniques on a given dataset by measuring how much an individual's record from the sanitized dataset influences the inference of his/her own sensitive attribute. Our intent is not to accurately predict any sensitive attribute but rather to measure the impact of a single record on the inference of sensitive information. We demonstrate our approach by sanitizing two real datasets in different privacy models and evaluate/compare each sanitized dataset in our framework.

In this paper, RBF-based multistage auto-encoders are used to detect IDS attacks. RBF has numerous applications in various actual life settings. The planned technique involves a two-part multistage auto-encoder and RBF. The multistage auto-encoder is applied to select top and sensitive features from input data. The selected features from the multistage auto-encoder is wired as input to the RBF and the RBF is trained to categorize the input data into two labels: attack or no attack. The experiment was realized using MATLAB2018 on a dataset comprising 175,341 case, each of which involves 42 features and is authenticated using 82,332 case. The developed approach here has been applied for the first time, to the knowledge of the authors, to detect IDS attacks with 98.80% accuracy when validated using UNSW-NB15 dataset. The experimental results show the proposed method presents satisfactory results when compared with those obtained in this field.

We study the power of interactivity in local differential privacy. First, we focus on the difference between fully interactive and sequentially interactive protocols. Sequentially interactive protocols may query users adaptively in sequence, but they cannot return to previously queried users. The vast majority of existing lower bounds for local differential privacy apply only to sequentially interactive protocols, and before this paper it was not known whether fully interactive protocols were more powerful. We resolve this question. First, we classify locally private protocols by their compositionality, the multiplicative factor by which the sum of a protocol's single-round privacy parameters exceeds its overall privacy guarantee. We then show how to efficiently transform any fully interactive compositional protocol into an equivalent sequentially interactive protocol with a blowup in sample complexity linear in this compositionality. Next, we show that our reduction is tight by exhibiting a family of problems such that any sequentially interactive protocol requires this blowup in sample complexity over a fully interactive compositional protocol. We then turn our attention to hypothesis testing problems. We show that for a large class of compound hypothesis testing problems - which include all simple hypothesis testing problems as a special case - a simple noninteractive test is optimal among the class of all (possibly fully interactive) tests.

Failure to detect malware at its very inception leaves room for it to post significant threat and cost to cyber security for not only individuals, organizations but also the society and nation. However, the rapid growth in volume and diversity of malware renders conventional detection techniques that utilize feature extraction and comparison insufficient, making it very difficult for well-trained network administrators to identify malware, not to mention regular users of internet. Challenges in malware detection is exacerbated since complexity in the type and structure also increase dramatically in these years to include source code, binary file, shell script, Perl script, instructions, settings and others. Such increased complexity offers a premium on misjudgment. In order to increase malware detection efficiency and accuracy under large volume and multiple types of malware, this research adopts Convolutional Neural Networks (CNN), one of the most successful deep learning techniques. The experiment shows an accuracy rate of over 90% in identifying malicious and benign codes. The experiment also presents that CNN is effective with detecting source code and binary code, it can further identify malware that is embedded into benign code, leaving malware no place to hide. This research proposes a feasible solution for network administrators to efficiently identify malware at the very inception in the severe network environment nowadays, so that information technology personnel can take protective actions in a timely manner and make preparations for potential follow-up cyber-attacks.