Biblio

The popularity of Unmanned Aerial Vehicles (UAV) or more commonly known as Drones is increasing recently. UAVs have tremendous potential in various industries, e.g., military, agriculture, transportation, movie, supply chain, and surveillance. UAVs are also popular among hobbyists for photography, racing, etc. Despite the possibilities, many UAV related security incidents are reported nowadays. UAVs can be targeted by malicious parties and if compromised, life-threatening activities can be performed using them. As a result, governments around the world have started to regulate the use of UAVs. We believe that UAVs need an intelligent and automated defense mechanism to ensure the safety of humans, properties, and the UAVs themselves. A major component where we can incorporate the defense mechanism is the operating system. In this paper, we investigate the security of existing operating systems used in consumer and commercial UAVs. We then survey various security issues of UAV operating systems and possible solutions. Finally, we discuss several research challenges for developing a secure operating system for UAVs.

Recommender systems in software engineering provide developers with a wide range of valuable items to help them complete their tasks. Among others, API recommender systems have gained momentum in recent years as they became more successful at suggesting API calls or code snippets. While these systems have proven to be effective in terms of prediction accuracy, there has been less attention for what concerns such recommenders’ resilience against adversarial attempts. In fact, by crafting the recommenders’ learning material, e.g., data from large open-source software (OSS) repositories, hostile users may succeed in injecting malicious data, putting at risk the software clients adopting API recommender systems. In this paper, we present an empirical investigation of adversarial machine learning techniques and their possible influence on recommender systems. The evaluation performed on three state-of-the-art API recommender systems reveals a worrying outcome: all of them are not immune to malicious data. The obtained result triggers the need for effective countermeasures to protect recommender systems against hostile attacks disguised in training data.

There is an increasing trend in cloud adoption of enterprise applications in, for example, manufacturing, healthcare, and finance. Such applications are routinely subject to targeted cyberattacks, which result in significant loss of sensitive data (e.g., due to data exfiltration in advanced persistent threats) or valuable utilities (e.g., due to resource the exfiltration of power in cryptojacking). There is a critical need to train highly skilled cybersecurity professionals, who are capable of defending against such targeted attacks. In this article, we present the design, development, and evaluation of the Mizzou Cyber Range, an online platform to learn basic/advanced cyber defense concepts and perform training exercises to engender the next-generation cybersecurity workforce. Mizzou Cyber Range features flexibility, scalability, portability, and extendability in delivering cyberattack/defense learning modules to students. We detail our “research-inspired learning” and “learn-apply-create” three-phase pedagogy methodologies in the development of four learning modules that include laboratory exercises and self-study activities using realistic cloud-based application testbeds. The learning modules allow students to gain skills in using latest technologies (e.g., elastic capacity provisioning, software-defined everything infrastructure) to implement sophisticated “attack defense by pretense” techniques. Students can also use the learning modules to understand the attacker-defender game in order to create disincentives (i.e., pretense initiation) that make the attacker's tasks more difficult, costly, time consuming, and uncertain. Lastly, we show the benefits of our Mizzou Cyber Range through the evaluation of student learning using auto-grading, rank assessments with peer standing, and monitoring of students' performance via feedback from prelab evaluation surveys and postlab technical assessments.

Deep reinforcement learning has recently been adopted for robot behavior learning, where robot skills are acquired and adapted from data generated by the robot while interacting with its environment through a trial-and-error process. Despite this success, most model-free deep reinforcement learning algorithms learn a task-specific policy from a clean slate and thus suffer from high sample complexity (i.e., they require a significant amount of interaction with the environment to learn reasonable policies and even more to reach convergence). They also suffer from poor initial performance due to executing a randomly initialized policy in the early stages of learning to obtain experience used to train a policy or value function. Model based deep reinforcement learning mitigates these shortcomings. However, it suffers from poor asymptotic performance in contrast to a model-free approach. In this work, we investigate knowledge transfer from a model-based teacher to a task-specific model-free learner to alleviate executing a randomly initialized policy in the early stages of learning. Our experiments show that this approach results in better asymptotic performance, enhanced initial performance, improved safety, better action effectiveness, and reduced sample complexity.

Information leakage estimation for practical wiretap codes is a challenging task for which existing solutions are either too complex or suboptimal, and don't scale for large blocklengths. In this paper we present a new method, based on a modified version of the successive cancellation decoder in order to compute the information leakage for the wiretap polar code which improves upon existing methods in terms of complexity and accuracy. Results are presented for classical binary-input symmetric channels alike the Binary Erasure Channel (BEC), the Binary Symmetric Channel (BSC) and Binary Input Additive White Gaussian Noise channel (BI-AWGN).

Human multi-robot system (MRS) collaboration is demonstrating potentials in wide application scenarios due to the integration of human cognitive skills and a robot team’s powerful capability introduced by its multi-member structure. However, due to limited human cognitive capability, a human cannot simultaneously monitor multiple robots and identify the abnormal ones, largely limiting the efficiency of the human-MRS collaboration. There is an urgent need to proactively reduce unnecessary human engagements and further reduce human cognitive loads. Human trust in human MRS collaboration reveals human expectations on robot performance. Based on trust estimation, the work between a human and MRS will be reallocated that an MRS will self-monitor and only request human guidance in critical situations. Inspired by that, a novel Synthesized Trust Learning (STL) method was developed to model human trust in the collaboration. STL explores two aspects of human trust (trust level and trust preference), meanwhile accelerates the convergence speed by integrating active learning to reduce human workload. To validate the effectiveness of the method, tasks "searching victims in the context of city rescue" were designed in an open-world simulation environment, and a user study with 10 volunteers was conducted to generate real human trust feedback. The results showed that by maximally utilizing human feedback, the STL achieved higher accuracy in trust modeling with a few human feedback, effectively reducing human interventions needed for modeling an accurate trust, therefore reducing human cognitive load in the collaboration.

Recently, with the increasing number of large-scale remote sensing images, the demand for large-scale remote sensing image object classification is growing and attracting the interest of many researchers. Hashing, because of its low memory requirements and high time efficiency, has widely solve the problem of large-scale remote sensing image. Supervised hashing methods mainly leverage the label information of remote sensing image to learn hash function, however, the similarity of the original feature space cannot be well preserved, which can not meet the accurate requirements for object classification of remote sensing image. To solve the mentioned problem, we propose a novel method named Optimized Projection Supervised Discrete Hashing(OPSDH), which jointly learns a discrete binary codes generation and optimized projection constraint model. It uses an effective optimized projection method to further constraint the supervised hash learning and generated hash codes preserve the similarity based on the data label while retaining the similarity of the original feature space. The experimental results show that OPSDH reaches improved performance compared with the existing hash learning methods and demonstrate that the proposed method is more efficient for operational applications.

This paper presents experimental results of evaluating the effect of network delay on the synchronization time of three parity machines. The possibility of using a hash function to confirm the synchronization of parity tree machines has been investigated. Three parity machines have been proposed as a modification of the symmetric encryption algorithm. One advantage of the method is the possibility to use the phenomenon of mutual synchronization of neural networks to generate an identical encryption key for users without the need to transfer it. As a result, the degree of influence of network delay and the type of hash function used on the synchronization time of neural networks was determined. The degree of influence of the network delay and hash function was determined experimentally. The hash function sha512 showed the best results. The tasks for further research have been defined.

We propose a human-operator guided planning approach to pushing-based manipulation in clutter. Most recent approaches to manipulation in clutter employs randomized planning. The problem, however, remains a challenging one where the planning times are still in the order of tens of seconds or minutes, and the success rates are low for difficult instances of the problem. We build on these control-based randomized planning approaches, but we investigate using them in conjunction with human-operator input. In our framework, the human operator supplies a high-level plan, in the form of an ordered sequence of objects and their approximate goal positions. We present experiments in simulation and on a real robotic setup, where we compare the success rate and planning times of our human-in-the-loop approach with fully autonomous sampling-based planners. We show that with a minimal amount of human input, the low-level planner can solve the problem faster and with higher success rates.

We present a scalable architecture based on a trained filter bank for input pre-processing and a recurrent neural network (RNN) for the detection of atrial fibrillation in electrocardiogram (ECG) signals, with the focus on enabling a very efficient hardware implementation as application-specific integrated circuit (ASIC). Our already very efficient base architecture is further improved by replacing the RNN with a delta-encoded gated recurrent unit (GRU) and adding a confidence measure (CM) for terminating the computation as early as possible. With these optimizations, we demonstrate a reduction of the processing load of 58 % on an internal dataset while still achieving near state-of-the-art classification results on the Physionet ECG dataset with only 1202 parameters.

Human-in-the-Loop has been receiving special attention from the data science and machine learning community. It is essential to realize the advantages of human feedback and the pressing need for manual annotation to improve machine learning performance. Recent advancements in natural language processing (NLP) and machine learning have created unique challenges and opportunities for digital humanities research. In particular, there are ample opportunities for NLP and machine learning researchers to analyze data from literary texts and use these complex source texts to broaden our understanding of human sentiment using the human-in-the-loop approach. This paper presents our understanding of how human annotators differ from machine annotators in sentiment analysis tasks and how these differences can contribute to designing systems for the "human in the loop" sentiment analysis in complex, unstructured texts. We further explore the challenges and benefits of the human-machine collaboration for sentiment analysis using a case study in Greek tragedy and address some open questions about collaborative annotation for sentiments in literary texts. We focus primarily on (i) an analysis of the challenges in sentiment analysis tasks for humans and machines, and (ii) whether consistent annotation results are generated from multiple human annotators and multiple machine annotators. For human annotators, we have used a survey-based approach with about 60 college students. We have selected six popular sentiment analysis tools for machine annotators, including VADER, CoreNLP's sentiment annotator, TextBlob, LIME, Glove+LSTM, and RoBERTa. We have conducted a qualitative and quantitative evaluation with the human-in-the-loop approach and confirmed our observations on sentiment tasks using the Greek tragedy case study.

Deep learning has undergone tremendous advancements in computer vision studies. The training of deep learning neural networks depends on a considerable amount of ground truth datasets. However, labeling ground truth data is a labor-intensive task, particularly for large-volume video analytics applications such as video surveillance and vehicles detection for autonomous driving. This paper presents a rapid and accurate method for associative searching in big image data obtained from security monitoring systems. We developed a semi-automatic moving object annotation method for improving deep learning models. The proposed method comprises three stages, namely automatic foreground object extraction, object annotation in subsequent video frames, and dataset construction using human-in-the-loop quick selection. Furthermore, the proposed method expedites dataset collection and ground truth annotation processes. In contrast to data augmentation and data generative models, the proposed method produces a large amount of real data, which may facilitate training results and avoid adverse effects engendered by artifactual data. We applied the constructed annotation dataset to train a deep learning you-only-look-once (YOLO) model to perform vehicle detection on street intersection surveillance videos. Experimental results demonstrated that the accurate detection performance was improved from a mean average precision (mAP) of 83.99 to 88.03.

The rapid progress experienced in the Internet of Things (IoT) space is one that has introduced new and unique challenges for cybersecurity and IoT-Forensics. One of these problems is how digital forensics and incident response (DFIR) are handled in IoT. Since enormous users use IoT platforms to accomplish their day to day task, massive amounts of data streams are transferred with limited hardware resources; conducting DFIR needs a new approach to mitigate digital evidence and incident response challenges owing to the facts that there are no unified standard or classified principles for IoT forensics. Today's IoT DFIR relies on self-defined best practices and experiences. Given these challenges, IoT-related incidents need a more structured approach in identifying problems of DFIR. In this paper, we examined the major DFIR challenges in IoT by exploring the different phases involved in a DFIR when responding to IoT-related incidents. This study aims to provide researchers and practitioners a road-map that will help improve the standards of IoT security and DFIR.

Network embedding, which aims to map the discrete network topology to a continuous low-dimensional representation space with the major topological properties preserved, has emerged as an essential technique to support various network inference tasks. However, incorporating both the evolutionary nature and the network's heterogeneity remains a challenge for existing network embedding methods. In this study, we propose a novel Translation-Based Dynamic Heterogeneous Network Embedding (TransDHE) approach to consider both the aspects simultaneously. For a dynamic heterogeneous network with a sequence of snapshots and multiple types of nodes and edges, we introduce a translation-based embedding module to capture the heterogeneous characteristics (e.g., type information) of each single snapshot. An orthogonal alignment module and RNN-based aggregation module are then applied to explore the evolutionary patterns among multiple successive snapshots for the final representation learning. Extensive experiments on a set of real-world networks demonstrate that TransDHE can derive the more informative embedding result for the network dynamic and heterogeneity over state-of-the-art network embedding baselines.

Online news articles are an important source of information for decisions makers to understand the causal relation of events that happened. However, understanding the causality of an event or between events by traditional machine learning-based techniques from natural language text is a challenging task due to the complexity of the language to be comprehended by the machines. In this study, the Knowledge-oriented convolutional neural network (K-CNN) technique is used to extract the causal relation from online news articles related to the South China Sea (SCS) dispute. The proposed K-CNN model contains a Knowledge-oriented channel that can capture the causal phrases of causal relationships. A Data-oriented channel that captures the position information was added to the K-CNN model in this phase. The online news articles were collected from the national news agency and then the sentences which contain relation such as causal, message-topic, and product-producer were extracted. Then, the extracted sentences were annotated and converted into lower form and base form followed by transformed into the vector by looking up the word embedding table. A word filter that contains causal keywords was generated and a K-CNN model was developed, trained, and tested using the collected data. Finally, different architectures of the K-CNN model were compared to find out the most suitable architecture for this study. From the study, it was found out that the most suitable architecture was the K-CNN model with a Knowledge-oriented channel and a Data-oriented channel with average pooling. This shows that the linguistic clues and the position features can improve the performance in extracting the causal relation from the SCS online news articles. Keywords-component; Convolutional Neural Network, Causal Relation Extraction, South China Sea.

Attack surface detection for the complex software is needed to find targets for the fuzzing, because testing the whole system with many inputs is not realistic. Researchers that previously applied taint analysis for dealing with different security tasks in the virtual machines did not examined how to apply it for attack surface detection. I.e., getting the program modules and functions, that may be affected by input data. We propose using taint tracking within a virtual machine and virtual machine introspection to create a new approach that can detect the internal module interfaces that can be fuzz tested to assure that software is safe or find the vulnerabilities.

The front-end text processing module is considered as an essential part that influences the intelligibility and naturalness of a Mandarin text-to-speech system significantly. For commercial text-to-speech systems, the Mandarin front-end should meet the requirements of high accuracy and low time latency while also ensuring maintainability. In this paper, we propose a universal BERT-based model that can be used for various tasks in the Mandarin front-end without changing its architecture. The feature extractor and classifiers in the model are shared for several sub-tasks, which improves the expandability and maintainability. We trained and evaluated the model with polyphone disambiguation, text normalization, and prosodic boundary prediction for single task modules and multi-task learning. Results show that, the model maintains high performance for single task modules and shows higher accuracy and lower time latency for multi-task modules, indicating that the proposed universal front-end model is promising as a maintainable Mandarin front-end for commercial applications.

Internet of Things (IoT) and its applications are becoming commonplace with more devices, but always at risk of network security. It is therefore crucial for an IoT network design to identify attackers accurately, quickly and promptly. Many solutions have been proposed, mainly concerning secure IoT architectures and classification algorithms, but none of them have paid enough attention to reducing the complexity. Our proposal in this paper is an edge-cloud architecture that fulfills the detection task right at the edge layer, near the source of the attacks for quick response, versatility, as well as reducing the cloud's workload. We also propose a multi-attack detection mechanism called LCHA (Low-Complexity detection solution with High Accuracy) , which has low complexity for deployment at the edge zone while still maintaining high accuracy. The performance of our proposed mechanism is compared with that of other machine learning and deep learning methods using the most updated BoT-IoT data set. The results show that LCHA outperforms other algorithms such as NN, CNN, RNN, KNN, SVM, KNN, RF and Decision Tree in terms of accuracy and NN in terms of complexity.

Recently, smart video surveillance at the edge has become a trend in developing security applications since edge computing enables more image processing tasks to be implemented on the decentralised network note of the surveillance system. As a result, many security applications such as behaviour recognition and prediction, employee safety, perimeter intrusion detection and vandalism deterrence can minimise their latency or even process in real-time when the camera network system is extended to a larger degree. Technically, human detection is a key step in the implementation of these applications. With the advantage of high detection rates, deep learning methods have been widely employed on edge devices in order to detect human objects. However, due to their high computation costs, it is challenging to apply these methods on resource limited edge devices for real-time applications. Inspired by the You Only Look Once (YOLO), residual learning and Spatial Pyramid Pooling (SPP), a novel form of real-time human detection is presented in this paper. Our approach focuses on designing a network structure so that the developed model can achieve a good trade-off between accuracy and processing time. Experimental results show that our trained model can process 2 FPS on Raspberry PI 3B and detect humans with accuracies of 95.05 % and 96.81 % when tested respectively on INRIA and PENN FUDAN datasets. On the human COCO test dataset, our trained model outperforms the performance of the Tiny-YOLO versions. Additionally, compare to the SSD based L-CNN method, our algorithm achieves better accuracy than the other method.

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

Network Centric Warfare (NCW) is a design that supports information excellence for the concept of military operations. Network Centric Warfare is currently being developed as the basis for the operating concept, namely multidimensional operations. TNI operations do not rely on conventional warfare. TNI operations must work closely with the TNI Puspen team, territorial intelligence, TNI cyber team, and support task force. Sending digital images sent online requires better techniques to maintain confidentiality. The purpose of this research is to design digital image security with AES cryptography and discrete wavelet transform method on interoperability and to utilize and study discrete wavelet transform method and AES algorithm on interoperability for digital image security. The AES cryptography technique in this study is used to protect and maintain the confidentiality of the message while the Discrete Wavelet Transform in this study is used to reduce noise by applying a discrete wavelet transform, which consists of three main steps, namely: image decomposition, thresholding process and image reconstruction. The result of this research is that Digital Image Security to support TNI interoperability has been produced using the C \# programming language framework. NET and Xampp to support application development. Users can send data in the form of images. Discrete Wavelet Transformation in this study is used to find the lowest value against the threshold so that the resulting level of security is high. Testing using the AESS algorithm to encrypt and decrypt image files using key size and block size.

Web application firewalls (WAF) are the last line of defense in protecting web applications from application layer security threats like SQL injection and cross-site scripting. Currently, most evasion techniques from WAFs are still developed manually. In this work, we propose a solution, which automatically scans the WAFs to find payloads through which the WAFs can be bypassed. Our solution finds out rules defects, which can be further used in rule tuning for rule-based WAFs. Also, it can enrich the machine learning-based dataset for retraining. To this purpose, we provide a framework based on reinforcement learning with an environment compatible with OpenAI gym toolset standards, employed for training agents to implement WAF evasion tasks. The framework acts as an adversary and exploits a set of mutation operators to mutate the malicious payload syntactically without affecting the original semantics. We use Q-learning and proximal policy optimization algorithms with the deep neural network. Our solution is successful in evading signature-based and machine learning-based WAFs.

Nowadays, web vulnerability is a critical issue in web applications. Web developers develop web applications, but sometimes they are not very well-versed with security concerns, thereby creating loopholes for the vulnerabilities. If a web application is developed without considering security, it is harmful for the client and the company. Different types of vulnerabilities encounter during the web application development process. Therefore, vulnerability identification is a crucial and critical task from a web application development perspective. It is vigorous to secure them from the earliest development life cycle process. In this paper, we have analyzed and classified vulnerabilities related to web application security during the development phases. Here, the concern is to identify a weakness, countermeasure, confidentiality impact, access complexity, and severity level, which affect the web application security.

A real or potential threat to various large and small security objects, which comes from both internal and external attackers, determines one or another activities to ensure internal and external security. These actions depend on the spheres of life of state and society, which are targeted by the security threats. These threats can be conveniently classified into political threats (or threats to the existing constitutional order), economic, military, informational, technogenic, environmental, corporate, and other threats. The article discusses a model of an information system, which main criterion is the system security based on the concept of risk. When considering the model, it was determined that it possess multi-criteria aspects. Therefore the establishing the quantitative and qualitative characteristics is a complex and dynamic task. The paper proposes to use the mathematical apparatus of the teletraffic theory in one of the elements of the protected system, namely, in the end-to-end security subsystem.

Software-defined networking (SDN) nowadays is extensively being used in a variety of practical settings, provides a new way to manage networks by separating the data plane from its control plane. However, SDN is particularly vulnerable to Distributed Denial of Service (DDoS) attacks because of its centralized control logic. Many studies have been proposed to tackle DDoS attacks in an SDN design using machine-learning-based schemes; however, these feature-based detection schemes are highly resource-intensive and they are unable to perform reliably in such a large-scale SDN network where a massive amount of traffic data is generated from both control and data planes. This can deplete computing resources, degrade network performance, or even shut down the network systems owing to being exhausting resources. To address the above challenges, this paper proposes a big data framework to overcome traditional data processing limitations and to exploit distributed resources effectively for the most compute-intensive tasks such as DDoS attack detection using machine learning techniques, etc. We demonstrate the robustness, scalability, and effectiveness of our framework through practical experiments.