Biblio

Modern consumer electronic devices often provide intelligence services with deep neural networks. We have started migrating the computing locations of intelligence services from cloud servers (traditional AI systems) to the corresponding devices (on-device AI systems). On-device AI systems generally have the advantages of preserving privacy, removing network latency, and saving cloud costs. With the emergence of on-device AI systems having relatively low computing power, the inconsistent and varying hardware resources and capabilities pose difficulties. Authors' affiliation has started applying a stream pipeline framework, NNStreamer, for on-device AI systems, saving developmental costs and hardware resources and improving performance. We want to expand the types of devices and applications with on-device AI services products of both the affiliation and second/third parties. We also want to make each AI service atomic, re-deployable, and shared among connected devices of arbitrary vendors; we now have yet another requirement introduced as it always has been. The new requirement of “among-device AI” includes connectivity between AI pipelines so that they may share computing resources and hardware capabilities across a wide range of devices regardless of vendors and manufacturers. We propose extensions of the stream pipeline framework, NNStreamer, for on-device AI so that NNStreamer may provide among-device AI capability. This work is a Linux Foundation (LF AI & Data) open source project accepting contributions from the general public.

This paper discusses the outcome of combining insider threat agent taxonomies with the aim of enhancing insider threat detection. The objectives sought to explore taxonomy combinations and investigate threat sophistication from the taxonomy combinations. Investigations revealed the plausibility of combining the various taxonomy categories to derive a new taxonomy. An observation on category combinations yielded the introduction of the concept of a threat path. The proposed taxonomy tree consisted of more than a million threat-paths obtained using a formula from combinatorics analysis. The taxonomy category combinations thus increase the insider threat landscape and hence the gap between insider threat agent sophistication and countermeasures. On the defensive side, knowledge of insider threat agent taxonomy category combinations has the potential to enhance defensive countermeasure tactics, techniques and procedures, thus increasing the chances of insider threat detection.

Key management for self-organized wireless ad-hoc networks using peer-to-peer (P2P) keys is the primary goal of this article (SOWANs). Currently, wireless networks have centralized security architectures, making them difficult to secure. In most cases, ad-hoc wireless networks are not connected to trusted authorities or central servers. They are more prone to fragmentation and disintegration as a result of node and link failures. Traditional security solutions that rely on online trusted authorities do not work together to protect networks that are not planned. With open wireless networks, anyone can join or leave at any time with the right equipment, and no third party is required to verify their identity. These networks are best suited for this proposed method. Each node can make, distribute, and revoke its keying material in this paper. A minimal amount of communication and computation is required to accomplish this task. So that they can authenticate one another and create shared keys, nodes in the self-organized version of the system must communicate via a secure side channel between the users' devices.

Recent works have empirically shown that neural network interpretability is susceptible to malicious manipulations. However, existing attacks against Interpretable Deep Learning Systems (IDLSes) all focus on the white-box setting, which is obviously unpractical in real-world scenarios. In this paper, we make the first attempt to attack IDLSes in the decision-based black-box setting. We propose a new framework called Dual Black-box Adversarial Attack (DBAA) which can generate adversarial examples that are misclassified as the target class, yet have very similar interpretations to their benign cases. We conduct comprehensive experiments on different combinations of classifiers and interpreters to illustrate the effectiveness of DBAA. Empirical results show that in all the cases, DBAA achieves high attack success rates and Intersection over Union (IoU) scores.

Axie infinity is a complicated card game with a huge-scale action space. This makes it difficult to solve this challenge using generic Reinforcement Learning (RL) algorithms. We propose a hybrid RL framework to learn action representations and game strategies. To avoid evaluating every action in the large feasible action set, our method evaluates actions in a fixed-size set which is determined using action representations. We compare the performance of our method with two baseline methods in terms of their sample efficiency and the winning rates of the trained models. We empirically show that our method achieves an overall best winning rate and the best sample efficiency among the three methods.

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

The vast proliferation of the connected devices makes the operation of the traditional networks so complex and drops the network performance, particularly, failure cases. In fact, a novel solution is proposed to enable the management of the network resources and services named software defined network (SDN). SDN splits the data plane and the control plane by centralizing all the control plane on one common platform. Further, SDN makes the control plane programmable by offering high flexibility for the network management and monitoring mostly in failure cases. However, the main challenge in SDN is security that is presented as the first barrier for its development. Security in SDN is presented at various levels and forms, particularly, the communication between the data plane and control plane that presents a weak point in SDN framework. In this article, we suggest a new security framework focused on the combination between the trust and awareness concepts (TAS-SDN) for a dynamic southbound communication SDN. Further, TAS-SDN uses trust levels to establish a secure communication between the control plane and data plane. As a result, we discuss the implementation and the performance of TAS-SDN which presents a promote security solution in terms of time execution, complexity and scalability for SDN.

The success of the web and the consequent rise in data sharing have made network security a challenge. Attackers from all around the world target PC installations. When an attack is successful, an electronic device's security is jeopardised. The intrusion implicitly includes any sort of behaviours that purport to think twice about the respectability, secrecy, or accessibility of an asset. Information is shielded from unauthorised clients' scrutiny by the integrity of a certain foundation. Accessibility refers to the framework that gives users of the framework true access to information. The word "classification" implies that data within a given frame is shielded from unauthorised access and public display. Consequently, a PC network is considered to be fully completed if the primary objectives of these three standards have been satisfactorily met. To assist in achieving these objectives, Intrusion Detection Systems have been developed with the fundamental purpose of scanning incoming traffic on computer networks for malicious intrusions.

Digitalization has occurred in almost all industries, one of them is health industry. Patients” medical records are now easier to be accessed and managed as all related data are stored in data storages or repositories. However, this system is still under development as number of patients still increasing. Lack of standardization might lead to patients losing their right to control their own data. Therefore, implementing private blockchain system with Self-Sovereign Identity (SSI) concept for identity management in health industry is a viable notion. With SSI, the patients will be benefited from having control over their own medical records and stored with higher security protocol. While healthcare providers will benefit in Know You Customer (KYC) process, if they handle new patients, who move from other healthcare providers. It will eliminate and shorten the process of updating patients' medical records from previous healthcare providers. Therefore, we suggest several flows in implementing blockchain for digital identity in healthcare industry to help overcome lack of patient's data control and KYC in current system. Nevertheless, implementing blockchain on health industry requires full attention from surrounding system and stakeholders to be realized.

Today, social communication through the Internet has become more popular and has become a crucial part of our daily life. Naturally, sending and receiving various data through the Internet has also grown a lot. Keeping important data secure in transit has become a challenge for individuals and even organizations. Therefore, the trinity of confidentiality, integrity, and availability will be essential, and encryption will definitely be one of the best solutions to this problem. Of course, for image data, it will not be possible to use conventional encryption methods for various reasons, such as the redundancy of image data, the strong correlation of adj acent pixels, and the large volume of image data. Therefore, special methods were developed for image encryption. Among the prevalent methods for image encryption is the use of DNA sequences as well as chaos signals. In this paper, a cycling 3D chaotic map and DNA sequences are used to present a new method for color image encryption. Several experimental analyses were performed on the proposed method, and the results proved that the presented method is secure and efficient.

Over earlier years of huge technical developments, the need for a communication system has risen tremendously. Inrecent times, public realm interaction has been a popular area, hence the research group is emphasizing the necessity of quick and efficient broadband speeds, as well as upgraded security protocols. The main objective of this project work is to combine conventional Li-Fi and VLC techniques for video communication. VLC is helping to deliver fast data speeds, bandwidth efficiency, and a relatively secure channel of communication. Li-Fi is an inexpensive wireless communication (WC) system. Li-Fi can transmit information (text, audio, and video) to any electronic device via the LEDs that are positioned in the space to provide lighting. Li-Fi provides more advantages than Wi-Fi, such as security, high efficiency, speed, throughput, and low latency. The information can be transferred based on the flash property of the LED. Communication is accomplished by turning on and off LED lights at a faster pace than the human visual system can detect.

The paper aims to discover vulnerabilities by application of supervisory control theory and to design a defensive supervisor against vulnerability attacks. Supervisory control restricts the system behavior to satisfy the control specifications. The existence condition of the supervisor, sometimes results in undesirable plant behavior, which can be regarded as a vulnerability of the control specifications. We aim to design a more robust supervisor against this vulnerability.

Structured Query Language Injection (SQLi) is a client-side application vulnerability that allows attackers to inject malicious SQL queries with harmful intents, including stealing sensitive information, bypassing authentication, and even executing illegal operations to cause more catastrophic damage to users on the web application. According to OWASP, the top 10 harmful attacks against web applications are SQL Injection attacks. Moreover, based on data reports from the UK's National Fraud Authority, SQL Injection is responsible for 97% of data exposures. Therefore, in order to prevent the SQL Injection attack, detection SQLi system is essential. The contribution of this research is securing web applications by developing a browser extension for Google Chrome using Long Short-Term Memory (LSTM), which is a unique kind of RNN algorithm capable of learning long-term dependencies like SQL Injection attacks. The results of the model will be deployed in static analysis in a browser extension, and the LSTM algorithm will learn to identify the URL that has to be injected into Damn Vulnerable Web Application (DVWA) as a sample-tested web application. Experimental results show that the proposed SQLi detection model based on the LSTM algorithm achieves an accuracy rate of 99.97%, which means that a reliable client-side can effectively detect whether the URL being accessed contains a SQLi attack or not.

The study focused on assessing and testing Windows 10 to identify possible vulnerabilities and their ability to withstand cyber-attacks. CVE data, alongside other vulnerability reports, were instrumental in measuring the operating system's performance. Metasploit and Nmap were essential in penetration and intrusion experiments in a simulated environment. The study applied the following testing procedure: information gathering, scanning and results analysis, vulnerability selection, launch attacks, and gaining access to the operating system. Penetration testing involved eight attacks, two of which were effective against the different Windows 10 versions. Installing the latest version of Windows 10 did not guarantee complete protection against attacks. Further research is essential in assessing the system's vulnerabilities are recommending better solutions.

Fruit-80, an ultra-lightweight stream cipher with 80-bit secret key, is oriented toward resource constrained devices in the Internet of Things. In this paper, we propose area and speed optimization architectures of Fruit-80 on FPGAs. The area optimization architecture reuses NFSR&LFSR feedback functions and achieves the most suitable ratio of look-up-tables and flip-flops. The speed optimization architecture adopts a hybrid approach for parallelization and reduces the latency of long data paths by pre-generating primary feedback and inserting flip-flops. In conclusion, the optimal throughput-to-area ratio of the speed optimization architecture is better than that of Grain v1. The area optimization architecture occupies only 35 slices on Xilinx Spartan-3 FPGA, smaller than that of Grain and other common stream ciphers. To the best of our knowledge, this result sets a new record of the minimum area in lightweight cipher implementations on FPGA.

Advanced metamorphic malware and ransomware use techniques like obfuscation to alter their internal structure with every attack. Therefore, any signature extracted from such attack, and used to bolster endpoint defense, cannot avert subsequent attacks. Therefore, if even a single such malware intrudes even a single device of an IoT network, it will continue to infect the entire network. Scenarios where an entire network is targeted by a coordinated swarm of such malware is not beyond imagination. Therefore, the IoT era also requires Industry-4.0 grade AI-based solutions against such advanced attacks. But AI-based solutions need a large repository of data extracted from similar attacks to learn robust representations. Whereas, developing a metamorphic malware is a very complex task and requires extreme human ingenuity. Hence, there does not exist abundant metamorphic malware to train AI-based defensive solutions. Also, there is currently no system that could generate enough functionality preserving metamorphic variants of multiple malware to train AI-based defensive systems. Therefore, to this end, we design and develop a novel system, named X-Swarm. X-Swarm uses deep policy-based adversarial reinforcement learning to generate swarm of metamorphic instances of any malware by obfuscating them at the opcode level and ensuring that they could evade even capable, adversarial-attack immune endpoint defense systems.

Data privacy is a key concern for smart contracts handling sensitive data. The existing work zkay addresses this concern by allowing developers without cryptographic expertise to enforce data privacy. However, while zkay avoids fundamental limitations of other private smart contract systems, it cannot express key applications that involve operations on foreign data.We present ZeeStar, a language and compiler allowing non-experts to instantiate private smart contracts and supporting operations on foreign data. The ZeeStar language allows developers to ergonomically specify privacy constraints using zkay’s privacy annotations. The ZeeStar compiler then provably realizes these constraints by combining non-interactive zero-knowledge proofs and additively homomorphic encryption.We implemented ZeeStar for the public blockchain Ethereum. We demonstrated its expressiveness by encoding 12 example contracts, including oblivious transfer and a private payment system like Zether. ZeeStar is practical: it prepares transactions for our contracts in at most 54.7s, at an average cost of 339k gas.

With the increasing number of catastrophic weather events and resulting disruption in the energy supply to essential loads, the distribution grid operators’ focus has shifted from reliability to resiliency against high impact, low-frequency events. Given the enhanced automation to enable the smarter grid, there are several assets/resources at the disposal of electric utilities to enhances resiliency. However, with a lack of comprehensive resilience tools for informed operational decisions and planning, utilities face a challenge in investing and prioritizing operational control actions for resiliency. The distribution system resilience is also highly dependent on system attributes, including network, control, generating resources, location of loads and resources, as well as the progression of an extreme event. In this work, we present a novel multi-stage resilience measure called the Anticipate-Withstand-Recover (AWR) metrics. The AWR metrics are based on integrating relevant ‘system characteristics based factors’, before, during, and after the extreme event. The developed methodology utilizes a pragmatic and flexible approach by adopting concepts from the national emergency preparedness paradigm, proactive and reactive controls of grid assets, graph theory with system and component constraints, and multi-criteria decision-making process. The proposed metrics are applied to provide decision support for a) the operational resilience and b) planning investments, and validated for a real system in Alaska during the entirety of the event progression.

Cancelable biometric is a new era of technology that deals with the protection of the privacy content of a person which itself helps in protecting the identity of a person. Here the biometric information instead of being stored directly on the authentication database is transformed into a non-invertible coded format that will be utilized for providing access. The conversion into an encrypted code requires the provision of an encryption key from the user side. Both invertible and non-invertible coding techniques are there but non-invertible one provides additional security to the user. In this paper, a non-invertible cancelable biometric method has been proposed where the biometric image information is canceled and encoded into a code using a user-provided encryption key. This code is generated from the image histogram after continuous bin updation to the maximal value and then it is encrypted by the Hill cipher. This code is stored on the database instead of biometric information. The technique is applied to a set of retinal information taken from the Indian Diabetic Retinopathy database.

Static analysis tools help to detect common pro-gramming errors but generate a large number of false positives. Moreover, when applied to evolving software systems, around 95 % of alarms generated on a version are repeated, i.e., they have also been generated on the previous version. Version-aware static analysis techniques (VSATs) have been proposed to suppress the repeated alarms that are not impacted by the code changes between the two versions. The alarms reported by VSATs after the suppression, called delta alarms, still constitute 63% of the tool-generated alarms. We observe that delta alarms can be further postprocessed using their corresponding code changes: the code changes due to which VSATs identify them as delta alarms. However, none of the existing VSATs or alarms postprocessing techniques postprocesses delta alarms using the corresponding code changes. Based on this observation, we use the code changes to classify delta alarms into six classes that have different priorities assigned to them. The assignment of priorities is based on the type of code changes and their likelihood of actually impacting the delta alarms. The ranking of alarms, obtained by prioritizing the classes, can help suppress alarms that are ranked lower, when resources to inspect all the tool-generated alarms are limited. We performed an empirical evaluation using 9789 alarms generated on 59 versions of seven open source C applications. The evaluation results indicate that the proposed classification and ranking of delta alarms help to identify, on average, 53 % of delta alarms as more likely to be false positives than the others.

Concurrency vulnerabilities caused by synchronization problems will occur in the execution of multi-threaded programs, and the emergence of concurrency vulnerabilities often cause great threats to the system. Once the concurrency vulnerabilities are exploited, the system will suffer various attacks, seriously affecting its availability, confidentiality and security. In this paper, we extract 839 concurrency vulnerabilities from Common Vulnerabilities and Exposures (CVE), and conduct a comprehensive analysis of the trend, classifications, causes, severity, and impact. Finally, we obtained some findings: 1) From 1999 to 2021, the number of concurrency vulnerabilities disclosures show an overall upward trend. 2) In the distribution of concurrency vulnerability, race condition accounts for the largest proportion. 3) The overall severity of concurrency vulnerabilities is medium risk. 4) The number of concurrency vulnerabilities that can be exploited for local access and network access is almost equal, and nearly half of the concurrency vulnerabilities (377/839) can be accessed remotely. 5) The access complexity of 571 concurrency vulnerabilities is medium, and the number of concurrency vulnerabilities with high or low access complexity is almost equal. The results obtained through the empirical study can provide more support and guidance for research in the field of concurrency vulnerabilities.

Researchers have investigated the dark web for various purposes and with various approaches. Most of the dark web data investigation focused on analysing text collected from HTML pages of websites hosted on the dark web. In addition, researchers have documented work on dark web image data analysis for a specific domain, such as identifying and analyzing Child Sexual Abusive Material (CSAM) on the dark web. However, image data from dark web marketplace postings and forums could also be helpful in forensic analysis of the dark web investigation.The presented work attempts to conduct image classification on classes other than CSAM. Nevertheless, manually scanning thousands of websites from the dark web for visual evidence of criminal activity is time and resource intensive. Therefore, the proposed work presented the use of quantum computing to classify the images using a Quantum Convolutional Neural Network (QCNN). Authors classified dark web images into four categories alcohol, drugs, devices, and cards. The provided dataset used for work discussed in the paper consists of around 1242 images. The image dataset combines an open source dataset and data collected by authors. The paper discussed the implementation of QCNN and offered related performance measures.

This paper deals with the problem of image forgery detection because of the problems it causes. Where The Fake im-ages can lead to social problems, for example, misleading the public opinion on political or religious personages, de-faming celebrities and people, and Presenting them in a law court as evidence, may Doing mislead the court. This work proposes a deep learning approach based on Deep CNN (Convolutional Neural Network) Architecture, to detect fake images. The network is based on a modified structure of Xception net, CNN based on depthwise separable convolution layers. After extracting the feature maps, pooling layers are used with dense connection with Xception output, to in-crease feature maps. Inspired by the idea of a densenet network. On the other hand, the work uses the YCbCr color system for images, which gave better Accuracy of %99.93, more than RGB, HSV, and Lab or other color systems.

Server-side web applications are vulnerable to request races. While some previous studies of real-world request races exist, they primarily focus on the root cause of these bugs. To better combat request races in server-side web applications, we need a deep understanding of their characteristics. In this paper, we provide a complementary focus on race effects and fixes with an enlarged set of request races from web applications developed with Object-Relational Mapping (ORM) frameworks. We revisit characterization questions used in previous studies on newly included request races, distinguish the external and internal effects of request races, and relate requestrace fixes with concurrency control mechanisms in languages and frameworks for developing server-side web applications. Our study reveals that: (1) request races from ORM-based web applications share the same characteristics as those from raw-SQL web applications; (2) request races violating application semantics without explicit crashes and error messages externally are common, and latent request races, which only corrupt some shared resource internally but require extra requests to expose the misbehavior, are also common; and (3) various fix strategies other than using synchronization mechanisms are used to fix request races. We expect that our results can help developers better understand request races and guide the design and development of tools for combating request races.

Distributed Denial of Service (DDoS) attacks aim to make a server unresponsive by flooding the target server with a large volume of packets (Volume based DDoS attacks), by keeping connections open for a long time and exhausting the resources (Low and Slow DDoS attacks) or by targeting protocols (Protocol based attacks). Volume based DDoS attacks that flood the target server with a large number of packets are easier to detect because of the abnormality in packet flow. Low and Slow DDoS attacks, however, make the server unavailable by keeping connections open for a long time, but send traffic similar to genuine traffic, making detection of such attacks difficult. This paper proposes a solution to detect and mitigate one such Low and slow DDoS attack, Slowloris in an SDN (Software Defined Networking) environment. The proposed solution involves communication between the detection and mitigation module and the controller of the Software Defined Network to get data to detect and mitigate low and slow DDoS attack.