Biblio

Malicious cybersecurity activities have become increasingly worrisome for individuals and companies alike. While machine learning methods like Graph Neural Networks (GNNs) have proven successful on the malware detection task, their output is often difficult to understand. Explainable malware detection methods are needed to automatically identify malicious programs and present results to malware analysts in a way that is human interpretable. In this survey, we outline a number of GNN explainability methods and compare their performance on a real-world malware detection dataset. Specifically, we formulated the detection problem as a graph classification problem on the malware Control Flow Graphs (CFGs). We find that gradient-based methods outperform perturbation-based methods in terms of computational expense and performance on explainer-specific metrics (e.g., Fidelity and Sparsity). Our results provide insights into designing new GNN-based models for cyber malware detection and attribution.

Tourism is one of the main sources of income in Australia. The number of tourists will affect airlines, hotels and other stakeholders. Predicting the arrival of tourists can make full preparations for welcoming tourists. This paper selects Queensland Tourism data as intelligent data. Carry out data visualization around the intelligent data, establish seasonal ARIMA model, find out the characteristics and predict. In order to improve the accuracy of prediction. Based on the tourism data around Queensland, build a 10 layer Back Propagation neural network model. It is proved that the network shows good performance for the data prediction of this paper.

With the development of software defined network and network function virtualization, network operators can flexibly deploy service function chains (SFC) to provide network security services more than before according to the network security requirements of business systems. At present, most research on verifying the correctness of SFC is based on whether the logical sequence between service functions (SF) in SFC is correct before deployment, and there is less research on verifying the correctness after SFC deployment. Therefore, this paper proposes a method of using Colored Petri Net (CPN) to establish a verification model offline and verify whether each SF deployment in SFC is correct after online deployment. After the SFC deployment is completed, the information is obtained online and input into the established model for verification. The experimental results show that the SFC correctness verification method proposed in this paper can effectively verify whether each SF in the deployed SFC is deployed correctly. In this process, the correctness of SF model is verified by using SF model in the model library, and the model reuse technology is preliminarily discussed.

Insider threats have high risk and concealment characteristics, which makes traditional anomaly detection methods less effective in insider threat detection. Existing detection methods ignore the logical relationship between user behaviors and the consistency of behavior sequences among homogeneous users, resulting in poor model effects. We propose an insider threat detection method based on internal user heterogeneous graph embedding. Firstly, according to the characteristics of CERT data, comprehensively consider the relationship between users, the time sequence, and logical relationship, and construct a heterogeneous graph. In the second step, according to the characteristics of heterogeneous graphs, the embedding learning of graph nodes is carried out according to random walk and Word2vec. Finally, we propose an Insider Threat Detection Design (ITDD) model which can map and the user behavior sequence information into a high-dimensional feature space. In the CERT r5.2 dataset, compared with a variety of traditional machine learning methods, the effect of our method is significantly better than the final result.

A malicious insider threat is more vulnerable to an organization. It is necessary to detect the malicious insider because of its huge impact to an organization. The occurrence of a malicious insider threat is less but quite destructive. So, the major focus of this paper is to detect the malicious insider threat in an organization. The traditional insider threat detection algorithm is not suitable for real time insider threat detection. A supervised learning-based anomaly detection technique is used to classify, predict and detect the malicious and non-malicious activity based on highest level of anomaly score. In this paper, a framework is proposed to detect the malicious insider threat using supervised learning-based anomaly detection. It is used to detect the malicious insider threat activity using One-Class Support Vector Machine (OCSVM). The experimental results shows that the proposed framework using OCSVM performs well and detects the malicious insider who obtain huge anomaly score than a normal user.

Named in-network computing is an emerging technology of Named Data Networking (NDN). Through deploying the named computing services/functions on NDN router, the router can utilize its free resources to provide nearby computation for users while relieving the pressure of cloud and network edge. Benefitted from the characteristic of named addressing, named computing services/functions can be easily discovered and migrated in the network. To implement named in-network computing, integrating the computing services as Virtual Machines (VMs) into the software router is a feasible way, but how to effectively deploy the service VMs to optimize the local processing capability is still a challenge. Focusing on this problem, we first give the design of NDN-enabled software router in this paper, then propose a service earning based named service deployment scheme (SE-NSD). For available service VMs, SE-NSD not only considers their popularities but further evaluates their service earnings (processed data amount per CPU cycle). Through modelling the deployment problem as the knapsack problem, SE-NSD determines the optimal service VMs deployment scheme. The simulation results show that, comparing with the popularity-based deployment scheme, SE-NSD can promote about 30% in-network computing capability while slightly reducing the service invoking RTT of user.

Based on the characteristics of the new power system with many points, wide range and unattended, this paper studies the specific Cyberspace security risks faced by the disease control side, the station side and the site side, and proposes a new power system Cyberspace security assurance system of “integration of collection, network, side, end, industry and people”. The site side security access measures, the site side civil air defense technology integration measures, the whole business endogenous security mechanism, the whole domain communication security mechanism, the integrated monitoring and early warning and emergency response mechanism are specifically adopted to form a comprehensive integrated security mechanism for the new power system, form a sustainable protection model, effectively improve the security capability, while taking into account the cost and operational complexity of specific implementation links, Provide comprehensive guarantee capability for the safe operation of the new power system.

As an important pillar of social informatization, e-government not only provides more convenient services for the public, but also effectively improves administrative efficiency. At the same time, the application of cloud computing technology also urgently requires the government to improve the level of digital construction. This paper proposes the concept of e-government based on cloud computing, analyze the possible hidden dangers that cloud computing brings to e-government in management, technology, and security, and build cloud computing e-government information security system from three aspects: cloud security management, cloud security technology, and cloud security assurance.

Digital identity management system is the securi-ty infrastructure of computer and internet applications. However, currently, most of the digital identity management systems are faced with problems such as the difficulty of cross-domain authentication and interoperation, the lack of credibility of identity authentication, the weakness of the security of identity data. Although the advantages of block-chain technology have attached the attentions of experts and scholars in the field of digital identity management and many digital identity management systems based on block-chain have been built, the systems still can't completely solve the problems mentioned above. Therefore, in this pa-per, an effective digital identity management system model is proposed which combines technologies of self-sovereign identity and oracle with blockchain so as to pave a way in solving the problems mentioned above and constructing a secure and reliable digital identity management system.

In recent years, generative adversarial networks (GAN) have become a research hotspot in the field of deep learning. Researchers apply them to the field of anomaly detection and are committed to effectively and accurately identifying abnormal images in practical applications. In anomaly detection, traditional supervised learning algorithms have limitations in training with a large number of known labeled samples. Therefore, the anomaly detection model of unsupervised learning GAN is the research object for discussion and research. Firstly, the basic principles of GAN are introduced. Secondly, several typical GAN-based anomaly detection models are sorted out in detail. Then by comparing the similarities and differences of each derivative model, discuss and summarize their respective advantages, limitations and application scenarios. Finally, the problems and challenges faced by GAN in anomaly detection are discussed, and future research directions are prospected.

With the progress of cryptography computer science, designing cryptographic algorithms using deep learning is a very innovative research direction. Google Brain designed a communication model using generation adversarial network and explored the encrypted communication algorithm based on machine learning. However, the encrypted communication model it designed lacks quantitative evaluation. When some plaintexts and keys are leaked at the same time, the security of communication cannot be guaranteed. This model is optimized to enhance the security by adjusting the optimizer, modifying the activation function, and increasing batch normalization to improve communication speed of optimization. Experiments were performed on 16 bits and 64 bits plaintexts communication. With plaintext and key leak rate of 0.75, the decryption error rate of the decryptor is 0.01 and the attacker can't guess any valid information about the communication.

We show that the most common flavors of noisy leakage can be simulated in the information-theoretic setting using a single query of bounded leakage, up to a small statistical simulation error and a slight loss in the leakage parameter. The latter holds true in particular for one of the most used noisy-leakage models, where the noisiness is measured using the conditional average min-entropy (Naor and Segev, CRYPTO’09 and SICOMP’12). Our reductions between noisy and bounded leakage are achieved in two steps. First, we put forward a new leakage model (dubbed the dense leakage model) and prove that dense leakage can be simulated in the information-theoretic setting using a single query of bounded leakage, up to small statistical distance. Second, we show that the most common noisy-leakage models fall within the class of dense leakage, with good parameters. Third, we prove lower bounds on the amount of bounded leakage required for simulation with sub-constant error, showing that our reductions are nearly optimal. In particular, our results imply that useful general simulation of noisy leakage based on statistical distance and mutual information is impossible. We also provide a complete picture of the relationships between different noisy-leakage models. Our result finds applications to leakage-resilient cryptography, where we are often able to lift security in the presence of bounded leakage to security in the presence of noisy leakage, both in the information-theoretic and in the computational setting. Remarkably, this lifting procedure makes only black-box use of the underlying schemes. Additionally, we show how to use lower bounds in communication complexity to prove that bounded-collusion protocols (Kumar, Meka, and Sahai, FOCS’19) for certain functions do not only require long transcripts, but also necessarily need to reveal enough information about the inputs.

In order to control users’ access to the information system, it is necessary to develop a security system that can work in real time and easily reconfigure. This problem can be solved using a fuzzy logic. In this paper the authors propose a fuzzy distribution system for access to the student assessment system, which takes into account the level of user access, identifier and the risk of attack during the request. This approach allows process fuzzy or incomplete information about the user and implement a sufficient level of confidential information protection.

In Wireless Sensor Networks (WSNs), energy and security are two critical concerns that must be addressed. Because of the scarcity of energy, several security measures are restricted. For secure data routing in WSN, it becomes vital to identify insider packet drop attacks. The trust mechanism is an effective strategy for detecting this assault. Each node in this system validates the trustworthiness of its neighbors before transmitting packets, ensuring that only trust-worthy nodes get packets. With such a trust-aware scheme, however, there is a risk of false alarm. This work develops an adaptive trust computation model (TCM)which is implemented in our already proposed Chimp Optimization Algorithm-based Energy-Aware Secure Routing Protocol (COA-EASRP) for WSN. The proposed technique computes the optimal path using the hybrid combination of COA-EASRP and AODV as well as TCM is used to indicate false alarms in detecting selfish nodes. Our Proposed approach provides the series of Simulation outputs carried out based on various parameters

Securing optical edge devices across an optical network is a critical challenge for the technological capabilities of fog/cloud computing. Locating and blocking rogue devices from transmitting data frames in an optical network is a significant security problem due to their widespread distribution over the optical fog cloud. A malicious actor might simply compromise such a device and execute assaults that degrade the optical channel’s Quality. In this study, we advocate an innovative framework for the use of an optical network to facilitate cloud and fog computing in a safe environment. This framework is sustainable and able to detect hostile equipment in optical fog and cloud and redirect it to a honeypot, where the assault may be halted and analyzed. To do this, it employs a model based on a two-stage hidden Markov, a fog manager based on an intrusion detection system, and an optical virtual honeypot. An internal assault is mitigated by simulated testing of the suggested system. The findings validate the adaptable and affordable access for cloud computing and optical fog.

Under the background of rapid development of campus informatization, the information management of college students' innovative activities is slightly outdated, and the operation of the traditional innovative activity record system has gradually become rigid. In response to this situation, this paper proposes a B/S architecture-based information management system for college students' innovative activities based on the current situation that the network and computers are widely used, which is designed for the roles of relevant managers of students on campus, such as class teachers, teachers and counselors, and has developed various functions to meet the needs of such users as class teachers, including user The system is designed to meet the needs of classroom teachers, classroom teachers and tutors. In order to meet the requirements of generality, expandability and ease of development, the overall architecture of the system is based on the javaEE platform, with JSP technology as the main development technology.

Aim: Data is secured in the cloud using Elliptic Curve Cryptography (ECC) compared with Advanced Encryption Standard (AES) with improved time efficiency. Materials and Methods: Encryption and decryption time is performed with files stored in the cloud. Protecting data with improved time efficiency is carried out using ECC where the number of samples (\textbackslashmathrmN=6) and AES (\textbackslashmathrmN=6), obtained using the G-power value of 80%. Results: Mean time of ECC is 0.1683 and RSA is 0.7517. Significant value for the proposed system is 0.643 (\textbackslashmathrmp \textgreater 0.05). Conclusion: Within the limit of study, ECC performs faster in less consumption time when compared to AES.

New advancements in cloud computing technology enable the usage of cloud platforms for business purposes rapidly increasing every day. Data accumulation related to business transactions, Communications, business model architecture and much other information are stored in the cloud platform and access Dubai the business Associates commonly. Considering the security point of view data stored in the cloud need to be highly secured and accessed through authentication. The proposed system is focused on evaluating a cloud integrity auditing model in which the security and privacy preserving system is being audited, privacy is decided using a machine learning algorithm. The proposed model is developed using a hybrid CatBoost algorithm (HCBA) in which the input data is stored into the cloud platform using Bring your own encryption Key (BYOEK). The security of BYOEK model is evaluated and validated with respect to the given test model in terms of Execution time comparison Vs. Data transactions.

Attribute Based Encryption that solely decrypts the cipher text's secret key attribute. Patient information is maintained on trusted third party servers in medical applications. Before sending health records to other third party servers, it is essential to protect them. Even if data are encrypted, there is always a danger of privacy violation. Scalability problems, access flexibility, and account revocation are the main security challenges. In this study, individual patient health records are encrypted utilizing a multi-authority ABE method that permits a multiple number of authorities to govern the attributes. A strong key generation approach in the classic Attribute Based Encryption is proposed in this work, which assures the robust protection of health records while also demonstrating its effectiveness. Simulation is done by using CloudSim Simulator and Statistical reports were generated using Cloud Reports. Efficiency, computation time and security of our proposed scheme are evaluated. The simulation results reveal that the proposed key generation technique is more secure and scalable.

Security troubles of restricted sources communications are vital. Existing safety answers aren't sufficient for restricted sources gadgets in phrases of Power Area and Ef-ficiency‘. Elliptic curves cryptosystem (ECC) is area efficent for restricted sources gadgets extra than different uneven cryp-to systems because it gives a better safety degree with equal key sizes compared to different present techniques. In this paper, we studied a lightweight hybrid encryption technique that makes use of set of rules primarily based totally on AES for the Plain text encription and Elliptic Curve Diffie-Hellman (ECDH) protocol for Key encryption. The simplicity of AES implementation makes it light weight and the complexity of ECDH make it secure. The design is simulated using Spyder Tool, Modelsim and Implemented using Xilinx Vivado the effects display that the proposed lightweight Model offers a customary security degree with decreased computing capacity. we proposed a key authentication system for enhanced security along with an Idea to implement the project with multimedia input on FPGA

The problem of information privacy has grown more significant in terms of data storage and communication in the 21st century due to the technological explosion during which information has become a highly important strategic resource. The idea of employing DNA cryptography has been highlighted as a potential technology that offers fresh hope for unbreakable algorithms since standard cryptosystems are becoming susceptible to assaults. Due to biological DNA's outstanding energy efficiency, enormous storage capacity, and extensive parallelism, a new branch of cryptography based on DNA computing is developing. There is still more study to be done since this discipline is still in its infancy. This work proposes a DNA encryption strategy based on cryptographic key generation techniques and chaotic diffusion operation.

This paper examines how the extent of the network has expanded from the traditional computer Internet to the field of edge computing based on mobile communication technology with the in-depth development of the mobile Internet and the Internet of Things. In particular, the introduction of 5G has enabled massive edge computing nodes to build a high-performance, energy-efficient and low-latency mobile edge computing architecture. Traditional network security technologies and methods are not fully applicable in this environment. The focus of this paper is on security protection for edge networks. Using virtualized networks builds a dynamic heterogeneous redundancy security model (i.e., DHR). It first designs and evaluates the DHR security model, then constructs the required virtualized heterogeneous entity set, and finally constructs a DHR-based active defense scheme. Compared with existing network security solutions, the security protection technology of the edge network studied this time has a better protective effect against the unknown security threats facing the edge network.

Train operation is highly influenced by the rail track state and the surrounding environment. An abnormal obstacle on the rail track will pose a severe threat to the safe operation of urban rail transit. The existing general obstacle detection approaches do not consider the specific urban rail environment and requirements. In this paper, we propose an edge intelligence (EI)-based obstacle intrusion detection system to detect accurate obstacle intrusion in real-time. A two-stage lightweight deep learning model is designed to detect obstacle intrusion and obtain the distance from the train to the obstacle. Edge computing (EC) and 5G are used to conduct the detection model and improve the real-time detection performance. A multi-agent reinforcement learning-based offloading and service migration model is formulated to optimize the edge computing resource. Experimental results show that the two-stage intrusion detection model with the reinforcement learning (RL)-based edge resource optimization model can achieve higher detection accuracy and real-time performance compared to traditional methods.

The safety of the power system is inherently vital, due to the high risk of the electronic power system. In the wave of digitization in recent years, many power systems have been digitized to a certain extent. Under this circumstance, network security is particularly important, in order to ensure the normal operation of the power system. However, with the development of the Internet, network security issues are becoming more and more serious. Among all kinds of network attacks, the Distributed Denial of Service (DDoS) is a major threat. Once, attackers used huge volumes of traffic in short time to bring down the victim server. Now some attackers just use low volumes of traffic but for a long time to create trouble for attack detection. There are many methods for DDoS detection, but no one can fully detect it because of the huge volumes of traffic. In order to better detect DDoS and make sure the safety of electronic power system, we propose a novel detection method based on neural network. The proposed model and its service are deployed to the edge cloud, which can improve the real-time performance for detection. The experiment results show that our model can detect attacks well and has good real-time performance.

In recent years, differential privacy has gradually become a standard definition in the field of data privacy protection. Differential privacy does not need to make assumptions about the prior knowledge of privacy adversaries, so it has a more stringent effect than existing privacy protection models and definitions. This good feature has been used by researchers to solve the in-depth learning problem restricted by the problem of privacy and security, making an important breakthrough, and promoting its further large-scale application. Combining differential privacy with BEGAN, we propose the DP-BEGAN framework. The differential privacy is realized by adding carefully designed noise to the gradient of Gan model training, so as to ensure that Gan can generate unlimited synthetic data that conforms to the statistical characteristics of source data and does not disclose privacy. At the same time, it is compared with the existing methods on public datasets. The results show that under a certain privacy budget, this method can generate higher quality privacy protection data more efficiently, which can be used in a variety of data analysis tasks. The privacy loss is independent of the amount of synthetic data, so it can be applied to large datasets.