International Journal of Network Management最新文献

Cyber threat intelligence relies on network telescopes to detect attacks and emerging threats, traditionally utilizing a substantial portion of the IPv4 address space. However, the escalating scarcity and value of this resource force companies and research centers to grapple with the challenge of repurposing their address spaces, potentially impacting cybersecurity effectiveness and hindering research efforts. In this article, we first investigate the historical usage of IPv4 address space in network telescopes and the current reduction trend in several initiatives. Then, we examine the impact of reducing the allocated space on the ability of these systems to identify attackers and attack campaigns. We explore two network telescopes with the intention of assessing the impact of this reduction by quantifying the losses in several ways. Our findings reveal that even halving the allocated space for a network telescope may still permit the detection of 80% of unique cyberattack sources and the address allocation schema has little to no influence on this detection. We also found that most of the proportions and patterns remain present, albeit with reduced intensity.

The rapid growth of networking technology has generated several situations and issues in the field of safeguarding critical brand design data in the present hyper connected context, particularly with the arrival of the 6^th Generation (6G). As brand development relies more on cloud-based services, protecting client data and intellectual property (IP) is essential. By using 6G network slicing architecture, which contains dedicated, secure network sections for brand design services, improved encryption, and anomaly detection systems, the research suggested a solution to such issues. The data includes features such as network performance, security measurements, and user data privacy measures. The methodology entails pre-processing brand design data with Z-score normalization to standardize feature distributions, followed by Principal Component Analysis (PCA) for a decrease of dimensions. The proposed method uses a Fully Homomorphic Encryption Driven Quantum Support Vector Machine (FHE-QSVM) to detect anomalies in real time while assuring safe and efficient resource allocation in dedicated slices. FHE-QSVM anomaly detection model produced significant metrics, with accuracy (98%), recall (96%), precision (97%), and F1-score (96%) data by accurately categorizing threats while maintaining data confidentiality. The finding shows the FHE-QSVM enhances both the security and privacy of brand design data by accurately categorizing threats while maintaining data confidentiality. Overall, this strategy offers a scalable solution for secure AI-powered brand design services, highlighting the importance of creative encryption, real-time monitoring, and 6G network slicing to meet contemporary data security standards.

Rapid growth in intelligent digital media interaction systems (IDMIS) has created new difficulties in controlling and optimizing content distribution and engagement, especially with the impending 6G networks. The purpose of the investigate is to create an intelligent system that uses 6G network slicing to increase digital media communication and user experience through seamless connectivity, dynamic content distribution, and real-time engagement. The structure includes a dynamic, multilayered architecture for IDMIS, and network capital is allocated through 6G network slicing based on user demand and content type. The system includes machine learning (ML) algorithms that predict user behavior and optimize media delivery in real time. To correctly predict user behavior, the research gathers data that capture users' performance and preference (historical interaction data, demographics, contextual data, and user feedback). Once collected, data are processed to reduce dimensionality using principal component analysis (PCA). Refined Support Vector Machine Integrated with Flying Fox Optimization (RSVM-FFO) predicts user behavior and optimizes media delivery in real time. Metrics are used to evaluate the RSVM-FFO approach, such as F1-score (98.12%), accuracy (98.59%), precision (98.57%), and recall (98.17%). The results reveal that the suggested systems considerably improve media interaction effectiveness by reducing latency and bandwidth usage while providing a highly responsive user experience. Finally, advancement in the delivery of high-performance, customized media services is the combination of an IDMIS with 6G network slicing.

Many smart gadgets are connecting to the Internet, and Internet-of-Things (IoT) technologies are enabling a variety of applications. Artificial intelligence (AI) of Things (AIoT) devices are anticipated to possess human-like decision-making, reasoning, perception, and other capacities with the combination of AI and IoT. AIoT gadgets are expected to be extensively utilized across several domains, as anticipated by 6G networks. With AI's steady advancements in speech recognition, computer vision, and natural language processing—not to mention its ability to analyze large amounts of data—semantic communication is now feasible. A new paradigm in wireless communication is opened by semantic communication, which seeks to explore the meaning behind the bits and only transmits the information that may be used, as opposed to attaining error-free transmission. The combination of IoT with AI provides prominent features to overcome various important issues in cloud computing networks. However, there is bottleneck of delay and precision. Therefore, this paper proposed a new method to overcome this problem. First, the network slicing feature maps were extracted by convolutional neural networks. Next, the processing delay is reduced by semantic compression. Simulation results show that the proposed approach makes 99.2% reduction in communication complexity and an 80% reduction in transmission delay as compared with traditional methods. Taking the Resnet18 network as an example, the running time of the semantic communication method is only 0.8% of the traditional method.

The introduction of 6G networks presents substantial challenges for network security, particularly in multi-virtual network topologies. The combination of network function virtualization (NFV) and software-defined networking (SDN) in 6G is designed to increase scalability and flexibility; nevertheless, these advances complicate network security management. The goal is to identify risks to network security and develop defense solutions for 6G multi-virtual network situations. SDN's virtualized network functions (VNFs) are utilized to provide stateful firewall services that provide scalable and dynamic threat prevention. The SDN controller is critical in developing a set of rules to prevent risky network connectivity and decrease possible risks. 6G multi-virtual network domains—attacking threats that involve different socket addresses so complex that usually applicable protection measures hardly tackle that scenario, machine learning (ML) algorithms, and Intelligent Osprey Optimized Versatile Random Forest (IOO-VRF) model—have been proposed for potentially harmful connection detection and predicting cyber threats accessing the network. Multiple open-access sources can be exploited to gather diverse data for collecting valuable information on studying network traffic and cyber threats. The experimental results indicate that IOO-VRF achieved prediction accuracy comparable to that of other traditional algorithms. The proposed model is assessed on various types of metrics, including accuracy (98%), precision (97.4%), recall (94%), and F1-score (93%). The results emphasized the importance of ML in combination with SDN and NFV for security in the case of resilient, expandable, and flexible security measures for future multi-virtual 6G network communications networks.

The rapid growth of vehicle networks in the Internet of Vehicles (IoV) needs novel approaches to optimizing network resource allocation and enhancing traffic management. Sixth-generation (6G) network slicing, when paired with artificial intelligence (AI), has enormous potential in this field. The purpose of this research is to investigate the use of AI-driven 6G network slicing (NS) for efficient usage of resources and accurate traffic prediction in IoV systems. A unique network design is suggested, combining data-driven approaches and dynamic network slicing. Data are acquired from vehicular sensors and traffic monitoring systems, and log transformation is used to handle exponential growth patterns like vehicle counts and congestion levels. The Fourier transform (FT) is used to extract frequency-domain information from traffic data, which allows for the detection of periodic patterns, trends, and anomalies such as vehicle velocity and traffic density. The Dipper Throated Optimized Efficient Elman Neural Network (DTO-EENN) is used to forecast traffic and optimize resources. This technology allows the system to predict traffic patterns and dynamically alter network slices to ensure optimal resource allocation while reducing latency. The results show that the suggested AI-driven NS technique increases forecast accuracy and network performance while dramatically reducing congestion levels. The research indicates that AI-driven 6G based NS offers a solid framework for optimizing IoV performance.

Although many studies have conducted the traffic scheduling of time-sensitive networks, most focus on small-scale static scheduling for specific scenarios, which cannot cope with dynamic and rapid scheduling of time-triggered (TT) flows generated in scalable scenarios in the Industrial Internet of Things. In this paper, we propose a Scalable TT flow scheduling method based on Dynamic Online Grouping in industrial time-sensitive networks (SDOG). To achieve that, we establish an undirected weighted flow graph based on the conflict index between TT flows and divide available time into equally spaced time windows. We dynamically group the TT flows within each window locally. When the number of flows to be scheduled doubles, we can achieve scalable and efficient solutions to efficiently schedule dynamic TT flows, avoiding unnecessary conflicts during data communication. In addition, a topology pruning strategy is adopted to prune the network topology of each group, reducing unnecessary link variables and further effectively shortening the scheduling time. Experimental results validated our expected performance and demonstrated that our proposed SDOG scheduling method has advantages in terms of overall traffic schedulability and average time for scheduling individual traffic.