Journal of Network and Computer Applications最新文献

Multi-access edge computing (MEC), deploying cloud infrastructures proximate to end-devices and reducing latency, takes pivotal roles for mission-critical services such as smart grids, self-driving cars, and healthcare. Ensuring fault-tolerance is paramount for mission-critical services, as failures in these services can lead to fatal accidents and blackouts. However, the distributed nature of MEC architectures makes them more susceptible to failures than traditional cloud systems. Existing research in this field has focused on enhancing robustness to prevent failures in MEC systems rather than restoring them from failure conditions. To bridge this gap, we introduce Fatriot, a SmartNIC-based architecture designed to ensure fault-tolerance in MEC systems. Fatriot actively monitors for anomalies on MEC hosts and seamlessly redirects incoming service traffic to backup hosts upon detecting failures. Operating as a stand-alone solution on a SmartNIC, Fatriot guarantees the continuous operation of its fault-tolerance mechanism, even during severe errors (e.g., kernel failure) on the MEC host, maintaining uninterrupted service in mission-critical services. Our prototype of Fatriot, implemented on the NetFPGA-SUME, demonstrates effective mitigation of various failure scenarios, achieving this with minimal overhead to services (less than 1%).

Modern radio access networks (RANs) are both highly complex and potentially vulnerable to unauthorized security setting changes. A RAN is studied in a proof-of-concept experiment to demonstrate that an unauthorized network state is detectable at layers in the RAN architecture away from the source of the state setting. Specifically, encryption state is set at the packet data convergence protocol (PDCP) layer in the Long-Term Evolution (LTE) network model and an anomalous cipher-$\mathrm{OFF}$ state is shown to be detectable at the physical layer. Three tranches of experimental data totaling 1,987 runs and each involving 285 measurands were collected and used to construct and demonstrate single-feature, multi-feature, and multi-run encryption state detectors. These detectors show a range of performances with the single-feature detector based on reference signal received quality achieving near-0% false alarms and near-100% true detections. Multi-run averaging detectors show similar low-error performance, even just based on marginally effective detector features. The detectors’ performances are studied across the three tranches of experimental data and found by multiple complementary measures to be generalizable provided the testbed protocol is carefully controlled. Essential to these results was an automated, comprehensively instrumented experiment testbed in which measurands were treated as distributions.

The advent of blockchain technology has brought major changes to traditional centralized storage. Therefore, various fields have begun to study the application and development of blockchain. However, blockchain technology has a serious shortcoming of data bloating. The reason is that blockchain technology achieves decentralization by storing complete blockchain data at each node, incurring a significant amount of blockchain data. Therefore, each node must spend significant amount of storage space and initialization synchronization time. To solve the above problems, in this research, we propose a secure and agile synchronization framework for low storage blockchains. First, we design a K-extreme segment algorithm, which reduces the synchronization time by returning only the first and last $k$ blocks of each block segment at once to the local storage. Next, we decentrally store the block data of the blockchain by IPFS and establish a backup mechanism by IPFS-cluster. Finally, due to use of distributed storage, the nodes must request un-stored block data from IPFS, causing an increase in the throughput of the blockchain network. To avoid network congestion, we propose the working set algorithm to improve the hit ratio of the local storage and reduce the number of requests to decrease throughput. In summary, our experiments demonstrate that the ratio of full nodes to low storage nodes is significantly lower for nodes with higher storage limits compared to those with lower storage limits. In other words, a higher storage limit results in more low storage nodes which can be permitted to ensure that the blockchain network is robust and reliable. Therefore, our proposed framework can provide reliable low storage nodes for the blockchain. The node can reduce the local storage pressure and can still maintain the full functionality of blockchains.

Intrusion detection systems (IDSs) that continuously monitor data flow and take swift action when attacks are identified safeguard networks. Conventional IDS exhibit limitations, such as reduced detection rates and increased computational complexity, attributed to the redundancy and substantial correlation of network data. Ensemble learning (EL) is effective for detecting network attacks. Nonetheless, network traffic data and memory space requirements are typically significant. Therefore, deploying the EL approach on Internet-of-Things (IoT) devices with limited memory is challenging. In this paper, we use feature importance (FI), a filter-based feature selection technique for feature dimensionality reduction, to reduce the feature dimensions of an IoT/IIoT network traffic dataset. We also employ lightweight stacking ensemble learning (SEL) to appropriately identify network traffic records and analyse the reduced features after applying FI to the dataset. Extensive experiments use the Edge-IIoTset dataset containing IoT and IIoT network records. We show that FI reduces the storage space needed to store comprehensive network traffic data by 86.9%, leading to a significant decrease in training and testing time. Regarding accuracy, precision, recall, training and test time, our classifier that utilised the eight best dataset features recorded 87.37%, 90.65%, 77.73%, 80.88%, 16.18 s and 0.10 s for its overall performance. Despite the reduced features, our proposed SEL classifier shows insignificant accuracy compromise. Finally, we pioneered the explanation of SEL by using a decision tree to analyse its performance gain against single learners.

In recent years, we have seen a significant enhancement in the performance of standard Transmission Control Protocol (TCP) congestion control algorithms. The number of packet drops and high round-trip time (RTT) are indications of network congestion. Many congestion control mechanisms have been proposed to overcome the challenge of achieving increased throughput and reduced latency. We have reviewed many TCP congestion control algorithms which are discussed in the literature. The limitation of the existing work is a trade-off between throughput, loss ratio, and delay. It is not possible for any algorithm to outperform the existing algorithm in terms of all the performance measures. We attempt to achieve the best performance while our proposed algorithm competes with CUBIC and Bottleneck Bandwidth and Round-trip propagation time (BBR). According to the observed results in the literature, TCP Veno dominates among other existing algorithms. We have proposed a bandwidth-delay product (BDP) based TCP (BDP-Veno) congestion control algorithm by modifying Veno to incorporate the information of BDP of the bottleneck. The proposed algorithm is implemented using ns-2. Moreover, we have analyzed the performances of standard TCP congestion control algorithms by considering different network scenarios. Our proposed algorithm performs better compared to other existing TCP congestion control schemes such as Reno, Newreno, BIC, CUBIC, Vegas, Veno, and Compound TCP in terms of average throughput in most of the scenarios. In Scenario 1, our proposed algorithm enhances the throughput with respect to Veno by 57%. Further, we have also compared the throughput with BBR using ns3 where we receive comparable throughput with BBR.

The Internet of Things (IoT) has emerged as a promising paradigm which facilitates the seamless integration of physical devices and digital systems, thereby transforming multiple sectors such as healthcare, transportation, and urban planning. This paradigm is also known as ad-hoc networks. IoT is characterized by several pieces of equipment called objects. These objects have different and limited capacities such as battery, memory, and computing power. These limited capabilities make it difficult to design routing protocols for IoT networks because of the high number of objects in a network. In IoT, objects often have data which does not belong to them and which should be sent to other objects, then leading to a problem known as permutation routing problems. The solution to that problem is found when each object receives its items. In this paper, we propose a new approach to addressing the permutation routing problem in single-hop IoT networks. To this end, we start by representing an IoT network as an oriented graph, and then, based on a reservation channel protocol, we first define a permutation routing protocol for an IoT in a single channel. Secondly, we generalize the previous protocol to make it work in multiple channels. Routing is done using graph theory approaches. The obtained results show that the wake-up times and activities of IoT objects are greatly reduced, thus optimizing network lifetime. This is an effective solution for the permutation routing problem in IoT networks. The proposed approach considerably reduces energy consumption and computation time. It saves 5.2 to 32.04% residual energy depending on the number of items and channels used. Low energy and low computational cost demonstrate that the performance of circuit-based and oriented graph theory is better than the state-of-the-art protocol and therefore is a better candidate for the resolution of the permutation routing problem in single-hop environment.

The cloud computing infrastructure is a distributed environment and the existing research works have some provisioning problems such as suboptimal resource utilization and high execution time. The Heterogeneity Resource Management Framework for Dynamic Resource Provisioning (HRMF-DRP) is proposed for focusing on task scheduling and workload management. This framework incorporates advanced algorithms for dataset preprocessing, task clustering, workload prediction, and dynamic resource provisioning. For data preprocessing, the real-world workload traces were captured from the Planet Lab dataset that are taken as input for the preprocessing stage. The data preprocessing is responsible for ensuring data quality and reliability by using different models like missing data handling, outlier detection and removal as well as standardization and normalization. In this paper, the tasks are grouped into clusters by utilizing Density-Based Spatial Clustering of Applications with Noise (DBSCAN) model and this model categorizes the data points into border points, core points and noise points based on their density. The temporal dependencies are captured for the workload prediction by using Long Short-Term Memory (LSTM) neural network model. A Gaussian Mixture Model (GMM) model is responsible for estimating the number of Virtual machines (VMs) present in the workload prediction process. The Self-Adaptive Genetic Algorithm (SAGA) is implemented for task mapping that adjusts the parameters to change workload patterns for contributing adaptability and robustness. The different experimental evaluations are conducted based on the task completion time, workload balance index, resource utilization efficiency and workload prediction accuracy. The proposed model achieved the workload prediction accuracy of 98.5%, cost of $89.6, execution time of 125ms, Task Completion Time (TCT) of 40ms, Workload Balance Index (WBI) of 0.96 and Resource Utilization Efficiency (RUE) of 0.93. The quantitative results collectively position HRMF-DRP as a practical and efficient solution, promising advancements in dynamic resource provisioning for cloud computing, particularly within the Infrastructure as a Service (IaaS) cloud model.

In recent years, the application of blockchain technology in the Unmanned Aerial Vehicle (UAV) industry has shown promise in making a substantial impact on various aspects of the field. Blockchain can provide key solutions to several challenges related to security, data integrity, and operational efficiency within UAV systems. In this paper, we conduct an in-depth investigation of the transformative role of blockchain in the UAV industry. Through a comprehensive literature review, we examine the potential impact and applications of blockchain technology in this field, with a particular focus on its capacity to address challenges across the manufacturing, planning, and operational phases of UAV systems. We explore how blockchain implementation within UAV networks enhances secure data traceability within supply chain processes and facilitates more efficient flight operations management. Our findings reveal that blockchain technology significantly improves data traceability and operational efficiency in UAV systems, offering robust solutions to challenges related to trust, transparency, data integrity, and access control within UAV networks, thereby enhancing overall system reliability and performance. Furthermore, we highlight some of the future potential opportunities and use cases for blockchain in the UAV industry, including real-time data management and decentralized verification mechanisms. We discuss the primary challenges obstructing the widespread adoption of blockchain in this industry and also propose some future research directions.

Nowadays, web robots are predominantly used for auto-accessing web content, sharing almost one-third of the total web traffic and often posing threats to various web applications’ security, privacy, and performance. Detecting these robots is essential, and both online and offline methods are employed. One popular offline method is the use of weblog feature-based automated learning. However, this method alone cannot accurately identify web robots that continuously evolve and camouflage. Web content features combined with weblog features are used to detect such robots based on the assumption that human users exhibit specific interests while robots randomly navigate web pages. State-of-the-art web content-based feature methods lack the ability to generate coherent topics, which can confound the performance of classification models. Therefore, we propose a new content semantic feature extraction method that uses the LDA2Vec topic model, combining the strengths of LDA and the Word2Vec model to produce more semantically coherent topics by exploiting website content for a web session. To effectively detect web robots, web resource content semantic features are combined with log-based features in the proposed web robot detection approach. The proposed approach is evaluated in an e-commerce website access logs and content data. The F-score, balanced accuracy, G-mean, and Jaccard similarity are used for performance measures, and the coherence score metric is used to determine the number of topics for a session. Experimental results demonstrate that a combination of weblogs and content semantic features is effective in web robot detection.

Analyzing the geographical information for the Plateau Lake region with remote sensing images (RSI) is an emerging technology to monitor the changes of the ecological environment. To alleviate the requirement of abundant labels for supervised RSI segmentation, the Cross-consistency Guiding Semi-supervised Learning (SSL) Semantic Segmentation Network is proposed, and it can perform high-quality multi-category semantic segmentation for complex remote sensing scenes with limited quantity of labeled images. Firstly, based on the SSL semantic segmentation framework, through the cross-consistency method training a teacher model with less annotated images and plentiful unannotated images, then generating higher-quality pseudo labels to guide the learning process of the student model. Secondly, dense conditional random field and mask hole repair are used to patch and fill the flaw areas of pseudo-labels based on the pixel features of position, color, and texture, further improving the granularity and reliability of the student model training dataset. Additionally, to improve the accuracy of the model, we designed a strong data augmentation (SDA) method based on a stochastic cascaded strategy, which connects multiple augmentation techniques in random order and probability cascade to generate new training samples. It mimics a variety of image transformations and noise conditions that occur in the real world to enhance the robustness in complex scenarios. To validate the effectiveness of CGSNet in complex remote sensing scenes, extended experiments are conducted on the self-built plateau lake RSI dataset and two public multi-category RSI datasets. The experiment results demonstrate that, compared with other state-of-the-art SSL methods, the proposed CGSNet achieves the highest 77.47% mIoU and 87.06% F1 scores with a limited quantity of annotated data.