Development Trend Of Network Research Articles

Temporal Heterogeneous Information Network Embedding via Semantic Evolution

Real-world networks are often heterogeneous and constantly changing over time. Evolution reveals the trend of network development, which is vital for predicting its future state, and network embedding can effectively learn the information from it. Nevertheless, previous works only consider the impact of meta-path instances or node neighbors on the network dynamics but ignore the relationship between them, and hence the hidden semantic information is missed, which will result in performance deterioration. Therefore, we propose a novel temporal heterogeneous information network embedding method (SemE), which abstracts the instance of the meta-path as semantic units and then considers the interaction between them to discover deeper semantic information. Specifically, we first construct semantic networks by the Ethernet topology and the interaction between semantic units. The semantic units are sampled based on a pre-designed meta-path-guided random walk. To further capture the semantic evolution of the semantic network, we learn the embedding of nodes by the attention-Hawkes process. Finally, we generate the final embedding by aggregating the structure, semantic and temporal information with the attention mechanism. Experiments on three real-world temporal heterogeneous information networks show that our model performs better than competitive counterparts.

Trusted Data Sharing Mechanism Based on Blockchain and Federated Learning in Space-Air-Ground Integrated Networks

Network data is distributed data on electricity, with the explosive growth of network data, and it has become an inevitable trend of network development to synergized and shared crossdomain scattered data and enhances the value transmission of network data. Federated learning, as a technology that combines data value delivery and data privacy security, is widely concerned in the process of data sharing. However, currently federated learning is used within a single business system. In the process of crossdomain data sharing, how to ensure the data trust, model trust, and result trust of federated learning is still an urgent problem to be solved. To this end, we designed to use blockchain structure to record each behavior of data sharing. Based on its tamper-proof and traceability, combined with cryptography technology, we constructed an endogenous trusted architecture for crossdomain data sharing. In addition, a reverse auction node incentive mechanism based on high credit preference is designed to solve the common problems in data sharing, such as low enthusiasm of users in sharing, unstable data quality of contributions, and unreasonable distribution of data sharing benefits. Through theoretical analysis and experimental verification, it can be seen that the incentive mechanism designed in this paper can meet the authenticity, user rationality, and budget feasibility. On this basis, it can motivate users to participate in data sharing, improve the average quality of data shared by users, and ensure security and trustworthiness and resist malicious attacks to a certain extent.

Resource Allocation Algorithm Based on Modeling of Ubiquitous Network Node Capability

With the continuous development and progress of the Internet, computing power is gradually moving from the center to the edge. In order to perceive, measure, schedule and manage the distributed computing resources, it is necessary to model and evaluate the node capability. Based on the development trend of network computing convergence and the demand of network computing service scheduling, this paper studies the unified measurement model and resource allocation algorithm for the integration of network computing and heterogeneous resources. The node resources are divided into four dimensions: computing, communication, memory and storage, and on this basis, the comprehensive ability evaluation index is proposed. Through simulation analysis, the effectiveness of the resource allocation algorithm based on node capability modeling is verified.

The blockchain based access control scheme for the Internet of Things

The Internet of Things (IoT) is an information carrier based on the Internet and traditional telecommunications networks. Access control is an important part of the service architecture, which can effectively solve the security issues in IoT. Therefore, many solutions adopt the access control to solve some security problems. However, these existing solutions cannot easily deploy access control in the IoT. As a decentralized, non-tamperable, and highly credible distributed ledger technology, blockchain is in line with the current trend of network development. To improve the transparency, non-tampering of the access control process and traceability and auditability of access flow changes, we propose a blockchain-based access control scheme for the IoT, which uses a combination of on-chain and off-chain to reduce the storage pressure of the blockchain. On the one hand, we deploy policy decision points (PDP) and other components on the smart contract in the form of chain code to better judge and execute decision-making. On the other hand, we put policy administration points (PAP) in the off-chain organization to reduce the storage pressure of the blockchain and make better access flow transformation and traceability. Ultimately, we concluded through feasibility analysis that our solution can maintain high throughput in a large-scale transaction volume environment.

SDRM‐LDP: A Recommendation Model Based on Local Differential Privacy

The development of 5G technology has driven the rise of e‐commerce, social networking, and the Internet of Things. Under the high‐speed transmission, the data volume increases, and the user demand also changes. Personalized customization has become the mainstream trend of network development. However, as the speed of the Internet increases, a series of problems also arise. The increase in data volume results in a reduction of bandwidth, a growth of the central processor’s pressure, and a higher risk of data leakage. A search system and a recommendation platform are the tools to improve people’s search efficiency. However, providing personalized recommendations to different users according to their needs is still an urgent problem. Simultaneously, the big data volume means that attackers can also get more information. They can use background knowledge and various reasoning methods to deduce the user’s private information using nonprivate items. In this paper, the solutions to safe and reliable recommendation services are the main problem explored. Based on this idea, this paper proposed short‐term dynamic recommendation model based on local differential privacy (SDRM‐LDP). This model uses a small amount of user information to construct short‐term user preference behaviors and provides recommendations for users based on the similarity between items. We consider that an attacker uses nonprivate items to derive privacy items. Therefore, we randomly replace the original data in the same category. At the same time, the local differential privacy (LDP) is added to the privacy item query to make the private data available and protect the privacy information. In this paper, two real‐world datasets, ML‐100K and ML‐10M, are used for experiments. Experimental results show that the results of SDRM‐LDP are superior to other models.

When Network Operation Meets Blockchain: An Artificial-Intelligence-Driven Customization Service for Trusted Virtual Resources of IoT

The Internet of Things (IoT) is evolving from a stovepipe architecture to a collaborative and shared converged architecture. In the future, it will become a network development trend to construct a network-based virtual operating platform by virtualizing the IoT infrastructure to support different IoT services. The trust issues encountered during resource aggregation in heterogeneous networks and state measurements limit the applicability of the virtual network operating platform. In this article, we propose a framework for an endogenous trusted network to support virtual network operation, provide shared services externally, allocate resources internally, and implement artificial-intelligence-driven customization services for trusted virtual resources for the IoT. We use the consortium blockchain to establish the endogenous trusted framework for heterogeneous networks through software-defined networking and network function virtualization. In this framework, we design a mechanism for on-chain and off-chain collaborative resource allocation. The framework integrates time prediction algorithms and resource allocation algorithms to predict resource demand and allocate resources to achieve customization and dynamic adjustment. The application of this framework in specific scenarios is described using application examples.

Review of Physical Layer Security Technology in Battlefield Network

Global openness of battlefield network makes it face many security threats, the physical security technology can provide an important support to traditional security system and strong protection to messages transmission, studying its application in battlefield network is of great significance. In view of the current and future network development trend, this paper studies the physical layer security problem from the perspective of battlefield network under modern warfare conditions, summarizes and analyzes the basic model of physical layer security, sorts out and introduces the physical layer security technology of wireless communication, and finally summarizes the risks and challenges still faced by the physical layer of the current battlefield network.

Intelligent resource allocation management for vehicles network: An A3C learning approach

With the increasing demand of users for high-speed, low-delay and high-reliability services in connected vehicles network, wireless networks with communication, caching and computing convergence become the trend of network development in the future. To improve the quality of services of vehicles network, we propose a virtualized framework for mobile vehicle services, which using a learning-based resource allocation scheme. The dynamic change processes are modeled as Markov chains without making assumptions about the optimization goal and reducing the complexity of resource allocation computing. A high performance asynchronous advantage actor–critic learning algorithm is proposed to solve the complex dynamic resource allocation problem. Base on software-defined networking and information-centric networking, the method can dynamic orchestration of computing and communication resources to enhance the performance of virtual wireless networks. Simulation results verify that the proposed scheme can converge at a fast speed and improve the network operator’s total rewards.

AI Routers & Network Mind: A Hybrid Machine Learning Paradigm for Packet Routing

With the increasing complexity of network topologies and architectures, adding intelligence to the network control plane through Artificial Intelligence and Machine Learning (AI&ML) is becoming a trend in network development. For large-scale geo-distributed systems, determining how to appropriately introduce intelligence in networking is the key to high-efficiency operation. In this treatise, we explore two deployment paradigms (centralized vs. distributed) for AI-based networking. To achieve the best results, we propose a hybrid ML paradigm that combines a distributed intelligence, based on units called "AI routers," with a centralized intelligence, called the "network mind", to support different network services. In the proposed paradigm, we deploy centralized AI control for connection-oriented tunneling-based routing protocols (such as multiprotocol label switching and segment routing) to guarantee a high QoS, whereas for hop-by-hop IP routing, we shift the intelligent control responsibility to each AI router to ease the overhead imposed by centralized control and use the network mind to improve the global convergence.

An assessment method for aviation network optimization based on time-varying small world model

The optimization of aviation networks is of great significance for optimizing the allocation of resources, improving transport efficiency, and enhancing the competitiveness among airline companies. There have been a lot of researches which combine the theory of complex network and the actual situations to analyze the air transportation system. The present work provides a certain theoretical basis for the plan of airline schedule. Firstly, we regard an airport as a node, flight frequency as a link weight, and build a heterogeneous network. Through empirical analysis, we find that the aviation network has small-world and scale-free properties. In addition, considering that the instant network consists of current flights changing over time, time-varying is another important characteristic of aviation network. Also, a spatiotemporal correspondence between the flight frequency and route geometric distance is demonstrated to be τij~rij-C. Secondly, by Monte Carlo simulation, we know that the time-ordered topologies influence the optimal navigation structure and make it different from those from traditional static models. Specially, we can obtain a unique restriction between C and optimal structural exponent α, which unveils a new optimization principle in route design and schedule arrangement. Applying these features to the cost-minimized optimization model, a method to evaluate the optimization of network is proposed, by which we can directly predict the overall optimal distribution of flight distances and corresponding flight frequencies only based on the information about the passenger flow assignment. Thirdly, China aviation network data from 2001 to 2010 are used for empirical study. It is found that the predictions consist with the actual data. Compared with traditional optimization methods, it can simplify the computational complexity, and therefore it takes full advantage of the structural convenience and provides a new perspective for the overall scheduling of air transportation system. In this case, companies are able to estimate route adjustments easily to see whether they are reasonable and analyze the development trend of network to provide suggestions for future optimization.

Intercity Rail Transit and Integrated Development of the Pearl River Delta Urban Cluster: Based on the Perspective of Network Analysis

Intercity rail transit is an important power promoting the integrated development of urban clusters. This paper adopts a social network analysis to analyze the impact of the intercity rail development on the integration of the Pearl River Delta urban cluster in 2010, 2012, and 2020, respectively. The results show that the development of intercity rail transit is generally conducive to enhancing the network development trend of the Pearl River Delta urban cluster, strengthening the inter links and promoting the division of labor within the urban cluster, reinforcing the status of the center cities, thus promoting the integrated development of the Pearl River Delta urban cluster. However, the development of intercity rail transit still cannot break the former core–periphery structure, which may aggravate the unbalanced development among cities. Therefore, on the one hand, accelerating the construction of intercity rail transit is an important path to promote the integrated development of the Pearl River Delta urban cluster; on the other hand, it is necessary to optimize the intercity rail network planning, so that the adverse impact could be mitigated.

Study on DSM Energy-Saving Effect Evaluation under the Smart Grid Environment

China's power conflicts between supply and demand have become increasingly prominent. The smart grid has become an inevitable trend of network development at the same time, and DSM implementation effect in smart grid research increasingly attracts people’s attention. Through the analysis of different subjects in demand side management, the paper established demand side management effectiveness evaluation system in smart grid from the aspects of economic , environmental and social. Finally, empirical research verified the validity of the index system ,which provides a good reference for improving the level of demand side management.

Exploring the green features of converged networks

The converged network has been the trend of network development. It can not only provide advanced new features to the end users, but also provide a novel way of green computing to the network operators. This paper presented the idea of utilising converged networks for green network operations. The converged network-based solution is presented in detail and the benefits of the proposal are discussed. Mathematical analyses showed that the converged networks consume less network operation resources compared with non-converged networks and converged networks also perform better when the network operation scale is increased. Some open research issues and the future work are further discussed in the conclusion part of this paper.

Connection provisioning with guaranteed recovery time in survivable WDM optical networks

Optical networks employing wavelength- division multiplexing (WDM) are prone to failures. In case of a failure, the time to recovery is an important metric, since the amount of lost information and the performance of carried application strongly depends on the time of service interruption. Moreover, as the current trend in network development is moving toward a unified solution providing support for voice, data and various multimedia services, different application and users may tolerate different time of service interruption. Therefore, it is imperative to provide recovery-time guaranteed connections in WDM optical networks. In this paper, we propose a new shared-segment protection algorithm, called AGBSP-QoP, to provision connection requests according to their differentiated recovery-time requirements. Given a working path, AGBSP-QoP first computes all possible recovery-time guaranteed working-backup segment pairs, and then dynamically selects a least-cost eligible subset based on a dexterously constructed auxiliary graph. We conduct simulations under single class of service and multiple classes of service, and compare AGBSP-QoP with other protection algorithms. The results show that AGBSP-QoP yields the best performance and can properly provision connection requests according to their differentiated recovery-time requirements.

Differentiated QoS for survivable WDM optical networks

Optical networks based on WDM technology have become a promising solution to realize transport networks that can meet the ever-increasing demand for bandwidth. As WDM networks carry a huge volume of traffic, maintaining a high level of survivability is an important and critical issue. The. development of GMPLS switching technology led to the direct integration of IP and WDM. In these IP-over-WDM networks different applications/end users need different levels of fault tolerance and differ in how much they are willing to pay for the service they get. The current trend in network development is moving toward a unified solution providing support for voice, data, and various multimedia services. Therefore, it imperative that WDM networks incorporate fault tolerance to single or multiple component failures, protection bandwidth, recovery time, and recovery granularity besides resource utilization and call acceptance ratio. This article presents a survey of various methods that have been proposed for providing service differentiation in survivable WDM networks and discuss their performance. Such methods are broadly classified under various paradigms such as differentiated reliability, R-connections, quality of protection, and quality of recovery.