With the emphasis on user privacy and communication security, encrypted traffic has increased dramatically, which brings great challenges to traffic classification. The classification method of encrypted traffic based...With the emphasis on user privacy and communication security, encrypted traffic has increased dramatically, which brings great challenges to traffic classification. The classification method of encrypted traffic based on GNN can deal with encrypted traffic well. However, existing GNN-based approaches ignore the relationship between client or server packets. In this paper, we design a network traffic topology based on GCN, called Flow Mapping Graph (FMG). FMG establishes sequential edges between vertexes by the arrival order of packets and establishes jump-order edges between vertexes by connecting packets in different bursts with the same direction. It not only reflects the time characteristics of the packet but also strengthens the relationship between the client or server packets. According to FMG, a Traffic Mapping Classification model (TMC-GCN) is designed, which can automatically capture and learn the characteristics and structure information of the top vertex in FMG. The TMC-GCN model is used to classify the encrypted traffic. The encryption stream classification problem is transformed into a graph classification problem, which can effectively deal with data from different data sources and application scenarios. By comparing the performance of TMC-GCN with other classical models in four public datasets, including CICIOT2023, ISCXVPN2016, CICAAGM2017, and GraphDapp, the effectiveness of the FMG algorithm is verified. The experimental results show that the accuracy rate of the TMC-GCN model is 96.13%, the recall rate is 95.04%, and the F1 rate is 94.54%.展开更多
Maximize the resource utilization efficiency and guarantee the quality of service(QoS)of users by selecting the network are the key issues for heterogeneous network operators,but the resources occupied by users in dif...Maximize the resource utilization efficiency and guarantee the quality of service(QoS)of users by selecting the network are the key issues for heterogeneous network operators,but the resources occupied by users in different networks cannot be compared directly.This paper proposes a network selection algorithm for heterogeneous network.Firstly,the concept of equivalent bandwidth is proposed,through which the actual resources occupied by users with certain QoS requirements in different networks can be compared directly.Then the concept of network applicability is defined to express the abilities of networks to support different services.The proposed network selection algorithm first evaluates whether the network has enough equivalent bandwidth required by the user and then prioritizes network with poor applicability to avoid the situation that there are still residual resources in entire network,but advanced services can not be admitted.The simulation results show that the proposed algorithm obtained better performance than the baselines in terms of reducing call blocking probability and improving network resource utilization efficiency.展开更多
As the economy grows, environmental issues are becoming increasingly severe, making the promotion of green behavior more urgent. Information dissemination and policy regulation play crucial roles in influencing and am...As the economy grows, environmental issues are becoming increasingly severe, making the promotion of green behavior more urgent. Information dissemination and policy regulation play crucial roles in influencing and amplifying the spread of green behavior across society. To this end, a novel three-layer model in multilayer networks is proposed. In the novel model, the information layer describes green information spreading, the physical contact layer depicts green behavior propagation, and policy regulation is symbolized by an isolated node beneath the two layers. Then, we deduce the green behavior threshold for the three-layer model using the microscopic Markov chain approach. Moreover, subject to some individuals who are more likely to influence others or become green nodes and the limitations of the capacity of policy regulation, an optimal scheme is given that could optimize policy interventions to most effectively prompt green behavior.Subsequently, simulations are performed to validate the preciseness and theoretical results of the new model. It reveals that policy regulation can prompt the prevalence and outbreak of green behavior. Then, the green behavior is more likely to spread and be prevalent in the SF network than in the ER network. Additionally, optimal allocation is highly successful in facilitating the dissemination of green behavior. In practice, the optimal allocation strategy could prioritize interventions at critical nodes or regions, such as highly connected urban areas, where the impact of green behavior promotion would be most significant.展开更多
Quantum machine learning is an important application of quantum computing in the era of noisy intermediate-scale quantum devices.Domain adaptation(DA)is an effective method for addressing the distribution discrepancy ...Quantum machine learning is an important application of quantum computing in the era of noisy intermediate-scale quantum devices.Domain adaptation(DA)is an effective method for addressing the distribution discrepancy problem between the training data and the real data when the neural network model is deployed.In this paper,we propose a variational quantum domain adaptation method inspired by the quantum convolutional neural network,named variational quantum domain adaptation(VQDA).The data are first uploaded by a‘quantum coding module',then the feature information is extracted by several‘quantum convolution layers'and‘quantum pooling layers',which is named‘Feature Extractor'.Subsequently,the labels and the domains of the samples are obtained by the‘quantum fully connected layer'.With a gradient reversal module,the trained‘Feature Extractor'can extract the features that cannot be distinguished from the source and target domains.The simulations on the local computer and IBM Quantum Experience(IBM Q)platform by Qiskit show the effectiveness of the proposed method.The results show that VQDA(with 8 quantum bits)has 91.46%average classification accuracy for DA task between MNIST→USPS(USPS→MNIST),achieves 91.16%average classification accuracy for gray-scale and color images(with 10 quantum bits),and has 69.25%average classification accuracy on the DA task for color images(also with 10 quantum bits).VQDA achieves a 9.14%improvement in average classification accuracy compared to its corresponding classical domain adaptation method with the same parameter scale for different DA tasks.Simultaneously,the parameters scale is reduced to 43%by using VQDA when both quantum and classical DA methods have similar classification accuracies.展开更多
Next-generation 6G networks seek to provide ultra-reliable and low-latency communications,necessitating network designs that are intelligent and adaptable.Network slicing has developed as an effective option for resou...Next-generation 6G networks seek to provide ultra-reliable and low-latency communications,necessitating network designs that are intelligent and adaptable.Network slicing has developed as an effective option for resource separation and service-level differentiation inside virtualized infrastructures.Nonetheless,sustaining elevated Quality of Service(QoS)in dynamic,resource-limited systems poses significant hurdles.This study introduces an innovative packet-based proactive end-to-end(ETE)resource management system that facilitates network slicing with improved resilience and proactivity.To get around the drawbacks of conventional reactive systems,we develop a cost-efficient slice provisioning architecture that takes into account limits on radio,processing,and transmission resources.The optimization issue is non-convex,NP-hard,and requires online resolution in a dynamic setting.We offer a hybrid solution that integrates an advanced Deep Reinforcement Learning(DRL)methodology with an Improved Manta-Ray Foraging Optimization(ImpMRFO)algorithm.The ImpMRFO utilizes Chebyshev chaotic mapping for the formation of a varied starting population and incorporates Lévy flight-based stochastic movement to avert premature convergence,hence facilitating improved exploration-exploitation trade-offs.The DRL model perpetually acquires optimum provisioning strategies via agent-environment interactions,whereas the ImpMRFO enhances policy performance for effective slice provisioning.The solution,developed in Python,is evaluated across several 6G slicing scenarios that include varied QoS profiles and traffic requirements.The DRL model perpetually acquires optimum provisioning methods via agent-environment interactions,while the ImpMRFO enhances policy performance for effective slice provisioning.The solution,developed in Python,is evaluated across several 6G slicing scenarios that include varied QoS profiles and traffic requirements.Experimental findings reveal that the proactive ETE system outperforms DRL models and non-resilient provisioning techniques.Our technique increases PSSRr,decreases average latency,and optimizes resource use.These results demonstrate that the hybrid architecture for robust,real-time,and scalable slice management in future 6G networks is feasible.展开更多
Fine-grained Image Recognition(FGIR)task is dedicated to distinguishing similar sub-categories that belong to the same super-category,such as bird species and car types.In order to highlight visual differences,existin...Fine-grained Image Recognition(FGIR)task is dedicated to distinguishing similar sub-categories that belong to the same super-category,such as bird species and car types.In order to highlight visual differences,existing FGIR works often follow two steps:discriminative sub-region localization and local feature representation.However,these works pay less attention on global context information.They neglect a fact that the subtle visual difference in challenging scenarios can be highlighted through exploiting the spatial relationship among different subregions from a global view point.Therefore,in this paper,we consider both global and local information for FGIR,and propose a collaborative teacher-student strategy to reinforce and unity the two types of information.Our framework is implemented mainly by convolutional neural network,referred to Teacher-Student Based Attention Convolutional Neural Network(T-S-ACNN).For fine-grained local information,we choose the classic Multi-Attention Network(MA-Net)as our baseline,and propose a type of boundary constraint to further reduce background noises in the local attention maps.In this way,the discriminative sub-regions tend to appear in the area occupied by fine-grained objects,leading to more accurate sub-region localization.For fine-grained global information,we design a graph convolution based Global Attention Network(GA-Net),which can combine extracted local attention maps from MA-Net with non-local techniques to explore spatial relationship among subregions.At last,we develop a collaborative teacher-student strategy to adaptively determine the attended roles and optimization modes,so as to enhance the cooperative reinforcement of MA-Net and GA-Net.Extensive experiments on CUB-200-2011,Stanford Cars and FGVC Aircraft datasets illustrate the promising performance of our framework.展开更多
To ensure the access security of 6G,physical-layer authentication(PLA)leverages the randomness and space-time-frequency uniqueness of the channel to provide unique identity signatures for transmitters.Furthermore,the ...To ensure the access security of 6G,physical-layer authentication(PLA)leverages the randomness and space-time-frequency uniqueness of the channel to provide unique identity signatures for transmitters.Furthermore,the introduction of artificial intelligence(AI)facilitates the learning of the distribution characteristics of channel fingerprints,effectively addressing the uncertainties and unknown dynamic challenges in wireless link modeling.This paper reviews representative AI-enabled PLA schemes and proposes a graph neural network(GNN)-based PLA approach in response to the challenges existing methods face in identifying mobile users.Simulation results demonstrate that the proposed method outperforms six baseline schemes in terms of authentication accuracy.Furthermore,this paper outlines the future development directions of PLA.展开更多
The advent of 6G networks is poised to drive a new era of intelligent,privacy-preserving distributed learning by leveraging advanced communication and AI-driven edge intelligence.Federated Learning(FL)has emerged as a...The advent of 6G networks is poised to drive a new era of intelligent,privacy-preserving distributed learning by leveraging advanced communication and AI-driven edge intelligence.Federated Learning(FL)has emerged as a promising paradigm to enable collaborative model training without exposing raw data.However,its deployment in 6G networks faces significant obstacles,including vulnerabilities to inference attacks,the complexities of heterogeneous and dynamic network environments,and the inherent trade-off between privacy protection and model performance.In response to these challenges,we introduce DP-Fed6G,a novel FL framework that integrates differential privacy(DP)to fortify data security while ensuring high-quality learning outcomes.Specifically,DPFed6G employs an adaptive noise injection strategy that dynamically adjusts privacy protection levels based on real-time 6G network conditions and device heterogeneity,ensuring robust data security while maximizing model performance and optimizing the trade-off between privacy and utility.Extensive experiments on three real-world healthcare datasets demonstrate that DP-Fed6G consistently outperforms existing baselines(DP-Fed SGD and DPFed Avg),achieving up to 10.3%higher test accuracy under the same privacy budget.The proposed framework thus provides a practical solution for secure and privacy-preserving AI in 6G,supporting intelligent decisionmaking in privacy-sensitive applications.展开更多
With the rise of encrypted traffic,traditional network analysis methods have become less effective,leading to a shift towards deep learning-based approaches.Among these,multimodal learning-based classification methods...With the rise of encrypted traffic,traditional network analysis methods have become less effective,leading to a shift towards deep learning-based approaches.Among these,multimodal learning-based classification methods have gained attention due to their ability to leverage diverse feature sets from encrypted traffic,improving classification accuracy.However,existing research predominantly relies on late fusion techniques,which hinder the full utilization of deep features within the data.To address this limitation,we propose a novel multimodal encrypted traffic classification model that synchronizes modality fusion with multiscale feature extraction.Specifically,our approach performs real-time fusion of modalities at each stage of feature extraction,enhancing feature representation at each level and preserving inter-level correlations for more effective learning.This continuous fusion strategy improves the model’s ability to detect subtle variations in encrypted traffic,while boosting its robustness and adaptability to evolving network conditions.Experimental results on two real-world encrypted traffic datasets demonstrate that our method achieves a classification accuracy of 98.23% and 97.63%,outperforming existing multimodal learning-based methods.展开更多
The next-generation RAN,known as Open Radio Access Network(ORAN),allows for several advantages,including cost-effectiveness,network flexibility,and interoperability.Now ORAN applications,utilising machine learning(ML)...The next-generation RAN,known as Open Radio Access Network(ORAN),allows for several advantages,including cost-effectiveness,network flexibility,and interoperability.Now ORAN applications,utilising machine learning(ML)and artificial intelligence(AI)techniques,have become standard practice.The need for Federated Learning(FL)for ML model training in ORAN environments is heightened by the modularised structure of the ORAN architecture and the shortcomings of conventional ML techniques.However,the traditional plaintext model update sharing of FL in multi-BS contexts is susceptible to privacy violations such as deep-leakage gradient assaults and inference.Therefore,this research presents a novel blockchain-assisted improved cryptographic privacy-preserving federated learning(BICPPFL)model,with the help of ORAN,to safely carry out federated learning and protect privacy.This model improves on the conventional masking technique for sharing model parameters by adding new characteristics.These features include the choice of distributed aggregators,validation for final model aggregation,and individual validation for BSs.To manage the security and privacy of FL processes,a combined homomorphic proxy-reencryption(HPReE)and lattice-cryptographic method(HPReEL)has been used.The upgraded delegated proof of stake(Up-DPoS)consensus protocol,which will provide quick validation of model exchanges and protect against malicious attacks,is employed for effective consensus across blockchain nodes.Without sacrificing performance metrics,the BICPPFL model strengthens privacy and adds security layers while facilitating the transfer of sensitive data across several BSs.The framework is deployed on top of a Hyperledger Fabric blockchain to evaluate its effectiveness.The experimental findings prove the reliability and privacy-preserving capability of the BICPPFL model.展开更多
In 6G era,service forms in which computing power acts as the core will be ubiquitous in the network.At the same time,the collaboration among edge computing,cloud computing and network is needed to support edge computi...In 6G era,service forms in which computing power acts as the core will be ubiquitous in the network.At the same time,the collaboration among edge computing,cloud computing and network is needed to support edge computing service with strong demand for computing power,so as to realize the optimization of resource utilization.Based on this,the article discusses the research background,key techniques and main application scenarios of computing power network.Through the demonstration,it can be concluded that the technical solution of computing power network can effectively meet the multi-level deployment and flexible scheduling needs of the future 6G business for computing,storage and network,and adapt to the integration needs of computing power and network in various scenarios,such as user oriented,government enterprise oriented,computing power open and so on.展开更多
In MEC-enabled vehicular network with limited wireless resource and computation resource,stringent delay and high reliability requirements are challenging issues.In order to reduce the total delay in the network as we...In MEC-enabled vehicular network with limited wireless resource and computation resource,stringent delay and high reliability requirements are challenging issues.In order to reduce the total delay in the network as well as ensure the reliability of Vehicular UE(VUE),a Joint Allocation of Wireless resource and MEC Computing resource(JAWC)algorithm is proposed.The JAWC algorithm includes two steps:V2X links clustering and MEC computation resource scheduling.In the V2X links clustering,a Spectral Radius based Interference Cancellation scheme(SR-IC)is proposed to obtain the optimal resource allocation matrix.By converting the calculation of SINR into the calculation of matrix maximum row sum,the accumulated interference of VUE can be constrained and the the SINR calculation complexity can be effectively reduced.In the MEC computation resource scheduling,by transforming the original optimization problem into a convex problem,the optimal task offloading proportion of VUE and MEC computation resource allocation can be obtained.The simulation further demonstrates that the JAWC algorithm can significantly reduce the total delay as well as ensure the communication reliability of VUE in the MEC-enabled vehicular network.展开更多
Cell discontinuous transmission(Cell DTx)is a key technology to mitigate inter-cell interference(ICI)in ultra-dense networks(UDNs).The aim of this work is to understand the impact of Cell DTx on physical-layer sum rat...Cell discontinuous transmission(Cell DTx)is a key technology to mitigate inter-cell interference(ICI)in ultra-dense networks(UDNs).The aim of this work is to understand the impact of Cell DTx on physical-layer sum rates of SBSs and link-layer quality-of-service(QoS)performance in multiuser UDNs.In this work,we develop a cross-layer framework for capacity analysis in multiuser UDNs with Cell DTx.In particular,we first extend the traditional one-dimensional effective capacity model to a new multidimensional effective capacity model to derive the sum rate and the effective capacity.Moreover,we propose a new iterative bisection search algorithm that is capable of approximating QoS performance.The convergence of this new algorithm to a unique QoS exponent vector is later proved.Finally,we apply this framework to the round-robin and the max-C/I scheduling policies.Simulation results show that our framework is accurate in approximating 1)queue length distribution,2)delay distribution and 3)sum rates under the above two scheduling policies,and further show that with the Cell DTx,systems have approximately 30% higher sum rate and 35% smaller average delay than those in full-buffer scenarios.展开更多
A K-tier uplink heterogeneous cellular network is modelled and analysed by accounting for both truncated channel inversion power control and biased user association. Each user has a maximum transmit power constraint a...A K-tier uplink heterogeneous cellular network is modelled and analysed by accounting for both truncated channel inversion power control and biased user association. Each user has a maximum transmit power constraint and transmits data when it has sufficient transmit power to perform channel inversion. With biased user association, each user is associated with a base station(BS) that provides the maximum received power weighted by a bias factor, but not their nearest BS. Stochastic geometry is used to evaluate the performances of the proposed system model in terms of the outage probability and ergodic rate for each tier as functions of the biased and power control parameters. Simulations validate our analytical derivations. Numerical results show that there exists a trade-off introduced by the power cut-off threshold and the maximum user transmit power constraint. When the maximum user transmit power becomes a binding constraint, the overall performance is independent of BS densities. In addition, we have shown that it is beneficial for the outage and rate performances by optimizing different network parameters such as the power cut-off threshold as well as the biased factors.展开更多
Compared with accurate diagnosis, the system’s selfdiagnosing capability can be greatly increased through the t/kdiagnosis strategy at most k vertexes to be mistakenly identified as faulty under the comparison model,...Compared with accurate diagnosis, the system’s selfdiagnosing capability can be greatly increased through the t/kdiagnosis strategy at most k vertexes to be mistakenly identified as faulty under the comparison model, where k is typically a small number. Based on the Preparata, Metze, and Chien(PMC)model, the n-dimensional hypercube network is proved to be t/kdiagnosable. In this paper, based on the Maeng and Malek(MM)*model, a novel t/k-fault diagnosis(1≤k≤4) algorithm of ndimensional hypercube, called t/k-MM*-DIAG, is proposed to isolate all faulty processors within the set of nodes, among which the number of fault-free nodes identified wrongly as faulty is at most k. The time complexity in our algorithm is only O(2~n n~2).展开更多
Active Queue Management (AQM) is an active research area in the Internet community. Random Early Detection (RED) is a typical AQM algorithm, but it is known that it is difficult to configure its parameters and its ave...Active Queue Management (AQM) is an active research area in the Internet community. Random Early Detection (RED) is a typical AQM algorithm, but it is known that it is difficult to configure its parameters and its average queue length is closely related to the load level. This paper proposes an effective fuzzy congestion control algorithm based on fuzzy logic which uses the pre- dominance of fuzzy logic to deal with uncertain events. The main advantage of this new congestion control algorithm is that it discards the packet dropping mechanism of RED, and calculates packet loss according to a preconfigured fuzzy logic by using the queue length and the buffer usage ratio. Theo- retical analysis and Network Simulator (NS) simulation results show that the proposed algorithm achieves more throughput and more stable queue length than traditional schemes. It really improves a router's ability in network congestion control in IP network.展开更多
Network intrusion poses a severe threat to the Internet.However,existing intrusion detection models cannot effectively distinguish different intrusions with high-degree feature overlap.In addition,efficient real-time ...Network intrusion poses a severe threat to the Internet.However,existing intrusion detection models cannot effectively distinguish different intrusions with high-degree feature overlap.In addition,efficient real-time detection is an urgent problem.To address the two above problems,we propose a Latent Dirichlet Allocation topic model-based framework for real-time network Intrusion Detection(LDA-ID),consisting of static and online LDA-ID.The problem of feature overlap is transformed into static LDA-ID topic number optimization and topic selection.Thus,the detection is based on the latent topic features.To achieve efficient real-time detection,we design an online computing mode for static LDA-ID,in which a parameter iteration method based on momentum is proposed to balance the contribution of prior knowledge and new information.Furthermore,we design two matching mechanisms to accommodate the static and online LDA-ID,respectively.Experimental results on the public NSL-KDD and UNSW-NB15 datasets show that our framework gets higher accuracy than the others.展开更多
Cooperative spectrum sensing in cog- nitive radio is investigated to improve the det- ection performance of Primary User （PU）. Meanwhile, cluster-based hierarchical coop- eration is introduced for reducing the overh...Cooperative spectrum sensing in cog- nitive radio is investigated to improve the det- ection performance of Primary User （PU）. Meanwhile, cluster-based hierarchical coop- eration is introduced for reducing the overhead as well as maintaining a certain level of sens- ing performance. However, in existing hierar- chically cooperative spectrum sensing algo- rithms, the robustness problem of the system is seldom considered. In this paper, we pro- pose a reputation-based hierarchically coop- erative spectrum sensing scheme in Cognitive Radio Networks （CRNs）. Before spectrum sensing, clusters are grouped based on the location correlation coefficients of Secondary Users （SUs）. In the proposed scheme, there are two levels of cooperation, the first one is performed within a cluster and the second one is carried out among clusters. With the reputa- tion mechanism and modified MAJORITY rule in the second level cooperation, the pro- posed scheme can not only relieve the influ- ence of the shadowing, but also eliminate the impact of the PU emulation attack on a rela- tively large scale. Simulation results show that, in the scenarios with deep-shadowing or mul- tiple attacked SUs, our proposed scheme ach- ieves a better tradeoff between the system robustness and the energy saving compared with those conventionally cooperative sensing schemes.展开更多
The network reliability is difficult to be evaluated because of the complex relationship among the network components.It can be quite different for different users running different applications on the same network.Th...The network reliability is difficult to be evaluated because of the complex relationship among the network components.It can be quite different for different users running different applications on the same network.This paper proposes a new concept and a model of application reliability.Different from the existing models that ignores the effects of applications,the proposed application reliability model considers the effects of different applications on the network performance and different types of network faults and makes the analysis of network components relationship possible.This paper also provides a method to evaluate the application reliability when the data flow satisfies Markov properties.Finally,a case study is presented to illustrate the proposed network reliability model and the analysis method.展开更多
Based on the sequence entropy of Shannon information theory, we work on the network coding technology in Wireless Sensor Network (WSN). In this paper, we take into account the similarity of the transmission sequences ...Based on the sequence entropy of Shannon information theory, we work on the network coding technology in Wireless Sensor Network (WSN). In this paper, we take into account the similarity of the transmission sequences at the network coding node in the multi-sources and multi-receivers network in order to compress the data redundancy. Theoretical analysis and computer simulation results show that this proposed scheme not only further improves the efficiency of network transmission and enhances the throughput of the network, but also reduces the energy consumption of sensor nodes and extends the network life cycle.展开更多
基金supported by the National Key Research and Development Program of China No.2023YFA1009500.
文摘With the emphasis on user privacy and communication security, encrypted traffic has increased dramatically, which brings great challenges to traffic classification. The classification method of encrypted traffic based on GNN can deal with encrypted traffic well. However, existing GNN-based approaches ignore the relationship between client or server packets. In this paper, we design a network traffic topology based on GCN, called Flow Mapping Graph (FMG). FMG establishes sequential edges between vertexes by the arrival order of packets and establishes jump-order edges between vertexes by connecting packets in different bursts with the same direction. It not only reflects the time characteristics of the packet but also strengthens the relationship between the client or server packets. According to FMG, a Traffic Mapping Classification model (TMC-GCN) is designed, which can automatically capture and learn the characteristics and structure information of the top vertex in FMG. The TMC-GCN model is used to classify the encrypted traffic. The encryption stream classification problem is transformed into a graph classification problem, which can effectively deal with data from different data sources and application scenarios. By comparing the performance of TMC-GCN with other classical models in four public datasets, including CICIOT2023, ISCXVPN2016, CICAAGM2017, and GraphDapp, the effectiveness of the FMG algorithm is verified. The experimental results show that the accuracy rate of the TMC-GCN model is 96.13%, the recall rate is 95.04%, and the F1 rate is 94.54%.
文摘Maximize the resource utilization efficiency and guarantee the quality of service(QoS)of users by selecting the network are the key issues for heterogeneous network operators,but the resources occupied by users in different networks cannot be compared directly.This paper proposes a network selection algorithm for heterogeneous network.Firstly,the concept of equivalent bandwidth is proposed,through which the actual resources occupied by users with certain QoS requirements in different networks can be compared directly.Then the concept of network applicability is defined to express the abilities of networks to support different services.The proposed network selection algorithm first evaluates whether the network has enough equivalent bandwidth required by the user and then prioritizes network with poor applicability to avoid the situation that there are still residual resources in entire network,but advanced services can not be admitted.The simulation results show that the proposed algorithm obtained better performance than the baselines in terms of reducing call blocking probability and improving network resource utilization efficiency.
基金Project supported by the National Natural Science Foundation of China (Grant No. 62371253)the Postgraduate Research and Practice Innovation Program of Jiangsu Province, China (Grant No. KYCX24_1179)。
文摘As the economy grows, environmental issues are becoming increasingly severe, making the promotion of green behavior more urgent. Information dissemination and policy regulation play crucial roles in influencing and amplifying the spread of green behavior across society. To this end, a novel three-layer model in multilayer networks is proposed. In the novel model, the information layer describes green information spreading, the physical contact layer depicts green behavior propagation, and policy regulation is symbolized by an isolated node beneath the two layers. Then, we deduce the green behavior threshold for the three-layer model using the microscopic Markov chain approach. Moreover, subject to some individuals who are more likely to influence others or become green nodes and the limitations of the capacity of policy regulation, an optimal scheme is given that could optimize policy interventions to most effectively prompt green behavior.Subsequently, simulations are performed to validate the preciseness and theoretical results of the new model. It reveals that policy regulation can prompt the prevalence and outbreak of green behavior. Then, the green behavior is more likely to spread and be prevalent in the SF network than in the ER network. Additionally, optimal allocation is highly successful in facilitating the dissemination of green behavior. In practice, the optimal allocation strategy could prioritize interventions at critical nodes or regions, such as highly connected urban areas, where the impact of green behavior promotion would be most significant.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.62375140 and 61871234)。
文摘Quantum machine learning is an important application of quantum computing in the era of noisy intermediate-scale quantum devices.Domain adaptation(DA)is an effective method for addressing the distribution discrepancy problem between the training data and the real data when the neural network model is deployed.In this paper,we propose a variational quantum domain adaptation method inspired by the quantum convolutional neural network,named variational quantum domain adaptation(VQDA).The data are first uploaded by a‘quantum coding module',then the feature information is extracted by several‘quantum convolution layers'and‘quantum pooling layers',which is named‘Feature Extractor'.Subsequently,the labels and the domains of the samples are obtained by the‘quantum fully connected layer'.With a gradient reversal module,the trained‘Feature Extractor'can extract the features that cannot be distinguished from the source and target domains.The simulations on the local computer and IBM Quantum Experience(IBM Q)platform by Qiskit show the effectiveness of the proposed method.The results show that VQDA(with 8 quantum bits)has 91.46%average classification accuracy for DA task between MNIST→USPS(USPS→MNIST),achieves 91.16%average classification accuracy for gray-scale and color images(with 10 quantum bits),and has 69.25%average classification accuracy on the DA task for color images(also with 10 quantum bits).VQDA achieves a 9.14%improvement in average classification accuracy compared to its corresponding classical domain adaptation method with the same parameter scale for different DA tasks.Simultaneously,the parameters scale is reduced to 43%by using VQDA when both quantum and classical DA methods have similar classification accuracies.
文摘Next-generation 6G networks seek to provide ultra-reliable and low-latency communications,necessitating network designs that are intelligent and adaptable.Network slicing has developed as an effective option for resource separation and service-level differentiation inside virtualized infrastructures.Nonetheless,sustaining elevated Quality of Service(QoS)in dynamic,resource-limited systems poses significant hurdles.This study introduces an innovative packet-based proactive end-to-end(ETE)resource management system that facilitates network slicing with improved resilience and proactivity.To get around the drawbacks of conventional reactive systems,we develop a cost-efficient slice provisioning architecture that takes into account limits on radio,processing,and transmission resources.The optimization issue is non-convex,NP-hard,and requires online resolution in a dynamic setting.We offer a hybrid solution that integrates an advanced Deep Reinforcement Learning(DRL)methodology with an Improved Manta-Ray Foraging Optimization(ImpMRFO)algorithm.The ImpMRFO utilizes Chebyshev chaotic mapping for the formation of a varied starting population and incorporates Lévy flight-based stochastic movement to avert premature convergence,hence facilitating improved exploration-exploitation trade-offs.The DRL model perpetually acquires optimum provisioning strategies via agent-environment interactions,whereas the ImpMRFO enhances policy performance for effective slice provisioning.The solution,developed in Python,is evaluated across several 6G slicing scenarios that include varied QoS profiles and traffic requirements.The DRL model perpetually acquires optimum provisioning methods via agent-environment interactions,while the ImpMRFO enhances policy performance for effective slice provisioning.The solution,developed in Python,is evaluated across several 6G slicing scenarios that include varied QoS profiles and traffic requirements.Experimental findings reveal that the proactive ETE system outperforms DRL models and non-resilient provisioning techniques.Our technique increases PSSRr,decreases average latency,and optimizes resource use.These results demonstrate that the hybrid architecture for robust,real-time,and scalable slice management in future 6G networks is feasible.
基金supported by the National Natural Science Foundation of China,China (Grants No.62171232)the Priority Academic Program Development of Jiangsu Higher Education Institutions,China。
文摘Fine-grained Image Recognition(FGIR)task is dedicated to distinguishing similar sub-categories that belong to the same super-category,such as bird species and car types.In order to highlight visual differences,existing FGIR works often follow two steps:discriminative sub-region localization and local feature representation.However,these works pay less attention on global context information.They neglect a fact that the subtle visual difference in challenging scenarios can be highlighted through exploiting the spatial relationship among different subregions from a global view point.Therefore,in this paper,we consider both global and local information for FGIR,and propose a collaborative teacher-student strategy to reinforce and unity the two types of information.Our framework is implemented mainly by convolutional neural network,referred to Teacher-Student Based Attention Convolutional Neural Network(T-S-ACNN).For fine-grained local information,we choose the classic Multi-Attention Network(MA-Net)as our baseline,and propose a type of boundary constraint to further reduce background noises in the local attention maps.In this way,the discriminative sub-regions tend to appear in the area occupied by fine-grained objects,leading to more accurate sub-region localization.For fine-grained global information,we design a graph convolution based Global Attention Network(GA-Net),which can combine extracted local attention maps from MA-Net with non-local techniques to explore spatial relationship among subregions.At last,we develop a collaborative teacher-student strategy to adaptively determine the attended roles and optimization modes,so as to enhance the cooperative reinforcement of MA-Net and GA-Net.Extensive experiments on CUB-200-2011,Stanford Cars and FGVC Aircraft datasets illustrate the promising performance of our framework.
文摘To ensure the access security of 6G,physical-layer authentication(PLA)leverages the randomness and space-time-frequency uniqueness of the channel to provide unique identity signatures for transmitters.Furthermore,the introduction of artificial intelligence(AI)facilitates the learning of the distribution characteristics of channel fingerprints,effectively addressing the uncertainties and unknown dynamic challenges in wireless link modeling.This paper reviews representative AI-enabled PLA schemes and proposes a graph neural network(GNN)-based PLA approach in response to the challenges existing methods face in identifying mobile users.Simulation results demonstrate that the proposed method outperforms six baseline schemes in terms of authentication accuracy.Furthermore,this paper outlines the future development directions of PLA.
基金supported in part by the Research and Development Project of China Railway Information Technology Group under Grant WJZG-CKY-2024040(2024P01)the National Natural Science Foun-dation of China under Grant 62272100the Consulting Project of Chinese Academy of Engineering under Grant 2023-XY-09。
文摘The advent of 6G networks is poised to drive a new era of intelligent,privacy-preserving distributed learning by leveraging advanced communication and AI-driven edge intelligence.Federated Learning(FL)has emerged as a promising paradigm to enable collaborative model training without exposing raw data.However,its deployment in 6G networks faces significant obstacles,including vulnerabilities to inference attacks,the complexities of heterogeneous and dynamic network environments,and the inherent trade-off between privacy protection and model performance.In response to these challenges,we introduce DP-Fed6G,a novel FL framework that integrates differential privacy(DP)to fortify data security while ensuring high-quality learning outcomes.Specifically,DPFed6G employs an adaptive noise injection strategy that dynamically adjusts privacy protection levels based on real-time 6G network conditions and device heterogeneity,ensuring robust data security while maximizing model performance and optimizing the trade-off between privacy and utility.Extensive experiments on three real-world healthcare datasets demonstrate that DP-Fed6G consistently outperforms existing baselines(DP-Fed SGD and DPFed Avg),achieving up to 10.3%higher test accuracy under the same privacy budget.The proposed framework thus provides a practical solution for secure and privacy-preserving AI in 6G,supporting intelligent decisionmaking in privacy-sensitive applications.
基金supported by the National Key Research and Development Program of China No.2023YFB2705000.
文摘With the rise of encrypted traffic,traditional network analysis methods have become less effective,leading to a shift towards deep learning-based approaches.Among these,multimodal learning-based classification methods have gained attention due to their ability to leverage diverse feature sets from encrypted traffic,improving classification accuracy.However,existing research predominantly relies on late fusion techniques,which hinder the full utilization of deep features within the data.To address this limitation,we propose a novel multimodal encrypted traffic classification model that synchronizes modality fusion with multiscale feature extraction.Specifically,our approach performs real-time fusion of modalities at each stage of feature extraction,enhancing feature representation at each level and preserving inter-level correlations for more effective learning.This continuous fusion strategy improves the model’s ability to detect subtle variations in encrypted traffic,while boosting its robustness and adaptability to evolving network conditions.Experimental results on two real-world encrypted traffic datasets demonstrate that our method achieves a classification accuracy of 98.23% and 97.63%,outperforming existing multimodal learning-based methods.
文摘The next-generation RAN,known as Open Radio Access Network(ORAN),allows for several advantages,including cost-effectiveness,network flexibility,and interoperability.Now ORAN applications,utilising machine learning(ML)and artificial intelligence(AI)techniques,have become standard practice.The need for Federated Learning(FL)for ML model training in ORAN environments is heightened by the modularised structure of the ORAN architecture and the shortcomings of conventional ML techniques.However,the traditional plaintext model update sharing of FL in multi-BS contexts is susceptible to privacy violations such as deep-leakage gradient assaults and inference.Therefore,this research presents a novel blockchain-assisted improved cryptographic privacy-preserving federated learning(BICPPFL)model,with the help of ORAN,to safely carry out federated learning and protect privacy.This model improves on the conventional masking technique for sharing model parameters by adding new characteristics.These features include the choice of distributed aggregators,validation for final model aggregation,and individual validation for BSs.To manage the security and privacy of FL processes,a combined homomorphic proxy-reencryption(HPReE)and lattice-cryptographic method(HPReEL)has been used.The upgraded delegated proof of stake(Up-DPoS)consensus protocol,which will provide quick validation of model exchanges and protect against malicious attacks,is employed for effective consensus across blockchain nodes.Without sacrificing performance metrics,the BICPPFL model strengthens privacy and adds security layers while facilitating the transfer of sensitive data across several BSs.The framework is deployed on top of a Hyperledger Fabric blockchain to evaluate its effectiveness.The experimental findings prove the reliability and privacy-preserving capability of the BICPPFL model.
基金This work was supported by the National Key R&D Program of China No.2019YFB1802800.
文摘In 6G era,service forms in which computing power acts as the core will be ubiquitous in the network.At the same time,the collaboration among edge computing,cloud computing and network is needed to support edge computing service with strong demand for computing power,so as to realize the optimization of resource utilization.Based on this,the article discusses the research background,key techniques and main application scenarios of computing power network.Through the demonstration,it can be concluded that the technical solution of computing power network can effectively meet the multi-level deployment and flexible scheduling needs of the future 6G business for computing,storage and network,and adapt to the integration needs of computing power and network in various scenarios,such as user oriented,government enterprise oriented,computing power open and so on.
基金This work was supported in part by the National Key R&D Program of China under Grant 2019YFE0114000in part by the National Natural Science Foundation of China under Grant 61701042+1 种基金in part by the 111 Project of China(Grant No.B16006)the research foundation of Ministry of EducationChina Mobile under Grant MCM20180101.
文摘In MEC-enabled vehicular network with limited wireless resource and computation resource,stringent delay and high reliability requirements are challenging issues.In order to reduce the total delay in the network as well as ensure the reliability of Vehicular UE(VUE),a Joint Allocation of Wireless resource and MEC Computing resource(JAWC)algorithm is proposed.The JAWC algorithm includes two steps:V2X links clustering and MEC computation resource scheduling.In the V2X links clustering,a Spectral Radius based Interference Cancellation scheme(SR-IC)is proposed to obtain the optimal resource allocation matrix.By converting the calculation of SINR into the calculation of matrix maximum row sum,the accumulated interference of VUE can be constrained and the the SINR calculation complexity can be effectively reduced.In the MEC computation resource scheduling,by transforming the original optimization problem into a convex problem,the optimal task offloading proportion of VUE and MEC computation resource allocation can be obtained.The simulation further demonstrates that the JAWC algorithm can significantly reduce the total delay as well as ensure the communication reliability of VUE in the MEC-enabled vehicular network.
文摘Cell discontinuous transmission(Cell DTx)is a key technology to mitigate inter-cell interference(ICI)in ultra-dense networks(UDNs).The aim of this work is to understand the impact of Cell DTx on physical-layer sum rates of SBSs and link-layer quality-of-service(QoS)performance in multiuser UDNs.In this work,we develop a cross-layer framework for capacity analysis in multiuser UDNs with Cell DTx.In particular,we first extend the traditional one-dimensional effective capacity model to a new multidimensional effective capacity model to derive the sum rate and the effective capacity.Moreover,we propose a new iterative bisection search algorithm that is capable of approximating QoS performance.The convergence of this new algorithm to a unique QoS exponent vector is later proved.Finally,we apply this framework to the round-robin and the max-C/I scheduling policies.Simulation results show that our framework is accurate in approximating 1)queue length distribution,2)delay distribution and 3)sum rates under the above two scheduling policies,and further show that with the Cell DTx,systems have approximately 30% higher sum rate and 35% smaller average delay than those in full-buffer scenarios.
基金supported by the National Natural Science Foundation of China (61401225, 61571234)the National Science Foundation of Jiangsu Province (BK20140894, BK20140883, BK20160899)+4 种基金the Six Talented Eminence Foundation of Jiangsu Province (XYDXXJS-044)the National Science Foundation of the Higher Education Institutions of Jiangsu Province (14KJD510007, 16KJB510035)the Jiangsu Planned Projects for Postdoctoral Research Funds (1501125B)China Postdoctoral Science Foundation funded project (2015M581844)the Introduction of Talent Scientific Research Fund of Nanjing University of Posts Telecommunications project (NY213104, NY214190)
文摘A K-tier uplink heterogeneous cellular network is modelled and analysed by accounting for both truncated channel inversion power control and biased user association. Each user has a maximum transmit power constraint and transmits data when it has sufficient transmit power to perform channel inversion. With biased user association, each user is associated with a base station(BS) that provides the maximum received power weighted by a bias factor, but not their nearest BS. Stochastic geometry is used to evaluate the performances of the proposed system model in terms of the outage probability and ergodic rate for each tier as functions of the biased and power control parameters. Simulations validate our analytical derivations. Numerical results show that there exists a trade-off introduced by the power cut-off threshold and the maximum user transmit power constraint. When the maximum user transmit power becomes a binding constraint, the overall performance is independent of BS densities. In addition, we have shown that it is beneficial for the outage and rate performances by optimizing different network parameters such as the power cut-off threshold as well as the biased factors.
基金supported by the National Natural Science Foundation of China(61363002)
文摘Compared with accurate diagnosis, the system’s selfdiagnosing capability can be greatly increased through the t/kdiagnosis strategy at most k vertexes to be mistakenly identified as faulty under the comparison model, where k is typically a small number. Based on the Preparata, Metze, and Chien(PMC)model, the n-dimensional hypercube network is proved to be t/kdiagnosable. In this paper, based on the Maeng and Malek(MM)*model, a novel t/k-fault diagnosis(1≤k≤4) algorithm of ndimensional hypercube, called t/k-MM*-DIAG, is proposed to isolate all faulty processors within the set of nodes, among which the number of fault-free nodes identified wrongly as faulty is at most k. The time complexity in our algorithm is only O(2~n n~2).
基金Supported by the National High Technology Research and Development of China (863 Program) (No.2003AA121560)the High Technology Research and Development Program of Jiangsu Province (No.BEG2003001).
文摘Active Queue Management (AQM) is an active research area in the Internet community. Random Early Detection (RED) is a typical AQM algorithm, but it is known that it is difficult to configure its parameters and its average queue length is closely related to the load level. This paper proposes an effective fuzzy congestion control algorithm based on fuzzy logic which uses the pre- dominance of fuzzy logic to deal with uncertain events. The main advantage of this new congestion control algorithm is that it discards the packet dropping mechanism of RED, and calculates packet loss according to a preconfigured fuzzy logic by using the queue length and the buffer usage ratio. Theo- retical analysis and Network Simulator (NS) simulation results show that the proposed algorithm achieves more throughput and more stable queue length than traditional schemes. It really improves a router's ability in network congestion control in IP network.
基金supported by the National Natural Science Foundation of China(Grant No.U1636208,No.61862008,No.61902013)the Beihang Youth Top Talent Support Program(Grant No.YWF-21-BJJ-1039)。
文摘Network intrusion poses a severe threat to the Internet.However,existing intrusion detection models cannot effectively distinguish different intrusions with high-degree feature overlap.In addition,efficient real-time detection is an urgent problem.To address the two above problems,we propose a Latent Dirichlet Allocation topic model-based framework for real-time network Intrusion Detection(LDA-ID),consisting of static and online LDA-ID.The problem of feature overlap is transformed into static LDA-ID topic number optimization and topic selection.Thus,the detection is based on the latent topic features.To achieve efficient real-time detection,we design an online computing mode for static LDA-ID,in which a parameter iteration method based on momentum is proposed to balance the contribution of prior knowledge and new information.Furthermore,we design two matching mechanisms to accommodate the static and online LDA-ID,respectively.Experimental results on the public NSL-KDD and UNSW-NB15 datasets show that our framework gets higher accuracy than the others.
基金ACKNOWLEDGEMENT This work was partially supported by the Na- tional Natural Science Foundation of China under Grant No. 61071127 and the Science and Technology Department of Zhejiang Pro- vince under Grants No. 2012C01036-1, No. 2011R10035.
文摘Cooperative spectrum sensing in cog- nitive radio is investigated to improve the det- ection performance of Primary User （PU）. Meanwhile, cluster-based hierarchical coop- eration is introduced for reducing the overhead as well as maintaining a certain level of sens- ing performance. However, in existing hierar- chically cooperative spectrum sensing algo- rithms, the robustness problem of the system is seldom considered. In this paper, we pro- pose a reputation-based hierarchically coop- erative spectrum sensing scheme in Cognitive Radio Networks （CRNs）. Before spectrum sensing, clusters are grouped based on the location correlation coefficients of Secondary Users （SUs）. In the proposed scheme, there are two levels of cooperation, the first one is performed within a cluster and the second one is carried out among clusters. With the reputa- tion mechanism and modified MAJORITY rule in the second level cooperation, the pro- posed scheme can not only relieve the influ- ence of the shadowing, but also eliminate the impact of the PU emulation attack on a rela- tively large scale. Simulation results show that, in the scenarios with deep-shadowing or mul- tiple attacked SUs, our proposed scheme ach- ieves a better tradeoff between the system robustness and the energy saving compared with those conventionally cooperative sensing schemes.
基金supported by the Beijing Natural Science Foundation(4113074)
文摘The network reliability is difficult to be evaluated because of the complex relationship among the network components.It can be quite different for different users running different applications on the same network.This paper proposes a new concept and a model of application reliability.Different from the existing models that ignores the effects of applications,the proposed application reliability model considers the effects of different applications on the network performance and different types of network faults and makes the analysis of network components relationship possible.This paper also provides a method to evaluate the application reliability when the data flow satisfies Markov properties.Finally,a case study is presented to illustrate the proposed network reliability model and the analysis method.
基金Supported by Major Projects of the National Science and Technology (2010ZX03003-003-02) National 973 Key Project (2011CB302903)
文摘Based on the sequence entropy of Shannon information theory, we work on the network coding technology in Wireless Sensor Network (WSN). In this paper, we take into account the similarity of the transmission sequences at the network coding node in the multi-sources and multi-receivers network in order to compress the data redundancy. Theoretical analysis and computer simulation results show that this proposed scheme not only further improves the efficiency of network transmission and enhances the throughput of the network, but also reduces the energy consumption of sensor nodes and extends the network life cycle.
