The rapid expansion of the Internet of Things(IoT)and Edge Artificial Intelligence(AI)has redefined automation and connectivity acrossmodern networks.However,the heterogeneity and limited resources of IoT devices expo...The rapid expansion of the Internet of Things(IoT)and Edge Artificial Intelligence(AI)has redefined automation and connectivity acrossmodern networks.However,the heterogeneity and limited resources of IoT devices expose them to increasingly sophisticated and persistentmalware attacks.These adaptive and stealthy threats can evade conventional detection,establish remote control,propagate across devices,exfiltrate sensitive data,and compromise network integrity.This study presents a Software-Defined Internet of Things(SD-IoT)control-plane-based,AI-driven framework that integrates Gated Recurrent Units(GRU)and Long Short-TermMemory(LSTM)networks for efficient detection of evolving multi-vector,malware-driven botnet attacks.The proposed CUDA-enabled hybrid deep learning(DL)framework performs centralized real-time detection without adding computational overhead to IoT nodes.A feature selection strategy combining variable clustering,attribute evaluation,one-R attribute evaluation,correlation analysis,and principal component analysis(PCA)enhances detection accuracy and reduces complexity.The framework is rigorously evaluated using the N_BaIoT dataset under k-fold cross-validation.Experimental results achieve 99.96%detection accuracy,a false positive rate(FPR)of 0.0035%,and a detection latency of 0.18 ms,confirming its high efficiency and scalability.The findings demonstrate the framework’s potential as a robust and intelligent security solution for next-generation IoT ecosystems.展开更多
The rise of time-sensitive applications with broad geographical scope drives the development of time-sensitive networking(TSN)from intra-domain to inter-domain to ensure overall end-to-end connectivity requirements in...The rise of time-sensitive applications with broad geographical scope drives the development of time-sensitive networking(TSN)from intra-domain to inter-domain to ensure overall end-to-end connectivity requirements in heterogeneous deployments.When multiple TSN networks interconnect over non-TSN networks,all devices in the network need to be syn-chronized by sharing a uniform time reference.How-ever,most non-TSN networks are best-effort.Path delay asymmetry and random noise accumulation can introduce unpredictable time errors during end-to-end time synchronization.These factors can degrade syn-chronization performance.Therefore,cross-domain time synchronization becomes a challenging issue for multiple TSN networks interconnected by non-TSN networks.This paper presents a cross-domain time synchronization scheme that follows the software-defined TSN(SD-TSN)paradigm.It utilizes a com-bined control plane constructed by a coordinate con-troller and a domain controller for centralized control and management of cross-domain time synchroniza-tion.The general operation flow of the cross-domain time synchronization process is designed.The mecha-nism of cross-domain time synchronization is revealed by introducing a synchronization model and an error compensation method.A TSN cross-domain proto-type testbed is constructed for verification.Results show that the scheme can achieve end-to-end high-precision time synchronization with accuracy and sta-bility.展开更多
Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not ...Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not in a time-efficient manner.The time efficiency becomes even worse when the flows to be classified are sampled by flow entry timeout over Software-Defined Networks(SDNs)to achieve a better resource efficiency.This paper addresses this situation by combining co-training and Reinforcement Learning(RL)to enable a closed-loop classification approach that divides the entire classification process into episodes,each involving two elephant models.One predicts elephants and is retrained by a selection of flows automatically labeled online by the other.RL is used to formulate a reward function that estimates the values of the possible actions based on the current states of both models and further adjusts the ratio of flows to be labeled in each phase.Extensive evaluation based on real traffic traces shows that the proposed approach can stably predict elephants using the packets received in the first 10% of their lifetime with an accuracy of over 80%,and using only about 10% more control channel bandwidth than the baseline over the evolved SDNs.展开更多
Zero Trust Network(ZTN)enhances network security through strict authentication and access control.However,in the ZTN,optimizing flow control to improve the quality of service is still facing challenges.Software Define...Zero Trust Network(ZTN)enhances network security through strict authentication and access control.However,in the ZTN,optimizing flow control to improve the quality of service is still facing challenges.Software Defined Network(SDN)provides solutions through centralized control and dynamic resource allocation,but the existing scheduling methods based on Deep Reinforcement Learning(DRL)are insufficient in terms of convergence speed and dynamic optimization capability.To solve these problems,this paper proposes DRL-AMIR,which is an efficient flow scheduling method for software defined ZTN.This method constructs a flow scheduling optimization model that comprehensively considers service delay,bandwidth occupation,and path hops.Additionally,it balances the differentiated requirements of delay-critical K-flows,bandwidth-intensive D-flows,and background B-flows through adaptiveweighting.Theproposed framework employs a customized state space comprising node labels,link bandwidth,delaymetrics,and path length.It incorporates an action space derived fromnode weights and a hybrid reward function that integrates both single-step and multi-step excitation mechanisms.Based on these components,a hierarchical architecture is designed,effectively integrating the data plane,control plane,and knowledge plane.In particular,the adaptive expert mechanism is introduced,which triggers the shortest path algorithm in the training process to accelerate convergence,reduce trial and error costs,and maintain stability.Experiments across diverse real-world network topologies demonstrate that DRL-AMIR achieves a 15–20%reduction in K-flow transmission delays,a 10–15%improvement in link bandwidth utilization compared to SPR,QoSR,and DRSIR,and a 30%faster convergence speed via adaptive expert mechanisms.展开更多
Distributed denial of service(DDoS)attacks are common network attacks that primarily target Internet of Things(IoT)devices.They are critical for emerging wireless services,especially for applications with limited late...Distributed denial of service(DDoS)attacks are common network attacks that primarily target Internet of Things(IoT)devices.They are critical for emerging wireless services,especially for applications with limited latency.DDoS attacks pose significant risks to entrepreneurial businesses,preventing legitimate customers from accessing their websites.These attacks require intelligent analytics before processing service requests.Distributed denial of service(DDoS)attacks exploit vulnerabilities in IoT devices by launchingmulti-point distributed attacks.These attacks generate massive traffic that overwhelms the victim’s network,disrupting normal operations.The consequences of distributed denial of service(DDoS)attacks are typically more severe in software-defined networks(SDNs)than in traditional networks.The centralised architecture of these networks can exacerbate existing vulnerabilities,as these weaknesses may not be effectively addressed in this model.The preliminary objective for detecting and mitigating distributed denial of service(DDoS)attacks in software-defined networks(SDN)is to monitor traffic patterns and identify anomalies that indicate distributed denial of service(DDoS)attacks.It implements measures to counter the effects ofDDoS attacks,and ensure network reliability and availability by leveraging the flexibility and programmability of SDN to adaptively respond to threats.The authors present a mechanism that leverages the OpenFlow and sFlow protocols to counter the threats posed by DDoS attacks.The results indicate that the proposed model effectively mitigates the negative effects of DDoS attacks in an SDN environment.展开更多
Accurate calibration of China's new generation ground-based polarimetric radar(GR) network is crucial yet challenging. Although application of the Dual-frequency Precipitation Radar(DPR) of the Global Precipitatio...Accurate calibration of China's new generation ground-based polarimetric radar(GR) network is crucial yet challenging. Although application of the Dual-frequency Precipitation Radar(DPR) of the Global Precipitation Measurement Core Observatory for GR assessment is well-established, current methodologies are inherently limited. Focusing on three GRs in the Guangdong-Hong Kong-Macao Greater Bay Area(GBA)—strategically selected for their high overlapping coverage(>65%) and distinct from single GR or less dense coverage studies—this work introduces key refinements by integrating innovative enhancements into the volume-matching method(VMM), reflecting a systematic approach to mitigating potential error sources. Specifically, we integrate: 1) a novel frequency correction method that adapts to DPR-observed precipitation phase and type, replacing assumption-based polynomial fitting;and 2) a precise beam time-difference matching approach(accuracy < 1 s) to minimize temporal mismatch errors, which improves upon coarser time averaging methods. Furthermore, we developed statistically robust, optimized threshold criteria based on systematic sensitivity analyses using 11 quality control factors, including precipitation type, bright band effects, and attenuation correction limitations. These criteria establish an enhanced protocol designed to minimize errors arising from instrumental, frequency, and scanning differences. Application of this enhanced methodology to the GBA GRs(2021–2023) yielded a significantly improved matching accuracy(correlation coefficient, CC: 0.90–0.95;standard deviation,STD: 1.2–1.6 dB). A unique contribution of this work is the quantitative estimation of historical calibration errors and operational stability, which was achieved by linking VMM results with operational GR calibration and maintenance records. This analysis revealed decreasing STD trends and identified specific calibration-related events, such as an underestimation of approximately 2.43 dB for the Shenzhen radar following calibration in 2023. Consequently, the refined methodology provides a robust framework for ongoing GR network monitoring and offers a validated pathway for authenticating China's Fengyun-3G(FY-3G) satellite precipitation measurement radar(PMR) data.展开更多
China launched its first spaceborne Precipitation Measurement Radar(PMR)on the FY-3G satellite in April 2023.To achieve the scientific goal of measuring the three-dimensional precipitation structure,evaluating the qua...China launched its first spaceborne Precipitation Measurement Radar(PMR)on the FY-3G satellite in April 2023.To achieve the scientific goal of measuring the three-dimensional precipitation structure,evaluating the quantitative measurement ability of the PMR is critical.China operates more than 250 weather radars over the mainland.Consistency of the spaceborne radar with ground-based radars will enhance precipitation measurement ability,especially over oceans and mountains where observations are sparse.Additionally,the spaceborne radar can be used to evaluate the spatial and temporal homogeneity of the ground-based radar network.This paper focuses on comparing the PMR onboard the FY-3G satellite with S-band China New Generation Weather Radars(CINRADs).A comparison algorithm between the PMR and CINRADs has been developed,incorporating detailed quality control,attenuation correction,data optimization,spatiotemporal matching,non-uniform beam filling constraint,uniformity constraint,and frequency correction.The matched data in typical months of four seasons were selected to carry out the comparison.The data consistency between the PMR and CINRADs was analyzed.The correlation coefficient is 0.87,the deviation is 0.89 dB,and the standard deviation is 2.50 dB,based on 98226 matching samples.The results show the radar reflectivity of the PMR is quite comparable to that of the CINRADs,demonstrating that the PMR data quality is satisfactory and can be used to verify and correct data consistency among multiple ground-based radars.This work also paves the way for data fusion and joint application of satellite and ground radars in the future.展开更多
Dual-polarization(dual-pol)radar variables provide information about the quantity,type,size,and water content of hydrometeors.Assimilating these dual-pol radar variables into numerical weather prediction models can en...Dual-polarization(dual-pol)radar variables provide information about the quantity,type,size,and water content of hydrometeors.Assimilating these dual-pol radar variables into numerical weather prediction models can enhance forecast accuracy.Observation operators are essential for radar data assimilation.This study focuses on applying a realistic dual-pol radar observation operator to more accurately calculate dual-pol radar variables.Previously reported dual-pol radar observation operators tended to overestimate radar variables near 0℃ in convective precipitation and simulate unrealistic dual-pol radar variables in subfreezing regions.To address this,the improved operator(KNU dual-pol radar observation operator;K-DROP)limits the distribution of mixed-phase hydrometeors,which have both solid and liquid properties,in areas with strong updrafts and downdrafts,improving the overestimation of radar variables near the melting layer.Additionally,by applying the observed snow axis ratio during winter to K-DROP,the issue of differential reflectivity(Z_(DR))being calculated as a constant value in subfreezing regions has been improved.By incorporating the observed maximum radius of hydrometeors into K-DROP,the overestimation of reflectivity(Z_(H))in subfreezing regions,the overestimation of Z_(DR)in warm regions,and the underestimation of specific differential phase(K_(DP))in subfreezing regions and overestimation in warm regions,are improved.Compared to previous operators,the enhanced version reported in the present work produces more realistic dual-pol radar variables.展开更多
Ground-based radar is the primary means by which severe storms are monitored and tracked;however, due to limited coverage, important data is often missed over ocean and mountainous areas. On the other hand, geostation...Ground-based radar is the primary means by which severe storms are monitored and tracked;however, due to limited coverage, important data is often missed over ocean and mountainous areas. On the other hand, geostationary(GEO)weather satellites provide continuous observations with seamless coverage with advanced imager, despite their limited capability to penetrate clouds. Combining satellite and ground-radar observations could exploit the advantages of both techniques, providing tracking capability close to that of ground radar while maintaining full spatial coverage. This study presents a novel method called Multi-dimensional satellite Observation information for Radar Estimation(MORE) to reconstruct radar composite reflectivity(CREF). Deep learning techniques are important components of MORE for estimating CREF from China's Fengyun-4B(FY-4B) GEO satellite observations. Two models are developed: an infraredonly(IR-Single) model available for all times, and a visible-infrared(VIS+IR) model for daytime applications. These models incorporate multi-dimensional satellite observation information, including temporal, spatial, spectral, and viewing angle information, to enhance the accuracy of radar echo reconstruction. Results demonstrate that the VIS+IR model outperforms the IR-Single model, and both models achieves a root-mean-square error(RMSE) of less than 6 dBZ and a coefficient of determination(R~2) of greater than 0.7. The models effectively reconstruct radar echoes, including strong echoes exceeding 50 dBZ, and show good agreement with precipitation data in radar-blind areas. This study offers a valuable solution for severe weather monitoring and tracking in regions lacking ground-based radar observations, and provides a potential tool for enhanced data assimilation in numerical weather prediction(NWP) models.展开更多
This paper addresses weak target detection problem for bistatic radar via a pseudo-spectrum(PS)based track-before-detect(TBD).Generally,PS-TBD estimates target position and velocity by means of pseudo-spectrum constru...This paper addresses weak target detection problem for bistatic radar via a pseudo-spectrum(PS)based track-before-detect(TBD).Generally,PS-TBD estimates target position and velocity by means of pseudo-spectrum construction in the discrete measurement space and accurate energy accumulation in mixed coordinates.However,the grids within the polar sensing region of the receivers in the bistatic radar are not aligned.Traditional PS-TBD can not directly process these measurements.In this paper,a PS-TBD method for bistatic radar is proposed to overcome this problem.Each cell in the measurement space of the receivers is mapped to the aligned Cartesian coordinates and predicted to the integration frame according to the assumed filter velocity.A PS is formulated centered on the predicted Cartesian position.Then the samples of the pseudo-spectra are accumulated to the nearest cell around the predicted Cartesian position.The procedure of the energy integration is derived in detail.Simulation results validate the efficacy of the proposed method in terms of detection accuracy and parameter estimation.展开更多
To address the issues of poor adaptability in resource allocation and low multi-agent cooperation efficiency in Joint Radar and Communication(JRC)systems under dynamic environments,an intelligent optimization framewor...To address the issues of poor adaptability in resource allocation and low multi-agent cooperation efficiency in Joint Radar and Communication(JRC)systems under dynamic environments,an intelligent optimization framework integrating Deep Reinforcement Learning(DRL)and Graph Neural Network(GNN)is proposed.This framework models resource allocation as a Partially Observable Markov Game(POMG),designs a weighted reward function to balance radar and communication efficiencies,adopts the Multi-Agent Proximal Policy Optimization(MAPPO)framework,and integrates Graph Convolutional Networks(GCN)and Graph Sample and Aggregate(Graph-SAGE)to optimize information interaction.Simulations show that,compared with traditional methods and pure DRL methods,the proposed framework achieves improvements in performance metrics such as communication success rate,Average Age of Information(AoI),and policy convergence speed,effectively enabling resource management in complex environments.Moreover,the proposed GNN-DRL-based intelligent optimization framework obtains significantly better performance for resource management in multi-agent JRC systems than traditional methods and pure DRL methods.展开更多
In high-intensity electromagnetic warfare,radar systems are persistently subjected to multi-jammer attacks,including potentially novel unknown jamming types that may emerge exclusively under wartime conditions.These j...In high-intensity electromagnetic warfare,radar systems are persistently subjected to multi-jammer attacks,including potentially novel unknown jamming types that may emerge exclusively under wartime conditions.These jamming signals severely degrade radar detection performance.Precise recognition of these unknown and compound jamming signals is critical to enhancing the anti-jamming capabilities and overall reliability of radar systems.To address this challenge,this article proposes a novel open-set compound jamming cognition(OSCJC)method.The proposed method employs a detection-classification dual-network architecture,which not only overcomes the false alarm and misdetection issues of traditional closed-set recognition methods when dealing with unknown jamming but also effectively addresses the performance bottleneck of existing open-set recognition techniques focusing on single jamming scenarios in compound jamming environments.To achieve unknown jamming detection,we first employ a consistency labeling strategy to train the detection network using diverse known jamming samples.This strategy enables the network to acquire highly generalizable jamming features,thereby accurately localizing candidate regions for individual jamming components within compound jamming.Subsequently,we introduce contrastive learning to optimize the classification network,significantly enhancing both intra-class clustering and inter-class separability in the jamming feature space.This method not only improves the recognition accuracy of the classification network for known jamming types but also enhances its sensitivity to unknown jamming types.Simulations and experimental data are used to verify the effectiveness of the proposed OSCJC method.Compared with the state-of-the-art open-set recognition methods,the proposed method demonstrates superior recognition accuracy and enhanced environmental adaptability.展开更多
In this paper,the joint design of transmit and receive beamformers for transmit subaperturing multiple-input-multiple-output(TS-MIMO)radar is investigated,aiming to enhance its low probability of intercept(LPI)capabil...In this paper,the joint design of transmit and receive beamformers for transmit subaperturing multiple-input-multiple-output(TS-MIMO)radar is investigated,aiming to enhance its low probability of intercept(LPI)capability.The main objective is to simultaneously minimize the transmission power,suppress the transmit sidelobe levels,and minimize the probability of intercept,thus bolstering the LPI performance of the radar system while maintaining the desired target detection performance.An alternative optimization method is proposed to jointly optimize the transmit and receive beamformers,yielding an unified LPI optimization framework.Particularly,the proposed iterative algorithm based on the Lagrange duality theory for transmit beamforming is more efficient than the conventional convex optimization method.Numerical experiments highlight the effectiveness of the proposed approach in sidelobe suppression and computational efficiency.展开更多
Integrated sensing and communication(ISAC)is an appealing approach to address spectrum congestion and beamforming is an effective method to realize ISAC.In this paper,we investigate the beamforming design problem for ...Integrated sensing and communication(ISAC)is an appealing approach to address spectrum congestion and beamforming is an effective method to realize ISAC.In this paper,we investigate the beamforming design problem for multiple-input multipleoutput(MIMO)ISAC systems and propose to maximize the radar beampattern gain of the target direction while ensuring the signal-to-interference-plus-noise ratio(SINR)constraints of communication users.Particularly,we discuss two cases of ISAC transmit beamforming,i.e.,Case-Ⅰand Case-Ⅱ,which do not have and do have the dedicated probing signal,respectively.For these two cases of transmit beamforming design problems,we start from the single-user scenario and provide the closed-form solutions for MIMO ISAC beamforming vectors.Then,we consider the multiuser scenario and utilize the semidefinite relaxation technique to convert the beamforming design problems into convex semidefinite programming problems.Furthermore,we investigate the impact of the channel correlation between radar and communication on the performance gain of MIMO ISAC systems and characterize the performance tradeoff.Numerical results validate that the dedicated probing signal is unnecessary in the single-user scenario,whereas it has a slight improvement in target detection performance at low SINR thresholds in the multi-user scenario.It is also shown that the stronger the correlation between radar and communication channels,the greater the performance gain of the system.展开更多
Within the domain of Intelligent Group Systems(IGSs),this paper develops a resourceaware multitarget Constant False Alarm Rate(CFAR)detection framework for multisite MIMO radar systems.It underscores the necessity of ...Within the domain of Intelligent Group Systems(IGSs),this paper develops a resourceaware multitarget Constant False Alarm Rate(CFAR)detection framework for multisite MIMO radar systems.It underscores the necessity of managing finite transmit and receive antennas and transmit power systematically to enhance detection performance.To tackle the multidimensional resource optimization challenge,we introduce a Cooperative Transmit-Receive Antenna Selection and Power Allocation(CTRSPA)strategy.It employs a perception-action cycle that incorporates uncertain external support information to optimize worst-case detection performance with multiple targets.First,we derive a closed-form expression that incorporates uncertainty for the noncoherent integration squared-law detection probability using the Neyman-Pearson criterion.Subsequently,a joint optimization model for antenna selection and power allocation in CFAR detection is formulated,incorporating practical radar resource constraints.Mathematically,this represents an NPhard problem involving coupled continuous and Boolean variables.We propose a three-stage method—Reformulation,Node Picker,and Convex Power Allocation—that capitalizes on the independent convexity of the optimization model for each variable,ensuring a near-optimal result.Simulations confirm the approach's effectiveness,efficiency,and timeliness,particularly for large-scale radar networks,and reveal the impact of threat levels,system layout,and detection parameters on resource allocation.展开更多
Motor imbalance is a critical failure mode in rotating machinery,potentially causing severe equipment damage if undetected.Traditional vibration-based diagnostic methods rely on direct sensor contact,leading to instal...Motor imbalance is a critical failure mode in rotating machinery,potentially causing severe equipment damage if undetected.Traditional vibration-based diagnostic methods rely on direct sensor contact,leading to installation challenges and measurement artifacts that can compromise accuracy.This study presents a novel radar-based framework for non-contact motor imbalance detection using 24 GHz continuous-wave radar.A dataset of 1802 experimental trials was sourced,covering four imbalance levels(0,10,20,30 g)across varying motor speeds(500–1500 rpm)and load torques(0–3 Nm).Dual-channel in-phase and quadrature radar signals were captured at 10,000 samples per second for 30-s intervals,preserving both amplitude and phase information for analysis.A multi-domain feature extraction methodology captured imbalance signatures in time,frequency,and complex signal domains.From 65 initial features,statistical analysis using Kruskal–Wallis tests identified significant descriptors,and recursive feature elimination with Random Forest reduced the feature set to 20 dimensions,achieving 69%dimensionality reduction without loss of performance.Six machine learning algorithms,Random Forest,Extra Trees Classifier,Extreme Gradient Boosting,Categorical Boosting,Support Vector Machine with radial basis function kernel,and k-Nearest Neighbors were evaluated with grid-search hyperparameter optimization and five-fold cross-validation.The Extra Trees Classifier achieved the best performance with 98.52%test accuracy,98%cross-validation accuracy,and minimal variance,maintaining per-class precision and recall above 97%.Its superior performance is attributed to its randomized split selection and full bootstrapping strategy,which reduce variance and overfitting while effectively capturing the nonlinear feature interactions and non-normal distributions present in the dataset.The model’s average inference time of 70 ms enables near real-time deployment.Comparative analysis demonstrates that the radar-based framework matches or exceeds traditional contact-based methods while eliminating their inherent limitations,providing a robust,scalable,and noninvasive solution for industrial motor condition monitoring,particularly in hazardous or space-constrained environments.展开更多
By decoupling control plane and data plane,Software-Defined Networking(SDN) approach simplifies network management and speeds up network innovations.These benefits have led not only to prototypes,but also real SDN dep...By decoupling control plane and data plane,Software-Defined Networking(SDN) approach simplifies network management and speeds up network innovations.These benefits have led not only to prototypes,but also real SDN deployments.For wide-area SDN deployments,multiple controllers are often required,and the placement of these controllers becomes a particularly important task in the SDN context.This paper studies the problem of placing controllers in SDNs,so as to maximize the reliability of SDN control networks.We present a novel metric,called expected percentage of control path loss,to characterize the reliability of SDN control networks.We formulate the reliability-aware control placement problem,prove its NP-hardness,and examine several placement algorithms that can solve this problem.Through extensive simulations using real topologies,we show how the number of controllers and their placement influence the reliability of SDN control networks.Besides,we also found that,through strategic controller placement,the reliability of SDN control networks can be significantly improved without introducing unacceptable switch-to-controller latencies.展开更多
Software-Defined Networking (SDN) has been a hot topic for future network development, which implements the different layers of control plane and data plane respectively. Despite providing high openness and programmab...Software-Defined Networking (SDN) has been a hot topic for future network development, which implements the different layers of control plane and data plane respectively. Despite providing high openness and programmability, the “three-layer two-interface” architecture of SDN changes the traditional network and increases the network attack nodes, which results in new security issues. In this paper, we firstly introduced the background, architecture and working process of SDN. Secondly, we summarized and analyzed the typical security issues from north to south: application layer, northbound interface, control layer, southbound interface and data layer. Another contribution is to review and analyze the existing solutions and latest research progress of each layer, mainly including: authorized authentication module, application isolation, DoS/DDoS defense, multi-controller deployment and flow rule consistency detection. Finally, a conclusion about the future works of SDN security and an idealized global security architecture is proposed.展开更多
The controller is indispensable in software-defined networking(SDN).With several features,controllers monitor the network and respond promptly to dynamic changes.Their performance affects the quality-of-service(QoS)in...The controller is indispensable in software-defined networking(SDN).With several features,controllers monitor the network and respond promptly to dynamic changes.Their performance affects the quality-of-service(QoS)in SDN.Every controller supports a set of features.However,the support of the features may be more prominent in one controller.Moreover,a single controller leads to performance,single-point-of-failure(SPOF),and scalability problems.To overcome this,a controller with an optimum feature set must be available for SDN.Furthermore,a cluster of optimum feature set controllers will overcome an SPOF and improve the QoS in SDN.Herein,leveraging an analytical network process(ANP),we rank SDN controllers regarding their supporting features and create a hierarchical control plane based cluster(HCPC)of the highly ranked controller computed using the ANP,evaluating their performance for the OS3E topology.The results demonstrated in Mininet reveal that a HCPC environment with an optimum controller achieves an improved QoS.Moreover,the experimental results validated in Mininet show that our proposed approach surpasses the existing distributed controller clustering(DCC)schemes in terms of several performance metrics i.e.,delay,jitter,throughput,load balancing,scalability and CPU(central processing unit)utilization.展开更多
Software-Defined Networking(SDN)is an emerging architecture that enables a computer network to be intelligently and centrally controlled via software applications.It can help manage the whole network environment in a ...Software-Defined Networking(SDN)is an emerging architecture that enables a computer network to be intelligently and centrally controlled via software applications.It can help manage the whole network environment in a consistent and holistic way,without the need of understanding the underlying network structure.At present,SDN may face many challenges like insider attacks,i.e.,the centralized control plane would be attacked by malicious underlying devices and switches.To protect the security of SDN,effective detection approaches are indispensable.In the literature,challenge-based collaborative intrusion detection networks(CIDNs)are an effective detection framework in identifying malicious nodes.It calculates the nodes'reputation and detects a malicious node by sending out a special message called a challenge.In this work,we devise a challenge-based CIDN in SDN and measure its performance against malicious internal nodes.Our results demonstrate that such a mechanism can be effective in SDN environments.展开更多
基金supported by Princess Nourah bint Abdulrahman University Researchers Supporting ProjectNumber(PNURSP2025R97),PrincessNourah bint AbdulrahmanUniversity,Riyadh,Saudi Arabia.
文摘The rapid expansion of the Internet of Things(IoT)and Edge Artificial Intelligence(AI)has redefined automation and connectivity acrossmodern networks.However,the heterogeneity and limited resources of IoT devices expose them to increasingly sophisticated and persistentmalware attacks.These adaptive and stealthy threats can evade conventional detection,establish remote control,propagate across devices,exfiltrate sensitive data,and compromise network integrity.This study presents a Software-Defined Internet of Things(SD-IoT)control-plane-based,AI-driven framework that integrates Gated Recurrent Units(GRU)and Long Short-TermMemory(LSTM)networks for efficient detection of evolving multi-vector,malware-driven botnet attacks.The proposed CUDA-enabled hybrid deep learning(DL)framework performs centralized real-time detection without adding computational overhead to IoT nodes.A feature selection strategy combining variable clustering,attribute evaluation,one-R attribute evaluation,correlation analysis,and principal component analysis(PCA)enhances detection accuracy and reduces complexity.The framework is rigorously evaluated using the N_BaIoT dataset under k-fold cross-validation.Experimental results achieve 99.96%detection accuracy,a false positive rate(FPR)of 0.0035%,and a detection latency of 0.18 ms,confirming its high efficiency and scalability.The findings demonstrate the framework’s potential as a robust and intelligent security solution for next-generation IoT ecosystems.
基金supported in part by National Key R&D Program of China(Grant No.2022YFC3803700)in part by the National Natural Science Foundation of China(Grant No.92067102)in part by the project of Beijing Laboratory of Advanced Information Networks.
文摘The rise of time-sensitive applications with broad geographical scope drives the development of time-sensitive networking(TSN)from intra-domain to inter-domain to ensure overall end-to-end connectivity requirements in heterogeneous deployments.When multiple TSN networks interconnect over non-TSN networks,all devices in the network need to be syn-chronized by sharing a uniform time reference.How-ever,most non-TSN networks are best-effort.Path delay asymmetry and random noise accumulation can introduce unpredictable time errors during end-to-end time synchronization.These factors can degrade syn-chronization performance.Therefore,cross-domain time synchronization becomes a challenging issue for multiple TSN networks interconnected by non-TSN networks.This paper presents a cross-domain time synchronization scheme that follows the software-defined TSN(SD-TSN)paradigm.It utilizes a com-bined control plane constructed by a coordinate con-troller and a domain controller for centralized control and management of cross-domain time synchroniza-tion.The general operation flow of the cross-domain time synchronization process is designed.The mecha-nism of cross-domain time synchronization is revealed by introducing a synchronization model and an error compensation method.A TSN cross-domain proto-type testbed is constructed for verification.Results show that the scheme can achieve end-to-end high-precision time synchronization with accuracy and sta-bility.
基金supported by the National Natural Science Foundation of China(61962016)the Ministry of Science and Technology of China(G2022033002L)+1 种基金National Natural Science Foundation of Guangxi(2022JJA170057)Guangxi Education Department’s Project on Improving the Basic Research Ability of Young and Middleaged Teachers in Universities(2023ky0812,Research on Statistical Network Delay Predictions in Large-scale SDNs).
文摘Accurate early classification of elephant flows(elephants)is important for network management and resource optimization.Elephant models,mainly based on the byte count of flows,can always achieve high accuracy,but not in a time-efficient manner.The time efficiency becomes even worse when the flows to be classified are sampled by flow entry timeout over Software-Defined Networks(SDNs)to achieve a better resource efficiency.This paper addresses this situation by combining co-training and Reinforcement Learning(RL)to enable a closed-loop classification approach that divides the entire classification process into episodes,each involving two elephant models.One predicts elephants and is retrained by a selection of flows automatically labeled online by the other.RL is used to formulate a reward function that estimates the values of the possible actions based on the current states of both models and further adjusts the ratio of flows to be labeled in each phase.Extensive evaluation based on real traffic traces shows that the proposed approach can stably predict elephants using the packets received in the first 10% of their lifetime with an accuracy of over 80%,and using only about 10% more control channel bandwidth than the baseline over the evolved SDNs.
基金supported in part by Scientific Research Fund of Zhejiang Provincial Education Department under Grant Y202351110in part by Huzhou Science and Technology Plan Project under Grant 2024YZ23+1 种基金in part by Research Fund of National Key Laboratory of Advanced Communication Networks under Grant SCX23641X004in part by Postgraduate Research and Innovation Project of Huzhou University under Grant 2024KYCX50.
文摘Zero Trust Network(ZTN)enhances network security through strict authentication and access control.However,in the ZTN,optimizing flow control to improve the quality of service is still facing challenges.Software Defined Network(SDN)provides solutions through centralized control and dynamic resource allocation,but the existing scheduling methods based on Deep Reinforcement Learning(DRL)are insufficient in terms of convergence speed and dynamic optimization capability.To solve these problems,this paper proposes DRL-AMIR,which is an efficient flow scheduling method for software defined ZTN.This method constructs a flow scheduling optimization model that comprehensively considers service delay,bandwidth occupation,and path hops.Additionally,it balances the differentiated requirements of delay-critical K-flows,bandwidth-intensive D-flows,and background B-flows through adaptiveweighting.Theproposed framework employs a customized state space comprising node labels,link bandwidth,delaymetrics,and path length.It incorporates an action space derived fromnode weights and a hybrid reward function that integrates both single-step and multi-step excitation mechanisms.Based on these components,a hierarchical architecture is designed,effectively integrating the data plane,control plane,and knowledge plane.In particular,the adaptive expert mechanism is introduced,which triggers the shortest path algorithm in the training process to accelerate convergence,reduce trial and error costs,and maintain stability.Experiments across diverse real-world network topologies demonstrate that DRL-AMIR achieves a 15–20%reduction in K-flow transmission delays,a 10–15%improvement in link bandwidth utilization compared to SPR,QoSR,and DRSIR,and a 30%faster convergence speed via adaptive expert mechanisms.
基金supported by the Deanship of Graduate Studies and Scientific Research at Qassim University for financial support(QU-APC-2025).
文摘Distributed denial of service(DDoS)attacks are common network attacks that primarily target Internet of Things(IoT)devices.They are critical for emerging wireless services,especially for applications with limited latency.DDoS attacks pose significant risks to entrepreneurial businesses,preventing legitimate customers from accessing their websites.These attacks require intelligent analytics before processing service requests.Distributed denial of service(DDoS)attacks exploit vulnerabilities in IoT devices by launchingmulti-point distributed attacks.These attacks generate massive traffic that overwhelms the victim’s network,disrupting normal operations.The consequences of distributed denial of service(DDoS)attacks are typically more severe in software-defined networks(SDNs)than in traditional networks.The centralised architecture of these networks can exacerbate existing vulnerabilities,as these weaknesses may not be effectively addressed in this model.The preliminary objective for detecting and mitigating distributed denial of service(DDoS)attacks in software-defined networks(SDN)is to monitor traffic patterns and identify anomalies that indicate distributed denial of service(DDoS)attacks.It implements measures to counter the effects ofDDoS attacks,and ensure network reliability and availability by leveraging the flexibility and programmability of SDN to adaptively respond to threats.The authors present a mechanism that leverages the OpenFlow and sFlow protocols to counter the threats posed by DDoS attacks.The results indicate that the proposed model effectively mitigates the negative effects of DDoS attacks in an SDN environment.
基金National Key Research and Development Program of China (2023YFB3905801)。
文摘Accurate calibration of China's new generation ground-based polarimetric radar(GR) network is crucial yet challenging. Although application of the Dual-frequency Precipitation Radar(DPR) of the Global Precipitation Measurement Core Observatory for GR assessment is well-established, current methodologies are inherently limited. Focusing on three GRs in the Guangdong-Hong Kong-Macao Greater Bay Area(GBA)—strategically selected for their high overlapping coverage(>65%) and distinct from single GR or less dense coverage studies—this work introduces key refinements by integrating innovative enhancements into the volume-matching method(VMM), reflecting a systematic approach to mitigating potential error sources. Specifically, we integrate: 1) a novel frequency correction method that adapts to DPR-observed precipitation phase and type, replacing assumption-based polynomial fitting;and 2) a precise beam time-difference matching approach(accuracy < 1 s) to minimize temporal mismatch errors, which improves upon coarser time averaging methods. Furthermore, we developed statistically robust, optimized threshold criteria based on systematic sensitivity analyses using 11 quality control factors, including precipitation type, bright band effects, and attenuation correction limitations. These criteria establish an enhanced protocol designed to minimize errors arising from instrumental, frequency, and scanning differences. Application of this enhanced methodology to the GBA GRs(2021–2023) yielded a significantly improved matching accuracy(correlation coefficient, CC: 0.90–0.95;standard deviation,STD: 1.2–1.6 dB). A unique contribution of this work is the quantitative estimation of historical calibration errors and operational stability, which was achieved by linking VMM results with operational GR calibration and maintenance records. This analysis revealed decreasing STD trends and identified specific calibration-related events, such as an underestimation of approximately 2.43 dB for the Shenzhen radar following calibration in 2023. Consequently, the refined methodology provides a robust framework for ongoing GR network monitoring and offers a validated pathway for authenticating China's Fengyun-3G(FY-3G) satellite precipitation measurement radar(PMR) data.
基金jointly supported by the National Natural Science Foundation of China(Grant U2442214)the China Meteorological Administration Youth Innovation Team(Grant No.CMA2024QN10)+1 种基金the National Defense Science and Technology Bureau’s 14th Five-Year Civil Aerospace Preresearch Project(Grant Nos.D030303 and D040204)the International Space Water Cycle Observation Constellation Program(Grant No.183311KYSB20200015).
文摘China launched its first spaceborne Precipitation Measurement Radar(PMR)on the FY-3G satellite in April 2023.To achieve the scientific goal of measuring the three-dimensional precipitation structure,evaluating the quantitative measurement ability of the PMR is critical.China operates more than 250 weather radars over the mainland.Consistency of the spaceborne radar with ground-based radars will enhance precipitation measurement ability,especially over oceans and mountains where observations are sparse.Additionally,the spaceborne radar can be used to evaluate the spatial and temporal homogeneity of the ground-based radar network.This paper focuses on comparing the PMR onboard the FY-3G satellite with S-band China New Generation Weather Radars(CINRADs).A comparison algorithm between the PMR and CINRADs has been developed,incorporating detailed quality control,attenuation correction,data optimization,spatiotemporal matching,non-uniform beam filling constraint,uniformity constraint,and frequency correction.The matched data in typical months of four seasons were selected to carry out the comparison.The data consistency between the PMR and CINRADs was analyzed.The correlation coefficient is 0.87,the deviation is 0.89 dB,and the standard deviation is 2.50 dB,based on 98226 matching samples.The results show the radar reflectivity of the PMR is quite comparable to that of the CINRADs,demonstrating that the PMR data quality is satisfactory and can be used to verify and correct data consistency among multiple ground-based radars.This work also paves the way for data fusion and joint application of satellite and ground radars in the future.
基金supported by the National Research Foundation(NRF)funded by the Korean government(MSIT)(Grant Nos.2022R1A2C1012361,2022R1A6A3A 13073165 and RS-2025-02242970).
文摘Dual-polarization(dual-pol)radar variables provide information about the quantity,type,size,and water content of hydrometeors.Assimilating these dual-pol radar variables into numerical weather prediction models can enhance forecast accuracy.Observation operators are essential for radar data assimilation.This study focuses on applying a realistic dual-pol radar observation operator to more accurately calculate dual-pol radar variables.Previously reported dual-pol radar observation operators tended to overestimate radar variables near 0℃ in convective precipitation and simulate unrealistic dual-pol radar variables in subfreezing regions.To address this,the improved operator(KNU dual-pol radar observation operator;K-DROP)limits the distribution of mixed-phase hydrometeors,which have both solid and liquid properties,in areas with strong updrafts and downdrafts,improving the overestimation of radar variables near the melting layer.Additionally,by applying the observed snow axis ratio during winter to K-DROP,the issue of differential reflectivity(Z_(DR))being calculated as a constant value in subfreezing regions has been improved.By incorporating the observed maximum radius of hydrometeors into K-DROP,the overestimation of reflectivity(Z_(H))in subfreezing regions,the overestimation of Z_(DR)in warm regions,and the underestimation of specific differential phase(K_(DP))in subfreezing regions and overestimation in warm regions,are improved.Compared to previous operators,the enhanced version reported in the present work produces more realistic dual-pol radar variables.
基金supported by the National Natural Science Foundation of China (NSFC) (Grant No.42205044)Feng Yun Application Pioneering Project (FY-APP) Innovation Center for Feng Yun Meteorological Satellite (FYSIC) Special Project (FY-APP-XC-2023.04)the Wuxi University Research Start-up Fund for Recruited Talent。
文摘Ground-based radar is the primary means by which severe storms are monitored and tracked;however, due to limited coverage, important data is often missed over ocean and mountainous areas. On the other hand, geostationary(GEO)weather satellites provide continuous observations with seamless coverage with advanced imager, despite their limited capability to penetrate clouds. Combining satellite and ground-radar observations could exploit the advantages of both techniques, providing tracking capability close to that of ground radar while maintaining full spatial coverage. This study presents a novel method called Multi-dimensional satellite Observation information for Radar Estimation(MORE) to reconstruct radar composite reflectivity(CREF). Deep learning techniques are important components of MORE for estimating CREF from China's Fengyun-4B(FY-4B) GEO satellite observations. Two models are developed: an infraredonly(IR-Single) model available for all times, and a visible-infrared(VIS+IR) model for daytime applications. These models incorporate multi-dimensional satellite observation information, including temporal, spatial, spectral, and viewing angle information, to enhance the accuracy of radar echo reconstruction. Results demonstrate that the VIS+IR model outperforms the IR-Single model, and both models achieves a root-mean-square error(RMSE) of less than 6 dBZ and a coefficient of determination(R~2) of greater than 0.7. The models effectively reconstruct radar echoes, including strong echoes exceeding 50 dBZ, and show good agreement with precipitation data in radar-blind areas. This study offers a valuable solution for severe weather monitoring and tracking in regions lacking ground-based radar observations, and provides a potential tool for enhanced data assimilation in numerical weather prediction(NWP) models.
基金supported by the National Natural Science Foundation of China(62371155)the Heilongjiang Outstanding Youth Science Fund(JQ2022F002)。
文摘This paper addresses weak target detection problem for bistatic radar via a pseudo-spectrum(PS)based track-before-detect(TBD).Generally,PS-TBD estimates target position and velocity by means of pseudo-spectrum construction in the discrete measurement space and accurate energy accumulation in mixed coordinates.However,the grids within the polar sensing region of the receivers in the bistatic radar are not aligned.Traditional PS-TBD can not directly process these measurements.In this paper,a PS-TBD method for bistatic radar is proposed to overcome this problem.Each cell in the measurement space of the receivers is mapped to the aligned Cartesian coordinates and predicted to the integration frame according to the assumed filter velocity.A PS is formulated centered on the predicted Cartesian position.Then the samples of the pseudo-spectra are accumulated to the nearest cell around the predicted Cartesian position.The procedure of the energy integration is derived in detail.Simulation results validate the efficacy of the proposed method in terms of detection accuracy and parameter estimation.
基金funded by Shandong Provincial Natural Science Foundation,grant number ZR2023MF111.
文摘To address the issues of poor adaptability in resource allocation and low multi-agent cooperation efficiency in Joint Radar and Communication(JRC)systems under dynamic environments,an intelligent optimization framework integrating Deep Reinforcement Learning(DRL)and Graph Neural Network(GNN)is proposed.This framework models resource allocation as a Partially Observable Markov Game(POMG),designs a weighted reward function to balance radar and communication efficiencies,adopts the Multi-Agent Proximal Policy Optimization(MAPPO)framework,and integrates Graph Convolutional Networks(GCN)and Graph Sample and Aggregate(Graph-SAGE)to optimize information interaction.Simulations show that,compared with traditional methods and pure DRL methods,the proposed framework achieves improvements in performance metrics such as communication success rate,Average Age of Information(AoI),and policy convergence speed,effectively enabling resource management in complex environments.Moreover,the proposed GNN-DRL-based intelligent optimization framework obtains significantly better performance for resource management in multi-agent JRC systems than traditional methods and pure DRL methods.
文摘In high-intensity electromagnetic warfare,radar systems are persistently subjected to multi-jammer attacks,including potentially novel unknown jamming types that may emerge exclusively under wartime conditions.These jamming signals severely degrade radar detection performance.Precise recognition of these unknown and compound jamming signals is critical to enhancing the anti-jamming capabilities and overall reliability of radar systems.To address this challenge,this article proposes a novel open-set compound jamming cognition(OSCJC)method.The proposed method employs a detection-classification dual-network architecture,which not only overcomes the false alarm and misdetection issues of traditional closed-set recognition methods when dealing with unknown jamming but also effectively addresses the performance bottleneck of existing open-set recognition techniques focusing on single jamming scenarios in compound jamming environments.To achieve unknown jamming detection,we first employ a consistency labeling strategy to train the detection network using diverse known jamming samples.This strategy enables the network to acquire highly generalizable jamming features,thereby accurately localizing candidate regions for individual jamming components within compound jamming.Subsequently,we introduce contrastive learning to optimize the classification network,significantly enhancing both intra-class clustering and inter-class separability in the jamming feature space.This method not only improves the recognition accuracy of the classification network for known jamming types but also enhances its sensitivity to unknown jamming types.Simulations and experimental data are used to verify the effectiveness of the proposed OSCJC method.Compared with the state-of-the-art open-set recognition methods,the proposed method demonstrates superior recognition accuracy and enhanced environmental adaptability.
基金supported by the National Natural Science Foundation of China(62271247)the Natural Science Foundation of Jiangsu Province(BK20240181)+4 种基金the Dreams Foundation of Jianghuai Advance Technology Center(2023-ZM01D001)the National Aerospace Science Foundation of China(20220055052001)the Qing Lan Project of Jiangsu Provincethe Fund of Prospective Layout of Scientific Research for Nanjing University of Aeronautics and Astronauticsthe Key Laboratory of Radar Imaging and Microwave Photonics(Nanjing University of Aeronautics and Astronautics),Ministry of Education。
文摘In this paper,the joint design of transmit and receive beamformers for transmit subaperturing multiple-input-multiple-output(TS-MIMO)radar is investigated,aiming to enhance its low probability of intercept(LPI)capability.The main objective is to simultaneously minimize the transmission power,suppress the transmit sidelobe levels,and minimize the probability of intercept,thus bolstering the LPI performance of the radar system while maintaining the desired target detection performance.An alternative optimization method is proposed to jointly optimize the transmit and receive beamformers,yielding an unified LPI optimization framework.Particularly,the proposed iterative algorithm based on the Lagrange duality theory for transmit beamforming is more efficient than the conventional convex optimization method.Numerical experiments highlight the effectiveness of the proposed approach in sidelobe suppression and computational efficiency.
基金National Natural Science Foundation of China under Grant 62571248 and Grant 62201266Key Laboratory of Intelligent Space TTC&O(Space Engineering University),Ministry of Education under Grant CYK2025-01-12。
文摘Integrated sensing and communication(ISAC)is an appealing approach to address spectrum congestion and beamforming is an effective method to realize ISAC.In this paper,we investigate the beamforming design problem for multiple-input multipleoutput(MIMO)ISAC systems and propose to maximize the radar beampattern gain of the target direction while ensuring the signal-to-interference-plus-noise ratio(SINR)constraints of communication users.Particularly,we discuss two cases of ISAC transmit beamforming,i.e.,Case-Ⅰand Case-Ⅱ,which do not have and do have the dedicated probing signal,respectively.For these two cases of transmit beamforming design problems,we start from the single-user scenario and provide the closed-form solutions for MIMO ISAC beamforming vectors.Then,we consider the multiuser scenario and utilize the semidefinite relaxation technique to convert the beamforming design problems into convex semidefinite programming problems.Furthermore,we investigate the impact of the channel correlation between radar and communication on the performance gain of MIMO ISAC systems and characterize the performance tradeoff.Numerical results validate that the dedicated probing signal is unnecessary in the single-user scenario,whereas it has a slight improvement in target detection performance at low SINR thresholds in the multi-user scenario.It is also shown that the stronger the correlation between radar and communication channels,the greater the performance gain of the system.
基金supported by the National Natural Science Foundation of China(Nos.62071482 and 62471348)the Shaanxi Association of Science and Technology Youth Talent Support Program Project,China(No.20230137)+1 种基金the Innovative Talents Cultivate Program for Technology Innovation Team of Shaanxi Province,China(No.2024RS-CXTD-08)the Youth Innovation Team of Shaanxi Universities,China。
文摘Within the domain of Intelligent Group Systems(IGSs),this paper develops a resourceaware multitarget Constant False Alarm Rate(CFAR)detection framework for multisite MIMO radar systems.It underscores the necessity of managing finite transmit and receive antennas and transmit power systematically to enhance detection performance.To tackle the multidimensional resource optimization challenge,we introduce a Cooperative Transmit-Receive Antenna Selection and Power Allocation(CTRSPA)strategy.It employs a perception-action cycle that incorporates uncertain external support information to optimize worst-case detection performance with multiple targets.First,we derive a closed-form expression that incorporates uncertainty for the noncoherent integration squared-law detection probability using the Neyman-Pearson criterion.Subsequently,a joint optimization model for antenna selection and power allocation in CFAR detection is formulated,incorporating practical radar resource constraints.Mathematically,this represents an NPhard problem involving coupled continuous and Boolean variables.We propose a three-stage method—Reformulation,Node Picker,and Convex Power Allocation—that capitalizes on the independent convexity of the optimization model for each variable,ensuring a near-optimal result.Simulations confirm the approach's effectiveness,efficiency,and timeliness,particularly for large-scale radar networks,and reveal the impact of threat levels,system layout,and detection parameters on resource allocation.
基金funded by Princess Nourah bint Abdulrahman University Researchers Support-ing Project number(PNURSP2026R346)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Motor imbalance is a critical failure mode in rotating machinery,potentially causing severe equipment damage if undetected.Traditional vibration-based diagnostic methods rely on direct sensor contact,leading to installation challenges and measurement artifacts that can compromise accuracy.This study presents a novel radar-based framework for non-contact motor imbalance detection using 24 GHz continuous-wave radar.A dataset of 1802 experimental trials was sourced,covering four imbalance levels(0,10,20,30 g)across varying motor speeds(500–1500 rpm)and load torques(0–3 Nm).Dual-channel in-phase and quadrature radar signals were captured at 10,000 samples per second for 30-s intervals,preserving both amplitude and phase information for analysis.A multi-domain feature extraction methodology captured imbalance signatures in time,frequency,and complex signal domains.From 65 initial features,statistical analysis using Kruskal–Wallis tests identified significant descriptors,and recursive feature elimination with Random Forest reduced the feature set to 20 dimensions,achieving 69%dimensionality reduction without loss of performance.Six machine learning algorithms,Random Forest,Extra Trees Classifier,Extreme Gradient Boosting,Categorical Boosting,Support Vector Machine with radial basis function kernel,and k-Nearest Neighbors were evaluated with grid-search hyperparameter optimization and five-fold cross-validation.The Extra Trees Classifier achieved the best performance with 98.52%test accuracy,98%cross-validation accuracy,and minimal variance,maintaining per-class precision and recall above 97%.Its superior performance is attributed to its randomized split selection and full bootstrapping strategy,which reduce variance and overfitting while effectively capturing the nonlinear feature interactions and non-normal distributions present in the dataset.The model’s average inference time of 70 ms enables near real-time deployment.Comparative analysis demonstrates that the radar-based framework matches or exceeds traditional contact-based methods while eliminating their inherent limitations,providing a robust,scalable,and noninvasive solution for industrial motor condition monitoring,particularly in hazardous or space-constrained environments.
基金supported in part by the National High Technology Research and Development Program(863 Program)of China under Grant No.2011AA01A101the National High Technology Research and Development Program(863 Program)of China under Grant No.2013AA01330the National High Technology Research and Development Program(863 Program)of China under Grant No.2013AA013303
文摘By decoupling control plane and data plane,Software-Defined Networking(SDN) approach simplifies network management and speeds up network innovations.These benefits have led not only to prototypes,but also real SDN deployments.For wide-area SDN deployments,multiple controllers are often required,and the placement of these controllers becomes a particularly important task in the SDN context.This paper studies the problem of placing controllers in SDNs,so as to maximize the reliability of SDN control networks.We present a novel metric,called expected percentage of control path loss,to characterize the reliability of SDN control networks.We formulate the reliability-aware control placement problem,prove its NP-hardness,and examine several placement algorithms that can solve this problem.Through extensive simulations using real topologies,we show how the number of controllers and their placement influence the reliability of SDN control networks.Besides,we also found that,through strategic controller placement,the reliability of SDN control networks can be significantly improved without introducing unacceptable switch-to-controller latencies.
基金supported by the Wuhan Frontier Program of Application Foundation (No.2018010401011295)National High Technology Research and Development Program of China (“863” Program) (Grant No. 2015AA016002)
文摘Software-Defined Networking (SDN) has been a hot topic for future network development, which implements the different layers of control plane and data plane respectively. Despite providing high openness and programmability, the “three-layer two-interface” architecture of SDN changes the traditional network and increases the network attack nodes, which results in new security issues. In this paper, we firstly introduced the background, architecture and working process of SDN. Secondly, we summarized and analyzed the typical security issues from north to south: application layer, northbound interface, control layer, southbound interface and data layer. Another contribution is to review and analyze the existing solutions and latest research progress of each layer, mainly including: authorized authentication module, application isolation, DoS/DDoS defense, multi-controller deployment and flow rule consistency detection. Finally, a conclusion about the future works of SDN security and an idealized global security architecture is proposed.
基金supported by the MSIT(Ministry of Science and ICT),Korea,under the ITRC(Information Technology Research Center)support program(IITP-2020-2018-0-01431)supervised by the IITP(Institute for Information&Communications Technology Planning&Evaluation).
文摘The controller is indispensable in software-defined networking(SDN).With several features,controllers monitor the network and respond promptly to dynamic changes.Their performance affects the quality-of-service(QoS)in SDN.Every controller supports a set of features.However,the support of the features may be more prominent in one controller.Moreover,a single controller leads to performance,single-point-of-failure(SPOF),and scalability problems.To overcome this,a controller with an optimum feature set must be available for SDN.Furthermore,a cluster of optimum feature set controllers will overcome an SPOF and improve the QoS in SDN.Herein,leveraging an analytical network process(ANP),we rank SDN controllers regarding their supporting features and create a hierarchical control plane based cluster(HCPC)of the highly ranked controller computed using the ANP,evaluating their performance for the OS3E topology.The results demonstrated in Mininet reveal that a HCPC environment with an optimum controller achieves an improved QoS.Moreover,the experimental results validated in Mininet show that our proposed approach surpasses the existing distributed controller clustering(DCC)schemes in terms of several performance metrics i.e.,delay,jitter,throughput,load balancing,scalability and CPU(central processing unit)utilization.
基金This work was supported by National Natural Science Foundation of China(No.61802080 and 61802077)Guangdong General Colleges and Universities Research Project(2018GkQNCX105)+1 种基金Zhongshan Public Welfare Science and Technology Research Project(2019B2044)Keping Yu was supported in part by the Japan Society for the Promotion of Science(JSPS)Grants-in-Aid for Scientific Research(KAKENHI)under Grant JP18K18044.
文摘Software-Defined Networking(SDN)is an emerging architecture that enables a computer network to be intelligently and centrally controlled via software applications.It can help manage the whole network environment in a consistent and holistic way,without the need of understanding the underlying network structure.At present,SDN may face many challenges like insider attacks,i.e.,the centralized control plane would be attacked by malicious underlying devices and switches.To protect the security of SDN,effective detection approaches are indispensable.In the literature,challenge-based collaborative intrusion detection networks(CIDNs)are an effective detection framework in identifying malicious nodes.It calculates the nodes'reputation and detects a malicious node by sending out a special message called a challenge.In this work,we devise a challenge-based CIDN in SDN and measure its performance against malicious internal nodes.Our results demonstrate that such a mechanism can be effective in SDN environments.
