Search

Network function virtualization achieves cost-effective network service provisioning through exploitation of virtualization and automation by decoupling network functions (software) from dedicated hardware. The software of the devices can then be hosted by low-cost computation devices rather than by more expensive dedicated devices. To obtain a specific network service, the traffic flow is steered to go through a specific order of network functions that are hosted by cloud computing, and this network function sequence is known as a service function chaining (SFC). To allocate computation resources for network functions and bandwidth resources between network functions in a physical network is the SFC embedding problem. In this paper, we consider the SFC embedding problem in multi-domain networks, where no domain information, like domain topology and network resource, is disclosed among domains. We propose a new optimization algorithm based on column generation method to solve this problem, which is distributedly computed in domains. To further improve the scalability, we provide two heuristic algorithms. We selected two networks one large (158 nodes) and one small (18 nodes) to numerically validate the proposed algorithms and demonstrate that the acceptance ratio obtained by the heuristic algorithms is close (within 5.6%) to that of the optimal algorithm.

With the increasing demand for intelligent transportation systems, short-term traffic flow prediction has become an important research direction. The memory unit of a Long Short-Term Memory (LSTM) neural network can store data characteristics over a certain period of time, hence the suitability of this network for time series processing. This paper uses an improved Gate Recurrent Unit (GRU) neural network to study the time series of traffic parameter flows. The LSTM short-term traffic flow prediction based on the flow series is first investigated, and then the GRU model is introduced. The GRU can be regarded as a simplified LSTM. After extracting the spatial and temporal characteristics of the flow matrix, an improved GRU with a bidirectional positive and negative feedback called the Bi-GRU prediction model is used to complete the short-term traffic flow prediction and study its characteristics. The Rectified Adaptive (RAdam) model is adopted to improve the shortcomings of the common optimizer. The cosine learning rate attenuation is also used for the model to avoid converging to the local optimal solution and for the appropriate convergence speed to be controlled. Furthermore, the scientific and reliable model learning rate is set together with the adaptive learning rate in RAdam. In this manner, the accuracy of network prediction can be further improved. Finally, an experiment of the Bi-GRU model is conducted. The comprehensive Bi-GRU prediction results demonstrate the effectiveness of the proposed method.

Empowered by sufficient robotics and haptic hardware at the edges of a communication network system, Tactile Internet (TI) is the upcoming transformation that will empower the control of IoT in real-time and facilitate a true paradigm shift in making a dream into a reality. However, one of the most crucial challenges in realising TI is to achieve round trip time (RTT) of 1ms or less. This RTT includes transmission, queuing, processing (operator’s end), and acknowledgement times (controlled end). Another challenge is to achieve ultra-high reliability for establishing haptic communications for TI. This paper proposes a novel fog-based traffic flow framework employing Software-Defined Networking (SDN) and Fog Computing (FC) to address these challenges. An efficient traffic flow algorithm is developed to effectively manage complex and critical traffic in the network to reduce extra processing and waiting times at each level of clouds. The SDN and FC approaches provide an effective solution to route the traffic efficiently in the system. Hence, the traffic flow paths from the master to slave sections are reduced, and consequently, RTT is also reduced. The performance of the proposed system is evaluated by an iFogSim simulator in terms of throughput, RTT, energy consumption, and reliability. The simulation results obtained show that the proposed system outperforms the existing edge, cloud, and cellular networks.

Software-defined networking (SDN) has become an attractive solution to carry out centralized and efficient control in industrial internet of things (IIoT). However, its security has received little attention when applied to IIoT, and no comprehensive consideration has been given to attacks against forwarding nodes (FNs), the basic elements of the data plane. Therefore, in this paper, we aim to investigate attacks against FNs from multiple perspectives in software-defined IIoT. To the best of our knowledge, we are the first to systematically consider this kind of attacks. Since it is difficult to pre-deploy all defence methods against various attacks, we propose a deep reinforcement learning (DRL)-based general attack tolerance scheme to guide the benign traffic flow bypass the attacked FNs. Furthermore, in view of the situation that the real dataset is rare and the standard model-based dataset is likely to be impractical, we use generative adversarial network (GAN), a representative deep generative model (DGM), to flexibly generate real-like network traffic for more sufficient and effective experimental verification on the attack tolerance scheme. Experimental results show that our proposed scheme can significantly improve the successful arrival rate of IIoT traffic and achieve near-optimal results.

Benefiting from dense connections in data center networks (DCNs), load balancing algorithms are capable of steering traffic into multiple paths for the sake of preventing traffic congestion. However, given each path’s time-varying and asymmetrical traffic state, this may also lead to worse congestion when some paths are overutilised. Especially in the two-tier hybrid optical/electrical DCNs (Hoe-DCNs), the port contentions and large-grained optical packets of the fast optical switch (FOS) require the top-of-rack (TOR) switch to have microsecond-level load balancing capability for microburst traffic. This paper establishes a leaf-spine Hoe-DCN model to illustrate the principal characteristic of dynamic load balancing in TOR switches for the first time. Moreover, we propose the dynamic distributed multi-path (DDMP) load balancing algorithm that relies on dynamic hashing computing for network flow distribution in DCNs, which dynamically adjusts traffic flow distribution at microsecond level according to the inverse ratio of the buffer occupancy. The simulation results show that our proposed algorithm reduces the TOR-to-TOR latency by 15.88% and decreases the packet loss by 22.06% compared to conventional algorithms under regular load conditions, which effectively improves the overall performance of the Hoe-DCNs. Moreover, our proposed algorithm prevents more than 90% packet loss under low load conditions.

Space-air-ground Internet of vehicle networks is a promising network paradigm to support diverse vehicular services. Exploiting the unique advantages of spatial, aerial, and terrestrial network segments, the quality of service (QoS) of different services can be met by smart access network selection. However, the integrated network inevitably has to face many challenges, such as dynamic network topology, heterogeneous resources, and long propagation latency, which can seriously degrade the QoS performance. In order to provide QoS guarantee to vehicles, we propose a novel architecture called QoS-guaranteed access assistance (QA2) serve vehicles with access assistance. To be specific, a virtualized layer is established in the network, which contains the logical resources of ground network infrastructures. Using that layer, Access Assistants are flexibly deployed at a dynamic set of ground infrastructures, and then help vehicles to acquire network services with satisfied QoS requirements. Furthermore, to accommodate temporal-spatial-varying network demands of vehicles, QA2 adopts a cost-effective deployment scheme, which updates the locations of Assistants based on the real-time traffic flow with the minimum operation cost. The evaluation results demonstrate the effectiveness of QA2, which increases request success rate by more than 50%. The results also prove that the cost-effective deployment scheme can efficiently adapt to the change of traffic flow, and achieve a good balance between QoS guarantee and cost saving. The operation cost is saved by more than 25%, and the QoS performance is improved by more than 22%.

The Internet of Vehicles (IoV) may enhance road safety, improve traffic flow, etc. However, internet connected intelligent vehicles (IVs) are vulnerable to cyber-attacks. One of the important challenges in IoV is thus, verifying data integrity with strict latency requirements. The conventional way of providing data integrity in the Internet cannot be applied to IoV due to excessive overhead and latency. Therefore, most commercially available IVs do not use any security mechanisms for delay sensitive traffic. However, if a road side unit (RSU) has been compromised, it can tamper with the data sent or received by IVs. To solve this issue, this paper presents a light-weight mechanism called DRiVe to establish data integrity for the IVs and detect malicious RSUs. DRiVe is based on a probabilistic model to identify malicious RSUs using specially constructed authentication techniques. The authentication parameters are only sent when a vehicle leaves the coverage area of one RSU and enters that of another. DRiVe does not employ any computationally intensive cryptographic primitives. This significantly reduces the security overhead introduced by sending message authentication codes (MACs) with each packet. A security and performance analysis shows that DRiVe can not only identify malicious RSUs effectively but can do so without introducing any significant communication overhead or latency. The proposed scheme reduces the number of bits transmitted by approximately 7% and decreases the latency incurred by 7.5%. For the scenario where malicious vehicles are present, the proposed scheme achieves a probability of detection close to 99%.

Many real-world problems, such as power line transmission planning, computer network management, water pipe planning, and traffic flow planning, can be modeled using multistate network systems. One of the most interesting aspects of these problems is that one typically wishes to know the reliability of such a plan, namely the network reliability. To evaluate the reliability of a specific network system, the three-stage method (TSM) is typically applied. TSM consists of 1) finding all minimal paths, 2) finding all $d$ -system vectors, and 3) calculating the union probability for reliability, where 1) is one of the most challenging problems discussed in the literature. This article proposes an improved merge search method that combines the advantages of both enumeration and searching technologies. A step-by-step exploration of the proposed method is presented and several benchmark comparisons are performed using well-known algorithms to demonstrate the efficiency of the proposed method. The results show that the computation time of the proposed algorithm can save as much as 10 times against the well-known methods in the literature.

Hybrid hydrogen-electric vehicles (H2EVs) develop rapidly in recent years. Evidently, the planning of H2EVs energy supplying facilities (HVESFs) is important. It involves coupled power flow of power network (PN), mass flow of hydrogen network (HN), and traffic flow of traffic network (TN), simultaneously. These coupled flows would indeed challenge secure and economic performances of HVESFs planning, especially when uncertain renewable energy is introduced. Therefore, a multi-network framework considering coupled flow constraints is proposed for the HVESFs planning, involving all the PN, HN, and TN. Specifically, we first propose a TN based modified maximum covering location method, in order to satisfy energy demand of H2EVs. On this basis, a many-objective optimization based bilevel HVESFs planning model is developed. Then, we use a hybrid differential evolution-interior point algorithm to solve this model, and obtain the planning solutions. Simulation results verify that we could effectively provide solutions for HVESFs planning, and obtain better performances compared to the existing approach.

Accurate and timely traffic flow prediction is crucial for intelligent transportation systems (ITS). Recent advances in graph-based neural networks have achieved promising prediction results. However, some challenges remain, especially regarding graph construction and the time complexity of models. In this paper, we propose a multi-stream feature fusion approach to extract and integrate rich features from traffic data and leverage a data-driven adjacent matrix instead of the distance-based matrix to construct graphs. We calculate the Spearman rank correlation coefficient between monitor stations to obtain the initial adjacent matrix and fine-tune it while training. As to the model, we construct a multi-stream feature fusion block (MFFB) module, which includes a three-channel network and the soft-attention mechanism. The three-channel networks are graph convolutional neural network (GCN), gated recurrent unit (GRU) and fully connected neural network (FNN), which are used to extract spatial, temporal and other features, respectively. The soft-attention mechanism is utilized to integrate the obtained features. The MFFB modules are stacked, and a fully connected layer and a convolutional layer are used to make predictions. We conduct experiments on two real-world traffic prediction tasks and verify that our proposed approach outperforms the state-of-the-art methods within an acceptable time complexity.

Reliable traffic flow prediction is of great value in the field of transportation, which, for example, contributes to traffic control and public safety. The key of achieving better performance is to well capture the non-linear spatial-temporal dependency. The state-of-the-art works consider both aspects, but they ignore the effect of the global trend on local dynamics and fail to capture long-term dynamic dependencies. In this article, we propose a novel Global-Local Temporal Convolutional Network (GL-TCN) to break through these limitations. Specifically, a novel local temporal convolutional mechanism is proposed to capture the long-term local dynamics effectively. Meanwhile, the global and local flow patterns are integrated to handle the effect of the global flow trend on local dynamics. To the best of our knowledge, this is the first work to utilize the temporal convolutional network for traffic flow prediction. Experiments on two real-world datasets demonstrate the superior performance of our method over several state-of-the-art baselines.

Deep learning has achieved good performance in short-term traffic forecasting recently. However, the stochasticity and distribution imbalance are main characteristics to traffic flow, and these will bring the uncertainty and induce the network overfitting problem during deep learning. To deal with the problems, a new end-to-end hybrid deep learning network model, named M-B-LSTM, is proposed for short-term traffic flow forecasting in this paper. In the M-B-LSTM model, an online self-learning network is constructed as a data mapping layer to learn and equalize the traffic flow statistic distribution for reducing the effect of distribution imbalance and overfitting problem during network learning. Besides, the deep bidirectional long short-term memory network (DBLSTM) is introduced to reduce the uncertainty problem by forward and reverse contexts approximation process in the stochasticity reducing layer, and then the long short-term memory network (LSTM) is used to forecast the next traffic flow state in the forecasting layer. Furthermore, sufficient comparative experiments have been conducted and the results show the proposed model has better ability on solving uncertainty and overfitting problems than the state-of-art methods.

In this paper, we propose a novel and adaptive flow rule placement system based on deep reinforcement learning, namely DeepPlace, in Software-Defined Internet of Things (SDIoT) networks. DeepPlace can provide a fine-grained traffic analysis capability while assuring QoS of traffic flows and proactively avoiding the flow-table overflow issue in the data plane. Specifically, we first investigate the traffic forwarding process in an SDIoT network, i.e., routing and flow rule placement tasks. We design a cost function for the routing to set up traffic flow paths in the data plane. Next, we propose an adaptive flow rule placement approach to maximize the number of match-fields in a flow rule at SDN switches. To deal with the dynamics of IoT traffic flows, we model the system operation by using the Markov decision process (MDP) with a continuous action space and formulate its optimization problem. Subsequently, we develop a deep deterministic policy gradient-based algorithm to help the system obtain the optimal policy. The evaluation results demonstrate that DeepPlace can efficiently maintain a significant number of match-fields in a flow rule, i.e., approximately 86% of the maximum level, while minimizing the QoS violation ratio of traffic flows, i.e., 6.7%, in a highly dynamic traffic scenario, which outperforms three other existing solutions, i.e., FlowMan, FlowStat, and DeepMatch.

With the vast expansion of data loads across numerous safety and infotainment applications, the software-defined paradigm has become a critical component for 5G-enabled and intelligent vehicular networks. By transforming the network’s rigid hardware devices into programmable units, SDVN offers a broad view of the entire network status and a standard interface between heterogeneous wireless access technologies. However, deploying a centralized control unit carries several problems, such as bottleneck issues and densification issues. A distributed control plane is a potential alternative to a centralized control plane. Still, it poses many questions about the optimal location of local control units and the number of control units required for a given network structure. In this article, we propose an efficient proactive controller deployment and assignment strategy for distributed 5G-enabled software-defined Vehicular Networks based on predictive vehicles’ traffic flow information and communication delays between switch-enabled roadside units. A thorough performance evaluation is conducted under different vehicular network’s mobility scenarios. We assess the performance of the proposed method in terms of communication latency and load on the subsequent control units and their assigned set of switches. The proposed proactive clustering method is compared to different forms of controller placement methods. The results showed that the proposed approach reduced the network’s communication delays while maintaining a balanced distributed load among controllers over the simulation time.

OBJECTIVES There is a pressing need to improve safety and efficiency in the operating room (OR). Postsurgical adverse events, such as surgical site infections and surgical flow disruption, occur at a significant rate in industrial countries where a considerable portion of such complications result in death. The aim of the study was to identify an ideal room design that improves the flow of staff members using risk and safety performance measures. METHODS Operating room designs were compared by using computer simulation modeling to analyze traffic flow inside an OR. The study was conducted in two phases. A historical data set was first created based on surgical flow data obtained from 23 video observations of actual surgical procedures. A detailed simulation-based model was then developed. RESULTS As room size increases, staff members have more available space to maneuver in the room, resulting in more distance walked but far fewer undesirable contacts. An angled table orientation is preferred with the circulating nurse workstation at the foot of the OR table, as it provides more space for staff to move across the room without increasing the number of contacts. Furthermore, when the nurse workstation is near the wall, staff members experience fewer undesirable contacts. CONCLUSIONS Simulation modeling was used to assess the impact of OR layout alternatives on three performance metrics, and the medium-sized OR prototype performs well across the metrics. Future research will consider the relative influence of several factors on traffic-based safety and efficiency performance metrics, resulting in a more predictive simulation design model.

Due to that multivariate time series, multistep forecasting technology has a guiding role in many fields, such as electricity consumption, traffic flow detection, and stock price prediction, many approaches have been proposed, seeking to realize accurate prediction based on historical data. However, multivariate time series in real-world applications often contain complex and non-linear interdependencies between time steps and series, for which traditional approaches that just model dependencies in time dimension may fail to make forecasting accurately. In addition, the traditional cyclic multistep prediction strategy leads to error accumulation and consequently results in the reduction of prediction performance. To address these limitations, we propose a novel Deep Multivariate Time Series Multistep Forecasting Network based on the Encoder-Decoder model, abbreviated as DualMNet. DualMNet uses the temporal patterns module to capture long-term patterns between time steps, and, simultaneously, employs the spatial patterns module to discover the interdependencies between time series. Furthermore, the decoder utilizes the historical data of the target sequence to predict the long time-series sequences at the final forward operation, significantly ameliorating the issue of error accumulation. Finally, the conducted t-test results, based upon the extensive experimental results on the three benchmark multivariate time series datasets, they demonstrate that the multistep predictive performance of DualMNet is significantly superior to those of the comparison models, and the ablation study shows that integrating all the components of DualMNet together results in robust forecasting performance.

International audience; Recent development in smart devices has lead us to an explosion in data generation and heterogeneity, which requires new network solutions for better analysing and understanding traffic. These solutions should be intelligent and scalable in order to handle the huge amount of data automatically. With the progress of high-performance computing (HPC), it becomes feasible easily to deploy machine learning (ML) to solve complex problems and its efficiency has been validated in several domains (e.g., healthcare or computer vision). At the same time, network slicing (NS) has drawn significant attention from both industry and academia as it is essential to address the diversity of service requirements. Therefore, the adoption of ML within NS management is an interesting issue. In this paper, we have focused on analyzing network data with the objective of defining network slices according to traffic flow behaviors. For dimensionality reduction, the feature selection has been applied to select the most relevant features (15 out of 87 features) from a real dataset of more than 3 million instances. Then, a K-Means clustering is applied to better understand and distinguish behaviors of traffic. The results demonstrated a good correlation among instances in the same cluster generated by the unsupervised learning. This solution can be further integrated in a real environment using network function virtualization.

A standout amongst the most dangers to the cyber security is known as Botnet since it offers a conveyed stage for many undesirable activities. From the network traffic flow, the identification of Botnet is a fundamental test. Artificial Neural Network–Particle Swarm Optimization (ANN–PSO) based botnet discovery is proposed in this paper. In this paper, ISCX dataset is utilized for botnet location. The features are classified as botnet flow and normal flow by giving the features separated from the dataset as a contribution to the grouping. For grouping, we have displayed ANN–PSO which lessens the false classification ratio and time multifaceted nature to 3.3% and 14 s. We contrast our proposed work with other existing work and demonstrate that our work is superior to anything that of alternate works in the simulation results.

The current non-connected autonomous vehicle scheme for speed changing along the road has limitations. Other alternatives require a central control or a complex communication system between the veh...

Energy costs have dramatically increased in data center networks as an increasing number of large-scale Internet applications are used. In software-defined vehicular networks (SDVN), the communication delay between two vehicles and between vehicles and the controller will dramatically climb up as the number of vehicles increases. This requires more controllers to provide communication service to minimize the latency. More controllers lead to high energy costs. Therefore, the number of controllers and their placement, the so-called controller placement problem (CPP), should be addressed. The appropriate placement of controllers can decrease the energy cost, enabling green communication in SDVN. Although CPP has been studied for static networks, it has not been effectively solved in highly dynamic and complex networks. In this article, a novel cost-effective CPP scheme for SDVN is proposed. First, our proposed minimum controller selection mechanism (MOSA) can reduce the number of controllers and guarantee the coverage of the area. Besides, an improved multiobjective artificial bee colony algorithm (IMABC) is proposed based on the original artificial bee colony algorithm. The IMABC can judge which controller should be switched on for data transmission based on real-time traffic flow. A route computation mechanism is proposed to evaluate the performance of our CPP scheme. The experimental results confirm that compared to other existing CPP schemes, our scheme can achieve a higher packet delivery ratio while greatly reducing energy consumption and latency.

Traffic flow prediction is a very challenging task in traffic networks. Travel order quantity prediction is of great value to the analysis of traffic flow. However, the number of travel orders is closely related to time, leading to its short-term proximity and long-term cyclical dependence. Therefore, it is difficult to predict travel order quantity using traditional methods. To capture both the long-term and short-term correlations, in this paper, we propose an attention-based bidirectional long short-term memory network (AT-BLSTM) model, which consists of bidirectional long short-term memory (BLSTM) layer and attention layer. The BLSTM layer contains forward and backward long short-term memory (LSTM), which uses a novel method to combine forward and backward output. The attention layer utilizes novel self-attention algorithm to assign different weights according to the correlation between features. In this case, AT-BLSTM can predict travel order quantity more accurately than other time series models. For example, AT-BLSTM achieves the lowest MAE (0.0647), the lowest RMSE (0.0836) and the lowest MAPE (0.1239) among all the methods on dataset Xiuying. Extensive experiments on real-world travel datasets offer evidence that the proposed approach matches or outperforms state-of-the-art methods.

Filtering refers to an operation to determine whether the concerned data should be accepted and transferred, or be blocked and marked as a malicious traffic flow. It mitigates the inter-domain bandwidth overhead, local computational cost and storage cost for data identification. In many sensitive applications, the identity of the data provider needs to be hidden. This creates challenges how to filter the transmitted data packet with an encrypted form. It is non-trivial to hide this data provider's identity while enabling filtering, as the policy used as a matching criteria will need to determine whether the data needs to be transferred or not without knowing the origin of that data. In this work, we design PPFilter , a privacy-aware encrypted filtering mechanism which allows the filtering to be conducted without the need to know the identity of the data provider. PPFilter achieves the integrity protection of the data packets and the provider privacy Level 3. PPFilter is built on top of a novel notion called identity-based encryption with sender search (IESS) , which supports anonymous sender identity in an encrypted searching. We present a provably secure IESS instantiation, and apply it to achieve a PPFilter protocol. PPFilter allows the data provider's identity to be hidden from both the transferred data and policy while enabling the filtering capability, which solves the aforementioned problem. The analysis and evaluation show that PPFilter maintains cost-reasonable filtering while preserving provider privacy, and hence it guarantees its practicality.

The traditional centralized network architecture can lead to a bandwidth bottleneck in the core network. In contrast, in the information-centric network, decentralized in-network caching can alleviate the traffic flow pressure from the network center to the edge. In this paper, a popularity-aware in-network caching policy, namely, Pop, is proposed to achieve an optimal caching of network contents in the resource-constrained edge networks. Specifically, Pop senses content popularity and distributes content caching without adding additional hardware and traffic overhead. We conduct extensive performance evaluation experiments by using ndnSIM. The experiments showed that the Pop policy achieves 54.39% cloud service hit reduction ratio and 22.76% user request average hop reduction ratio and outperforms other policies including Leave Copy Everywhere, Leave Copy Down, Probabilistic Caching, and Random choice caching. In addition, we proposed an ideal caching policy (Ideal) as a baseline whose popularity is known in advance; the gap of Pop and Ideal in cloud service hit reduction ratio is 4.36%, and the gap in user request average hop reduction ratio is only 1.47%. More simulation results further show the accuracy of Pop in perceiving popularity of contents, and Pop has good robustness in different request scenarios.

Metro system has been increasingly recognized as a backbone of urban transportation system in many cities around the world. To improve the demand management and operation efficiency, it is crucial to have accurate prediction of real-time metro passenger flow. However, the forecast performance is often subject to the complex spatial and temporal distributions of the metro passenger flow data. To this end, we developed a novel dual attentive graph neural network that can effectively predict the distribution of metro traffic flow considering the spatial and temporal influences. Specifically, two directed complete metro graphs (i.e., inbound and outbound graphs) and the weighted matrix of them are proposed to characterize the inbound (entering the system) and outbound (leaving the system) passenger flow, respectively. The weighted matrix of inbound graph is estimated based on the historical origin-destination demand and that of the outbound graph is estimated based on the similarity metrics between every two stations. Moreover, to capture the dependencies between inbound and outbound flows, multi-layer graph spatial attention networks that incorporate the spatial context are applied to exploit the dynamic inter-station correlations. Then, the acquired dependency features integrated with external factors, such as weather conditions, are filtered by temporal attention and fed into a sequence decoder to produce short-term and long-term passenger flow predictions. Finally, a series experiments are conducted based on a comprehensive empirical dataset. Findings indicated that the proposed model does not only well predict the metro passenger flow, but also effectively detect the emergencies and incidents of metro system.

The existing aircraft center track extraction methods only extract the position information of the trajectory, which cannot meet the requirements of abnormal trajectory detection and trajectory prediction. This paper innovatively proposes a center locus extraction algorithm based on multidimensional hierarchical clustering. Firstly, to solve the problem that trajectory resampling is easy to lose the original trajectory features, an equal arc length interpolation resampling method is proposed to process the original trajectory data. Then, the weighted Euclidean distance matrix of the trajectory set is calculated. The calculation model of the weighted Euclidean distance matrix is novel and takes into account the influence of multidimensional features. Finally, multidimensional hierarchical clustering is used to get the traffic flow distribution and output the center trajectory. 703 departure trajectory data from the terminal area of an airport are used for example verification. The results show that compared with the traditional hierarchical clustering, this method has a significant advantage in accurately dividing traffic flow. Moreover, the extracted center locus can retain the multidimensional features of locus, which has certain practical significance.

Graph Convolutional Neural Networks are more and more widely used in traffic flow parameter prediction tasks by virtue of their excellent non-Euclidean spatial feature extraction capabilities. However, most graph convolutional neural networks are only used to predict one type of traffic flow parameter. This means that the proposed graph convolutional neural network may only be effective for specific parameters of specific travel modes. In order to improve the universality of graph convolutional neural networks. By embedding time feature and spatio-temporal attention layer, we propose a spatio-temporal attention graph convolutional neural network based on the attention mechanism of the neural network. Through experiments on passenger flow data and vehicle speed data of two different travel modes (Hangzhou Metro Data and California Highway Data), it is verified that the proposed spatio-temporal attention graph convolutional neural network can be used to predict passenger flow and vehicle speed simultaneously. Meanwhile, the error distribution range of the proposed model is minimum, and the overall level of prediction results is more accurate.

The French standard governing occupanckes dedicated for gatherings of individuals and institutions receiving the public, including railway stations and tunnels was used in the design of Cairo’s transit network, which consists of three lines connected through three existing interchange connections and one that is currently under construction. There is a lack of studies quantifying passenger evacuation in relation to Cairo transit, especially with regard to the interchange connections. This study assessed the viability of the evacuation methods applied to the design of interchange connections in Egypt. The emergency evacuation procedures for the future interchange connection layout between Lines 1 and 3 were assessed using the French standard and NFPA 130 Standard for Fixed Guideway Transit and Passenger Rail Systems analytical methods. The analysis accounts for the traffic forecast as presented in the Egyptian company for operation and maintenance report for Line 1 and the feasibility study of Line 3. The results demonstrated that the connection was compliant with the French standard, but it did not fully meet NFPA requirements. It was concluded that there is still room for enhancement and for gaps to be considered. Recommendations include undertaking investigations into passenger characteristics and behavior in the context of passenger flow in emergency situations using approaches such as virtual research and field surveys, and simulation techniques.

With the rapid development of in-vehicle communication technology and the integration of big data intelligent technology, intelligent algorithms for vehicle communication used to predict traffic flow and location information have been widely used. Aiming at the problem that the gravitational algorithm is difficult to minimize the complex function and easily fall into the local optimum, this paper proposes an improved IGSA algorithm. First, a gridding algorithm is introduced to initialize the population, and under the premise of ensuring the randomness of the initial individuals, improving the ergodicity of the population is conducive to improving the quality of the solution; then, an adaptive location-based update strategy of decreasing inertia weights is proposed. this strategy inherits the advantages of linearly decreasing weights, and adaptively adjusts the weights according to the fitness value to further improve the optimization performance. The optimization simulation of 8 classic test functions shows that the IGSA algorithm is an effective algorithm for solving complex optimization problems. Finally, the IGSA algorithm is used to predict the geographic location problem in the vehicle GPS data. The IGSA algorithm is used to optimize the extreme learning method to optimize the hyperparameters and establish a vehicle GPS data prediction model. Simulation results verify the feasibility of the method.

With the emergence of a large number of computational resource-intensive applications and various content delivery services, there is an explosion of data growth in the Internet of Vehicles (IoV). To improve the transmission performance of the IoV, caching content on the edge of the network is considered as a potential solution to reduce the content transmission delay. In this article, we investigate the content caching decisions optimization method in the IoV to minimize the content fetching delay for vehicles, which is based on the vehicle-to-vehicle (V2V) collaboration. A delay-aware content caching (DCC) algorithm in the IoV is proposed, which consists of vehicle associations, content caching, and precaching decisions optimization. First, a delay-aware vehicle associations (DVAs) algorithm is proposed to optimize the vehicle associations. Consequently, based on the vehicle associations results, the content caching decisions are optimized in two network scenarios according to the existence of the handover vehicles. Finally, a practical scenario of Shanghai with time-varying traffic flow is used for simulations and the effectiveness of the proposed DCC algorithm is verified.

In this paper, we propose a novel flow rule matching framework, DeepMatch, in Software-Defined Networking (SDN) to provide a fine-grained traffic flow measurement capability. Specifically, the flow rule matching control at a particular SDN switch is examined to maximize the traffic flow granularity degree while proactively protecting the flow-table in the switch from being overflowed. This control process is supervised by a control module referred to as DeepMatch instance. Regarding this instance, an optimization problem is formulated based on a Markov decision process (MDP) and a Partially Observable Markov decision process (POMDP), respectively. We develop a deep dueling neural network based flow rule matching control algorithm to solve the optimization problem, thereby quickly attaining a significant traffic flow granularity level and eliminating the switch flow-table overflow problem. Furthermore, we propose an experience data sharing (EDS) mechanism that enables a new instance to learn faster about the flow rule matching control. The results of our performance evaluation show that, by applying the DeepMatch framework in a highly dynamic traffic scenario, the traffic flow granularity degree at the access and the core switches increases by 24.0% and 31.63%, respectively, compared to the FlowStat method. DeepMatch is also highly outperforming the ReWiFlow, SDN-Mon, and Exact-Match approaches. In addition, by employing the EDS mechanism, a new instance can reduce its learning time up to 46.42% for supervising an access switch and up to 37.50% for supervising a core switch.

In this article, a learning-based proactive resource sharing scheme is proposed for the next-generation core communication networks, where the available forwarding resources at a switch are proactively allocated to the traffic flows in order to maximize the efficiency of resource utilization with delay satisfaction. The resource sharing scheme consists of two joint modules, estimation of resource demands and allocation of available resources. For service provisioning, resource demand of each traffic flow is estimated based on the predicted packet arrival rate. Considering the distinct features of each traffic flow, a linear regression algorithm is developed for resource demand estimation, utilizing the mapping relation between traffic flow status and required resources, upon which a network switch makes decision on allocating available resources for delay satisfaction and efficient resource utilization. To learn the implicit relation between the allocated resources and delay, a multi-armed bandit learning-based resource allocation scheme is proposed, which enables fast resource allocation adjustment to traffic arrival dynamics. The proposed algorithm is proved to be asymptotically approaching the optimal strategy, with polynomial time complexity. Extensive simulation results are presented to demonstrate the effectiveness of the proposed resource sharing scheme in terms of delay satisfaction, traffic adaptiveness, and resource allocation gain.

With the widely used Internet of Things, 5G, and smart city technologies, we are able to acquire a variety of vehicle trajectory data. These trajectory data are of great significance which can be used to extract relevant information in order to, for instance, calculate the optimal path from one position to another, detect abnormal behavior, monitor the traffic flow in a city, and predict the next position of an object. One of the key technology is to cluster vehicle trajectory. However, existing methods mainly rely on manually designed metrics which may lead to biased results. Meanwhile, the large scale of vehicle trajectory data has become a challenge because calculating these manually designed metrics will cost more time and space. To address these challenges, we propose to employ network representation learning to achieve accurate vehicle trajectory clustering. Specifically, we first construct the k-nearest neighbor-based internet of vehicles in a dynamic manner. Then we learn the low-dimensional representations of vehicles by performing dynamic network representation learning on the constructed network. Finally, using the learned vehicle vectors, vehicle trajectories are clustered with machine learning methods. Experimental results on the real-word dataset show that our method achieves the best performance compared against baseline methods.

An important part of transportation network vulnerability analysis is identifying critical links where failure may lead to severe consequences, and the potential of such incidents cannot be considered negligible. Existing transportation network vulnerability assessment can be categorized as topological, or traffic based. Topological-based assessment identifies the most critical components in the network by considering network structure and connectivity. Traffic-based assessment identifies the most critical components in the network by full-scan analysis and takes into consideration effects of link failures to traffic flow assignment. The former approach does not consider traffic flow dynamics and fails to capture the non-linearity performance function of transport systems while the latter, even though accurate and robust, requires significant computational power and time and may not always be feasible for real life size networks. The primary objective of this paper is to propose new link criticality measures and evaluate their accuracy for transportation network vulnerability assessment. These measures combine characteristics of traffic equilibrium and network topology to balance accuracy and computational complexity. Nine measures are proposed, and their accuracy is compared with three existing traffic-based measures using three case study transportation networks from the literature. Results indicate that three of the proposed measures show strong correlation to the three traffic-based measures while requiring significantly less computational power and time.

Routing discovery is a pivotal component of the communication stack in low-power and lossy networks (LLNs). The IPv6 routing protocol for LLNs, termed RPL, has been recently standardized to provide routing discovery in a wide range of LLN-based deployments realizing the Internet-of-Things (IoT) vision. RPL was mainly designed with the assumption that the predominant traffic flow would be gathering data toward a single destination, typically the root node. However, node-to-node (N2N) communication, where the root is neither a source nor destination, is a prime requirement in most of the LLN-based applications, such as actuating, decision making, and controlling applications. RPL in its current form does not cater well to such applications where there is a good number of inward N2N traffic flows. In this article, we propose a hybrid routing mechanism based on RPL for high N2N communications in LLNs referred to by HRPL. HRPL is implemented on Contiki and evaluated through extensive simulations on Cooja. HRPL is fully backward compatible with RPL where HRPL and RPL nodes can work together seamlessly in a hybrid network. Results show that HRPL provides a considerable improvement in the packet delivery ratio relative to the standardized RPL-based modes and to the RPL-based opportunistic routing approach. While HRPL requires a slightly larger memory footprint more than that for RPL, HRPL manages to significantly reduce the control plane packets, number of hops, and MAC transmissions needed to successfully deliver N2N data packets. Accordingly, this is translated to lower packet delay and less energy consumption.

Due to the epidemic of COVID-19, more social activities have been moved to the internet, such as online education and online learning. The education management to avoid burst events is a basic requirement of online education, especially when a huge number of persons are visiting at the same time. In order to monitor the abnormal and burst access in online education systems, this paper proposes an anomaly detection method by using data flow to mining high frequency events among massive network traffic data during online education. First, the data flow in traffic network is described as a special structure which is used to establish an efficient high frequent event detection algorithm. Second, the network traffic flow is reduced to make it possible to monitor large-scale concurrent network visiting. The effectiveness of the abnormal network behavior detection method is verified through the experiment on a real network environment for online education. © 2021 IGI Global. All rights reserved.

As accurate and timely traffic flow information is extremely important for traffic management, traffic flow prediction has become a vital component of intelligent transportation systems. However, existing traffic flow prediction methods based on centralized machine learning need to gather raw data for model training, which involves serious privacy exposure risks. To address these problems, federated learning that shares model updates without exchanging raw data, has recently been introduced as an efficient solution for achieving privacy protection. However, the existing federated learning frameworks are based on a centralized model coordinator that still suffers from severe security challenges, such as a single point of failure. Thereby, a consortium blockchain-based federated learning framework is proposed to enable decentralized, reliable, and secure federated learning without a centralized model coordinator. In the proposed framework, the model updates from distributed vehicles are verified by miners to prevent unreliable model updates and are then stored on the blockchain. In addition, to further protect model privacy on the blockchain, a differential privacy method with a noise-adding mechanism is applied for the blockchain-based federated learning framework. Numerical results illustrate that the proposed schemes can effectively prevent data poisoning attacks and improve the privacy protection of model updates for secure and privacy-preserving traffic flow prediction.

Deploying an efficient number and locations of Roadside Units (RSUs) plays an important role in vehicular ad-hoc network. In this work, the realistic urban area “JAWAHARLAL NEHRU UNIVERSITY” (JNU), Delhi India, and its dense areas of macroscopic traffic flow have been used for RSUs’ placement purpose. From the various intersections of the area, only the dense areas of JNU intersections have been considered for RSUs deployment. Further, the random vehicular mobility was generated on the area and studied the communication pattern among vehicle-to-infrastructure (V2I) communication. Finally, the performance of V2I communication was evaluated in terms of Packet Delivery Ratio, Packet Loss, Routing overhead and End-to-End Delay as statistical measures.

This paper proposes a signal timing scheme through a two-stage fuzzy logic controller. The controller first determines the signal phase and then adjusts the green time. At the first stage, the adaptive membership function of vehicle arrival rate is improved to adapt to the changing traffic flow. In addition to arrival rate and queue vehicles, a specific phase order rule is considered to avoid disordered phase selection in fuzzy control. At the second stage, the green time detection module decides whether to extend the current green time or switch phases every few seconds and the vehicle arrival rate is not required as the input to controller in real-time detection. Differential evolution algorithm with low space complexity and fast convergence is applied to optimize the fuzzy rules for avoiding artificial uncertainty. Simulation experiments are designed to compare traditional fuzzy controller, fixed-time controller, and fuzzy controller without flow prediction. Results show that the current proposed method in this paper can reduce vehicle delay significantly.

Network traffic analysis has been increasingly used in various applications to either protect or threaten people, information, and systems. Website fingerprinting is a passive traffic analysis attack which threatens web navigation privacy. It is a set of techniques used to discover patterns from a sequence of network packets generated while a user accesses different websites. Internet users (such as online activists or journalists) may wish to hide their identity and online activity to protect their privacy. Typically, an anonymity network is utilized for this purpose. These anonymity networks such as Tor (The Onion Router) provide layers of data encryption which poses a challenge to the traffic analysis techniques. Although various defenses have been proposed to counteract this passive attack, they have been penetrated by new attacks that proved the ineffectiveness and/or impracticality of such defenses. In this work, we introduce a novel defense algorithm to counteract the website fingerprinting attacks. The proposed defense obfuscates original website traffic patterns through the use of double sampling and mathematical optimization techniques to deform packet sequences and destroy traffic flow dependency characteristics used by attackers to identify websites. We evaluate our defense against state-of-the-art studies and show its effectiveness with minimal overhead and zero-delay transmission to the real traffic.

As the density of cellular networks grows, it becomes exceedingly difficult to provide traditional fiber backhaul access to each cell site. Millimeter wave communication coupled with beamforming can be used to provide high-speed wireless backhaul to such cell sites. Therefore, the 3rd generation partnership project (3GPP) defines an integrated access and backhaul (IAB) architecture, in which the same infrastructure and spectral resources are shared to provide access and backhaul. However, this complicates the design of topologies in such networks as they need to enable efficient traffic flow and minimize congestion or increase robustness to backhaul link failure. We formulate this problem as a graph optimization problem that maximizes the lower bound of the network capacity and propose a topology formation approach based on a combination of deep reinforcement learning and graph embedding. Our proposed approach is significantly less complex, more scalable, and yields very close performance compared to the optimal dynamic programming approach and significant gains when compared with baseline approaches.

Engineered systems get smarter with computing capabilities, particularly through a multitude of Internet-of-Things (IoT) devices. IoT devices, however, are prone to be compromised as they are often resource limited and optimized for a certain task. They lack computing power for security software hence, they have become a major target of malicious activities. In order to secure a network, administrators may isolate vulnerable devices and limit traffic to a device based on its communication needs. In this article, we introduce a novel approach to identify an IoT device based on the locality-sensitive hash of its traffic flow. Different from previous studies that employ machine learning, the proposed approach does not require feature extraction from the data, operates in all states of the device (e.g., setup, idle, and active), and does not require to retrain the model when a new device type/version is introduced. The evaluation results on different data sets show that our approach achieves precision and recall above 90% on average and performs equally well compared to the state-of-the-art machine learning-based methods.

Short-term traffic flow forecasting at isolated points is a fundamental yet challenging task in many intelligent transportation systems. We present a novel long short-term memory (LSTM) network enhanced by temporal-aware convolutional context (TCC) blocks and a new loss-switch mechanism (LSM) to carry out this task. Compared with conventional recurrent neural networks (RNN) or LSTM networks, the proposed network can capture much more distinguishable temporal features and effectively counteracting noise and outliers for more accurate prediction. The proposed TCC blocks, leveraging dilated convolution, produce an enlarged receptive field in temporal contexts, and formulate a temporal-aware attention mechanism to learn the complicated and subtle temporal features from the traffic flows. We further cascade multiple TCC blocks in the network to learn more temporal features at different scales. To deal with the noise and outliers, we propose a novel loss-switch mechanism (LSM) by combining the traditional mean square error loss and the generalized correntropy induced metric (GCIM), which is capable of effectively counteracting non-Gaussian disturbances. The whole network is trained in an end-to-end manner guided by the loss-switch mechanism. Extensive experiments are conducted on two typical benchmark datasets and the experimental results corroborate the superiority of the proposed model over state-of-the-art methods.

The emergence of the Internet of Things (IoT) concept as a new direction of technological development raises new problems such as valid and timely identification of such devices, security vulnerabilities that can be exploited for malicious activities, and management of such devices. The communication of IoT devices generates traffic that has specific features and differences with respect to conventional devices. This research seeks to analyze the possibilities of applying such features for classifying devices, regardless of their functionality or purpose. This kind of classification is necessary for a dynamic and heterogeneous environment, such as a smart home where the number and types of devices grow daily. This research uses a total of 41 IoT devices. The logistic regression method enhanced by the concept of supervised machine learning (logitboost) was used for developing a classification model. Multiclass classification model was developed using 13 network traffic features generated by IoT devices. Research has shown that it is possible to classify devices into four previously defined classes with high performances and accuracy (99.79%) based on the traffic flow features of such devices. Model performance measures such as precision, F-measure, True Positive Ratio, False Positive Ratio and Kappa coefficient all show high results (0.997–0.999, 0.997–0.999, 0.997–0.999, 0–0.001 and 0.9973, respectively). Such a developed model can have its application as a foundation for monitoring and managing solutions of large and heterogeneous IoT environments such as Industrial IoT, smart home, and similar.

Traffic accidents have become a major issue for researchers, academia, government and vehicle manufacturers over the last few years. Many accidents and emergency situations frequently occur on the road. Unfortunately, accidents lead to health injuries, destruction of some infrastructure, bad traffic flow, and more importantly these events cause deaths of hundreds of thousands of people due to not getting treatment in time. Thus, we need to develop an efficient and smart emergency system to ensure the timely arrival of an ambulance service to the place of the accident in order to provide timely medical help to those injured. In addition, we also need to communicate promptly with other entities such as hospitals so that they can make appropriate arrangements and provide timely medical information to emergency personnel on the scene including alerting those related to the injured person(s). In this paper, we have developed an intelligent protocol that uses connected and autonomous vehicles’ scenarios in Intelligent Transportation System (ITS) so that prompt emergency services can be provided to reduce the death rate caused. The proposed protocol smartly connects with all the relevant entitles during the emergency while maintaining a smooth traffic flow for the arrival of the ambulance service. Moreover, our protocol also mitigates the broadcasting of messages circulating over the network for delay sensitive tasks. The evaluation results, based on the performance metrics such as channel collision, average packet delay, packet loss, and routing-overhead demonstrate that our proposed protocol outperforms previously proposed protocols such as Emergency Message Dissemination for Vehicular (EMDV), Contention Based Broadcasting (CBB), and Particle Swarm Optimization Contention-based Broadcast (PCBB) protocols. Finally, we discuss several issues and challenges that need to be addressed in the network in order to achieve more a reliable, efficient, connected, and autonomous vehicular network.

Graph neural networks have shown promising results on representing and analyzing diverse graph-structured data, such as social networks, traffic flow, drug discovery, and recommendation systems. Existing approaches for graph neural networks typically suffer from the oversmoothing issue that results in indistinguishable node representation, as recursive and simultaneous neighborhood aggregation deepens. Also, most methods focus on transductive scenarios that are limited to fixed graphs, which do not generalize properly to unseen graphs. To address these issues, we propose a novel graph neural network that considers both edge-based neighborhood relationships and node-based entity features with multiple steps, i.e. Graph Entities with Step Mixture via random walk (GESM). GESM employs a mixture of various steps through random walk to alleviate the oversmoothing problem, an attention mechanism to dynamically reflect interrelations depending on node information, and structure-based regularization to enhance embedding representation. With intensive experiments, we show that our proposed GESM achieves state-of-the-art or comparable performances on eight benchmark graph datasets in both transductive and inductive learning tasks. Furthermore, we empirically demonstrate the superiority of our method on the oversmoothing issue with rich graph representations. Our source code is available at https://github.com/ShinKyuY/GESM.

With the rapid growth of data exfiltration carried out by cyber attacks, Covert Timing Channels (CTC) have become an imminent network security risk that continues to grow in both sophistication and utilization. These types of channels utilize inter-arrival times to steal sensitive data from the targeted networks. CTC detection relies increasingly on machine learning techniques, which utilize statistical-based metrics to separate malicious (covert) traffic flows from the legitimate (overt) ones. However, given the efforts of cyber attacks to evade detection and the growing column of CTC, covert channels detection needs to improve in both performance and precision to detect and prevent CTCs and mitigate the reduction of the quality of service caused by the detection process. In this article, we present an innovative image-based solution for fully automated CTC detection and localization. Our approach is based on the observation that the covert channels generate traffic that can be converted to colored images. Leveraging this observation, our solution is designed to automatically detect and locate the malicious part (i.e., set of packets) within a traffic flow. By locating the covert parts within traffic flows, our approach reduces the drop of the quality of service caused by blocking the entire traffic flows in which covert channels are detected. We first convert traffic flows into colored images, and then we extract image-based features for detection covert traffic. We train a classifier using these features on a large data set of covert and overt traffic. This approach demonstrates a remarkable performance achieving a detection accuracy of 95.83% for cautious CTCs and a covert traffic accuracy of 97.83% for 8 bit covert messages, which is way beyond what the popular statistical-based solutions can achieve.

Edge computing can provide network services with low latency and real-time processing by operating cloud services on network edges. Edge computing has numerous advantages such as low latency, local characteristics, and network congestion localization, however, associated resource management is becoming a significant challenge because of its features such as hierarchy, decentralization, and heterogeneity. Thus, a joint resource operation and management scheme for edge computing that can greatly reduce the traffic load in the network by considering the generated traffic load and computing resources is proposed in this paper. It is based on the concept of a virtual service flow (VSF) consisting of clients, data sources, a server instance, and external network entities for a specific service. The VSF consists of several drafts according to the expected server location, and each draft estimates a traffic load that can occur according to its characteristics. The location of the edge server of the VSF is determined for each VSF, and it is performed based on the VSF rejection and reconfiguration algorithm using the weighted vector bin packing algorithm that considers the client node coverage of the VSF. The proposed scheme is evaluated based on simulations that consider the actual characteristics of the network services and devices.

Bridge health assessment has been a challenging problem due to great evaluation errors caused by heterogeneous and different dimensional bridge factors. Most approaches employ monitoring data of one bridge factor to simplify the evaluation problem resulting in poor assessment performance. To address this issue, this article proposes an end-to-end framework to evaluate the health of bridges by exploring objective features and correlations of multiple monitoring factors. This model aims to learn representative features from raw monitoring data of bridge factors (i.e., strain, temperature, traffic flow, and heavy vehicle number) and classify the comprehensive features into different health degrees in a single framework. Especially, in terms of characteristics of bridge factors (e.g., time sequence and heterogeneity), a hierarchical learning structure with multiple convolutional, pooling, and dense layers is presented to learn the representations of the whole bridge monitoring data. The structure contributes to capturing rich information from different observed factors and improving feature learning ability through designing particular neural networks for each bridge factor in accordance with their corresponding data structure. In addition, a classification scheme with multiple fully connection layers and support vector machines in the framework is designed to achieve higher evaluation accuracy. Experiments on real-world monitoring data of a specific bridge validate the superiority of the proposed model.

Recent advancements in network technology and associated services have led to a rapid increase in the amount of data traffic. However, the detrimental effects caused by cyber-attacks have also significantly increased. Network attacks are evolving in various forms. Two primary approaches exist for addressing such threats: signature-based detection and anomaly detection. Although the aforementioned approaches can be effective, they also have certain drawbacks. Signature-based detection is vulnerable to variant attacks, while anomaly detection cannot be used for real-time data traffic. For resolving such issues, this paper proposes a two-level classifier that can simultaneously achieve high performance and real-time classification. It employs level 1 and 2 classifiers internally. The level 1 classifier initially performs real-time detection with moderate accuracy for incoming data traffic. If the data cannot be classified with high probability by the classifier, the classification is delayed until the traffic flow terminates. The level 2 classifier then collects the statistical features of the traffic flow for performing precise classification. Compared to existing techniques, the proposed two-level classification method can achieve superior performance in terms of accuracy and detection time.

In order to optimize the existing signal control period division method and improve signal control effect, a new period division method based on Locally Linear Embedding and Particle Swarm Optimization combined with K-means clustering (LLE-PSO-K) algorithm is proposed in this paper. Firstly, traffic flow characteristics of signal-controlled intersections are fully considered, and a multi-dimensional flow matrix is constructed based on the phase traffic flow. In order to reduce the computational complexity of the model and improve the operating efficiency of the method, manifold learning Locally Linear Embedding (LLE) algorithm is brought in to reduce the dimension of the multidimensional phase flow matrix. Then, the dimensionality reduction matrix is used as input data, and signal control period is divided by using Particle Swarm Optimization combined with K-means clustering (PSO-K) algorithm. Finally, an actual intersection in a city is selected to verify the performance of the proposed method. For comparative analysis, control periods are divided based on the phase traffic flow data with 15min, 30min and 1h interval respectively. Results show that for different time intervals, the division of the proposed method is better than other methods, of which the invalid control periods are less. Besides, the optimal clustering number can be obtained, which proves the effectiveness of the new proposed method.