Mobile Networking Beyond 5G

Mobile Networking Beyond 5G Liang Zhao May 2020 沈阳航空航天大学 Shenyang Aerospace University Table of Contents 1. 2. 3. 4. Self-Introduction Multi-hop Networking Air-Ground Networking And future… About the Speaker: Liang Zhao  Academic achievements ▪ Research Interests: vehicular networks, space-air-ground, mobile edge computing. ▪ authored/co-authored 70+ scientific papers (24 first authored, 23 non-first author corresponding author), 32 IEEE/European journals (27 SCI or SCIE), and over 30 in international conferences., including IEEE TPDS, TWC, TON, ICC, etc. ▪ 40+ granted/applied patents, 1 monograph and 2 edited chapters.  Recognition in the international arena ▪ Program Co-Chair (IEEE IUCC 2019), Publicity Chair (Scalcom 2020), Poster&Demo Co-Chair (IEEE Scalcom 2019), Workshop Funder and General Chair (NGDN 2018&2019), Vice-President of CCF-YOCSEF Shenyang, and member of INGR Satellite WG. ▪ Guest Editor: IEEE Transactions on Network Science and Engineering (Lead), Journal of Computing (Lead), and Internet Technology Letters. ❑ Supervision • co-supervised to successful completion of 6 SAU MPhil students and is currently supervising/co-supervising 9 MPhil at SAU including Chinese and oversea students Future Network- Beyond 5G or 6G According to IEEE Future Roadmap, the examples of new revolutionary application heralded with the advent of 5G:  Immersive media and education  Predictive Policing  Mobile Healthcare via wearables  Industrial IoT (IIoT) and autonomous manufactruring  Smart agriculture (drones and field robots)  Environment monitoring  Smart logistics  Telemedicine  Virtual Reality (VR) and Augmented Reality (AR)  … Future Network- Beyond 5G or 6G According to IEEE Future Roadmap, the examples of new revolutionary application heralded with the advent of 5G:  Immersive media and education  Predictive Policing  Mobile Healthcare via wearables  Industrial IoT (IIoT) and autonomous manufacturing  Smart agriculture (drones and field robots)  Environment monitoring  Smart logistics  Telemedicine  Virtual Reality (VR) and Augmented Reality (AR)  … Multi-hop (vehicular) Networking AIR-Ground Networking We need ubiquitous intelligent connectivity and computing to realize the above applications. Multi-hop Networking -Wireless Mesh Network  A network architecture to provide metropolitan broadband access.  Attractive features: rapid rollout, low capital costs, low power consumption, self configuration and organization, high integration with other networks, and easy installation.  WMN is capable of using the basic radio frequency to provide robust, flexible mobile broadband communications to different communities through the readily attainable multihop connection A.Al-Dubai, L.Zhao, A.Zomaya, G.Min, QoS-Aware Inter-Domain Multicast for Scalable Wireless Community Networks, IEEE Transactions on Parallel and Distributed Systems, 2015, 3136-314 L. Zhao, A. Al-Dubai, and G. Min, “GLBM: A New QoS Aware Multicast Scheme for Wireless Mesh Networks,” Performance Evaluation and Optimization of Ubiquitous Computing and Networked Systems, Journal of Systems and Software, vol. 83, iss. 8, pp. 1318-1326, August 2010. Multi-hop Networking -Wireless Mesh Network  GMR shifts the role of the gateway from a simple packet forwarder to a routing orchestrating node. GMR is a hybrid routing in which both proactive and reactive components cooperate concurrently. This combination mode makes GMR suitable for implementation on a variety of different network configurations.  Built upon GMR, a novel Multicast Gateway Multi-hop Routing algorithm is proposed to handle high-bandwidth applications. Unlike the existing work, the MGMR considers the QoS provisioning, load balancing and the capability of gateway-based management in largescale networks. A.Al-Dubai, L.Zhao, A.Zomaya, G.Min, QoS-Aware Inter-Domain Multicast for Scalable Wireless Community Networks, IEEE Transactions on Parallel and Distributed Systems, 2015, 3136-314 L. Zhao, A. Al-Dubai, and G. Min, “GLBM: A New QoS Aware Multicast Scheme for Wireless Mesh Networks,” Performance Evaluation and Optimization of Ubiquitous Computing and Networked Systems, Journal of Systems and Software, vol. 83, iss. 8, pp. 1318-1326, August 2010. Multi-hop Networking -Wireless Mesh Network A.Al-Dubai, L.Zhao, A.Zomaya, G.Min, QoS-Aware Inter-Domain Multicast for Scalable Wireless Community Networks, IEEE Transactions on Parallel and Distributed Systems, 2015, 3136-314 L. Zhao, A. Al-Dubai, and G. Min, “GLBM: A New QoS Aware Multicast Scheme for Wireless Mesh Networks,” Performance Evaluation and Optimization of Ubiquitous Computing and Networked Systems, Journal of Systems and Software, vol. 83, iss. 8, pp. 1318-1326, August 2010. Multi-hop Networking -Wireless Mesh Network  We propose two new routing metrics, namely, the Packet Priority-Oriented routing metric (PPO) and Packet Priority-oriented QoS routing metric (PP-QoS), to enhance the QoS of WMNs.  Our schemes are based on the different communication demands of mesh applications, indirectly providing superior paths in order to provide QoS provisioning services to realtime applications. L.Zhao, A.Al-Dubai, X.Li, G.Chen, G.Min, A New Efficient Cross-layer Relay Node Selection Model for Wireless Community Mesh Networks, Computers & Electrical Engineering, 2017. L.Zhao, A.Al-Dubai, X.Li and G.Chen, “A New Relay Node Selection Model for Wireless Community Mesh Networks”, Accepted by the 14th IEEE International Conference on Ubiquitous Computing and Communications (IUCC-2015), 2015, Liverpool UK. (EI) (Best Paper Award) Multi-hop Networking -Wireless Mesh Network Random Grid Heavy-loaded networks L.Zhao, A.Al-Dubai, X.Li, G.Chen, G.Min, A New Efficient Cross-layer Relay Node Selection Model for Wireless Community Mesh Networks, Computers & Electrical Engineering, 2017. L.Zhao, A.Al-Dubai, X.Li and G.Chen, “A New Relay Node Selection Model for Wireless Community Mesh Networks”, Accepted by the 14th IEEE International Conference on Ubiquitous Computing and Communications (IUCC-2015), 2015, Liverpool UK. (EI) (Best Paper Award) Multi-hop Networking -Wireless Mesh Network Random Grid Light traffic networks L.Zhao, A.Al-Dubai, X.Li, G.Chen, G.Min, A New Efficient Cross-layer Relay Node Selection Model for Wireless Community Mesh Networks, Computers & Electrical Engineering, 2017. L.Zhao, A.Al-Dubai, X.Li and G.Chen, “A New Relay Node Selection Model for Wireless Community Mesh Networks”, Accepted by the 14th IEEE International Conference on Ubiquitous Computing and Communications (IUCC-2015), 2015, Liverpool UK. (EI) (Best Paper Award) Multi-hop Networking -Vehicular Ad-hoc Networks  Vehicular Ad-hoc Networks (VANETs)- Features: (1) the size of the network: the number of vehicles is enormous; (2) node mobility: The node moving speed of the vehicle along the road network; (3) the topology and network density: topology and density nodes can be very frequent in the entire network or part of the network, due to changes in fast moving speed of the vehicle; (4) hardware: there is no energy constraint in vehicles node.  The selection of next-hop forwarding node in some cases is not optimal. For these reasons, we propose a new algorithm namely, Greedy Machine Learning Routing (GMLR) by applying a machine learning algorithm (Support Vector Machines, hereinafter referred to as SVM) to improve the routing metric model in location-based routing protocols like GPSR. L. Zhao, Y. Li, C. Meng, C. Gong and X. Tang, "A SVM based routing scheme in VANETs," 2016 16th International Symposium on Communications and Information Technologies (ISCIT), Qingdao, 2016, 380-383. Multi-hop Networking -Vehicular Ad-hoc Networks  The inter-path process (i.e., selecting a sequence of successive road segments) is modeled as a multi-objective function that aggregates multiple attributes, density, and shortest distance. The density is defined as a linguistic variable with three fuzzy sets, Low, Medium and High.  The intra-path process (i.e., selecting the relay vehicles on a road segment) is mathematically modeled as a multi-objective function, which is utterly driven by multiple attributes such as speed difference, movement direction of vehicles, signal fading or path loss and transmission distance. Each of these attributes is modeled as a fuzzy set independently.  To identify the relationships among the captured attributes in intra-path and inter-path processes, Analytical Hierarchy Process is applied. In addition, to obtain the forwarding decision, TSK inference system is used. A. Hawbani, E. Torbosh, X. Wang, P. Sincak, L. Zhao, and A. Al-Dubai, “Fuzzy based Distributed Protocol for Vehicle to Vehicle Communication,” IEEE Transactions on Fuzzy Systems, 2019. (CA) Multi-hop Networking -Vehicular Ad-hoc Networks A. Hawbani, E. Torbosh, X. Wang, P. Sincak, L. Zhao, and A. Al-Dubai, “Fuzzy based Distributed Protocol for Vehicle to Vehicle Communication,” IEEE Transactions on Fuzzy Systems, 2019. (CA) Multi-hop Networking -Vehicular Ad-hoc Networks A. Hawbani, E. Torbosh, X. Wang, P. Sincak, L. Zhao, and A. Al-Dubai, “Fuzzy based Distributed Protocol for Vehicle to Vehicle Communication,” IEEE Transactions on Fuzzy Systems, 2019. (CA) Multi-hop Networking -Vehicular Ad-hoc Networks  SDVN is a promising vehicular networking paradigm which can provide extensible and flexible means to manage networks to enable V2V and V2I communications. With SDVN, new routing schemes can be deployed easily.  SDVN decouples the data plane and the control plane so that it can separate data forwarding functions and network functions. The SDVN-based architecture consists of two main components, controller and device. Devices transfer packets based on strategies dictated to each of them from the controller which has global knowledge about devices. L.Zhao, A. Al-Dubai, A. Y. Zomaya, G. Min, A. Hawbani, and J. Li, “Routing Schemes in Software-defined Vehicular Networks: Design, Open Issues and Challenges”, IEEE Intelligent Transportation Systems Magazine (ITSM), 2020. Multi-hop Networking -Vehicular Ad-hoc Networks L.Zhao, A. Al-Dubai, A. Y. Zomaya, G. Min, A. Hawbani, and J. Li, “Routing Schemes in Software-defined Vehicular Networks: Design, Open Issues and Challenges”, IEEE Intelligent Transportation Systems Magazine (ITSM), 2020. Multi-hop Networking -Vehicular Ad-hoc Networks Open Issues  Lack of Research in Dedicated Trajectory Prediction Algorithm: The trajectory prediction is to obtain the future status of the vehicle based on its current status. The traditional trajectory prediction algorithms normally apply different movement analysis models.  Further Reduction of Communication Overhead: The status beacons and routing messages sent from/to the controller generate high uplink/downlink communication overhead.  Lack of Research in Routing Algorithms in Controller: Most existing works apply the static shortest path algorithm like Dijsktra’s algorithm, in the controller to compute the routes for routing queries. However, most links between vehicle pairs are only valid for a certain amount of time or the weight of the links are dynamic in SDVNs.  Lack of Applying AI (Artificial Intelligence) in Routing Management in Controller: So far existing solutions lack the research of using AI to enhance the routing performance of SDVNs. In particular, exploring growing data traffic to manage network routing is a very promising approach to deal with dynamic and large-scale SDVNs. L.Zhao, A. Al-Dubai, A. Y. Zomaya, G. Min, A. Hawbani, and J. Li, “Routing Schemes in Software-defined Vehicular Networks: Design, Open Issues and Challenges”, IEEE Intelligent Transportation Systems Magazine (ITSM), 2020. Multi-hop Networking -Vehicular Ad-hoc Networks Open Issues  Lack of Research in Multicast Routing: Most current studies focus on providing the unicast routing. However, multicasting is also a fundamental technology for many key vehicular applications such as collision avoidance, cooperative driving and so forth.  Lack of Considering Security in Routing: Security is a particular important issue in vehicular communications related to the safety of in-car passengers, vehicles, pedestrians and other public entities. With the existence of controller, SDVN could be less vulnerable to cyber-attacks than other types of wireless vehicular networks, by enabling the central coordination in controller. L.Zhao, A. Al-Dubai, A. Y. Zomaya, G. Min, A. Hawbani, and J. Li, “Routing Schemes in Software-defined Vehicular Networks: Design, Open Issues and Challenges”, IEEE Intelligent Transportation Systems Magazine (ITSM), 2020. Multi-hop Networking -Vehicular Ad-hoc Networks Future Direction L.Zhao, A. Al-Dubai, A. Y. Zomaya, G. Min, A. Hawbani, and J. Li, “Routing Schemes in Software-defined Vehicular Networks: Design, Open Issues and Challenges”, IEEE Intelligent Transportation Systems Magazine (ITSM), 2020. Multi-hop Networking -Vehicular Ad-hoc Networks  Hybrid SDVN: The local controller acts as a mobile edge node for collecting, processing the local feature data, and switching routing schemes based on the central controller's decisionmaking model. The central controller is responsible for merging the data and training the decision-making model in real time.  We propose a method for extracting road network information. By obtaining real-time road network information, road network characteristics and traffic conditions can be described.  Adaptive routing and switching scheme: We applied the OS-ELM for real-time model training. This kind of artificial neural network can train data chunk-by-chunk or one-by-one (a special case of chunk), so it can be used for real-time applications. L. Zhao, W. Zhao, A. Al-Dubai, G. Min, “A Novel Adaptive Routing and Switching Scheme for Software-Defined Vehicular Networks,” 2019 IEEE International Conference on Communications (ICC), Shanghai, 2019. Multi-hop Networking -Vehicular Ad-hoc Networks L. Zhao, W. Zhao, A. Al-Dubai, G. Min, “A Novel Adaptive Routing and Switching Scheme for Software-Defined Vehicular Networks,” 2019 IEEE International Conference on Communications (ICC), Shanghai, 2019. Multi-hop Networking -Vehicular Ad-hoc Networks L. Zhao, W. Zhao, A. Al-Dubai, G. Min, “A Novel Adaptive Routing and Switching Scheme for Software-Defined Vehicular Networks,” 2019 IEEE International Conference on Communications (ICC), Shanghai, 2019. Multi-hop Networking -Vehicular Ad-hoc Networks  An efficient optimal routing algorithm for the temporal graph is proposed by applying the properties of temporal graphs. Within linear computation time cost, the single-source shortest path can be achieved.  HMM-based network: all parameters are adaptively adjusted according to the source vehicle, destination vehicle, and the current status of VN.  A novel prediction strategy: predict enough REIs in the HMM. With the constructed HMM, a proper number of qualified REIs are generated. Routing path among these REIs guarantees the reachability and the limited delay. A temporal graph corresponding to the current vehicular network is constructed. All the edges in it are possible REIs in future. L. Zhao, Z. Li, J. Li, A. Al-Dubai, G. Min, A. Zomaya, “A Temporal-information-based Adaptive Routing Algorithm for Software Defined Vehicular Networks,” 2019 IEEE International Conference on Communications (ICC), Shanghai, 2019. (Extended version submitted to IEEE TMC) Multi-hop Networking -Vehicular Ad-hoc Networks L. Zhao, Z. Li, J. Li, A. Al-Dubai, G. Min, A. Zomaya, “A Temporal-information-based Adaptive Routing Algorithm for Software Defined Vehicular Networks,” 2019 IEEE International Conference on Communications (ICC), Shanghai, 2019. (Extended version submitted to IEEE TMC) Multi-hop Networking -Vehicular Ad-hoc Networks L. Zhao, Z. Li, J. Li, A. Al-Dubai, G. Min, A. Zomaya, “A Temporal-information-based Adaptive Routing Algorithm for Software Defined Vehicular Networks,” 2019 IEEE International Conference on Communications (ICC), Shanghai, 2019. (Extended version submitted to IEEE TMC) AIR-Ground Networking UAV Path Planning  The improved artificial potential field (APF) method is adopted to accelerate the convergence of the bat’s position update process  The optimal success rate strategy is proposed to improve the adaptive inertia weight of bat algorithm. It also balances the global search and the local search and makes the algorithm with great robustness.  The chaos strategy is adopted in the initial contribution of bat swarms. It makes the search process avoid from local optimum and updates the convergence rate. Na Lin, Jiacheng Tang, Xianwei Li, and Liang Zhao, “A Novel Improved Bat Algorithm in UAV Path Planning,” Computers, Materials & Continua, 2019 (CA) AIR-Ground Networking UAV Path Planning APF Problem: UAV path planning is defined as the process of finding a path from the start point to the end point while meeting with the performance requirements of the UAV under some specific UAV flight constraints. It aims to search the extreme value of multi-objective function under the condition of multiple constraints. Solution: The APF method (Artificial Potential Field, APF) was first proposed for mobile robot path planning and obstacle avoidance problems. The APF method is inspired by the principle of the gravity force and the repulsive force. The gravity force is commonly generated by the heterogeneous charge with the different type of electrostatic charge between the target point and UAV. Na Lin, Jiacheng Tang, Xianwei Li, and Liang Zhao, “A Novel Improved Bat Algorithm in UAV Path Planning,” Computers, Materials & Continua, 2019 (CA) AIR-Ground Networking UAV Path Planning The adaptive inertia weight Problem: Similar with the exploration process and the exploit process in the standard heuristic search algorithm, the swarm intelligent algorithm has the process of global search, and local search in the whole optimize the process. Global search is aimed to determine the approximate range of the optimal solution, and local search is aimed to calculate the optimization fitness. Solution: The adaptive inertia weight based on optimal success rate transform the occasion of global search and local search. It reflects on the development of a globally optimal solution. Compared with other linear inertia weight, our proposed method has great robustness. Simulation and experiment results will prove our views. The improved APF method accelerates the convergence rate of the path planning process. Compared with the standard APF method, we redefine the attractive potential field function and the repulsive potential field function. After the derivation process, we get the attractive force function and the repulsive force function, which mainly influence the movement of UAV in the potential field. The standard APF method has shortcomings as low robustness and easy to fall into local optimum. Our proposed APF method set the threshold value to modify the attractive force and the repulsive force, which makes the fight of UAV with high-efficiency and matching with reality. Na Lin, Jiacheng Tang, Xianwei Li, and Liang Zhao, “A Novel Improved Bat Algorithm in UAV Path Planning,” Computers, Materials & Continua, 2019 (CA) AIR-Ground Networking UAV Path Planning Chaos strategy Problem: In order to traverse the solution space completely, it requires that the initial bat's population should be distributed randomly. Solution: Chaos strategy satisfies this demand and can be combined with the improved bat algorithms to reallocate the initial distribution of the bat’s population. Chaos strategy is a pseudo-random phenomenon with the feature of random distribution. Na Lin, Jiacheng Tang, Xianwei Li, and Liang Zhao, “A Novel Improved Bat Algorithm in UAV Path Planning,” Computers, Materials & Continua, 2019 (CA) AIR-Ground Networking Multi-UAV Clustering  The combination of cooperative control and secure communication, which guarantees the secure communication in the scenario of multi-UAV cooperative control flight.  The flocking algorithm is proposed by Olfati-Saber, enabling multi-UAV to fly cooperatively and reach a relatively stable position. In the flocking process, communication links are produced suitable for the cooperative control of multi-UAV. J. Wu, L. Zou, L. Zhao, A. Al-Dubai, L. Mackenzie, G. Min, “A Multi-UAV Clustering Strategy for Reducing Insecure Communication Range,” Computer Networks, vol.158, pp. 132-142, 2019. (CA) AIR-Ground Networking Multi-UAV Clustering  the UAV dynamic model is used to discretion UAV trajectories and the position of each UAV, velocity and other information. We use the graph theory to describe the topological structure of multi-UAV groups in flight and obtain information about each individual. The clustering algorithm improves flocking to control the UAVs cooperatively. The flocking algorithm itself is derived from the flight behavior of birds in nature where a dynamic hierarchical network is formed. J. Wu, L. Zou, L. Zhao, A. Al-Dubai, L. Mackenzie, G. Min, “A Multi-UAV Clustering Strategy for Reducing Insecure Communication Range,” Computer Networks, vol.158, pp. 132-142, 2019. (CA) AIR-Ground Networking Multi-UAV Clustering  The HVCR is presented in this paper to address the communication strategy. First, the UAV group is stratified to find the boundary UAVs. Then, the insecure range decreases by reducing the communication radius of these. Finally, the movement algorithm is used to move the boundary UAVs to communicate with the UAV group. J. Wu, L. Zou, L. Zhao, A. Al-Dubai, L. Mackenzie, G. Min, “A Multi-UAV Clustering Strategy for Reducing Insecure Communication Range,” Computer Networks, vol.158, pp. 132-142, 2019. (CA) AIR-Ground Networking Multi-UAV Clustering J. Wu, L. Zou, L. Zhao, A. Al-Dubai, L. Mackenzie, G. Min, “A Multi-UAV Clustering Strategy for Reducing Insecure Communication Range,” Computer Networks, vol.158, pp. 132-142, 2019. (CA) AIR-Ground Networking UAV Transmission  Based on the constraints of limited time for maximum transmission, an energy consumption model for UAVs data transmission is established. This is then converted into the optimal stopping problem to find the optimal data-energy efficiency.  In order to reduce the energy consumption of data transmission, the optimal stopping based target UAV selection mechanism is proposed.  By analyzing the impacts on the time limit of different maximum transmission, safety communication radius and arrival time interval parameters of target UAV on the optimal data-energy efficiency are given. J. Wu, J. Ma, Y. Rou, L. Zhao and R. Ahmad, "An Energy-Aware Transmission Target Selection Mechanism for UAV Networking," in IEEE Access, 2019. (CA) AIR-Ground Networking UAV-Assisted VANET  A novel collaborative network architecture integrates the drones with VANETs. For a certain position, we define the detailed criteria to evaluate the demand for drones quantitatively. It considers multiple objectives of VANETs and builds an evaluation function for optimization. The purpose of this evaluation function is to find the optimal distribution of multiple drones to assist VANET, which is modeled as a multimodal optimization problem. In order to reduce the energy consumption of data transmission, the optimal stopping based target UAV selection mechanism is proposed. N. Lin, L. Fu, L. Zhao, G. Min, A. Al-Dubai, H. Gacanin, “A Novel Multimodal Collaborative-based Drone-assisted VANET Networking Model,” IEEE Transactions on Wireless Communications, 2020. (CA) AIR-Ground Networking UAV-Assisted VANET  To improve the above model, we design a specific multimodal optimization algorithm, namely, Multimodal Nomad Algorithm (MNA), inspired by the migratory behavior of the nomadic tribes on the Mongolia prairie. MNA enables the instant dispatching of multiple drones to the best service positions in order to enhance the efficiency of drone-assisted VANETs. N. Lin, L. Fu, L. Zhao, G. Min, A. Al-Dubai, H. Gacanin, “A Novel Multimodal Collaborative-based Drone-assisted VANET Networking Model,” IEEE Transactions on Wireless Communications, 2020. (CA) AIR-Ground Networking UAV-Assisted VANET N. Lin, L. Fu, L. Zhao, G. Min, A. Al-Dubai, H. Gacanin, “A Novel Multimodal Collaborative-based Drone-assisted VANET Networking Model,” IEEE Transactions on Wireless Communications, 2020. (CA) AIR-Ground Networking UAV-Assisted VANET N. Lin, L. Fu, L. Zhao, G. Min, A. Al-Dubai, H. Gacanin, “A Novel Multimodal Collaborative-based Drone-assisted VANET Networking Model,” IEEE Transactions on Wireless Communications, 2020. (CA) Future Networks IDT-SDVN  The physical SDVN is the actual network running in the real world. In this case, controllers are set up to calculate the routing requests as well as scheduling vehicles when acting as a Mobile Edge Computing (MEC) server. On the other hand, the IDT part as the virtual network(s) is constructed by the controller to model and verify the instant learned functional model. L. Zhao, G. Han, Z. Li, L. Shu, “Intelligent Digital Twin-based Software-Defined Vehicular Networks”, IEEE Network, 2020 Future Networks IDT-SDVN: services  Data Aggregation and Virtualization: With the global view, each controller can collect and store vehicular and road data within its region.  IDT Networking: Controller offers higher computational power compared to vehicles which enable intelligent algorithm operated to refine the networking schemes of SDVN from time to time.  IDT Verification: Before applying to the real physical world, the learned network schemes, including routing schemes and policies, will be assessed in the virtual SDVNs with predicted future state. L. Zhao, G. Han, Z. Li, L. Shu, “Intelligent Digital Twin-based Software-Defined Vehicular Networks”, IEEE Network, 2020 Future Networks IDT-SDVN: opportunities  Learning Algorithm: Although learning algorithms have been well studied in the fields of computer vision and natural language processing, the existing study does not consider the special demand of intelligent algorithms for networking.  Switching from Virtual to Practical: In IDT-SDVN, the learned network schemes should be applied to the real physical environment at a proper time, which can also be known as a tipping point.  Adjusting from Failure: In the IDT side, most verifications are based on the predicted data where prediction cannot be one hundred percent accurate. Hence, even with the best prediction methods, the testing results can be inaccurate inevitably, which may lead to the immature learned scheme employed in the real network.  Validation of Intelligent Strategies: In the IDT side, it is essential to validate the virtual simulation environment before doing any verification. Without validation, the effectiveness verification of networking schemes will be useless in which such verified schemes are uncertain and can be harmful to the network. L. Zhao, G. Han, Z. Li, L. Shu, “Intelligent Digital Twin-based Software-Defined Vehicular Networks”, IEEE Network, 2020 Future Networks IDT-SDVN: Case Study L. Zhao, G. Han, Z. Li, L. Shu, “Intelligent Digital Twin-based Software-Defined Vehicular Networks”, IEEE Network, 2020 Conclusions and Future Work  We introduce the wireless mesh network for extending the urban broadband coverage for less developed areas.  We show the vehicular networking solutions, including VANET, SDVN, for extending the coverage of vehicular communication.  We also show the collaborative multiple UAV including communication, networking, path planning and UAV-assisted VANET.  As the future of networking, IDT-SDVN is also introduced as an exemplary to help us further virtualize the network and world.  For our future work, we will focus on vehicular networking, and also participate in the space-air-ground networking to further enhance and realize the B5G and 6G, to allow the connectivity of everything in the world. Thank You. Q&A