于(yu)全、Sherman Shen、Nei Kato以及Wei Zhang四位专家(jia)学(xue)术报告通知
ABSTRACT:Space information network is proposed based on national strategic planning andthe present development status of associated technologies. In the network, alow earth orbit (LEO) constellation communication system is proposed for thecollection and distribution of the wide-area information. Besides, thesatellite system can be employed for the information monitoring of maritimebuoy networks. In this talk, we will give some considerations for the new LEOsystem and the maritime buoy network. Furthermore, key technologies that needto be solved and the core system functions are presented. Finally, we will alsodiscuss the future of civil-military integration.
报告(gao)二：Automated Driving and Connected Vehicles for Modern Transportation Systems
——加拿大(da)工程院(yuan)院(yuan)士、IEEE ComSoc副主(zhu)席、IEEE IoT 期刊主编、IEEE Fellow。
ABSTRACT:Modern society depends on faster, safer, and environment friendlytransportation system. Vehicular communications network in terms of vehicle tovehicle, vehicle to infrastructure, vehicle to pedestrian, vehicle to cloud,and vehicle to sensor, can provide a solution to modern transportation system.In this talk, we first introduce the all connected vehicles. We then presentthe applications, challenges and scientific research issues of vehicularcommunications network. We also explain the role of vehicular networking in theautomated driving era. We conclude the talk by discuss the futureSpace-Air-Ground (SAG) Integration.
报告三：OnRemoving Routing Protocol from Future Wireless Networks: A Real-time DeepLearning Approach for Intelligent Traffic Control
——IEEE ComSoc副主席、IEEE Network期刊主编、IEEETVT期(qi)刊主编、IEEE Fellow。
ABSTRACT:Recently, deep learning, an emergingmachine learning technique, is garnering a lot of research attention in severalcomputer science areas. However, to the best of our knowledge, its applicationto improve heterogeneous network traffic control which is an important andchallenging area for IoT by its own merit has yet to appear because of thedifficult challenge in characterizing the appropriate input and output patternsfor a deep learning system to correctly reflect the highly dynamic nature oflarge-scale heterogeneous networks. In this talk, an appropriate input andoutput characterizations of heterogeneous network traffic will be introducedand a supervised deep neural network system will be proposed. I will describehow our proposed system works and how it differs from traditional neuralnetworks. Also, preliminary results will be discussed and I will demonstratethe encouraging performance of our proposed deep learning system compared to abenchmark routing strategy (Open Shortest Path First (OSPF)) in terms ofsignificantly better signaling overhead, throughput, and delay. In addition,can a new intelligent traffic control system be designed without any benchmarktraining data, and can learn by itself to replace existing non-intelligentrouting protocols? I will also address this issue by some of our preliminaryresults and look toward the future.
报告四：TheDrones Are Coming!
ABSTRACT：Thereare expected to be millions of unmanned aerial vehicles or drones in the worldby 2020. Apart from being used for delivering package, Drones have been used tosave lives, monitor animals, protect environment, survey the trafficconditions, and help journalists tell the headline events, etc. As flyingrobots, drones complete the tasks by remote controllers. Thus, wirelesscommunications for drones is crucial. This talk will discuss research challengesand opportunities for wireless communications for drones, including spectrumsharing, channels, networking, safety and privacy issues.