摘要
1.Introduction Emerging networking paradigms,including Information-Centric Networking(ICN)[1],Software-Defined Networking(SDN)[2],Mobile Satellite Communication Networks(MSCN)[3],and Internet of Vehicles(IoV)[4],have faced some severe challenges.For example,the dynamic network environment makes it very hard to optimize resource allocation.In addition,these networking paradigms usually have heterogeneous features,making it difficult to schedule traffic among different kinds of networks.These challenges can be addressed by the adaptive learning of Artificial Intelligence(AI)[5,6]and the edge caching of edge computing.AI can also help establish a relatively optimal routing strategy and perform congestion control by learning the dynamic network state.Just like AI,edge computing[7–10]can help provide fast response to users,and deploy edge servers with strong computing and storage capabilities can greatly improve the performance of 4K/8K and VR/AR.However,despite their ability to improve network performance,there are still many challenges.For example,the integrated architectures and frameworks need to be clearly identified,and the related protocols need to be better defined.
基金
supported by the National Natural Science Foundation of China(Granted No.62202247).
