Research On Mobility Management In Ultra-dense Edge Computing

The widespread use of various smart mobile devices and Internet of Things devices has created new access and transmission requirements for wireless network.The research of ultradense network is carried in this context,which can achieve thousandfold improvement of network capacity in the hot spots by intensive deployment of wireless access points.The emergence of delay-sensitive,computing-intensive mobile applications brought by the development of smart devices have brought high computing requirements to the edge network environment.Mobile edge computing has become a viable solution to these explosively growing computing requirements.The integration of ultra-dense network and mobile edge computing,so-called ultra-dense edge computing,is considered to be the most promising solution for the emerging explosive growth requirements of communication and computation,which allows mobile devices to run demanding applications with low latency by densely deploying communication and computing resources near users.However,due to the mobility of edge users,when the users move away from the current service wireless access point and edge server,the service experience may be declined or the performing service may be interrupted.Therefore,mobility management becomes one of the bottlenecks in the implementation of ultra-dense edge computing network architecture.At present,most existing mobility management works in ultra-dense edge computing considered that wireless handover and service migration should be performed simultaneously,and ignored the difference between wireless handover and service migration.In addition,there are a few works focused on mobility management issues in ultra-dense edge computing with multi-point collaboration enabled.In view of the shortcomings of the existing works,this paper first proposes a mobility management scheme that considers both wireless handover and service migration optimization.Then a mobility management scheme for ultra-dense edge computing with multi-point collaboration technology enabled is proposed.The main research results of this paper are shown below.(1)Firstly,this paper researches the mobility management in ultra-dense edge computing considering the wireless handover and service migration,and proposes a mobility management scheme for this scenario based on deep reinforcement learning.The proposed scheme makes decisions focusing on the user's service experience and service migration cost,and determines the appropriate wireless access points and edge servers for the user with an online and dynamic way during the user's movement to ensure the reliability of the offloading service.(2)Secondly,this paper researches the mobility management issue in ultra-dense edge computing with multi-point collaboration technology and proposes a mobility aware mobility management scheme applying to this scenario.The proposed scheme estimates the future performance of access points and edge servers according to the user's future movement trend range,and determines the suitable access points and edge servers as the collaboration set to provide reliable communication and offloading services.(3)In addition,for the two proposed mobility management schemes,we separately set up simulation verification environments and compared them with other available schemes.The results show that the proposed schemes in this paper can provide reliable communication and computing offloading services for users during mobility,while reducing the signal cost effectively.