| With the development and maturity of mobile communication technology,the 5G access network will be more closely integrated with vertical industries,and the range of access terminals will be expanded to different scenarios such as the Internet of Vehicles,Industrial Internet,power network,and smart home.Network slicing technology can provide logical networks with different performance characteristics for differentiated services.This technology has been widely studied due to its advantages of saving network construction costs and simplifying network operation and maintenance.However,the key challenge of deploying access network slices is that,on the one hand,the scarcity of wireless spectrum resources makes the allocation strategy of access network slice resources need to consider the trade-off between resource utilization and isolation between slices,and it is difficult to take into account the complexity and effectiveness of the algorithm.On the other hand,a perfect slice isolation metric can assist resource scheduling decisions when resources are dynamically adjusted.However,there are various measurement methods for the setting of this indicator.There is a lack of unified comparison standards,and it is difficult to comprehensively evaluate the isolation requirements of each slice.In response to the above challenges,this paper conducts an in-depth study.Firstly,this paper studies the optimization strategy of access network slice resource allocation for the problem of inter-slice interference caused by the shared spectrum resources of each base station in the multi-base station network scenario.In order to minimize the interference between slices,this paper designs a heuristic algorithm for resource allocation of access network slices.The main idea of this algorithm is to regard idle resources as objects of resource allocation,and adjust their time-frequency positions.While meeting the resource requirements of each slice,resource blocks at the same time-frequency location in different base stations are allocated to the same access network slice to the greatest extent possible.The simulation verification shows that the access network slice resource allocation algorithm proposed in this paper can effectively reduce the interference degree between each access network slice in the multi-base station network scenario with low time cost.Secondly,based on the extensive requirements of access network slicing in terms of independence,reliability and service performance assurance,this paper considers the design of isolation metrics for access network slicing service quality assurance.In the first step,this paper considers the independent relationship between the physical resource allocation results and the operation of slice services,so as to design parameters to quantify the degree of physical isolation of slices.In the second step,this paper considers the impact of resource sharing of each slice on the quality of slice services under the background of periodic resource dynamic adjustment,so as to design parameters to quantify the degree of slice service isolation.Then,this paper uses the normalization function to normalize the above two quantitative parameters and combine them into a single form of isolated quantitative index.The simulation data verifies that the slice isolation metric proposed in this paper can correctly reflect the isolation level of slices.It also verifies that the resource allocation strategy for access network slices based on the minimization of inter-slice interference proposed in this paper can maintain the stability of the isolation of each slice in the context of rapid resource dynamic adjustment.In summary,this paper focuses on the optimization of resource allocation for access network slices and the design of isolation metrics for slice quality assurance.It provides ideas for solving the trade-off between resource utilization and slice service guarantee,and contributes to the design of indicators that comprehensively quantify the degree of slice isolation. |
PDF Full Text Request