Online self-adaptive virtual resource allocation method for network slices based on service prediction

Abstract

The invention relates to an online self-adaptive virtual resource allocation algorithm based on service prediction and belongs to the field of mobile communication. The algorithm is online dynamic resource scheduling carried out by taking the minimum network slice resource mapping cost as a target on the premise of guaranteeing QoS of virtual network slices. The service of each network slice is predicted on a large time scale through a self-regression sliding average model, and the optimal buffer space size which is reserved in the next period is selected according to the predicted maximum probability load value; and the priority of each network slice is adjusted on a small time scale according to a queue overflow probability or a residual time value of each network slice, and appropriatephysical resource blocks are allocated to each network slice to ensure the service rate of each network slice on the premise of guaranteeing the queue overflow probability of each network slice. According to the online dynamic self-adaptive virtual resource allocation method provided by the invention, the QoS of each network slice can be effectively guaranteed, and meanwhile, the resource mappingcost of the network slices is minimized.

Description

technical field [0001] The invention belongs to the field of mobile communication, and relates to an online self-adaptive virtual resource allocation method based on service prediction. Background technique [0002] With the rapid increase in the number of mobile users and the widespread popularization of smart terminals, next-generation network technologies need to provide higher system capacity, spectrum utilization, and lower latency. Therefore, wireless network virtualization technology has emerged as the times require. In a wireless network virtualization scenario, traditional operators are decoupled into two independent roles: an infrastructure provider (Infrastructure Provider, InP) and a service provider (ServiceProvider, SP). InP abstracts and slices physical resources in the network (including spectrum, power, antenna, backhaul, etc.) to provide SPs with virtual networks with different requirements; SPs can lease physical resources from different InPs to provide us...

AI Technical Summary

Problems solved by technology

In addition, the existing research mainly focuses on static resource allocation strategies or real-time resource scheduling based on the network status of the current scheduling period. However, due to sudden changes in network load, it is very likely to cause network overload or resource oversupply. Therefore, there is a need for a resource allocation strategy that proactively and dynamically adjusts resources according to the changing characteristics of the load.

Images