?
The Increasing of Resource Sharing Efficiency in Network Slicing Implementation
Network slicing has wide range of applications in resource sharing especially for creating differentiated servicing scenarios for heterogeneous traffic environments supported by 5G networks. Despite of such potential features there are numerous problems in implementation of network slicing and hence present ETSI standards only give definitions and key principles of corresponding architecture. Such situations signifies the importance of mathematical modeling of network slicing procedures to develop efficient instruments for their implementation. An analytical framework to model the resource allocation procedures for transmission of multiservice traffic has been constructed and analyzed. The model consists of arbitrary number of traffic streams originating from different types of real time applications. All random variables used in the model have exponential distribution with corresponding mean values. Two different scenario of resource sharing has been taken into consideration for incoming traffic streams: Network Slicing, when resources are strictly divided among incoming traffic streams, and Filtering, when the access to resource is restricted depending on the amount of resource occupied by all traffic streams. It has been shown how to use both scenarios for creating conditions for differentiated servicing of heterogeneous traffic. Given numerical assessment proves that scenario based on Filtering is more efficient to solve the formulated task in contrast to the analogous scenario based on slicing.
