Research And Implementation Of Network Slicing System For 5G Industrial Internet Of Things

With the development of industrial manufacturing to digitalization,informatization and intelligence,the industrial Internet of Things will face increasingly complex application scenarios,such as kerosene exploration,aviation control,factory intelligence,intelligent security,real-time monitoring,etc.This not only requires the industrial Internet of Things technology to support the universal interconnection of million-level intelligent terminal devices,but also requires the network transmission bandwidth,delay,jitter,packet loss and other fine-grained performance,as well as the network security and other endogenous requirements to be constantly improved.In recent years,the fifth generation mobile communication technology(5G),with its new transmission characteristics of "large bandwidth,low delay and wide connection",has gradually become a new solution to solve the bottleneck of the development of industrial Internet of Things.Firstly,the spectrum range of 5G includes FR1(450 MHz-6GHz)and FR2(24.25 GHz-52.6 GHz).This provides a favorable prerequisite for alleviating the fierce competition of frequency band resources in the industrial Internet of Things scenario.Secondly,5G introduces network slicing technology and is committed to dividing a single physical network into multiple logical networks to meet the performance requirements of different business scenarios.Although the 5G network slicing technology has a broad application prospect,this thesis finds that the current 5G network slicing technology has problems such as low resource sharing rate between slices and low user delay requirements.Therefore,this thesis studies the allocation of 5G network slice frequency band resources in the industrial Internet of Things scenario,and completes the following work:(1)We design a differentiated slicing resource division algorithm based on over-booking to meet the long-term slicing resource demand and improve the resource sharing rate between slices in the short term;(2)We propose an end-to-end user resource allocation algorithm with fair proportion to introduce application layer information into the process of resource allocation,so as to ensure low delay transmission of user data;(3)Based on the open-source network slicing framework,we deploy a network slicing system for 5G industrial Internet of Things and implement the above algorithms in the system.In this thesis,the above algorithms are tested in the above systems.The experimental results show that compared with the traditional network slicing scheme,the proposed scheme reduces the user delay by 43.98%and increases the total bandwidth by 10.85%,which greatly meets the requirements of 5G network slicing system in industrial Internet of Things scenarios.