Artificial intelligence-based network slice management system and method on the ran side

Abstract

The invention discloses a RAN side network slice management system and method based on artificial intelligence. The system includes an AI slice algorithm platform and several base stations; the AI ​​slice algorithm platform provides AI slice algorithms and initiates and terminates AI slice functions. The module implements the interaction between the base station and the AI ​​slicing algorithm platform, as well as the collection and reporting of base station status information. The AI ​​slicing algorithm implements the network resource slice allocation scheme, and the base station performs AI slice-based network resource scheduling. The present invention realizes AI-based RAN side network slicing without adding existing base station hardware equipment, supports multiple artificial intelligence algorithms and is also compatible with traditional old base stations that do not support AI slicing, which is helpful for rapid network deployment and improvement. Maintenance; at the same time, the AI ​​slicing algorithm platform is connected to multiple base stations, which can jointly manage and optimize the wireless resources of multiple base stations, and improve the overall utilization of network resources and user experience on the basis of making full use of artificial intelligence algorithms.

Description

technical field [0001] The present invention relates to the technical field of wireless communication, in particular to an artificial intelligence-based RAN side network slice management system and method. Background technique [0002] With the rapid advancement of 5G commercial use, application scenarios such as high-definition video, intelligent security, and autonomous driving are constantly emerging, and the requirements for networks are becoming more differentiated. 5G business scenarios are mainly divided into three categories: enhanced mobile broadband (eMBB), massive machine-to-machine communication (mMTC), and ultra-reliable and ultra-low-latency communication (uRLLC). Among them, the eMBB scenario is mainly for high-speed requirements, such as real-time video conferencing, high-speed download, etc.; mMTC is for large-scale connected device scenarios, such as the networking requirements of a large number of IoT devices in industrial scenarios; uRLLC scenario is for ...

AI Technical Summary

Problems solved by technology

However, the current artificial intelligence-based network slicing method is difficult to implement on real wireless access network base station equipment products

The main reasons are as follows: 1) The network slicing method based on artificial intelligence does not consider the actual architecture and software and hardware systems of wireless access network base station equipment

Current base station equipment often uses dedicated hardware systems, and does not have AI acceleration hardware modules that artificial intelligence algorithms highly rely on

At the same time, due to cost and power consumption constraints, the software and hardware systems of base station equipment do not have additional computing power to support artificial intelligence algorithms

At this stage, the energy consumption of base stations has brought huge economic and social responsibility pressure on operators. In the future, it will be difficult to increase energy consumption by expanding hardware equipment in exchange for additional computing power.

2) The existing artificial intelligence network slicing algorithm only targets the network resources of a single base station, and does not consider the joint management and optimization of the network resources of multiple base stations

Images