Self-organised admission control for multi-tenant 5G networks
The vision of the future 5G corresponds to a highly heterogeneous network at different levels, including multiple Radio Access Technologies (RATs), multiple cell layers, multiple spectrum bands, multiple types of devices and services, etc. Consequently, the overall RAN planning and optimization proc...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Fecha de publicación: | 2017 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/114740 |
| Acceso en línea: | https://hdl.handle.net/2117/114740 |
| Access Level: | acceso abierto |
| Palabra clave: | Reinforcement Learning Artificial intelligence Self-organizing systems Self-Organizing Network 5G Fuzzy Q-Learning Aprenentatge per reforç Intel·ligència artificial Sistemes autoorganitzatius Àrees temàtiques de la UPC::Enginyeria de la telecomunicació |
| Sumario: | The vision of the future 5G corresponds to a highly heterogeneous network at different levels, including multiple Radio Access Technologies (RATs), multiple cell layers, multiple spectrum bands, multiple types of devices and services, etc. Consequently, the overall RAN planning and optimization processes that constitute a key point for the success of the 5G concept will exhibit tremendous complexity. In this direction, legacy systems such as 2G/3G/4G already started the path towards a higher degree of automation in the planning and optimization processes through the introduction of SON functionalities. SON refers to a set of features and capabilities designed to reduce or remove the need for manual activities in the lifecycle of the network. With the introduction of SON, classical manual planning, deployment, optimization and maintenance activities of the network can be replaced and/or supported by more autonomous and automated processes, operating costs can be reduced and human errors minimized. In this work, a self-organizing admission control algorithm for multi-tenant 5G networks is proposed and developed with novel artificial intelligence techniques. A simulation-based analysis is presented to assess the improvements of the proposed approach with respect to a baseline scheme. |
|---|