Development of AI/ML Methods for Advanced Device Localization in Beyond 5G Systems

Abstract

This master thesis aims to investigate the positioning accuracy improvements of device localization by the implementation of AI/ML functionalities. In this project, we use the fifth-generation(5G) New Radio (NR) system and focus on indoor factories. This
setup held under study is a relevant scenario for the Industrial Internet of Things within the industrial sector (IIoT), as the specific characteristics of this scenario create a disadvantageous environment for effective positioning. It is a typical deployment
scenario standardized in Third Generation Partnership Project (3GPP) releases.

The study is based on using a provided MATLAB simulator for generating the required information about the User Equipment (UE) locations and their channel measurements with the 5G base stations, the gNodeBs (gNBs). This simulator creates the channel model, environment geometry, and position reference signals aligned with the 3GPP. Pursuing the goal of mitigating the multipath propagation effects, different AI/ML methods have been developed in Python. Several AI/ML models have been explored investigating different inputs, such as the Channel Impulse Response or other significant channel features, as well as various model outputs, such as the direct UE position or intermediate angles or times of the radio signals.

Consequently, this project evaluates the positioning performance of the assisted and direct AI/ML positioning versus the legacy methods in terms of accuracy and complexity while considering different deployment strategies. Different scenario configurations have been simulated regarding the generalization ability of the AI/ML methods evaluation.

Furthermore, another objective has been studying the actual viability of AI/ML for 5G device positioning and in that case, which direction is more worthwhile for future investigation. Finally, based on the results of this simulation-based evaluation, it has been demonstrated that deploying AI/ML methods on the UE side is advantageous for improving the existing 5G location services in this particular scenario without requiring an excessive computational capacity.

This thesis project has investigated several models of different natures and complexity, comparing their performance. Besides, the best generated AI/ML models show a general performance improvement versus the legacy methods from 80% of the distance error CDF. In the most severe Non-Line-of-Sight (NLOS) scenarios, the AI/ML methods have achieved an improvement of more than 10 meters for the 95% CDF compared with the legacy. To conclude, the AI/ML models achieve greater performance for most of the devices than legacy, also offering great results for heavily NLOS situations.

Popular Abstract

Everyone has found themselves lost in the city at some point, having had to use a map app from their smartphone to locate themselves. Now try to recall the same situation, but you are lost inside a building and tried to use the same app. It was telling you that you were outside the building when you were on the second floor of a mall.

You may ask yourself, how can we solve this? One possible solution for the indoor positioning problem is very popular in modern days, Artificial Intelligence (AI) and Machine Learning (ML). New communication technologies need to be followed by new methods that fix this problem. For 5G and beyond technologies, it is being studied the implementation of AI for enhancing different functionalities, one of them being device localization.

The AI methods are not referred to as a robot that can talk to us but more of an intelligent model installed in the smart devices you are using. This intelligent algorithm can be trained to predict the exact location of the device. The AI/ML models have the ability to learn from the environment, in this case from the network signals. Using this promising technology this project aims to help and improve the nowadays network positioning systems, to get better results in some scenarios.

This master thesis ”Development of AI/ML Methods for Advanced Device Localization in Beyond 5G Systems” aims to explain the research done for solving the positioning problem, describing the steps taken to achieve the goals. By the end of this report, the goal is that the reader has more understanding of how device localization is currently done in 5G, acquire more knowledge about AI/ML, and how it can be implemented to improve the traditional location methods. Apart from illustrating these topics, the results of the developed AI/ML are shown, giving promising possibilities to further develop the explored AI/ML approach. Finally, some next steps are proposed for future researchers to have a starting point for improving the investigated solution.

• author: Costa Andrés, Eros Iván and Aizkorreta Carro, Jon
• supervisor: Fredrik Tufvesson
• organization: Department of Electrical and Information Technology
• course: EITM01 20231
• year: 2023
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: Artificial Intelligence, localization, IIoT devices, Python, radio signals, 5G, 3GPP
• report number: LU/LTH-EIT 2023-940
• language: English
• id: 9134603
• date added to LUP: 2023-08-29 10:42:50
• date last changed: 2023-08-29 10:42:50