Publication: A Generalized Channel Dataset Generator for 5G New Radio Systems Based on Ray-Tracing 

Authors:
Zhang, Y. ,  Sun, J. ,  Gui, G. ,  Gacanin, H. ,  Sari, H.
Journal:
IEEE Wireless Communications Letters
Volume:
10
Page(s):
2402-2406
number:
11
Date:
Nov. 2021
DOI:
10.1109/LWC.2021.3101908
hsb:
RWTH-2021-10450
Language:
English

Abstract

Deep learning is considered one of promising tools to develop intelligent wireless techniques in the fifth-generation (5G) wireless communication systems. However, existing researches are conducted based on the channel datasets in fourth-generation (4G) wireless communications systems. Also, some 5G nonstandard channel dataset generators are proposed for frontier technology research. However, these datasets cannot be applied in real 5G new radio (NR) systems. In this letter, we propose a generalized channel dataset generator for 5G NR systems. Furthermore, this letter also proposes a data sampling scheme called RB replacement, which improves the resolution of the dataset and greatly reduces the size of the dataset. The dataset generator can set different channel parameters according to different needs of users, and also can generate massive multiple input multiple output (MIMO) channel. The data generator is open source at GitHub, 1 which can be downloaded and used by researchers for free. 1

The code of this letter can be downloaded from GitHub link:https://github.com/CodeDwan/5G-NR-data-generato.git.

Download

BibTeX