A Deep Learning-Based End-to-End Algorithm for 5G Positioning

Fifth generation (5G) new radio (NR) signals present unique opportunities for users and devices localization due to their higher bandwidth and more antenna elements. We propose an efficient deep learning-based end-to-end algorithm for 5G or beyond 5G (B5G) positioning. Both the channel geometric par...

Full description

Saved in:
Bibliographic Details
Published in:IEEE sensors letters 2022-04, Vol.6 (4), p.1-4
Main Authors: Lv, Ning, Wen, Fuxi, Chen, Yanping, Wang, Zhongmin
Format: Article
Language:English
Subjects:
5G positioning
Algorithms
Azimuth
complex impulse response
Complexity
Computer architecture
Convolution
Deep learning
Feature extraction
Impulse response
Location awareness
Machine learning
multifeature fusion
new radio
Numerical models
Radio signals
Sensor networks
Wireless communication
Wireless sensor networks
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Fifth generation (5G) new radio (NR) signals present unique opportunities for users and devices localization due to their higher bandwidth and more antenna elements. We propose an efficient deep learning-based end-to-end algorithm for 5G or beyond 5G (B5G) positioning. Both the channel geometric parameters and complex impulse response are utilized to exploit the connection between propagation channel and environment. A new multifeature fusion network architecture is proposed. Simulation results and complexity analysis are provided to demonstrate the efficacy of the proposed algorithm.
ISSN:2475-1472
2475-1472
DOI:10.1109/LSENS.2022.3148910