Open Access

Dynamic channel model estimation based on gradient descent method and its optimization in massive MIMO

Jinhui Chen

Jinhui Chen

College of Information and Communication Engineering, Beijing Information Science and Technology UniversityBeijing, China
Search for this author on
Sciendo|Google Scholar
Chen, Jinhui
, 
Zhan Xu

Zhan Xu

College of Information and Communication Engineering, Beijing Information Science and Technology UniversityBeijing, China
Search for this author on
Sciendo|Google Scholar
Xu, Zhan
 and 
Ruxin Zhi

Ruxin Zhi

College of Information and Communication Engineering, Beijing Information Science and Technology UniversityBeijing, China
Search for this author on
Sciendo|Google Scholar
Zhi, Ruxin
  
Mar 21, 2025
Dynamic channel model estimation based on gradient descent method and its optimization in massive MIMO's Cover Image
About this article
Previous Article
Next Article
Cite
Download Cover
Published Online: Mar 21, 2025
Received: Oct 26, 2024
Accepted: Feb 10, 2025
DOI: https://doi.org/10.2478/amns-2025-0653
Keywords
© 2025 Jinhui Chen et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1.

Schematic diagram of time-varying impulse response of dynamic channel
Schematic diagram of time-varying impulse response of dynamic channel

Figure 2.

Block diagram of multi-channel dynamic channel simulation platform
Block diagram of multi-channel dynamic channel simulation platform

Figure 3.

Flowchart of user parameter configuration unit
Flowchart of user parameter configuration unit

Figure 4.

Frame structure of channel parameters
Frame structure of channel parameters

Figure 5.

Block diagram of a massive MIMO system
Block diagram of a massive MIMO system

Figure 6.

Test area in LOS scenario
Test area in LOS scenario

Figure 7.

NLOS scenario
NLOS scenario

Figure 8.

Beamforming training curve of SVGD model
Beamforming training curve of SVGD model

Figure 9.

The beam obtained by training SVGD model in LOS scene
The beam obtained by training SVGD model in LOS scene

Figure 10.

Min-Sum·0bjective function simulation effect of LOS scene
Min-Sum·0bjective function simulation effect of LOS scene

Figure 11.

NLO scenario simulation results using Min-Max objective function
NLO scenario simulation results using Min-Max objective function

Figure 12.

Schematic diagram of virtual antenna pair selection in spherical anechoic darkroom
Schematic diagram of virtual antenna pair selection in spherical anechoic darkroom

Figure 13.

Monte Carlo simulation results of reference algorithm and SVGD algorithm
Monte Carlo simulation results of reference algorithm and SVGD algorithm

Network parameters of SVGD model

Argument Numerical value
Learning rate 0.001
Weight attenuation 0.02
Playback buffer size 10000
Lot size 1036

Simulation parameters of NLOS scenarios

Argument Numerical value
Carrier frequency 30 GHz
System bandwidth 0.5 GHz
Number of subcarriers 1
Multipath quantity 5
Base station number 2
Base station antenna array (1,32,2)
Antenna spacing 0.5
Phase shifter resolution 5 bit
User area number R551-R650, R651-R750, R751-R850

Simulation parameters of LOS scenarios

Argument Numerical value
Carrier frequency 50 GHz
System bandwidth 0.5 GHz
Number of subcarriers 1
Multipath quantity 10
Base station number 1
Base station antenna array (1,64,32)
Antenna spacing 0.5
Phase shifter resolution 5 bit
User area number R551-R654, R655-R789, R790-R920
Language:
English
Publication timeframe:
1 times per year
Journal Subjects:
Life Sciences, Life Sciences, other, Mathematics, Applied Mathematics, General Mathematics, Physics, Physics, other
Journal RSS Feed