Uneingeschränkter Zugang

Research on topology control method of multi-hop wireless communication network based on graph neural network

  
21. März 2025

Zitieren
COVER HERUNTERLADEN

Figure 1.

Cellular modeling
Cellular modeling

Figure 2.

Convergence curve
Convergence curve

Figure 3.

Convergence result
Convergence result

Figure 4.

The performance of GCNTO varies with the number of devices
The performance of GCNTO varies with the number of devices

Figure 5.

Test results for offline models
Test results for offline models

Figure 6.

Sum rate of the interference channel of 25 users in the 5 community
Sum rate of the interference channel of 25 users in the 5 community

Figure 7.

Sum rate of the interference channel of 24 users in the 6 community
Sum rate of the interference channel of 24 users in the 6 community

Equal weight and Random weight

Index 10 log10 σ2 WMMSE IWFA MLP GCN GCNTO
Equal weight 10dB 30.747 26.578 21.069 28.362 29.052
5dB 29.856 25.843 19.991 27.619 28.292
0dB 28.012 24.375 18.222 26.174 26.72
-5dB 24.814 21.828 15.754 23.609 23.975
-10dB 20.313 18.192 12.729 19.814 19.974
Random weight 10dB 17.354 12.671 11.914 15.504 16.585
5dB 16.81 12.308 11.401 15.019 16.022
0dB 15.711 11.59 10.518 14.083 14.967
-5dB 13.815 10.371 9.211 12.536 13.294
-10dB 11.209 8.636 7.508 10.328 10.754

Hardware environment

Central processing 12th Gen Intel(R)Core(TM)i7-12700H
CPU main frequency 2.70GHz
Memory 32GB
Operating system Windows11
Simulation software PyCharm 2022.2.2
Sprache:
Englisch
Zeitrahmen der Veröffentlichung:
1 Hefte pro Jahr
Fachgebiete der Zeitschrift:
Biologie, Biologie, andere, Mathematik, Angewandte Mathematik, Mathematik, Allgemeines, Physik, Physik, andere