Power_Control_GNN

These codes are for results presented in the paper - Power Control for 6G Industrial Wireless Subnetworks: A Graph Neural Network Approach https://arxiv.org/abs/2212.14051

Bibtex:

@INPROCEEDINGS{10118984,
  author={Abode, Daniel and Adeogun, Ramoni and Berardinelli, Gilberto},
  booktitle={2023 IEEE Wireless Communications and Networking Conference (WCNC)}, 
  title={Power Control for 6G Industrial Wireless Subnetworks: A Graph Neural Network Approach}, 
  year={2023},
  volume={},
  number={},
  pages={1-6},
  doi={10.1109/WCNC55385.2023.10118984}}

About the code

How to use the code

Download all the files in the same folder

The file main.py contains the run_experiment(f_metric, train_num_subn, test_num_subn, trainsh_sd, testsh_sd, device) function. Run this function and call the function in a console with appropriate arguments, or import main in a python file and run main.run_experiment(f_metric, train_num_subn, test_num_subn, trainsh_sd, testsh_sd, device). See appropriate options for the function arguments below.

f_metric options - hH, dD, hD :The type of graph attribute; hH - use full channel gain, dD - use only distance information, hD - use desired link channel gain and interfering link distances

train_num_subn - number of subnetworks in training deployment - choose between values 20, 25, 10, for larger values, there is a need to increase self.deploy_length

test_num_subn - number of subnetworks in testing deployments - choose between values 20, 25, 10, for larger values, there is a need to increase self.deploy_length

trainsh_sd - the shadowing standard deviation of the training deployments - reasonable values between 4 and 10

testsh_sd - the shadowing standard deviation of the testing environment - reasonable values between 4 and 10

device - Choose appropriate device - 'cuda', 'cpu'

An example:

run_experiment('hH',20,10,7,4,'cuda')

The results are returned as plots with appropriate titles.

Dependencies

Pytorch, pytorch_geometric, numpy, matplotlib, Sum_rate_power_allocator, subnetwork_generator

You can view the evaluation for the WMMSE in the WMMSE_eval.ipynb