This code is for the following paper:
H. He, C. Wen, S. Jin, and G. Y. Li, “Deep learning-based channel estimation for beamspace mmwave massive MIMO systems,” IEEE Wireless Commun. Lett., vol. 7, no. 5, pp. 852–855, Oct. 2018.
Please cite the paper when use this code.
-
Install Matcovnet
Choose the gpu version and verify you have a supported GPU and that the latest driver is installed.
-
Run main program
Location: ..\LDAMP_for_Rice\D-AMP_Toolbox-master\SCAMPI-MATLAB\scampi-vs-ssd\
main_new_old.m: main function, run and call trained network and denoiser to produce simulation results
test.m: Generate the Saleh-Valenzuela channel model
Trained network is saved at ..\Code_WCL_2018\LDAMP_for_Rice\D-AMP_Toolbox-master\Packages\DnCNN\BestNets_20
- Trainin DnCNN network
Location: ..\LDAMP_for_Rice\D-AMP_Toolbox-master\Packages\DnCNN\Training
Demo_Train.m: main function for training the DnCNN
Demo_test: main fuction for test the DnCNN
Training_h.mat: Training data
Vali_h.mat: Validata and Test data
rescaleImage.m: rescale the channel into [0,1]
channel_gen.m: generate the training and test data
Many trained network will be saved at ..\LDAMP_for_Rice\D-AMP_Toolbox-master\Packages\DnCNN\Training\NewNetworks. Copy the best network with different SNR and rename to ..\LDAMP_for_Rice\D-AMP_Toolbox-master\Packages\DnCNN\BestNets_20.
- LDAMP network
Location: ..\LDAMP_for_Rice\D-AMP_Toolbox-master\SCAMPI-MATLAB and ..\LDAMP_for_Rice\D-AMP_Toolbox-master\Algorithms
DAMP_SNR1.m: D-AMP algorithm
Acknoledge:
Many thanks for Dr.Christopher A. Metzler share the code selflessly. Instructions about installing the matcovnet and using LDAMP network can refer to
website: https://github.com/ricedsp/D-AMP_Toolbox
Questions/suggestions/comments about LDAMP-based Channel estimation network: hehengtao@seu.edu.cn