Skip to content

Dataset for "Downlink Transmit Design for Massive MIMO LEO Satellite Communications"

Notifications You must be signed in to change notification settings

likexin1415/Dataset_LEO_DL_transmit_design

Repository files navigation

Guide

The dataset consists of $200000$ samples. The number of users $K$ is given by $K=100$.

Please remember that the rows are for different samples, while the columns are for different users.

The input for the neural network has $3K+1=301$ parameters. The dataset for the input is given by:

  1. Data_xsort_beta_nvar_Tx_12x12_Rx_6x6_NumUser_100

​ Dimension: (200000,100) Variable: $ \tilde{\beta}_k $

  1. Data_xsort_tdtheta_x_Tx_12x12_Rx_6x6_NumUser_100

    Dimension: (200000,100) Variable: $ \tilde{\theta}_k^{\mathrm{x}} $

  2. Data_xsort_tdtheta_y_Tx_12x12_Rx_6x6_NumUser_100

    Dimension: (200000,100) Variable: $ \tilde{\theta}_k^{\mathrm{y}} $

  3. Data_xsort_power_Tx_12x12_Rx_6x6_NumUser_100

    Dimension: (200000,1) Variable: $ P $

The output for the neural network has $K=100$ parameters. The dataset for the output is given by:

​ Data_xsort_lambda_Tx_12x12_Rx_6x6_NumUser_100

​ Dimension: (200000,100) Variable: $ \tilde{\lambda}_k^{\mathrm{opt}} $

About

Dataset for "Downlink Transmit Design for Massive MIMO LEO Satellite Communications"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published