# masadipradeeprao/deep_learning_mimo architecture

> Architecture diagram of the GitHub repository masadipradeeprao/deep_learning_mimo, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-01). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/masadipradeeprao/deep_learning_mimo
- Repository: https://github.com/masadipradeeprao/deep_learning_mimo

## Overview

The repository combines MATLAB-based DeepMIMO data preparation with a Python beam-selection learning workflow. The documented path is to obtain a DeepMIMO scenario, generate model input/output data, train per-transmitter models, and save predictions and corresponding optimal beams. The tree also contains codebook-generation and hybrid-learning assets, but their runtime connections are not shown. MATLAB source was not excerpted, so those relationships are omitted rather than inferred.

Main entry points: [`Deep_learning_MIMO/DL_model_python.py`](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/DL_model_python.py) (Training and evaluation), [`Deep_learning_MIMO/Generate_DL_data.m`](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/Generate_DL_data.m) (DL data generation)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_data["Dataset preparation"]
  node_matlab_entry["Dataset generator"]
  node_mimo_generator["DeepMIMO generator"]
  node_raytracing["Ray-tracing reader<br/>[read_raytracing.m]"]
  node_channel["Channel construction"]
  node_td_channel["Time-domain channel"]
  node_users["User selection<br/>[find_users.m]"]
  node_parameters["Generator parameters"]
  node_dl_data["DL data generation<br/>[Generate_DL_data.m]"]
end

subgraph group_beam["Beamforming assets"]
  node_codebook["UPA codebook"]
  node_cbf_parameters["Codebook parameters<br/>[CBF_parameters.m]"]
  node_hybrid_notebook["Hybrid learning"]
end

subgraph group_learn["Learning workflow"]
  node_mat_files[("Input/output data")]
  node_python_runner["Training and evaluation<br/>[DL_model_python.py]"]
  node_per_bs_models["Per-transmitter models<br/>[DL_model_python.py]"]
end

subgraph group_results["Evaluation outputs"]
  node_predictions["Beam predictions"]
end

node_researcher(("Researcher"))
node_scenario[("DeepMIMO scenario")]

node_researcher -->|"downloads"| node_scenario
node_researcher -->|"runs"| node_dl_data
node_scenario -.->|"provides dataset"| node_dl_data
node_dl_data -->|"generates"| node_mat_files
node_mat_files -->|"loads"| node_python_runner
node_python_runner -->|"trains"| node_per_bs_models
node_python_runner -->|"saves results"| node_predictions
node_per_bs_models -->|"predicts beams"| node_predictions

click node_matlab_entry "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_Dataset_Generator.m"
click node_mimo_generator "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/DeepMIMO_generator.m"
click node_raytracing "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/read_raytracing.m"
click node_channel "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/construct_DeepMIMO_channel.m"
click node_td_channel "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/construct_DeepMIMO_channel_TD.m"
click node_users "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/find_users.m"
click node_parameters "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/validate_parameters.m"
click node_dl_data "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/Generate_DL_data.m"
click node_codebook "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/BF_codebook/UPA_codebook_generator.m"
click node_cbf_parameters "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/CBF_parameters.m"
click node_hybrid_notebook "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/Hybrid_Deep_learning.ipynb"
click node_python_runner "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/DL_model_python.py"
click node_per_bs_models "https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/DL_model_python.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_matlab_entry,node_mimo_generator,node_raytracing,node_channel,node_td_channel,node_users,node_parameters,node_dl_data toneBlue
class node_codebook,node_cbf_parameters,node_hybrid_notebook,node_scenario toneAmber
class node_mat_files,node_python_runner,node_per_bs_models toneMint
class node_predictions toneRose
class node_researcher toneIndigo
```

## Components

### Dataset preparation

- [Dataset generator](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_Dataset_Generator.m): `DeepMIMOv2/DeepMIMO_Dataset_Generator.m`
- [DeepMIMO generator](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/DeepMIMO_generator.m): `DeepMIMOv2/DeepMIMO_functions/DeepMIMO_generator.m`
- [Ray-tracing reader](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/read_raytracing.m): `DeepMIMOv2/DeepMIMO_functions/read_raytracing.m`
- [Channel construction](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/construct_DeepMIMO_channel.m): `DeepMIMOv2/DeepMIMO_functions/construct_DeepMIMO_channel.m`
- [Time-domain channel](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/construct_DeepMIMO_channel_TD.m): `DeepMIMOv2/DeepMIMO_functions/construct_DeepMIMO_channel_TD.m`
- [User selection](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/find_users.m): `DeepMIMOv2/DeepMIMO_functions/find_users.m`
- [Generator parameters](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/DeepMIMOv2/DeepMIMO_functions/validate_parameters.m): `DeepMIMOv2/DeepMIMO_functions/validate_parameters.m`
- [DL data generation](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/Generate_DL_data.m): `Deep_learning_MIMO/Generate_DL_data.m`

### Beamforming assets

- [UPA codebook](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/BF_codebook/UPA_codebook_generator.m): `Deep_learning_MIMO/BF_codebook/UPA_codebook_generator.m`
- [Codebook parameters](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/CBF_parameters.m): `Deep_learning_MIMO/CBF_parameters.m`
- [Hybrid learning](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/Hybrid_Deep_learning.ipynb): `Deep_learning_MIMO/Hybrid_Deep_learning.ipynb`

### Learning workflow

- **Input/output data**
- [Training and evaluation](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/DL_model_python.py): `Deep_learning_MIMO/DL_model_python.py`
- [Per-transmitter models](https://github.com/masadipradeeprao/deep_learning_mimo/blob/main/Deep_learning_MIMO/DL_model_python.py): `Deep_learning_MIMO/DL_model_python.py`

### Evaluation outputs

- **Beam predictions**

### Other components

- **Researcher**
- **DeepMIMO scenario**

## Connections

- Researcher → DeepMIMO scenario: downloads (evidence: `README.md`)
- Researcher → DL data generation: runs (evidence: `README.md`)
- DeepMIMO scenario → DL data generation: provides dataset
- DL data generation → Input/output data: generates (evidence: `README.md`)
- Input/output data → Training and evaluation: loads (evidence: `Deep_learning_MIMO/DL_model_python.py`)
- Training and evaluation → Per-transmitter models: trains (evidence: `Deep_learning_MIMO/DL_model_python.py`)
- Training and evaluation → Beam predictions: saves results (evidence: `Deep_learning_MIMO/DL_model_python.py`)
- Per-transmitter models → Beam predictions: predicts beams (evidence: `Deep_learning_MIMO/DL_model_python.py`)
