FM-EAC/
├── Urban/                                       # Urban scenario modeling and communication simulation
│   ├── city_map/                                # Original TIFF map dataset
│   ├── city_map_cropped/                        # Cropped map images used for simulation
│   ├── trace/                                   # Pedestrian movement trajectory dataset
│   ├── antenna.py                               # Beamforming and antenna simulation
│   ├── cap.py                                   # Generates QoS-SINR relationship figures
│   ├── channel.py                               # Wireless channel capacity calculation
│   ├── communication.py                         # UAV communication model with the environment
│   ├── physical_model.py                        # UAV kinematic and battery consumption modeling
│   ├── crop_fig.py                              # Automated TIFF cropping script
│   ├── store_dsm.py                             # Store DSM (Digital Surface Model) height maps
│   ├── plot_antenna.py                          # Visualization of antenna patterns
│   ├── plot_building.py                         # 2D building distribution visualization
│   ├── plot_terrain.py                          # 3D rendering of terrain including buildings and elevation
│   ├── plot_traffic.py                          # Visualization of pedestrian traffic flow
│   ├── signal_map.py                            # Generate SINR heatmaps based on simulation data
│   ├── EAC-PAN/                                 # PAN-based reinforcement learning for urban UAV deployment
│   │   ├── outputs/
│   │   │   ├── photos/                          # Output of various visualizations
│   │   │   ├── path/                            # UAV trajectory images
│   │   │   ├── reward/                          # Reward curve plots during training
│   │   │   └── results/                         # Numerical results: inference time and reward
│   │   ├── model/                               # Trained neural network weights and checkpoints
│   │   ├── feature_extraction.py                # Feature extraction and PAN model training
│   │   ├── pointarray_feature_extractor.pth     # Saved feature extraction model weights
│   │   ├── env.py                               # UAV environment simulation for reinforcement learning
│   │   ├── urban_eac_pan_model.py               # PAN + RL network architecture
│   │   ├── train.py                             # Main training pipeline
│   │   └── test.py                              # Model evaluation and testing
│   └── EAC-GNN/                                 # GNN-based reinforcement learning for urban UAVs
│       ├── outputs/
│       │   ├── photos/                          # Visualization outputs
│       │   ├── path/                            # UAV path plots
│       │   ├── reward/                          # Reward plots
│       │   └── results/                         # Inference and performance results
│       ├── model/                               # GNN model weights
│       ├── env.py                               # UAV flight environment for GNN-based learning
│       ├── urban_eac_gnn_model.py               # GNN + RL model architecture
│       ├── train.py                             # Training pipeline
│       └── test.py                              # Testing and evaluation
│
├── Agriculture/                                 # Agricultural UAV monitoring and communication simulation
│   ├── communication.py                         # UAV-WS (Wireless Station) communication logic
│   ├── ws_generation.py                         # Generator for spatial WS distributions
│   ├── physical_model.py                        # UAV dynamics and energy model for agriculture
│   ├── terrain.py                               # Procedural terrain generation
│   ├── EAC-PAN/                                 # PAN-based learning for dual agricultural UAV tasks
│   │   ├── outputs_visit/                       # Output for visitation subtask
│   │   │   ├── photos/                          # Visual outputs
│   │   │   ├── 2Dpath/                          # Trajectory visualizations
│   │   │   ├── 3Dpath/                          # Trajectory visualizations
│   │   │   └── results/                         # Evaluation metrics
│   │   ├── outputs_return/                      # Output for return subtask
│   │   │   ├── photos/                          # Visual outputs
│   │   │   ├── 2Dpath/                          # Trajectory visualizations
│   │   │   ├── 3Dpath/                          # Trajectory visualizations
│   │   │   └── results/                         # Evaluation metrics
│   │   ├── model/                               # Trained PAN model weights
│   │   ├── feature_extraction.py                # PAN training script
│   │   ├── pointarray_feature_extractor.pth     # Trained extractor weights
│   │   ├── sample_for_return.py                 # Dataset preparation for BPN
│   │   ├── pre-train-predict.py                 # Energy predictor training
│   │   ├── energy_predictor.pth                 # Trained BPN model
│   │   ├── time-print.py                        # Inference time evaluation
│   │   ├── visit_env.py                         # Environment for visitation task
│   │   ├── return_env.py                        # Environment for return task
│   │   ├── main_env.py                          # Combined UAV task environment
│   │   ├── agri_eac_pan_model.py                # PAN + RL structure for agriculture
│   │   ├── visit_train.py                       # Subtask-specific training
│   │   ├── return_train.py                      # Subtask-specific training
│   │   ├── main_train.py                        # Full task training
│   │   ├── visit_test.py                        # Subtask evaluation
│   │   ├── return_test.py                       # Subtask evaluation
│   │   └── main_test.py                         # Full task evaluation
│   └── EAC-GNN/                                 # GNN-based UAV control in agriculture
│       ├── outputs_visit/                       # Same structure and purpose as PAN variant
│       ├── outputs_return/                      # Same structure and purpose as PAN variant
│       ├── model/                               # GNN model weights
│       ├── sample_for_return.py                 # Dataset preparation
│       ├── pre-train-predict.py                 # BPN training
│       ├── energy_predictor.pth                 # Trained BPN weights
│       ├── time-print.py                        # Timing analysis
│       ├── visit_env.py                         # Environment simulations
│       ├── return_env.py                        # Environment simulations
│       ├── main_env.py                          # Environment simulations
│       ├── agri_eac_gnn_model.py                # GNN-RL model
│       ├── visit_train.py                       # Subtask & full training
│       ├── return_train.py                      # Subtask & full training
│       ├── main_train.py                        # Subtask & full training
│       ├── visit_test.py                        # Subtask & full evaluation
│       ├── return_test.py                       # Subtask & full evaluation
│       └── main_test.py                         # Subtask & full evaluation
