🧑‍💻 Intermediate Tutorial: Customizing Placements and Frequencies¶

This tutorial builds upon the Basic Tutorial by demonstrating how to customize antenna placements and simulation frequencies in GOLIAT. You'll learn to modify configuration files to explore different exposure scenarios.

🎯 What you'll achieve¶

By the end of this tutorial, you will be able to: - Create a custom configuration file for intermediate studies. - Define multiple frequencies for a single study. - Specify custom antenna placements relative to the phantom. - Run a study with your customized settings.

📋 Prerequisites¶

Ensure you have completed the Basic Tutorial and have a working GOLIAT environment.

Step 1: Create a custom configuration file¶

Instead of modifying the default near_field_config.json, we'll create a new configuration file to keep your changes organized.

1. Copy configs/near_field_config.json to configs/my_intermediate_study.json.

   cp configs/near_field_config.json configs/my_intermediate_study.json
   

2. Open configs/my_intermediate_study.json in your code editor.

Step 2: Define multiple frequencies¶

Let's add another frequency to our study. We'll simulate at both 700 MHz and 900 MHz.

Locate the "frequencies_mhz" array and modify it:

{
  "extends": "base_config.json",
  "study_type": "near_field",
  "phantoms": ["thelonious"],
  "frequencies_mhz": [700, 900], // Added 900 MHz
  "execution_control": {
    "do_setup": true,
    "do_run": true,
    "do_extract": true
  },
  // ... rest of the config
}

Step 3: Specify custom antenna placements¶

GOLIAT allows you to define various antenna placement scenarios. We'll add a new placement called "by_ear" and enable it for the "thelonious" phantom.

1. Add a "placement_scenarios" block to your my_intermediate_study.json (if it doesn't exist, or modify if it does). This block defines the base positions and orientations.

   {
     "extends": "base_config.json",
     "study_type": "near_field",
     "phantoms": ["thelonious"],
     "frequencies_mhz": [700, 900],
     "execution_control": {
       "do_setup": true,
       "do_run": true,
       "do_extract": true
     },
     "placement_scenarios": {
       "by_cheek": {
         "positions": {"base": [0, 0, 0]},
         "orientations": {"base": [], "up": [{"axis": "X", "angle_deg": 10}]}
       },
       "by_ear": { // New placement scenario
         "positions": {"base": [0, 50, 0]}, // Example: 50mm along Y-axis from origin
         "orientations": {"base": [], "rotate_z": [{"angle_deg": 90}]} // Example: Rotate 90 deg around Z
       }
     },
     // ... rest of the config
   }
   

   Note: The exact coordinates and rotations will depend on your antenna model and desired position. These are illustrative examples.

2. Enable the new placement for the "thelonious" phantom in the "phantom_definitions" section:

   {
     // ...
     "phantom_definitions": {
       "thelonious": {
         "placements": {
           "do_by_cheek": true,
           "do_by_ear": true // Enable the new placement
         },
         "distance_from_cheek": 8,
         "distance_from_ear": 5 // Example: 5mm distance for "by_ear"
       }
     }
   }
   

Step 4: Run the customized study¶

Save your configs/my_intermediate_study.json file. Now, execute the study using your new configuration:

python run_study.py --config configs/my_intermediate_study.json

GOLIAT will now run simulations for: - "thelonious" phantom at 700 MHz with "by_cheek" placement. - "thelonious" phantom at 700 MHz with "by_ear" placement. - "thelonious" phantom at 900 MHz with "by_cheek" placement. - "thelonious" phantom at 900 MHz with "by_ear" placement.

You will see the GUI update as each simulation scenario is processed.

Step 5: Examine the results¶

After the study completes, navigate to your results/ directory. You will find new subdirectories corresponding to the additional frequency and placement:

• results/near_field/thelonious/700MHz/by_ear/
• results/near_field/thelonious/900MHz/by_cheek/
• results/near_field/thelonious/900MHz/by_ear/

Each directory will contain its own sar_results.json, sar_stats_all_tissues.pkl, and plots.

Run the analysis script to aggregate all results from this multi-scenario study:

python run_analysis.py --config configs/my_intermediate_study.json

This will generate aggregated CSVs and plots that include data from all frequencies and placements you defined.

⚠️ Troubleshooting¶

• "Placement not found": Double-check that the placement name in "phantom_definitions" (e.g., "do_by_ear": true) exactly matches a key in your "placement_scenarios" block (e.g., "by_ear").
• JSON Syntax Errors: Ensure your JSON file is correctly formatted. Use a JSON validator if you encounter issues.
• Simulation Time: Running multiple frequencies and placements will increase the total simulation time. Consider using fewer scenarios for quick tests.

➡️ Next steps¶

• Advanced Batching: Learn how to run many simulations in parallel using oSPARC cloud batching in the Advanced Tutorial.
• Far-Field Studies: Explore environmental exposure scenarios by configuring a far-field study.
• Developer Guide: If you're interested in extending GOLIAT, refer to the Developer Guide.

Last updated: {date}