RF Propagation & Terrain Modeling with HEAVY.AI

Objective: Build and analyze a high-resolution Digital Surface Model (DSM) from LiDAR and visualize RF propagation in the Dallas area using HEAVY.AI. 

Downloadable Notebook: aws s3 cp s3://batteries-included/usgs_coppell_texas_rf_rprop/dallas-greenfield.ipynb . --no-sign-request

RF Propagation Tutorial Video - Part 1

RF Propogation Tutorial Video - Part 2

Module 1: Set Up Environment and Dependencies

• Tools Used:
  • heavyai Python library (for DB connection and query execution)
  • geopandas, boto3, pyproj, dotenv 
• Initial Tasks: 
  • Install and verify Python libraries. 
  • Create a working import directory and set file permissions. 
  • Configure AWS S3 client for LiDAR file downloads. 

Module 2: Define Area of Interest (AOI)

• Task: Materialize a bounding box for Coppell, TX. 
• Tools: pyproj to convert from WGS84 to UTM. 
• Heavy.AI SQL: - Create a polygon geometry table using latitude and longitude boundaries. 

Module 3: Enrich Digital Elevation Model (DEM)

• Goal: Join base DEM tiles by aligning with UTM floor coordinates. 
• Output: Unified elevation surface, suitable for RF modeling. 

Module 4: Load and Process LiDAR Data 

• Tools: Heavy.AI table creation and import from S3. 
• Product: Point cloud data ready for DSM generation. 

Module 5: Generate 1-Meter DSM (Digital Surface Model) 

• Goal: Produce rasterized elevation model at 1m resolution. 
• Option: Restore DSM table from backup for faster iteration. 

Module 6: RF Propagation Table Creation 

• Pre-requisite: Copy antenna_types.csv to import directory. 
• Action: Use CREATE TABLE AS SELECT (CTAS) to build rf_prop_omnidirectional table. 

Final Step: Validate and Visualize 

If you've reached this point: - You've created and processed terrain and point cloud data. - You've modeled RF propagation using real-world elevation data. - You can now restore tables and dashboards to visualize propagation outcomes.