SQL-Geospatial-Clustering-with-DBSCAN

Overview

This repository demonstrates the clustering of geographic entities using the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm using Big Query SQL. DBSCAN is particularly effective for identifying spatial patterns and grouping nearby points based on density, making it well-suited for geospatial analysis.

Example Use Case

The example provided in this repository focuses on grouping contiguous hexbins in a map grid based on shared radio frequency (RF) characteristics. By leveraging DBSCAN, we can efficiently identify clusters of hexbins that exhibit similar RF properties, which can be useful for applications such as:

• Identifying coverage areas in wireless networks
• Detecting signal interference zones = Analyzing spatial distribution of RF performance metrics
• Adaptability

While this example is designed for contiguous hexbins, the methodology can be adapted to:

• Non-contiguous geographic entities, such as scattered infrastructure points or sensor locations
• Alternative input attributes, including environmental, demographic, or infrastructure-related data

This flexibility makes the approach suitable for a wide range of spatial clustering problems beyond RF analysis.

NB - The code in this project is for guidance as it is very much still project-specific.