ML-Based Drive Test Prioritization for 5G Coverage Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional drive testing methods for 5G networks are cumbersome and resource-intensive, requiring tedious trial-and-error approaches to ensure network coverage meets performance requirements, straining computational, communication, and human resources.
Innovation Solution
A computer-implemented method using a trained machine learning model that integrates estimated radio frequency measurements with past drive test data to prioritize drive tests, determining execution priorities based on predicted success rates and geographic data, and adjusting network configurations as needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional trial-and-error drive testing is performed to ensure network coverage meets performance requirements, then network coverage quality is improved, but resource consumption (computational, communication, hardware, and human resources) increases
Solution Approach 1:
The system performs preliminary actions by training a machine learning model on historical drive test data and using it to predict outcomes of candidate drive tests before execution. This allows prioritization of high-value tests and avoidance of low-value ones, reducing actual resource consumption while maintaining coverage quality.
Solution Approach 2:
The system uses automatically collected historical drive test data and estimated RF measurements to train the machine learning model, which then autonomously prioritizes future drive tests. This self-service approach eliminates manual trial-and-error planning and reduces human resource consumption.
2Measurement precision
If comprehensive drive testing is performed across all network areas, then measurement completeness is improved, but time consumption increases
Solution Approach 1:
The system applies partial action by selectively executing only the most valuable drive tests based on machine learning predictions. Instead of performing exhaustive testing across all areas, it focuses resources on tests with highest predicted success rates and information value, achieving adequate measurement completeness with reduced time consumption.
Solution Approach 2:
The system uses historical drive test results as feedback to train the machine learning model, which then predicts outcomes of candidate tests. This feedback loop enables intelligent prioritization that maintains measurement completeness by focusing on areas most likely to provide valuable information.
3Productivity
If machine learning based prioritization is implemented, then resource efficiency is improved, but system complexity increases
Solution Approach 1:
The machine learning model serves multiple functions: it predicts drive test outcomes, prioritizes test sequences, and implicitly identifies high-value measurement areas. This multi-functionality justifies the added system complexity by delivering comprehensive resource efficiency improvements across the entire drive testing process.
Data Source
AI summary
Technologies for network drive test prioritization based on machine learning are disclosed. An example method includes feeding a representation of routes of a target candidate drive test to a trained machine learning model to obtain a drive test prediction, wherein the trained machine learning model is trained based on integrating radio frequency (RF) estimations or predictions with past drive test data. The method also includes sorting a set of candidate drive tests for the communications network including the target candidate drive test, based on drive test predictions associated with each candidate drive test, to determine priorities for executing drive tests; and determining expectation of network usability in accordance with network availability and performance metrics.