DBSCAN and Hierarchical Clustering for Poor-Coverage Deployment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing telecommunications systems struggle to accurately identify and deploy solutions to improve service quality and coverage in geographic areas with impaired or degraded service, lacking efficient methods for precise location determination and solution optimization.
Innovation Solution
A machine learning-based approach utilizing DBSCAN and hierarchical clustering algorithms to analyze telecommunications data, generating clusters of impaired service areas and determining optimal solution types and locations for deployment, such as tilt, low-band, lease, and small cell solutions, based on geographic data and network metrics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods are used to identify and deploy solutions for impaired service areas, then deployment can proceed with existing processes, but accuracy in identifying and determining precise deployment locations is insufficient
Solution Approach 1:
The patent replaces traditional manual or rule-based methods for identifying impaired service areas with machine learning-based clustering algorithms (DBSCAN and hierarchical clustering). This substitution enables automated, data-driven identification of poor quality areas with higher precision, eliminating the need for human analysts to manually interpret network performance data and manually determine deployment locations.
Solution Approach 2:
The system enables self-service by allowing the clustering algorithms to autonomously identify impaired service areas and determine optimal deployment locations without human intervention. The DBSCAN algorithm automatically clusters poor quality data points, and the hierarchical clustering algorithm subsequently identifies precise deployment locations within those clusters, creating a self-directed optimization process.
2Productivity
If manual analysis and deployment planning is used, then system complexity remains low, but productivity in identifying and deploying solutions is reduced
Solution Approach 1:
The patent replaces manual analysis and deployment planning with automated machine learning systems. The DBSCAN clustering algorithm processes network performance data to identify poor quality areas, and the hierarchical clustering algorithm determines optimal deployment locations, dramatically increasing productivity compared to manual methods while the system handles the computational complexity internally.
Solution Approach 2:
The clustering algorithms serve as intermediaries between raw network performance data and deployment decisions. Rather than humans directly analyzing data and making deployment decisions, the DBSCAN and hierarchical clustering algorithms process the data, identify patterns, and recommend deployment locations, acting as intelligent mediators that bridge data and action.
3Measurement precision
If comprehensive data analysis is performed to accurately identify deployment locations, then deployment precision is improved, but computational resources and processing time increase
Solution Approach 1:
The patent segments the analysis process into two distinct phases: first, DBSCAN clustering identifies poor quality service areas by grouping similar data points; second, hierarchical clustering determines precise deployment locations within those identified areas. This segmentation allows each algorithm to focus on a specific task, improving overall precision while managing computational resources more efficiently than a single comprehensive analysis would require.
Data Source
AI summary
Aspects herein provide a system, media, and methods for/of an application that utilizes and leverages machine learning techniques, infrastructure information (e.g., existing and planned cell sites, lease agreement sites, fiber-optic networks, geographic landmarks), and collected telecommunication data, to accurately identify and determine specific solutions and specific locations for those solutions to be deployed in a geographic area. In embodiments, the application autonomously identifies an optimized specific solution type for various clusters of poor-service coverages areas within the geographic areas. The application also determines a precise location for deployment of each optimized specific solution type for each cluster.