Wireless Cell Location Correction Using Network Usage Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing datasets mapping the locations of wireless equipment in cellular networks, such as cell towers, often contain inaccurate or missing location information due to data write-in issues, software bugs, and hardware problems, which hinders the ability of cellular network providers to derive accurate demographic and user location insights.
Innovation Solution
Utilizing machine learning models trained on network usage data and artificially introduced noise to identify incorrect or missing locations of wireless equipment, and generate accurate estimates for these locations, thereby updating the dataset with precise location information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If crowdsourced datasets are used to map wireless equipment locations, then location coverage is improved, but location accuracy deteriorates due to data write-in issues, software bugs, and hardware problems
Solution Approach 1:
The system uses machine learning models that analyze network usage data to provide feedback on the accuracy of location information in crowdsourced datasets. The models identify inconsistencies between reported locations and actual network behavior patterns, enabling continuous improvement of location accuracy while maintaining broad coverage
Solution Approach 2:
Machine learning models serve as an intermediary between crowdsourced location data and network operations. The models process and validate location information from multiple sources including OpenCelliD, cross-referencing it with network usage patterns to filter out inaccurate data while preserving comprehensive location coverage
2Measurement precision
If machine learning models are used to identify and correct location errors, then location accuracy is improved, but data processing complexity increases
Solution Approach 1:
The machine learning models automatically identify and correct location errors without requiring manual intervention. The system self-services by training on network usage data and autonomously determining accurate locations of wireless cells, reducing the need for complex manual data validation processes
Solution Approach 2:
The system changes parameters by training machine learning models on multiple features including cell IDs, geographic coordinates, and network usage patterns. By adjusting and analyzing multiple parameters simultaneously, the system achieves high location accuracy while managing processing complexity through automated model training and inference
3Measurement precision
If complete location data is obtained for all wireless cells, then demographic analysis accuracy is improved, but data collection requirements increase
Solution Approach 1:
Machine learning models act as intermediaries that infer missing location data from available network usage patterns. Instead of requiring complete direct measurements for all cells, the models interpolate and predict locations based on surrounding cell data and user connection patterns, reducing data collection requirements while maintaining analysis accuracy
Solution Approach 2:
The system creates copies of location information by training machine learning models on existing network usage data and then using these trained models to generate location estimates for cells with missing data. This copying approach allows demographic analysis with near-complete location coverage without requiring exhaustive data collection
Data Source
AI summary
A method includes receiving first data indicative of network usage of multiple users of a wireless network and receiving second data indicative of one or more locations of a set of wireless cells. The first data includes information representing sequences of wireless cells that are connected to by user devices of the multiple users. The method also includes identifying, using one or more machine learning models, a portion of the one or more locations that are likely to be incorrect based on the first data and the second data. The method also includes generating estimates of a revised location for each wireless cell corresponding to the identified portion of the one or more locations that are likely to be incorrect, wherein the estimates are generated by one or more additional machine learning models. The method also includes updating the second data to include the generated estimates.