Positioning Service Experimentation Framework for Beacon Model Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing positioning services rely on crowd-sourced data, which can be noisy and unreliable, and lack systematic analysis of data quality and algorithm performance, leading to inconsistent accuracy in determining device positions.
Innovation Solution
A systematic positioning service experimentation framework that divides data into training and test datasets to evaluate and adjust beacon models and algorithms, improving accuracy by assigning observations to geographic areas and recalculating aggregate accuracy values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If crowd-sourced data is used for positioning, then the system can provide position information to multiple devices, but the accuracy deteriorates due to noisy and unreliable data
Solution Approach 1:
The patent segments the positioning system into multiple independent components: data collection module, data quality analysis module, beacon model generation module, and position inference module. This segmentation allows each component to be optimized independently, with the data quality analysis specifically addressing noise filtering to maintain accuracy while processing crowd-sourced data from multiple devices.
Solution Approach 2:
The patent introduces beacon models as intermediary representations that mediate between raw crowd-sourced data and position inference. These beacon models act as a filtering layer that captures essential spatial relationships while eliminating noise, enabling accurate position determination from unreliable crowd-sourced observations.
2Device complexity
If a single modeling algorithm is used, then the system is simple to implement, but the accuracy deteriorates because different algorithms perform better in different geographic areas
Solution Approach 1:
The patent implements a dynamic algorithm selection mechanism that adapts the modeling approach based on geographic location. The system evaluates data characteristics in different regions and selects or weights algorithms accordingly, allowing the system to maintain simplicity in uniform areas while achieving high accuracy in diverse geographic conditions through adaptive behavior.
3Quantity of substance
If crowd-sourced data is collected from multiple devices, then more position information can be obtained, but the reliability deteriorates due to differences in devices, locations, and observation conditions
Solution Approach 1:
The patent implements a feedback mechanism through data quality analysis that evaluates the reliability of crowd-sourced data before incorporating it into beacon models. The system analyzes observation conditions, device characteristics, and data consistency, providing feedback that weights or filters unreliable observations, thereby maintaining high data volume while ensuring reliability through continuous quality assessment.
Data Source
AI summary
Embodiments respond to a position inference request from a computing device to determine a location of a computing device. The position inference request received from the computing device identifies a set of beacons observed by the computing device. A geographic area is estimated in which the computing device is located using the set of beacons. At least one location method is selected to identify a location of the computing device within the geographic area. In some cases two or more location methods may be employed and their results combined using, for example, a weighting function. The location of the computing device is determined within the geographic area using the set of beacons and the selected location method(s). The location that is determined is communicated to the computing device.