Adaptive Indoor Positioning Algorithm Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Location estimation in indoor environments using Wi-Fi beacons is challenging due to unpredictable signal propagation, leading to inaccuracies in position calculation, especially when beacon density is low or RF characteristics are unpredictable.
Innovation Solution
A system that selects between two positioning algorithms and data models based on device capabilities, observed beacons, and data availability, using 2D and 3D Wi-Fi fingerprint models to calculate positions, with the 2D model being less sensitive to RF environment and the 3D model providing better results in high-density environments with predictable RF characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If Wi-Fi beacon signal strength is used to estimate position, then position can be determined in indoor environments where GPS is not accessible, but signal propagation unpredictability leads to measurement errors and inaccuracies
Solution Approach 1:
The system dynamically selects between different positioning algorithms ( fingerprinting, trilateration, hybrid) based on environmental conditions, beacon density, and data availability. This dynamic adaptation allows the system to optimize for reliability in low-density areas while achieving better precision in high-density areas with predictable RF characteristics.
Solution Approach 2:
The system changes parameters such as algorithm selection, data model choice (2D vs 3D), and weighting factors based on observed beacon density and signal characteristics. By adjusting these parameters adaptively, the system resolves the contradiction between reliability and precision across different indoor environments.
2Measurement precision
If multiple positioning algorithms and data models are provided to handle different environmental conditions, then positioning accuracy improves across diverse scenarios, but system complexity increases
Solution Approach 1:
The positioning system is segmented into multiple independent algorithms (fingerprinting, trilateration, hybrid) and data models (2D, 3D) that can be selectively executed. Each algorithm handles specific scenarios, reducing the complexity burden on any single component while collectively improving overall precision across diverse environments.
Solution Approach 2:
The system includes automated mechanism that selects the most appropriate algorithm and data model based on observed beacon data and environmental characteristics. This self-service selection reduces the need for manual configuration and simplifies operation despite the presence of multiple complex algorithms.
3Measurement precision
If adaptive selection between positioning algorithms is implemented, then position calculation accuracy improves under varying conditions, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary assessments of environmental conditions (beacon density, signal characteristics, data availability) before selecting the positioning algorithm. This preliminary action enables informed algorithm selection that balances precision requirements with processing time constraints, avoiding unnecessary computational overhead.
Solution Approach 2:
The system applies partial processing by selecting only the most suitable algorithm for the current environment rather than executing all available algorithms. This partial action approach achieves sufficient precision for each scenario while minimizing unnecessary computational time and resource consumption.
Data Source
AI summary
A system and method for calculating a position in response to a position request. Observed beacon data associated with the request is used to select a calculation method based on available data for a venue and device capabilities. If sufficient venue data based on previously verified beacon positions is available, a position calculation can resolve floor and venue information. If insufficient previously observed data is available for a venue, the position is calculated using 2D data based on GPS observations. Following the choice a calculation model, the calculation position is returned in response to the position request.