Indoor Positioning via Crowdsourced Wireless AP Fingerprint Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing indoor positioning systems face challenges in accurately determining device location within buildings due to the labor-intensive process of manually creating and updating wireless access point (AP) fingerprint maps, which is inefficient and prone to errors.
Innovation Solution
The system employs crowd-sourcing by utilizing position-inference data from multiple devices to determine probable locations and refine wireless AP fingerprints, leveraging data from various sources such as security systems, calendar data, and VoIP systems to associate wireless AP signal measurements with specific locations, thereby improving the accuracy and reliability of indoor positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual fingerprint mapping is used, then positioning system can be established, but labor consumption and time required increase significantly
Solution Approach 1:
The system enables automatic fingerprint map creation by having mobile devices themselves contribute location data and wireless signal measurements. Users don't need manual intervention - the devices self-report their position (from GPS or other sources) and the system automatically captures wireless AP measurements at those locations to build fingerprints
Solution Approach 2:
The system uses feedback from multiple mobile devices reporting their locations and signal measurements to continuously refine and update the fingerprint database. This creates a self-improving system where more device participation leads to better positioning accuracy over time
2Measurement precision
If manual fingerprint updating is performed, then positioning accuracy can be maintained, but operational complexity and labor requirements increase
Solution Approach 1:
The system automatically updates fingerprints by receiving new location and signal data from mobile devices. The server autonomously processes incoming data, identifies location changes, and refreshes the fingerprint database without requiring manual triggering or complex operational procedures
Solution Approach 2:
Mobile devices serve multiple functions: they are both the positioning targets and the data collection instruments. The same device that needs positioning also contributes to creating and updating the fingerprint database, eliminating the need for separate manual surveying equipment and procedures
3Productivity
If crowd-sourcing approach is used, then manual labor is reduced, but data processing complexity increases
Solution Approach 1:
The system divides data processing into discrete segments: each mobile device independently reports its location and signal measurements as separate data units. The server processes these segmented data points individually, matching locations to fingerprints and updating only the specific entries that need refinement, rather than processing entire datasets at once
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An indoor positioning system and method that correlates map locations to respective wireless access point (AP) fingerprints, and wherein each wireless AP fingerprint is a plurality of wireless AP signal measurements associated with the correlated map location. The method includes receiving position-inference data associated with a user; determining a probable location inside the building based on the position-inference data; receiving wireless AP signal measurements detected by a wireless device associated with the user; associating the probable location with the wireless device based on the association between the position-inference data and the user of the wireless device; and updating the wireless AP fingerprint correlated to the probable location using the received wireless AP signal measurements. The method and system use crowd-sourcing to build and refine the wireless AP fingerprints.