On-Device Positioning Using Compressed RF Fingerprint Archives
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing outdoor positioning systems, such as GPS, are ineffective in environments with limited sky visibility, necessitating alternative methods for mobile devices to determine their location indoors.
Innovation Solution
The use of compressed fingerprint archives based on radio frequency (RF) fingerprint models, where a mobile device calculates and stores data triples representing access points, allowing for efficient on-device positioning by matching current RF fingerprints with pre-defined geographical locations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If GPS is used for outdoor positioning, then positioning accuracy is improved, but it becomes ineffective in indoor environments with limited sky visibility
Solution Approach 1:
The patent introduces RF access points as intermediary objects to enable positioning in indoor environments. Instead of directly using GPS satellites, the system uses locally deployed access points that emit RF signals, serving as intermediaries that allow the mobile device to determine its location through fingerprint matching when GPS is unavailable
Solution Approach 2:
The system creates copies of the physical environment in the form of RF fingerprints. Each location is characterized by a unique fingerprint consisting of signal strengths from multiple access points, creating a digital representation of the physical space that can be stored and matched against current readings to determine position
2Measurement precision
If complete fingerprint data is stored for all access points, then positioning accuracy is maintained, but memory requirements increase significantly
Solution Approach 1:
The patent extracts only the essential features from complete fingerprint data by identifying and removing redundant access points. Instead of storing data for all observed access points, the system selectively retains only those access points that contribute meaningfully to location discrimination, eliminating unnecessary data while preserving positioning accuracy
Solution Approach 2:
The system applies different levels of data retention to different access points based on their local importance. Each access point is evaluated individually, and only those with significant contribution to fingerprint uniqueness are retained, creating a non-uniform, optimized data structure where quality varies by local relevance rather than applying a uniform retention policy
3Reliability
If more access points are monitored for fingerprint creation, then positioning reliability improves, but device complexity and processing requirements increase
Solution Approach 1:
The system monitors all access points temporarily during the survey phase (excessive action) but then selectively retains only the necessary subset for actual positioning operations. This partial retention strategy maintains reliability by keeping sufficient access points while reducing complexity by discarding redundant ones
Data Source
AI summary
A method and apparatus are provided for on-device positioning using compressed fingerprint archives. The method and apparatus may be configured to provide compression of localization fingerprints, to facilitate efficient on-device positioning based on the RF fingerprint model, and to estimate the physical distance between two or more devices. Embodiments of the method may receive a space-to-access point histogram that corresponds to an access point as observed in a space. The histogram may be normalized and a mean and standard deviation may be calculated from the histogram. A weight may be assigned to the access point corresponding to the frequency with which the access point is observed within the space.