Indoor Location Recommendation via Access Point Fingerprinting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing indoor positioning techniques require pre-constructed infrastructure and known access point locations, making it difficult to provide accurate positioning and location recommendations without initial infrastructure information.
Innovation Solution
A device and method that detect access point locations and store locations using fingerprints collected from multiple user terminals, generating a metamap to recommend locations based on user preference and current location, even without initial infrastructure information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional indoor positioning techniques are used that require pre-constructed infrastructure and known access point locations, then positioning accuracy can be maintained, but the system cannot provide positioning services in buildings without initial infrastructure information
Solution Approach 1:
The system performs preliminary actions by collecting fingerprint data from multiple user terminals before actual positioning is needed. This advance data collection builds a probabilistic map of access point locations that enables subsequent positioning without requiring pre-installed infrastructure or known AP locations
Solution Approach 2:
The patent introduces fingerprint data as an intermediary element that mediates between the user terminal and the positioning system. By using fingerprints collected from multiple terminals as intermediate information, the system can infer access point locations and provide positioning services without direct knowledge of AP positions or pre-constructed infrastructure
2Reliability
If access points are installed at regular distances to ensure coverage, then positioning coverage is improved, but the aesthetic appearance of the interior is spoiled and installation flexibility is reduced
Solution Approach 1:
The patent inverts the conventional approach by not requiring known access point locations for positioning. Instead of installing APs at specific locations and then using those known locations for positioning, the system first collects fingerprint data and then infers AP locations probabilistically, allowing APs to be installed freely without compromising positioning coverage
Solution Approach 2:
The system changes the parameter of access point location from a fixed, predetermined value to a probabilistic, inferred value. By transitioning from deterministic AP locations to probabilistic distributions based on fingerprint data, the system enables flexible AP installation while maintaining positioning reliability through statistical methods
3Ease of operation
If access points are installed in locations that minimize interference and maintain aesthetic appearance rather than at regular distances, then installation flexibility and aesthetic appearance are improved, but positioning coverage and reliability may deteriorate
Solution Approach 1:
The system implements feedback by continuously collecting fingerprint data from multiple user terminals and using this feedback to build and update the probabilistic map of access point locations. This feedback mechanism ensures that even irregularly placed APs are accurately located through aggregated data, maintaining positioning coverage without requiring regular installation patterns
Solution Approach 2:
The patent merges information from multiple user terminals to compensate for the irregular placement of access points. By combining fingerprint data from multiple sources, the system creates a robust probabilistic model that maintains positioning reliability even when individual APs are placed irregularly for aesthetic or interference-minimization reasons
Data Source
AI summary
Disclosed are a device for recommending locations in a building using fingerprints of access points and a method using the device. The locations of access points are detected by collecting fingerprints for the access points from terminals of multiple users, the locations of stores in the building are detected by using the fingerprints and payment histories of the multiple users, and locations may be recommended to a recommendation target user based on a meta-path, using a metamap generated by integrating the locations of the access points and stores with the indoor plan of the building. Indoor positioning is enabled by detecting the locations of access points using the information obtained from the terminals of multiple users, whereby constructing infrastructure information for the access points in the building in advance may not be required.


