Link-Type Grid Fingerprint Database for Indoor Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fingerprint-based indoor positioning systems face challenges in creating accurate databases efficiently, leading to decreased positioning resolution and quality due to large grid sizes and complex indoor environments, resulting in 'jumping' or 'bouncing' location results.
Innovation Solution
A link-type grid fingerprint database is created by simplifying indoor spaces into connection networks of links, collecting positioning resource information along these paths, and including link attribute information to improve positioning accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the indoor space is divided into smaller grid sections to improve positioning resolution, then positioning accuracy is improved, but data collection time and expense increase significantly
Solution Approach 1:
The patent applies segmentation by dividing the indoor space into grid sections but then further segmenting the data collection process by identifying and prioritizing frequent movement paths within these sections. This allows selective collection of positioning resource information only along these paths rather than uniformly across all grid points, resolving the contradiction between achieving fine positioning resolution and reducing data collection time.
2Measurement precision
If the indoor space is divided into smaller grid sections to improve positioning resolution, then positioning accuracy is improved, but creation cost of the database increases
Solution Approach 1:
The patent segments the database creation process by identifying frequent movement paths and collecting positioning resource information only along these paths rather than at all grid points. This selective approach maintains high positioning resolution where users actually move while significantly reducing the overall data collection effort and database creation cost.
Solution Approach 2:
The patent applies local quality by concentrating data collection efforts on specific locations (frequent movement paths) rather than uniformly across the entire space. This ensures high-quality positioning data is collected where it is most needed while reducing unnecessary data collection in areas with low user traffic, thereby reducing overall database creation cost while maintaining positioning accuracy.
3Area of stationary object
If positioning is performed by comparing patterns in a 2-dimensional grid, then comprehensive coverage is achieved, but positioning results show jumping or bouncing due to environmental changes
Solution Approach 1:
The patent segments the positioning approach by focusing on 1-dimensional movement paths rather than 2-dimensional grid cells. This segmentation along frequent movement paths provides more stable positioning results by reducing the impact of environmental changes that cause jumping or bouncing in traditional 2-dimensional grid-based approaches.
Solution Approach 2:
The patent applies dynamics by adaptively identifying frequent movement paths based on user behavior patterns and updating the positioning model accordingly. This dynamic approach improves positioning reliability by adapting to changing environmental conditions and user movement patterns, reducing jumping or bouncing effects compared to static 2-dimensional grid approaches.
Data Source
AI summary
An apparatus for creating a link-type grid fingerprint database includes a collected data processing unit for receiving positioning resource information collected from a link, which represents a path along which a pedestrian can move in an indoor space, and storing the positioning resource information in a collection DB, and a database creation unit for creating a link-type fingerprint database based on the positioning resource information and information about the link, stored in the collection DB, and storing the created link-type grid fingerprint database in a positioning DB.