Iterative Radio Map Generation Using Mobile Device Fingerprints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current indoor positioning technologies face challenges in accurately creating radio maps without user interaction, as satellite signals are unreliable indoors, and sensor-based solutions suffer from noise, drift, misalignment, and poor initial position estimates, making it difficult to georeference radio measurements and maintain accurate positioning.
Innovation Solution
An iterative process that determines fingerprints from multiple mobile devices, estimates radio transmitter characteristics, and refines position data using these characteristics, leveraging strong signal points to stitch together device tracks and improve radio map generation, incorporating existing infrastructure like WLAN access points and Bluetooth beacons for scalable and low-maintenance solutions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor-based tracking is used for indoor positioning, then positioning can be achieved without satellite signals, but the system suffers from noise, drift, misalignment, and poor initial position estimates
Solution Approach 1:
The patent implements an iterative feedback mechanism where radio measurements and sensor data are continuously processed to refine position estimates. The system uses measured radio signal characteristics (fingerprints) compared against a radio map to correct sensor-based position drift, creating a closed-loop feedback system that maintains positioning accuracy over time.
Solution Approach 2:
The system performs preliminary actions by pre-collecting radio measurements at known reference positions to create a radio map before actual positioning occurs. This pre-established database of radio fingerprints serves as a reference framework that enables accurate position determination without requiring real-time satellite signals.
2Productivity
If crowd-sourcing is used to collect training data, then exhaustive survey of sites can be achieved in short time at limited costs, but the system requires user participation consent and device functionality
Solution Approach 1:
The patent implements self-service by enabling mobile devices to automatically collect and upload radio measurement data without requiring explicit user action during data collection. The system leverages existing device sensors and radio interfaces to autonomously gather training data, reducing the operational burden on users while maintaining data quality.
3Measurement precision
If iterative refinement process is implemented, then radio map accuracy is improved, but processing time and computational resources increase
Solution Approach 1:
The patent applies partial action by implementing a limited number of iterative refinement cycles rather than continuous processing. The system performs one or more passes of fingerprint matching and position refinement, achieving sufficient accuracy without the computational overhead of exhaustive iterative optimization, thus balancing precision with processing efficiency.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A system obtains from each of a plurality of mobile devices a batch of data for a site. A batch includes data enabling a determination of a reference position, sensor data of motion sensors and sets of results of measurements performed by the mobile device on radio signals at different locations, each set including at least one indication of a received signal strength of radio signals and an associated identifier of at least one radio transmitter transmitting the radio signals. In at least two iterations, the system processes each batch to determine for each mobile device a (refined) sequence of fingerprints associating a respective position, which is based on the at least one reference position, the sensor data (and estimated characteristics of radio transmitters), with each set of results of measurements; and estimates characteristics of the radio transmitters based on the (refined) sequences of fingerprints.