Mobile Device Location Tracking Using Probabilistic Motion Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile device location tracking technologies face challenges in urban areas due to interference from landscape obstructions, requiring improvements in accuracy and reliability, especially when satellite signals are intermittent and environment data is not fully utilized.
Innovation Solution
A method and system that use probabilistic temporospatial motion models and probability map structures to determine the most probable location of a mobile device by combining cellular network data with geographical location data, incorporating sensor data and motion models to account for various types of motion and environmental factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If GPS-based location determination is used, then location tracking capability is provided, but signal reception is interrupted by landscape obstructions such as buildings, geographic features, and urban canyons
Solution Approach 1:
The system segments the location determination process into multiple independent data sources: GPS satellite signals, cellular base station signals, Wi-Fi access point signals, and inertial sensor measurements. Each segment operates independently and contributes to the overall location estimate, so failure of one segment (e.g., GPS blocked by buildings) does not prevent location tracking through other segments.
Solution Approach 2:
The system creates a composite location estimation by fusing data from heterogeneous sources with different characteristics: satellite-based GPS, terrestrial cellular networks, local Wi-Fi infrastructure, and device-mounted inertial sensors. This composite approach leverages the strengths of each source while compensating for their individual weaknesses, particularly in urban environments where GPS signals are obstructed.
2Loss of time
If assisted GPS with cellular base station data is used, then start-up time is reduced, but location estimate accuracy deteriorates when base station locations are not exactly available
Solution Approach 1:
The system introduces inertial sensors (accelerometers, gyroscopes, magnetometers) as intermediary measurement devices that provide independent location and motion information. These sensors do not rely on external infrastructure like GPS satellites or cellular base stations, and their measurements are fused with the other data sources to compensate for inaccuracies when base station locations are uncertain or unavailable.
Solution Approach 2:
The system continuously fuses incoming measurements from all sources with the current location estimate using probabilistic filtering algorithms. This feedback mechanism allows the system to dynamically adjust the weight of each data source based on its current reliability, and to correct accumulated errors in real-time as new measurements become available, thereby maintaining accuracy even when individual sources have uncertainties.
3Measurement precision
If multiple data sources are integrated for location determination, then positioning accuracy in challenging environments is improved, but system complexity increases
Solution Approach 1:
The system employs a universal probabilistic filtering framework (such as Kalman filtering or particle filtering) that can process and fuse data from multiple different sources with varying characteristics. This single unified algorithm handles GPS, cellular, Wi-Fi, and inertial sensor data through a common mathematical structure, reducing the need for separate processing pipelines for each sensor type and thereby managing system complexity despite the diversity of inputs.
Solution Approach 2:
The system dynamically adjusts the weighting parameters and covariance matrices of different data sources based on their current quality and reliability. When GPS signals are strong, the system increases the weight of GPS measurements; when cellular base station locations are uncertain, it reduces their weight and increases reliance on inertial sensors. This adaptive parameter adjustment optimizes positioning accuracy while managing computational complexity by focusing processing on the most reliable data sources at any given moment.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for locating an electronic mobile device capable of wireless communication, the method comprising obtaining a probability map structure for the area of interest, said probability map structure associating environment data values, preferably including cellular network based data, with location data, obtaining a number of probabilistic temporospatial motion models for different types of motion, said models at least partly defining transition probabilities between location estimates relative to time, obtaining at least environment data captured or determined by the mobile device at a plurality of sequential time instances, and determining the most probable location estimate of the mobile device based on combining probability data from both the probability map structure and said motion models, wherein the at least environment data is best fitted in the motion models and probability map structure according to predefined criteria. Related two arrangements and a computer program are presented.