Geolocation Stop Period Detection Using Kalman Filter Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to accurately identify stop periods of a mobile device in real-time using noisy and partially missing geolocation data, as they are not adapted for real-time processing and are unreliable due to data irregularity and uncertainty.
Innovation Solution
A method utilizing a discrete-time Kalman Filter to predict and filter geolocation data, adapting prediction error thresholds to spatial uncertainty levels, and employing exponential smoothing to determine stop periods, which can handle geolocation data of varying quality and trigger actions accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If geolocation data are used to identify stop periods by verifying locations are within place limits, then stop periods can be identified, but the solution is unreliable due to noisy and irregular data
Solution Approach 1:
The system performs preliminary actions by predicting future geolocation data points based on historical movement patterns before actual data arrives. This allows the system to prepare expected locations and compare them with actual measurements, improving reliability by having reference points ready in advance for comparison with noisy data
Solution Approach 2:
The patent introduces an intermediary prediction model that acts as a mediator between raw geolocation data and stop period identification. The prediction model smooths out noise and fills gaps in irregular data, providing a cleaner intermediate representation that makes reliable stop period detection possible despite poor measurement precision
2Reliability
If spatial clustering methods are used on stored log files, then stop periods can be identified, but the solution can only be implemented as post-processing and not in real-time
Solution Approach 1:
The system performs preliminary clustering and pattern recognition continuously on incoming data streams, maintaining updated models of movement patterns and stop behaviors in real-time. This preliminary processing ensures that when a stop period needs to be identified, the computational work has already been done, enabling immediate detection without post-processing delays
Solution Approach 2:
The patent implements continuous real-time processing of geolocation data through ongoing prediction and pattern matching operations. Instead of batch processing stored logs, the system continuously analyzes incoming data points, maintaining uninterrupted analysis that enables immediate stop period identification while preserving the accuracy benefits of clustering methods
3Ease of operation
If conventional methods assume regular location intervals, then processing is simplified, but the methods fail when data are missing or irregular
Solution Approach 1:
The system dynamically adapts its processing approach based on the actual characteristics of incoming data. Instead of assuming fixed regular intervals, the prediction model adjusts to variable time gaps between location points, modifying its expectations and comparisons in real-time to match the actual irregular data pattern while maintaining operational simplicity
Solution Approach 2:
The patent changes key parameters of the processing system based on observed data characteristics. When data irregularity is detected, the system adjusts prediction horizons, comparison thresholds, and smoothing parameters to accommodate the irregular pattern, maintaining ease of operation through automatic parameter adaptation rather than complex manual configuration
Data Source
Figure 1~2
Figure 3~4
AI summary
The invention concerns means for determining a stop period of a mobile device in geographical locations, according to geolocation data related to said mobile device, and for triggering an action accordingly, comprising: • means for determining a predicted position of the mobile device and an associated level of uncertainty related to the predicted position, according to the geolocation data and a first position related to the mobile device; the first position being either a current position of the mobile device if it is possible to obtain said current position from the geolocation data, or an artificial position of the mobile device according to a previously known current position of the mobile device; • means for, • determining a normalized prediction error according to the difference between the predicted position and the first position; • if the normalized prediction error is smaller than a prediction error threshold, generating a stop message to indicate that the stop period is starting ; • means for triggering an action on the mobile device, according to the stop message.