Adaptive Geo-fence Event Detection Using Confidence Levels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing location-based functionality in mobile devices faces challenges in accurately determining device location, leading to incorrect actions being taken or not taken, resulting in user frustration due to the uncertainty in pinpointing the exact position of the device within geo-fences.
Innovation Solution
A system that identifies the size of a geo-fence and a position uncertainty area based on the estimated accuracy error, using varying confidence levels to determine geo-fence events such as entering, exiting, or staying within the geo-fence, by adapting to the size of the geo-fence and the uncertainty area, thereby reducing false alarms and missed events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a fixed confidence level threshold is used for geo-fence event detection, then the detection process is simple and fast, but the accuracy of event detection deteriorates due to position uncertainty leading to false alarms and missed events
Solution Approach 1:
The patent applies dynamics by making the confidence level threshold adaptive rather than fixed. The threshold dynamically adjusts based on the ratio between geo-fence size and position uncertainty area. When the geo-fence is large relative to uncertainty, a higher threshold is used; when the geo-fence is small relative to uncertainty, a lower threshold is applied. This dynamic adjustment resolves the contradiction by maintaining detection accuracy across different scenarios without requiring overly complex fixed-threshold systems.
Solution Approach 2:
The patent changes the parameter of confidence level threshold based on the relationship between geo-fence dimensions and position uncertainty. By modifying this critical parameter according to contextual factors (geo-fence size versus uncertainty area), the system achieves accurate event detection while avoiding the complexity of multiple fixed-threshold systems or overly sophisticated algorithms.
2Reliability
If a high confidence level threshold is used to reduce false alarms, then false positive detection is reduced, but the system becomes less responsive and may miss legitimate geo-fence events
Solution Approach 1:
The system dynamically adjusts the confidence level threshold based on the geo-fence size to uncertainty area ratio. For large geo-fences where position uncertainty is less significant, a high threshold (e.g., 95%) is applied to ensure reliability and reduce false alarms. For small geo-fences where uncertainty has greater impact, a lower threshold is used to maintain responsiveness and detect legitimate events. This dynamic approach resolves the contradiction between reliability and responsiveness.
Solution Approach 2:
The patent applies local quality by using different confidence level thresholds for different geo-fence scenarios rather than a single global threshold. Each geo-fence event evaluation uses a locally optimized threshold appropriate to that specific geo-fence's size and the associated position uncertainty, thereby achieving both high reliability where needed and high responsiveness where appropriate.
3Measurement precision
If position uncertainty area is not considered in geo-fence event detection, then the detection process is simpler, but the accuracy of determining device location within geo-fences deteriorates
Solution Approach 1:
The patent applies preliminary action by calculating the position uncertainty area in advance and using it to determine the appropriate confidence level threshold before evaluating geo-fence events. This pre-computation of uncertainty metrics allows the system to accurately determine device location within geo-fences without adding complex real-time calculations during event detection, thus improving accuracy while maintaining process simplicity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The location of a computing device is determined, and the location of an area of interest that is a geographic area referred to as a geo-fence is identified. The accuracy of the determined location of the computing device has an associated uncertainty, so the exact position of the computing device cannot typically be pinpointed. In light of this, the uncertainty associated with the determined location is evaluated relative to the size of the geo-fence in order to determine whether the computing device is inside the geo-fence or outside the geo-fence. Based on this determination, various actions can be taken if the user is entering the geo-fence, exiting the geo-fence, remaining in the geo-fence for at least a threshold amount of time, and so forth.