In-transit detection using WiFi heuristics and accelerometer fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for detecting the drive state in mobile devices using accelerometers often result in false positives due to similarities in acceleration signals between pedestrian and vehicular motion, leading to battery drain and erroneous warnings, as GPS usage is power-intensive and not always available.
Innovation Solution
The proposed solution utilizes WiFi connectivity information as heuristics in conjunction with accelerometer signals to improve the detection of the in-transit state, employing a heuristics-based approach that sets probabilities for motion states, reducing false positives by differentiating between in-transit, stationary, and walking states using distributed or hierarchical architectures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If accelerometer-based motion classification is used to detect drive state, then low power consumption is achieved, but false positives occur due to signal similarity between pedestrian and vehicular motion
Solution Approach 1:
The patent segments the motion detection task into multiple specialized modules: a drive state detector specifically tuned for vehicular motion, a pedestrian motion detector for walking/running states, and a general motion classifier. Each module processes accelerometer signals independently and provides probabilistic outputs that are combined to determine the final motion state, thereby improving detection accuracy while maintaining low power consumption through targeted processing.
Solution Approach 2:
The patent introduces a new dimension of decision-making by incorporating probabilistic confidence scores from multiple specialized detectors alongside the traditional accelerometer signal analysis. This multi-dimensional approach combines signal characteristics with detection confidence levels to resolve ambiguities between similar motion patterns, improving accuracy without requiring additional sensors or excessive power consumption.
2Measurement precision
If GPS is used to resolve motion state ambiguities, then measurement precision is improved, but power consumption increases significantly
Solution Approach 1:
The patent introduces an intermediary probabilistic decision framework that mediates between accelerometer signal analysis and final motion state determination. Instead of directly relying on power-intensive GPS, the system uses specialized detectors that output confidence scores, which serve as an intermediary layer to resolve ambiguities. This intermediary approach achieves high detection accuracy by combining multiple low-power signal sources with probabilistic reasoning, avoiding the need for continuous GPS operation.
3Measurement precision
If additional sensors are added to improve motion state detection, then measurement precision is improved, but device cost and power consumption increase
Solution Approach 1:
The patent implements dynamic processing where the system adapts its analysis depth based on the confidence levels of specialized detectors. When the drive state detector and pedestrian motion detector provide high-confidence results, the system can make quick determinations without exhaustive analysis. This dynamic approach maximizes the utilization of existing accelerometer capabilities, achieving high accuracy through intelligent signal processing rather than additional hardware sensors.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
System and methods are disclosed to use information available on the state of mobile devices in a heuristics-based approach to improve motion state detection. In one or more embodiments, information on the WiFi connectivity of mobile de vices may be used to improve the detection of the in-transit state. The WiFi connectivity information may be used with sensor signal such as accelerometer signals in a motion classifier to reduce the false positives of the in-transit state. In one or more embodiments, information that a mobile device is connected to a WiFi access point (AP) may be used as heuristics to reduce the probability of falsely classifying the mobile device in the in-transit state when mobile device is actually in the hand of a relatively stationary user. Information on the battery charging state or (he wireless connectivity of the mobile devices may also be used to improve the detection of in-transit state.