Mobility Mode State Detection Using HMM and GNSS Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing GNSS-based mobility mode state detection in wireless mobile devices is often ambiguous due to poor real-time calibration of MEMS sensors, leading to high probabilities of incorrect mobility mode state conclusions when relying solely on these sensors.
Innovation Solution
A system incorporating a GNSS receiver, memory with location and mobility mode state detection algorithms, and FIFO queues, utilizing a processor to acquire satellite navigation data, estimate position, and detect mobility mode using a Hidden Markov Model (HMM) and dynamic programming algorithm to determine the most probable mobility mode state, thereby controlling position data output based on confidence limits and transition probabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If MEMS sensors are used for mobility mode state detection, then the device can detect movement types, but the detection accuracy deteriorates due to poor real-time calibration and ambiguous results
Solution Approach 1:
The patent introduces GNSS position data as an intermediary to verify and correct MEMS sensor-based mobility mode detection. The system uses position changes over time as a reference truth to validate whether MEMS sensor interpretations are correct, thereby resolving the ambiguity and improving accuracy without sacrificing the adaptability of mobility mode detection.
2Measurement precision
If continuous position data output is maintained, then location accuracy is preserved, but power consumption increases
Solution Approach 1:
The patent dynamically adjusts the position data output rate based on detected mobility mode states. When the device is stationary or in low-mobility modes, position updates are reduced or suspended. When high mobility is detected, the system increases update frequency to maintain accuracy. This dynamic adaptation resolves the contradiction by matching power consumption to actual positional change needs.
Solution Approach 2:
The system changes the output parameter (position update rate) based on mobility mode state. Different mobility modes correspond to different optimal output rates, allowing the system to optimize power consumption while maintaining sufficient position data accuracy for each specific mobility context.
3Productivity
If mobility mode detection is performed without verification, then processing speed is maintained, but detection reliability deteriorates due to wrong conclusions
Solution Approach 1:
The patent implements a feedback mechanism where GNSS position data serves as verification feedback for MEMS sensor-based mobility mode detection. The system continuously compares detected mobility states against position change evidence, correcting errors when discrepancies are found. This feedback loop maintains reliability without significantly impacting processing speed, as the verification operates on already-collected position data.
Data Source
AI summary
Systems and methods are provided for detecting a mobility mode of a mobile device. The method, in response to receiving an observation decision, comprises finding (e.g., recursively) the most probable mobility mode state of the wireless mobile device using a dynamic programming algorithm based on a Hidden Markov Model that comprises: calculating the probability for the observation by the knowledge of the observation and probability of the previous state using the transition probability and multiplying by the emission probability of observation for the state, obtaining the maximum probability for the detected mobility mode states, determining a mobility mode state with the maximum probability, and storing the mobility mode state data of the wireless mobile device in a given FIFO queue of the one or more FIFO queues. The method future comprises controlling the output of the position data by acquiring the satellite navigation data on an interval basis.