Wireless Tracking via Graph-Based Particle Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current motion tracking systems, particularly those using Inertial Measurement Units (IMUs) and radio signals, face significant errors and limitations in accurately measuring moving distance, heading direction, and rotating angle, especially in rich-scattering environments, which hinders applications like indoor tracking and virtual reality.
Innovation Solution
A wireless monitoring system that transmits probe signals through a wireless multipath channel, allowing a processor to extract time series of channel information (TSCI) and use this data to monitor an object's motion relative to a map, determining incremental distance and computing the object's next location based on its current location and motion direction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Inertial Measurement Units (IMUs) are used for motion tracking, then motion measurements can be obtained, but significant errors and drifts occur in measuring moving distance, heading direction, and rotating angle
Solution Approach 1:
The patent introduces wireless channel information as an intermediary medium to bridge the gap between radio-based positioning and inertial measurement. By extracting channel information (multipath components, time delays, phase rotations) from wireless signals and using graph-based particle filtering to fuse this with motion models, the system achieves accurate motion tracking without relying solely on drift-prone IMUs or requiring precise AP installations. The wireless channel acts as a mediator that provides geometric information about the environment and object motion.
2Measurement precision
If radio signal-based tracking systems are used, then location can be determined, but significant errors occur and the systems cannot track in-place angular motion or multiple motion parameters simultaneously
Solution Approach 1:
The patent makes the wireless monitoring system multi-functional by enabling it to simultaneously track multiple motion parameters (location, moving distance, heading direction, rotating angle) and detect various object states (presence, motion, rotation) using a single system architecture. The graph-based particle filtering algorithm processes wireless channel information to extract multiple motion parameters from successive location estimates, allowing the system to serve diverse tracking needs without requiring separate specialized systems for each function.
3Measurement precision
If traditional radio-based tracking systems are used, then location estimation can be performed, but the systems require precisely installed Access Points and accurate information about their locations and orientations
Solution Approach 1:
The patent replaces the requirement for expensive, precisely installed Access Points with ordinary wireless devices that can be easily deployed. Instead of requiring carefully positioned infrastructure equipment, the system uses standard wireless communications to extract channel information that inherently contains geometric information about the environment. This eliminates the need for complex installation procedures while maintaining tracking accuracy.
4Measurement precision
If radio-based tracking systems are used, then location can be determined, but the systems face accuracy limitations dictated by frequency bandwidth, antenna amount, and synchronization errors
Solution Approach 1:
The patent changes the approach by extracting multiple geometric parameters (time delays, phase rotations, multipath components) from the wireless channel rather than relying on traditional radio-based location estimation methods. By analyzing these channel parameters over time and using graph-based particle filtering to fuse them with motion models, the system achieves higher tracking accuracy without increasing the number of antennas or requiring complex synchronization protocols.
Data Source
AI summary
Methods, apparatus and systems for wireless tracking with graph-based particle filtering are described. A described wireless monitoring system comprises a transmitter transmitting a series of probe signals, a receiver, and a processor. The receiver is configured for: receiving the series of probe signals modulated by the wireless multipath channel and an object moving in a venue, and obtaining a time series of channel information (TSCI) of the wireless multipath channel from the series of probe signals. The processor is configured for: monitoring a motion of the object relative to a map based on the TSCI, determining an incremental distance travelled by the object in an incremental time period based on the TSCI, and computing a next location of the object at a next time in the map based on at least one of: a current location of the object at a current time, the incremental distance, and a direction of the motion during the incremental time period.