Wireless Sensing Calibration to Distinguish Motion From Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless sensing technologies face challenges in accurately distinguishing between environmental changes caused by motion and noise, leading to unreliable detection of events such as motion in a room.
Innovation Solution
A calibration scheme is implemented to estimate statistics of sensing noise and set detection thresholds by using multiple channel estimation measurements with different time intervals, allowing differentiation between noise and actual environmental changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If wireless sensing is used to detect motion in the environment, then motion detection capability is provided, but reliability of detection is reduced due to inability to distinguish between noise and actual motion
Solution Approach 1:
The system performs preliminary calibration by collecting channel estimation measurements during a calibration period when no motion is present. This preliminary action establishes a baseline of environmental noise characteristics before actual motion detection begins, enabling the system to later distinguish between noise and genuine motion events.
Solution Approach 2:
The system uses feedback from multiple channel estimation measurements taken at different time intervals to continuously refine noise statistics. By comparing new measurements against the established noise model and adjusting detection thresholds based on observed noise patterns, the system improves its ability to reliably detect motion while filtering out false positives caused by environmental noise.
2Measurement precision
If multiple channel estimation measurements are taken with different time intervals, then noise statistics can be estimated accurately, but time consumption increases
Solution Approach 1:
The system takes a limited number of channel estimation measurements during calibration (e.g., 5-10 measurements) rather than continuously collecting data. This partial action provides sufficient statistical information to characterize noise patterns without requiring excessive time, balancing measurement precision with time efficiency.
Solution Approach 2:
Channel estimation measurements are taken periodically at specific time intervals during calibration rather than continuously. This periodic sampling approach captures the temporal characteristics of noise while minimizing the total calibration time required to establish accurate noise statistics.
Data Source
AI summary
For example, a first device may calibrate a reference channel estimation based on a plurality of first channel estimation measurements, the plurality of first channel estimation measurements corresponding to first PPDUs received from a second device over a wireless channel, wherein two consecutive channel estimation measurements of the plurality of first channel estimation measurements are separated by no more than a first time interval; and determine a plurality of second channel estimation measurements for detection of a change in an environment of the wireless channel based on the reference channel estimation, wherein the plurality of second channel estimation measurements corresponds to a plurality of second PPDUs received from the second device over the wireless channel, wherein two consecutive channel estimation measurements of the plurality of second channel estimation measurements are separated by at least a second time interval, the second time interval is longer than the first time interval.