Indoor RF Localization Using Multipath Point-Cloud Fingerprints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radar systems struggle with multipath interference in indoor environments, leading to inaccurate localization of individuals due to multiple reflections from surfaces, which traditional tracking algorithms like Kalman filters often misinterpret as ghost objects, distorting size and contour.
Innovation Solution
A method utilizing multipath signals for fingerprinting, combining micro-location and user actions, employs machine learning algorithms like K-Nearest Neighbor (KNN) to create unique signatures for each location and action, enabling precise localization and activity detection by comparing point clouds and Doppler shifts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multipath signals are filtered out to prevent interference, then signal accuracy is improved, but location information is lost
Solution Approach 1:
The patent converts the previously harmful multipath signals into beneficial location fingerprints. By treating multipath reflections as informative signals rather than noise, the system creates unique signatures for different locations within the indoor space, enabling precise location determination while utilizing the full signal information.
Solution Approach 2:
Instead of filtering out multipath signals as traditionally done, the patent inverts the approach by retaining and utilizing these signals to create location fingerprints. This inversion allows the system to determine user location by comparing current signal patterns against stored fingerprints of known locations.
2Device complexity
If traditional tracking algorithms are used, then processing simplicity is maintained, but localization accuracy deteriorates due to ghost objects
Solution Approach 1:
The patent creates a database of stored signal fingerprints representing different locations and user actions. During operation, the system compares current signal measurements against these stored copies to identify matching patterns, enabling accurate localization without complex real-time processing algorithms.
Solution Approach 2:
The system performs preliminary training by collecting and storing signal fingerprints for various locations and actions before actual operation. This pre-computed database eliminates the need for complex real-time tracking algorithms, simplifying processing while maintaining high accuracy through pattern matching against stored references.
3Reliability
If multipath is treated as interference, then conventional radar performance is maintained, but indoor localization capability is lost
Solution Approach 1:
The patent enables the radar system to perform multiple functions simultaneously: traditional target detection and indoor location determination. By processing multipath signals through fingerprinting, the system adapts to indoor environments while maintaining conventional radar capabilities, achieving versatility without sacrificing reliability.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach provides high-fidelity, real-time mapping of individuals within indoor spaces, distinguishing between multiple users and pets, and detecting specific actions with high accuracy by leveraging multipath reflections as informative signatures.
Implementation Method 1
receiving backscattered training radiofrequency signals, including multipath
Implementation Method 2
converting the received training data into a point cloud for each location
Implementation Method 3
look at the Doppler signature of the point cloud to identify the action
Data Source
AI summary
In one aspect, a method of determining a location of a user within an indoor space, includes emitting a radiofrequency signal into the indoor space, receiving backscattered training radiofrequency signals, including multipath, for at least one location within the indoor space, converting the received training signals into a point cloud for each location of the at least one location, assigning a signature for each location based on the point cloud for each location, receiving additional radiofrequency signals, including multipath, converting the additional radiofrequency signals into an additional point cloud, and determining a location of the user by comparing the additional point cloud to the assigned signatures.