UWB Location Guided Image Classification for Wireless Device Labeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices face challenges in efficiently discovering and connecting to other wireless devices, particularly in environments with numerous devices, due to the lack of descriptive wireless identifiers.
Innovation Solution
The implementation of an electronic device equipped with ultra-wideband (UWB) locating capabilities and visual image classification, which allows for autonomous device labeling of UWB-capable devices by determining their location, receiving UWB signals, and visually classifying objects using an image stream.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If wireless devices broadcast their identifiers for discovery, then device discoverability is improved, but the complexity of identifying and connecting to a specific device increases due to non-descriptive identifiers
Solution Approach 1:
The patent introduces an intermediary system consisting of image capture devices and machine learning models that act as a mediator between wireless device identifiers and human understanding. The system captures images of devices, processes them through ML models, and generates descriptive labels that bridge the gap between technical identifiers and user comprehension, thereby simplifying device selection without compromising discoverability
Solution Approach 2:
The patent replaces the manual mechanical process of scanning and interpreting wireless identifiers with an automated optical system. Instead of users manually examining technical identifiers, the system uses image capture devices to photograph devices and machine learning models to automatically generate descriptive labels, substituting automated AI processing for manual identification tasks
2Measurement precision
If manual device identification from wireless addresses is required, then connection accuracy is improved, but the time required for device discovery and connection increases
Solution Approach 1:
The patent performs preliminary actions by capturing images of wireless devices and pre-processing them through machine learning models before the user needs to make a selection. The system proactively generates descriptive labels and presents multiple candidate labels in advance, allowing users to quickly identify devices without manually examining technical identifiers at the moment of connection need
Solution Approach 2:
The system performs self-service by automatically capturing device images, processing them through machine learning models, and generating descriptive labels without requiring user intervention. The electronic device autonomously manages the entire workflow from image capture to label presentation, reducing the time users need to spend on device identification while maintaining accurate identification
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 solution enables accurate and efficient device labeling, facilitating easy connection and tracking of UWB-capable devices, thereby simplifying the process of discovering and connecting to specific devices in a crowded wireless environment.
Implementation Method 1
The controller receives, via a UWB transceiver and an antenna assembly of the electronic device, UWB signals from a first device
Implementation Method 2
The antenna assembly includes two or more antennas spaced in a support structure as one of a two-dimensional (2D) or a three-dimensional (3D) antenna array
Implementation Method 3
The controller receives, from an image capturing device, an image stream depicting an area captured within a field of view that includes the location
Data Source
AI summary
An electronic device, a method, and a computer program product facilitate device labeling of devices that are ultra-wideband (UWB) capable. A controller of the electronic device identifies, via a wireless transceiver, proximity to device(s) including a first device that are UWB capable. The controller receives, via a UWB transceiver, UWB signals from a first device. The controller determines a location of the first device based on a range and an angle of arrival of the UWB signals. The controller receives an image stream from an image capturing device. The controller visually classifies an object corresponding to the first device in the image stream using a library of classification images that includes one or more classification images of one or more devices. The controller identifies an object classification associated with a device label for the first device. The controller assigns the device label to the first device.