Wireless Power Sensing With Neural Distance Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless power transmission systems struggle to accurately determine the distance of wireless receivers, which is crucial for controlling power levels and avoiding unwanted absorption of RF power signals.
Innovation Solution
A neural network framework, specifically a convolutional neural network, is used to estimate distance information by processing phase data from reflected beacon signals, allowing for real-time control of wireless power output.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If phase-based information from reflected signals is used to determine transmission paths, then the system can avoid unwanted absorption of RF power signals, but the distance estimation accuracy is insufficient for precise power level control
Solution Approach 1:
The patent introduces a neural network framework as an intermediary between the raw phase-based signal information and the distance estimation output. This neural network processes the reflected signal phases through multiple layers (convolutional layers, fully connected layers) to transform the phase information into accurate distance estimates, resolving the insufficiency of direct phase-based distance calculation while maintaining path determination reliability
Solution Approach 2:
The patent replaces traditional mechanical or mathematical signal processing methods with a neural network-based computational system. Instead of using conventional algorithms to interpret phase information, the system employs trained neural networks that have learned optimal transformation patterns from training data, achieving superior distance estimation accuracy without sacrificing the reliability of transmission path identification
2Length of stationary object
If wireless power transmission is performed over relatively long distances using RF signals, then power can be delivered to remote devices, but stray power may be transmitted causing energy loss and safety concerns
Solution Approach 1:
The patent implements a feedback mechanism where the neural network continuously estimates the distance to the wireless power receiver based on reflected signal phases, and this distance information is fed back to control the power transmission level. The system adjusts the transmitted power dynamically based on the estimated distance, reducing stray power transmission when the receiver is far away while maintaining adequate power delivery when the receiver is closer, thus resolving the energy loss problem while preserving long-distance transmission capability
Solution Approach 2:
The patent makes the power transmission system dynamic by enabling real-time adjustment of transmission parameters based on neural network distance estimates. Instead of fixed power levels, the system dynamically adapts the transmitted power according to the current distance to the receiver, optimizing energy efficiency while maintaining the ability to transmit over varying distances as needed
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 enables precise distance estimation and control of wireless power transmission, reducing stray power and ensuring safe and efficient energy delivery.
Implementation Method 1
Directed antenna systems can transmit wireless power over radio frequency (RF) signals to various devices coupled to a wireless power receiver
Implementation Method 2
The wireless power receiver comprises an antenna (one or more antenna elements) and circuitry that converts the received RF signals into power
Implementation Method 3
the wireless transmission device transmits a wireless beacon signal that is reflected back to wireless transceivers
Data Source
AI summary
Various wireless power transmission systems are provided for sensing an environment, e.g., using a neural network. For instance, phase information corresponding to wireless power transmission is input into a neural network framework, and then, distance information representative of a distance from a wireless power transmitter to an object is obtained as output from the neural network framework. Based on the distance information, a power of a subsequent wireless power transmission can be modified, or an environment comprising the object can be mapped.