Tomographic Imaging Antenna Calibration for Multiple Scattering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing tomographic imaging systems face challenges in reconstructing the internal structure of objects due to multiple scattering of signals, leading to noisy and inaccurate images, especially when using extended-source antennas, which are complex to analyze and design.
Innovation Solution
A tomographic imaging system utilizing a neural network that synthesizes measurements for extended-source antennas by first simulating point-source antennas and then calibrating them to extended-source antennas, simplifying training and improving image reconstruction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If extended-source antennas are used for tomographic imaging, then the practical applicability and radiation characteristics are improved, but the device complexity and analysis difficulty increase
Solution Approach 1:
The patent introduces a calibration neural network as an intermediary component that bridges the simplified point-source antenna model and the complex extended-source antenna. The calibration network takes point-source measurements and transforms them into extended-source antenna responses, allowing the system to use simple theoretical models while achieving accurate practical imaging with real antennas
Solution Approach 2:
The patent creates a virtual copy of the extended-source antenna response through the calibration neural network. Instead of directly measuring and analyzing complex extended-source antenna data, the system synthesizes equivalent responses from simple point-source measurements through the calibration model, simplifying the analysis while preserving the practical antenna characteristics
2Measurement precision
If multiple scattering effects are accounted for in image reconstruction, then the measurement precision is improved, but the computational complexity and reconstruction difficulty increase
Solution Approach 1:
The patent replaces the traditional mechanical/numerical iterative reconstruction process with a neural network-based approach. The neural network operator learns the complex inverse scattering problem including multiple scattering effects during training, and then performs rapid reconstruction by simply processing measurements through the trained network, substituting complex computational mechanics with learned patterns
Solution Approach 2:
The patent performs preliminary action by training the neural network operator in advance using simulated data that includes multiple scattering effects. This pre-trained operator then handles multiple scattering corrections during actual reconstruction without requiring complex real-time computations, as the network has already learned the scattering patterns during offline training
3Ease of operation
If point-source antenna models are used for simplification, then the analysis ease is improved, but the manufacturing precision and radiation pattern accuracy deteriorate
Solution Approach 1:
The calibration neural network serves as an intermediary that corrects the simplified point-source antenna model to match the actual extended-source antenna radiation characteristics. The network learns the differences between point-source and extended-source responses and applies corrections to achieve accurate radiation patterns while maintaining the analytical simplicity of point-source models
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
The system effectively reconstructs the internal structure of objects with enhanced precision by recursively updating images until synthesized measurements match actual measurements, overcoming the complexity of extended-source antennas and multiple scattering issues.
Implementation Method 1
a transmitter emits a signal such as an electromagnetic (EM), light, or acoustic pulse, which propagates through the object
Implementation Method 2
a deep learning operator that models the physics of wave propagation
Implementation Method 3
a deep learning operator that calibrates the received waveform to a desired antenna configuration
Data Source
AI summary
A tomographic imaging system including an extended source antenna, which produces a wavefield scattered by the internal structure of an object. A processor recursively reconstructs the internal structure by processing a current image of the internal structure of the object with a neural network operator trained to synthesize measurements of a point-source antenna corresponding to a wavefield scattered by the current image of the internal structure of the object, processing the synthesized measurements of the point-source antenna with a calibration neural network to estimate measurements of the extended-source antenna, and updating the current image of the internal structure of the object based on a difference between the measurements of the extended-source antenna and the estimation of the measurements produced by the calibration neural network.