Virtual Transducer Array Scene Reconstruction via Joint Sparsity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing scene reconstruction systems using transducer arrays are sensitive to positioning errors, especially in virtual arrays where accurate relative positioning of transducers is difficult to achieve, leading to reduced reconstruction accuracy.
Innovation Solution
A method that reconstructs a scene by treating each configuration of a virtual array as a separate problem, sharing information to determine a common sparsity pattern across all configurations, allowing for accurate reconstruction despite positioning errors, using compressive sensing techniques such as greedy sparse recovery or model-based methods to combine solutions from multiple array instances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a virtual array of transducers is used to reduce the number of physical transducers, then device complexity and cost are reduced, but positioning errors increase leading to reduced measurement precision
Solution Approach 1:
The virtual array is segmented into multiple configurations or snapshots, where each configuration is processed separately through compressive sensing. This segmentation allows the system to handle positioning errors in each snapshot independently while reconstructing the complete scene through information sharing across all configurations.
Solution Approach 2:
The system changes the operational parameters by using compressive sensing reconstruction algorithms that can tolerate positioning errors. By transforming the reconstruction problem into a sparse signal recovery problem, the system maintains measurement precision despite variations in transducer positioning across different virtual array configurations.
2Measurement precision
If multiple pulse transmissions are used to scan different directions, then measurement precision improves, but loss of time increases
Solution Approach 1:
The system performs preliminary processing by decomposing each received pulse into frequency coefficients and organizing them into linear systems for each configuration. This preliminary organization enables parallel processing of multiple configurations, reducing the overall reconstruction time while maintaining accuracy.
Solution Approach 2:
The compressive sensing reconstruction process continuously shares information across all linear systems from different configurations, allowing the system to progressively build up the scene reconstruction without discrete interruptions. This continuous information sharing accelerates convergence while maintaining precision.
3Measurement precision
If a large number of transducers are used to form an effective array, then measurement precision improves, but device complexity increases
Solution Approach 1:
The system creates multiple virtual copies of the transducer array through different configurations or snapshots. Each virtual configuration acts as a copy that provides redundant measurement information, enabling accurate scene reconstruction with fewer physical transducers by leveraging the information from multiple virtual instances.
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 accurate scene reconstruction with fewer transducers and improved robustness to positioning errors, allowing for efficient object detection in applications like driver assistance and surveillance.
Implementation Method 1
preferred embodiments of the invention use a small number of transducers that transmit and receive wideband ultrasound pulses
Implementation Method 2
The scene reflects the pulses. The reflected pulses are received by a virtual array of transducers
Data Source
AI summary
A scene is reconstructed by receiving pulses transmitted by a virtual array of transducers. The pulses are reflected by the scene. The virtual array has a set of configurations subject to positioning errors. Each received pulse is sampled and decomposed to produce frequency coefficients stacked in a set of linear systems modeling a reflectivity of the scene. There is one linear system for each configuration. A reconstruction method is applied to the set of linear systems. The reconstruction method solves each linear system separately to obtain a corresponding solution. The corresponding solutions share information during the solving. Then, the solutions are combined to reconstruct the scene.