Radar Processing With Haar Wavelets for Data Retention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radar signal processing methods in automotive radar systems discard a significant amount of information, leading to sparse data representation, despite advancements in hardware capabilities that allow for capturing large amounts of data, and require inefficient handling of large 3D arrays to minimize computing power and memory.
Innovation Solution
A method involving a hierarchical filter bank and discrete wavelet transform is applied to radar signals, allowing for efficient extraction of Doppler information with higher time resolution and retention of data, using a Haar wavelet transform to simplify processing and reduce computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If classical radar processing methods are used to handle large 3D arrays, then computing power and memory requirements are minimized, but a significant amount of information is discarded leading to sparse data representation
Solution Approach 1:
The patent applies segmentation by dividing the 3D radar data array into multiple 2D slices along the range dimension. Each slice corresponds to a specific range bin and contains Doppler and angle information. This segmentation allows processing of smaller 2D arrays instead of large 3D arrays, reducing computational complexity while retaining information through the slice-based representation of the complete radar scene.
Solution Approach 2:
The patent transforms the problem from 3D array processing to 2D array processing by slicing the 3D data along the range dimension. This dimensional reduction enables efficient processing while maintaining information integrity, as each 2D slice preserves the essential Doppler and angular information for objects at a specific range.
2Productivity
If classical radar processing reduces large maps to point targets, then detection lists are generated, but a large amount of input radar data is processed to produce a relatively small amount of usable information
Solution Approach 1:
The patent applies local quality by processing each 2D slice independently with appropriate processing parameters tailored to that slice's content. Each slice can be processed to extract local features and characteristics specific to objects at that range, preserving detailed information about each local region rather than reducing everything to uniform point targets.
Solution Approach 2:
The patent performs preliminary actions by first organizing radar data into 2D slices before applying detection algorithms. This preliminary organization preserves the structural information of the data and enables subsequent processing to maintain both detection efficiency and information retention by working with the pre-organized 2D representations.
3Measurement precision
If image-like processing of radar data is performed, then denser perception of the environment is achieved, but handling of large sized 3D arrays requires significant computing power and memory
Solution Approach 1:
The patent segments the 3D radar data into multiple 2D slices, each representing a specific range bin. This segmentation enables image-like processing of each slice with high spatial resolution in the angle-Doppler plane, while the computational burden is distributed across smaller 2D arrays rather than concentrated in a single large 3D array processing operation.
Solution Approach 2:
By transforming the processing task from 3D to 2D through slicing, the patent achieves image-like representation capability with reduced computational requirements. The 2D slices can be processed using efficient image processing algorithms that provide dense spatial perception without the full computational burden of 3D array processing.
Data Source
Figure 1
Figure 2
Figure 3a~3b
AI summary
Method and apparatus for processing radar signals. A plurality of radio waves reflected from at least one object are received (201) at an antenna array. Each of the plurality of reflected radio waves are processed (203) to determine the range from the antenna array to the at least one object. Each of the plurality of processed radio waves are adjusted (205) based on an azimuth value corresponding to an angle of incidence of the reflected radio wave, and the motion of the antenna array. Each of the plurality of adjusted processed radio waves are filtered (207) through a hierarchical filter bank comprising a plurality of filters to receive a plurality of output images representing the motion of the object.