Progressive Peak Detection in Automotive Radar Cubes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automotive radar systems face memory constraints due to the form factor limitations, necessitating the use of expensive SRAM memory, and conventional processing methods require significant memory resources to handle large radar cube datasets, which is costly and inefficient.
Innovation Solution
The radar signal processing is decomposed into iterative partial steps, allowing progressive detection of peaks in subsections of the radar cube data, reducing memory requirements by processing and discarding each subsection after analysis, and combining candidate peaks into a final dataset using a threshold-based criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional processing methods are used to handle large radar cube datasets, then object detection capability is maintained, but memory requirements increase significantly
Solution Approach 1:
The patent divides the radar cube dataset into multiple smaller subsections that are processed iteratively. Each subsection is processed independently through the peak detection algorithm, and only the candidate peaks from each subsection are stored in memory, rather than storing the entire radar cube. This segmentation approach reduces memory requirements while maintaining detection capability.
Solution Approach 2:
The patent performs preliminary processing by detecting peaks in each subsection before combining results. The peak detection is performed on subsections as they become available, and candidate peaks are identified and stored temporarily. This preliminary action allows the system to process data in smaller chunks rather than requiring the entire dataset to be in memory simultaneously.
2Quantity of substance
If iterative processing of subsections is implemented, then memory footprint is reduced, but processing time may increase
Solution Approach 1:
The patent implements continuous processing where subsections are processed as they become available in a streaming manner. The peak detection algorithm operates continuously on incoming subsections without interruption, and candidate peaks are immediately stored and combined with previous results. This continuous approach minimizes idle time and ensures efficient use of processing resources while maintaining reduced memory footprint.
Solution Approach 2:
The patent processes radar data in periodic intervals by dividing the complete radar cube into multiple subsections. Each subsection is processed in turn through the peak detection algorithm, and the results are accumulated. This periodic processing approach allows the system to manage memory efficiently by only having one subsection in memory at a time while maintaining overall processing throughput.
3Measurement precision
If peak detection is performed on complete radar cubes, then detection accuracy is maximized, but memory resources are exhausted
Solution Approach 1:
The patent segments the radar cube into smaller subsections that can be processed individually. The peak detection algorithm is applied to each subsection separately, and candidate peaks are identified and stored. This segmentation allows the system to maintain detection accuracy by processing all subsections through the same algorithm while using significantly less memory resources.
Solution Approach 2:
The patent creates a simplified representation of the radar data by detecting and storing only the candidate peaks from each subsection rather than storing the complete radar cube. This copying approach retains the essential detection information while reducing memory requirements, as only the peak locations and characteristics need to be stored rather than the entire dataset.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
An automotive radar system includes at least one transmitter and at least one receiver and a processor configured to receive, from the at least one receiver, received radar signals, generate a first subsection of a radar cube using the received radar signals, detect a first set of candidate peaks in the first subsection of the radar cube, generate a second subsection of the radar cube using the received radar signals, detect a second set of candidate peaks in the second subsection of the radar cube, determine a set of locations in a candidate peak dataset, wherein each location in the set of locations is associated with candidate peaks in both the first set of candidate peaks and the second set of candidate peaks, and estimate a direction of arrival of an object using the candidate peaks associated with the set of locations.