Radar Processor Using 1D Convolutions for 2D CA-CFAR
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional 2D CFAR algorithms require significant processing operations and 2D convolution capabilities, which are not feasible in systems with limited computational resources, such as IoT, automotive, and mobile applications.
Innovation Solution
A method to perform 2D CFAR radar processing using one-dimensional convolutions by spatially separating 2D CA-CFAR radar processing kernels into pairs of one-dimensional vectors and applying superposition principles to reduce the number of operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional 2D CFAR algorithms are used, then CFAR processing capability is achieved, but processing complexity and computational operations increase significantly
Solution Approach 1:
The patent segments the 2D CFAR processing into multiple 1D convolution operations. Instead of performing a single complex 2D convolution, the kernel is decomposed and applied through sequential 1D convolutions along different dimensions, reducing the computational burden while maintaining the same CFAR detection capability.
Solution Approach 2:
The patent transforms the 2D convolution operation into a series of 1D convolution operations by applying convolutions along rows and columns separately. This dimensionality reduction allows the system to achieve 2D CFAR processing using only 1D convolution capabilities, suitable for resource-constrained devices.
2Reliability
If conventional 2D CFAR algorithms are used, then CFAR processing is performed, but memory requirements increase
Solution Approach 1:
The patent segments the 2D kernel into multiple 1D kernels that can be stored and applied separately. This segmentation reduces the memory footprint by storing smaller 1D kernel vectors instead of a large 2D kernel matrix, while still enabling complete 2D CFAR processing through sequential application.
3Measurement precision
If 2D convolution capabilities are required, then accurate CFAR processing is achieved, but system requirements and computational resources increase
Solution Approach 1:
The patent transforms the 2D convolution operation into a series of 1D convolution operations by applying convolutions along rows and columns separately. This dimensionality reduction allows the system to achieve 2D CFAR processing using only 1D convolution capabilities, suitable for resource-constrained devices.
Data Source
AI summary
A cell-averaging constant false alarm rate (CA-CFAR) radar processor configured to: receive a two-dimensional radar data array; for each of a plurality of pairs of one-dimensional vectors: perform a convolution between the two-dimensional radar data array and a first one-dimensional vector of the pair of one-dimensional vectors to obtain an intermediate convolution result; and perform a convolution between the intermediate convolution result and a second one-dimensional vector of the pair of one-dimensional vectors to obtain an intermediate CA-CFAR result; and combine the intermediate CA-CFAR results to provide a CA-CFAR output.