Automotive Radar Sparse-Array DOA Estimation With FFT Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional BLRC-based DOA estimation algorithms for automotive radar systems require substantial memory and computing power, limiting their applicability to small-sized arrays and incompatible with modern vehicle radar systems that demand high angular resolution.
Innovation Solution
A novel implementation of the BLRC algorithm, referred to as fastBLRC, reduces memory and computer requirements by using Fast Fourier Transform (FFT) equivalent operations and inverting a smaller N × N matrix, rather than the larger M × M matrix, thereby achieving reduced complexity and memory footprint.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional BLRC-based DOA estimation algorithms are used, then measurement precision is improved, but device complexity increases due to substantial memory and computing power requirements
Solution Approach 1:
The patent transforms the original BLRC algorithm parameters and computational structure by using FFT-based operations instead of direct matrix operations. This changes the computational approach from O(M³) complexity to O(M log M), significantly reducing the computational burden while preserving the algorithm's ability to achieve high angular resolution in DOA estimation
Solution Approach 2:
The patent replaces the conventional matrix inversion and multiplication operations with Fast Fourier Transform operations. This substitution leverages the efficiency of FFT algorithms to perform the same mathematical transformations required for BLRC, thereby reducing the computational complexity and memory requirements while maintaining measurement precision
2Measurement precision
If conventional BLRC-based DOA estimation algorithms are used, then measurement precision is improved, but productivity decreases due to high computational overhead
Solution Approach 1:
The patent modifies the computational parameters of the BLRC algorithm by replacing iterative matrix operations with closed-form FFT-based solutions. This parameter change reduces the number of computational steps required, thereby increasing processing speed and productivity while maintaining the algorithm's precision in DOA estimation
Solution Approach 2:
The patent skips the computationally intensive intermediate steps of conventional matrix inversion by directly applying FFT operations to achieve the same result. This allows the algorithm to rush through the computational process more efficiently, reducing processing time and improving productivity without sacrificing measurement precision
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method for processing received radar signals is presented. A steering matrix for the radar system is determined. A measurement vector is determined using signals received by a plurality of receiver modules. An output amplitude vector is initialized using the steering matrix and the measurement vector. An optimized output amplitude vector is determined by determining a diagonal loading vector using the output amplitude vector, calculating a weighting matrix using a first fast Fourier transform of the diagonal loading vector, determining an inverse matrix using the weighting matrix and an identity matrix, and executing a second fast Fourier transform using the inverse matrix to determine the optimized output amplitude vector, wherein the optimized output amplitude vector is determined by a relationship between the steering matrix and the measurement vector. An objects direction of arrival is determined by correlating the optimized output amplitude vector to the measurement vector.