Vector Processor SAR Imaging with Accelerated Memory Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional synthetic aperture radar (SAR) systems for automotive applications face challenges due to the computational complexity and cost associated with high-resolution SAR imaging, particularly in achieving efficient processing without significant degradation in performance.
Innovation Solution
The implementation of high-performance and high-efficiency oversampled FFT processes without zero-padding, combined with streamlined range computation and vector processor engines with accelerators, reduces processing load and improves image quality in SAR systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional back projection SAR processing with DFT computations is used to achieve high-resolution imaging, then image quality is improved, but computational complexity and cost increase significantly
Solution Approach 1:
The patent changes the computational parameters by replacing DFT with FFT algorithms, transforming the mathematical approach from O(N²) to O(N log N) complexity. This parameter change in the processing algorithm maintains high-resolution imaging capability while dramatically reducing computational burden and system cost
Solution Approach 2:
The patent substitutes the mechanical/computational process of back projection with an interferometric measurement system. Instead of computationally intensive DFT operations on collected data, the system uses hardware-based interferometric processing that achieves the same imaging resolution with significantly reduced computational complexity
2Measurement precision
If zero-padding is applied to increase FFT oversampling for better image quality, then quantization error is reduced, but processing time and computational load increase
Solution Approach 1:
The patent applies preliminary windowing functions (such as Hamming or Hanning windows) to the input data before FFT processing. This preliminary action pre-conditiones the data to reduce spectral leakage and minimize quantization errors, achieving better image quality without requiring zero-padding that would increase processing time
Solution Approach 2:
The patent optimizes the FFT oversampling parameter by determining the minimum necessary oversampling factor based on the specific imaging requirements and signal characteristics. This parameter optimization avoids excessive zero-padding while still achieving adequate quantization accuracy, thereby reducing processing time without significant degradation in image quality
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
Embodiments are disclosed that for synthetic aperture radar (SAR) systems and methods that process radar image data to generate radar images using vector processor engines, such as single-instruction-multiple-data (SIMD) processor engines. The vector processor engines can be further augmented with accelerators that vectorize element selection thereby expediting memory accesses required for interpolation operations performed by the vector processor engines.