High-Efficiency Super-Resolution FMCW Radar Algorithm Based on FFT Estimation

Area of Science:

• Radar Systems Engineering
• Signal Processing

Background:

• Frequency-modulated continuous-wave (FMCW) radar systems often use a fixed number of samples, determined by maximum detectable distance, which can be inefficient for closer targets.
• Conventional super-resolution algorithms, like MUSIC, require significant computational resources.

Purpose of the Study:

• To propose a high-efficiency super-resolution algorithm for FMCW radar that reduces computational complexity.
• To maintain or improve performance compared to existing super-resolution techniques.

Main Methods:

• Developed a novel FMCW radar algorithm utilizing Fast Fourier Transform (FFT) for coarse target range estimation.
• Adaptively selects a reduced number of samples based on FFT-estimated ranges for super-resolution processing.
• Replaces the maximum sample count with the reduced sample count for super-resolution input.

Main Results:

• The proposed algorithm achieves performance comparable to the Multiple Signal Classification (MUSIC) algorithm.
• Demonstrates an average complexity reduction of 88% compared to the conventional MUSIC algorithm.
• Validated through simulations and practical experimental results.

Conclusions:

• The proposed algorithm offers a significant reduction in computational complexity for FMCW radar super-resolution.
• It achieves similar high performance to conventional methods, making it suitable for practical applications.
• Adaptive sample selection is key to improving efficiency without compromising accuracy.