Scholarlink

<p>Addressing the dual challenges of high computational complexity in traditional algorithms and spectral leakage in discrete Fourier transform (DFT)-based methods for single-snapshot direction of arrival (DOA) estimation in massive uniform linear array (ULA), this study proposes a rapid, high-accuracy DOA estimation algorithm that leverages the enhanced all-phase fast Fourier transform (ApFFT). The process begins with an initial estimation using enhanced ApFFT, which provides preliminary DOA estimates with higher accuracy compared to traditional algorithms based on DFT. Then, a refinement process involving a Taylor expansion of the steering vector and Newton iteration for angle offset adjustment follows. The initial estimation based on the enhanced ApFFT confines the estimated angles within a tighter range around the true DOA, enabling the optimization phase to yield exact estimates quickly. Simulation results confirm the superior performance of our algorithm over existing DFT-based methods and Root-MUSIC algorithm.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Efficient and Accurate Single-Snapshot DOA Estimation for Massive ULA via Enhanced ApFFT

  • 期刊论文

  • 发表日期：

    2025年3月12日

查看全文
  • Liangang Qi,
  • Jiakang He,
  • Qiang Guo,
  • Lichao Liu,
  • Mykola Kaliuzhnyi

摘要

Addressing the dual challenges of high computational complexity in traditional algorithms and spectral leakage in discrete Fourier transform (DFT)-based methods for single-snapshot direction of arrival (DOA) estimation in massive uniform linear array (ULA), this study proposes a rapid, high-accuracy DOA estimation algorithm that leverages the enhanced all-phase fast Fourier transform (ApFFT). The process begins with an initial estimation using enhanced ApFFT, which provides preliminary DOA estimates with higher accuracy compared to traditional algorithms based on DFT. Then, a refinement process involving a Taylor expansion of the steering vector and Newton iteration for angle offset adjustment follows. The initial estimation based on the enhanced ApFFT confines the estimated angles within a tighter range around the true DOA, enabling the optimization phase to yield exact estimates quickly. Simulation results confirm the superior performance of our algorithm over existing DFT-based methods and Root-MUSIC algorithm.