Frequency Estimation Using Complex-Valued Shifted Window Transformer

TL;DR

This paper introduces novel complex-valued and real-valued shifted window transformer models, SwinFreq and CVSwinFreq, for high-precision frequency estimation of complex signals, outperforming classical and deep learning methods especially at low SNR.

Contribution

It presents the first complex-valued Swin transformer module for frequency estimation, demonstrating superior performance and efficiency over existing algorithms and deep learning approaches.

Findings

SwinFreq outperforms CVSwinFreq in several tasks.

Both models surpass traditional algorithms like MUSIC and OMP.

Models are effective for radar super-resolution on real data.

Abstract

Estimating closely spaced frequency components of a signal is a fundamental problem in statistical signal processing. In this letter, we introduce 1-D real-valued and complex-valued shifted window (Swin) transformers, referred to as SwinFreq and CVSwinFreq, respectively, for line-spectra frequency estimation on 1-D complex-valued signals. Whereas 2-D Swin transformer-based models have gained traction for optical image super-resolution, we introduce for the first time a complex-valued Swin module designed to leverage the complex-valued nature of signals for a wide array of applications. The proposed approach overcomes the limitations of the classical algorithms such as the periodogram, MUSIC, and OMP in addition to state-of-the-art deep learning approach cResFreq. SwinFreq and CVSwinFreq boast superior performance at low signal-to-noise ratio SNR and improved resolution capability while requiring fewer model parameters than cResFreq, thus deeming it more suitable for edge and mobile applications. We find that the real-valued Swin-Freq outperforms its complex-valued counterpart CVSwinFreq for several tasks while touting a smaller model size. Finally, we apply the proposed techniques for radar range profile super-resolution using real data. The results from both synthetic and real experimentation validate the numerical and empirical superiority of SwinFreq and CVSwinFreq to the state-of-the-art deep learning-based techniques and traditional frequency estimation algorithms. The code and models are publicly available at https://github.com/josiahwsmith10/spectral-super-resolution-swin.