Key points are not available for this paper at this time.
Millimeter-wave (mmWave) radar has been deemed a key enabling technology for Internet of Things (IoT) applications and offers robust and non-invasive sensing capabilities of intelligent systems. However, conventional Fourier transform (FT)-based range estimation methods suffer from limited resolution, which can restrict performance, especially in dynamic IoT scenarios. To address this, we propose a fractional Fourier transform (FrFT)-based range estimation scheme that achieves higher resolution and accuracy. The proposed method incorporates a recursive clutter removal mechanism to enhance signal quality and employs a random forest regressor to dynamically determine the optimal FrFT angle parameter α for improved performance. Experimental results demonstrate that our proposed new FrFT-based approach achieves a mean absolute percentage error (MAPE) of 16.67%, which significantly outperforms the conventional FT-based method with a MAPE of 37.13%. Furthermore, we validate the effectiveness of our new approach by presenting its application to human height measurement to showcase its potential for real-world applications.
Fang et al. (Mon,) studied this question.