24 July 2020 in IEEE Transactions on Instrumentation and Measurement
This paper presents a novel signal technique for Ku-band automobile frequency-modulated continuous-wave fully polarimetric synthetic aperture radar (FMCW PolSAR) system using triangular linear frequency modulation (LFM). Our proposed system shows the first utilizations of triangular LFM for a FMCW PolSAR. The proposed signal processing algorithm is based on the range Doppler algorithm (RDA). We developed a FMCW PolSAR system that transmits triangular LFM signals, which are used less frequently than sawtooth LFM in a SAR sensor. Using a theoretical background, we describe its configuration and how it works. We propose the novel processing solution, which forms two kinds of single-polarization images from a raw data set and is suitable for our system. We obtained all four kinds of single-polarization images from two raw data sets while using the triangular LFM. In comparison, when using sawtooth LFM, we obtained the four images from four raw data sets by repeating the RDA four times. The proposed method simplifies the FMCW PolSAR system configuration, and the processing algorithm. We collected FMCW PolSAR raw data from an experimentally equipped automobile while maintaining a constant speed on a highway. The proposed algorithm and system were validated by processing a high-resolution FMCW PolSAR image.
2 July 2020 in IEEE Transactions on Signal Processing
The signal processing technologies of synthetic aperture radar (SAR) have been vastly studied as diverse measurement modes for various applications. One of the technologies is Video-SAR (ViSAR). It is a land-imaging mode where a sequence of images is continuously formed on the spotlight mode. We propose a new signal processing method to generate the multi-frames of ViSAR using Doppler shifting technique via sub-aperture processing. The method is based on our designed wide-angle antenna, which is a corrugated horn shape with low side lobe and Gaussian beam pattern. This wide-angle antenna receives a wide Doppler band along the azimuth direction. Based on the characteristics of the antenna, we separate the wide Doppler band into each chirp pulse signal. Each of these separated signals has a different center time. Through each chirp pulse signal in Doppler domain, the proposed method generates each frame of video at a high frame rate. The proposed ViSAR processing can show high efficiency with low computation and without a fixed coordinate system. Moreover, it can also visualize a wider area on stripmap mode than the conventional ViSAR mode, and can derive the video results based on Range Doppler Algorithm (RDA). These advantages can lead to reduce economic costs and simplify the operating ViSAR system. We have performed the practical experiments using X-band chirp pulse SAR system mounted on an airplane to verify the proposed method.
7 May 2019 in IEEE Microwave and Wireless Components Letters
Our laboratory has developed a drone detection radar that involves a mechanical scanning antenna. Radar and antenna systems that have mechanical scanning require an additional component called a rotary joint (RJ). The purpose of RJ in scanning assemblies is to provide an uninterrupted link between the antenna and the transceiver. In our system, antenna and transceiver are linearly polarized and use TE₁₀ mode along the rectangular waveguide section. TE₁₀ modes are asymmetric in nature making them incompatible with the rotating motion of the antenna. An RJ is presented with back-to-back mode transitions using symmetric modes along the circular waveguide and contactless branch line joint. The mode used along the circular waveguide for symmetry purpose is TM₀1. An I-type linear RJ topology is selected due to lack of space along the lateral directions and system configuration. The problem of antenna matching while scanning was solved at 14.0-14.5 GHz at Ku-band using single-channel linear RJ.
11 July 2024 in IEEE Sensors Journal
Frequency-modulated continuous wave (FMCW) radar sensors are widely used to measure range, velocity, and target status. As the range of applications for FMCW radar sensors expands, there is an increasing demand for smaller FMCW radar sensors suitable for use in confined spaces. To reduce the form factor of the FMCW radar sensor, the distance between the transmit (Tx) and receive (Rx) antennas is also reduced. This decrease in antenna separation increases the magnitude of the leakage signal transmitted from the Tx to the Rx antenna. However, the leakage signal makes it difficult for the FMCW radar sensor to detect the target by increasing the noise floor. In this article, we present an advanced method to mitigate the increased leakage signal power in the miniaturized FMCW radar sensor. We have fabricated a miniaturized homodyne FMCW radar with consistent leakage signal characteristics. We also propose a method to minimize the influence of the leakage signal by storing the leakage signal characteristics and using the stored data when operating the radar. Experimental results show significant leakage signal power reduction. This technique not only preserves the power of the received target signals but also improves the overall performance of compact FMCW radar sensors. The proposed method provides a basis for miniaturized radar sensors to contribute to various industries by providing an effective solution to the leakage signal problem of small radar technology.
12 January 2026 in IEEE Sensors Journal
This paper presents a projection-based interference suppression method for FMCW radar sensors under single-tone continuous wave jamming. The proposed approach operates in the baseband domain by exploiting the orthogonality of I and Q channels. First, a CFAR-based thresholding process localizes jamming-contaminated segments, from which the dominant phase trajectory is estimated in the IQ polar domain. Leveraging this trajectory, the signal is phase-rotated to concentrate the jamming energy into a single channel (I or Q). Subsequently, variance analysis identifies the interference-minimized “optimum channel”, to which FFT processing is exclusively applied. Experimental validation using a 24.6 GHz jamming source demonstrates that the proposed method reduces the noise floor by up to 16 dB, effectively recovering target detectability degraded by interference. These results confirm the method’s computational efficiency and robustness, thereby improving detection reliability in spectrally congested environments.