This paper investigates two approaches for real-time object detection and classification using Frequency-Modulated Continuous-Wave (FMCW) radar data from the Radical Dataset. The first method uses the Constant False Alarm Rate (CFAR) algorithm and manual annotations to extract patches from Range-Azimuth (RA) heatmaps, which are then processed by a simple Convolutional Neural Network (CNN) named RADICAL-CNN. The second approach evaluates the performance of the RAMP-CNN model on the Radical Dataset, which incorporates RA, Range-Doppler (RD) and Doppler-Azimuth (DA) data and utilizes 3D convolution to integrate temporal information. This comparative analysis of two innovative neural network architectures not only highlights their strengths and weaknesses, but also contributes to the ongoing optimization of object detection techniques for FMCW radar data in dynamic, real-world scenarios such as autonomous driving, robotics, zone monitoring and safety.
Widely used in automotive safety, industrial automation, and surveillance, Frequency-Modulated Continuous-Wave (FMCW) radar provides accurate object tracking (OT). Kalman filter (KF)-based algorithms are proven to work reliably in industry for OT with FMCW radar. Extending the KF functionality, Probabilistic Data Association Filter (PDAF) is a method that also enables the association of detected reflections with the observed target. Integrated PDAF (IPDAF) offers formulations for both the probability of track existence and data association simultaneously. This paper shows how parameters for IPDAF tracker can be selected to optimize the First, FMCW radar is simulated to generate precise distance, velocity, and angle measurements through a multiple-input multiple-output (MIMO) antenna setup providing ground-truth labeled data. Next, a practical method for calculating optimal tracking parameters (OTP), particularly for short-range applications, based on Optuna optimizer is introduced. This method aims to improve tracking accuracy in scenarios such as indoor environments and pedestrian safety systems, considering scenarios with different target movement maneuvers and different dynamic movement models.
Accurate and reliable vital signs monitoring has the potential to detect different health conditions, driver fatigue, or even depression. Contactless approaches for heartbeat and respiration rate monitoring are more comfortable than conventional contact-based monitoring devices. In this paper, heartbeat and breathing data were estimated using Frequency-Modulated Continuous-Wave (FMCW) radar (AWR1843, Texas Instruments, USA) placed underneath the bed on which the participant was lying on the back for 5 minutes. The Wearable Wellness System (Smartex, Italy) was used to obtain the reference signal for breathing and heart rate. FMCW data from three participants were acquired during the resting state scenario. First, a slow-time phase correlation method was applied to FMCW data to increase the signal-to-noise ratio. Subsequently, an Adaptive Nonlinear Least Square (ANLS) framework was used to explore the frequency characteristics of the phase signal. Additionally, the Adaptive Notch filter (ANF) removed the second and third breathing harmonics. Finally, the Kalman filter (KF) was applied for better heart rate accuracy. The average root mean square error (RMSE) for estimation from 10-second time windows was 1.66 BPM for the breathing rate estimation and 1.61 BPM for the heart rate estimation.
Numerous Frequency-Modulated Continuous-Wave (FMCW) radar applications require Angle Of Arrival (AOA) estimation through the use of antenna arrays. Due to manufacturing tolerances, the antennas not being ideal, and interference between the antennas, calibration is needed to achieve theoretical angular resolution. The most commonly used method is Amplitude and Phase Compensation (APC) which compensates the amplitudes and phases at reference angle. However, since this method does not take into account Mutual Coupling (MC) between the antennas, performance benefits of Mutual Coupling Compensation (MCC) algorithm over conventional APC are investigated. A model for MC is presented and two methods for MCC are evaluated. A dataset was created using an automated measurement system that sweeps through a range of angles to record the required measurement set needed for coefficient calculation. Radar corner reflector at fixed distance was used as a reference target. The results of the three methods (APC and two MCC) were evaluated for different positions of the reflector and different boards to determine performance relative to ideal theoretical response.