Cardiovascular diseases claim over 10 million lives annually, highlighting the critical need for long-term monitoring and early detection of cardiac abnormalities. Existing techniques like electrocardiograms (ECG) and Holter are accurate but suffer from discomfort caused by body-attached electrodes. While wearable devices using photoplethysmography offer more convenience, they sacrifice accuracy and are susceptible to environmental interference. Here we present a radio frequency (RF)-based (60 to 64 GHz) sensing system that monitors long-term heart rate variability (HRV) with clinical-grade accuracy. Our system successfully overcomes the orders-larger interference from respiration motion in far-field conditions without any model training. By identifying previously undiscovered frequency ranges (beyond 10-order heartbeat harmonics) where heartbeat information predominates over other motions, we generate prominent heartbeat patterns with harmonics typically considered detrimental. Extensive evaluations, including a large-scale outpatient setting involving 6,222 eligible participants and a long-term daily life scenario, where sleep data was collected over 5 separate random nights over two months and a continuous 21-night period, demonstrate that our system can monitor HRV and identify abnormalities with comparable performance to clinical-grade ECG-based systems. This RF-based HRV sensing system has the potential to support active self-assessment and revolutionize medical prevention with long-term and precise health monitoring.

Despite the enormous potential of millimeter-wave (mmWave) imaging, the high cost of large-scale antenna arrays or stringent prerequisites of the synthetic aperture radar (SAR) principle impedes its widespread application. Here, we report a portable, affordable, and high-resolution 3D mmWave imaging system by overcoming the destructive motion error of handheld SAR imaging. This is achieved by revealing two important phenomenons: spatial asymmetry of motion errors in different directions, and local similarity of phase errors exhibited by different targets, based on which we formulate the challenging phase error estimation problem as a tractable point spread function optimization problem. Experiments demonstrate that our approach can recover high-fidelity 3D mmWave images from severely distorted signals and augment the aperture size by over 50 times. Since our system does not rely on costly massive antennas or bulky motion controllers, it can be applied for diverse applications including security inspection, autonomous driving, and medical monitoring.

Infrared (IR) small target detection in sky scenes is crucial for aerospace, border security, and atmospheric monitoring. Most current works are typically designed for generalized IR scenes, which may not be optimal for the specific scenario of sky backgrounds, particularly for detecting small and dim targets at long ranges. In these scenarios, the presence of heavy clouds usually causes significant false alarms due to factors such as strong edges, streaks, large undulations, and isolated floating clouds. To address these challenges, we propose an infrared dim and small target detection algorithm based on morphological filtering with dual-structure elements. First, we design directional dual-structure element morphological filters, which enhance the grayscale difference between the target and the background in various directions, thus highlighting the region of interest. The grayscale difference is then normalized in each direction to mitigate the interference of false alarms in complex cloud backgrounds. Second, we employ a dynamic scale awareness strategy, effectively preventing the loss of small targets near cloud edges. We enhance the target features by multiplying and fusing the local response values in all directions, which is followed by threshold segmentation to achieve target detection results. Experimental results demonstrate that our method achieves strong detection performance across various complex cloud backgrounds. Notably, it outperforms other state-of-the-art methods in detecting targets with a low signal-to-clutter ratio (MSCR ≤ 2). Furthermore, the algorithm does not rely on specific parameter settings and is suitable for parallel processing in real-time systems.

This paper demonstrates human synthesis based on the Radio Frequency (RF) signals, which leverages the fact that RF signals can record human movements with the signal reflections off the human body. Different from existing RF sensing works that can only perceive humans roughly, this paper aims to generate fine-grained optical human images by introducing a novel cross-modal RFGAN model. Specifically, we first build a radio system equipped with horizontal and vertical antenna arrays to transceive RF signals. Since the reflected RF signals are processed as obscure signal projection heatmaps on the horizontal and vertical planes, we design a RF-Extractor with RNN in RFGAN for RF heatmap encoding and combining to obtain the human activity information. Then we inject the information extracted by the RF-Extractor and RNN as the condition into GAN using the proposed RF-based adaptive normalizations. Finally, we train the whole model in an end-to-end manner. To evaluate our proposed model, we create two cross-modal datasets (RF-Walk & RF-Activity) that contain thousands of optical human activity frames and corresponding RF signals. Experimental results show that the RFGAN can generate target human activity frames using RF signals. To the best of our knowledge, this is the first work to generate optical images based on RF signals.

Existing RF-based human pose estimation methods usually require intensive computations and cannot meet the real-time processing and portability requirements for mobile devices. To tackle the limitation, in this paper, we introduce a lightweight RF-based pose estimation model, i.e., MobiRFPose, to construct the portable RF-based pose camera. Different from traditional optical-based cameras, the RF-based camera does not capture visual information, which means the privacy-preserving characteristic. Specifically, we only utilize a horizontal antenna array to transceive RF signals, then estimate the human locations on the RF signal heatmap and crop the human location regions, and finally estimate the fine-grained human poses based on the cropped small RF signal heatmaps. To evaluate the performance, we compare MobiRFPose with state-of-the-art methods. Experimental results demonstrate that MobiRFPose can achieve accurate 3D human pose estimation with fewer parameters and computations. We also test the trained MobiRFPose model using mobile computing devices, where the model structures and parameters only take up 268 KB and 3226 KB of disk space, and MobiRFPose can achieve 66 FPS processing speed. The pose estimation error is 11.05 cm in the case of a single person and 11.29 cm in the case of multiple people. All experimental results indicate that our proposed method can construct a portable RF camera to estimate human poses accurately.

Atrial fibrillation (AF) has been a prevalent and serious arrhythmia associated with increased morbidity and mortality worldwide. The Electrocardiogram (ECG) is considered as the golden standard for AF diagnosis. However, current ECG is primarily used only when symptoms arise or for occasional checkups due to the necessity of contact-based measurements. This limitation results in difficulty of capturing early-stage AF episodes and missed opportunities for timely intervention. Here we introduce a contactless, operation-free, and device-free AF detection framework utilizing artificial intelligence (AI)-powered radio technology. Our approach analyzes the mechanical motion of the heart using radar sensing and leverages AI-powered knowledge transfer from established clinical ECG diagnostic practices to read AF-associated motion patterns precisely. Our system is evaluated on 6258 outpatient visitors, including 229 with AF, and achieves AF detection with a sensitivity of 0.844 (95% Confidence Interval (CI), 0.790-0.884) and a specificity of 0.995 (95% CI, 0.993-0.997), which is comparable to the performance of ECG-based methods. We also provide initial evidence that this system could be deployed in a practical daily life scenario, detecting AF before traditional clinical diagnosis routines. These results highlight its potential to support feasible lifelong proactive monitoring, covering the full spectrum of AF progression.