Continuous tracking is often thought to require specialised, actively powered sensors. Yet energy harvesters already embedded in commercial devices, such as Garmin solar-powered smartwatches, generate energy signals that inherently carry continuous variations linked to user motion and environment. Prior studies have shown that these signals are sufficient for classification tasks such as human activity and gesture recognition by exploiting class-distinguishing cues. However, whether they can support continuous trajectory tracking has remained an open question-until now.In this paper, we present the first fundamental study of continuous hand trajectory tracking with wearable solar harvesters. Through a novel radiometric model, we analytically link photovoltaic (PV) power to the solar cell’s geometric configuration, exposing both the promise of energy signals for tracking and their core limitations: the positional ambiguities and distortions that arise when power is used directly for positioning. To resolve this ambiguity, we propose SolarTrack, a framework that embeds a radiometric model as a physical backbone within a sequence-learning pipeline, enforcing cycle consistency between data-driven predictions and physical feasibility. This yields the first standalone solar-based tracker capable of estimating continuous hand motion directly from harvested energy signals.To validate this, we built a wearable prototype with a solar panel and IMU and collected the first dataset pairing motion-capture ground truth with harvested energy from 15 participants (700k samples). Results show that solar signals alone achieve subdecimeter tracking accuracy, outperforming purely data-driven baselines and only 1.6 cm worse compared to the IMU tracker despite having access to only 1D power signal. Furthermore, when fused with IMU, it further boosts IMU performance by 13%.

Real-world sensing challenges such as sensor failures, communication issues, and power constraints lead to data intermittency. An issue that is known to undermine the traditional classification task that assumes a continuous data stream. Previous works addressed this issue by designing bespoke solutions (i.e. task-specific and/or modality-specific imputation). These approaches, while effective for their intended purposes, had limitations in their applicability across different tasks and sensor modalities. This raises an important question: Can we build a task-agnostic imputation pipeline that is transferable to new sensors without requiring additional training? In this work, we formalise the concept of zero-shot imputation and propose a novel approach that enables the adaptation of pre-trained models to handle data intermittency. This framework, named NeuralPrefix, is a generative neural component that precedes a task model during inference, filling in gaps caused by data intermittency. NeuralPrefix is built as a continuous dynamical system, where its internal state can be estimated at any point in time by solving an Ordinary Differential Equation (ODE). This approach allows for a more versatile and adaptable imputation method, overcoming the limitations of task-specific and modality-specific solutions. We conduct a comprehensive evaluation of NeuralPrefix on multiple sensory datasets, demonstrating its effectiveness across various domains. When tested on intermittent data with a high 50% missing data rate, NeuralPreifx accurately recovers all the missing samples, achieving SSIM score between 0.93-0.96. Zero-shot evaluations show that NeuralPrefix generalises well to unseen datasets, even when the measurements come from a different modality. Project page: https://neuralprefix.github.io/

We introduce BeaMsteerX (BMX), a novel mmWave hand hygiene gesture recognition technique that improves accuracy in longer ranges (1.5m). BMX steers a mmWave beam towards multiple directions around the subject, generating multiple views of the gesture that are then intelligently combined using deep learning to enhance gesture classification. We evaluated BMX using off-the-shelf mmWave radars and collected a total of 7,200 hand hygiene gesture data from 10 subjects performing a 6-step hand-rubbing procedure, as recommended by the World Health Organization, using sanitizer, at 1.5m---over 5 times longer than in prior works. BMX outperforms state-of-the-art approaches by 31--43% and achieves 91% accuracy at boresight by combining only two beams, demonstrating superior gesture classification in low SNR scenarios. BMX maintained its effectiveness even when the subject was positioned 30° away from the boresight, exhibiting a modest 5% drop in accuracy.

Recent research has demonstrated the huge potential of WiFi for contactless sensing of human activities. Unfortunately, such sensing is highly sensitive to the relative orientation between the user and the WiFi receivers. To overcome this problem, existing solutions deploy multiple WiFi receivers at precise positions to capture orientation-independent view of the human activity. Orientation-independent single-receiver WiFi sensing is still considered an open problem. In this article, we propose a deep neural network architecture that uses radar data during training to learn high-precision Doppler features of human activities from the noisy channel states observed by a single WiFi receiver. Once trained with radars, the network can be used to detect human activities at any arbitrary orientations based only on WiFi signals. Using extensive experiments with millimeter-wave radars, we demonstrate that the proposed approach, called WiFi2Radar in this article, significantly outperforms state-of-the-art for detecting human activities in untrained orientations using only a single WiFi receiver. Our results show that WiFi2Radar can detect orientation-independent human activities with up to 91% accuracy, which outperforms the state of the art by 19%.

Hand Pose Estimation (HPE) is crucial to many applications, but conventional cameras-based CM-HPE methods are completely subject to Line-of-Sight (LoS), as cameras cannot capture occluded objects. In this paper, we propose to exploit Radio-Frequency-Vision (RF-vision) capable of bypassing obstacles for achieving occluded HPE, and we introduce OCHID-Fi as the first RF-HPE method with 3D pose estimation capability. OCHID-Fi employs wideband RF sensors widely available on smart devices (e.g., iPhones) to probe 3D human hand pose and extract their skeletons behind obstacles. To overcome the challenge in labeling RF imaging given its human incomprehensible nature, OCHID-Fi employs a cross-modality and cross-domain training process. It uses a pre-trained CM-HPE network and a synchronized CM/RF dataset, to guide the training of its complex-valued RF-HPE network under LoS conditions. It further transfers knowledge learned from labeled LoS domain to unlabeled occluded domain via adversarial learning, enabling OCHID-Fi to generalize to unseen occluded scenarios. Experimental results demonstrate the superiority of OCHID-Fi: it achieves comparable accuracy to CM-HPE under normal conditions while maintaining such accuracy even in occluded scenarios, with empirical evidence for its generalizability to new domains.

Each year, hundreds of thousands of people contract Healthcare Associated Infections (HAI). Poor hand hygiene compliance among healthcare workers is thought to be the leading cause of HAIs and methods were developed to measure compliance. Surprisingly, human observation is still considered the gold standard for measuring compliance by World Health Organization (WHO). Moreover, no automated solutions exist for monitoring hand hygiene techniques, such as "how to hand rub" technique by WHO. In this work, we introduce RFWash; the first radio-based device-free system for monitoring Hand Hygiene (HH) technique. On the technical level, HH gestures are performed back-to-back in a continuous sequence and pose a significant challenge to conventional two-stage gesture detection and recognition approaches. We propose a deep model that can be trained on unsegmented naturally-performed HH gesture sequences. RFWash evaluation demonstrates promising results for tracking HH gestures, achieving gesture error rate of≈67% compared to fully supervised approach. The work is a step towards practical RF sensing that can reliably operate inside future healthcare facilities.

Sensitivity of WiFi channel measurements to the transceiver placement is a major limitation for on-demand deployment of device-free WiFi sensing in environments where the transmitting/receiving devices may move. Using publicly available datasets, we show that even slight deviations of transmitter/receiver placements from the reference values can degrade device-free gesture recognition accuracy significantly. We design a convolutional autoencoder to translate WiFi spectrograms from arbitrary receiver placements to a reference placement configuration in a given area of interest with minimal human effort. Our experiments with the public datasets reveal that the proposed autoencoder can successfully reduce WiFi measurement variability caused by transmitter/receiver movement, which ultimately increases gesture recognition accuracy by up to 58%.

Each year, hundreds of thousands of people contract Healthcare Associated Infections (HAIs). Poor hand hygiene compliance among healthcare workers is thought to be the leading cause of HAIs and methods were developed to measure compliance. Surprisingly, human observation is still considered the gold standard for measuring compliance by World Health Organization (WHO). Moreover, no automated solutions exist for monitoring hand hygiene techniques, such as "how to hand rub" technique by WHO. In this paper, we introduce RFWash; the first radio-based device-free system for monitoring Hand Hygiene (HH) technique. On the technical level, HH gestures are performed back-to-back in a continuous sequence and pose a significant challenge to conventional two-stage gesture detection and recognition approaches. We propose a deep model that can be trained on unsegmented naturally-performed HH gesture sequences. RFWash evaluation demonstrates promising results for tracking HH gestures, achieving gesture error rate of < 8% when trained on 10-second segments, which reduces manual labelling overhead by ≈ 67% compared to fully supervised approach. The work is a step towards practical RF sensing that can reliably operate inside future healthcare facilities.

We present Heartbeats-2-Bits (H2B), which is a system for securely pairing wearable devices by generating a shared secret key from the skin vibrations caused by heartbeat. This work is motivated by potential power saving opportunity arising from the fact that heartbeat intervals can be detected energy-efficiently using inexpensive and power-efficient piezo sensors, which obviates the need to employ complex heartbeat monitors such as Electrocardiogram or Photoplethysmogram. Indeed, our experiments show that piezo sensors can measure heartbeat intervals on many different body locations including chest, wrist, waist, neck and ankle. Unfortunately, we also discover that the heartbeat interval signal captured by piezo vibration sensors has low Signal-to-Noise Ratio (SNR) because they are not designed as precision heartbeat monitors, which becomes the key challenge for H2B. To overcome this problem, we first apply a quantile function-based quantization method to fully extract the useful entropy from the noisy piezo measurements. We then propose a novel Compressive Sensing-based reconciliation method to correct the high bit mismatch rates between the two independently generated keys caused by low SNR. We prototype H2B using off-the-shelf piezo sensors and evaluate its performance on a dataset collected from different body positions of 23 participants. Our results show that H2B has a pairing success rate of 95.6%. We also analyze and demonstrate H2B's robustness against three types of attacks. Finally, our power measurements show that H2B is very power-efficient.

Heart rate is one of the most important vital signals for personal health tracking. A number of approaches were proposed to monitor heart rate, ranging from wearables to device-less systems. While WiFi has been shown to track heart rate accurately, existing solutions rely on directional antennas to improve the signal quality and ultimately the accuracy of heart rate estimation. Special hardware used in these approaches limits their applicability and truly device-less and ubiquitous heart rate monitoring is yet to be achieved.