Search for: All records

Abstract 24/7 continuous recording of in-home daily trajectories is informative for health status assessment (e.g., monitoring Alzheimer’s, dementia based on behavior patterns). Indoor device-free localization/tracking are ideal because no user efforts on wearing devices are needed. However, prior work mainly focused on improving the localization accuracy. They relied on well-calibrated sensor placements, which require hours of intensive manual setup and respective expertise, feasible only at small scale and by mostly researchers themselves. Scaling the deployments to tens or hundreds of real homes, however, would incur prohibitive manual efforts, and become infeasible for layman users. We presentSCALING, a plug-and-play indoor trajectory monitoring system that layman users can easily set up by walking a one-minute loop trajectory after placing radar nodes on walls. It uses a self calibrating algorithm that estimates sensor locations through their distance measurements to the person walking the trajectory, a trivial effort without taxing layman users physically or cognitively. We evaluateSCALINGvia simulations and two testbeds (in lab and home configurations of sizes 3$$\times$$ $\times$6 sq m and 4.5$$\times$$ $\times$8.5 sq m). Experimental results demonstrate thatSCALINGoutperformed the baseline using the approximate multidimensional scaling (MDS, the most relevant method in the context of self calibration) by 3.5 m/1.6 m in 80-percentile error of self calibration and tracking, respectively. Notably, only 1% degradation in performance has been observed withSCALINGcompared to the classical multilateration with known sensor locations (anchors), which costs hours of intensive calibrating effort. In addition, we conduct Monte Carlo experiments to numerically analyze the impact of sensor placements and develop practical guidelines for deployment in real life scenarios. 

Human activity recognition provides insights into physical and mental well-being by monitoring patterns of movement and behavior, facilitating personalized interventions and proactive health management. Radio Frequency (RF)-based human activity recognition (HAR) is gaining attention due to its less privacy exposure and non-contact characteristics. However, it suffers from data scarcity problems and is sensitive to environment changes. Collecting and labeling such data is laborintensive and time consuming. The limited training data makes generalizability challenging when the sensor is deployed in a very different relative view in the real world. Synthetic data generation from abundant videos presents a potential to address data scarcity issues, yet the domain gaps between synthetic and real data constrain its benefit. In this paper, we firstly share our investigations and insights on the intrinsic limitations of existing video-based data synthesis methods. Then we present M4X, a method using metric learning to extract effective viewindependent features from the more abundant synthetic data despite their domain gaps, thus enhancing cross-view generalizability. We explore two main design issues in different mining strategies for contrastive pairs/triplets construction, and different forms of loss functions. We find that the best choices are offline triplet mining with real data as anchors, balanced triplets, and a triplet loss function without hard negative mining for higher discriminative power. Comprehensive experiments show that M4X consistently outperform baseline methods in cross-view generalizability. In the most challenging case of the least amount of real training data, M4X outperforms three baselines by 7.9- 16.5% on all views, and 18.9-25.6% on a view with only synthetic but no real data during training. This proves its effectiveness in extracting view-independent features from synthetic data despite their domain gaps. We also observe that given limited sensor deployments, a participant-facing viewpoint and another at a large angle (e.g. 60◦) tend to produce much better performance.