The deeply penetrated WiFi signals not only provide fundamental communications for the massive Internet of Things (IoT) devices, but also enable cognitive sensing ability in many other applications, such as human activity recognition. State-of-the-art WiFi-based device-free systems leverage the correlations between signal changes and body movements for human activity recognition. They have demonstrated reasonably good recognition results with a properly placed transceiver pair, or, in other words, when the human body is within a certain sweet zone. Unfortunately, the sweet zone is not ubiquitous. When the person moves out of the area and enters a dead zone, or even just the orientation changes, the recognition accuracy can quickly decay. In this paper, we closely examine such spatial diversity in WiFi-based human activity recognition. We identify the dead zones and their key influential factors, and accordingly present WiSDAR, a WiFi based Spatial Diversity-aware device-free Activity Recognition system. WiSDAR overshadows the dead zones yet with only one physical WiFi sender and receiver. The key innovation is extending the multiple antennas of modern WiFi devices to construct multiple separated antenna pairs for activity observing. Profiling activity features from multiple spatial dimensions can be more complicated and offer much richer information for further recognition. To this end, we propose a deep learning based framework that integrates the hidden features from both temporal and spatial dimensions, achieving highly accurate and reliable recognition results. WiSDAR is fully compatible with commercial off-the-shelf WiFi devices, and we have implemented it on the commonly available Intel WiFi 5300 cards. Our real-world experiments demonstrate that it recognizes human activities with a stable accuracy of around 96%.

With the multi-antenna design of WiFi interfaces, phased array has become a promising mechanism for accurate WiFi localization. State-of-the-art WiFi-based solutions using AoA (Angle-of-Arrival), however, face a number of critical challenges. First, their localization accuracy degrades dramatically due to low Signal-to-Noise Ratio (SNR) and incoherent processing. Second, they tend to produce outliers when the available number of packets is low. Moreover, the prior phase calibration schemes are not multipath robust and accurate enough. All the above degrade the robustness of localization systems. In this paper, we present ROArray, a RObust Array based system that accurately localizes a target even with low SNRs. The key insight of ROArray is to use sparse recovery and coherent processing across all available domains, including time, frequency, and spatial domains. Also, ROArray includes an online phase calibration technique that can eliminate random phase offsets while keeping communication uninterrupted. Our implementation using off-the-shelf WiFi cards demonstrates that, with low SNRs, ROArray significantly outperforms state-of-the-art solutions in terms of localization accuracy; when medium or high SNRs are present, it achieves comparable accuracy.

WiFi-based human activity recognition explores the correlations between body movement and the reflected WiFi signals to classify different activities. State-of-the-art solutions mostly work on a single WiFi channel and hence are quite sensitive to the quality of a particular channel. Co-channel interference in an indoor environment can seriously undermine the recognition accuracy. In this paper, we for the first time explore wideband WiFi information with advanced deep learning towards more accurate and robust activity recognition. We present a practical Channel Selective Activity Recognition system (CSAR) with Commercial Off-The-Shelf (COTS) WiFi devices. The key innovation is to actively select available WiFi channels with good quality and seamlessly hop among adjacent channels to form an extended channel. The wider bandwidth with more subcarriers offers stable information with a higher resolution for feature extraction. Conventional classification tools, e.g., hidden Markov model and k-nearest neighbours, however, are not only sensitive to feature distortion but also not smart enough to explore the time scale correlations from the extracted spectrogram. We accordingly explore advanced deep learning tools for this application context. We demonstrate an integration of channel selection and long short term memory network (LSTM), which seamlessly combine the richer time and frequency features for activity recognition. We have implemented a CSAR prototype using Intel 5300 WiFi cards. Our realworld experiments show that CSAR achieves a stable recognition accuracy around 95% even in crowded wireless environments (compared to 80% with state-of-the-art solutions that highly depend on the quality of the working channel). We have also examined the impact of environments and persons, and the results reaffirm its robustness.

With rapid development of RFID technology, ever-increasing research effort has been dedicated to devising RFID-enabled services. The missing tag identification, which is to identify all missing tags, is one of the most important services in Internet-of-Things applications such as inventory management. Prior work on missing tag detection rely on hash functions at tags. However, hash functions are not supported by COTS RFID tags. To bridge this gap between theory and practice, this paper is devoted to detecting missing tags with COTS Gen2 devices. We first introduce a point-to-multipoint protocol, named P2M that works in an analog frame slotted Aloha paradigm to interrogate tags and collect their EPCs. A missing tag will be found if its EPC is not present in the collected ones. To reduce time cost of P2M resulted from tag response collisions, we further present a collision-free point-to-point protocol, named P2P that selectively specifies a tag to reply with its EPC in each slot. If the EPC is not received, this tag is regarded to be missing. We develop bitmask selection methods to enable the selective query while reducing communication overhead. We implement P2M and P2P with COTS RFID devices and evaluate their performance under diverse settings.

In-car human activity recognition opens a new opportunity toward intelligent driving behavior detection and touchless human-car interaction. Among the many sensing technologies (e.g., using cameras and wearable sensors), radio frequency identification (RFID) exhibits unique advantages given its low cost, easy deployment, and less privacy concerns. Existing RFID-based solutions for activity recognition are mostly confined to working in stable indoor spaces. The inside space of a car however is much more compact and complex, not to mention the fast-changing driving conditions. All these introduce non-negligible noises that pollute the activity-related information, and the existence of various car models in the market further complicates the problem. In this article, we for the first time closely examine the distinct factors that affect the RFID-based in-car activity recognition. We present RF-CAR , a novel RFID-based tag-free solution that well adapts to different in-car environments. RF-CAR smartly filters the domain-specific features in RF signals and retains activity-related features to the maximum extent. It then integrates a deep learning architecture and an advanced multi-adversarial domain adaptation network for training and prediction. With only one-time pre-training, RF-CAR can adapt to new data domains such as new driving conditions, car models, and human subjects for robust activity recognition. We also demonstrate that it is readily deployable in cars with commercial off-the-shelf (COTS) RFID devices. Our extensive experiments suggest that RF-CAR achieves an overall recognition accuracy of around 95 percent, which significantly outperforms the state-of-the-art solutions.

Wireless tracing technologies have seen great potentials in many applications, including drawing and writing, gesture-based commanding, and gaming. Many state-of-the-art systems have recently been proposed along this line. However, none of them can strike a balance among hardware complexity, time delay, and accuracy in real-world scenarios. In this paper, we propose RF-Pen, a practical and complete RFID tracking system that achieves centimeter-level real-time tracing with 4 antennas. To do so, RF-Pen mainly employs two key designs, namely selective hologram and hybrid voting. Our selective hologram places antennas in large separation, which not only expedites the tracking process by producing a handful of good-quality candidate points but also maximizes tracing resolution. Nevertheless, a big challenge is ambiguity. To address this, we introduce hybrid voting that effectively integrates RSSI and phase measurements to evaluate the likelihood of all candidate points. This way, a precise initial position and fine-resolution tracing beams are located. We implement RF-Pen using off-the-shelf readers and tags and compare it against state-of-the-art systems. Results show that with a single reader of 4 antennas, RF-pen achieves a median trajectory error of 2.15 cm and a median position error of 12.8 cm, which are $3.7\times$ and $4.1\times$ better than RF-IDraw, respectively.