Recent advancements in Text-Attributed Graphs (TAGs) have attracted significant attention for their wide-ranging applications in domains such as social networks, academics, and e-commerce. The powerful text-processing capabilities of pre-trained language models offer a promising avenue for effectively integrating textual attributes with graph structures. However, existing methods exhibit two key limitations: (1) reliance on rigid graph construction processes that fail to capture a comprehensive view of the text-attributed graph data; (2) insufficient fusion of textual semantics and graph topology, leading to information loss, unstable training, and limited generalization across diverse downstream tasks. In this work, we propose GA-Clip, a novel semantic-aware graph augmentation contrastive learning model. We leverage the pre-trained language model to generate semantic edges that extract the fine-grained topology within the text feature space to augment the graph structure. We then separately employ the graph and text encoders to sufficiently fuse the different modalities through a modified self-supervised contrastive learning approach. This augmentation mitigates the dependency on cumbersome graph construction processes and integrates information from different modalities, which jointly enables scalable graph learning on coarse-grained, large-scale source data. Experimental results demonstrated that our approach achieved a 2-4% improvement over the SOTA methods in accuracy across multiple datasets, validating the effectiveness of the proposed method.

Radio Frequency IDentification (RFID) is emerging as a vital technology of the Internet of Things (IoT). Billions of RFID tags have been deployed to locate daily objects such as equipment, pharmaceuticals, vehicles, and so on. Unlike previous solutions that focus on localizing tagged objects in the world coordinate system in reference to reader antennas, this work exploits a system, called RFCamera, that can identify and locate RFID-tagged objects in images with pixel dimensions. Our core insight is that an image is a visual AoA profile in terms of lights, which is resulted from the pinhole camera model. Similarly, we generate an RF image derived from the AoA profile of a tag using the same pinhole model as the camera. Consequently, the locations of visual entities corresponding to tagged objects are highlighted by comparing two types of images. To this end, we customized a camera system equipped with a pair of rotatable reader antennas. Our experimental evaluation demonstrates that RFCamera enables a mean error of 5.7∘ and 2.9∘ at azimuth and elevation angle estimation, respectively. It can locate a visual entity with a mean error of 51 pixels (i.e., ≈1.3 cm at 96 dpi) in a 640× 480 image.

Single twisted pair Ethernet becomes popular in the industrial internet of thing (IIoT), since it can use only one twisted pair to provide high speed data transmission while the cables of the field bus can be reused. However, since its transmission medium is inferior to traditional Ethernet, it is easier to generate delay jitter that greatly impacts the accuracy of time synchronization. To solve this problem, in this paper, an offset estimation method based on AutoRegressive Integrated Moving Average (ARIMA) and Long Short-Term Memory (LSTM) is proposed to estimate the clock offset when the delay jitter appears. The offset estimation model is firstly obtained by training the ARIMA-LSTM with offline offset data. When the delay jitter is detected, the offset can be estimated by the model to replace the unreliable offset obtained by the time synchronization protocol. Experiments are executed in the testbed, and the results prove that the proposed method can improve the time synchronization accuracy in the single twisted pair Ethernet.

An ultrasonic positioning system (UPS) has demonstrated its high accuracy for years. However, few of the developed solutions have been deployed in practice to satisfy the localization demand of today’s smart devices, which lack ultrasonic sensors and were considered as being “deaf” to ultrasound. A recent finding demonstrates that ultrasound may be audible to the smart devices under certain conditions due to their microphone’s nonlinearity. Inspired by this insight, this work revisits the ultrasonic positioning technique and builds a practical UPS, called UPS+, for ultrasound-incapable smart devices. The core concept is to deploy two types of indoor beacon devices, which will advertise ultrasonic beacons at two different ultrasonic frequencies respectively. Their superimposed beacons are downconverted to a low-frequency by exploiting the nonlinearity effect at the receiver’s microphone. This underlying property functions as an implicit ultrasonic downconverter without inflicting harm to the hearing system of humans. We demonstrate UPS+, a fully functional UPS prototype, with centimeter-level localization accuracy using custom-made beacon hardware and well-designed algorithms.

Radio Frequency IDentification (RFID) is emerging as a vital technology of the Internet of Things. Billions of RFID tags have been deployed to locate daily objects such as equipment, pharmaceuticals, and vehicles, and so on. Unlike previous solutions that focus on localizing tagged objects in the world coordinate system in reference to reader antennas, this work exploits a system, called RFCamera, that can identify and locate RFID-tagged objects in images with pixel dimensions. Many applications would benefit from RFCamera. For instance, the RF-aware image annotation system is able to generate rich annotations for RFID-tagged entities in images at the pixel level for the deep learning; the RF-aware auto-focus allows surveillance camera to exactly focalize the burglar who carries the stolen tagged-property out of a crowd. Our core insight is that an image is a visual AoA profile in terms of lights, which is resulted from the pinhole camera model. Similarly, we generate an RF image, derived from the AoA profile of a tag using the same pinhole model as the camera. Consequently, the locations of visual entities corresponding to tagged objects are highlighted by comparing two types of images. To this end, we customized a camera system equipped with a pair of rotatable reader antennas. Our experimental evaluation demonstrates that RFCamera enables a mean error of 5.7° and 2.9° at azimuth and elevation angle estimation. It can locate a visual entity with a mean error of 51 pixels (i.e., ≈ 1.3 cm at 96 dpi) in a 640 × 480 image.

Radio frequency identification technology has been recently employed as an effective solution for anti-counterfeiting. By attaching a tag on each product, a reader can provide a fast and automatic authentication. However, most of existing approaches adopt a per-tag approach to validate each product, which will incur a long scanning time and communication traffic when there are many products to be concurrently authenticated. To address these issues, batch authentication, e.g., SEBA, is proposed in recent literature. Batch authentication is an identification-free and probabilistic approach and aims to authenticate the validity of a batch of tags. Unfortunately, the existing batch authentication still suffers a serious scalability problem where the system cost linearly depends on the total number of genuine tags, e.g.,${\mathcal {O}(N)}$. Its efficiency will significantly decrease and even be worse than the per-tag approach when there is a large number of genuine tags in the system. This limitation hinders the batch authentication to be widely used in practice because there are up to millions of total genuine tags in usual. In this paper, we propose FISH, which is also probabilistic batch authentication-based, to overcome the scalability problem by reducing the dependence to${\mathcal {O}(\log (N))}$. Moreover, FISH owns the following two salient advantages: 1) FISH can identify the genuine products when there is no counterfeit product in the batch although it is identification-free in essence and 2) FISH can defend against silence attack by utilizing Pearson Chi-square test. Last, we also conduct the theoretical analysis and extensive simulation with the real logistics trace

Radio frequency identification (RFID) technology has been widely adopted in a variety of applications from logistics to access control. Many applications gain benefits from knowing the exact position of an RFID-tagged object. Existing localization algorithms in wireless network, however, can hardly be directly employed due to tag's limited capabilities in terms of energy and memory. In this paper, we propose BackPos, a fine-grained backscatter positioning technique using the commercial off-the-shelf (COTS) RFID products with detected phases. Our studies show that the phase is a stable indicator highly related to tag's position and preserved over frequency or tag orientation, but challenged by its periodicity and tag's diversity. We attempt to infer the distance differences from phases detected by antennas under triangle constraint. Further, hyperbolic positioning using the distance differences is employed to shrink the tag's candidate positions until finding out the real one. In combination with interrogation zone, we finally relax the triangle constraint and allow arbitrary deployment of antennas by sacrificing the feasible region. We implement a prototype of BackPos with COTS RFID products and evaluate this design in various scenarios. The results show that BackPos achieves the mean accuracy of 12:8cm with variance of 3:8 cm.

Screen locking/unlocking is important for modern smart phones to avoid the unintentional operations and secure the personal stuff. Once the phone is locked, the user should take a specific action or provide some secret information to unlock the phone. The existing unlocking approaches can be categorized into four groups: motion, password, pattern, and fingerprint. Existing approaches do not support smart phones well due to the deficiency of security, high cost, and poor usability. We collect 200 users' handwaving actions with their smart phones and discover an appealing observation: the waving pattern of a person is kind of unique, stable and distinguishable. In this paper, we propose OpenSesame, which employs the users' waving patterns for locking/unlocking. The key feature of our system lies in using four fine-grained and statistic features of handwaving to verify users. Moreover, we utilize support vector machine (SVM) for accurate and fast classification. Our technique is robust compatible across different brands of smart phones, without the need of any specialized hardware. Results from comprehensive experiments show that the mean false positive rate of OpenSesame is around 15 percent, while the false negative rate is lower than 8 percent.

Existing methods in RFID systems often employ presence or absence fashion to detect the tags' motions, so they cannot meet motion detection requirement in many applications. Our recent observations suggest that the signal strength backscattered from the tag is hypersensitive to its position, inspiring us to perceive the tag motion through its radio signal strength changes. Motion perception is not trivial and challenged by weak stability of strength in that any other interference or noise may incur significant changes as well, resulting in high false positives. To tackle this issue, we propose to model the strength via the Mixture of Gaussian Model (MoG). The problem is thus converted to foreground segment in computer vision with the help of Strength Image, where the technique of MoG based background subtraction is employed. We then implement a prototype using commercial off-the-shelf products. The evaluation results show that the slightest tag motion (~10 cm) can be precisely perceived, and the accuracy is up to 92.34 percent while the false positive is suppressed under 0.5 percent.

Indoor navigation has received much attention in academics and industry in recent years. Previous methods often attempt to locate users with various localization algorithms in combination with an indoor map, so they need expensive infrastructures deployed in advance. In this study, we propose to utilize existing indoor objects attached RFID tags and the reader to navigate the user to the destination, without need of any extra hardware. The key insight is that the personal movement takes an impact on the Doppler frequency shift values collected from the indoor objects when getting close to the tag. Such local human-item spatial relation is leveraged to infer the users position and further navigate user to destination step by step. We implement a prototype navigation system, called RollCaller, and conduct comprehensive experiments to examine its performance.

RFID technology has been widely adopted in a variety of applications from logistics to access control. Many applications gain benefits from knowing the exact position of an RFID-tagged object. Existing localization algorithms in wireless network, however, can hardly be directly employed due to tag's limited capabilities in terms of energy and memory. For example, the RSS based methods are vulnerable to both distance and tag orientation, while AOA based methods put a strict constraint on the antennas' spacing that reader's directional antennas are too large to meet. In this paper, we propose BackPos, a fine-grained backscatter positioning technique using the COTS RFID products with detected phase. Our study shows that the phase is indeed a stable indicator highly related to tag's position and preserved over frequency or tag orientation, but challenged by its periodicity and tag's diversity. We attempt to infer the distance differences from phases detected by antennas under triangle constraint. Further, hyperbolic positioning using the distance differences is employed to shrink the tag's candidate positions until filtering out the real one. In combination with interrogation zone, we finally relax the triangle constraint and allow arbitrary deployment of antennas by sacrificing the feasible region. We implement a prototype of BackPos with COTS RFID products and evaluate this design in various scenarios. The results show that BackPos achieves the mean accuracy of 12.8cm with variance of 3.8cm.

In many applications, we have to identify an object and then locate the object to within high precision (centimeter- or millimeter-level). Tracking mobile RFID tags in real time has been a daunting task, especially challenging for achieving high precision. We achieve these three goals by leveraging the phase value of the backscattered signal, provided by the COTS RFID readers, to estimate the location of the object. To illustrate the basic idea of our system, we firstly focus on a simple scenario where the tag is moving along a fixed track known to the system. We propose Differential Augmented Hologram (DAH) which will facilitate the instant tracking of the mobile RFID tag to a high precision. We then devise a comprehensive solution to accurately recover the tag's moving trajectory and its locations, relaxing the assumption of knowing tag's track function in advance.

In this paper, we model the data dissemination in online social networks (OSNs) and study the scaling laws of traffic load. We propose a three-layered system model to formulate data dissemination sessions for social applications in OSNs. The layered model consists of the physical network layer, social relationship layer, and application session layer. By analyzing mutual relevances among these three layers, we investigate the geographical distribution feature of dissemination sessions in OSNs. Based on this, we derive the traffic load of OSNs under a realistic assumption that every source sustains a data generating rate of constant order. To the best of our knowledge, this is the first work to address the issue of traffic load scaling for OSNs by modeling the social data dissemination from a layered perspective.

Screen locking/unlocking is important for modern smart phones to avoid the unintentional operations and secure the personal stuff. Once the phone is locked, the user should take a specific action or provide some secret information to unlock the phone. Existing approaches do not support smart phones well due to the deficiency of security, high cost, and poor usability. We collect 200 users' handshaking actions with their smart phones and discover an appealing observation: the shaking pattern of a person is kind of unique, stable and distinguishable. In this paper, we propose OpenSesame, which employs the users' shaking patterns for locking/unlocking. The key feature of our system lies in using four fine-grained and statistic features of handshaking to verify users. Moreover, we utilize support vector machine (SVM) for accurate classification. Results from comprehensive experiments show that our technique is robust compatible across different brands of smart phones, without the need of any specialized hardware.

This paper is concerned with the routing protocol design for large-scale wireless sensor and actor networks (WSANs). The actor-sensor-actor communication (ASAC) strategy is first proposed to guarantee the reliability of persistent actor-actor communication. To keep network connectivity and prolong network lifetime, we propose a dynamic gradient-based routing protocol (DGR) to balance the energy consumption of the network. With the different communication ranges of sensors and actors, the DGR protocol uses a data load expansion strategy to significantly prolong the network lifetime. The balance coefficient and the routing re-establishment threshold are also introduced to make the tradeoff between network lifetime and routing efficiency. Simulation results show the effectiveness of the proposed DGR protocol for unbalanced and persistent data transmission.