Location-awareness has become a fundamental requirement for multiple emerging applications with the rapid development of wireless technologies. The high-accuracy ranging enabled by ultra-wide bandwidth (UWB) signals is often deteriorated by clocks imperfections and non-line-of-sight (NLOS) propagation. Existing supervised learning methods for NLOS identification and mitigation are time-consuming, labor-intensive, and cost-inefficient due to the need for training data acquisition and label assignment. This paper presents an indoor localization system that enables NLOS mitigation based on self-training. The system provides a general information fusion framework that integrates map, inertial sensors, and UWB measurements, where the weak labels for UWB measurements are produced and iteratively refined by multi-sensory information fusion for self-training. In addition, the system utilizes the maximum likelihood ranging estimator that considers the impact of clock drift. The effectiveness of the proposed system is demonstrated via extensive experimentation in multiple real-world environments, e.g., the proposed methods reduce the NLOS ranging error by 80% and result in a 90th localization error percentile of 0.5 meters in a complex indoor environment.

Sentiment analysis in social media is challenging since posts are short of context. As a popular way to express emotion on social media, stickers related to these posts can supplement missing sentiments and help identify sentiments precisely. However, research about stickers has not been investigated further. To this end, we present a Chinese sticker-based multimodal dataset for the sentiment analysis task (CSMSA). Compared with previous real-world photo-based multimodal datasets, the CSMSA dataset focuses on stickers, conveying more vivid and moving emotions. The sticker-based multimodal sentiment analysis task is challenging in three aspects: inherent multimodality of stickers, significant inter-series variations between stickers, and complex multimodal sentiment fusion. We propose SAMSAM to address the above three challenges. Our model introduces a flexible masked self-attention mechanism to allow the dynamic interaction between post texts and stickers. The experimental results indicate that our model performs best compared with other models. More researches need to be devoted to this field. The dataset is publicly available at https://github.com/Logos23333/CSMSA.

The ultra-wideband (UWB) technology has been widely used in high-accuracy localization systems benefiting from its superior time-resolution and stability. In recent years, the single-anchor localization (SAL) scheme has attracted intense interests with its high accuracy, low system complexity, and low deployment costs compared to traditional UWB localization schemes. In this paper, we develop a SAL system which achieves 3D high-accuracy localization using time and wrapped phase measurements of UWB signals. We expand the array aperture to improve the localization accuracy and address the phase wrapping problem using information fusion. Statistical characterization of ambiguous position estimates is encapsulated by soft positional information (SPI), which is approximated by Gaussian mixture model (GMM) to accelerate the filtering process. The parameters of GMM are estimated by an efficient parallel algorithm and two filtering methods are proposed to track the agent in different scenarios using approximate SPI. Finally, we implement the system on a low-cost platform and experimental results show that the proposed SAL system can achieve decimeter-level 3D localization accuracy in outdoor and indoor environments.

Accurate indoor localization has attracted much research interest in the past decades. Utilizing existing WiFi infrastructures and inertial measurement units on smartphones is a promising method for indoor localization. In this paper, we propose a multi-grained grid-based filter to jointly estimate the location and the heading direction which can efficiently use map information for localization. Our method calculates the transition probability of location on fine-grained grids and only keeps location probabilities on coarse grids. The on/off model and the wall attenuation factor model are applied to generate observation likelihood from WiFi RSSI measurements. We implemented the algorithm on smartphones and it can run in real time. Experimental results have shown that the root mean squared error of the proposed method is 1.46m and the 90th percentile error is 2.18m.

Significant progress has been made in the past twenty years on the implementation of software-defined radios (SDRs) based on general purpose processors (GPPs). In contrast, there has been limited work on the analysis of the computational requirements of such SDRs. Quantifying the signal processing complexity in SDR is critical to ensuring that the system has sufficient computational resources to support reliable digital communications. Our contribution is to analyze and measure the computational cost of a Gaussian minimum shift keying (GMSK) transceiver based on GNU Radio, a widely used GPP-based SDR. Further, we introduce GNU Radio's working mechanism. To the best of our knowledge, our work is the first to provide detailed computational analysis of digital communications based on GNU Radio.

Significant progress has been made in the past few years on Dynamic Spectrum Access (DSA) wireless networks which seek to use RF spectrum more efficiently and dynamically. For example, many measurements of current spectrum utilization are available and theoretical analyses and computational simulations of DSA networks abound. In sharp contrast, few network systems, particularly those with a decentralized structure, have been built even at a small scale to investigate the performance, behavior, and dynamics of DSA networks. Our contribution is designing a decentralized and asynchronous DSA network and building a prototype based on software radio technologies, signal detection and classification methods, distributed cooperative spectrum sensing systems, and mobile ad-hoc network (MANET) protocols. This paper details the network''s design and implementation as well as its enabling technologies. Through systematic experiments, we identify several factors influencing performance for decentralized DSA networks.

This paper introduces the Virginia Tech Center for Wireless Telecommunications' Vehicular Public Safety Cognitive Radio (VPSCR), a vehicle-based extension to our previously fielded cognitive radio prototype for the public safety community. Specifically, the VPSCR can scan the radio spectrum over multiple public safety frequency bands, classify commonly used public safety waveforms and networks, and reconfigure itself for network interoperation. The VPSCR is designed for vehicular assembly, powering itself from the onboard car battery. Such a platform, however, has a significant power requirement, which prevents it from being easily downsized into a handheld mobile device, thereby limiting its mobility. We introduce a new architecture which uses a personal digital assistant (PDA) to remotely control and access services from the VPSCR. Following a modular approach, we designed a remote agent on the VPSCR to advertise services that a PDA can discover and subscribe. This provides a PDA with the capability to remotely control the VPSCR according to a public safety officers own needs. The data link between the PDA and the VPSCR utilizes existing fixed infrastructures such as the 802.11 Wi-Fi or Bluetooth standards, enabling the PDA to extend its operational range.

Image segmentation and its performance evaluation are very difficult but important problems in computer vision. A major challenge in segmentation evaluation comes from the fundamental conflict between generality and objectivity: For general-purpose segmentation, the ground truth and segmentation accuracy may not be well defined, while embedding the evaluation in a specific application, the evaluation results may not be extended to other applications. We present in this paper a new benchmark for evaluating image segmentation. Specifically, we formulate image segmentation as identifying the single most perceptually salient structure from an image. We collect a large variety of test images that conforms to this specific formulation, construct unambiguous ground truth for each image, and define a reliable way to measure the segmentation accuracy. We then present two special strategies to further address two important issues: (a) the most salient structures in some real images may not be unique or unambiguously defined, and (b) many available image-segmentation methods are not developed to directly extract a single salient structure. Finally, we apply this benchmark to evaluate and compare the performance of several state-of-the-art image-segmentation methods, including the normalized-cut method, the level-set method, the efficient graph-based method, the mean-shift method, and the ratio-contour method.