Search

A novel scheme of edge detection based on the physical law of diffusion is presented in this paper. Though the most current studies are using data based methods such as deep neural networks, these methods on machine learning need big data of labeled ground truth as well as a large amount of resources for training. On the other hand, the widely used traditional methods are based on the gradient of the grayscale or color of images with using different sorts of mathematical tools to accomplish the mission. Instead of treating the outline of an object in an image as a kind of gradient of grayscale or color, our scheme deals with the edge detection as a character of an energy diffusing in the space of media such as charge-coupled device. By using the characteristic function of diffusion, the information of the energy will be extracted. The scheme preserves the structural information of images very well. Because it comes from the inhere law of images’ physical property, it has a unified mathematical framework for images’ edge detection under different conditions, for example, multiscales, diferent light conditions, and so on. Moreover, it has low computational complexity. [ABSTRACT FROM AUTHOR]

It is of significant importance to provide network-wide Quality of Service (QoS) for a wide range of event detection applications in wireless sensor networks. This paper investigates the important problem of QoS provision to abnormal event detection. For event detection applications, there are two key performance metrics, i.e., detection probability and detection latency. This paper considers both metrics, aiming to provide statistical QoS for abnormal event detection. It is, however, a challenging issue because of stringent resource constraints of sensor nodes and unpredictable randomness of physical events. We propose a probabilistic approach to statistical QoS provision for event detection in sensor networks. We propose a distributed algorithm that iteratively determines the active probability of each sensor node. The probability is kept small for energy efficiency but sufficiently large to provide the required detection QoS. Our approach is flexible and can detection QoS customized by applications. Comprehensive simulation experiments have been conducted, which demonstrate that our approach is able to provide the required detection QoS for event detection and achieves considerably longer the system lifetime compared with other competing schemes. [ABSTRACT FROM AUTHOR]

Convolutional sparse coding (CSC) can learn representative shift-invariant patterns from data. However, existing CSC methods assume the Gaussian noise, which can be restrictive in some challenging applications. In this paper, we propose a generalized CSC model capable of handling complicated unknown noise. The noise is modeled by the Gaussian mixture model, which can approximate any continuous probability density function. The Expectation-Maximization algorithm is used to solve the resultant learning problem. For efficient optimization, the crux is to speed up the convolution in the frequency domain while keeping the other computations involving the weight matrix in the spatial domain. We design an efficient solver for the weighted CSC problem in the M-step. The dictionary and codes are updated simultaneously by an efficient nonconvex accelerated proximal gradient algorithm. The resultant procedure, called generalized convolutional sparse coding (GCSC), obtains the same space complexity and a smaller running time than existing CSC methods (which are limited to the Gaussian noise). Extensive experiments on synthetic and real-world noisy data sets validate that GCSC can model the noise effectively and obtain high-quality filters and representations. [ABSTRACT FROM AUTHOR]

Learning human mobility behaviors from location-sensing data are crucial to mobility data mining because of its potential to address a range of analytical purposes in mobile context reasoning, including exploration, inference, and prediction. However, existing approaches suffer from two practical problems: temporal and spatial sparsity. To address these shortcomings, we present two unsupervised learning methods to model the mobility behaviors of multiple users (i.e., a population), considering efficiency and accuracy. These methods intelligently overcome the sparsity in individual data by seeking temporal commonality among users' heterogeneous location behaviors. The advantages of our models are highlighted through experiments on several real-world mobility data sets, which also show how our methods can realize the three analytical purposes in a unified manner. [ABSTRACT FROM AUTHOR]

Measuring learner cognition based on their problem-solving performance is a joint discipline of cognitive psychology and machine learning. In the case of learner problem-solving, the interaction between learner and knowledge forms a typical type of signed interaction graph. Interaction graphs are a widely used and effective solution to model the relationships between interacting entities. However, most of previous interaction graph methods are inclined to the observed interactions as positive links but they often fail to consider unobserved and negative links, which leads to an insufficiency in capturing the complete cognition/mis-cognition proximity information. To address this limitation, we propose a knowledge graph representation learning method that is based on signed knowledge interaction network (SKIN). We explicitly model the correct/incorrect cognitive performance as the positively + /negatively − signed links in the graph. The model simultaneously measures the nodes' local and global proximity , and then preserves them in the learned knowledge embedding. We architect a pairwise neural network that is based on a tri-sampling strategy and a sign-driven distance measuring objective function. The network generates knowledge representations by maximizing the knowledge distance between oppositely-signed pairs and minimizing the distance between identically-signed pairs. Our experimental results show the learned knowledge embedding demonstrates a desired Euclidean property and can be visualized with clear classification boundaries. We also show it can power downstream tasks such as learner-performance-prediction. The learned embeddings generate promising prediction scores on this task when compared to several methods in network sign prediction and learner-performance-prediction. [ABSTRACT FROM AUTHOR]

Map-matching is a process that aligns location points on a digital map and it is an essential step in location-based services. However, regular map-matching methods cannot archive very high accuracy due to the errors in raw location data and the complexity of road networks. Hence, the final resort for map matching is often through manual annotation, which is human labour intensive. Therefore, we propose iMatching , an interactive system for map-matching which greatly reduces annotation cost and achieves a high accuracy through an active learning approach. Specifically, we model the mapping of a sequence of location points to a road network as a hidden Markov model and automatically generate an initial result. Then, we select error-prone points on the trajectory and guide the annotator to review, and possibly correct, the results. Our evaluation on both real-world and synthetic data demonstrates that iMatching has a better performance comparing with the existing methods. [ABSTRACT FROM AUTHOR]

Machine learning has been highly successful in data-intensive applications but is often hampered when the data set is small. Recently, Few-shot Learning (FSL) is proposed to tackle this problem. Using prior knowledge, FSL can rapidly generalize to new tasks containing only a few samples with supervised information. In this article, we conduct a thorough survey to fully understand FSL. Starting from a formal definition of FSL, we distinguish FSL from several relevant machine learning problems. We then point out that the core issue in FSL is that the empirical risk minimizer is unreliable. Based on how prior knowledge can be used to handle this core issue, we categorize FSL methods from three perspectives: (i) data, which uses prior knowledge to augment the supervised experience; (ii) model, which uses prior knowledge to reduce the size of the hypothesis space; and (iii) algorithm, which uses prior knowledge to alter the search for the best hypothesis in the given hypothesis space. With this taxonomy, we review and discuss the pros and cons of each category. Promising directions, in the aspects of the FSL problem setups, techniques, applications, and theories, are also proposed to provide insights for future research. [ABSTRACT FROM AUTHOR]

Autonomous learning diagnostics, where the students' strengths and weaknesses are disclosed from their observed performance data, is a challenging task in e-learning systems. Current student knowledge models can alleviate some of the problems in learning (i.e. predicting student performance) but they neglect learning diagnostics, which is based on causal reasoning. To this end, we propose a novel heterogeneous attention interpreter with a maximum entropy regularizer on top of a student knowledge model to achieve explainable learning diagnostics. Our model segregates the impact of the homogeneous knowledge points, while promoting the heterogeneous relatives by maximizing their chance to contribute to the prediction. We also propose a multi-spatial knowledge representation that is readily generalizable to other data-driven educational tasks. Extensive experiments on real-world datasets reveal that the proposed method is able to enhance the model's explanatory power, hence increases the trustworthiness towards learning diagnostics. It also brings notable improvement in accuracy in the student performance prediction task. The findings in this paper are adoptable to various types of e-learning systems to assist teachers to gain insights into student learning states and diagnose learning problems. • Autonomously diagnosing learning problems in e-learning systems can be challenging due to the lack of teacher resources (teachers) in e-learning. This work proposes to tackle the explainable learning diagnostics problem using attention-based explanation mechanisms by performing target-source relation prediction. The findings are adoptable to various types of e-learning systems to gain insights into their learner states and diagnose their learning problems. • This paper identifies the importance to use the 'close relatives of knowledge to generate prediction decisions in a knowledge tracing model. We propose a Heterogeneous Attention Relative Detector and a Maximum Entropy Regularizer to detect those relatives, i.e. knowledge relation discovery. • We propose a multi-spatial representation of knowledge in expressing knowledge relations in finer-granularity and low-dimensionality, which can be readily generalized to other data-driven educational tasks. • We provide a different perspective on knowledge graph completion and/or construction. In this perspective, we exploit the interaction data to uncover inner knowledge relations (links) where there is no existing relation data to learn. Our method is effective in settings where the knowledge space is relatively small while the interaction space is very large. [ABSTRACT FROM AUTHOR]

In modern big data warehouse systems, we observe a common phenomenon that a column of data values can be derived from one or several other columns by transforming and concatenating these columns. We call this relationship between columns a Soft Concatenation Mapping (SCM). SCMs imply significant redundancy in the schema or data, and therefore can be exploited for data integration or data compression. In this paper, we formalize the problem of SCM detection and prove it is NP-hard. We then propose efficient approximate algorithms to detect all SCMs or an optimal set of SCMs in a table. Our experiments on both real-world and synthetic datasets show promising results. [ABSTRACT FROM AUTHOR]

Abstract Animation is a commonly used technique in information visualization for smooth transitions between different views. When observing animations of moving objects, people often need to track several specific objects while identify the major trend of movement simultaneously. In this paper, we propose a novel focus+context grouping technique to facilitate target tracking and trend identification. It divides objects into several groups based on a comprehensive tree cut algorithm and generates a staggering animation in which groups are animated sequentially. A balance between efficiency and accuracy is achieved for an effective animation planning. To evaluate the effectiveness of the proposed technique, a carefully designed user study is conducted. The results indicate that focus+context grouping is effective for users to track targets without losing context (i.e., major trend of movement). Based on the study, we discuss advantages and limitations of the proposed grouping technique and conclude with design implications. [ABSTRACT FROM AUTHOR]

Ocean and water basically cover the major parts of our planet. To obtain the best utilization of the underlying resources on these parts of the Earth, people have made some research advancements. Specifically, the research on underwater wireless acoustic sensor networks (UWA-SNs) has made great progress. However, wide deployment of UWA-SNs is far from a reality due to several reasons. One important reason is that offshore deployment and field-level experiments of ocean-centric applications are both expensive and labor intensive. Other alternatives to attain this objective are to conduct simulation or experimentation that can reduce cost and accelerate the research activities and their outcomes. However, designing efficient and reliable simulation and experimentation platforms have proven to be more challenging beyond the expectation. In this article, we explore the main techniques (including their pros and cons) and components to develop simulation and experimentation platforms and provide a comprehensive survey report in this area. We classify simulation and experimentation platforms based on some typical criteria and then provide useful guidelines for researchers on choosing suitable platforms in accordance with their requirements. Finally, we address some open and un-resolved issues in this context and provide some suggestions on future research. [ABSTRACT FROM AUTHOR]

Wi-Fi, which enables convenient wireless access to Internet services, has become integral to our modern lives. With widely-deployed Wi-Fi infrastructure, modern people can enjoy a variety of online services such as web browsing, online shopping, social interaction, and e-commerce almost at any time and any place. Traditionally, the most significant functionality of Wi-Fi is to enable high-throughput data communication between terminal devices and the Internet. However, beyond that, we observe that a novel type of system based on commodity Wi-Fi is increasingly attracting intense academic interest. Without hardware modification and redeployment, researchers are exploiting channel state information output by commodity Wi-Fi and transforming existing Wi-Fi systems into radar-like ones that can recognize human behavior along with data communication. This fancy functionality is tremendously expanding the boundaries of Wi-Fi to a new realm and triggering revolutionary applications in the context of the Internet of Things. In this article, we provide a guide to and introduce the impressive landscape of this new realm. [ABSTRACT FROM AUTHOR]

Deploying wireless sensor networks in the ocean poses many challenges due to the harsh conditions of the ocean and the nonnegligible node mobility. In this paper, we propose hybrid ocean sensor networks called drifting restricted floating ocean sensor networks (DR-OSNs) for long-term maritime surveillance monitoring tasks, which combines both the advantages of wireless sensor networks and underwater wireless acoustic sensor networks. We present a localization scheme termed localization for double-head maritime sensor networks (LDSN) for DR-OSNs, which leverages the unique characteristics of DR-OSNs to establish the whole localization system after the network is deployed from a plane or a ship, and it does not need the presence of designated anchor nodes deployed underwater. The whole localization process consists of three steps with algorithms self-moored node localization (SML), underwater sensor localization (USD), and floating-node localization algorithm (FLA). The first step is for the super group nodes to localize their underwater moored nodes via an SML algorithm by leveraging the free-drifting movement of their surface nodes. Once the moored nodes in the super group nodes have localized themselves, they turn into anchor nodes underwater. Thus, in the second step, with the help of these new anchor nodes, the unlocalized underwater moored nodes use the USD algorithm to localize their positions. In the last step, when the free-drifting floating nodes without a Global Positioning System (GPS) module need to know their instant position, they apply the FLA to figure out their position. We conduct extensive simulations to evaluate the scheme, with the results indicating that LDSN achieves high localization accuracy and is an effective localization scheme for DR-OSNs. [ABSTRACT FROM PUBLISHER]

With the marvelous development of wireless techniques and ubiquitous deployment of wireless systems indoors, myriad indoor location-based services (ILBSs) have permeated into numerous aspects of modern life. The most fundamental functionality is to pinpoint the location of the target via wireless devices. According to how wireless devices interact with the target, wireless indoor localization schemes roughly fall into two categories: device based and device free. In device-based localization, a wireless device (e.g., a smartphone) is attached to the target and computes its location through cooperation with other deployed wireless devices. In device-free localization, the target carries no wireless devices, while the wireless infrastructure deployed in the environment determines the target's location by analyzing its impact on wireless signals. This article is intended to offer a comprehensive state-of-the-art survey on wireless indoor localization from the device perspective. In this survey, we review the recent advances in both modes by elaborating on the underlying wireless modalities, basic localization principles, and data fusion techniques, with special emphasis on emerging trends in (1) leveraging smartphones to integrate wireless and sensor capabilities and extend to the social context for device-based localization, and (2) extracting specific wireless features to trigger novel human-centric device-free localization. We comprehensively compare each scheme in terms of accuracy, cost, scalability, and energy efficiency. Furthermore, we take a first look at intrinsic technical challenges in both categories and identify several open research issues associated with these new challenges. [ABSTRACT FROM AUTHOR]

Most existing connectivity-based localization algorithms require high node density which is unavailable in many large-scale sparse mobile networks. By analyzing large datasets of real user traces from Dartmouth and MIT, we observe that user mobility exhibits high spatiotemporal regularity and, more importantly, that user mobility is strongly correlated with the user' s social encounters (including so called Familiar Strangers). Motivated by these important observations, we propose a distributed localization scheme called SOMA that is particularly suitable for sparse mobile networks. To exploit the correlation between mobility and social encounters, we formulate the localization process as an optimization problem with the objective of maximizing the probability of visiting a sequence of locations when the user witnesses the given set of social encounters at different time. Employing the Hidden Markov Model, we design an efficient algorithm based on dynamic programming for solving the optimization problem. SOMA is fully distributed, in which each user only makes use of the connectivity information with other users. Since different users may have varying levels of mobility regularity, one critical challenge with SOMA is that a user with weak mobility regularity may result in poor localization accuracy. We introduce the concept of mobility irregularity to distinguish users. Then, one optimization is made to SOMA that allows a user with weak mobility regularity to leverage the locations from the users encounters. Experimental results based on large-scale real traces demonstrate that SOMA achieves much smaller localization error than many state-of-the-art localization schemes, but requires it minimal running time. [ABSTRACT FROM AUTHOR]

Multicast is a crucial routine operation for vehicular networks, which underpins important functions such as message dissemination and group coordination. As vehicles may distribute over a vast area, the number of vehicles in a given region can be limited which results in sparse node distribution in part of the vehicular network. This poses several great challenges for efficient multicast, such as network disconnection, scarce communication opportunities and mobility uncertainty. Existing multicast schemes proposed for vehicular networks typically maintain a forwarding structure assuming the vehicles have a high density and move at low speed while these assumptions are often invalid in a practical vehicular network. As more and more vehicles are equipped with GPS enabled navigation systems, the trajectories of vehicles are becoming increasingly available. In this work, we propose an approach called TMC to exploit vehicle trajectories for efficient multicast in vehicular networks. The novelty of TMC includes a message forwarding metric that characterizes the capability of a vehicle to forward a given message to destination nodes, and a method of predicting the chance of inter-vehicle encounter between two vehicles based only on their trajectories without accurate timing information. TMC is designed to be a distributed approach. Vehicles make message forwarding decisions based on vehicle trajectories shared through inter-vehicle exchanges without the need of central information management. We have performed extensive simulations based on real vehicular GPS traces and compared our proposed TMC scheme with other existing approaches. The performance results demonstrate that our approach can achieve a delivery ratio close to that of the flooding-based approach while the cost is reduced by over 80 percent. [ABSTRACT FROM PUBLISHER]

Geocast in vehicular networks aims to deliver a message to a target geographical region, which is useful for many applications such as geographic advertising. This is a highly challenging task in vehicular network environments due to the rare encounter opportunities and uncertainty caused by vehicular mobility. As more vehicles are equipped with on-board navigation systems, vehicle trajectories are ready for exploitation. We observe that a vehicle has a higher capability of delivering a message to the target region if its own future trajectory or trajectories of those vehicles to be encountered overlap the target region. Motivated by this observation, we develop a message forwarding metric, called coverage capability, to characterize the capability of a vehicle to successfully geocast the message. When calculating the coverage capability, we are facing the major challenge raised by the absence of accurate vehicle arrival time. Through an empirical study using real vehicular GPS traces of 2,600 taxis, we verify that the travel time of a vehicle, which is modeled as a random variable, follows the Gamma distribution. The travel time modeling helps us to make accurate predictions for inter-vehicle encounters. We perform extensive trace-driven simulations and the results show that our approach achieves 37.4 percent higher delivery ratio and 43.1 percent lower transmission overhead comparing with GPSR which is a representative geographic routing protocol. [ABSTRACT FROM AUTHOR]

Radio frequency (RF) based technologies play an important role in indoor localization, since Radio Signal Strength (RSS) can be easily measured by various wireless devices without additional cost. Among these, radio map based technologies (also referred as fingerprinting technologies) are attractive due to high accuracy and easy deployment. However, these technologies have not been extensively applied on real environment for two fatal limitations. First, it is hard to localize multiple objects. When the number of target objects is unknown, constructing a radio map of multiple objects is almost impossible. Second, environment changes will generate different multipath signals and severely disturb the RSS measurement, making laborious retraining inevitable. Motivated by these, in this paper, we propose a novel approach, called Line-of-sight radio map matching, which only reserves the LOS signal among nodes. It leverages frequency diversity to eliminate the multipath behavior, making RSS more reliable than before. We implement our system MODLoc based on TelosB sensor nodes and commercial 802.11 NICs with Channel State Information (CSI) as well. Through extensive experiments, it shows that the accuracy does not decrease when localizing multiple targets in a dynamic environment. Our work outperforms the traditional methods by about 60 percent. More importantly, no calibration is required in such environment. Furthermore, our approach presents attractive flexibility, making it more appropriate for general RF-based localization studies than just the radio map based localization. [ABSTRACT FROM AUTHOR]

Spectrum sensing is essential for the success of the cognitive radio networks. In traditional spectrum sensing schemes, secondary users (SUs) are responsible for the spectrum sensing which could be very time and resource consuming. It leads to a great deal of inefficiency in spectrum usage and introduces many practical challenges. To tackle these challenges and leverage the spectrum opportunity more efficiently, we propose a new system that provides a spectrum sensing service for SUs using dedicated wireless spectrum sensor networks (WSSNs). In this paper we focus on the sensing channel assignment problem in WSSNs and formulate the problem as a sensing effectiveness maximization problem (SEMP). We prove that SEMP is NP-complete under the ideal case, and show that more challenges arise in real environments. To address the issue, we systematically study the design tradeoff and critical factors when maximizing the sensing effectiveness. Based on these study results we propose a sensing channel assignment algorithm (SCAS). We conduct test-bed empirical investigations as well as comprehensive simulations. Performance evaluation results show that for both the scenarios of given deployments and manual deployments, SCAS is able to sense more channels to improve the sensing effectiveness. The improvement is up to 300 % and the average improvement is 150 % compared with other simple alternatives. [ABSTRACT FROM AUTHOR]