Graphs are widely used as a popular representation of the network structure of connected data. Graph data can be found in a wide spectrum of application domains such as social systems, ecosystems, biological networks, knowledge graphs, and information systems. With the continuous penetration of artificial intelligence technologies, graph learning (i.e., machine learning on graphs) has been gaining rapidly increasing attention from both researchers and practitioners. Graph learning has proved to be effective for many tasks, such as classification, link prediction, matching, and anomaly detection. Generally, graph learning methods extract relevant features of graphs by taking advantage of machine learning algorithms. In this survey, we present a comprehensive overview on the state of the art of graph learning. Special attention is paid to four categories of existing graph learning methods, including graph signal processing, matrix factorization, random walk, and deep learning. Major models and algorithms under these categories are reviewed respectively. Graph learning applications are examined. In addition, we discuss several promising research directions in this field.

A random walk is known as a random process which describes a path including a succession of random steps in the mathematical space. It has increasingly been popular in various disciplines such as mathematics and computer science. Furthermore, in quantum mechanics, quantum walks can be regarded as quantum analogues of classical random walks. Classical random walks and quantum walks can be used to calculate the proximity between nodes and extract the topology in the network. Various random walk related models can be applied in different fields, which is of great significance to downstream tasks such as link prediction, recommendation, computer vision, semi-supervised learning, and network embedding. In this article, we aim to provide a comprehensive review of classical random walks and quantum walks. We first review the knowledge of classical random walks and quantum walks, including basic concepts and some typical algorithms. We also compare the algorithms based on quantum walks and classical random walks from the perspective of time complexity. Then we introduce their applications in the field of computer science. Finally we discuss the open issues from the perspectives of efficiency, main-memory volume, and computing time of existing algorithms. This study aims to contribute to this growing area of research by exploring random walks and quantum walks together.

As a promising research area in Internet of Things (IoT), Internet of Vehicles (IoV) has attracted much attention in wireless communication and network. In general, vehicle localization can be achieved by the Global Positioning Systems (GPS). However, in some special scenarios, such as cloud cover, tunnels or some places where the GPS signals are weak, GPS cannot perform well. The continuous and accurate localization services cannot be guaranteed. In order to improve the accuracy of vehicle localization, an assistant vehicle localization method based on Direction-of-Arrival (DOA) estimation is proposed in this paper. The assistant vehicle localization system is composed of three Base Stations (BS) equipped with a Multiple Input Multiple Output (MIMO) array. The locations of vehicles can be estimated if the positions of the three BSs and the DOAs of vehicles estimated by the BSs are known. However, the DOA estimated accuracy maybe degrade dramatically when the electromagnetic environment is complex. In the proposed method, a Sparse Bayesian Learning (SBL) based robust DOA estimation approach is first proposed to achieve the off-grid DOA estimation of the target vehicles under the condition of non-uniform noise, where the covariance matrix of non-uniform noise is estimated by a Least Squares (LS) procedure, and a grid refinement procedure implemented by finding the roots of a polynomial is performed to refine the grid points to reduce the off-grid error. Then, according to the DOA estimation results, the target vehicle is cross-located once by each two BSs in the localization system. Finally, robust localization can be realized based on the results of three-time cross-location. Plenty of simulation results demonstrate the effectiveness and superiority of the proposed method.

In this paper, a novel system architecture including a massive multi-input multi-output (MIMO) or a reconfigurable intelligent surface (RIS) and multiple autonomous vehicles is considered in vehicle location systems. The location parameters of autonomous vehicles can be estimated based on the deep unfolding technique, which is a recent advance of deep learning. Traditional vehicle location methods such as the global position system (GPS) can only locate the target vehicles with relatively low accuracy. The super resolution cannot be achieved when two vehicles are too close, which means that the safety incidents exist when autonomous vehicles are deployed in future intelligent transportation systems (ITS). Different from the existing massive MIMO or RIS equipped with a regular array such as uniform rectangular array (URA) and uniform circular array (UCA), we exploit a massive MIMO or a RIS equipped with a conformal array extended from traditional regular array. First, the rotation from the global coordinate system to the local coordinate system is achieved based on geometric algebra. Second, 2D-DOA estimation of autonomous vehicles is modeled as a novel block sparse recovery problem. Third, the deep network architecture SBLNet is implemented to learn the nonlinear characteristic from the DOAs of autonomous vehicles and the data received by massive MIMOs or RISs. The 2D-DOA and polarization parameters can be estimated based on SBLNet with relatively low computational complexity. Simulation results demonstrate that SBLNet performs better than the state-of-the-art methods in terms of estimation accuracy and successful probability. The SBLNet is also suitable for the practical scenario considering fast moving autonomous vehicles, while, the traditional block sparse recovery methods fail in this complex scenario.

Forest fire monitoring plays an important role in forest resource protection. Although satellite remote sensing is an effective way for forest fire monitoring, satellite-based methods can only monitor huge scale forest areas, and they are weak in predicting the specific areas of forest fires. In this paper, we first propose an unmanned aerial vehicle (UAV)-enabled system architecture consisting of multiple industrial internet of things (IIoTs), in which the collected data by sensors in IIoTs can be delivered to UAVs for processing directly. As the sensors of IIoTs are deployed to monitor different indexes of forest fires, fully considering the priority constraints among sensors can guarantee a quick response of forest fire monitoring. Thus, the priority constraints among the sensors are taken into consideration in this system architecture, and the objective to minimize the maximum response time of forest fire monitoring. To search for the optimal UAV resource allocation strategy, a learning based cooperative particle swarm optimization algorithm (LCPSO) with Markov random field (MRF) based decomposition strategy is proposed. The solution space of UAV resource allocation is decomposed into sub solution space according to the decomposed decision variables by the MRF network structure, and the optimal resource allocation strategy is searched by LCPSO in multiple sub solution space cooperatively. Three simulation experiments on two datasets are designed, and the simulation results compared with the state-of-the-art methods verify the validity of LCPSO, which are reflected by the quickest response time of forest fire monitoring.

Local administrations and governments aim at leveraging wireless communications and Internet of Things (IoT) technologies to manage the city infrastructures and enhance the public services in an efficient and sustainable manner. Furthermore, they strive to adopt smart and cost-effective mobile applications to deal with major urbanization problems, such as natural disasters, pollution, and traffic congestion. Mobile crowdsourcing (MCS) is known as a key emerging paradigm for enabling smart cities, which integrates the wisdom of dynamic crowds with mobile devices to provide decentralized ubiquitous services and applications. Using MCS solutions, residents (i.e., mobile carriers) play the role of active workers who generate a wealth of crowdsourced data to significantly promote the development of smart cities. In this article, we present an overview of state-of-the-art technologies and applications of MCS in smart cities. First, we provide an overview of MCS in smart cities and highlight its major characteristics. Second, we introduce the general architecture of MCS and its enabling technologies. Third, we study novel applications of MCS in smart cities. Finally, we discuss several open problems and future research challenges in the context of MCS in smart cities.