Xidian University

About: Xidian University is a education organization based out in Xi'an, China. It is known for research contribution in the topics: Antenna (radio) & Synthetic aperture radar. The organization has 32099 authors who have published 38961 publications receiving 431820 citations. The organization is also known as: University of Electronic Science and Technology at Xi'an & Xīān Diànzǐ Kējì Dàxué.

An Integrated Architecture for Software Defined and Virtualized Radio Access Networks with Fog Computing

Abstract: Today, billions of communication devices connecting to wireless networks impose serious challenges to network deployment, management, and data processing. Among all emerging technologies tackling these challenges, SDNs decouple the control plane from the data plane to provide network programmability, and virtualization can share network and radio resources among various applications. On the other hand, fog computing offloads computing services from the cloud to the edge of networks, offering real-time data services to nearby data terminals. In this article, we present an integrated architecture for software defined and virtualized radio access networks with fog computing. We propose a design of software as a service called OpenPipe, which enables network-level virtualization. To integrate SDNs and network virtualization with fog computing, we adopt a hybrid control model with two hierarchical control levels, where an SDN controller forms the higher level and local controllers comprise the lower level. Typical use cases of the proposed network architecture are validated through laboratory demonstrations.

Unsupervised Feature Extraction in Hyperspectral Images Based on Wasserstein Generative Adversarial Network

Abstract: Feature extraction (FE) is a crucial research area in hyperspectral image (HSI) processing. Recently, due to the powerful ability of deep learning (DL) to extract spatial and spectral features, DL-based FE methods have shown great potentials for HSI processing. However, most of the DL-based FE methods are supervised, and the training of them suffers from the absence of labeled samples in HSIs severely. The training issue of supervised DL-based FE methods limits their application on HSI processing. To address this issue, in this paper, a novel modified generative adversarial network (GAN) is proposed to train a DL-based feature extractor without supervision. The designed GAN consists of two components, which are a generator and a discriminator. The generator can focus on the learning of real probability distributions of data sets and the discriminator can extract spatial–spectral features with superior invariance effectively. In order to learn upsampling and downsampling strategies adaptively during FE, the proposed generator and discriminator are designed based on a fully deconvolutional subnetwork and a fully convolutional subnetwork, respectively. Moreover, a novel min–max cost function is designed for training the proposed GAN in an end-to-end fashion without supervision, by utilizing the zero-sum game relationship between the generator and discriminator. Besides, the proposed modified GAN replaces the original Jensen–Shannon divergence with the Wasserstein distance, aiming to mitigate the unstability and difficulty of the training of GAN frameworks. Experimental results on three real data sets validate the effectiveness of the proposed method.

Holes Detection in Anisotropic Sensornets: Topological Methods

Abstract: Wireless sensor networks (WSNs) are tightly linked with the practical environment in which the sensors are deployed. Sensor positioning is a pivotal part of main location-dependent applications that utilize sensornets. The global topology of the network is important to both sensor network applications and the implementation of networking functionalities. This paper studies the topology discovery with an emphasis on boundary recognition in a sensor network. A large mass of sensor nodes are supposed to scatter in a geometric region, with nearby nodes communicating with each other directly. This paper is thus designed to detect the holes in the topological architecture of sensornets only by connectivity information. Existent edges determination methods hold the high costs as assumptions. Without the help of a large amount of uniformly deployed seed nodes, those schemes fail in anisotropic WSNs with possible holes. To address this issue, we propose a solution, named PPA based on Poincare-Perelman Theorem, to judge whether there are holes in WSNs-monitored areas. Our solution can properly detect holes on the topological surfaces and connect them into meaningful boundary cycles. The judging method has also been rigorously proved to be appropriate for continuous geometric domains as well as discrete domains. Extensive simulations have been shown that the algorithm even enables networks with low density to produce good results.