检索结果-内蒙古大学图书馆

Safeml: Safety monitoring of machine learning classifiers through statistical difference measure

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Aslansefat, Koorosh Sorokos, Ioannis Whiting, Declan Kolagari, Ramin Tavakoli Papadopoulos, Yiannis Department of Computer Science University of Hull Kingston upon HullHU6 7RX United Kingdom Fraunhofer Institute for Experimental Software Engineering IESE Fraunhofer-Gesellschaft Fraunhofer-Platz 1 Kaiserslautern67663 Germany Department of Computer Science University of Hull APD Communications Kingston upon HullHU6 7RX United Kingdom Faculty of Computer Science Nuremberg Tech Keßlerplatz 12 Nürnberg90489 Germany

Ensuring safety and explainability of machine learning (ML) is a topic of increasing relevance as data-driven applications venture into safety-critical application domains, traditionally committed to high safety standards that are not satisfied with an exclusive testing approach of otherwise inaccessible black-box systems. Especially the interaction between safety and security is a central challenge, as security violations can lead to compromised safety. The contribution of this paper to addressing both safety and security within a single concept of protection applicable during the operation of ML systems is active monitoring of the behavior and the operational context of the data-driven system based on distance measures of the Empirical Cumulative Distribution Function (ECDF). We investigate abstract datasets (XOR, Spiral, Circle) and current security-specific datasets for intrusion detection (CICIDS2017) of simulated network traffic, using statistical distance measures including the Kolmogorov-Smirnov, Kuiper, Anderson-Darling, Wasserstein and mixed Wasserstein-Anderson-Darling measures. Our preliminary findings indicate that there is a meaningful correlation between ML decisions and the ECDF-based distances measures of the input features. Thus, they can provide a confidence level that can be used for a) analyzing the applicability of the ML system in a given field (safety/security) and b) analyzing if the field data was maliciously manipulated1. Copyright © 2020, The Authors. All rights reserved.

关键词： Intrusion detection

Towards Ubiquitous AI in 6G with Federated Learning

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Xiao, Yong Shi, Guangming Krunz, Marwan School of Electronic Information and Communications Huazhong University of Science and Technology Wuhan430074 China School of Artificial Intelligence Xidian University Xi’an Shaanxi710071 China Department of Electrical and Computer Engineering University of Arizona TucsonAZ85710 United States

—With 5G cellular systems being actively deployed worldwide, the research community has started to explore novel technological advances for the subsequent generation, i.e., 6G. It is commonly believed that 6G will be built on a new vision of ubiquitous AI, an hyper-flexible architecture that brings human-like intelligence into every aspect of networking systems. Despite its great promise, there are several novel challenges expected to arise in ubiquitous AI-based 6G. Although numerous attempts have been made to apply AI to wireless networks, these attempts have not yet seen any large-scale implementation in practical systems. One of the key challenges is the difficulty to implement distributed AI across a massive number of heterogeneous devices. Federated learning (FL) is an emerging distributed AI solution that enables data-driven AI solutions in heterogeneous and potentially massive-scale networks. Although it still in an early stage of development, FL-inspired architecture has been recognized as one of the most promising solutions to fulfill ubiquitous AI in 6G. In this article, we identify the requirements that will drive convergence between 6G and AI. We propose an FL-based network architecture and discuss its potential for addressing some of the novel challenges expected in 6G. Future trends and key research problems for FL-enabled 6G are also discussed. Copyright © 2020, The Authors. All rights reserved.

关键词： 5G mobile communication systems

Perfect absorption in a metasurface-programmable complex scattering enclosure

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Imani, Mohammadreza F. Smith, David R. del Hougne, Philipp Department of Electrical and Computer Engineering Duke University DurhamNC27708 United States Institut d'Electronique et de Télécommunications de Rennes CNRS UMR 6164 Université de Rennes 1 Rennes35000 France

Achieving the very special condition of perfect absorption (PA) in a complex scattering enclosure promises to enable a wealth of applications in secure communication, precision sensing, wireless power transfer, analog signal processing and random lasing. Consequently, a lot of recent research effort was dedicated to proposing wavefront shaping protocols to implement coherent PA in a complex scattering environment. Here, instead of shaping the impinging wave front and tuning the system's attenuation, we propose the conceptually different route of solely tweaking the randomness of the complex scattering environment in order to achieve PA. We provide an experimental proof-of-concept in the microwave domain where we tune the randomness of a three-dimensional chaotic cavity with a programmable metasurface. We systematically investigate the achievability and extreme sensitivity of the PA condition. Our technique can impose a PA condition at over hundred distinct frequencies within a small frequency band, which allows us to propose and experimentally demonstrate a concrete practical application: receiver-powered secure wireless communication in a complex scattering enclosure. Our fundamentally new perspective on PA is applicable to all types of wave phenomena and our results foreshadow the large potential of this novel tool for minute wave control in sensing, communication and energy transfer. Copyright © 2020, The Authors. All rights reserved.

关键词： Enclosures

Preparing a commercial quantum key distribution system for certification against implementation loopholes

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Makarov, Vadim Abrikosov, Alexey Chaiwongkhot, Poompong Fedorov, Aleksey K. Huang, Anqi Kiktenko, Evgeny Petrov, Mikhail Ponosova, Anastasiya Ruzhitskaya, Daria Tayduganov, Andrey Trefilov, Daniil Zaitsev, Konstantin Russian Quantum Center Skolkovo Moscow121205 Russia Vigo Quantum Communication Center University of Vigo VigoE-36310 Spain NTI Center for Quantum Communications National University of Science and Technology MISiS Moscow119049 Russia Department of Physics Faculty of Science Mahidol University Bangkok10400 Thailand Institute for Quantum Computing University of Waterloo WaterlooONN2L 3G1 Canada Department of Physics and Astronomy University of Waterloo WaterlooONN2L 3G1 Canada Bangkok10110 Thailand QRate Skolkovo Moscow143026 Russia Institute for Quantum Information State Key Laboratory of High Performance Computing College of Computer Science and Technology National University of Defense Technology Changsha410073 China Steklov Mathematical Institute Russian Academy of Sciences Moscow119991 Russia atlanTTic Research Center University of Vigo VigoE-36310 Spain School of Telecommunication Engineering Department of Signal Theory and Communications University of Vigo VigoE-36310 Spain National Research University Higher School of Economics Moscow101000 Russia

A commercial quantum key distribution (QKD) system needs to be formally certified to enable its wide deployment. The certification should include the system’s robustness against known implementation loopholes and attacks that exploit them. Here we ready a fiber-optic QKD system for this procedure. The system has a prepare-and-measure scheme with decoy-state BB84 protocol, polarisation encoding, qubit source rate of 312.5 MHz, and is manufactured by QRate. We detail its hardware and post-processing. We analyse the hardware for known implementation loopholes, search for possible new loopholes, and discuss countermeasures. We then amend the system design to address the highest-risk loopholes identified. We also work out technical requirements on the certification lab and outline its possible structure. Copyright © 2023, The Authors. All rights reserved.

关键词： Quantum cryptography

Physical layer security enhancement using artificial noise in cellular vehicle-to-everything (C-V2X) networks

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Wang, Chao Li, Zan Xia, Xiang-Gen Shi, Jia Si, Jiangbo Zou, Yulong Integrated Service Networks Lab Xidian University Xi’an710071 China College of Communications Engineering Xidian University Xi’an710071 China Department of Electrical and Computer Engineering University of Delaware NewarkDE19716 United States School of Telecommunications and Information Engineering Nanjing University of Posts and Telecommunications Nanjing210003 China

The secure transmission of confidential information in cellular vehicle-to-everything (C-V2X) communication networks is vitally important for user’s personal safety. However, for C-V2X there have not been much studies on the physical layer security (PLS). Since artificial noise (AN) and secure beamforming are popular PLS techniques for cellular communications, in this paper we investigate the potential of these PLS techniques for enhancing the security of C-V2X networks. In particular, leveraging stochastic geometry, we study the PLS of an AN assisted C-V2X network, where the locations of legitimate vehicular nodes, malicious vehicular nodes and road side units (RSUs) are modeled by Cox processes driven by a common Poisson line process (PLP), and the locations of cellular base stations (BSs) are modeled by a two-dimensional (2D) Poisson point process (PPP). Based on the maximum signal-to-interference-ratio (SIR) association scheme, we calculate the coverage probability of the network. We also derive bounds on the secrecy probability, which are validated by simulation results. Moreover, we obtain an analytical result of the effective secrecy throughput for characterizing the reliability and security of wiretap channels. Simulation results are given to validate the analytical result, and provide interesting insights into the impact of network parameters on the achievable secrecy performance. Simulation results show that a larger array antenna can provide a better robustness of the secure transmission strategy, and the optimal power allocation ratio between confidential information and AN remains almost unchanged for different numbers of antennas. Copyright © 2020, The Authors. All rights reserved.

关键词： Geometry

Review on secure data aggregation in Wireless Sensor Networks

学校读者我要写书评

暂无评论

IOP Conference Series: Materials Science and engineering 2021年第1期1076卷

作者： Mohammed Salah Abood Hua Wang(王华) Hussain Falih Mahdi Mustafa Maad Hamdi Ahmad S. Abdullah School of Information and Electronics Beijing Institute of Technology Haidian District Beijing 100081 China Computer Engineering Department Faculty of Engineering University of Diyala Baquba Diyala Province Iraq. Department of Computer Engineering Techniques Al-Maarif University College Al-Anbar Iraq. Department of Communications College of Engineering University of Diyala Baqubah Iraq.

Recent advances in wireless sensor networks ( WSNs) have inspired various new applications, including target tracking and monitoring of ecosystems. Nevertheless, a large amount of data energy consumption of WSNs is utilized by data communication. However, data aggregation techniques can eliminate redundant data being transferred back to the base station and thereby lessening the consumption of energy. Security features like data integrity, confidentiality and accumulation of latest data is of great importance if WSN is utilized in extremely hostile or remote settings whereby the sensors are vulnerable to node failure and breach. Currently, there is prospect for securing data aggregation using WSN. In light of this, security features in data aggregation using WSN will be examined in current study. Next, the adversarial model that can be used in any aggregation scheme is then evaluated. Moreover, the latest and most sophisticated secured data aggregation scheme, proposed will be examined and categorized into two groups which are classified according to the number of aggregator nodes and the existence of the verification term. In this paper, created a table that showed the advantages and limitations of each algorithm that was adopted by researchers, shown in Table 3. Last, a computational approach is recommended to build a modern design with minimal protection preconditions against various types of challenges. The conceptual framework aims to provide an advanced understanding and enable the process of evaluation.

关键词：

Single-Shot Diagnosis of Electron Energy Evolution via Streaked Betatron X Rays in a Curved Laser Wakefield Accelerator

学校读者我要写书评

暂无评论

Physical Review Letters 2024年第22期132卷 225001-225001页

作者： Y. Ma J. A. Cardarelli P. T. Campbell S. Fourmaux R. Fitzgarrald M. D. Balcazar A. F. Antoine N. F. Beier Q. Qian A. E. Hussein B. Kettle S. R. Klein K. Krushelnick Y. F. Li S. P. D. Mangles G. Sarri D. Seipt V. Senthilkumaran M. J. V. Streeter L. Willingale A. G. R. Thomas Gérard Mourou Center for Ultrafast Optical Science University of Michigan Ann Arbor Michigan 48109-2099 USA Institut National de la Recherche Scientifique - Énergie Matériaux et Télécommunications Université du Québec (INRS-EMT) 1650 Lionel Boulet Varennes Québec J3X 1P7 Canada Department of Electrical and Computer Engineering University of Alberta Edmonton Alberta T6G 1H9 Canada The John Adams Institute for Accelerator Science Imperial College London London SW7 2AZ United Kingdom Institute of Physics Chinese Academy of Sciences Beijing 100190 China School of Mathematics and Physics Queen’s University Belfast Belfast BT7 1NN United Kingdom Helmholtz Institut Jena Fröbelstieg 3 Jena 07743 Germany GSI Helmholtzzentrum für Schwerionenforschung GmbH Planckstraße 1 64291 Darmstadt Germany

We report on an experimental observation of the streaking of betatron x rays in a curved laser wakefield accelerator. The streaking of the betatron x rays was realized by launching a laser pulse into a plasma with a transverse density gradient. By controlling the plasma density and the density gradient, we realized the steering of the laser driver, electron beam, and betatron x rays simultaneously. Moreover, we observed an energy-angle correlation of the streaked betatron x rays and utilized it in diagnosing the electron acceleration process in a single-shot mode. Our work could also find applications in advanced control of laser beam and particle propagation. More importantly, the angular streaked betatron x ray has an intrinsic spatiotemporal correlation, which makes it a promising tool for single-shot pump-probe applications.

关键词： Laser wakefield acceleration Plasma acceleration & new acceleration techniques X-ray generation in plasmas X-ray techniques

BiCANet: Bi-directional contextual aggregating network for image semantic segmentation

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhou, Quan Cong, Dechun Kang, Bin Wu, Xiaofu Zheng, Baoyu Lu, Huimin Latecki, Longin Jan National Engineering Research Center of Communications and Networking Nanjing University of Posts & Telecommunications Nanjing Jiangsu China Department of Internet of Things Nanjing University of Posts & Telecommunications Nanjing Jiangsu China Department of Mechanical and Control Engineering Kyushu Institute of Technology Kitakyushu Fukuoka Japan Department of Computer and Information Sciences Temple University PhiladelphiaPA United States

Exploring contextual information in convolutional neural networks (CNNs) has gained substantial attention in recent years for semantic segmentation. This paper introduces a Bi-directional Contextual Aggregating Network, called BiCANet, for semantic segmentation. Unlike previous approaches that encode context in feature space, BiCANet aggregates contextual cues from a categorical perspective, which mainly consists of three parts: contextual condensed projection block (CCPB), bidirectional context interaction block (BCIB), and muti-scale contextual fusion block (MCFB). More specifically, CCPB learns a category-based mapping through a split-transform-merge architecture, which condenses contextual cues with different receptive fields from intermediate layer. BCIB, on the other hand, employs dense skipped-connections to enhance the class-level context exchanging. Finally, MCFB integrates multi-scale contextual cues by investigating short- and long-ranged spatial dependencies. To evaluate BiCANet, we have conducted extensive experiments on three semantic segmentation datasets: PASCAL VOC 2012, Cityscapes, and ADE20K. The experimental results demonstrate that BiCANet outperforms recent state-of-the-art networks without any postprocess techniques. Particularly, BiCANet achieves the mIoU score of 86.7%, 82.4% and 38.66% on PASCAL VOC 2012, Cityscapes and ADE20K testset, respectively. Our code is open-source and is publicly available at https://***/cdcnjupt/BCANet. Copyright © 2020, The Authors. All rights reserved.

关键词： Semantic Segmentation

Robust Precoding in Massive MIMO: A Deep Learning Approach

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Shi, Junchao Wang, Wenjin Yi, Xinping Gao, Xiqi Li, Geoffrey Ye National Mobile Communications Research Laboratory Southeast University Nanjing210096 China Department of Electrical Engineering and Electronics University of Liverpool L69 3BX United Kingdom School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA30332 United States

In this paper, we consider massive multiple-input-multiple-output (MIMO) communication systems with a uniform planar array (UPA) at the base station (BS) and investigate the downlink precoding with imperfect channel state information (CSI). By exploiting both instantaneous and statistical CSI, we aim to design precoding vectors to maximize the ergodic rate (e.g., sum rate, minimum rate and etc.) subject to a total transmit power constraint. To maximize an upper bound of the ergodic rate, we leverage the corresponding Lagrangian formulation and identify the structural characteristics of the optimal precoder as the solution to a generalized eigenvalue problem. As such, the high-dimensional precoder design problem turns into a low-dimensional power control problem. The Lagrange multipliers play a crucial role in determining both precoder directions and power parameters, yet are challenging to be solved directly. To figure out the Lagrange multipliers, we develop a general framework underpinned by a properly designed neural network that learns directly from CSI. To further relieve the computational burden, we obtain a low-complexity framework by decomposing the original problem into computationally efficient subproblems with instantaneous and statistical CSI handled separately. With the off-line pretrained neural network, the online computational complexity of precoding is substantially reduced compared with the existing iterative algorithm while maintaining nearly the same performance. Copyright © 2020, The Authors. All rights reserved.

关键词： Lagrange multipliers