检索结果-内蒙古大学图书馆

Favorable-Propagation-Exploited Variational Inference for Massive MIMO Detection

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2024年第9期73卷 14074-14079页

作者： He, Lanxin Wang, Zheng Gao, Zhen Liu, Lei Huang, Yongming Southeast Univ Sch Informat Sci & Engn Nanjing 210096 Peoples R China Beijing Inst Technol Zhuhai 519088 Peoples R China Zhejiang Univ Coll Informat Sci & Elect Engn Hangzhou 310027 Peoples R China

In this paper, we investigate the massive MIMO detection under the framework of mean-field variational inference (VI), which leads to a better detection trade-off between performance and complexity. First of all, by fully taking advantages of the favorable propagation characteristic of massive MIMO, the favorable-propagation-exploited variational inference (FPE-VI) algorithm is proposed for the low-complexity detection. Secondly, with respect to the system with KK transmitting and NN receiving antennas, the linear version of the FPE-VI detection is studied in detail, where its convergence is ensured when N/K > 1/(root 2-1)(2). Thirdly, by examining the evidence lower bound (ELBO) of the proposed FPE-VI, further optimization via the application of discrete Gaussian distribution is presented for extra performance gain. Finally, all the related theoretical analysis and the improved performance-complexity trade-off of FPE-VI are demonstrated by numerical results.

关键词： Massive MIMO Convergence approximation algorithms Receiving antennas Inference algorithms Complexity theory Transmitting antennas Massive MIMO detection approximate inference variational inference favorable propagation

来源：评论

学校读者我要写书评

暂无评论

Generalized Bilinear Factorization via Hybrid Vector Message Passing

引用

IEEE TRANSACTIONS ON SIGNAL PROCESSING 2024年 72卷 5675-5690页

作者： Jiang, Hao Yuan, Xiaojun Guo, Qinghua Univ Elect Sci & Technol China Natl Key Lab Wireless Commun Chengdu 611731 Peoples R China Univ Wollongong Sch Elect Comp & Telecommun Engn Wollongong NSW 2522 Australia

Generalized bilinear factorization (GBF), in which two matrices are recovered from noisy and typically compressed measurements of their product, arises in various applications such as blind channel-and-signal estimation, image completion, and compressed video foreground and background separation. In this paper, we formulate the GBF problem by unifying several existing bilinear inverse problems, and establish a novel hybrid vector message passing (HVMP) algorithm for GBF. The GBF-HVMP algorithm integrates expectation propagation (EP) and variational message passing (VMP) via variational free energy minimization, and exchanges matrix-variable messages in closed form. GBF-HVMP is advantageous over its counterparts in several aspects. For example, a matrix-variable message can characterize the correlations between the elements of the matrix, which is not possible in scalar-variable message passing;the hybrid of EP and VMP yields closed-form Gaussian messages associated with the bilinear constraints inherent in the GBF problem. We show that damping is unnecessary for GBF-HVMP to ensure convergence. We also show that GBF-HVMP performs close to the replica bound, and significantly outperforms state-of-the-art approaches in terms of both normalized mean squared error (NMSE) performance and computational complexity.

关键词： Message passing approximation algorithms Signal processing algorithms Sparse matrices Covariance matrices Correlation Vectors Inverse problems Inference algorithms Convergence Generalized bilinear factorization message passing variational free energy minimization expectation propagation variational message passing

来源：评论

学校读者我要写书评

暂无评论

Approximating Energy Market Clearing and Bidding With Model-Based Reinforcement Learning

引用

IEEE ACCESS 2024年 12卷 145106-145117页

作者： Wolgast, Thomas Niesse, Astrid Carl von Ossietzky Univ Oldenburg Dept Comp Sci D-26129 Oldenburg Germany

Energy market rules should incentivize market participants to behave in a market and grid conform way. However, they can also provide incentives for undesired and unexpected strategies if the market design is flawed. MARL is a promising new approach to predicting the expected profit-maximizing behavior of energy market participants in simulation. However, reinforcement learning requires many interactions with the system to converge, and the power system environment often consists of extensive computations, e.g., optimal power flow (OPF) calculation for market clearing. To tackle this complexity, we provide a model of the energy market to a basic multi-agent reinforcement learning (MARL) algorithm in the form of a learned OPF approximation and explicit market rules. The learned OPF surrogate model makes an explicit solving of the OPF completely unnecessary. Our experiments demonstrate that the model additionally reduces training time by about one order of magnitude but at the cost of a slightly worse performance. Potential applications of our method are market design, more realistic modeling of market participants, and analysis of manipulative behavior. Source code available at https://***/Digitalized-Energy-Systems/marl_clearing_and_bidding.

关键词： Training Economics Computational modeling approximation algorithms Prediction algorithms Optimization Artificial neural networks Stochastic processes Load modeling Load flow Agent-based modeling Game theory Nash equilibrium Reinforcement learning economic dispatch game theory gaming model-based nash equilibrium reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Advanced Multi-Sensor Person-Following System on a Mobile Robot: Design, Construction and Measurements

引用

IEEE INSTRUMENTATION & MEASUREMENT MAGAZINE 2024年第5期27卷 38-44页

作者： Zhang, Chi-Zhou Li, Ze-Hao Chen, Ming-Song Lin, Yong-Cheng Wang, Guan-Qiang Wang, Qiu Zeng, Wei-Dong Cent South Univ Changsha Peoples R China Cent South Univ Light Alloy Res Inst Changsha Peoples R China Cent South Univ Sch Mech & Elect Engn Changsha Peoples R China Cent South Univ machine vis deep learning & control Syst design Changsha Peoples R China

A wide range of collaborative human-robot applications across diverse domains, encompassing manufacturing, industry, and social interactions, necessitate an autonomous robot to follow its human companion. However, the great degree of flexibility and unpredictability of human mobility makes the integration of multi-sensor measurements for human-following difficult. This paper addresses these challenges by demonstrating the design and implementation of a multi-sensor person-following system. On the one hand, a target tracking control method based on detection of center boxes is proposed. On the other hand, the positioning, navigation and tracking modules are integrated into the robot system. Exhaustive functional verification is conducted on an Ackermann-steering mobile robot, in which an embedded device is the core algorithm board. Notably, we discuss the degradation phenomenon of Light Detection and Ranging (LiDAR) odometry. The positioning error of our system is 2.42%, which is much smaller than other advanced methods. In indoor experiments, our system enables the robot to stably track the target at a distance of approximately 1.5 m. The effectiveness of the entire system is verified, demonstrating its advantage in real-time performance.

关键词： Tracking loops Target tracking Laser radar Simultaneous localization and mapping approximation algorithms Real-time systems Sensors

来源：评论

学校读者我要写书评

暂无评论

A Velocity-Guided Grey Wolf Optimization Algorithm With Adaptive Weights and Laplace Operators for Feature Selection in Data Classification

引用

IEEE ACCESS 2024年 12卷 39887-39901页

作者： Zhang, Li Chen, Xiaobo Hainan Normal Univ Key Lab Data Sci & Intelligence Educ Minist Educ Hainan 571158 Peoples R China Jiangsu Univ Technol Sch Comp Engn Changzhou 213001 Jiangsu Peoples R China Peoples Bank China Changzhou City Ctr Branch Changzhou 213001 Jiangsu Peoples R China Jilin Univ Key Lab Symbol Computat & Knowledge Engn Minist Educ Changchun 130012 Peoples R China

The rapid growth of data quantity directly leads to the increasing feature dimension, which challenges machine learning and data mining. Wrapper-based intelligent swarm algorithms are effective solution techniques. The Grey Wolf Optimization (GWO) algorithm is a novel intelligent population algorithm. Simple principles and few parameters characterize it. However, the basic GWO has disadvantages, such as difficulty coordinating exploration and exploitation capabilities and premature convergence. As a result, GWO fails to identify many irrelevant and redundant features. To improve the performance of the basic GWO algorithm, this paper proposes a velocity-guided grey wolf optimization algorithm with adaptive weights and Laplace operators (VGWO-AWLO). Firstly, by introducing a uniformly distributed dynamic adaptive weighting mechanism, the control parameters $a$ are guided to undergo nonlinear dynamic changes to achieve a good transition from the exploratory phase to the development phase. Second, a velocity-based position update formula is designed with an individual memory function to enhance the local search capability of individual grey wolves and drive them to converge to the optimal solution. Thirdly, a Laplace cross-operator strategy is applied to increase the population diversity and help the grey wolf population escape from the local optimal solution. Finally, the VGWO-AWLO algorithm is evaluated for its comprehensive performance in terms of classification accuracy, dimensionality approximation, convergence, and stability in 18 classified datasets. The experimental results show that the classification accuracy and convergence speed of VGWO-AWLO are better than the basic GWO, GWO variants, and other state-of-the-art meta-heuristic algorithms.

关键词： Reflective binary codes Statistical analysis Social factors Machine learning algorithms Classification algorithms approximation algorithms Laplace equations Heuristic algorithms Nonlinear dynamical systems Metaheuristics Grey wolf optimization algorithm feature selection dynamic adaptive weighting mechanism velocity update equation mechanism Laplace operators

来源：评论

学校读者我要写书评

暂无评论

Parameterized approximation For Robust Clustering in Discrete Geometric Spaces 51

Parameterized Approximation For Robust Clustering in Discret...

引用

51st International Colloquium on Automata, Languages, and Programming, ICALP 2024

作者： Abbasi, Fateme Byrka, Jaroslaw Gadekar, Ameet Marx, Dániel Spoerhase, Joachim Banerjee, Sandip Chalermsook, Parinya Khodamoradi, Kamyar Sharma, Roohani University of Wroclaw Poland Bar-Ilan University Ramat-Gan Israel CISPA Helmholtz Center for Information Security Saarbrücken Germany University of Sheffield United Kingdom IDSIA USI-SUPSI Lugano Switzerland Aalto University Finland University of Regina Canada University of Bergen Norway

ISBN: (纸本)9783959773225

We consider the well-studied Robust (k, z)-Clustering problem, which generalizes the classic k-Median, k-Means, and k-Center problems and arises in the domains of robust optimization [Anthony, Goyal, Gupta, Nagarajan, Math. Oper. Res. 2010] and in algorithmic fairness [Abbasi, Bhaskara, Venkatasubramanian, 2021 & Ghadiri, Samadi, Vempala, 2022]. Given a constant z ≥ 1, the input to Robust (k, z)-Clustering is a set P of n points in a metric space (M, δ), a weight function w : P → R≥0 and a positive integer k. Further, each point belongs to one (or more) of the m many different groups S1, S2, . . ., Sm ⊆ P. Our goal is to find a set X of k centers such that maxi∈[m]Pp∈Si w(p)δ(p, X)z is minimized. Complementing recent work on this problem, we give a comprehensive understanding of the parameterized approximability of the problem in geometric spaces where the parameter is the number k of centers. We prove the following results: (i) For a universal constant η0 > 0.0006, we devise a 3z(1−η0)-factor FPT approximation algorithm for Robust (k, z)-Clustering in discrete high-dimensional Euclidean spaces where the set of potential centers is finite. This shows that the lower bound of 3z for general metrics [Goyal, Jaiswal, Inf. Proc. Letters, 2023] no longer holds when the metric has geometric structure. (ii) We show that Robust (k, z)-Clustering in discrete Euclidean spaces is (p3/2−o(1))-hard to approximate for FPT algorithms, even if we consider the special case k-Center in logarithmic dimensions. This rules out a (1 + ϵ)-approximation algorithm running in time f(k, ϵ)poly(m, n) (also called efficient parameterized approximation scheme or EPAS), giving a striking contrast with the recent EPAS for the continuous setting where centers can be placed anywhere in the space [Abbasi et al., FOCS’23]. (iii) However, we obtain an EPAS for Robust (k, z)-Clustering in discrete Euclidean spaces when the dimension is sublogarithmic (for the discrete problem, earlier work [Abbasi et al.

关键词： approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Hardness and approximation for the star p-Hub Routing Cost Problem in metric graphs

引用

THEORETICAL COMPUTER SCIENCE 2022年 922卷 13-24页

作者： Yeh, Hao-Ping Lu, Wei Chen, Li-Hsuan Hung, Ling-Ju Klasing, Ralf Hsieh, Sun-Yuan Natl Cheng Kung Univ Dept Comp Sci & Informat Engn 1 Univ Rd Tainan 701 Taiwan ProBrand Technol Inc Taoyuan 330 Taiwan Natl Taipei Univ Business Dept Creat Technol & Prod Design 100Sec 1Fulong Rd Taoyuan 32462 Taiwan Univ Bordeaux LaBRI CNRS 351 Cours Liberat F-33405 Talence France Natl Cheng Kung Univ Dept Comp Sci & Informat Engn Inst Med Informat Inst Mfg Informat & SystCtr Innovat FinTech Busi 1 Univ Rd Tainan 701 Taiwan Natl Cheng Kung Univ Int Ctr Sci Dev Shrimp Aquaculture 1 Univ Rd Tainan 701 Taiwan

Given a metric graph G = (V, E, w), a specific vertex c is an element of V, and an integer p, let Tbe a depth-2spanning tree of Grooted at csuch that cis adjacent to pvertices called hubs and each of the remaining vertices is adjacent to a hub. TheStar p-Hub Routing Cost Problemis to find a spanning tree Tof Gsatisfying the conditions stated above such that the sum of distances between all pairs of vertices in Tis minimized. In this paper, we prove that theStar p-Hub Routing Cost Problemis NP-hard. A 3-approximation algorithm running in time O(n(2)) is given for solving the same problem where nis the number of vertices in the input graph. Moreover, we give an example to show that the analysis of the approximation ratio cannot be better than 2 - epsilon for any epsilon > 0. (C) 2022 Elsevier B.V. All rights reserved.

关键词： Hub allocation Analysis of algorithms and problem complexity approximation algorithms Metric graphs

来源：评论

学校读者我要写书评

暂无评论

Joint Virtual Network Function Placement and Flow Routing in Edge-Cloud Continuum

引用

IEEE TRANSACTIONS ON COMPUTERS 2024年第3期73卷 872-886页

作者： Mao, Yingling Shang, Xiaojun Liu, Yu Yang, Yuanyuan SUNY Stony Brook Dept Elect & Comp Engn Stony Brook NY 11794 USA Univ Texas Arlington Dept Comp Sci & Engn Arlington TX 76019 USA

Network Function Virtualization (NFV) is becoming one of the most popular paradigms for providing cost-efficient, flexible, and easily-managed network services by migrating network functions from dedicated hardware to commercial general-purpose servers. Despite the benefits of NFV, it remains a challenge to deploy Service Function Chains (SFCs), placing virtual network functions (VNFs) and routing the corresponding flow between VNFs, in the edge-cloud continuum with the objective of jointly optimizing resource and latency. In this paper, we formulate the SFC Deployment Problem (SFCD). To address this NP-hard problem, we first introduce a constant approximation algorithm for a simplified SFCD limited at the edge, followed by a promotional algorithm for SFCD in the edge-cloud continuum, which also maintains a provable constant approximation ratio. Furthermore, we provide an online algorithm for deploying sequentially-arriving SFCs in the edge-cloud continuum and prove the online algorithm achieves a constant competitive ratio. Extensive simulations demonstrate that on average, the total costs of our offline and online algorithms are around 1.79 and 1.80 times the optimal results, respectively, and significantly smaller than the theoretical bounds. In addition, our proposed algorithms consistently outperform the popular benchmarks, showing the superiority of our algorithms.

关键词： Servers approximation algorithms Cloud computing Heuristic algorithms Routing Costs Edge computing Network function virtualization service function chain deployment edge computing cloud computing joint resource and latency optimization

来源：评论

学校读者我要写书评

暂无评论

An End-to-End Deep Reinforcement Learning Framework for Electric Vehicle Routing Problem

引用

IEEE INTERNET OF THINGS JOURNAL 2024年第20期11卷 33671-33682页

作者： Wang, Mengqin Wei, Yanling Huang, Xueliang Gao, Shan Southeast Univ Sch Automat Nanjing 210096 Peoples R China Southeast Univ Key Lab Measurement & Control Complex Syst Engn Nanjing 210096 Peoples R China Southeast Univ Sch Elect Engn Nanjing 210096 Peoples R China

Electric vehicles (EVs) have been increasingly used in the logistics and transportation industry due to their cost-effectiveness and sustainability. However, one of the major challenges in optimizing routes for EVs is the EV routing problem (EVRP), which arises from their limited battery capacity. This article proposes a reinforcement learning (RL)-based end-to-end framework to address EVRP with different sizes. The framework includes a graph attention network (GAT)-based encoder and an attention-based decoder. In particular, an improved GAT-based encoder is employed to encrypt node and edge information from the graph-structured EVRP instances, resulting in high-dimensional node embeddings and graph embedding for downstream tasks. The decoder comprises a dual-layer attention module, which generates solutions (a sequence of input nodes) based on the global state and the embeddings from the encoder. This encoder-decoder architecture constitutes the policy network, which takes instances as input and produces solutions in an auto-regressive manner. The policy network is trained using REINFORCE with a baseline. The experiments indicate that the proposed deep RL (DRL) method demonstrates more solvability efficiency than conventional methods (exact algorithms and heuristic algorithms) and shows superior performance than other DRL-based methods.

关键词： Decoding Routing Heuristic algorithms approximation algorithms Optimization Internet of Things Batteries Attention mechanism deep reinforcement learning (DRL) electric vehicle (EV) routing

来源：评论

学校读者我要写书评

暂无评论

Secure Communications via Active IRS Assisted SWIPT NOMA Networks

引用

IEEE SYSTEMS JOURNAL 2024年第2期18卷 1032-1043页

作者： Zhai, Qi Dong, Limeng Li, Yong Cheng, Wei Northwestern Polytech Univ Sch Elect & Informat Xian 710072 Peoples R China

Active intelligent reflecting surface (IRS) has attracted great attention due to the characteristic of changing the propagation channels by amplifying the amplitude and adjusting the phase shift of signals. In this article, we apply the active IRS into a simultaneous wireless information and power transfer multigroup nonorthogonal multiple access system and focus on improving the secure transmission performance. Specifically, from the overall system performance perspective, we first formulate a power minimization problem by jointly optimizing the beamforming at the base station, power allocation coefficient in each group, power splitting ratio at each user, and the IRS matrix. Then, we consider the fairness among users and aim to maximize the minimum secrecy rate (SR). In order to solve these two nonconvex problems, two efficient algorithms based on alternating optimization and block coordinate descent are proposed by applying the semidefinite relaxation, successive convex approximation, Schur complement, arithmetic-geometric mean inequality and convex upper bound substitution methods. Through simulations, we verify the convergence of the algorithms. The results show that both the power consumption and minimum SR performance can be enhanced by deploying the active IRS compared to the baseline schemes under passive IRS and random phase shift active IRS aided cases.

关键词： NOMA Array signal processing Optimization Picture archiving and communication systems Heuristic algorithms approximation algorithms Vectors Active intelligent reflecting surface (IRS) nonorthogonal multiple access (NOMA) simultaneous wireless information and power transfer (SWIPT) user fairness

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：