检索结果-内蒙古大学图书馆

Resource Optimization for Device-to-Device and Small Cell Uplink Communications Underlaying Cellular Networks

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2018年第2期67卷 1187-1201页

作者： Dai, Haibo Huang, Yongming Zhao, Rui Wang, Jiaheng Yang, Luxi Southeast Univ Sch Informat Sci & Engn Nanjing 210096 Jiangsu Peoples R China Nanjing Univ Posts & Telecommun Nanjing 210003 Jiangsu Peoples R China Huaqiao Univ Xiamen Key Lab Mobile Multimedia Commun Xiamen 361021 Peoples R China

This paper investigates the joint power control and subchannel allocation problem for device-to-device and small cell uplink communications in a cellular network, with the aim of improving the network throughput. For this throughput maximization problem, we leverage a game-theoretic learning approach to solve it. However, there is an intractable issue for the feasible region of the optimization problem, which is intermixed with the integer nature. To tackle this problem, we first present the closed-form expressions of the optimal power under different scenarios by presetting the subchannel allocation profile. Based on the optimal power control, we then formulate the subchannel allocation problem as a game theoretical framework, and define a welfare function which has the same optimum as the optimization objective. To optimize the welfare function, a distributed trial and error learning algorithm is proposed to converge to a stochastically stable state. Since the achieved stable state cannot guarantee to be the optimal solution, we reformulate this problem as an exact potential game model and propose another distributed learning algorithm to find a better Nash equilibrium which is the global optimal solution (or a superior solution). Moreover, to accelerate the convergence, these two algorithms are modified by getting rid of these unavailable strategic profiles. Finally, numerical results verify the effectiveness of the proposed schemes.

关键词： Device-to-device communication game theory heterogeneous network learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Using a Hybrid of Morphological Operation and Neuro-Fuzzy Filter for Transmission Map Refinement and Halation Elimination in Weather Degraded Images

引用

JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 2018年第3期34卷 701-720页

作者： Lin, Hsueh-Yi Lin, Cheng-Jian Huang, Mei-Ling Natl Chin Yi Univ Technol Dept Comp Sci & Informat Engn Taichung 411 Taiwan Natl Chin Yi Univ Technol Dept Ind Engn & Management Taichung 411 Taiwan

For photographs taken in outdoor environments, the air medium causes light attenuation, which reduces image quality;this effect is particularly obvious in hazy environments. To eliminate the hazy effect in images and improve the image quality, the present study proposes an efficient hybrid method. The proposed fuzzy estimator was adopted to estimate variations in light attenuation, and morphological erosion and a neuro-fuzzy filter proposed by this study were employed to refine the transmission map and eliminate the halation. Finally, an estimated mean value for atmospheric light was applied to calculate the color vector of atmospheric light to eliminate the color cast. Experimental results indicate that the proposed hybrid method is superior to other dehazing methods.

关键词： dehazing fuzzy estimator neuro-fuzzy filter learning algorithm transmission map halation

来源：评论

学校读者我要写书评

暂无评论

A Multiscale Concept Drift Detection Method for learning from Data Streams 14

A Multiscale Concept Drift Detection Method for Learning fro...

引用

14th IEEE International Conference on Automation Science and Engineering (IEEE CASE)

作者： Wang, XueSong Kang, Qi Zhou, MengChu Yao, SiYa Tongji Univ Sch Elect & Informat Engn Dept Control Sci & Engn Shanghai 201804 Peoples R China New Jersey Inst Technol Dept Elect & Comp Engn Newark NJ 07102 USA

ISBN: (纸本)9781538635933

Concept drifts can occur in data streams, which disable any models built on static data distribution to fit dynamic or recurrent concepts. If there is a feature that can monitor the stableness of such distribution, then we have a proper reference to adapt the model. Based on this idea, we propose a novel approach named Multiscale Drift Detection Test (MDDT) that localizes abrupt drift points when detection feature values fluctuate. MDDT is based on a resampling scheme and a paired student t-test. It applies a detection procedure on a broad and a narrow scale. This multiscale structure not only reduces massive time of a constant checking process, but also filters noise in the detection features. Experiments are performed via synthetic and real-world datasets. The results indicate that the proposed method outperforms the state-of-art algorithms in terms of computation cost and average accuracy.

关键词： Concept drift detection data stream learning algorithm optimization undersampling

来源：评论

学校读者我要写书评

暂无评论

Toward a soft computing-based correlation between oxygen toxicity seizures and hyperoxic hyperpnea

引用

SOFT COMPUTING 2018年第7期22卷 2421-2427页

作者： D'Angelo, Gianni Pilla, Raffaele Dean, Jay B. Rampone, Salvatore Univ Sannio Dept Sci & Technol Benevento Italy Univ S Florida Dept Mol Pharmacol & Physiol Morsani Coll Med Hyperbar Biomed Res Lab Tampa FL 33612 USA

Exposure to high levels of hyperbaric oxygen () can induce central nervous system oxygen toxicity in humans and animals, a phenomenon characterized by repeated tonic-clonic seizures. The risk of developing this type of convulsions represents the limiting factor in using for a number of clinical and diving applications. Previously, using radio-telemetry in awake rats, a significant increase in the mean value of the ventilatory responses in rats to approximately 5-8 min before onset of seizures has been observed. This response has been termed "hyperoxic hyperpnea," and it has been hypothesized it may serve as a predictor of an impending seizure while breathing . The purpose of the present study was to apply soft computing methods to determine whether there is a direct correlation between the onset of hyperoxic hyperpnea (i.e., the ventilatory response as defined by tidal volume and respiratory frequency) and the onset of seizure. In our experiments, we used Multilayer Perceptron, Naive Bayes, J48, and U-BRAIN aimed at evidencing the correlation between the respiratory feature vectors and the onset of an impending seizure in unanesthetized, freely behaving rats breathing 4, 5 or 6 atmospheres absolute (ATA) of oxygen. This strategy was also aimed to finding a set of rules relating physiological parameters and convulsive phenomena through inductive inference.

关键词： Oxygen toxicity Seizure Hyperoxia Diving physiology Rat learning algorithm Soft computing U-BRAIN

来源：评论

学校读者我要写书评

暂无评论

Domain Wall Motion-Based Dual-Threshold Activation Unit for Low-Power Classification of Non-Linearly Separable Functions

引用

IEEE TRANSACTIONS ON BIOMEDICAL CIRCUITS AND SYSTEMS 2018年第6期12卷 1410-1421页

作者： Deb, Suman Vatwani, Tarun Chattopadhyay, Anupam Basu, Arindam Fong, Xuanyao Nanyang Technol Univ Sch Comp Sci & Engn Singapore 639798 Singapore Nanyang Technol Univ Sch Elect & Elect Engn Singapore 639798 Singapore Natl Univ Singapore Dept Elect & Comp Engn Singapore 117583 Singapore

Recently, a great deal of scientific endeavour has been devoted to developing spin-based neuromorphic platforms owing to the ultra-low-power benefits offered by spin devices and the inherent correspondence between spintronic phenomena and the desired neuronal, synaptic behavior. While domain wall motion-based threshold activation unit has previously been demonstrated for neuromorphic circuits, it remains well known that neurons with threshold activation cannot completely learn nonlinearly separable functions. This paper addresses this fundamental limitation by proposing a novel domain wall motion-based dual-threshold activation unit with additional nonlinearity in its function. Furthermore, a new learning algorithm is formulated for a neuron with this activation function. We perform 100 trials of tenfold training and testing of our neural networks on real-world datasets taken from the UCI machine learning repository. On an average, the proposed algorithm achieves 1.04x -6.54x lower misclassification rate (MCR) than the traditional perceptron learning algorithm. In a circuit-level simulation, the neural networks with the proposed activation unit are observed to outperform the perceptron networks by as much as 2.98xMCR. The energy consumption of a neuron having the proposed domain wall motion-based activation unit averages to 35 fJ approximately.

关键词： ANN domain wall motion dual-threshold activation unit learning algorithm memristive crossbar array neuromorphic computing non-linearly separable function threshold function

来源：评论

学校读者我要写书评

暂无评论

Exergetic performance prediction of solar air heater using MLP, GRNN and RBF models of artificial neural network technique

引用

JOURNAL OF ENVIRONMENTAL MANAGEMENT 2018年 223卷 566-575页

作者： Ghritlahre, Harish Kumar Prasad, Radha Krishna Natl Inst Technol Dept Mech Engn Jamshedpur 831014 Jharkhand India

In the present study three different types of neural models: multi-layer perceptron (MLP), generalized regression neural network (GRNN) and radial basis function (RBF) has been used to predict the exergetic efficiency of roughened solar air heater. The experiments were conducted at NIT Jamshedpur, India, using two different types of absorber plate: arc shape wire rib roughened with relative roughness height 0.0395, relative roughness pitch 10 and angle of attack 60 degrees, and smooth absorber plates for 7 days. Total 210 data sets were collected from the experiments. Mass flow rate, relative humidity, wind speed, ambient air temperature, inlet air temperature, mean air temperature, average plate temperature and solar intensity were selected as input parameters in input layer to estimate the exergetic efficiency. In the first part of study, MLP model has been used. In this model 10-20 neurons with LM learning algorithm were used in hidden layer for optimal model selection. It has been found that LM-18 is an optimal model. In second part, GRNN model was used. The GRNN model was simulated experimentally at different spread constants and found that keeping spread constant as 1.5, optimal results have been obtained. In the third part, RBF model was used. For optimal model, 1-5 spread constant at interval of 0.5 have been used. It has been found that by taking spread constant 3.5, best results are obtained. In the last part of the study, all neural models are compared on the basis of statistical error analysis. It has been found that RBF model is better than GRNN and MLP models due to lowest value of RMSE and MAE and highest value of R-2 and ME. After RBF model, GRNN model performs better results as compared to MLP model. It has been found that the values of RMSE, MAE and R-2 were 0.001652, 2.86E-04 and 0.99999 respectively for RBF model.

关键词： Solar air heater Exergetic efficiency Artificial neural network learning algorithm Multi-layer perceptron Generalized regression neural network Radial basis function

来源：评论

学校读者我要写书评

暂无评论

A surrogate model for thermal characteristics of stratospheric airship

引用

ADVANCES IN SPACE RESEARCH 2018年第12期61卷 2989-3001页

作者： Zhao, Da Liu, Dongxu Zhu, Ming Beihang Univ Sch Aeronaut Sci & Engn Beijing 100191 Peoples R China Beihang Univ Minist Ind & Informat Technol Key Lab Aircraft Adv Design Technol Beijing 100191 Peoples R China

A simple and accurate surrogate model is extremely needed to reduce the analysis complexity of thermal characteristics for a stratospheric airship. In this paper, a surrogate model based on the Least Squares Support Vector Regression (LSSVR) is proposed. The Gravitational Search algorithm (GSA) is used to optimize hyper parameters. A novel framework consisting of a preprocessing classifier and two regression models is designed to train the surrogate model. Various temperature datasets of the airship envelope and the internal gas are obtained by a three-dimensional transient model for thermal characteristics. Using these thermal datasets, two-factor and multi-factor surrogate models are trained and several comparison simulations are conducted. Results illustrate that the surrogate models based on LSSVR-GSA have good fitting and generalization abilities. The pre-treated classification strategy proposed in this paper plays a significant role in improving the accuracy of the surrogate model. (C) 2018 COSPAR. Published by Elsevier Ltd. All rights reserved.

关键词： Stratospheric airship Surrogate model Thermal characteristics Least squares support vector regression Gravitational search algorithm learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Directed-Hypergraph-Based Channel Allocation for Ultradense Cloud D2D Communications With Asymmetric Interference

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2018年第8期67卷 7712-7718页

作者： Sun, Youming Du, Zhiyong Xu, Yuhua Zhang, Yuli Jia, Luliang Anpalagan, Alagan Army Engn Univ Nanjing 210000 Jiangsu Peoples R China Natl Univ Def Technol Changsha 410000 Hunan Peoples R China Ryerson Univ Dept Elect & Comp Engn Toronto ON M5B 2K3 Canada

In this paper, we investigate channel allocation for ultradense device-to-device (D2D) communications. Different from both binary graph and undirected hypergraph interference model, an improved directed hypergraph is applied to simultaneously represent cumulative and asymmetric interference aspects in the context of ultradense communications. We formulate the channel access problem in cloud D2D communication networks as a directed-hypergraph-based local altruistic game, which is proved to be an exact potential game. Then, a multiagent concurrent learning scheme in centralized-distributed fashion is proposed to search the optimal pure Nash equilibrium, which can also maximize the normalized network capacity. Finally, simulation results are presented to validate the proposed learning scheme.

关键词： Directed hypergraph Channel allocation Device-to-device communications Exact potential game learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Prediction of exergetic efficiency of arc shaped wire roughened solar air heater using ANN model

引用

INTERNATIONAL JOURNAL OF HEAT AND TECHNOLOGY 2018年第3期36卷 1107-1115页

作者： Ghritlahre, Harish K. Prasad, Radha K. Natl Inst Technol Dept Mech Engn Jamshedpur 831014 Jharkhand India

In present work, Artificial Neural Network (ANN) model has been structured to predict the exergetic efficiency of roughened solar air heater. Arc shaped wire rib roughened absorber plate with relative roughness pitch (P/e) =10, relative roughness height (e/D) =0.0395 and angle of attack (alpha) = 60 degrees is used in the experiments. Experiments were conducted at Jamshedpur (India) under the local weather conditions for 7 days between 09:00 and 16:00 h in the month of February. Total 105 data samples have been collected from the experiments. By the use of experimental data and calculated values of exergy efficiency were used to develop ANN model. The MLP model has been developed with eight parameters in input layer and one parameter in output layer. Levenberg-Marquardt (LM) learning algorithm has been used to training an ANN model. For optimal topology, 10-16 neurons were used to train the network and found that 14 numbers of neurons with single hidden layer is optimal on the basis of statistical error analysis. The values of R-2 for predicted exergetic efficiency is 0.99583 which shows that predicted values are very close to experimental data. The statistical analysis also show that the value of RMSE and COV were 0.021747 and 1.63766 respectively, which are very low as required for accurate prediction. The statistical results show that the proposed MLP model successfully predicts the exergetic performance of roughened solar air heater.

关键词： solar air heater artificial neural network exergy efficiency learning algorithm multi-layer perceptron

来源：评论

学校读者我要写书评

暂无评论

Automation of Roof Construction Management by Means Artificial Neural Network 19th

Automation of Roof Construction Management by Means Artifici...

引用

19th International Scientific Conference on Energy Management of Municipal Transportation Facilities and Transport (EMMFT)

作者： Lapidus, Azary Makarov, Alexander Moscow State Univ Civil Engn Yaroslayskoye Sh 26 Moscow 129337 Russia

ISBN: (纸本)9783319709871;9783319709864

To enable efficient managerial decisions in planning and generation of construction process, we need technologies to analyze and evaluate their conditions and to predict their further evolution. One of advanced methods to generate them is based on artificial neural networks (ANN). This paper represents the outcome from a research in construction process organization and management to build roof structures at the planning and operation stage, assisted by the ANN methodology. Relying on systems analysis and expert survey, the authors have designed a 4-layer feedforward ANN with a node pattern of 15-6-3-1. The authors used the concept of fuzzy sets to measure input and output of the network. The paper clearly illustrates the function of ANN and describes the learning algorithm by the backward propagation of errors relative to the resulting ANN. The authors have identified problems related to generation of the labeled training data in the researched field and developed a network learning approach that combines extraction of expert knowledge about the system with learning on training examples from real-life construction project.

关键词： Artificial neural network Organization of construction Decision making support Fuzzy sets Roofing Training data learning algorithm

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：