检索结果-内蒙古大学图书馆

A modified modeling approach of virtual power plant via improved federated learning

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS 2024年 158卷

作者： Taheri, Seyed Iman Davoodi, Mohammadreza Ali, Mohd. Hasan Univ Memphis Dept Elect & Comp Engn Memphis TN 38152 USA

The Virtual Power Plant (VPP) is a promising solution to power systems' challenges. However, its energy management system (EMS) faces challenges due to centralized Big-Data analysis, i.e., complexity and high computational cost. Decentralized strategies, e.g., federated learning (FL), have been proposed to mitigate these challenges. Although conventional FL can be implemented in VPPs, its effectiveness is limited by its typical structure, which relies on single-pattern modeling. Indeed, this particular structure leads to delays and inaccuracies in optimizing the heterogeneous generator dispatch patterns, particularly renewables. Hence, addressing this critical hindrance is essential for FL's efficient and effective deployment in VPPs. To overcome this drawback, this paper proposes novel VPP modeling utilizing multi-task FL modernized by multi-pattern modeling to decentralize the EMS processing. The proposed FL defines electricity generators' agents as clients collaborating to train the decentralized local data coordinated by a central server. The innovative idea is to create a novel multi-pattern-multi-task FL for electricity generation prediction so that the generator's agents learn to provide an optimized dispatch. Thus, the conventional multi-task FL is improved by proposing an optimized neural-network clustering technique. Simulation results demonstrate that multi-pattern-multi-task FL is 39%41% faster than the best-reported methods, i.e., teaching-learning-based and honey-bee-mating optimization algorithms. (c) 2017 Elsevier Inc. All rights reserved.

关键词： Virtual power plant Federated learning Optimization algorithm Energy management system distributed generation

来源：评论

学校读者我要写书评

暂无评论

Medical Remote Sensing and Internet of Things for Detection of Epileptic Seizure Using Deep neural networks 2

Medical Remote Sensing and Internet of Things for Detection ...

引用

2nd International Conference on distributed Computing and Optimization Techniques, ICDCOT 2024

作者： Vinuja, G. Devi, N. Bharatha Saveetha School of Engineering Simats Department of Cse Thandalam India

ISBN: (纸本)9798350382952

Medical Remote sensing and the Internet of Things (IoT) have emerged as powerful tools in the field of disease detection and monitoring. Early detection of infectious diseases is crucial in order to prevent outbreaks and minimize their impact on public health. This paper aims to explore suitable methods and algorithms that can effectively utilize medical remote sensing data and IoT devices to detect and identify infectious diseases at an early stage. In this research, the novel real-time personal health monitoring system WISE (Wearable IoT-cloud-based health monitoring system) is introduced. In the present research, the body area network (BAN) is used to collect the patient's EEG signal. These signals are sent from the patient's location to the cloud via wifi. To turn the EEG input signal into an image, the discrete wavelet transform (DWT) is initially employed. This EEG input image first undergoes data de-noising pre-processing. Wavelet decomposition of EEG data accurately captures and pinpoints transient patterns in both temporal and frequency domains. Classifiers based on multilayer perceptron neural networks (MLPNN) were created and their classification accuracy for EEG signals was tested.. As compared to the existing algorithm the accuracy rate in predicting an abnormal EEG was 98.1 %. The classifier is employed to assess whether a region is typical or anomalous, and it also gauges its effectiveness. © 2024 IEEE.

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Hybrid intrusion detection system using blockchain framework

引用

EURASIP JOURNAL ON WIRELESS COMMUNICATIONS AND networkING 2022年第1期2022卷 1-25页

作者： Khonde, S. R. Ulagamuthalvi, V. Sathyabama Inst Sci & Technol Dept Comp Sci & Engn Chennai Tamil Nadu India SP Pune Univ MES Coll Engn Dept Comp Engn Pune Maharashtra India

Data security and confidentiality are major goals now days due to the extensive use of the internet for data sharing. In modern era, most of the networks are compromised by intruders to grab access to private, confidential, and highly secured data. An intrusion detection system (IDS) is widely used to secure the network from getting compromised by intruders. Most of the IDS share the signatures of the novel attacks detected by anomaly approach for improving the detection rate and processing time. Security of signature shared by nodes is becoming a considerable problem. This paper presents a novel framework blockchain based hybrid intrusion detection system (BC-HyIDS), which uses the blockchain framework for exchanging signatures from one node to the other in distributed IDS. BC-HyIDS works in three phases where it uses both detection methods and blockchain in the third phase to provide security to data transferred through the network. This system makes use of a cryptosystem to encrypt the data stored in blocks to improve security one level higher. Hyperledger fabric v2.0 and Hyperledger sawtooth is used to implement system. Blockchain framework is created as a prototype using distributed ledger technology which helps in securing signature exchange. Performance of BC-HyIDS is evaluated in terms of accuracy, detection rate, and false alarm rate. From results, it is observed that a 2.8% increase in accuracy, 4.3% increase in detection rate, and a reduction of 2.6% in FAR is achieved. Blockchain performance is evaluated using Hyperledger fabric v2.0 and Hyperledger sawtooth on throughput, processing time, and average latency. BC-HyIDS shows improved performance when used with blockchain.

关键词： Blockchain Intrusion detection system Secured communication XGBoost Isolation random forest Artificial neural network Ensemble approach

来源：评论

学校读者我要写书评

暂无评论

Securing Health Records Using Quantum Convolutional neural network 3rd

Securing Health Records Using Quantum Convolutional Neural N...

引用

3rd International Conference on Sustainable Expert Systems, ICSES 2022

作者： Arulmozhi, B. Sheeba, J.I. Pradeep Devaneyan, S. Department of Computer Science and Engineering Pudhucherry Technologial University Puducherry605014 India Department of Mechanical Engineering Sri Venkateshwaraa College of Engineering and Technology Puducherry605012 India

ISBN: (纸本)9789811978739

In block chain technology, a distributed architecture, blocks are the data storage and processing unit. Data must be exchanged from one party to another safely and securely, and the block chain must be updated using smart contracts. But still, the problem facing was securing the healthcare data. So, this paper proposed securing health records using the QCNN method, which stands for a Quantum Convolutional neural network with Block Chain Technology, to safeguard medical data. Block chain technology's consensus process ensures that all data is correct and all transactions are safe. The datasets are collected from the data world repository. The datasets are pre-processed using label coding with the standard scalar method. The features are analyzed with QCNN and the patient healthcare data to increase the security and interoperability using block chain technology. The research specifically looks at Electronic Health Record (EHR) security issues to be overcome. This proposed quantum conventional neural network (QCNN) deep learning architecture provides higher accuracy and security than the other existing algorithms. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： network architecture

来源：评论

学校读者我要写书评

暂无评论

distributed Detection of Anomalies in the network Flow Using Generative Adversarial networks

Distributed Detection of Anomalies in the Network Flow Using...

引用

2022 International Russian Automation Conference, RusAutoCon 2022

作者： Grekov, Mikhail Sychugov, Alexey Tula State University Department of Information Security Tula Russia

ISBN: (数字)9781665466554

ISBN: (纸本)9781665466554

Ensuring information security in distributed information systems is a time-consuming task associated with processing large amounts of network data. Intrusion detection systems based on anomalies and machine learning algorithms are a promising tool for protecting distributed computing networks. The complexity of using such a means of protection at real industrial facilities lies in the high level of false positives. Another problem is the need to process a large number of network packets, which requires high computational costs. In this paper, a model of distributed processing and anomaly detection in network flows is proposed. To minimize the processed traffic, an attacker's behavior model is used. Improving the efficiency of anomaly detection is achieved through the use of generative-adversarial neural networks. Creating synthetic traffic based on the characteristics of a real network will allow us to balance data sets for training the anomaly detection core. Thus, the efficiency of detecting anomalies in network traffic and the resistance to adversarial attacks will increase. The distributed model allows to reduce the load on the communication lines and minimize the transmission of confidential data. The multilevel structure of this model will allow detecting multi-stage attacks by processing notification streams received from nodes of a distributed system. © 2022 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Telemetry Data Compression Algorithm Based on Balanced Cyclic neural network and Deep Learning

Telemetry Data Compression Algorithm Based on Balanced Cycli...

引用

2023 IEEE International Conference on Image processing and Computer Applications, ICIPCA 2023

作者： Huang, Min School of Computer Sichuan Technology and Business University Chengdu620000 China

ISBN: (纸本)9798350314670

With the rapid development of telemetry and telecontrol technology, the amount of information for acquiring telemetry data is increasing, which requires a large amount of storage space when storing, a large bandwidth when transmitting, and data processing as fast as possible. Therefore, data compression technology is needed to solve this problem. The redundancy of telemetry data is also very large. Statistics show that the redundancy of data transmitted in the standard telemetry system reaches more than 90%. In order to reduce the signal space and relieve the pressure of data processing, it is necessary and important to use data compression technology in telemetry system. Considering the space performance and time performance of the lossless compression algorithm, several commonly used lossless compression algorithms are compared for telemetry noise data. The compression space performance of each algorithm still needs to be improved. Considering that the measured engineering data is distributed in a typical balanced way, we studied the anomaly detection problem of time series data with inter class balance, and proposed an anomaly detection problem based on recurrent neural network (RNN). Finally, the overall test of the developed data acquisition and compression system is carried out. The test results show that the system has completed the real-time acquisition and compression of six channels of rapidly varying parameters, and the compression ratio is more than 1.8:1. All indicators meet the design requirements. © 2023 IEEE.

关键词： Redundancy

来源：评论

学校读者我要写书评

暂无评论

distributed LINK SPARSIFICATION FOR SCALABLE SCHEDULING USING GRAPH neural networkS 47

DISTRIBUTED LINK SPARSIFICATION FOR SCALABLE SCHEDULING USIN...

引用

47th IEEE International Conference on Acoustics, Speech and Signal processing (ICASSP)

作者： Zhao, Zhongyuan Swami, Ananthram Segarra, Santiago Rice Univ Houston TX 77251 USA US Armys DEVCOM Army Res Lab Adelphi MD USA

ISBN: (纸本)9781665405409

distributed scheduling algorithms for throughput or utility maximization in dense wireless multi-hop networks can have overwhelmingly high overhead, causing increased congestion, energy consumption, radio footprint, and security vulnerability. For wireless networks with dense connectivity, we propose a distributed scheme for link sparsification with graph convolutional networks (GCNs), which can reduce the scheduling overhead while keeping most of the network capacity. In a nutshell, a trainable GCN module generates node embeddings as topology-aware and reusable parameters for a local decision mechanism, based on which a link can withdraw itself from the scheduling contention if it is not likely to win. In mediumsized wireless networks, our proposed sparse scheduler beats classical threshold-based sparsification policies by retaining almost 70% of the total capacity achieved by a distributed greedy max-weight scheduler with 0.4% of the point-to-point message complexity and 2.6% of the average number of interfering neighbors per link.

关键词： Independent set graph neural networks scheduling overhead massive access distributed scheduling

来源：评论

学校读者我要写书评

暂无评论

A Novel Maize Futures Price Prediction Model based on EEMD-CNN-IGRU 22

A Novel Maize Futures Price Prediction Model based on EEMD-C...

引用

22nd IEEE International Symposium on Parallel and distributed processing with Applications, ISPA 2024

作者： Wang, Jingyang Han, Jingyu Liu, Wei Li, Ying Sun, Qiuhong Zhou, Xiaokang Hebei University of Science and Technology School of Information Science and Engineering Shijiazhuang China Kansai University Faculty of Business Data Science Osaka565-0823 Japan

ISBN: (纸本)9798331509712

Maize, a significant global food crop, is essential in agriculture and the economy. The price of maize futures is affected by many factors, and its data is a nonlinear, unstable, and long-term correlation, so it is difficult to predict it accurately. Accordingly, this paper presents a combined model based on Ensemble Empirical Mode Decomposition (EEMD), Convolutional neural network (CNN), and Improved Gated Recurrent Unit (IGRU). EEMD decomposes the maize price data to produce multiple Intrinsic Mode Function (IMF) components and residual sequences. Subsequently, the noisy IMF components are removed, and the remaining IMF components are reconstructed into low, medium, and high frequencies. The CNN is tasked with the extraction of eigenvalues for these components. The IGRU improves in two key respects on the original Gated Recurrent Unit (GRU) model. First, it enhances the update and reset gates. Second, it incorporates the Self-Attention (SA) module. Thereby, the model's predictive capabilities are improved. This study uses the primary maize futures trading data from the Dalian Commodity Exchange as the experimental data. A comparative analysis of the EEMD-CNN-IGRU model with seven baseline models shows that it outperforms other models in all evaluation indexes. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

ADSwitch: Adaptive CPU Scheduling for distributed Systems 21

ADSwitch: Adaptive CPU Scheduling for Distributed Systems

引用

21st IEEE International Symposium on Parallel and distributed processing with Applications, 13th IEEE International Conference on Big Data and Cloud Computing, 16th IEEE International Conference on Social Computing and networking and 13th International Conference on Sustainable Computing and Communications, ISPA/BDCloud/SocialCom/SustainCom 2023

作者： Zhong, Yidong Ao, Yang Li, Dingding Li, Jianguo Zeng, Deze South China Normal University School of Computer Science Guangdong Guangzhou510631 China China University of Geosciences School of Computer Science Hubei Wuhan430074 China

ISBN: (纸本)9798350329223

In traditional CPU scheduling systems, it is challenging to customize scheduling policies for datacenter workloads. Therefore, distributed cluster managers can only perform coarse-grained job scheduling rather than fine-grained CPU scheduling, although the latter is also critical to application performance in the clusters. Some scheduling systems have been proposed to facilitate the implementation and deployment of custom policies. However, these systems neither adequately capture the resource requirements of workloads nor cooperatively manage the CPU scheduling of distributed nodes, thus may failing to fully utilize CPU resources in the clusters. To tackle these issues, we present ADSwitch, a distributed CPU scheduling system that manages CPU resources adaptively in datacenter clusters. We train an efficient workload classification model based on a recurrent neural network (RNN), which enables ADSwitch to dynamically identify the demand of the running workload and switch to an appropriate scheduling policy through a reliable online update mechanism. We implement ADSwitch and evaluate it in a distributed cluster with synthetic workloads. The experimental results demonstrate that ADSwitch achieves up to 33% lower tail latency and 28% higher throughput than the existing scheduling systems without workload characteristic awareness. © 2023 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Code Representation Learning for Code Search with Abstract Code Semantics

Enhancing Code Representation Learning for Code Search with ...

引用

International Joint Conference on neural networks (IJCNN)

作者： Zhang, Shaojie Ding, Yiwei Hu, Enrui Yu, Yue Zhang, Yu Southern Univ Sci & Technol Dept Comp Sci & Engn Shenzhen Peoples R China Peng Cheng Lab Shenzhen Peoples R China Huawei Distributed & Parallel Software Lab Shenzhen Peoples R China

ISBN: (纸本)9798350359329;9798350359312

Code representation learning is an important way to encode the semantics of source code through pre-training. The learned representation supports a variety of downstream tasks, such as natural language code search and code defect detection. Inspired by pre-trained models for natural language representation learning, existing approaches often treat the source code or its structural information (e.g., Abstract Syntax Tree or AST) as a plain token sequence. Unlike natural language, programming language has its unique code unit information (e.g., identifiers and expressions) and logic information (e.g., the functionality of a code snippet). To further explore those properties, we propose Abstract Code Embedding (AbCE), a selfsupervised learning method that considers the abstract semantics of code logic. Instead of scattered tokens, AbCE treats an entire node or a subtree in an AST as a basic code unit during pre-training, which preserves the entirety of a coding unit. Moreover, AbCE learns the abstract semantics of AST nodes via a self-distillation way. Experimental results show that it achieves significant improvements over state-of-the-art baselines on code search tasks and comparable performance on code clone detection and defect detection tasks even without using contrastive learning or curriculum learning.

关键词： Natural Language processing Code Comprehension Self-distillation Code Search

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：