检索结果-内蒙古大学图书馆

2023 International Conference on Machine Learning and Cybernetics, ICMLC 2023

作者： Xu, Wang Zhuoming, Liang School of Computer Neusoft Institute Guangdong Department of Network Engineering China South China Nomal University Network Center China

ISBN: (纸本)9798350303780

As a new wireless transmission architecture technology, cSO can effectively reduce the influence of fading channels and provide cSO links for low channel quality. Secondly, in the ultra-large traffic cSO, The fitting of data and the constraint on the model must be considered at the same time. Generally speaking, regularization is the most common method, because in the process of inversion, it is far from enough to only use data fitting terms as constraints. In order to make the inversion results stable, model constraint terms should be added to the objective function on the basis of data fitting terms, and there are usually different model constraint methods for different practical problems. When the physical properties of anomalous body are different from those of surrounding rock, the minimum model constraint can be used. When inversion of the moderocess of EMM-connected state was consistent with the determination of the coverage enhancement level. © 2023 IEEE.

关键词： Fading channels

来源：评论

学校读者我要写书评

暂无评论

Twin delayed deep deterministic policy gradient-based intelligent computation offloading for IoT

引用

Digital Communications and networks 2023年第4期9卷 836-845页

作者： Siguang Chen Bei Tang Kun Wang School of Internet of Things Nanjing University of Posts and TelecommunicationsNanjing210003China Jiangsu Engineering Research Center of Communications and Network Technology Nanjing University of Posts and TelecommunicationsNanjing210003China Department of Electrical and Computer Engineering University of CaliforniaLos AngelesCA90095USA

In view of the randomness distribution of multiple users in the dynamic large-scale Internet of Things(IoT)scenario,comprehensively formulating available resources for fog nodes in the area and achieving computation services at low cost have become great *** a result,this paper studies an efficient and intelligent computation offloading mechanism with resource ***,an optimization problem is formulated to minimize the total energy consumption of all tasks under the joint optimization of computation offloading decisions,bandwidth resources and transmission ***,a Twin Delayed Deep Deterministic Policy Gradient-based Intelligent Computation Offloading(TD3PG-ICO)algorithm is proposed to solve this optimization *** combining the concept of the actor critic algorithm,the proposed algorithm designs two independent critic networks that can avoid the subjective prediction of a single critic network and better guide the policy network to generate the global optimal computation offloading ***,this algorithm introduces a continuous variable discretization operation to select the target offloading node with random *** available resources of the target node are dynamically allocated to improve the model decision-making ***,the simulation results show that this proposed algorithm has faster convergence speed and good *** can always approach the greedy algorithm with respect to the lowest total energy ***,compared with full local and Deep Q-learning network(DQN)-based computation offloading schemes,the total energy consumption can be reduced by an average of 15.53%and 6.41%,respectively.

关键词： Fog computing Computation offloading Deep reinforcement learning Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Developing Real-World Applications with Algorithmic Intelligence in Reinforcement Learning for Robotics 2

Developing Real-World Applications with Algorithmic Intellig...

引用

2nd IEEE International Conference on Trends in Quantum Computing and Emerging Business Technologies, TQCEBT 2024

作者： Bhanumathi, M. Sangeetha, M. Harin Fernandez, F. Mary Ganesh Ramachandran, A. Saravanan, S.K. Chennai India Department of Computer Engineering University of Technology and Applied Sciences Nizwa Oman Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai India Department of Computer Science Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai India Department of Computers Science and Engineering BigData and Network Security Saveetha School of Engineering Simats Tamil Nadu Chennai India

ISBN: (纸本)9798350384277

In the study, three leading reinforcement learning algorithms Deep Q-network (DQN), Proximal Policy Optimization (PPO) and Asynchronous Advantage Actor - Critic (A3Cc are applied in a number of real-world robotics tasks. The focus of our analysis is on performance indicators, computational efficiencies across multiple environments as well with respect to customer satisfaction rating. The DQN demonstrated good performance, with the success rate of up to 85% in object manipulation tasks and confirmed its reliability under controlled lab conditions where it made only a 5 percent error. PPO excelled in autonomous navigation (82%) and was the most energy-efficient option, requiring just 10 hours for training and using only twelve KWH of power. Compared with A3C, the method performed to a higher level of adaptability and robustness, especially in object manipulation (90% success rate) and dynamic environments (10% error rates applicable for outdoor variable weather conditions). But A3C was also the most automatable, which required 15 hours to train and consumed an astonishing amount of electricity - 18 kWh. Regarding customer satisfaction, PPO performs the best in end-user rating on user interface;A3C shows a preference with engineers due to its advanced features and flexibility. The focus of this study is significant trade-offs between performance, efficiency and user preference in real robotics applications that advocate for individual tailoring selections as well as deployments based on task specifics nature. © 2024 IEEE.

关键词： Customer satisfaction

来源：评论

学校读者我要写书评

暂无评论

VCApather: A network as a Service Solution for Video Conference Applications 34

VCApather: A Network as a Service Solution for Video Confere...

引用

34th Workshop on network and Operating System Support for Digital Audio and Video, NOSSDAV 2024

作者： He, Dongbiao Lin, Canshu Yu, Xian Westphal, Cedric Ming, Zhongxing Cui, Laizhong Zhou, Xu Garcia-Luna-Aceves, J.J. Sangfor Technologies Inc China College of Computer Science and Software Engineering Shenzhen University China Computer Network Information Center Chinese Academy of Sciences China China Electrical and Computer Engineering Department University of Toronto Canada Futurewei

ISBN: (纸本)9798400706134

We propose a network service as a solution for video conference applications by constructing network layer routing strategies. Our approach takes into account the characteristics of conferencing flows, addresses various self-customized metrics, and proactively ensures a positive user experience by preventing contention. The performance of VCApather is evaluated using a fully-meshed topology with five nodes and real-world video conference traffic. The results show that VCApather is capable of achieving high link utilization and balance, while also meeting predefined user metrics. Compared to other schemes, VCApather was found to satisfy 69.8% more QoE requirement and to yield an average bitrate improvement of 1.74×. © 2024 ACM.

关键词： network layers

来源：评论

学校读者我要写书评

暂无评论

Advancements in Object Detection for Accurate and Efficient Visual Recognition Using Machine Learning 2

Advancements in Object Detection for Accurate and Efficient ...

引用

2nd IEEE International Conference on Advances in Information Technology, ICAIT 2024

作者： Fernandez, F. Mary Harin Ramachandran, A. Ganesh Saravanan, S.K. Bhanumathi, M. Sangeetha, M. Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai India Department of Computer Science Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai India Department of Computers Science and Engineering BigData and Network Security Saveetha School of Engineering SIMATS Tamil Nadu Chennai India Chennai India Computer Engineering Department University of Technology and Applied Sciences Nizwa Oman

ISBN: (纸本)9798350383867

Object detection in images and videos is a vital task in computer vision where its application goes beyond the automobile driving and surveillance, to healthcare. The paper introduces an object detection model which is specific and powerful, that is able to recognize visually with machine learning. The framework is constructed on the basis of the recent innovations in deep learning and attentive mechanisms, which not only lead to the best performance in object detection today but also guarantee the widest applicability. Using an extensive experiment which reveals the described model superiority over the current state-of-the-art algorithms by evaluating mAP indices, whereas the latter has a normal complexity. The ablation study uncovers the impact of core parameters including attention mechanisms and semi-supervised techniques, underlining their importance in the growth of detection precision. Moreover, the illustrative comparison of outcomes from detecting objects on the framework calls for attention to the capacity of the networks to objectify in different categories. Our outcome also suggest that although the proposed framework has a human-level performance in task of object detection there is still a scope to enhance their performance above the human level. Summarizing, this work concludes the conversation and can be considered in the progress of research developments of the object detection systems aimed at practical object meeting situations. © 2024 IEEE.

关键词： computer Vision Deep Learning Machine Learning Object Detection Visual Recognition

来源：评论

学校读者我要写书评

暂无评论

Adaptive Defense Mechanisms Against Zero-Day Attacks in Wireless Sensor networks

Adaptive Defense Mechanisms Against Zero-Day Attacks in Wire...

引用

2024 International Conference on Distributed Systems, computer networks and Cybersecurity, ICDSCNC 2024

作者： Chaturvedi, Abhay Kukreti, Sanjeev Durga Bhavani, A. Nizampatnam, Venkata Naga Ravi Kiran Kaur, Davinder Natrayan, L. Gla University Department of Electronics & Communication Engineering Chaumuhan India Graphic Era Deemed to Be University Department of Computer Science & Engineering Dehradun India Bms Institute of Technology and Management Department of Computer Science and Engineering Bengaluru India Expert Network Security Engineer Acxiom Llc Department of Computer Networks AustinTX United States Symbiosis Law School Department of Law Pune India Saveetha School of Engineering Simats Department of Mechanical Engineering Kuthambakkam India

ISBN: (纸本)9798350375442

Important fields including healthcare, environmental monitoring, and military operations are seeing a rise in the use of wireless sensor networks (WSNs). But they are susceptible to complex cyber dangers, especially zero-day assaults, due to their dispersed nature and lack of resources. In order to protect WSNs against zero-day attacks, this research introduces an Adaptive Defense Mechanism (ADM). A dynamic response system that can adjust to changing attack patterns in real-time is integrated with machine learning-based anomaly detection in the proposed ADM. A hybrid model integrating Support Vector Machines (SVM) and Long Short-Term Memory (LSTM) networks is used by the anomaly detection module. It achieves a detection accuracy of 94.7% on simulated zero-day assault scenarios. In comparison to static defensive systems, the ADM's adaptive response mechanism, which is based on reinforcement learning, dynamically modifies the network's security regulations, resulting in a 23.5% reduction in false positives. Results from testing the proposed ADM in a simulated WSN setting show that it can greatly improve the network's ability to withstand zero-day assaults. With assaults present, the total network throughput increased by 15% and the average detection time dropped to 0.35 seconds. Based on these findings, the ADM provides a strong and effective defensive technique, guaranteeing the safety and dependability of WSNs in mission-critical applications. © 2024 IEEE.

关键词： Zero-day attack

来源：评论

学校读者我要写书评

暂无评论

Dual-Mode Resonant Sectorial Monopole Antenna with Stable Backfire Gain

引用

Chinese Journal of Electronics 2024年第6期33卷 1478-1486页

作者： Feiyan JI Heng ZHANG Xiuqiong XING Wenjun LU Lei ZHU Jiangsu Key Laboratory of Wireless Communications Nanjing University of Posts and Telecommunications Key Laboratory of Wireless Sensor Network & Communication Shanghai Institute of Microsystem and Information Technology Chinese Academy of Sciences Department of Electrical and Computer Engineering Faculty of Science and Technology University of Macau

A novel design approach to wideband, dual-mode resonant monopole antenna with stable, enhanced backfire gain is advanced. The sectorial monopole evolves from a linear, 0.75-wavelength electric prototype monopole under wideband dual-mode resonant operation. As theoretically predicted by the two resonant modes TE3/5,1and TE9/5,1within a 150° radiator, the operation principle is revealed at first. As have been numerically demonstrated and experimentally validated at 2.4-GHz band, the designed antenna exhibits a wide impedance bandwidth over 90.1%（i.e., 2.06–5.44 GHz）, in which the stable gain bandwidth in the backfire,-x-direction（θ = 90°, φ = 180°） with peak value of 3.2 dBi and fluctuation less than 3 dB is up to 45.3%（i.e., 3.74–5.44 GHz）. It is concluded that the stable wideband backfire gain frequency response should be owing to the high-order resonant mode in the unique sectorial monopole antennas.

关键词： Fluctuations Azimuth Prototypes Metals Resonance Frequency measurement Broadband antennas Impedance Gain Wideband

来源：评论

学校读者我要写书评

暂无评论

Modeling and Performance Analysis of UAV-Aided Millimeter Wave Cellular networks with Stochastic Geometry

引用

China Communications 2024年第6期21卷 146-162页

作者： Li Junruo Wang Yuanjie Cui Qimei Hou Yanzhao Tao Xiaofeng National Engineering Laboratory for Mobile Network Technologies Beijing University of Posts and TelecommunicationsBeijing 100876China National Computer Network Emergency Response Technical Team/Coordination Center of China(CNCERT/CC) Beijing 100029China Department of Broadband Communication Peng Cheng LaboratoryShenzhen 518055China

UAV-aided cellular networks,millimeter wave(mm-wave) communications and multi-antenna techniques are viewed as promising components of the solution for beyond-5G(B5G) and even 6G *** leveraging the power of stochastic geometry,this paper aims at providing an effective framework for modeling and analyzing a UAV-aided heterogeneous cellular network,where the terrestrial base stations(TBSs) and the UAV base stations(UBSs) coexist,and the UBSs are provided with mm-wave and multi-antenna *** modeling the TBSs as a PPP and the UBSs as a Matern hard-core point process of type Ⅱ(MPH-Ⅱ),approximated but accurate analytical results for the average rate of the typical user of both tiers are derived through an approximation method based on the mean interference-to-signal ratio(MISR) *** influence of some relevant parameters is discussed in detail,and some insights into the network deployment and optimization are *** results show that some trade-offs are worthy of being considered,such as the antenna array size,the altitude of the UAVs and the power control factor of the UBSs.

关键词： average rate downlink millimeter wave point process theory SIR stochastic geometry UAVaided cellular networks

来源：评论

学校读者我要写书评

暂无评论

Biometric Identification Systems with Both Chosen and Generated Secret Keys by Allowing Correlation

Biometric Identification Systems with Both Chosen and Genera...

引用

作者： Yachongka, Vamoua Yagi, Hideki Department of Electrical and Computer Engineering Yokohama National University Yokohama-shi240-8501 Japan Department of Computer and Network Engineering the University of Electro-Communications Chofushi182-8585 Japan

We propose a biometric identification system where the chosen- and generated-secret keys are used simultaneously, and investigate its fundamental limits from information theoretic perspectives. The system consists of two phases: enrollment and identification phases. In the enrollment phase, for each user, the encoder uses a secret key, which is chosen independently, and the biometric identifier to generate another secret key and a helper data. In the identification phase, observing the biometric sequence of the identified user, the decoder estimates index, chosen- and generated-secret keys of the identified user based on the helper data stored in the system database. In this study, the capacity region of such system is characterized. In the problem settings, we allow chosen- and generatedsecret keys to be correlated. As a result, by permitting the correlation of the two secret keys, the sum rate of the identification, chosen- and generatedsecret key rates can achieve a larger value compared to the case where the keys do not correlate. Moreover, the minimum amount of the storage rate changes in accordance with both the identification and chosen-secret key rates, but that of the privacy-leakage rate depends only on the identification rate. © 2023 The Institute of Electronics.

关键词： Biometrics

来源：评论

学校读者我要写书评

暂无评论

Explainable Artificial Intelligence in Drones: A Brief Review 14

Explainable Artificial Intelligence in Drones: A Brief Revie...

引用

14th International Conference on Information and Communication Systems, ICICS 2023

作者： Banimelhem, Omar Al-Khateeb, Baghdad Jordan University of Science and Technology Department of Network Engineering and Security Irbid Jordan Jordan University of Science and Technology Department of Computer Engineering Irbid Jordan

ISBN: (纸本)9798350307870

The lack of clarity of most artificial intelligence algorithms which have many applications has led to the emergence of explainable artificial intelligence (XAI), which includes machine learning algorithms that introduce a clear explanation of learning decisions for human users. In the field of using drones which have recently been expanded in several applications, the drone designers face many challenges due to the weather and the difficulty of making a decision during the flight. Therefore, explainable artificial intelligence may play a significant role in these models of unmanned aerial vehicle. This paper presents a brief review for the recent articles that were published in the field of using XAI in drone's applications. © 2023 IEEE.

关键词： Drones

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：