检索结果-内蒙古大学图书馆

OACR2: Online Admission Control and Resource Reservation for 5G Slice networks With Deep Reinforcement Learning

IEEE Transactions on Mobile Computing 2025年

作者： Li, Fang Hao, Yijun Yang, Shusen Zhao, Peng China China Xi'an Jiaotong University Ministry of Education Key Lab for Intelligent Networks and Network Security MOE KLINNS Lab China

Network slicing architecture is expected to fulfill network applications with heterogeneous requirements through efficient slice admission control (SAC) policies. Existing SAC approaches entirely rely on current limited observations to make admission decisions, ignoring the potential impact of future demands. The short-sighted behaviors lead to poor service performance and infrastructure providers' (InPs') revenue in practice. In this paper, we propose OACR2, an online SAC approach based on deep reinforcement learning (DRL) that can exploit predictable future requests to make more precise admission control decisions for the long-term revenue, and reserve proper resources accordingly. Specifically, we design three novel schemes: (i) a requirement predictor based on long short-term memory (LSTM) and a novel input-output way to predict future unforeseen requests, (ii) a DRL admission controller based on the partially observable Markov decision process model to make precise admission decisions without accurate future request information, with the convergence strictly proved, and (iii) a decision defender to guarantee decision reliability. Extensive experiments on real-world traces demonstrate that compared to the No-wait, Wait-queue, and Wait-earliest time approaches, OACR2 improves InPs' revenue and acceptance ratio by up to 40.9% and 16.7%, respectively, without sacrificing online inference time (within 0.9 milliseconds). © 2002-2012 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Poisoning Attacks and Defenses to Learned Bloom Filters for Malicious URL Detection

引用

IEEE Transactions on Dependable and Secure Computing 2025年

作者： Dong, Fangming Wang, Pinghui Li, Rundong Cui, Xueyao Zhao, Junzhou Tao, Jing Zhang, Chen Guan, Xiaohong Zhejiang CreateLink Technology Co. Ltd Guangdong Guangzhou China Xi'an Jiaotong University MOE Key Laboratory for Intelligent Networks and Network Security Shaanxi Xi'an710049 China Tsinghua University Center for Intelligent and Networked Systems Tsinghua National Lab for Information Science and Technology Beijing China

Approximate membership query (AMQ) structures represented by the Bloom Filter and its variants have been popularly researched in recent years. Researchers have recently combined machine learning with this type of structure to reduce space consumption and computation overhead further and make remarkable progress. However, with the booming performance in space or other metrics, researchers tend to ignore the security of the trained model. The machine learning model is vulnerable to poisoning attacks, and naturally, we infer that the learning-based filters also have the same deficiencies. Hence, in this paper, to confirm the inference mentioned above, experiments on the real-world datasets of URLs are conducted and prove that it is necessary to consider the security issue when using learning-based filters. We show that by data poisoning, the attacker can deflect learned Bloom Filters to make a false identification, which can lead to a significant loss in some cases. Aiming to solve this issue, we put forward a method named Defensive Learned Bloom Filter (DLBF) to diminish the influence of data poisoning and achieve a better performance compared to types of learned Bloom Filters. © 2004-2012 IEEE.

关键词： Wiener filtering

来源：评论

学校读者我要写书评

暂无评论

Autonomy on intelligent home network

引用

Information Technology Journal 2010年第2期9卷 282-289页

作者： Liu, Liguo MOE Key Lab. for Intelligent Networks and Network Security Xi'an Jiaotong University Xi'an 710049 China

Autonomy is an effective approach to cracking some hard nuts in an intelligent system. In this study on intelligent home network, the focus is placed on the implementation of autonomy of a device. For this reason, some related concepts such as Virtual Device are introduced in this study. Consisting of a family of function agents and possessing the ability for auto-configuration, auto-announcement, auto-service discovery and so on, a virtual device interacts with users or surrounding devices on behalf of the physical one. Based on the virtual device, the autonomy of a device as well as the entire home network is implemented. © 2010 Asian Network tor Scientific Information.

关键词： Home networks

来源：评论

学校读者我要写书评

暂无评论

Robust Quantum Control using Reinforcement Learning from Demonstration

arXiv

引用

arXiv 2025年

作者： Li, Shengyong Fan, Yidian Li, Xiang Ruan, Xinhui Zhao, Qianchuan Peng, Zhihui Wu, Re-Bing Zhang, Jing Song, Pengtao Department of Automation Tsinghua University Beijing100084 China Institute of Physics Chinese Academy of Sciences Beijing100190 China Department of Physics Synergetic Innovation Center for Quantum Effects and Applications Hunan Normal University Changsha410081 China School of Automation Science and Engineering Xi’an Jiaotong University Xi’an710049 China MOE Key Lab for Intelligent Networks and Network Security Xi’an Jiaotong University Xi’an710049 China

Quantum control requires high-precision and robust control pulses to ensure optimal system performance. However, control sequences generated with a system model may suffer from model bias, leading to low fidelity. While model-free reinforcement learning (RL) methods have been developed to avoid such biases, training an RL agent from scratch can be time-consuming, often taking hours to gather enough samples for convergence. This challenge has hindered the broad application of RL techniques to larger and more complex quantum control issues, limiting their adaptability. In this work, we use Reinforcement Learning from Demonstration (RLfD) to leverage the control sequences generated with system models and further optimize them with RL to avoid model bias. By avoiding learning from scratch and starting with reasonable control pulse shapes, this approach can increase sample efficiency by reducing the number of samples, which can significantly reduce the training time. Thus, this method can effectively handle pulse shapes that are discretized into more than 1000 pieces without compromising final fidelity. We have simulated the preparation of several high-fidelity non-classical states using the RLfD method. We also find that the training process is more stable when using RLfD. In addition, this method is suitable for fast gate calibration using reinforcement learning. © 2025, CC BY.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

AOD estimation in WSN localization system with synthetic aperture technique

引用

Science China(Information Sciences) 2012年第10期55卷 2216-2225页

作者： WANG WenJie,YAO BoBin & YIN QinYe moe key lab for intelligent networks and Network security,School of Electronic and Information Engineering,Xi’an Jiaotong University,Xi’an 710049,China 1. MOE Key Lab for Intelligent Networks and Network Security School of Electronic and Information Engineering Xi’an Jiaotong University Xi’an 710049 China

Angel of departure(AOD) based distributed localization scheme in wireless sensor networks allows the single antenna sensor node to estimate its own AOD information by an equivalent antenna array for further position ***,the condition that the number of antennas at the multi-antenna agent/anchor node should be larger than the number of multipath signal components(MPCs) constrains its actual *** this paper,we provide a more practical approach to achieve AOD information estimation under multipath *** that the sensor node can not identify the line of sight(LOS) signal component due to the insufficient degree of freedom,which is determined by the above limiting condition,we introduce a synthetic aperture technique with the aid of a fast moving multi-antenna agent/anchor *** operation greatly enlarges the degree of freedom of the equivalent antenna array constructed at sensor node end but without requiring other extra physical antennas equipped at agent/anchor node or sensor node end,which means that it relaxes the limiting ***,the AOD information of LOS signal component can be estimated and extracted successfully in complex multipath *** analysis proves the effectiveness of our proposed synthetic aperture technique and numerical simulations show great performance improvement.

关键词： angle of departure degree of freedom synthetic aperture wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

Exploiting Bitcoin Mining Pool for Stealthy and Flexible Botnet Channels 8

Exploiting Bitcoin Mining Pool for Stealthy and Flexible Bot...

引用

8th International Conference on Dependable Systems and Their Applications, DSA 2021

作者： Pan, Pengyu Ma, Xiaobo Bian, Huafeng Xi'an Jiaotong University MOE Key Lab for Intelligent Networks and Network Security Xi'an China

ISBN: (纸本)9781665443913

Botnets are used by hackers to conduct cyber attacks and pose a huge threat to Internet users. The key of botnets is the command and control (CC) channels. security researchers can keep track of a botnet by capturing and analyzing the communication traffic between CC servers and bots. Hence, the botmaster is constantly seeking more covert CC channels to stealthily control the botnet. This paper designs a new botnet dubbed mp-botnet wherein bots communicate with each other based on the Stratum mining pool protocol. The mp-botnet botnet completes information transmission according to the communication method of the Stratum protocol. The communication traffic in the botnet is disguised as the traffic between the mining pool and the miners in a Bitcoin network, thereby achieving better stealthiness and flexibility. © 2021 IEEE.

关键词： Botnet

来源：评论

学校读者我要写书评

暂无评论

Effectively and Efficiently Defending Shadowsocks against Website Fingerprinting Attacks 8

Effectively and Efficiently Defending Shadowsocks against We...

引用

8th International Conference on Dependable Systems and Their Applications, DSA 2021

作者： Huang, Shuangshuang Ma, Xiaobo Bian, Huafeng Xi'an Jiaotong University MOE Key Lab for Intelligent Networks and Network Security Xi'an China

ISBN: (纸本)9781665443913

Shadowsocks is one of the most popular encryption proxies. It is designed with few defense methods against website fingerprinting attack (WFP). WFP can identify user accessed websites by analyzing Shadowsocks traffic. Existing studies defending Shadowsocks against WFP generally pad or insert packets to minimize statistical differences between accesses to different websites. They can achieve promising denfense effects, yet at the cost of large time delay and bandwidth overhead. Therefore, it is necessary to design an effective and efficient algorithm to defend Shadowsocks against WFP. In this paper, we propose two algorithms, namely, PadS and PadI, which are based on the size and the true distribution of time series of traffic data packets, respectively. The results show that PadI and PadS can substantially degrade the attack performance while reducing bandwidth consumption to 34% and 32%, respectively, and the combination of the two algorithms outperforms existing defense algorithms. © 2021 IEEE.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

A Link Flooding Attack Detection Method based on Non-Cooperative Active Measurement 8

A Link Flooding Attack Detection Method based on Non-Coopera...

引用

8th International Conference on Dependable Systems and Their Applications, DSA 2021

作者： Wang, Xin Ma, Xiaobo Qu, Jian Xi'an Jiaotong University MOE Key Lab for Intelligent Networks and Network Security Xi'an China

ISBN: (纸本)9781665443913

In recent years, a new type of DDoS attacks against backbone routing links have appeared. They paralyze the communication network of a large area by directly congesting the key routing links concerning the network accessibility of the area. This new type of DDoS attacks make it difficult for traditional countermeasures to take effect. This paper proposes and implements an attack detection method based on non-cooperative active measurement. Experiments show that our detection method can efficiently perceive changes of network link performance and assist in identifying such new DDoS attacks. In our testbed, the network anomaly detection accuracy can reach 93.7%. © 2021 IEEE.

关键词： Denial-of-service attack

来源：评论

学校读者我要写书评

暂无评论

On the Performance Evaluation of Congestion Control Algorithms against DDoS Attacks 8

On the Performance Evaluation of Congestion Control Algorith...

引用

8th International Conference on Dependable Systems and Their Applications, DSA 2021

作者： Bian, Huafeng Ma, Xiaobo Pan, Pengyu Xi'an Jiaotong University MOE Key Lab for Intelligent Networks and Network Security Xi'an China

ISBN: (纸本)9781665443913

DoS attacks are currently one of the main security issues threatening the Internet. The harm caused by DDoS attacks and effective countermeasures have been research hot topics in the field of network security. However, how DDoS attacks affect different congest control mechanisms has not been investigated. In this paper, through theoretical analysis of congestion control algorithms in the TCP/IP protocol, we analyze the performance and fairness of multiple congestion control algorithms in the DDoS environment. Experiments show that the Cubic algorithm can make full use of network bandwidth and has good fairness under DDoS attacks. This finding is of great significance for further understanding of DDoS attacks and effective countermeasures. © 2021 IEEE.

关键词： Denial-of-service attack

来源：评论

学校读者我要写书评

暂无评论

In-depth measurement and analysis on densification power law of software execution 5

In-depth measurement and analysis on densification power law...

引用

5th International Workshop on Emerging Trends in Software Metrics, WETSoM 2014

作者： Qu, Yu Zheng, Qinghua Liu, Ting Li, Jian Guan, Xiaohong MOE Key Lab for Intelligent Networks and Network Security Xi'an Jiaotong University Xi'an China

ISBN: (纸本)9781450328548

Measuring software execution is important for many software engineering tasks. In this paper, Densification Power Law (DPL) of software execution is measured and studied as a feature of growing software complexity. Densification means that during a networked system's evolution, it usually becomes denser and the number of edges and nodes grows with a consistent super linear relation. This feature was discovered and reported in 2005. In this paper, based on a measurement of 15 open-source Java programs, we show that when software systems are modeled as a series of dynamic Call Graphs during their executions, they always obey DPL with very close correlation. Then a comparison between static Call Graph and DPL is presented, showing that DPL's properties cannot be derived statically. An explanation for DPL of software execution is given and verified. We believe the universality of DPL makes it an appropriate metric for software execution process. Copyright 2014 ACM.

关键词： Densification

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：