检索结果-南通市图书馆

Unsupervised Interpolation Recovery Method for Spectrum Anomaly Detection and Localization

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Space(Science & Technology) 2023年第1期3卷 560-571页

作者： Yishi Huang Shuai Yuan naijin liu Qing Li Wenyu Liang Lei liu Shaanxi Aerospace Flight Vehicle Design Key Laboratory School of AstronauticsNorthwestern Polytechnical UniversityXi'anChina Qian Xuesen Laboratory of Space Technology China Academy of Space TechnologyBeijingChina Department of Electrical and Computer Engineering National University of SingaporeSingaporeSingapore

With the growing efficiency of the use of unlicensed spectrum,the challenge of ensuring spectrum security has become increasingly *** managers aim to accurately and efficiently detect and recognize anomaly behaviors in the *** this study,we propose a novel framework for spectrum anomaly detection and localization by spectrum interpolation *** interpolation recovery refers to the recovery of the rest of the spectrum distribution based on a part of the spectrum distribution,which is achieved through a masked autoencoder(MAE)model with a core of multi-head self-attention(MHSA)*** spectrum interpolation recovery method restores the region where the masked abnormal signals are present,yielding anomaly-free results,with the difference between the restored and the masked representing the anomaly *** proposed method has been demonstrated to effectively reduce model-induced over-recovery of anomalous signals and dilute large-scale generation errors caused by anomalies,thereby improving the detection and localization performance of anomaly signals,and improving the area under the receiver operating characteristic curve(AUC)and the area under the precision-recall curve(AUPRC)by 0.0382(3.68%)and 0.1992(68.90%),*** a designed dataset containing 3 variables of interference-to-signal ratio(ISR),signal-to-noise ratio(SNR),and anomaly type,the total recall of anomaly detection and localization at a 5%false alarm rate reached 0.8799 and 0.5536,***,a comparative study among different methods demonstrates the effectiveness and rationality of the proposed method.

关键词： interpolation spectrum thereby

Decentralized Distributed Deep Learning with Low-Bandwidth Consumption for Smart Constellations

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Space(Science & Technology) 2021年第1期2021卷 56-65页

作者： Qingliang Meng Meiyu Huang Yao Xu naijin liu Xueshuang Xiang Qian Xuesen Laboratory of Space Technology China Academy of Space TechnologyBeijingChina

For the space-based remote sensing system,onboard intelligent processing based on deep learning has become an inevitable *** adapt to the dynamic changes of the observation scenes,there is an urgent need to perform distributed deep learning onboard to fully utilize the plentiful real-time sensing data of multiple satellites from a smart ***,the network bandwidth of the smart constellation is very ***,it is of great significance to carry out distributed training research in a low-bandwidth *** paper proposes a Randomized Decentralized Parallel Stochastic Gradient Descent(RD-PSGD)method for distributed training in a low-bandwidth *** reduce the communication cost,each node in RD-PSGD just randomly transfers part of the information of the local intelligent model to its *** further speed up the algorithm by optimizing the programming of random index generation and parameter *** the first time,we theoretically analyze the convergence property of the proposed RD-PSGD and validate the advantage of this method by simulation experiments on various distributed training tasks for image classification on different benchmark datasets and deep learning network *** results show that RD-PSGD can effectively save the time and bandwidth cost of distributed training and reduce the complexity of parameter selection compared with the TopK-based *** method proposed in this paper provides a new perspective for the study of onboard intelligent processing,especially for online learning on a smart satellite constellation.

关键词： Deep Smart smart

A multi-path routing algorithm of LEO satellite networks based on an improved ant colony system

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

High Technology Letters 2014年第3期20卷 253-260页

作者：王厚天 Zhang Qi Xin Xiangjun Tao Ying Chen Dong liu naijin School of Electronic Engineering Beijing University of Posts and Telecommunications State Key Lab of Information Photonics and Optical Communication

Geography rectangle is used to reduce signaling overhead of the LEO satellite ***,a multi-path routing algorithm based on an improved ant colony system(MPRA-AC) is *** indicating the importance of the link between satellites is introduced into MPRA-AC in order to find the optimal path more *** results show that MPRA-AC reduces the number of iterations to achieve a satisfactory *** the same time,the packet delivery ratio of LEO satellite networks when running MPRA-AC and DSR-LSN(dynamic source routing algorithm for LEO satellite networks) is *** packet delivery ratio is about 7.9%lower when running ***,because of the mechanism of active load balancing of MPRA-AC,simulation results show that MPRA-AC outperforms DSR-LSN in link utilization when data packets are transmitted in the networks.

关键词： ant colony algorithm low earth orbit (LEO) packet delivery ratio routing satellite networks

Unsupervised Spectrum Anomaly Detection Method for Unauthorized Bands

在线全文

学校读者我要写书评

暂无评论

Space(Science & Technology) 2022年第1期2022卷 462-471页

作者： Yu Tian Haihua Liao Jing Xu Ya Wang Shuai Yuan naijin liu Qian Xuesen Laboratory of Space Technology China Academy of Space TechnologyBeijingChina

With the rapid development of wireless communication,spectrum plays increasingly important role in both military and civilian *** anomaly detection aims at detecting emerging anomaly signals and spectrum usage behavior in the environment,which is indispensable to secure safety and improve spectrum ***,spectrum anomaly detection faces many difficulties,especially for unauthorized frequency *** unauthorized bands,the composition of spectrum is complex and the anomaly usage patterns are unknown in *** this paper,a Variational Autoencoder-(VAE-)based method is proposed for spectrum anomaly detection in unauthorized *** of all,we theoretically prove that the anomalies in unauthorized bands will introduce Background Noise Enhancement(BNE)effect and Anomaly Signal Disappearance(ASD)effects after VAE ***,we introduce a novel anomaly metric termed as percentile(PER)score,which focuses on capturing the distribution variation of reconstruction error caused by ASD and *** order to verify the effectiveness of our method,we developed an ISM Anomaly Detection(IAD)*** proposed PER score achieves superior performance against different type of *** QPSK type anomaly,our method increases the recall rate from 80%to 93%while keeping a false alarm rate of 5%.The proposed method is beneficial to broadband spectrum sensing and massive spectrum data *** code will be released at git@***:QXSLAB/vae_ism_***.

关键词： bands authorized false

Dynamic routing and wavelength assignment algorithm of optical satellite networks based on cross-layer design

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

High Technology Letters 2017年第3期23卷 252-259页

作者：文国莉 Zhang Qi Tian Qinghua Wang Houtian Tao Ying Chen Dong liu naijin Shen Yufei Zhou Na School of Electronic Engineering Beijing University of Posts and Telecommunications Beijing 100876 P. R. China China Academy of Space Technology Beijing 100094 P. R. China

In order to overcome the adverse effects of Doppler wavelength shift on data transmission in the optical satellite networks,a dynamic routing and wavelength assignment algorithm based on crosslayer design( CL-DRWA) is introduced which can improve robustness of the network. Above all,a cross-layer optimization model is designed,which considers transmission delay and wavelength-continuity constraint,as well as Doppler wavelength shift. Then CL-DRWA is applied to solve this model,resulting in finding an optimal light path satisfying the above constraints for every connection request. In CL-DRWA,Bellman-Ford method is used to find an optimal route and a distributed relative capacity loss method is implemented to get an optimal wavelength assignment result on the optimal route. Moreover,compared with the dynamic routing and wavelength assignment algorithm based on minimum delay strategy( MD-DRWA),CL-DRWA can make an improvement of 5. 3% on the communication success probability. Meanwhile,CL-DRWA can meet the requirement of transmission delay for real-time services.

关键词： cross-layer design Doppler wavelength shift dynamic routing and wavelength assignment optical satellite network

多小区MIMO系统中的一种主动式干扰抑制方案

在线全文

学校读者我要写书评

暂无评论

数据采集与处理 2008年第6期23卷 718-723页

作者：刘乃金邱玲朱近康中国科学技术大学个人通信与扩频实验室合肥230027

针对多小区下行链路中的同频干扰问题,提出了一种新的主动式干扰抑制方案。该方案通过在各基站独立地采用预编码的方式来主动抑制本小区基站对其他相邻小区所产生的干扰。文中提出了两种实现算法:(1)迫零预编码,它可以获取和无同频干扰... 详细信息

针对多小区下行链路中的同频干扰问题,提出了一种新的主动式干扰抑制方案。该方案通过在各基站独立地采用预编码的方式来主动抑制本小区基站对其他相邻小区所产生的干扰。文中提出了两种实现算法:(1)迫零预编码,它可以获取和无同频干扰系统同样的性能;(2)基于干扰信道最大特征子信道的线性预编码,它能够有效地实现抑制干扰和提高本小区发送自由度的折衷。仿真结果表明,本文方案提高了系统的性能和频谱效率。

关键词：多输入多输出(MIM0) 多小区干扰抑制预编码

在线全文

学校读者我要写书评

暂无评论

基于分布式深度学习的多星计算卸载策略

中国空间科学技术 2023年第2期43卷 73-80页

作者：周锦雯刘乃金陈清霞钱学森空间技术实验室北京100094

在边缘计算增强的低轨卫星网络场景下,低轨卫星集群协同处理地面任务能有效降低用户响应时延。对卫星集群的联合卸载决策和资源分配优化问题进行研究,将其描述为一个混合整数规划问题,并采用了一种基于分布式深度学习算法的卫星边缘计... 详细信息

在边缘计算增强的低轨卫星网络场景下,低轨卫星集群协同处理地面任务能有效降低用户响应时延。对卫星集群的联合卸载决策和资源分配优化问题进行研究,将其描述为一个混合整数规划问题,并采用了一种基于分布式深度学习算法的卫星边缘计算卸载算法(deep learning-based offloading algorithm, DLOA)。该算法使用多个并行DNN用于生成卸载决策并采用经验回放存储新生成的卸载决策,当采用隐藏层结构不同的DNN,收敛速度比同构DNN提升18%,收敛值与最优值的比值基本为1,可以认为已收敛至最优。此外,探讨了DNN的数量对所使用的算法的影响,仿真结果表明采用少量DNN就可以获得近优的收敛效果。通过对不同任务规模下采用不同算法的任务完成率进行研究,结果表明DLOA算法可通过采用异构DNN和优化资源分配方案显著提升完成率,其较单星运算方案任务完成率提升1倍,较二进制粒子群算法方案提升20%。

关键词：边缘计算计算卸载低轨卫星网络深度学习收敛性能

维普期刊数据库博看期刊

在线全文

学校读者我要写书评

暂无评论

一种基于计算资源度的星间路由协议

中国空间科学技术 2023年第3期43卷 134-143页

作者：苏安刘乃金陈清霞向雪霜刘佳钱学森空间技术实验室北京100094 航天东方红卫星有限公司北京100094

随着发射技术的提升和星载任务需求的扩大,低轨互联网星座逐渐成为航天产业的研究热点。在网络层进行算力资源感知有助于构建最短计算时延路径,有效利用边缘或路径上的节点资源进行任务调度。由于传统组网协议没有考虑邻居节点算力资源... 详细信息

随着发射技术的提升和星载任务需求的扩大,低轨互联网星座逐渐成为航天产业的研究热点。在网络层进行算力资源感知有助于构建最短计算时延路径,有效利用边缘或路径上的节点资源进行任务调度。由于传统组网协议没有考虑邻居节点算力资源情况,难以做到资源感知、星间协同组网完成卸载任务。为解决上述问题,提出了将低轨卫星组网建模为移动自组织网络(MANET),并在主动式协议优化链路状态路由协议(OLSR)中引入节点计算资源度来感知周边组网节点、CPU、内存和负载等计算能力情况,并且根据该指标修改OLSR中的多点中继(multi point relay,MPR)选择算法与路由表更新算法。通过仿真验证了在星间协同计算中,文章提出的路由协议在任务计算时延上降低了15%~30%,并通过与地面云计算的比较验证了星间协同计算的优势。

关键词：移动自组织网络 OLSR 算力网络 MPR选择算法计算资源感知

同方期刊数据库博看期刊

在线全文

学校读者我要写书评

暂无评论

一种重力测量卫星质心在轨标定改进算法

航天器工程 2015年第4期24卷 44-50页

作者：辛宁邱乐德张立华刘乃金中国空间技术研究院通信卫星事业部北京100094 航天东方红卫星有限公司北京100094

针对重力测量卫星质心在轨标定算法中存在的陀螺精度不高且易失效的问题,提出了一种应用星敏感器和静电加速度计的质心标定改进算法。在该算法中,将一个明显大于干扰力矩的周期性磁力矩,作用于卫星上;将星敏感器观测量代入预测滤波器中... 详细信息

针对重力测量卫星质心在轨标定算法中存在的陀螺精度不高且易失效的问题,提出了一种应用星敏感器和静电加速度计的质心标定改进算法。在该算法中,将一个明显大于干扰力矩的周期性磁力矩,作用于卫星上;将星敏感器观测量代入预测滤波器中,估计卫星的角速度及角加速度;利用静电加速度计信息设计扩展卡尔曼滤波,从而实现卫星质心的标定。数学仿真结果表明:此算法能够对卫星的角速度及质心位置进行实时估计,卫星质心三轴最佳标定精度均优于0.05mm,可实现陀螺失效情况下卫星质心较为精确的标定。

关键词：重力测量卫星质心在轨标定预测滤波扩展卡尔曼滤波星敏感器