您现在的位置: 首页> 研究主题> 隐Markov模型

隐Markov模型

隐Markov模型的相关文献在1998年到2021年内共计87篇,主要集中在自动化技术、计算机技术、机械、仪表工业、无线电电子学、电信技术 等领域,其中期刊论文77篇、会议论文8篇、专利文献152232篇;相关期刊60种,包括南京理工大学学报(社会科学版)、忻州师范学院学报、吉林大学学报(理学版)等; 相关会议8种,包括2016年全国设备监测诊断与维护学术会议、第十五届全国设备故障诊断学术会议、第十七届全国设备监测与诊断学术会议、2016年全国设备诊断工程会议、第二十二届全国信息保密学术会议(IS2012)、2010年全国电磁散射与逆散射学术年会等;隐Markov模型的相关文献由241位作者贡献,包括吴昭同、李志农、丁启全等。

隐Markov模型—发文量

期刊论文>

论文:77 占比:0.05%

会议论文>

论文:8 占比:0.01%

专利文献>

论文:152232 占比:99.94%

总计:152317篇

隐Markov模型—发文趋势图

隐Markov模型

-研究学者

  • 吴昭同
  • 李志农
  • 丁启全
  • 刘向明
  • 杨卫国
  • 林家瑞
  • 韩晓东
  • 冯长建
  • 刘国岁
  • 刘震
  • 期刊论文
  • 会议论文
  • 专利文献

搜索

排序:

年份

    • 吴建台; 乔翌峰; 朱赛凡; 刘光杰
    • 摘要: Cybersecurity situation evaluation and prediction is the vital issue of situation awareness. Following the ag-gregation of alert information,attack affairs are associated according to the attack pattern with four phases. The attack phase is identified and the threat level is obtained based on the vulnerabilities of cyber entities. Taking the threat level as the ob-servation variables of HMM,the situation value is successively figured out according to the estimation of HMM. The situa-tion prediction is ultimately performed via the composition of the neural-network-based predictor and support-vector-machine-based predictor. Experimental results based on DARPA2000 dataset indicate that the proposed method is able to achieve higher cybersecurity situation evaluation and prediction performance.%网络安全态势估计和预测是态势感知的重要过程.在告警信息聚合基础上,以典型攻击模式作为关联依据,结合网络资产的脆弱性识别网络实体所处受攻击阶段并将其转化威胁等级.以威胁等级作为观测值,基于隐Markov模型通过状态估计实现态势评估,并利用神经网络和支持向量机的组合模型实现态势预测.DARPA2000测试数据集上的相关实验表明,本文方法能更加准确地评估和预测网络态势.
    • 胡韶华; 谷振宇; 金迪文
    • 摘要: For poor diagnosis effect of existing fault diagnosis methods for axial flow fan based on spectrum analysis which correlated fault type with spectrum characteristic value simply,a fault diagnosis method for axial flow fan based on vector ellipsoid spectrum and hidden Markov model (HMM) was proposed.Firstly,two orthogonal vibration signals of axial flow fan in the same section are fused into a complex signal in time domain,and full-spectrum amplitudes of the vibration signals under multi characteristic frequencies are obtained by fast Fourier transform of the complex signal.Secondly,the fullspectrum amplitudes under different fault conditions are used to train HMM.Finally,full-spectrum amplitudes of real-time vibration signals are as input of HMM,and Viterbi algorithm is used to calculate likelihood probability outputted by HMM.Fault type is judged according to the maximum logarithm value of the likelihood probability,which avoids simple association between the vibration amplitude and fault type.The experimental result shows that correct rate of fault diagnosis of the method is above 90%.%针对现有基于谱分析的轴流通风机故障诊断方法只将故障类型和频谱特征值进行简单关联而导致诊断效果较差的问题,提出了一种基于矢椭谱和隐Markov模型的轴流通风机故障诊断方法.该方法首先将轴流通风机同一截面内互相垂直的2个振动信号在时域上直接融合为复信号,并对该复信号进行快速Fourier变换,获得多个特征频率下振动信号的全谱幅值;然后用不同故障状态下振动信号的全谱幅值训练隐Markov模型;最后以实时振动信号的全谱幅值作为隐Markov模型输入量,采用Viterbi算法计算隐Markov模型输出的似然概率,根据最大似然概率对数判断故障类型,避免了将振动幅值和故障类型进行简单关联.试验结果表明,该方法的故障诊断正确率达90%以上.
    • 廖雯竹; 崔诗好
    • 摘要: 针对传统的基于隐马尔科夫模型(HMM)的设备评估模型大多假定设备状态间的转移概率不变,忽略了实际运行中设备状态间转移概率会随使用时间的增加而变化的问题,提出一类考虑劣化因素的HMM,通过设计劣化因子来克服传统HMM的不足,并开发了一个双重的期望值最大算法来估算劣化因子和状态初始转移概率矩阵,从而对设备状态进行评估.最后,通过算例验证了模型的可行性和有效性.%The traditional Hidden Markov Model (HMM) usually assumed transition probability unfixed,which ignored the influence on transition probability by increasing usage.For this problem,an age-dependent HMM was proposed to eliminate this disadvantage so as to obtain accurate diagnosis result,in which a double Expectation Maximization (EM) algorithm was developed to estimate the aging factor and initial transition probability matrix.Through a numerical example,the computational results could prove the reliability and effectiveness of this proposed model.
    • 王平凡; 刘淑芬
    • 摘要: In order to realize the adaptive software ,we proposed an adaptive software decision model based on hidden Markov model (HMM ) for the complex and changeable operating environment . Firstly ,the Gaussian mixture model (GMM) was used to classify the initial environment .Secondly , the softmax regression was used to classify and divide the perceptual environment .Finally ,we used HMM instead of the manual intervention to make software decisions .The experimental results show that the adaptive software model can achieve the adaptive software decision well under the condition of the change of the perceptual environment .%为实现软件的自适应,针对复杂多变的运行环境,提出一个基于隐M arkov模型(HMM)的自适应软件决策模型.首先运用高斯混合模型(GMM)对初始环境进行分类,然后使用softmax回归对感知环境进行归类划分处理,最后利用HMM代替人工干预进行软件决策.实验结果表明,该自适应软件模型在感知环境发生变化的条件下,能很好地实现软件自适应决策.
    • 蒋海军; 谢钧; 段国仑; 王根春
    • 摘要: 提出了基于奇异值分解(Singular Value Decomposition,SVD)特征矩阵压缩和隐Markov模型(Hidden Markov Model,HMM)的动态手势识别方法.该方法通过SVD对特征矩阵进行时间维度的压缩,然后通过HMM的方法对提取的动态手势进行识别.通过对特征矩阵压缩可以显著地减少训练HMM的迭代计算量,提高模型的训练效率.采用Leap Motion体感控制器追踪并提取自定义的10个阿拉伯数字的动态手势特征.实验验证结果表明,该方法对这些动态手势在当前有限样本条件下的总识别率均在96%以上.
    • 蒋海军1; 谢钧1; 段国仑1; 王根春2
    • 摘要: 提出了基于奇异值分解(Singular Value Decomposition,SVD)特征矩阵压缩和隐Markov模型(Hidden Markov Model,HMM)的动态手势识别方法。该方法通过SVD对特征矩阵进行时间维度的压缩,然后通过HMM的方法对提取的动态手势进行识别。通过对特征矩阵压缩可以显著地减少训练HMM的迭代计算量,提高模型的训练效率。采用LeapMotion体感控制器追踪并提取自定义的10个阿拉伯数字的动态手势特征。实验验证结果表明,该方法对这些动态手势在当前有限样本条件下的总识别率均在96%以上。
    • 伊鹏; 周桥; 门浩崧
    • 摘要: 随着互联网的不断发展,大多数社会网络已逐渐显示出动态特性,动态社会网络社团分析对理解现实生活中社会网络结构和功能具有非常重要的意义.针对动态社会网络中的社团发现问题,提出一种基于隐Markov模型(hidden Markov model,HMM)的HMM DC算法.该算法考虑到社会网络的动态特性,结合历史信息,将社团发现转化为求解隐马尔可夫模型中的最优状态序列问题,将网络中的社团结构和节点信息分别采用状态链和观察链表示,在无须指定额外参数的情况下实现动态网络的社团结构发现.最后,利用该算法和其他算法对VAST数据集、ENRON数据集和Facebook social network数据集进行实验仿真.仿真结果表明:该算法能够快速、准确地发现真实动态网络中的社团,其模块度Q值和互信息NMI值有很大提升.%With the continuous development of the Internet,most social networks have gradually demonstrated dynamic characteristics,and dynamic analysis of social network community has a very important significance on the understanding of the structure and function of social networks in real life.The HMM_DC algorithm (hidden Markov model based on dynamic community detection) is proposed according to the HMM (hidden Markov model) to detect the community in dynamic social network.Firstly,the algorithm transforms the community detection problem to get the optimal status chain in hidden Markov model considering the history information and characteristics in dynamic social networks.And the algorithm uses the observed chain and status chain to represent the community structure and node information,and can identify the community structure without extra information.Finally,this algorithm and three other algorithms are used to make comparable simulation experiments with VAST data set,ENRON data set and Facebook social network data set.Experimental results show that HMM_DC algorithm performs effectively and accurately in identifying the community structure in the dynamic social network and the value of Q and NMI can be raised greatly compared with other three algorithms.
    • 张阳; 李琳国; 黄亦雅
    • 摘要: 为了充分利用桥梁历史数据,通过桥梁技术状况确定并预测承载能力,提出一种基于隐Markov模型的桥梁承载能力退化模型.首先对承载能力等级进行划分,再建立承载能力退化的Markov模型,最后引入桥梁技术状况与承载能力之间的概率模型,建立了承载能力退化模型.
    • 邱卫; 杨英杰; 汪永伟; 常德显
    • 摘要: As to solving the issues of insufficient training data and initial parameters sensitive in existing protocol anomaly de-tection based on hidden Markov model,this paper presented a new protocol anomaly detection method based on improved genet-ic algorithm and hidden Markov model.First,it used the local competitive selection strategy,arithmetic crossover and adaptive non-uniform mutation operator to improve the genetic algorithm,in order to avoid the premature and stagnation problemin in tra-ditional genetic algorithm.Then,it recommended the improved genetic algorithm to optimizethe initial parameters of hidden Markov model to avoid the initial model parameters sensitive issue.Finally,it took the keyword and keyword interval as training observations,described the behavior of protocol in detail to expand the training sample space.Experimental results on DARPA 1999 data set show that the method has higher detection rate and low false alarm rate.%针对现有基于隐Markov模型的协议异常检测方法中存在的训练样本不足和初始参数敏感问题,提出一种基于改进遗传算法和隐Markov模型的协议异常检测新方法。首先,采用局部竞争选择策略、算术交叉算子和自适应非均匀变异算子改进遗传算法,避免传统遗传算法在收敛过程中的早熟和停滞问题;然后,利用改进的遗传算法优化隐Markov模型的初始参数,解决模型对初始参数敏感的问题;最后,以协议关键词和关键词时间间隔作为训练观测值,细粒度地描述协议行为,扩大模型的训练样本空间。在DARPA 1999数据集上的实验结果表明,该方法具有很高的检测率和较低的误报率。
  • 查看更多

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号