社会网络应用已无处不在,在健康医疗领域也是如此.同时,传感器网络的发展也面临新的形势.在真实世界中,有许多因素(如社会关系、历史健康状态和个人属性信息)都能对健康状态检测?预测结果产生影响.然而,却很少有相关文献能够系统阐述新...社会网络应用已无处不在,在健康医疗领域也是如此.同时,传感器网络的发展也面临新的形势.在真实世界中,有许多因素(如社会关系、历史健康状态和个人属性信息)都能对健康状态检测?预测结果产生影响.然而,却很少有相关文献能够系统阐述新形势下在一个动态社会网络中节点用户健康状态如何进行检测?预测以及不同因素对用户健康状态影响到何种程度.首先描述一种新颖的医疗物联网:医疗社会网络(medical social networks,MSNs);然后统一考虑社会关系、历史健康状态和用户属性对网络用户健康状态检测结果的影响,提出一种新的基于时-空概率因子图模型(temporal-spatial factorgraph model,TS-FGM)的网络用户健康状态检测?预测方法.在Twitter数据集上对所提出的模型进行了验证,并在一个真实的临床医疗数据集上与SVM基线算法进行了对比实验.实验结果表明所提出的TS-FGM模型是有效的,健康状态检测方法也在一定程度上优于基线方法.展开更多
The key exposure problem is a practical threat for many security applications. In wireless sensor networks (WSNs), keys could be compromised easily due to its limited hardware protections. A secure group key managemen...The key exposure problem is a practical threat for many security applications. In wireless sensor networks (WSNs), keys could be compromised easily due to its limited hardware protections. A secure group key management scheme is responsible for secure distributing group keys among valid nodes of the group. Based on the key-insulated encryption (KIE), we propose a group key management scheme (KIE-GKMS), which integrates the pair-wise key pre-distribution for WSN. The KIE-GKMS scheme updates group keys dynamically when adding or removing nodes. Moreover, the security analysis proves that the KIE-GKMS scheme not only obtains the semantic security, but also provides the forward and backward security. Finally, the theoretical analysis shows that the KIE-GKMS scheme has constant performance on both communication and storage costs in sensor nodes.展开更多
Facing the problems lack of considering the non-uniform distribution of the static bias magnetic field and computing the panicle displacements in the simulation model of electromagnetic acoustic transducer （EMAT）, a...Facing the problems lack of considering the non-uniform distribution of the static bias magnetic field and computing the panicle displacements in the simulation model of electromagnetic acoustic transducer （EMAT）, a multi-field coupled model was established and the finite element method （FEM） was presented to calculate the entire transduction process. The multi-field coupled model included the static magnetic field, pulsed eddy current field and mechanical field. The FEM equations of the three fields were derived by Garlerkin FEM method. Thus, the entire transduction process of the EMAT was calculated through sequentially coupling the three fields. The transduction process of a Lamb wave EMAT was calculated according to the present model and method. The results show that, by the present method, it is valid to calculate the particle displacement under the given excitation signal and non-uniformly distributed static magnetic field. Calculation error will be brought about if the non-uniform distribution of the static bias magnetic field is neglected.展开更多
随着手机传感器的普遍使用,对人体日常行为动作识别需求越来越多,经典识别方法是利用启发式过程获得人工特征,再用机器学习方法识别动作。最新研究表明,Inception卷积结构在特征自动提取方面表现尤为突出,可避免人工提取特征带来的偏差...随着手机传感器的普遍使用,对人体日常行为动作识别需求越来越多,经典识别方法是利用启发式过程获得人工特征,再用机器学习方法识别动作。最新研究表明,Inception卷积结构在特征自动提取方面表现尤为突出,可避免人工提取特征带来的偏差问题。人体动作由复杂运动序列构成,捕捉该时间序列是动作识别必不可少的。基于此,首先对Inception结构进行了优化,提出了O-Inception结构,并将其与长短期记忆模型(long short term memory,LSTM)进行了融合,进而提出了OI-LSTM(optimization Inception-LSTM)动作识别模型。OI-LSTM模型一方面可以利用O-Inception结构实现对特征的自动提取,另一方面,还可以利用LSTM捕获动作时序,进而提高了动作识别准确率。在WISDM(wireless sensor data mining)和UCI(UC Irvine)两个数据集上进行了扩展性实验,实验结果表明,所提出的OI-LSTM动作识别模型,在WISDM和UCI两个数据集上其准确率比当前最先进的方法分别提高了约4%和1%。实验还证明,此模型拥有很强的容错性和实时性。展开更多
Based on principal component analysis, a multiple neural network was proposed. The principal component analysis was firstly used to reorganize the input variables and eliminate the correlativity. Then the reorganized ...Based on principal component analysis, a multiple neural network was proposed. The principal component analysis was firstly used to reorganize the input variables and eliminate the correlativity. Then the reorganized variables were divided into 2 groups according to the original information and 2 corresponding neural networks were established. A radial basis function network was used to depict the relationship between the output variables and the first group input variables which contain main original information. An other single-layer neural network model was used to compensate the error between the output of radial basis function network and the actual output variables. At last, The multiple network was used as soft sensor for the ratio of soda to aluminate in the process of high-pressure digestion of alumina. Simulation of industry application data shows that the prediction error of the model is less than 3%, and the model has good generalization ability.展开更多
文摘社会网络应用已无处不在,在健康医疗领域也是如此.同时,传感器网络的发展也面临新的形势.在真实世界中,有许多因素(如社会关系、历史健康状态和个人属性信息)都能对健康状态检测?预测结果产生影响.然而,却很少有相关文献能够系统阐述新形势下在一个动态社会网络中节点用户健康状态如何进行检测?预测以及不同因素对用户健康状态影响到何种程度.首先描述一种新颖的医疗物联网:医疗社会网络(medical social networks,MSNs);然后统一考虑社会关系、历史健康状态和用户属性对网络用户健康状态检测结果的影响,提出一种新的基于时-空概率因子图模型(temporal-spatial factorgraph model,TS-FGM)的网络用户健康状态检测?预测方法.在Twitter数据集上对所提出的模型进行了验证,并在一个真实的临床医疗数据集上与SVM基线算法进行了对比实验.实验结果表明所提出的TS-FGM模型是有效的,健康状态检测方法也在一定程度上优于基线方法.
基金Project(61100201) supported by National Natural Science Foundation of ChinaProject(12ZZ019) supported by Technology Innovation Research Program,Shang Municipal Education Commission,China+1 种基金Project(LYM11053) supported by the Foundation for Distinguished Young Talents in Higher Education of Guangdong Province,ChinaProject(NCET-12-0358) supported by New Century Excellent Talentsin University,Ministry of Education,China
文摘The key exposure problem is a practical threat for many security applications. In wireless sensor networks (WSNs), keys could be compromised easily due to its limited hardware protections. A secure group key management scheme is responsible for secure distributing group keys among valid nodes of the group. Based on the key-insulated encryption (KIE), we propose a group key management scheme (KIE-GKMS), which integrates the pair-wise key pre-distribution for WSN. The KIE-GKMS scheme updates group keys dynamically when adding or removing nodes. Moreover, the security analysis proves that the KIE-GKMS scheme not only obtains the semantic security, but also provides the forward and backward security. Finally, the theoretical analysis shows that the KIE-GKMS scheme has constant performance on both communication and storage costs in sensor nodes.
基金Project(10974115) supported by the National Natural Science Foundation of China
文摘Facing the problems lack of considering the non-uniform distribution of the static bias magnetic field and computing the panicle displacements in the simulation model of electromagnetic acoustic transducer （EMAT）, a multi-field coupled model was established and the finite element method （FEM） was presented to calculate the entire transduction process. The multi-field coupled model included the static magnetic field, pulsed eddy current field and mechanical field. The FEM equations of the three fields were derived by Garlerkin FEM method. Thus, the entire transduction process of the EMAT was calculated through sequentially coupling the three fields. The transduction process of a Lamb wave EMAT was calculated according to the present model and method. The results show that, by the present method, it is valid to calculate the particle displacement under the given excitation signal and non-uniformly distributed static magnetic field. Calculation error will be brought about if the non-uniform distribution of the static bias magnetic field is neglected.
文摘随着手机传感器的普遍使用,对人体日常行为动作识别需求越来越多,经典识别方法是利用启发式过程获得人工特征,再用机器学习方法识别动作。最新研究表明,Inception卷积结构在特征自动提取方面表现尤为突出,可避免人工提取特征带来的偏差问题。人体动作由复杂运动序列构成,捕捉该时间序列是动作识别必不可少的。基于此,首先对Inception结构进行了优化,提出了O-Inception结构,并将其与长短期记忆模型(long short term memory,LSTM)进行了融合,进而提出了OI-LSTM(optimization Inception-LSTM)动作识别模型。OI-LSTM模型一方面可以利用O-Inception结构实现对特征的自动提取,另一方面,还可以利用LSTM捕获动作时序,进而提高了动作识别准确率。在WISDM(wireless sensor data mining)和UCI(UC Irvine)两个数据集上进行了扩展性实验,实验结果表明,所提出的OI-LSTM动作识别模型,在WISDM和UCI两个数据集上其准确率比当前最先进的方法分别提高了约4%和1%。实验还证明,此模型拥有很强的容错性和实时性。
基金Project ( 2001AA411040 ) supported by the National High Technology Development Program of China project(2002CB312200) supported by the National Fundamental Research and Development Program of China
文摘Based on principal component analysis, a multiple neural network was proposed. The principal component analysis was firstly used to reorganize the input variables and eliminate the correlativity. Then the reorganized variables were divided into 2 groups according to the original information and 2 corresponding neural networks were established. A radial basis function network was used to depict the relationship between the output variables and the first group input variables which contain main original information. An other single-layer neural network model was used to compensate the error between the output of radial basis function network and the actual output variables. At last, The multiple network was used as soft sensor for the ratio of soda to aluminate in the process of high-pressure digestion of alumina. Simulation of industry application data shows that the prediction error of the model is less than 3%, and the model has good generalization ability.
