Security is a nonfunctional information system attribute that plays a crucial role in wide sensor network application domains. Security risk can be quantified as the combination of the probability that a sensor networ...Security is a nonfunctional information system attribute that plays a crucial role in wide sensor network application domains. Security risk can be quantified as the combination of the probability that a sensor network system may fail and the evaluation of the severity of the damage caused by the failure. In this paper, we devise a methodology of Rough Outlier Detection (ROD) for the detection of security-based risk factor, which originates from violations of attack requirements (namely, attack risks). The methodology elaborates dimension reduction method to analyze the attack risk probability from high dimensional and nonlinear data set, and combines it with rough redundancy reduction and the distance measurement of kernel function which is obtained using the ROD. In this way, it is possible to determine the risky scenarios, and the analysis feedback can be used to improve the sensor network system design. We illustrate the methodology in the DARPA case set study using step-by-step approach and then prove that the method is effective in lowering the rate of false alarm.展开更多
为解决乳腺肿瘤超声的定量分级问题,从超声射频信号的角度提出了一种乳腺肿瘤分级的评价算法。以乳腺影像报告和数据系统(Breast imaging reporting and data system,BI-RADS)作为分级依据,将提取的超声射频(Radio frequency,RF)信号进...为解决乳腺肿瘤超声的定量分级问题,从超声射频信号的角度提出了一种乳腺肿瘤分级的评价算法。以乳腺影像报告和数据系统(Breast imaging reporting and data system,BI-RADS)作为分级依据,将提取的超声射频(Radio frequency,RF)信号进行图像重建、图像分割并获取乳腺肿瘤感兴趣区(Region of interest,ROI)及其特征参数:熵和标准差。量化分析特征参数与病灶良恶性分级之间的关系,实现了对乳腺肿瘤的3级、4级、5级的分级,分类成功率达到84.9%。研究结果表明,超声射频信号对辅助临床诊断具有重要意义,熵和标准差可以有效地实现乳腺肿瘤超声分级。展开更多
As the network sizes continue to increase,network traffic grows exponentially.In this situation,how to accurately predict network traffic to serve customers better has become one of the issues that Internet service pr...As the network sizes continue to increase,network traffic grows exponentially.In this situation,how to accurately predict network traffic to serve customers better has become one of the issues that Internet service providers care most about.Current traditional network models cannot predict network traffic that behaves as a nonlinear system.In this paper,a long short-term memory(LSTM)neural network model is proposed to predict network traffic that behaves as a nonlinear system.According to characteristics of autocorrelation,an autocorrelation coefficient is added to the model to improve the accuracy of the prediction model.Several experiments were conducted using real-world data,showing the effectiveness of LSTM model and the improved accuracy with autocorrelation considered.The experimental results show that the proposed model is efficient and suitable for real-world network traffic prediction.展开更多
基金the Jiangsu 973 Scientific Project,the National Natural Science Foundation of China,the Jiangsu Natural Science Foundation,the Aerospace Innovation Fund,the Lianyungang Science & Technology Project
文摘Security is a nonfunctional information system attribute that plays a crucial role in wide sensor network application domains. Security risk can be quantified as the combination of the probability that a sensor network system may fail and the evaluation of the severity of the damage caused by the failure. In this paper, we devise a methodology of Rough Outlier Detection (ROD) for the detection of security-based risk factor, which originates from violations of attack requirements (namely, attack risks). The methodology elaborates dimension reduction method to analyze the attack risk probability from high dimensional and nonlinear data set, and combines it with rough redundancy reduction and the distance measurement of kernel function which is obtained using the ROD. In this way, it is possible to determine the risky scenarios, and the analysis feedback can be used to improve the sensor network system design. We illustrate the methodology in the DARPA case set study using step-by-step approach and then prove that the method is effective in lowering the rate of false alarm.
文摘为解决乳腺肿瘤超声的定量分级问题,从超声射频信号的角度提出了一种乳腺肿瘤分级的评价算法。以乳腺影像报告和数据系统(Breast imaging reporting and data system,BI-RADS)作为分级依据,将提取的超声射频(Radio frequency,RF)信号进行图像重建、图像分割并获取乳腺肿瘤感兴趣区(Region of interest,ROI)及其特征参数:熵和标准差。量化分析特征参数与病灶良恶性分级之间的关系,实现了对乳腺肿瘤的3级、4级、5级的分级,分类成功率达到84.9%。研究结果表明,超声射频信号对辅助临床诊断具有重要意义,熵和标准差可以有效地实现乳腺肿瘤超声分级。
基金supported by ZTE Industry-Academia-Research Cooperation Funds under Grant No.2016ZTE04-11National Key Research and Development Program:Key Projects of International Scientific and Technological Innovation Cooperation Between Governments under Grant No.2016YFE0108000+1 种基金Fundamental Research Funds for the Central Universities under Grant(30918012204)Jiangsu Province Key Research and Development Program under Grant(BE2017739)
文摘As the network sizes continue to increase,network traffic grows exponentially.In this situation,how to accurately predict network traffic to serve customers better has become one of the issues that Internet service providers care most about.Current traditional network models cannot predict network traffic that behaves as a nonlinear system.In this paper,a long short-term memory(LSTM)neural network model is proposed to predict network traffic that behaves as a nonlinear system.According to characteristics of autocorrelation,an autocorrelation coefficient is added to the model to improve the accuracy of the prediction model.Several experiments were conducted using real-world data,showing the effectiveness of LSTM model and the improved accuracy with autocorrelation considered.The experimental results show that the proposed model is efficient and suitable for real-world network traffic prediction.