The recognition of electroencephalogram (EEG) signals is the key of brain computer interface (BCI). Aimed at the problem that the recognition rate of EEG by using support vector machine (SVM) is low in BCI, based on t...The recognition of electroencephalogram (EEG) signals is the key of brain computer interface (BCI). Aimed at the problem that the recognition rate of EEG by using support vector machine (SVM) is low in BCI, based on the assumption that a well-defined physiological signal which also has a smooth form "hides" inside the noisy EEG signal, a Quasi-Newton-SVM recognition method based on Quasi-Newton method and SVM algorithm was presented. Firstly, the EEG signals were preprocessed by Quasi-Newton method and got the signals which were fit for SVM. Secondly, the preprocessed signals were classified by SVM method. The present simulation results indicated the Quasi-Newton-SVM approach improved the recognition rate compared with using SVM method; we also discussed the relationship between the artificial smooth signals and the classification errors.展开更多
In order to improve the precision of mining subsidence prediction, a mathematical model using Support Vector Machine （SVM） was established to calculate the displacement factor. The study is based on a comprehensive ...In order to improve the precision of mining subsidence prediction, a mathematical model using Support Vector Machine （SVM） was established to calculate the displacement factor. The study is based on a comprehensive analysis of factors affecting the displacement factor, such as mechanical properties of the cover rock, the ratio of mining depth to seam thickness, dip angle of the coal seam and the thickness of loose layer. Data of 63 typical observation stations were used as a training and testing sample set. A SVM regression model of the displacement factor and the factors affecting it was established with a kernel function, an insensitive loss factor and a properly selected penalty factor. Given an accurate calculation algorithm for testing and analysis, the results show that an SVM regression model can calcu- late displacement factor precisely and reliable precision can be obtained which meets engineering requirements. The experimental results show that the method to calculation of the displacement factor, based on the SVM method, is feasible. The many factors affecting the displacement factor can be consid- ered with this method. The research provides an efficient and accurate approach for the calculation of displacement in mining subsidence orediction.展开更多
针对综合孔径微波辐射计射频干扰(Radio Frequency Interference,RFI)定位与抑制中缺乏科学简便信源估计方法的问题,提出一种基于支持向量机(Support Vector Machine,SVM)与特征值差分的RFI信源估计方法。通过模拟综合孔径成像中的吉布...针对综合孔径微波辐射计射频干扰(Radio Frequency Interference,RFI)定位与抑制中缺乏科学简便信源估计方法的问题,提出一种基于支持向量机(Support Vector Machine,SVM)与特征值差分的RFI信源估计方法。通过模拟综合孔径成像中的吉布斯振荡现象,分别构建无干扰及不同干扰源数量的相关矩阵;采用矩阵分解获取特征值序列,计算特征值的二阶差分作为特征向量;基于多分类SVM构建干扰源数量识别模型,分别对T型、Y型和U型阵列进行仿真实验;通过对比传统信源估计方法与新方法在不同干扰数量、阵元配置及场景下的均方根误差(Root Mean Squared Error,RMSE),验证了方法的有效性;利用C波段7阵元实测数据(含强干扰及水陆交界弱干扰)验证了方法的有效性。仿真实验表明,所提方法在T/Y/U型阵列中的RMSE较传统方法在不同应用场景下都有提升。实测数据验证显示,训练后的SVM模型对强干扰和弱干扰场景都可以有效识别。所提出的SVM-特征值差分方法能准确识别综合孔径微波辐射计的RFI信源数量,性能明显优于传统方法,为复杂场景下的RFI抑制提供了新的技术途径。展开更多
基金The paper was supported by Jiangsu Education Nature Foundation(06KJD310050,06KJB520022)
文摘The recognition of electroencephalogram (EEG) signals is the key of brain computer interface (BCI). Aimed at the problem that the recognition rate of EEG by using support vector machine (SVM) is low in BCI, based on the assumption that a well-defined physiological signal which also has a smooth form "hides" inside the noisy EEG signal, a Quasi-Newton-SVM recognition method based on Quasi-Newton method and SVM algorithm was presented. Firstly, the EEG signals were preprocessed by Quasi-Newton method and got the signals which were fit for SVM. Secondly, the preprocessed signals were classified by SVM method. The present simulation results indicated the Quasi-Newton-SVM approach improved the recognition rate compared with using SVM method; we also discussed the relationship between the artificial smooth signals and the classification errors.
基金the Research and Innovation Program for College and University Graduate Students in Jiangsu Province (No.CX10B_141Z)the National Natural Science Foundation of China (No.41071273) for support of this project
文摘In order to improve the precision of mining subsidence prediction, a mathematical model using Support Vector Machine （SVM） was established to calculate the displacement factor. The study is based on a comprehensive analysis of factors affecting the displacement factor, such as mechanical properties of the cover rock, the ratio of mining depth to seam thickness, dip angle of the coal seam and the thickness of loose layer. Data of 63 typical observation stations were used as a training and testing sample set. A SVM regression model of the displacement factor and the factors affecting it was established with a kernel function, an insensitive loss factor and a properly selected penalty factor. Given an accurate calculation algorithm for testing and analysis, the results show that an SVM regression model can calcu- late displacement factor precisely and reliable precision can be obtained which meets engineering requirements. The experimental results show that the method to calculation of the displacement factor, based on the SVM method, is feasible. The many factors affecting the displacement factor can be consid- ered with this method. The research provides an efficient and accurate approach for the calculation of displacement in mining subsidence orediction.
文摘针对综合孔径微波辐射计射频干扰(Radio Frequency Interference,RFI)定位与抑制中缺乏科学简便信源估计方法的问题,提出一种基于支持向量机(Support Vector Machine,SVM)与特征值差分的RFI信源估计方法。通过模拟综合孔径成像中的吉布斯振荡现象,分别构建无干扰及不同干扰源数量的相关矩阵;采用矩阵分解获取特征值序列,计算特征值的二阶差分作为特征向量;基于多分类SVM构建干扰源数量识别模型,分别对T型、Y型和U型阵列进行仿真实验;通过对比传统信源估计方法与新方法在不同干扰数量、阵元配置及场景下的均方根误差(Root Mean Squared Error,RMSE),验证了方法的有效性;利用C波段7阵元实测数据(含强干扰及水陆交界弱干扰)验证了方法的有效性。仿真实验表明,所提方法在T/Y/U型阵列中的RMSE较传统方法在不同应用场景下都有提升。实测数据验证显示,训练后的SVM模型对强干扰和弱干扰场景都可以有效识别。所提出的SVM-特征值差分方法能准确识别综合孔径微波辐射计的RFI信源数量,性能明显优于传统方法,为复杂场景下的RFI抑制提供了新的技术途径。
