摘要
针对网络数据流量激增现象,提出一种5G蜂窝网络高维数据异构特征映射降维方法。构建异构框架提高蜂窝网络灵活性和可扩展性,使用随机矩阵完成对高维数据异构特征提取,并使用相关矩阵将符合预测的奇异值剔除,获得去除噪声后的特征数量,对特征数量进行奇异值分解,得到特征与类的相关性,分析二者之间冗余性完成特征选择;运用半监督正则化方法构建目标函数,通过处理矢量特征获得最小维数,完成特征映射降维。仿真结果表明:所提方法降维识别率较高,大量节省了运行所耗费的时间,相比其它方法具有较高准确性、优越性以及高效性。
In view of the surge of network data traffic, a method for reducing the heterogeneous feature-mapping dimension of high dimensional data in a 5 G cellular network was proposed. In order to improve the flexibility and scalability of the cellular network, a heterogeneous framework was constructed. Then, a random matrix was used to extract the heterogeneous features of high-dimensional data, and the correlation matrix was used to eliminate the singular values to obtain the number of features after noise removal. Moreover, the number of features was decomposed based on singular value, and the correlation between features and classes was obtained. The redundancy was analyzed to complete the feature selection. The semi-supervised regularization method was used to construct the objective function. Finally, the minimum dimension was obtained by processing vector features. Thus, the dimension reduction of feature mapping was completed. Simulation results show that the proposed method has a high recognition rate of dimension reduction, so it can save a lot of running time. Compared with other methods, the proposed method has higher accuracy, superiority, and efficiency.
作者
鞠瞻君
刘亚娟
JU Zhan-jun;LIU Ya-juan(Jilin University,Changchun Jilin 130012,China)
出处
《计算机仿真》
北大核心
2021年第6期335-338,361,共5页
Computer Simulation
基金
吉林省自然科学基金项目(81635571)。
关键词
蜂窝网络
高维数据异构
随机矩阵
半监督正则化
Cellular network
High-dimensional data heterogeneity
Random matrix
Semi-supervised regularization
