面向决策树目标路径编码的相空间嵌入维优化  

Optimization of Phase Space Embedding Dimension According to Decision Tree Target Path Coding

作　　者：常大俊[1] 

机构地区：[1]长春建筑学院电气信息学院,长春130607

出　　处：《科技通报》2015年第6期46-48,共3页Bulletin of Science and Technology

摘　　要：提出一种面向决策树目标路径编码的相空间嵌入维计算优化算法。构建云平台环境下的数据交互节点拓扑模型,通过部分链路失效多路径加密方法使得数据聚集具有很高的容错功能,然后采用决策树目标路径编码方案,在给定带宽约束和量化阈值的情况下,对决策树目标路径编码的相空间嵌入维数据进行自适应的量化分解,以实现对决策树目标路径编码的相空间嵌入维的准确估计,降低误码率。仿真结果表明,该算法能准确估计相空间嵌入维,提高估计精度,能有效降低误码率,提高数据动态交互通信的准确性,信号保真度较高。展示了其优越性和较好的应用价值。This paper presents an decision tree target path coding phase space embedding dimension calculation optimiza?tion algorithm. The construction of data interaction topology model of cloud platform environment, through encryption meth?ods part link failure multi path allows data aggregation has the functions of fault tolerance is high, then the decision tree tar?get path coding scheme in a given bandwidth constraints and quantization threshold, decomposition of quantization on deci?sion tree target path coding phase space embedding dimension data adaptive, accurate estimation of the decision tree to achieve the target path coding phase space embedding dimension, reducing the error rate. Simulation results show that the algorithm can accurately estimate the phase space embedding dimension, improve the estimation accuracy, can effectively reduce the error rate, improve the accuracy of dynamic interactive data communication, signal has a higher fidelity. It has demonstrated its superiority and good application value.

关 键 词：嵌入维 编码 决策树 

分 类 号：TN911[电子电信—通信与信息系统]