基于正交多项式的自适应提升格式  

Adaptive Lifting Scheme Based on Orthogonal Polynomials

作　　者：杨汉生[1,2] 孔鲲鹏[1] 许磊[1] 杨成梧[2] 

机构地区：[1]巢湖学院物理与电子科学系,安徽巢湖238000 [2]南京理工大学动力工程学院,江苏南京210094

出　　处：《南京理工大学学报》2009年第3期307-311,共5页Journal of Nanjing University of Science and Technology

基　　金：安徽省高校省级自然科学研究项目(KJ2009B034);巢湖学院自然科学研究项目(XLY-200607)

摘　　要：为确定提升格式的预测系数及更新系数,该文以小波分解的细节信号的平方和为目标函数,采用一组正交多项式确定预测系数,使预测系数能够反映分析数据的特征。根据分析数据的相关性来确定最优预测系数和更新系数,使小波能够较好地适应信号特征的变化。理论和仿真结果表明:正交多项式的自适应提升格式的滤波性能明显优于一般拟合方法的提升格式以及传统小波db5,尤其在低信噪比的情况更为明显。To determine the prediction and the sum of squared detail components as an thogonal polynomial, so that the prediction updated parameters of lifting scheme, this paper takes objective function, and designs predictors based on orparameter can represent the features of the given data. An optimal predictor and an optimal updater are chosen for the transforming sample according to the magnitude of the correlation factors, so the lifting scheme can fit the change of original signal better. The theory and simulation results demonstrate that the filtering performance of adaptive lifting scheme based on orthogonal polynomials is better than that based on general fitting method and traditional wavelet db5, especially with low signal-to-noise ratio.

关 键 词：提升格式 预测控制 自适应算法 正交多项式 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]