摘要
针对电厂数据种类多、数据量大,而实际存储空间有限的情况,提出了一种基于改进最小描述长度(MDL)的小波变换数据压缩方法。该方法根据用户关注的信息特征,对各频段的信息加上不同的权值,以改进最小描述长度函数为准则。通过对电厂负荷数据进行压缩和解压实验,不但选择了合适的小波基函数,而且结果显示出改进MDL方法能够获得优于MDL方法的数据压缩比和重构精度。
As a great number of sensors are installed in power plant, plenty of data is produced. It is difficult and unnecessary to save all of the data. A data compression technique is proposed based on wavelet transform and improved minimal description length. This method assigns different weights to the wavelet coefficients in distinct bands according the requirements of the user. Through compressing and decompressing experiments on load data in a power plant, a suitable wavelet base function is selected, and the experiment result shows that the improved MDL method achieves better compression ratio and higher reconstruction precision compared with ordinary MDL method.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2006年第10期1338-1341,共4页
Chinese Journal of Scientific Instrument
关键词
数据压缩
小波变换
最小描述长度
data compression wavelet transform minimum description length
