摘要
针对传统小波滤波器结构和系数进行改进,提出了一种高效的整数算法小波变换的构造方案。在此基础上,针对电力信号采集数据自身的特点,对小波变换后的低、高频数据分别进行二次压缩,低频数据采用RayPeriod压缩算法,高频数据采用改进的阈值压缩算法。压缩实验证实,该算法能在均方误差优于2.0×10-4的情况下,获得6.2%的压缩比,并且速度比传统小波变换的压缩算法提高了10倍以上,具有实用价值。
This paper proposes an efficient algorithm to build integer arithmetic wavelet transform is by improving the traditional wavelet filter structure and coefficients. Based on the above integer arithmetic, the low-frequency and high-frequency coefficients of wavelet transformation are further compressed respectively according to the data features of electric power signal. The low-frequency coefficients are compressed with the Ray-Period compression algorithm, while the high-frequency coefficients with the improved threshold value. The compression experiments 2.0×10^-4 prove that when the mean square error is better than , the data compression ratio can be as small as 6.2%,and the computing time is only tenth of that of the algorithm based on the traditional wavelet transform, which verifies that the proposed algorithm accommodates the requirement in engineering applications.
出处
《电测与仪表》
北大核心
2013年第8期20-25,共6页
Electrical Measurement & Instrumentation
基金
黑龙江省电力有限公司重点资助项目(黑电科信2011[437])
关键词
压缩与重构
小波变换
整数算法
Ray—Period
阈值
compression and reconstruction, wavelet transform, integer arithmetic, Ray-Period, threshold value
