首页> 外国专利> FUZZY FILTERING METHOD AND FUZZY FILTER

FUZZY FILTERING METHOD AND FUZZY FILTER

机译:模糊过滤方法和模糊过滤器

摘要

PROBLEM TO BE SOLVED: To accurately reproduce even signals accompanied by noise in simple constitution by specifying the front of the signals with the fluctuation of the signals, a distance between samples and a reproduction sample as variables and sectioning the fluctuation by the noise. SOLUTION: The respective minimum values and maximum values S.MIN, S.MAX, ERR.MIN and ERR.MAX of sample signals s(t) and noise sampled err(t) are obtained by a maximum value/minimum value decision unit 63 and stored in a second data storage device 64 and the maximum fluctuation values VX and VN of the signals and the noise are calculated in a difference unit 65. The respective difference levels of the true value of D for which the VX and VN are limit values are combined with respective samples in a fuzzy calculation unit 67 and the maximum and minimum true value of a distance value (k) between the samples is obtained. A fuzzy rule is applied, the weighting value P of the and is obtained in a weighting calculation unit 69. Each D and (k) are combined, sample reproduction output signals (o) are fed back from an output calculation unit 70 to the difference unit 65, a difference with the signals S is calculated and the noise is decided and sectioned.
机译:解决的问题:通过以信号的波动,信号样本与再现样本之间的距离为变量并指定信号的前部并根据噪声来分割波动,以简单的结构准确地再现带有噪声的偶数信号。解决方案:采样信号s(t)和噪声采样err(t)的最小值和最大值S.MIN,S.MAX,ERR.MIN和ERR.MAX由最大值/最小值确定单元63获得并存储在第二数据存储装置64中,并在差分单元65中计算信号和噪声的最大波动值VX和VN。VX和VN为极限值的D的真值的各个差分水平在模糊计算单元67中将各个样本与样本进行组合,并且获得样本之间的距离值(k)的最大和最小真实值。应用模糊规则,在加权计算单元69中获得和的加权值P。将每个D和(k)组合,将样本再现输出信号(o)从输出计算单元70反馈到差值。在单元65中,计算与信号S的差,并确定和分割噪声。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号