- 积分
- 123
- 在线时间
- 308 小时
- 最后登录
- 2013-10-9
- 阅读权限
- 50
- 精华
- 0
 
- UID
- 12413
- 帖子
- 61
- 精华
- 0
- 经验
- 123 点
- 金钱
- 58 ¥
- 注册时间
- 2006-9-16
|
#80楼的翻译
关于混叠产生的机理和
信号处理中的基础定理Nyquist-Shannon采样定理说被采样信号不能被清晰地表示出频率超过采样频率一半的组成信号。这个频率(采样频率的一半)称为Nyquist频率。超过Nyquist频率的频率 N 能够在数字信号中看到,但是它们的频率是不确定的。也就是说,一个频率为 f 的成份频率不能从其它的成份频率 2N-f、2N+f、4N-f 等中区分开来。这个不确定性称为混叠。为了更加完美地处理这个问题,许多模拟信号在转换成数字表示之前使用抗混叠滤波器(通常是低通滤波器)滤除高于Nyquist频率的频率分量。
According to the mechanism of how the Aliasing distortion is generated and the fundamental theorem in Information Theory, the Nyquist-Shannon sampling theorem, the frequency components with the frequency higher than half of the sampling frequency cannot be respresented clearly by the samples. The frequency which is half of the sampling frequency is called Nyquist frequency. The effect of these high frequency components can still be seen in the samples. But we cannot determine the exact frequency of these components from the samples. That is to say, for a frequency component with frequency f, we cannot differentiate it from other frequency components such as 2N-f, 2N+f, 4N-f. This uncertainty or distortion is called Alisasing. In order to solve this problem in a perfect way, analogue signal is treated with anti-aliasing filter (normally it is a low pass one) to filter out the frequency components higher than Nyquist frequency before converting to numeric samples.
事实上现在HDCD已经采用比48k更高的192k采样频率,
In fact, sampling frequency 192kHz, which is much higher than 48kHz one, is used in HDCD.
为什么需要过采样技术呢,
主要原因是后级滤波器.和抗混叠.
这个是一个复杂的过程
大家可以参考相关文章.
Why we need sampling technique with higher sampling frequency? The main reasons behind are related to the making of the output filter and the effect of anti-aliasing. This will be a fairly complicated process, please refer to the related reference paper. |
|