如何解释FFT频谱分析仪中的RBW和VBW

嗨，安德鲁，请参阅附件。
另外，在FFT模式下，没有视频信号，因此不存在正常的视频带宽功能。但是，可以通过平均FFT结果在软件中实现VBW功能。



     以下为原文

  Hi Andrew,

pls refer to the attached.

In addition,In FFT mode, there is no video signal, thus there can be no normal video bandwidth function.But  It can be implemented a VBW feature in software by averaging FFT results.

附件

• Performance SA Swept and FFT Analysis.pdf281.9 KB
  

嗨andyluo。在FFT分析仪中，他们使用FFT方法将信号从时域转移到频域，因此不能使用低通滤波器作为VBW滤波器。为了解决这个问题，使用了跟踪平均值。问题是，什么是
RBW / VBW比率和平均计数之间的数学关系？是否有一个函数可以用来根据RBW / VBW比率计算平均数？谢谢。



     以下为原文

  Hi andyluo.In FFT analyzers,they use FFT method to transfer signal from time domain to frequency domain,so it can't use low-pass filter as VBW filter.To solve the problem,trace average is used instead.The question is ,what's the mathematical relationship between RBW/VBW ratio and average count?Is there a function I can use to calculate average number according to RBW/VBW ratio?Thanks.

非常感谢。
本文和应用笔记150肯定为我清除了一些东西。
基于我所读到的，我在确定如何让频谱分析仪报告FFT的奈奎斯特积分时间时遇到了一些麻烦。
例如，假设我想确定信号的大小，该信号在某些RBW中可见为高于频谱分析仪本底噪声的3西格玛偏差。
我将使用以下等式：3 = P_sig /（k_B * T_N）* sqrt（t / RBW）其中t是积分时间。
我注意到t可能与测量时间不同，因为除了为处理添加的“盲时间”之外，足够宽的跨度FFT可能包括几个缝合在一起的BW“块”。
是否有一种简单的方法可以直接访问奈奎斯特积分，还是必须使用信号发生器校准自己？
再次感谢你的帮助。
干杯，安德鲁



     以下为原文

  Thanks very much. This article and Application Note 150 have definitely cleared a few things up for me. Based on what I've read I am having some trouble determining how to get the spectrum analyzer to report the Nyquist integration time for an FFT. For instance suppose I want to determine the size of a signal that would visible as a 3 sigma deviation above the spectrum analyzer noise floor in some RBW. I would use the equation: 3 = P_sig/(k_B*T_N)*sqrt(t/RBW) where t is the integration time. I've noted that t may not be the same thing as the measurement time since a wide enough span FFT may consist of several stitched together BW "chunks" in addition to the "blind time" added for processing. Is there an easy way to access the Nyquist integration directly or do you have to calibrate that your self using a signal generator? Thanks again for all your help. 

Cheers, 

Andrew