DSPRelated.com

GG (@all4dsp)

Electrical Engineer with interest in DSP for characterizing timing noise in signals.

Yes, it was the atan() versus atan2() phase errors. Thanks so much for pointing it out Cedron. 
Thanks Cedron, There is some noise in the waveform. I can try to get, say, 100 "whole" cycles, but the starting and ending points may not perfectly be aligned in...
Thanks Mark and Steve, My sample size is quite large (can go up to 1e9 pts). I'm currently using a flat-top window for accurate amplitude and it works great. I'm...
I'm trying to model each of the harmonics of a fundamental carrier as a sine wave with an accurate amplitude and phase.I know certain windows can be used to...
I posted the function below. I don't like the sound of: "The magnitude and phase of the resulting function cannot match those of your s-domain function at all frequencies..."...
Something of the form:(b1*s + b0)/(s*s + a1*s + a0), where s=2*pi*f
Thanks Sara, Besides computation time and the possibility of not converging, is there any technical downside to IIR vs FIR as far as the filtered result is concerned?...
Hello, I work with standards that specify first and second order PLL transfer-function (e.g. filtering) using s-domain (Laplace) equations. I know that Matlab...
So, is it that the Nyquist bin (e.g. Fs/2) satisfies the sampling criteria, but only if the measurement is infinitely long? If so, then does that mean the Nyquist...
Thanks jya, Can you explain "Fs/2 is the lowest frequency that is too high to meet all the sampling criteria"? I thought Nyquist's sampling criteria is that the...
Thanks Tim, Could you help me pinpoint where the algorithm breaks down for Nyquist? (the more specific the better) Any lower-frequency tone appears to work just...
Thank you JOS, When I perform an IFFT of the Nyquist tone (including Nyquist bin and its related apron), it reconstructs the original signal fine. When I apply H(s)...
Yes. I understand your perspective. Note that we're analyzing noise in the system. In the real world, it can appear at any frequency and any magnitude. A system...
Thanks JOS, Since H(s) isn't real at the Nyquist bin, I wasn't sure whether the filtered FFT coefficient at that bin should be real or imaginary, before IFFT'ing...
Thanks Tim.The sampling rate is fixed by the system, and has a dominant tone centered at Nyquist. The magnitude of this tone is often very much higher than all...
I'm performing frequency convolution by multiplying FFT coefficients by values computed from a Laplace equation (e.g. H(s)=wc/(s+wc)). The pseudo-code looks like:X=fft(data);for...
Great, thank you Tim!
Can you help me understand how the time-variant nature of a PLL translates to seeing the tone at 0.1*Fs?
Yes, good catch. There's a tone at DC as well. Same question though. Does the PLL see the spurious tone originally at 0.9*Fs in the roll-off region of its LPF (and...
Thanks Tim, Let me dig a little deeper, if I may. The example above is based on a clock (carrier) signal with frequency Fs. This clock is phase modulated with...
For the sake of discussion, let's say I sample a signal at a rate of Fs, and the signal contains a spurious tone at 0.9*Fs. The signal first passes through a low-pass...

Re: how to create filter kernel from Laplace equation?

Reply posted 8 years ago (07/29/2016)
Thanks so much Tim, you've been a tremendous help (really). 

Re: how to create filter kernel from Laplace equation?

Reply posted 8 years ago (07/29/2016)
For a high-pass filter, would the low-pass cutoff frequency be the "bandwidth" of interest, that I'd want to be restrict to be above 1 kHz (using my example above)? For...

Re: how to create filter kernel from Laplace equation?

Reply posted 8 years ago (07/29/2016)
Thanks Tim, There's no DC component. I initially apply a window to the signal, because the signal's start and finish points don't connect, and I need to analyze...

Re: how to create filter kernel from Laplace equation?

Reply posted 8 years ago (07/29/2016)
>There will be issues with numerical stability.Agreed. I've implemented checks to report potential issues to the user based on their input parameters, to ensure stability. The...

Re: how to create filter kernel from Laplace equation?

Reply posted 8 years ago (07/29/2016)
Thanks Tim, I have a software application where the user enters parameters a0, a1, a2, b0, b1, b2, and b3 for the Laplace equation H= (b3*s^3 + b2*s^2 + b1*s +...

Re: how to create filter kernel from Laplace equation?

Reply posted 8 years ago (07/28/2016)
Thanks Tim,Regarding (1), the Laplace equation I wrote is a 1st order LPF, but in reality it could be any 1st, 2nd, or 3rd order, low-pass, high-pass, or band-pass...

Re: how to create filter kernel from Laplace equation?

Reply posted 8 years ago (07/28/2016)
Thank you Tim. Can you help me flush out the procedure as a list of steps? Suppose my time series is 8 points, all real (to keep it simple). 1. Zero-pad the...
Hi, As explained in Chapter 32 of dspguide, "However, it is very important to remember that the (Laplace) values in the s-plane along the y-axis are exactly equal...

Re: Phase locked Loop Bandwidth for second order system

Reply posted 8 years ago (07/12/2016)
Thanks for straighting me out Tim.

Re: Phase locked Loop Bandwidth for second order system

Reply posted 8 years ago (07/12/2016)
The PLL bandwidth can be computed from the closed loop transfer function, where the roll-off passes through -3 dB. The 1/10 rule of thumb people talk about comes...

Use this form to contact all4dsp

Before you can contact a member of the *Related Sites:

  • You must be logged in (register here)
  • You must confirm you email address