>Are you sure you looked hard enough?I was sure... not anymore, so shame on me!>Robert Orban (yes, that Orban) >has posted quite a few examples of digital deemphasis filters on >comp.dsp. Here is one of his posts: > >http://groups.google.ch/group/comp.dsp/msg/4914352a383a2923Thank you a lot for that link! I just took a look at it and it seems to be perfect! However, I have still some problems using MatLab so in one plot it is perfect and in the other plot it isn't, but I will ask my way through ;-)
designing a preemphasis filter
Started by ●October 4, 2007
Reply by ●October 5, 20072007-10-05
Reply by ●October 5, 20072007-10-05
The filter looks nice, but as I took a closer look I could see that the time constant is 75us (for America). Too bad for me - I need 50us for Europe. There is another linke dealing with a 50us preemphasis filter (http://www.dsprelated.com/showmessage/42605/1.php). But this link does not help me much: As far as I know I should use an FIR filter to have a linear phase. Phase skipping could become a problem, because I am doing the FM-modulation in my program which is sensitive to the signal's phase.
Reply by ●October 5, 20072007-10-05
Narax wrote:> The filter looks nice, but as I took a closer look I could see that the > time constant is 75us (for America). Too bad for me - I need 50us for > Europe. > > There is another linke dealing with a 50us preemphasis filter > (http://www.dsprelated.com/showmessage/42605/1.php). But this link does > not help me much: As far as I know I should use an FIR filter to have a > linear phase.You should check that requirement (linear phase response), so that you can say: "I have checked with my supervisor. Even though it makes no sense, I am required to design a 15/50 emphasis filter with linear phase response". "As far as I know" will only get you into trouble. Regards, Andor
Reply by ●October 5, 20072007-10-05
Narax wrote:> The filter looks nice, but as I took a closer look I could see that the > time constant is 75us (for America). Too bad for me - I need 50us for > Europe. > > There is another linke dealing with a 50us preemphasis filter > (http://www.dsprelated.com/showmessage/42605/1.php). But this link does > not help me much: As far as I know I should use an FIR filter to have a > linear phase.You should check that requirement (linear phase response), so that you can say: "I have checked with my supervisor. Even though it makes no sense, I am required to design a 15/50 emphasis filter with linear phase response". "As far as I know" will only get you into trouble. Regards, Andor
Reply by ●October 5, 20072007-10-05
On Oct 5, 2:34 am, Andor <andor.bari...@gmail.com> wrote:> Greg Berchin also posted an example in another thread using his > excellent FDLS filter design method.That was for a deemphasis filter, but FDLS works for preemphasis, too. That thread can be found here: http://groups.google.com/group/comp.dsp/browse_frm/thread/59232b93211c041a FDLS can be found here: http://apollo.ee.columbia.edu/spm/external/tipsandtricks/files/TandT_Jan2007.zip Greg
Reply by ●October 5, 20072007-10-05
Andor wrote:> Narax wrote: >> The filter looks nice, but as I took a closer look I could see that the >> time constant is 75us (for America). Too bad for me - I need 50us for >> Europe. >> >> There is another linke dealing with a 50us preemphasis filter >> (http://www.dsprelated.com/showmessage/42605/1.php). But this link does >> not help me much: As far as I know I should use an FIR filter to have a >> linear phase. > > You should check that requirement (linear phase response), so that you > can say: "I have checked with my supervisor. Even though it makes no > sense, I am required to design a 15/50 emphasis filter with linear > phase response". > > "As far as I know" will only get you into trouble.Amplifying Andor's caveat: The standard deemphasis filter is minimum phase. That will exactly cancel the effect of a minimum-phase preemphasis. Linear-phase preemphasis will mess up the delivered audio. Jerry -- Engineering is the art of making what you want from things you can get. ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
Reply by ●October 5, 20072007-10-05
Jerry Avins <jya@ieee.org> writes:> Andor wrote: >> Narax wrote: >>> The filter looks nice, but as I took a closer look I could see that the >>> time constant is 75us (for America). Too bad for me - I need 50us for >>> Europe. >>> >>> There is another linke dealing with a 50us preemphasis filter >>> (http://www.dsprelated.com/showmessage/42605/1.php). But this link does >>> not help me much: As far as I know I should use an FIR filter to have a >>> linear phase. >> You should check that requirement (linear phase response), so that >> you >> can say: "I have checked with my supervisor. Even though it makes no >> sense, I am required to design a 15/50 emphasis filter with linear >> phase response". >> "As far as I know" will only get you into trouble. > > Amplifying Andor's caveat: The standard deemphasis filter is minimum > phase. That will exactly cancel the effect of a minimum-phase > preemphasis. Linear-phase preemphasis will mess up the delivered audio.Right, and I second Andor's suggestion to investigate Greg Berchin's FDLS filter design algorithm since it will create a filter that matches (with some possible excess phase) in amplitude AND phase. -- % Randy Yates % "So now it's getting late, %% Fuquay-Varina, NC % and those who hesitate %%% 919-577-9882 % got no one..." %%%% <yates@ieee.org> % 'Waterfall', *Face The Music*, ELO http://www.digitalsignallabs.com
Reply by ●October 8, 20072007-10-08
>> Amplifying Andor's caveat: The standard deemphasis filter is minimum >> phase. That will exactly cancel the effect of a minimum-phase >> preemphasis. Linear-phase preemphasis will mess up the deliveredaudio.> >Right, and I second Andor's suggestion to investigate Greg Berchin's >FDLS filter design algorithm since it will create a filter that matches >(with some possible excess phase) in amplitude AND phase.Thank you for correcting me; somehow I had in mind that a deemphasis filter is linear phase as well :-\ That a linear phase filter will mess up the audio is then beyond all question. Thank you for the link again. I'll work it through right now
Reply by ●October 8, 20072007-10-08
> That a linear phase filter will mess up the audio is then beyond all > question. >Not really, it is actually an interesting question.... Cascading a linear phase pre emphasis filter with a min phase de- emphasis will certainly "mess up" the __waveform__, but the prevailing theory is that effects would not be audible. Mark
Reply by ●October 8, 20072007-10-08
>but the prevailing theory is that effects would not be audible. >What I am doing is to simulate FM transmission. On the digital side (my PC) I take a .wav file and simulate the whole transmission incl. noise an rayleigh fading, which I can turn off. The signal is then transmitted via a DAQ board to an actual car radio. Right now I have turned off the noise and rayleigh fading so I can compare my signal without preemphasis and with preemphasis to the original wav file. The file with preemphasis should (nearly) sound like the original file, but it doesn't. Still, I am working with a FIR filter and I am really hoping that this was my mistake. If this should be the case, it will be a "pro" to: "the effect is audible". If not I must have made more mistakes... (Remember: the file without preemphasis sound fine, just a little dull (? I don't know the English word) as I expected)
