I need a digital filter or sequence of digital filters with the following response: Amplitude flat across all frequencies up to the Nyquist (folding) frequency. Phase starts at 0 degrees at low fregs linearly increases with frequency until it is 180 at the Nyquist (folding) frequency. I was thinking about do this with FFT e.g. multiply the spectrum by exp(j*phi*omega). Are there other ways of doing this? Thank you
Need Phase-Lead filter.
Started by ●April 1, 2004
Reply by ●April 1, 20042004-04-01
"HelpRaptor" <HelpRaptor@yahoo.com> wrote in message news:1ae6862a.0404011930.7798baed@posting.google.com...> I need a digital filter or sequence of digital filters with the > following > response: > > Amplitude flat across all frequencies up to the Nyquist (folding) > frequency. > > Phase starts at 0 degrees at low fregs linearly increases with > frequency until > it is 180 at the Nyquist (folding) frequency.You are asking for a filter that can predict the future. It doesn't exist. If you change the requirements in certain ways, like only requiring the response you give up to a small fraction of the Nyquist frequency, then it may be possible to find a reasonable approximation to the spec, but that's probably not what you really want. You would be better off telling us about your application and the practical difficulty.
Reply by ●April 2, 20042004-04-02
HelpRaptor wrote:> I need a digital filter or sequence of digital filters with the > following > response: > > Amplitude flat across all frequencies up to the Nyquist (folding) > frequency. > > Phase starts at 0 degrees at low fregs linearly increases with > frequency until > it is 180 at the Nyquist (folding) frequency. > > I was thinking about do this with FFT e.g. multiply the spectrum by > exp(j*phi*omega). Are there other ways of doing this? > > Thank youFor a phase lag proportional to frequency, it's easy: just delay one sample time. A lead of the same amount requires that the output appear one sample time before it has been input. Very hard! Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●April 2, 20042004-04-02
HelpRaptor wrote:> I need a digital filter or sequence of digital filters with the > following > response: > > Amplitude flat across all frequencies up to the Nyquist (folding) > frequency. > > Phase starts at 0 degrees at low fregs linearly increases with > frequency until > it is 180 at the Nyquist (folding) frequency. > > I was thinking about do this with FFT e.g. multiply the spectrum by > exp(j*phi*omega). Are there other ways of doing this?This is a trick question, think simple. I shouldnt even be telling you that, since it is obviously homework ... but hell, I needed to read it a couple of times before I saw through it so I feel for you. Marco
Reply by ●April 4, 20042004-04-04
Marco Al <m.f.al@student.utwente.nl> wrote in message news:<c4iq40$6hr$1@netlx020.civ.utwente.nl>...> HelpRaptor wrote: > > > I need a digital filter or sequence of digital filters with the > > following > > response: > > > > Amplitude flat across all frequencies up to the Nyquist (folding) > > frequency. > > > > Phase starts at 0 degrees at low fregs linearly increases with > > frequency until > > it is 180 at the Nyquist (folding) frequency. > > > > I was thinking about do this with FFT e.g. multiply the spectrum by > > exp(j*phi*omega). Are there other ways of doing this? > > This is a trick question, think simple. > > I shouldnt even be telling you that, since it is obviously homework ... > but hell, I needed to read it a couple of times before I saw through it > so I feel for you. > > MarcoYour smart remark is not called for. This is not a homework problem, nor am I in school. I posted because I had a need.
Reply by ●April 4, 20042004-04-04
HelpRaptor wrote:> Marco Al <m.f.al@student.utwente.nl> wrote in message news:<c4iq40$6hr$1@netlx020.civ.utwente.nl>... > >>HelpRaptor wrote: >> >> >>>I need a digital filter or sequence of digital filters with the >>>following >>>response: >>> >>>Amplitude flat across all frequencies up to the Nyquist (folding) >>>frequency. >>> >>>Phase starts at 0 degrees at low fregs linearly increases with >>>frequency until >>>it is 180 at the Nyquist (folding) frequency. >>> >>>I was thinking about do this with FFT e.g. multiply the spectrum by >>>exp(j*phi*omega). Are there other ways of doing this? >> >>This is a trick question, think simple. >> >>I shouldnt even be telling you that, since it is obviously homework ... >>but hell, I needed to read it a couple of times before I saw through it >>so I feel for you. >> >>Marco > > > > Your smart remark is not called for. This is not a homework problem, nor am I > in school. I posted because I had a need.It sure sounded like a student question, homework or not. One sample time is 180 degrees at Fs/2, 90 degrees at Fs/4, and so on*. Delay is doable. Constant delay is easy. Prediction is impossible. So what problem are you trying to address?. Jerry _____________________ * In fact, phase shift represented by one sample time = 360�*F/Fs. A sample time is Ts = 1/Fs. -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●April 4, 20042004-04-04
Jerry Avins <jya@ieee.org> writes:> Prediction is impossible.I'd say "exact prediction is impossible". Otherwise, linear predictive coders would have no use, nor would Kalman predictors. Ciao, Peter K. -- Peter J. Kootsookos "I will ignore all ideas for new works [..], the invention of which has reached its limits and for whose improvement I see no further hope." - Julius Frontinus, c. AD 84
Reply by ●April 4, 20042004-04-04
Jerry Avins wrote:> It sure sounded like a student question, homework or not.And that's just the problem with the haughty attitude around here. Who the hell do we think we are making judgements like that? As far as I'm concerned a question is a question. Even if a student wants to use this resource on a problem that has him stumped or is just being lazy that's on him and he'll pay the price. Or not. None of my goddamn business. Bob -- "Things should be described as simply as possible, but no simpler." A. Einstein
Reply by ●April 7, 20042004-04-07
Peter J. Kootsookos wrote:> Jerry Avins <jya@ieee.org> writes: > > >>Prediction is impossible. > > > I'd say "exact prediction is impossible". Otherwise, linear > predictive coders would have no use, nor would Kalman predictors. > > Ciao, >How about, "prediction is at best probabilistic"? Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●April 7, 20042004-04-07
Bob Cain wrote:> Jerry Avins wrote: > >> It sure sounded like a student question, homework or not. > > > And that's just the problem with the haughty attitude around here. Who > the hell do we think we are making judgements like that? As far as I'm > concerned a question is a question. Even if a student wants to use this > resource on a problem that has him stumped or is just being lazy that's > on him and he'll pay the price. Or not. None of my goddamn business. > > > BobPlease note that I made no judgment of any kind. My aim was pointing out that Marco wasn't completely unjustified. We have a collective problem. Most of us agree that handing out homework answers doesn't help the student or the profession in the long run. When a question seems like homework (as did this one with its sophisticated statement and an elementary solution) we can ignore it, ask the OP, take it at face value (as I did), or answer with a sneer (viz. Marco). Most of agree that obfuscation and hazing are inappropriate. The other options all have lesser faults. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
