bugGNU Octave - Bugs: bug #51310, [octave forge] (signal) firls.m...

 
 

bug #51310: [octave forge] (signal) firls.m modification to include all 4 FIR types, Hilbert transformer, and differentiator

Submitter:  Vlad <archbugaboo>
Submitted:  Sun 25 Jun 2017 07:00:13 PM UTC
   
 
Category:  Octave Package Severity:  1 - Wish
Priority:  3 - Low Item Group:  Feature Request
Status:  Patch Submitted Assigned to:  None
Originator Name:  Open/Closed:  * Open
Release:  * dev Operating System:  * Any
Fixed Release:  None Planned Release:  None
* Mandatory Fields

Add a New Comment Rich Markup
   

Jump to the original submission

Sun 02 May 2021 03:56:18 AM UTC, comment #19: 

I was unable to generate some type II filters using stock Octave 6.2 (my design requires an even number of coefficients), however I replaced firls.m with the file attached to comment 17 and then was able to generate the filters.

Thank you for making this available.

John Wehle <jwehle>
Sat 09 Mar 2019 08:11:57 PM UTC, comment #18: 

Thank you, I will take a look at this latest revision in the next week.

I sincerely appreciate all the work you have put into this function improvement.

I understand and appreciate the learning curve when submitting changes, especially when there are code style guidelines and conventions that you may not be familiar with. We try to document these things for new contributors, but we could use some help improving that. There's no need to apologize, our community should do a better job of documenting and supporting new contributors.

If you are talking about other functions in the signal package, yes, many of these files are 10 or 20 years old now, from before a lot of the Octave coding guidelines have been applied. As a background task I have been working to make these older functions conform to the current coding guidelines, but it's a slow process. If you are looking for functions to compare with, it's far better to look at Octave itself, not the signal package, as a standard.

Mike Miller <mtmiller>
Group Member
Fri 08 Mar 2019 09:39:20 AM UTC, comment #17: 

My apologies for the bad formatting, some of these I only hear about them for the first time:

- used "@end example" and "@end deftypefn" where needed
- used "@cite" in the reference section (used as it is in ellipord.m, the only one that has this in the signal package)
- the changelog is moved outside the doc part (for example FIXME in fir1.m)
- added a @seealso
- corrected differences between "@var{x}" and 'x'
- added "firls: " as a prefix to all printed messages
- used warning() instead of sprintf()
- reformatted the error messages
- relaxed tolerances for the test suite to account for differences between wxMaxima's fpprec>16 and bfloat(), and Octave's
- corrected single quotes to double quotes in string arguments

The errors were less than 1e-15 when I tested them, but that was with another implementation of expint.m, but it's not just that script, it's cumulative, within double precision error. If all these were said in the beginning, they would have been corrected right then (save my mistakes). IMO, that would have been easier than force the users to dig in the whole lot of scripts, trying to hunt down and concoct their own rules of formatting based on what they see, because, and I hope you do know this, too many are not at all as you had me make this, and some even differ between themselves in the same categories.

(file #46464)

Vlad <archbugaboo>
Thu 07 Mar 2019 06:40:40 PM UTC, comment #16: 

We are obviously not communicating very well, so I will stick to specifics. Here are some specific technical feedbacks on the latest revision of this file that I would like to fix before incorporating it into the signal package.




The following are what I referred to as "issues with the doc string":

  • The doc string emits this error message from makeinfo, probably because of a duplicate deftypefn somewhere.



>> help firls
/tmp/octave-help-QpBAq9:238: @bye seen before @end deftypefn


  • Strings options in the doc string should be written as @qcode{"d"} instead of 'd'. Strings in the examples should be written "d".


  • The doc string should not mention technical details like the changelog or that it requires a specific expint.m function.


  • The doc string should use @example markup for code examples, see other Octave functions for comparison.


  • The examples in the doc string should probably use 'b' for the return value to be consistent with the @deftypefn syntax.


  • Please use the @cite markup in the doc string for referencing a paper or book, see the Octave function lscov.m for an example using this.





The following are what I referred to as "the error reporting":

  • Error messages should be formatted according to the Octave conventions, instead of this



>> b = firls (-1, [0, 0.3, 0.4, 1], [1, 1, 0, 0])
error: The order (N) must be positive definite.


should be something like this


error: firls: N must be a positive integer


Every error message should start with "firls:" and the message should start with a lowercase letter, unless it is the name of a parameter. The error message should not end in a full stop/period. As another example, instead of this


>> b = firls (31, [0, 0.3, 0.4, 1], [1, 1, 0])
error: The sizes of the frequency and magnitude vectors must be equal.


should be something like


error: firls: A and F must be vectors of equal size


  • Warning messages should be raised using the warning function, not sprintf. Even better, warning with a warning ID argument:



  warning ("signal:firls-order-incremented",
           "Type II and IV FIRs can't have non-zero amplitude at Nyquist; N has been incremented");





The following are what I referred to as "unit test failures"


>> test firls.m
...
!!!!! test failed
ASSERT errors for:  assert (x,h,1e-15)

  Location  |  Observed  |  Expected  |  Reason
    (11)      -0.0073477   -0.0073477    Abs err 1.2299e-15 exceeds tol 1e-15 by 2e-16
    (13)      0.0011742    0.0011742     Abs err 1.9806e-15 exceeds tol 1e-15 by 1e-15
    (16)      -0.0066228   -0.0066228    Abs err 2.2647e-15 exceeds tol 1e-15 by 1e-15
    (17)      -0.011871    -0.011871     Abs err 1.2646e-15 exceeds tol 1e-15 by 3e-16
    (18)      0.0011213    0.0011213     Abs err 2.1053e-15 exceeds tol 1e-15 by 1e-15
...


Yes, these tests should pass. I think you are saying that the tests fail because the "expected" data came from another source, like Matlab or SciPy, and your algorithm produces different values that you consider more correct, is that right? If you think that was clear, sorry, I have a lot going on and I either forgot or read past that before. I think you think that I care about matching Matlab precisely, but I don't, I just want to see the tests pass. Either the "expected" values should be fudged or the tolerance should be loosened to allow them to pass by default.

Unit tests should also use "d" and "h" instead of 'd' and 'h'.

Mike Miller <mtmiller>
Group Member
Thu 07 Mar 2019 04:08:00 PM UTC, comment #15: 

It's not the categorization, it's the "wish" part, when I repeatedly said the script behaves, by default, like Matlab, that is, that flag that bothers you, is "yes", by default, unless explicitely added as "n" or "no"; it can be completely ignored. It's in the description, too.

And I never said "I must have it", even if I did press it as a matter of educational value, because if it really is not wanted, all you have to do is say "We don't want that flag". I will remove it. I'll wonder why you bothered saying that it's fine to add extra features, as long as the intended behaviour is not affected, but I'll remove it. I currently wonder what is it about that flag that hurts you, it's not like it breaks the Matlab compatibility, it simply adds a new feature, which can be ignored, and is ignored by default.

I only made the GitHub address because it was suggested, in the mails, that it may be easier to work with, but I never did work with it and it's quite foreign to me.

The errors? Last time it was that it's a perfect match except the 1/f^2 weighting, which I showed it to be a wrong answer from Matlab. It was also left, at your discretion, whether you want to emulate that mistake or you want to provide the correct answer. I don't know how Matlab does it, I only suspect (as mentioned), but if I have to make something willingly wrong, on the sole reason that the numbers don't coincide, I refuse. My motor functions will not let me. If they let you and you are fine with copying something wrong, go ahead.

As for the doc string, error reporting, unit test failures, no idea what you mean but the only errors I see in "test firls" are where the errors range between 1e-16..2e-13. I am willing to bet it's the expint() function. I could lessen the tolerances, if that's not frowned upon.

Finish the bug "hopefully" with my help? The reasons that this has been going on for two years are because you insisted I make a pristine script edition when many others don't even come close (but I did it, anyway, and never complained), and because last time (right before you released the last version) you didn't pick the latest upload. I'll upload the latest version here, with relaxed tolerances for the test suite, GitHub apparently won't let me. But, please, let's not point fingers. I've made mistakes, you've made mistakes, the purpose is to see this done.

(file #46454)

Vlad <archbugaboo>
Tue 05 Mar 2019 11:11:50 PM UTC, comment #14: 

Let's try to avoid debating bug tracker categorization details, ok?

I reviewed a batch of open signal package bugs to determine which ones can be done for the next version, and I am planning on finishing this bug, hopefully with your help.

In the process I briefly looked at the latest revision attached here, and I see some things that still need fixing. I see that you have this on GitHub, would it be easier to discuss and iterate there? I find a platform like GitLab or GitHub much easier to work with than Savannah when reviewing and making line-by-line comments.

I would appreciate it if you were open to dropping the 'MatlabCompat' flag entirely (reducing it to 'true' by default). If you are not, I respect your point of view, but honestly I will probably remove it and the associated complicated logic when incorporating this function into the signal package.

To be clear, that is not the reason for this taking so long and not being done yet, there are other issues with the doc string, the error reporting, unit test failures, all of which I would like to fix or see fixed with your help before this is committed.

Are you still willing to help get this ready for incorporating into the package? Here or on GitHub? If GitHub please push the latest changes you shared here as file #43795.

Mike Miller <mtmiller>
Group Member
Sun 03 Mar 2019 08:01:46 AM UTC, comment #13: 

I don't know how you judge the "low priority" part, but this is certainly not a "wish". I clearly showed throughout the mails that:

- it improves upon the existing script
- it has full Matlab compatibility
- it gives the correct output

The only "wish" is the extra added flag, which is disabled by default. This means that, not only it doesn't interfere with the compatibility, but it adds extra features (which you, yourselves, said it's fine as long as it respects compatibility). A "wish" is bug 53497.

But maybe this is your answer to my question: "do you want to emulate a wrong result, or make it correct?".

Vlad <archbugaboo>
Wed 04 Apr 2018 06:57:02 AM UTC, comment #12: 

That was a mistake I corrected with a newer upload, a short while after that. Still, you are right about expint(), so I uploaded now another one in which I explicitly mention it needs a newer version of expint() (from one of the links below: http://hg.savannah.gnu.org/hgweb/octave/file/tip/scripts/specfun/expint.m). I also corrected the ranges to be of form [...], rather than the "lazy" (...). Be sure to get this latest file.

(file #43795)

Vlad <archbugaboo>
Tue 03 Apr 2018 06:56:26 AM UTC, comment #11: 

I get 36 tests passing out of 43 tests included in the current revision of this function, do you see the same?

Five of the test failures are from tolerance errors, is there something still to be fixed in the code or should the tolerances be made less strict?

Two of the test failures are from an error


'E1' undefined near line 351 column 46


Is E1 a replacement for expint? Reason being that the expint in Octave 3.8 through Octave 4.2 is not accurate enough?

If I drop in expint for E1, with Octave 4.3.0+, I get tolerance errors on those two tests as well.

Mike Miller <mtmiller>
Group Member
Wed 21 Mar 2018 05:56:38 AM UTC, comment #10: 


> 1. This function is basically a rewrite, there's no need to keep the Copyright line from the original function there is basically no original code left.

Even if my whole contribution to a project would be one byte, I would be starting a bloody and very messy war if someone decided to brush it off. I don't know if Quentin Spencer behaves like me, but I'd rather not ignore his work, so I would keep that line, if it's alright with you.

> 2. Please don't use narginchk, use the original form that calls the print_usage function.
> 3. The entire body of the function (between "function" and "endfunction") should be indented by 2 spaces.

Alright, fixed, uploaded.

> I haven't read enough here about the reason for the "MatlabCompat" option, but I would rather it be compatible by default and not print a message when it does this compatible operation.

It defaults to Matlab compatibility. If one does not desire to be so, the extra argument 'no', or 'n', can be added, and then it will behave the modified way.

> I gather that you find this compatibility a little distasteful.

I have my reasons, sure, but I'd rather wait until my world domination is complete before I can impose my will unto the rest. My apologies for the wait.

(file #43611)

Vlad <archbugaboo>
Wed 21 Mar 2018 02:28:41 AM UTC, comment #9: 

Thanks for the work you have done so far on this firls replacement. I have glanced at the function but I haven't looked in detail at the code yet.

I have noticed a couple tiny things that could be fixed just by glancing at the code.

1. This function is basically a rewrite, there's no need to keep the Copyright line from the original function there is basically no original code left.

2. Please don't use narginchk, use the original form that calls the print_usage function.

3. The entire body of the function (between "function" and "endfunction") should be indented by 2 spaces.

I haven't read enough here about the reason for the "MatlabCompat" option, but I would rather it be compatible by default and not print a message when it does this compatible operation. I gather that you find this compatibility a little distasteful.

Mike Miller <mtmiller>
Group Member
Sat 01 Jul 2017 11:11:51 AM UTC, comment #8: 


> Ok. Was just thinking that if the output doesn't match MATLAB, someone in the future will probably report incompatibility. Some third party to compare against to justify 'goodness' of answer would be worthwhile


Well, for what it's worth, I'll upload two of the results to match these two (for now):

firls (42, [0, 0.3, 0.4, 1], [0, 0.3, 0, 0]*pi, 'd')
firls (43, [0, 0.3, 0.4, 1], [0, 0.3, 0, 0]*pi, [30, 1], 'd')

The rest of the examples are the same.

I also changed the script to allow a 'MatlabCompat' argument that takes 'no' or 'n', meaning it will not be compatible with Matlab's firls.m. If unspecified, it defaults to compatibility.

(file #41095, file #41096)

Vlad <archbugaboo>
Fri 30 Jun 2017 04:30:51 PM UTC, comment #7: 

Ok. Was just thinking that if the output doesn't match MATLAB, someone in the future will probably report incompatibility. Some third party to compare against to justify 'goodness' of answer would be worthwhile

Also, I'd recommend a more informative name than 'x'. 'prevent_order_increment' or... something? And I haven't looked but help and %!test should be updated to cover the new input if they haven't already.

Nicholas Jankowski <nrjank>
Group Member
Fri 30 Jun 2017 03:50:28 PM UTC, comment #6: 


> I think a posting a couple maxima comparisons would be instructive.


It might not be so, since the code in wxMaxima follows the one in Octave (actually the other way around, but it doesn't matter). It would only matter if wxMaxima had Matlab's implementation, which it doesn't. The only thing it will prove is the accuracy of expint().

I modified the script to allow an extra argument, 'x'. If left alone (no extra argument), the script will increment N for type II and IV FIRs with non-zero amplitude@Nyquist. If added, it will not increment it. The behaviour now matches Matlab's, while being able to be changed as desired. The changed version can also be found at the github address (in the OP), but I'll also upload it.

Vlad <archbugaboo>
Thu 29 Jun 2017 01:43:02 PM UTC, comment #5: 

I think a posting a couple maxima comparisons would be instructive.

Nicholas Jankowski <nrjank>
Group Member
Thu 29 Jun 2017 05:30:19 AM UTC, comment #4: 

The more I think of it, the better suited this seems for the increment N part: adding an extra parameter, say 'x', that, when added, makes the behaviour as originally intended, otherwise (no argument, default) mimicks Matlab. Later this day I'll make an example with what I mean.

About (2), I see that the results seem to get worse as N increases, at least based on the existent examples in the discussion list. For example, according to the results in https://lists.gnu.org/archive/html/help-octave/2017-05/msg00179.html and the followup https://lists.gnu.org/archive/html/help-octave/2017-05/msg00180.html, if one considers "h" to be Octave's output and "g" to be Matlab's, then plotting this should give an idea:

plot(diff(abs(fft(h,1024)(1:512))),"",diff(abs(fft(g,1024)(1:512))))

Zooming in on the passband region it can be seen that, as the order increases, the 1/f^2 weighting is clearly consistent in for "h", but not for "g". Going further with https://lists.gnu.org/archive/html/help-octave/2017-05/msg00138.html (the 2nd example, order 44) and the followup https://lists.gnu.org/archive/html/help-octave/2017-05/msg00139.html, the differences are more visible, and going even further with https://lists.gnu.org/archive/html/help-octave/2017-06/msg00259.html (the "firls_expected_output.txt" attachement) it can be seen that the passband for "g" gets more and more distorted, but the general magnitude is close to the requirements.

This is why I think Matlab must use some very small number for F(1)=0, probably 1e-6 or similar, which, when used for cos(1e-6)/1e-6 results in a six orders of magnitude number, which is then used to build the Q matrix which, in turn, is used for a=Q\b. This must have a negative effect on precision.

The results I had with wxMaxima were numerical confirmations, since their expintegral_e1(x) is accurate and can be used for bigfloats, as well.

Vlad <archbugaboo>
Wed 28 Jun 2017 02:38:05 PM UTC, comment #3: 

Mike? thoughts on either (2) or (3)?

Unless we really want to force compatibility here, my suggestion on (3) is to have the function produce a warning when the non-compatibile event occurs, document it in the help text, and move on.  One alternative would be would be to provide an optional compatibility flag so that it forces compatible behavior. (i.e., if (matlab_compatible) n++ )

See the help text for the currently released version of cov [1] for an example of such documentation.

Then, for (2) you said you were benchmarking against Maxima before. Can you provide a couple examples here showing that what you're referring to on your code produces a more accurate filter? I'm not disputing this, but having it documented here would be enough justification in my opinion to leave it as is and move on.  If someone later has a better idea for the algorithm they could always propose an update. What's more important is to be confident your version doesn't produce any wildly incorrect answers (for odd inputs, corner cases, etc.). I think another note in the help would be useful stating that because of algorithmic differences the output may not exactly match Matlab's but still meets "the definition of 'good' according to ??_".

Then if format/function are good, seems it's about ready to propose rolling it up into a patch for the signal package.

[1] https://octave.sourceforge.io/octave/function/cov.html

Nicholas Jankowski <nrjank>
Group Member
Mon 26 Jun 2017 05:49:19 AM UTC, comment #2: 


> So, apart from the accuracy issues in (2) and the compatibility decision in (3), is there anything else that needs to be done/fixed for it to be completely functional?


No, unless some bugs crept up the code, the script is functional.

> Is there another set of data we can evaluate (2) against? Maybe a symbolic routine for so.e simpler cases to which we can compare yours and matlabs outputs?


I'm afraid the results will always come visibly close, but not quite. I think, at this point, the question is: would it be preferred to mimick a wrong answer, or improve it? (because, based on my tests, I dare say Matlab is wrong, though not by much)

> Would maybe be worth get the package mmaintainers opinion on (3)


That's why I said it's a personal view. However, the results are correct for types I and III, which would be the resulting FIRs after the increment, so the results are the same, it's just that the script doesn't forcefully choose for you. Consider the educational point of view: you're a teacher and want to show your students what a type III or IV lowpass FIR looks like; could you do it with Matlab's script?

At any rate, if this contravenes so much with the intended way the script is meant to be working, then I'll change it. Maybe add an extra parameter?

Vlad <archbugaboo>
Sun 25 Jun 2017 11:23:11 PM UTC, comment #1: 

So, apart from the accuracy issues in (2) and the compatibility decision in (3), is there anything else that needs to be done/fixed for it to be completely functional?

Is there another set of data we can evaluate (2) against? Maybe a symbolic routine for so.e simpler cases to which we can compare yours and matlabs outputs?

Would maybe be worth get the package mmaintainers opinion on (3)

Nicholas Jankowski <nrjank>
Group Member
Sun 25 Jun 2017 07:00:13 PM UTC, original submission:  

Following the discussions in the Octave mailing lists, namely https://lists.gnu.org/archive/html/help-octave/2017-05/msg00136.html and https://lists.gnu.org/archive/html/help-octave/2017-06/msg00058.html, I managed to get this version for firls.m that allows creating all four types of FIR, plus Hilbert transformers and differentiators, the latter allowing either weighted bands or a special 1/f^2 weighting, all according to Matlab's description. The script has some differences between Matlab's, as follows:

1. It uses the more recent advances in Octave's builtin expint() (see http://hg.savannah.gnu.org/hgweb/octave/file/tip/scripts/specfun/expint.m), which corrects the previous function for completely imaginary arguments.

2. The derivation for the least-squares FIR is different than those mentioned in Matlab's online help (https://nl.mathworks.com/help/signal/ref/firls.html#bqih5y1-1 and https://nl.mathworks.com/help/signal/ref/firls.html#bqih5y9-1), and also the one in the original firls.m script (Ivan Selesnick's paper), and it involves sine and cosine integrals, and cos(x)/x terms. Ci(0) and cos(0)/0 result in Inf, and Matlab must use some sort of avoiding (most probably a small number for the frequency at zero) in orcer to circumvent the Inf problem. In my derivation, I considered their values as zero, in the same way type III FIRs require the impulse response be zero at h[M]. As a consequence, after verifying the resposnes side by side, and the derivatives of their FFTs, I conclude that my numbers are closer to reality, thus the results between Octave and Matlab are different, but not necessarily wrong.

3. This is more of a personal view. Types II and IV FIRs have a builtin zero at Nyquist, while types III an additional zero at DC, so having a type II highpass FIR will come with inherent downsides. But this doesn't mean that one cannot build a type II highpass FIR, since the books clearly warn you that making one will not yield what you would expect, but you can create one if you so wish. Because of this, my script does not increment the order behind your back in order to give what the script might think I want to see, instead, it lets you create your type II highpass FIR and enjoy the crippled magnitude, if that is what you wish. In short, it does not think it knows your thoughts better, but considers you know what you're doing (since you're using the scrpt), and you're a grown up that knows how to handle the responsabilities. If not, whatever would you be doing here in the first place?

The original script did its job, but only allowed the creation of type I FIRs.

As for the uploading of the script, I'm in doubt. I uploaded the script, but it can also be found at https://github.com./archbugaboo/Octave_signal.

Vlad <archbugaboo>

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach Files:
   
   
Comment:
   

Attached Files
file #46464:  firls.m added by archbugaboo (41KiB - text/x-objcsrc - 2019.03.08, corrections overall)
file #46454:  firls.m added by archbugaboo (40KiB - text/x-objcsrc - 2019.03.07, relaxed tolerances for test suite.)
file #43795:  firls.m added by archbugaboo (40KiB - text/x-objcsrc - converted the ranges from (...) to [...)
file #43641:  firls.m added by archbugaboo (40KiB - text/x-objcsrc - mistakenly added spaces for functions inside [])
file #43639:  firls.m added by archbugaboo (40KiB - text/x-objcsrc - forgot to replace E1 with expint, added a note saying it needs the latest expint.m)
file #43611:  firls.m added by archbugaboo (40KiB - text/x-objcsrc - replaced narginchk with print_usage, indented body.)
file #41095:  firls.m added by archbugaboo (39KiB - text/x-objcsrc - changed argument 'x' to something hopefully clearer; two comparative results in wxMaxima.)
file #41096:  wxMaxima_test.txt added by archbugaboo (2KiB - text/plain - changed argument 'x' to something hopefully clearer; two comparative results in wxMaxima.)
file #41083:  firls.m added by archbugaboo (37KiB - text/x-objcsrc - modified for extra argument, 'x')
file #41044:  firls.m added by archbugaboo (34KiB - text/x-objcsrc)

 

Depends on the following items: None found

Digest:
   bug dependencies.

 

Carbon-Copy List
  • -email is unavailable- added by jwehle (Posted a comment)
  • -email is unavailable- added by jwe (Updated the item)
  • -email is unavailable- added by nrjank (Posted a comment)
  • -email is unavailable- added by archbugaboo (Submitted the item)
  • -email is unavailable- added by archbugaboo
  •  

    There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.

    Only group members can vote.

     

    Follow 21 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2019-03-08 archbugaboo Attached File- Added firls.m, #46464
    2019-03-07 archbugaboo Attached File- Added firls.m, #46454
    2019-02-26 mtmiller Severity3 - Normal 1 - Wish
        Priority5 - Normal 3 - Low
        Releaseother dev
    2019-02-26 mtmiller Carbon-CopyRemoved 80942 -
    2019-02-26 mtmiller Carbon-CopyRemoved mtmiller -
    2018-04-04 archbugaboo Attached File- Added firls.m, #43795
    2018-04-03 mtmiller Dependencies- bugs #43605 is dependent
    2018-03-23 archbugaboo Attached File- Added firls.m, #43641
    2018-03-23 archbugaboo Attached File- Added firls.m, #43639
    2018-03-21 archbugaboo Attached File- Added firls.m, #43611
    2018-03-21 mtmiller StatusNone Patch Submitted
        Release4.2.1 other
    2017-08-12 jwe Summaryfirls.m modification to include all 4 FIR types, Hilbert transformer, and differentiator [octave forge] (signal) firls.m modification to include all 4 FIR types, Hilbert transformer, and differentiator
    2017-07-01 archbugaboo Attached File- Added firls.m, #41095
        Attached File- Added wxMaxima_test.txt, #41096
    2017-06-30 archbugaboo Attached File- Added firls.m, #41083
    2017-06-25 nrjank Carbon-Copy- Added mtmiller
    2017-06-25 archbugaboo Attached File- Added firls.m, #41044
        Carbon-Copy- Added nrjank

    Back to the top

    Powered by Savane 3.13-758e.
    Corresponding source code