"robert bristow-johnson" <rbj@audioimagination.com> escribi� en el mensaje news:c516eca8-5c69-4695-8cb6-327bfb650850@w34g2000hsg.googlegroups.com... On Dec 3, 8:24 am, "Jaime Andres Aranguren Cardona" <j...@nospam- sanjaac.com> wrote:> <subscribemeh...@yahoo.co.in> escribi� en el > mensajenews:cfb41a00-f558-4317-8a66-1e9ac1bebbe9@a35g2000prf.googlegroups.com... > > > > > Hi, > > > I am trying to implement Symmetric boost and cut digital Equlizer > > using bandpass filters. I found this link which has some help: > >http://72.14.235.104/search?q=cache:UIvZnXeDalgJ:www.freepatentsonlin... > > > In fact I did implement in Matlab. But for -ve gains, output is > > incorrect. (gain is applied after the bandpass filter.). Actually I > > am still not clear how to calculate "k1" as given in Fig.14 of that > > document. > > > I am using bandpass-butterworth-IIR-biquads; where b1=0 (in > > b0*x0+b1*x1+b2*x2-a1*y1-a2*y2) > > > Has anybody implemented this before? Or any other general symmetric > > cut-boost bandpass equalizer design? Help please! > > > Thanks, > > Balaji > >i tried the USPTO site: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=5524022.PN.&OS=PN/5524022&RS=PN/5524022 and i still couldn't view the images (i'm using linux, and Firefox says there is a missing plugin, but i can't figure out what it is and how to get it). --- Hi Robert, Vladimir, Original Poster, all: For those of you that can not access the patent, you can access it on www.freepatentsonline.com You can suscribe for FREE and will be able to get the patent in .pdf format, which you can read everywhere. Here is the link http://www.freepatentsonline.com/5524022.html Robert: I think that it is clear the equivalence on the various design methods for biquad IIR filters, regardless of their function. And you summarized many years of knowledge, BUT that is not the point on this thread; that problem is "solved". However, it works out fine ONLY for single filters, and not for a bank of parallel filters, like one needs to implement in several equalizer configurations, specially Interpolated Constant-Q Equalizers. The point here is: that patent seems to solve the problem of the Interpolated Constant-Q Equalizers, namely the problem of the feedback delay, which is needed to implement the cut filters. Remember: we are talking here of parallel BP filters which behave as a whole, and not single 2nd order IIR biquads. More specifically, the point is that the solution given by the authors seems to have some special trick that seems to be wrong, and the implementation of the original poster (and mine too!) does not seem to work, although the patent says that it works AND they even provide frequency response graphs to prove that. It could be due to an error on the patent itself, that is exactly the point that the original poster (and I) want to disucuss publicily with gurus here. Hopefully several of the audio gurus jump in to discuss that document. -- Jaime Andres Aranguren C. SanJaaC Electronics Soluciones en DSP www.sanjaac.com -- Posted via a free Usenet account from http://www.teranews.com
Symmetric boost and cut Equlizer using bandpass filters
Started by ●December 3, 2007
Reply by ●December 3, 20072007-12-03
Reply by ●December 4, 20072007-12-04
Hi All,>robert bristow-johnson: so, i would recommend designing the biquad boost/cut parametric EQ >using the simplest mathematical expressions and just try to get a good >definition for what you mean by Q or BW.Q = peak frequency of banpass/distance of 3dB points; What "Jaime Andres Aranguren" points is perfectly right. I have implemented the individual bandpass filters using "Robert Bristow-Johnson" method as in "http://www.musicdsp.org/files/Audio-EQ- Cookbook.txt". And there is no confusion about it - "individual" bandpass filters are perfect. As Robert points out, if peking filters are used, problem is solved long back. But as I went for "bandapass in parallel" instead of "peking in series" just to keep the gain outside the biquad loop. Input ------------------------------------------------------> + -------> Output| | |----------- BPF1 -----> Gain1 -----------> + |----------- BPF2 -----> Gain2 -----------> + And here problem comes when gain happens to be negative - problem is when the individual filter output is added with the input. As Robert suggests, I had a thought about tweaking the Q, which again works well. But wondering how do I find the "correct-Q" runtime (fixed-point implementation) - Quite challenging! Balaji.
Reply by ●December 4, 20072007-12-04
<subscribemehere@yahoo.co.in> escribi� en el mensaje news:3bd2ce8b-90fe-4294-bb3d-541f42f20cc6@s19g2000prg.googlegroups.com...> Hi All, > >>robert bristow-johnson: so, i would recommend designing the biquad >>boost/cut parametric EQ >>using the simplest mathematical expressions and just try to get a good >>definition for what you mean by Q or BW. > Q = peak frequency of banpass/distance of 3dB points; > > What "Jaime Andres Aranguren" points is perfectly right. > I have implemented the individual bandpass filters using "Robert > Bristow-Johnson" method as in "http://www.musicdsp.org/files/Audio-EQ- > Cookbook.txt". And there is no confusion about it - "individual" > bandpass filters are perfect. > > As Robert points out, if peking filters are used, problem is solved > long back. But as I went for "bandapass in parallel" instead of > "peking in series" just to keep the gain outside the biquad loop. > > Input ------------------------------------------------------> + ------- >> Output > | | > |----------- BPF1 -----> Gain1 -----------> + > |----------- BPF2 -----> Gain2 -----------> + > > And here problem comes when gain happens to be negative - problem is > when the individual filter output is added with the input. > As Robert suggests, I had a thought about tweaking the Q, which again > works well. But wondering how do I find the "correct-Q" runtime > (fixed-point implementation) - Quite challenging! > > Balaji.Hello guys, Well, the "tweaking" that has to be done to the Q in order to convert one boost filter to one cut filter with symmetric response is to change the Q for Q/k, and substract the filtered signal from the unfiltered one instead of adding, like on the boost case. For a reference, see "Equalization Methods with True Response using Discrete Filters", by Ray Miller, AES Convention Paper 6088, presented at the AES 116th convention in Berlin, Germany, May 8 - 11, 2004. It is mentioned in page 6, and proven in appendix II. But again: it ONLY works for the SINGLE filter case. If you have several BPFs in parallel, such as the case for Interpolated Constant-Q Equalizer which emulates the easy to implement analog case, this approach will NOT work either. It seemed like the Patent (the one originally introduced by Balaji (the OP) was going to solve the problem... But so far it solves nothing, there is some trick which does not match (at least my knowledge). Robert, all... any comment on the approach by the Patent's authors? Regards, -- Jaime Andres Aranguren C. SanJaaC Electronics Soluciones en DSP www.sanjaac.com -- Posted via a free Usenet account from http://www.teranews.com
Reply by ●December 4, 20072007-12-04
On Dec 3, 11:44 pm, "Jaime Andres Aranguren Cardona" <j...@nospam- sanjaac.com> wrote: ...> > Robert, all... any comment on the approach by the Patent's authors? >i'm overwelmed, at the moment. i got your stuff, but need more time. r b-j
Reply by ●December 4, 20072007-12-04
"robert bristow-johnson" <rbj@audioimagination.com> escribi� en el mensaje news:f6f83b14-0aa5-4f1f-a85e-11080c7b1a40@d21g2000prf.googlegroups.com...> On Dec 3, 11:44 pm, "Jaime Andres Aranguren Cardona" <j...@nospam- > sanjaac.com> wrote: > ... >> >> Robert, all... any comment on the approach by the Patent's authors? >> > > i'm overwelmed, at the moment. i got your stuff, but need more time. > > r b-jHello, Robert. Thanks for showing up. Your input on the subject is very instructive, take your time. Regards, -- Jaime Andres Aranguren C. SanJaaC Electronics Soluciones en DSP www.sanjaac.com -- Posted via a free Usenet account from http://www.teranews.com
