Hello, Currently I am searching about DC removal filter, I could understand that we can use an FIR high pass filter to remove the DC. Can any one please help me to get a document which gives the specifications required to design a FIR high pass for DC removal. It is very urgent. Please help me in this regard. Thanks, Neelufar.
Regarding DC removal filter.
Started by ●July 31, 2007
Reply by ●July 31, 20072007-07-31
neelufar wrote:> Hello, > > Currently I am searching about DC removal filter, I could understand that > we can use an FIR high pass filter to remove the DC. Can any one please > help me to get a document which gives the specifications required to > design a FIR high pass for DC removal. It is very urgent. Please help me > in this regard.The FIR with impulse response h = [1 -1] removes DC.
Reply by ●July 31, 20072007-07-31
neelufar.2005 schrieb:> Currently I am searching about DC removal filter, I could understand that > we can use an FIR high pass filter to remove the DC. Can any one please > help me to get a document which gives the specifications required to > design a FIR high pass for DC removal. It is very urgent. Please help me > in this regard.www.dspguide.com Depending on your needs a moving average filter with spectral inversion may be sufficient too. Marcel
Reply by ●July 31, 20072007-07-31
Andor wrote:> neelufar wrote: >> Hello, >> >> Currently I am searching about DC removal filter, I could understand that >> we can use an FIR high pass filter to remove the DC. Can any one please >> help me to get a document which gives the specifications required to >> design a FIR high pass for DC removal. It is very urgent. Please help me >> in this regard. > > The FIR with impulse response h = [1 -1] removes DC.[1 -1] also differentiates the signal, which I'm sure is not wanted. An FIR will be costly because the number of taps is proportional to the lowest significant period and the sharpness of the transition. Frequencies close to DC need lots of taps. Only a need for strict phase linearity down to near the cutoff justify an FIR. Any high-pass filter removes DC. The simplest is IIR. There's an informative thread at http://tinyurl.com/32gav8 http://tinyurl.com/32o2ng is a good tutorial, and there's a lead in http://tinyurl.com/387a37 Jerry -- Engineering is the art of making what you want from things you can get. ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
Reply by ●August 1, 20072007-08-01
"Jerry Avins" <jya@ieee.org> wrote in message news:v92dnUlFLOiIyTLbnZ2dnUVZ_qqgnZ2d@rcn.net...> Andor wrote: >> neelufar wrote: >>> Hello, >>> >>> Currently I am searching about DC removal filter, I could understand >>> that >>> we can use an FIR high pass filter to remove the DC. Can any one please >>> help me to get a document which gives the specifications required to >>> design a FIR high pass for DC removal. It is very urgent. Please help me >>> in this regard. >> >> The FIR with impulse response h = [1 -1] removes DC. > > [1 -1] also differentiates the signal, which I'm sure is not wanted. > > An FIR will be costly because the number of taps is proportional to the > lowest significant period and the sharpness of the transition. Frequencies > close to DC need lots of taps. Only a need for strict phase linearity down > to near the cutoff justify an FIR. > > Any high-pass filter removes DC. The simplest is IIR. > There's an informative thread at http://tinyurl.com/32gav8 > http://tinyurl.com/32o2ng is a good tutorial, and there's a lead in > http://tinyurl.com/387a37 > > Jerry > -- > Engineering is the art of making what you want from things you can get. > �����������������������������������������������������������������������Could the filtering be done in the analog domain before the A/D converter? Best wishes, --Phil Martel
Reply by ●August 1, 20072007-08-01
Philip Martel wrote:> "Jerry Avins" <jya@ieee.org> wrote in message > news:v92dnUlFLOiIyTLbnZ2dnUVZ_qqgnZ2d@rcn.net... >> Andor wrote: >>> neelufar wrote: >>>> Hello, >>>> >>>> Currently I am searching about DC removal filter, I could understand >>>> that >>>> we can use an FIR high pass filter to remove the DC. Can any one please >>>> help me to get a document which gives the specifications required to >>>> design a FIR high pass for DC removal. It is very urgent. Please help me >>>> in this regard. >>> The FIR with impulse response h = [1 -1] removes DC. >> [1 -1] also differentiates the signal, which I'm sure is not wanted. >> >> An FIR will be costly because the number of taps is proportional to the >> lowest significant period and the sharpness of the transition. Frequencies >> close to DC need lots of taps. Only a need for strict phase linearity down >> to near the cutoff justify an FIR. >> >> Any high-pass filter removes DC. The simplest is IIR. >> There's an informative thread at http://tinyurl.com/32gav8 >> http://tinyurl.com/32o2ng is a good tutorial, and there's a lead in >> http://tinyurl.com/387a37 >> >> Jerry >> -- >> Engineering is the art of making what you want from things you can get. >> ����������������������������������������������������������������������� > Could the filtering be done in the analog domain before the A/D converter? >Do you know of any ADCs with zero DC offset? It is usually a good idea to remove most of the DC in the analogue domain, so it doesn't eat up the range of the ADC. However, a final digital DC elimination is typically needed, too. Steve
Reply by ●August 2, 20072007-08-02
On 31 Jul., 17:24, Jerry Avins <j...@ieee.org> wrote:> Andor wrote: > > neelufar wrote: > >> Hello, > > >> Currently I am searching about DC removal filter, I could understand that > >> we can use an FIR high pass filter to remove the DC. Can any one please > >> help me to get a document which gives the specifications required to > >> design a FIR high pass for DC removal. It is very urgent. Please help me > >> in this regard. > > > The FIR with impulse response h = [1 -1] removes DC. > > [1 -1] also differentiates the signal, which I'm sure is not wanted.It was a provocation aimed at the OP to extract more information about the filter requirements. BTW, what makes you sure?
Reply by ●August 2, 20072007-08-02
Philip Martel wrote:> "Jerry Avins" <jya@ieee.org> wrote in message > news:v92dnUlFLOiIyTLbnZ2dnUVZ_qqgnZ2d@rcn.net... >> Andor wrote: >>> neelufar wrote: >>>> Hello, >>>> >>>> Currently I am searching about DC removal filter, I could understand >>>> that >>>> we can use an FIR high pass filter to remove the DC. Can any one please >>>> help me to get a document which gives the specifications required to >>>> design a FIR high pass for DC removal. It is very urgent. Please help me >>>> in this regard. >>> The FIR with impulse response h = [1 -1] removes DC. >> [1 -1] also differentiates the signal, which I'm sure is not wanted. >> >> An FIR will be costly because the number of taps is proportional to the >> lowest significant period and the sharpness of the transition. Frequencies >> close to DC need lots of taps. Only a need for strict phase linearity down >> to near the cutoff justify an FIR. >> >> Any high-pass filter removes DC. The simplest is IIR. >> There's an informative thread at http://tinyurl.com/32gav8 >> http://tinyurl.com/32o2ng is a good tutorial, and there's a lead in >> http://tinyurl.com/387a37...> Could the filtering be done in the analog domain before the A/D converter?Not if there's any offset in the converter. At least some bipolar converters are offset unipolar devices with the MSB inverted to convert offset binary to two's complement. If the offset isn't perfect, that's equivalent to DC injection. There other way for the fault to arise in the converter. After all, the signal is only digital after the converter has digitized it. The converter's front end is analog. Jerry -- Engineering is the art of making what you want from things you can get. ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
Reply by ●August 2, 20072007-08-02
Andor wrote:> On 31 Jul., 17:24, Jerry Avins <j...@ieee.org> wrote: >> Andor wrote: >>> neelufar wrote: >>>> Hello, >>>> Currently I am searching about DC removal filter, I could understand that >>>> we can use an FIR high pass filter to remove the DC. Can any one please >>>> help me to get a document which gives the specifications required to >>>> design a FIR high pass for DC removal. It is very urgent. Please help me >>>> in this regard. >>> The FIR with impulse response h = [1 -1] removes DC. >> [1 -1] also differentiates the signal, which I'm sure is not wanted. > > It was a provocation aimed at the OP to extract more information about > the filter requirements. > > BTW, what makes you sure?Experience. :-) Jerry -- Engineering is the art of making what you want from things you can get. ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
Reply by ●August 3, 20072007-08-03
On Tue, 31 Jul 2007 14:19:38 +0200, =?ISO-8859-15?Q?Marcel_M=FCller?= <news.5.maazl@spamgourmet.com> wrote:>neelufar.2005 schrieb: >> Currently I am searching about DC removal filter, I could understand that >> we can use an FIR high pass filter to remove the DC. Can any one please >> help me to get a document which gives the specifications required to >> design a FIR high pass for DC removal. It is very urgent. Please help me >> in this regard. > >www.dspguide.com > >Depending on your needs a moving average filter with spectral inversion >may be sufficient too. > >MarcelHi Marcel, I don't understand your DC-removal scheme. As far as I know, there's no low-computation linear-phase way to remove (attenuate) DC. If you have some sort of "neat" way to do so, we'd sure like to hear more about your method. Thanks, [-Rick-]
