Groups | Search | Server Info | Keyboard shortcuts | Login | Register


Groups > comp.dsp > #33829

Re: How to downsample this signal?

From Randy Yates <randyy@garnerundergroundinc.com>
Newsgroups comp.dsp
Subject Re: How to downsample this signal?
Organization Garner Underground, Inc.
References (1 earlier) <878tm0puch.fsf@digitalsignallabs.com> <ofamgn$bbs$1@node1.news.atman.pl> <ofbbkq$mit$1@dont-email.me> <ofbdtb$vg$1@node1.news.atman.pl> <ofcf3l$6o8$1@dont-email.me>
Date 2017-05-15 11:49 -0400
Message-ID <87h90mqfwv.fsf@garnerundergroundinc.com> (permalink)

Show all headers | View raw


rickman <gnuarm@gmail.com> writes:

> On 5/15/2017 1:21 AM, Piotr Wyderski wrote:
>> rickman wrote:
>>
>>> Are the 248 words for data or also program?
>>
>> Data only, there are dedicated code memories. The control part
>> is totally crazy: there are two separate code memories working
>> in parallel (they take both paths simultaneosly in order to
>> provide instructions with no delay in the case of a branch).
>> But there is no such thing as a program counter, the code
>> memories store instructions at consecutive addresses without
>> any deeper structure. A run of instructions is called a block
>> and is ended with the presence of a jump instruction. There
>> is also a third control memory which stores the information
>> where the i-th block begins and to which j-th state the FSM
>> should go in the case of a branch. In short, hardware basic
>> blocks. The data path is a VLIW with exposed pipelining,
>> which adds fun.
>>
>>> That should be plenty of room for coefficients.
>>
>> But doesn't the FIR require a lot of cells for
>> the past data values?
>
> Like I said, I wrote this a long time ago, so I am not the resource to
> be asking.  I can't picture how it works, but I specifically remember
> NOT needing to store previous data inputs.  I am thinking you store
> outputs which will be fewer because of the decimation.  As the input
> data comes in the output samples can be built up and when one is
> complete it is outputted and replaced by a new one.
>
> But read a proper reference.  If I wasn't busy today I'd dig this up
> for you.

Yes you do need to store N-1 previous inputs (plus 1 current input) for
a length N polyphase FIR decimation by M. Just think first principles:
for any output y[k], you need N inputs to compute it, even if k = n*M.

However, you don't have to input and shift every input as you do for a
normal-rate FIR. 
-- 
Randy Yates, Embedded Firmware Developer
Garner Underground, Inc.
866-260-9040, x3901
http://www.garnerundergroundinc.com

Back to comp.dsp | Previous | NextPrevious in thread | Next in thread | Find similar


Thread

How to downsample this signal? Piotr Wyderski <peter.pan@neverland.mil> - 2017-05-13 09:23 +0200
  Re: How to downsample this signal? Evgeny Filatov <filatov.ev@mipt.ru> - 2017-05-13 15:42 +0300
  Re: How to downsample this signal? bitterlemon40@yahoo.ie - 2017-05-13 07:53 -0700
  Re: How to downsample this signal? rickman <gnuarm@gmail.com> - 2017-05-13 11:48 -0400
  Re: How to downsample this signal? Randy Yates <yates@digitalsignallabs.com> - 2017-05-13 12:58 -0400
    Re: How to downsample this signal? rickman <gnuarm@gmail.com> - 2017-05-13 16:04 -0400
      Re: How to downsample this signal? Randy Yates <yates@digitalsignallabs.com> - 2017-05-14 14:19 -0400
    Re: How to downsample this signal? Piotr Wyderski <peter.pan@neverland.mil> - 2017-05-15 00:42 +0200
      Re: How to downsample this signal? rickman <gnuarm@gmail.com> - 2017-05-15 00:46 -0400
        Re: How to downsample this signal? Piotr Wyderski <peter.pan@neverland.mil> - 2017-05-15 07:21 +0200
          Re: How to downsample this signal? rickman <gnuarm@gmail.com> - 2017-05-15 10:51 -0400
            Re: How to downsample this signal? Randy Yates <randyy@garnerundergroundinc.com> - 2017-05-15 11:49 -0400
              Re: How to downsample this signal? Randy Yates <randyy@garnerundergroundinc.com> - 2017-05-15 17:20 -0400
              Re: How to downsample this signal? rickman <gnuarm@gmail.com> - 2017-05-16 02:59 -0400
                Re: How to downsample this signal? eric.jacobsen@ieee.org - 2017-05-16 15:59 +0000
                Re: How to downsample this signal? rickman <gnuarm@gmail.com> - 2017-05-16 15:36 -0400
                Re: How to downsample this signal? Randy Yates <randyy@garnerundergroundinc.com> - 2017-05-16 13:11 -0400
                Re: How to downsample this signal? eric.jacobsen@ieee.org - 2017-05-16 18:08 +0000
                Re: How to downsample this signal? Randy Yates <randyy@garnerundergroundinc.com> - 2017-05-16 15:35 -0400
                Re: How to downsample this signal? Randy Yates <randyy@garnerundergroundinc.com> - 2017-05-16 15:45 -0400
                Re: How to downsample this signal? Randy Yates <randyy@garnerundergroundinc.com> - 2017-05-16 15:50 -0400
                Re: How to downsample this signal? rickman <gnuarm@gmail.com> - 2017-05-16 21:36 -0400
                Re: How to downsample this signal? rickman <gnuarm@gmail.com> - 2017-05-16 15:35 -0400
                Re: How to downsample this signal? Randy Yates <randyy@garnerundergroundinc.com> - 2017-05-16 15:36 -0400
                Re: How to downsample this signal? Randy Yates <randyy@garnerundergroundinc.com> - 2017-05-16 15:40 -0400
                Re: How to downsample this signal? rickman <gnuarm@gmail.com> - 2017-05-16 22:06 -0400
                Re: How to downsample this signal? Randy Yates <yates@digitalsignallabs.com> - 2017-05-17 01:25 -0400
      Re: How to downsample this signal? Evgeny Filatov <filatov.ev@mipt.ru> - 2017-05-15 14:51 +0300
      Re: How to downsample this signal? Randy Yates <randyy@garnerundergroundinc.com> - 2017-05-15 10:37 -0400
  Re: How to downsample this signal? makolber@yahoo.com - 2017-05-15 06:27 -0700
    Re: How to downsample this signal? Piotr Wyderski <peter.pan@neverland.mil> - 2017-05-15 20:39 +0200

csiph-web