Hi Guys, Thanks for having a look at that conference paper and thanks for your thoughts. [-Rick-]
Looking for opinions on a DSP paper
Started by ●December 3, 2014
Reply by ●December 5, 20142014-12-05
Reply by ●December 5, 20142014-12-05
On Friday, December 5, 2014 6:44:36 AM UTC-5, Rick Lyons wrote:> Hi Guys, > Thanks for having a look at that > conference paper and thanks for > your thoughts. > > [-Rick-]Rick,. I've had a look at the paper. The concepts aren't clear. I can understand each sentence, but overall it's not clear what the point is. What does "distinct" mean in "If discrete-time (d.t.) signals and sequences are viewed as distinct ...". The abstract is only 2 sentences and one of those is very long winded. Symbols in the pictures are not explained. A sentence like "Now it's time for pictures" - I mean really. What is probably most disturbing to me is that is a submission for the IEEE Workshop on Signal Processing Education!!
Reply by ●December 5, 20142014-12-05
On Fri, 5 Dec 2014 07:01:18 -0800 (PST), Dave <dspguy2@netscape.net> wrote:>On Friday, December 5, 2014 6:44:36 AM UTC-5, Rick Lyons wrote: >> Hi Guys, >> Thanks for having a look at that >> conference paper and thanks for >> your thoughts. >> >> [-Rick-] > >Rick,. >I've had a look at the paper. The concepts aren't clear. I can understand each sentence, but overall it's not clear what the point is. What does "distinct" mean in "If discrete-time (d.t.) signals and sequences are viewed as >distinct ...". The abstract is only 2 sentences and one of those is very long winded. > >Symbols in the pictures are not explained. A sentence like "Now it's time for pictures" - I mean really. > >What is probably most disturbing to me is that is a submission for the IEEE Workshop on Signal Processing Education!!Ouch. Eric Jacobsen Anchor Hill Communications http://www.anchorhill.com
Reply by ●December 5, 20142014-12-05
On Fri, 05 Dec 2014 03:43:21 -0800, Rick Lyons <R.Lyons@_BOGUS_ieee.org> wrote:>On Thu, 04 Dec 2014 17:04:08 -0600, Tim Wescott ><seemywebsite@myfooter.really> wrote: > >>On Thu, 04 Dec 2014 14:56:08 -0800, Rick Lyons wrote: >> >>> On Wed, 3 Dec 2014 23:32:03 +0000 (UTC), glen herrmannsfeldt >>> <gah@ugcs.caltech.edu> wrote: >>> > [Snipped by Lyons] >>>> >>>>I think he shows that some resampling problems are easier to think of as >>>>trains of delta functions in continuous time, instead of just as samples >>>>it time. >>> >>> Hi Glen, >>> Unless I'm missing something significant, >>> I don't believe there exists any way to think of decimating a >>> discrete-time sequence of sample values by a factor of two as any kind >>> of continuous-time operation. >>> >>> It seems to me that continuous-time operations can only be performed on >>> continuous-time (analog) >>> signals. >> >>There are some aspects of sampling that only seem to make sense if you >>replace the Dirac deltas with functions of area = 1 but finite extent in >>time, and then take things in the limit as that time extent goes to zero. >> >>I'm sure that with careful application of the above, you could shoe-horn >>sub-sampling into that model. >> >>But -- why? > >Hi Tim, > The author's "Decimation" section was on page 4 >of the paper. > >If there's a way to show decimation (sub-sampling) >of a discrete sequence as some kind of 'multiplication >in the time domain' I'd sure like to see it.I've still only barely skimmed the text, but Figure 7 shows his intent for this. The discrete sequence is multiplied in time by a train of continuous sinx/x impulses where the nulls fall on the samples to be excluded and the main lobes align with the samples to be kept. That's not anything new or overly exciting, either.>In any case, it looks like we've beaten this subject >half to death. Thanks for your opinions Tim. > >[-Rick-] >PS. I hope your lawn mower is not an old-fashioned >"push mower". (Remember those?)Eric Jacobsen Anchor Hill Communications http://www.anchorhill.com
Reply by ●December 5, 20142014-12-05
On 12/5/14 11:31 AM, Eric Jacobsen wrote:> On Fri, 05 Dec 2014 03:43:21 -0800, Rick Lyons > <R.Lyons@_BOGUS_ieee.org> wrote: > >> On Thu, 04 Dec 2014 17:04:08 -0600, Tim Wescott >> <seemywebsite@myfooter.really> wrote: >> >>> On Thu, 04 Dec 2014 14:56:08 -0800, Rick Lyons wrote: >>> >>>> On Wed, 3 Dec 2014 23:32:03 +0000 (UTC), glen herrmannsfeldt >>>> <gah@ugcs.caltech.edu> wrote: >>>> >> [Snipped by Lyons] >>>>> >>>>> I think he shows that some resampling problems are easier to think of as >>>>> trains of delta functions in continuous time, instead of just as samples >>>>> it time. >>>> >>>> Hi Glen, >>>> Unless I'm missing something significant, >>>> I don't believe there exists any way to think of decimating a >>>> discrete-time sequence of sample values by a factor of two as any kind >>>> of continuous-time operation. >>>> >>>> It seems to me that continuous-time operations can only be performed on >>>> continuous-time (analog) >>>> signals. >>> >>> There are some aspects of sampling that only seem to make sense if you >>> replace the Dirac deltas with functions of area = 1 but finite extent in >>> time, and then take things in the limit as that time extent goes to zero. >>> >>> I'm sure that with careful application of the above, you could shoe-horn >>> sub-sampling into that model. >>> >>> But -- why? >> >> Hi Tim, >> The author's "Decimation" section was on page 4 >> of the paper. >> >> If there's a way to show decimation (sub-sampling) >> of a discrete sequence as some kind of 'multiplication >> in the time domain' I'd sure like to see it. > > I've still only barely skimmed the text, but Figure 7 shows his intent > for this. The discrete sequence is multiplied in time by a train of > continuous sinx/x impulses where the nulls fall on the samples to be > excluded and the main lobes align with the samples to be kept.uhm, what does this add up to? +inf "impulse train" = SUM{ sin(pi(t-n))/(pi(t-n)) } n=-inf multiplying by this train of sinc() impulses does exactly *what* to a signal?> > That's not anything new or overly exciting, either. >certainly not. Rick, despite this guy's resume, the paper is a whole lotta nothing. -- r b-j rbj@audioimagination.com "Imagination is more important than knowledge."
Reply by ●December 5, 20142014-12-05
robert bristow-johnson <rbj@audioimagination.com> writes:> uhm, what does this add up to? > > +inf > "impulse train" = SUM{ sin(pi(t-n))/(pi(t-n)) } > n=-infIt adds up to one if t is not an integer, otherwise it is undefined.> multiplying by this train of sinc() impulses does exactly *what* to a > signal?Beats me. I haven't read the text, nor am I sure that that will help after following this discussion. Scott -- Scott Hemphill hemphill@alumni.caltech.edu "This isn't flying. This is falling, with style." -- Buzz Lightyear
Reply by ●December 5, 20142014-12-05
On Fri, 05 Dec 2014 14:51:52 -0500, Scott Hemphill wrote:> robert bristow-johnson <rbj@audioimagination.com> writes: > >> uhm, what does this add up to? >> >> +inf >> "impulse train" = SUM{ sin(pi(t-n))/(pi(t-n)) } >> n=-inf > > It adds up to one if t is not an integer, otherwise it is undefined. > >> multiplying by this train of sinc() impulses does exactly *what* to a >> signal? > > Beats me. I haven't read the text, nor am I sure that that will help > after following this discussion.It will build your character, and give you an example of how not to write. -- Tim Wescott Wescott Design Services http://www.wescottdesign.com
Reply by ●December 5, 20142014-12-05
On Fri, 05 Dec 2014 11:46:31 -0500, robert bristow-johnson <rbj@audioimagination.com> wrote:>On 12/5/14 11:31 AM, Eric Jacobsen wrote: >> On Fri, 05 Dec 2014 03:43:21 -0800, Rick Lyons >> <R.Lyons@_BOGUS_ieee.org> wrote: >> >>> On Thu, 04 Dec 2014 17:04:08 -0600, Tim Wescott >>> <seemywebsite@myfooter.really> wrote: >>> >>>> On Thu, 04 Dec 2014 14:56:08 -0800, Rick Lyons wrote: >>>> >>>>> On Wed, 3 Dec 2014 23:32:03 +0000 (UTC), glen herrmannsfeldt >>>>> <gah@ugcs.caltech.edu> wrote: >>>>> >>> [Snipped by Lyons] >>>>>> >>>>>> I think he shows that some resampling problems are easier to think of as >>>>>> trains of delta functions in continuous time, instead of just as samples >>>>>> it time. >>>>> >>>>> Hi Glen, >>>>> Unless I'm missing something significant, >>>>> I don't believe there exists any way to think of decimating a >>>>> discrete-time sequence of sample values by a factor of two as any kind >>>>> of continuous-time operation. >>>>> >>>>> It seems to me that continuous-time operations can only be performed on >>>>> continuous-time (analog) >>>>> signals. >>>> >>>> There are some aspects of sampling that only seem to make sense if you >>>> replace the Dirac deltas with functions of area = 1 but finite extent in >>>> time, and then take things in the limit as that time extent goes to zero. >>>> >>>> I'm sure that with careful application of the above, you could shoe-horn >>>> sub-sampling into that model. >>>> >>>> But -- why? >>> >>> Hi Tim, >>> The author's "Decimation" section was on page 4 >>> of the paper. >>> >>> If there's a way to show decimation (sub-sampling) >>> of a discrete sequence as some kind of 'multiplication >>> in the time domain' I'd sure like to see it. >> >> I've still only barely skimmed the text, but Figure 7 shows his intent >> for this. The discrete sequence is multiplied in time by a train of >> continuous sinx/x impulses where the nulls fall on the samples to be >> excluded and the main lobes align with the samples to be kept. > >uhm, what does this add up to? > > +inf > "impulse train" = SUM{ sin(pi(t-n))/(pi(t-n)) } > n=-infIt's one at the discrete samples to be retained after decimation and zero at the discrete samples to be discarded. In between the discrete samples is zero so what's in the continuous sequence doesn't matter there.>multiplying by this train of sinc() impulses does exactly *what* to a >signal?He's using it to illustrate a conceptual approach to decimation. I've not sorted out why he thinks it's "better" yet, especially since this multiplication in the time domain results in convolution in the frequency domain which seems to me to be a fairly suboptimal way to explain decimation. I've been reading more of this paper and it's difficult to slog through. I haven't found any glaring errors, it's just very, very tedious and unclear.>> >> That's not anything new or overly exciting, either. >> > >certainly not. > >Rick, despite this guy's resume, the paper is a whole lotta nothing. > > > >-- > >r b-j rbj@audioimagination.com > >"Imagination is more important than knowledge." > >Eric Jacobsen Anchor Hill Communications http://www.anchorhill.com
Reply by ●December 5, 20142014-12-05
On Fri, 05 Dec 2014 07:01:18 -0800, Dave wrote:> On Friday, December 5, 2014 6:44:36 AM UTC-5, Rick Lyons wrote: >> Hi Guys, >> Thanks for having a look at that >> conference paper and thanks for your thoughts. >> >> [-Rick-] > > Rick,. > I've had a look at the paper. The concepts aren't clear. I can > understand each sentence, but overall it's not clear what the point is. > What does "distinct" mean in "If discrete-time (d.t.) signals and > sequences are viewed as distinct ...". The abstract is only 2 sentences > and one of those is very long winded. > > Symbols in the pictures are not explained. A sentence like "Now it's > time for pictures" - I mean really. > > What is probably most disturbing to me is that is a submission for the > IEEE Workshop on Signal Processing Education!!Do we know if it was accepted? If so, that would be more disturbing yet. -- www.wescottdesign.com
Reply by ●December 5, 20142014-12-05
On Fri, 05 Dec 2014 20:59:58 +0000, Eric Jacobsen wrote:> On Fri, 05 Dec 2014 11:46:31 -0500, robert bristow-johnson > <rbj@audioimagination.com> wrote: > >>On 12/5/14 11:31 AM, Eric Jacobsen wrote: >>> On Fri, 05 Dec 2014 03:43:21 -0800, Rick Lyons >>> <R.Lyons@_BOGUS_ieee.org> wrote: >>> >>>> On Thu, 04 Dec 2014 17:04:08 -0600, Tim Wescott >>>> <seemywebsite@myfooter.really> wrote: >>>> >>>>> On Thu, 04 Dec 2014 14:56:08 -0800, Rick Lyons wrote: >>>>> >>>>>> On Wed, 3 Dec 2014 23:32:03 +0000 (UTC), glen herrmannsfeldt >>>>>> <gah@ugcs.caltech.edu> wrote: >>>>>> >>>> [Snipped by Lyons] >>>>>>> >>>>>>> I think he shows that some resampling problems are easier to think >>>>>>> of as trains of delta functions in continuous time, instead of >>>>>>> just as samples it time. >>>>>> >>>>>> Hi Glen, >>>>>> Unless I'm missing something significant, >>>>>> I don't believe there exists any way to think of decimating a >>>>>> discrete-time sequence of sample values by a factor of two as any >>>>>> kind of continuous-time operation. >>>>>> >>>>>> It seems to me that continuous-time operations can only be >>>>>> performed on continuous-time (analog) >>>>>> signals. >>>>> >>>>> There are some aspects of sampling that only seem to make sense if >>>>> you replace the Dirac deltas with functions of area = 1 but finite >>>>> extent in time, and then take things in the limit as that time >>>>> extent goes to zero. >>>>> >>>>> I'm sure that with careful application of the above, you could >>>>> shoe-horn sub-sampling into that model. >>>>> >>>>> But -- why? >>>> >>>> Hi Tim, >>>> The author's "Decimation" section was on page 4 >>>> of the paper. >>>> >>>> If there's a way to show decimation (sub-sampling) >>>> of a discrete sequence as some kind of 'multiplication in the time >>>> domain' I'd sure like to see it. >>> >>> I've still only barely skimmed the text, but Figure 7 shows his intent >>> for this. The discrete sequence is multiplied in time by a train of >>> continuous sinx/x impulses where the nulls fall on the samples to be >>> excluded and the main lobes align with the samples to be kept. >> >>uhm, what does this add up to? >> >> +inf >> "impulse train" = SUM{ sin(pi(t-n))/(pi(t-n)) } >> n=-inf > > It's one at the discrete samples to be retained after decimation and > zero at the discrete samples to be discarded. In between the discrete > samples is zero so what's in the continuous sequence doesn't matter > there. > >>multiplying by this train of sinc() impulses does exactly *what* to a >>signal? > > He's using it to illustrate a conceptual approach to decimation. I've > not sorted out why he thinks it's "better" yet, especially since this > multiplication in the time domain results in convolution in the > frequency domain which seems to me to be a fairly suboptimal way to > explain decimation. > > I've been reading more of this paper and it's difficult to slog through. > I haven't found any glaring errors, it's just very, very tedious and > unclear.That was my overall impression, given the amount of it that I really read closely (which wasn't a lot). I think Rick needs to mow your lawn, too. -- Tim Wescott Wescott Design Services http://www.wescottdesign.com
