On 1/31/2012 10:14 PM, robert bristow-johnson wrote:> On 1/31/12 9:56 PM, Fred Marshall wrote: >> On 1/29/2012 9:39 PM, robert bristow-johnson wrote: >>> On 1/29/12 11:10 PM, HardySpicer wrote: >>>> https://springerlink3.metapress.com/content/048j81635412636v/resource-secured/?target=fulltext.pdf&sid=bif3ix3hbmktapj4ideervuf&sh=www.springerlink.com >>>> >>>> >> >>> >>> can someone tell me what "H∞ Filtering" or "H-infinity Filtering" is? >>> >> >> r b-j, >> >> It alludes to H-infinity filter *design* .. blah, blah, blah. >> >> Mimimax? Chebyshev? Remez objective? >> >> What P-M uses? >> >> i.e. if the error weighting is infinite then the peaks become equal or >> equal under some weighting function. >> >> Or, were you kidding? > > no, i wasn't kidding. i thought it might have something to do with the > L-infinity norm, and from what you say here, it appears it does. i had > previously heard of (or read of) "L-infinity" (as opposed to the L1 or > L2 norm) but not "H-infinity". > > i keep learning new things and forgetting other things. >It seems to come from control systems and perhaps means something slightly different than I said. But, I still see MAX in there.... Tim can surely tell us. Fred
Speaking of charges for papers
Started by ●January 30, 2012
Reply by ●February 1, 20122012-02-01
Reply by ●February 3, 20122012-02-03
On Feb 1, 1:14=C2=A0am, robert bristow-johnson <r...@audioimagination.com> wrote:> On 1/31/12 9:56 PM, Fred Marshall wrote: > > > > > > > > > > > On 1/29/2012 9:39 PM, robert bristow-johnson wrote: > >> On 1/29/12 11:10 PM, HardySpicer wrote: > >>>https://springerlink3.metapress.com/content/048j81635412636v/resource.=..> > >> can someone tell me what "H=E2=88=9E Filtering" or "H-infinity Filteri=ng" is?> > > r b-j, > > > It alludes to H-infinity filter *design* .. blah, blah, blah. > > > Mimimax? Chebyshev? Remez objective? > > > What P-M uses? > > > i.e. if the error weighting is infinite then the peaks become equal or > > equal under some weighting function. > > > Or, were you kidding? > > no, i wasn't kidding. =C2=A0i thought it might have something to do with =the> L-infinity norm, and from what you say here, it appears it does. =C2=A0i =had> previously heard of (or read of) "L-infinity" (as opposed to the L1 or > L2 norm) but not "H-infinity". > > i keep learning new things and forgetting other things. > > -- > > r b-j =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0r...@=audioimagination.com> > "Imagination is more important than knowledge."Robert, You can look into Dan Simon's works. He's publish a book on Kalman and H-infinity. Here's a link to a tutorial like paper he wrote for Embedded Systems Programming, Oct. 2001 "From Infinity to Here" http://academic.csuohio.edu/simond/courses/eec641/hinfinity.pdf Unfortunately some of the math symbols get garbled in the pdf. You can also read it online (the math is there but the formatting is poor - I'm used to Latex formatting :) ) http://eetimes.com/design/embedded/4023844/From-Here-to-Infinity Cheers, David
