home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cs.utexas.edu!sun-barr!sh.wide!wnoc-tyo-news!cs.titech!titccy.cc.titech!ss.titech!cooklev
- From: cooklev@ss.titech.ac.jp (COOKLEV)
- Newsgroups: sci.math
- Subject: Bernstein approximants
- Message-ID: <COOKLEV.92Dec25190257@shigep.ss.titech.ac.jp>
- Date: 25 Dec 92 10:02:57 GMT
- Sender: news@ss.titech.ac.jp
- Distribution: sci
- Organization: Tokyo Institute of Technology, JAPAN
- Lines: 33
- Nntp-Posting-Host: shigep.ss.titech.ac.jp
-
-
-
- What about the the following Bernstein approximant
-
- N
- ---- alpha(i) beta(i)-alpha(i)
- \ alpha(i) Gamma[alpha(i)+1] x (1-x)
- / f(---------) ---------------------------------------------
- ---- beta(i) Gamma[alpha(i)+1] Gamma[beta(i)-alpha(i)+1]
- i=0
-
- Using the property of the Gamma function Gamma (x+1)=x Gamma (x), this
- Bernstein approximant is readily identified as a generalization
- of the Bernstein polynomial for alpha (i) = i and beta (i)=N. The
- sampling pattern of the function is specified by the ratio
- alpha(i) / beta(i).
-
- The above Bernstein approximant has been found useful in some applications
- for simple functions alpha (i) and beta(i).
- However I am not a mathematician and for me it is difficult to evaluate it.
- My questions are
- 1. Is the suggested generalization of the Bernstein polynomials important?
- 2. Is it new?
- 3. What properties does it have?
-
- I would welcome any comments and suggestions.
-
- Merry X'mas and Happy New Year to all,
-
- Todor Cooklev, Dept. Physical Electronics |
- Tokyo Institute of Technolgy |
- 2-12-1 Ookayama, Meguro-ku, Tokyo 152, Japan |
- E-mail: cooklev@ss.titech.ac.jp |
-