Changeset 306 for trunk/doc/pslib/psLibADD.tex
- Timestamp:
- Mar 25, 2004, 3:28:34 PM (22 years ago)
- File:
-
- 1 edited
-
trunk/doc/pslib/psLibADD.tex (modified) (1 diff)
Legend:
- Unmodified
- Added
- Removed
-
trunk/doc/pslib/psLibADD.tex
r303 r306 434 434 quadratic which returns the 50\% value. 435 435 436 437 \section{Polynomials} 438 439 We will employ Chebyshev polynomials (NR \S 5.8) to approximate functions: 440 \begin{equation} 441 f(x) = \Sum_{i=0}^{n} c_i T_i(x) 442 \end{equation} 443 These have some desirable features: 444 \begin{itemize} 445 \item They are bounded on $-1 < x < 1$, with the maxima and minima 446 over this range being 1 and -1, respectively; 447 \item Truncation of the higher-order terms leaves one with the most accurate 448 lower-order polynomial representation of the desired function. 449 \end{itemize} 450 451 The first few Chebyshev polynomials are: 452 \begin{equation} 453 T_0(x) = 1 454 455 T_1(x) = x 456 457 T_2(x) = 2x^2 - 1 458 459 T_3(x) = 4x^3 - 3x 460 461 T_4(x) = 8x^4 - 8x^2 + 1 462 \end{equation} 463 Chebyshev polynomials follow the recurrence relation: 464 \begin{equation} 465 T_{n+1} = 2xT_n - T_{n-1} 466 \end{equation} 467 468 Practically, Chebyshev polynomials should be evaluated using Clenshaw's recurrence 469 formula (NR \S 5.5): 470 \begin{equation} 471 d_j = 2xd_{j+1} - d_{j+2} + c_j 472 473 f(x) = x*d_1 - d_2 + 1/2 c_0 474 \end{equation} 475 476 It shall be the responsibility of the user to convert the domain into the range 477 $-1 < x < 1$. 478 479 \subsection{Multi-dimensional polynomials} 480 481 Multi-dimensional polynomials shall be composed of multiplications of 1D polynomials. 482 483 436 484 \end{document}
Note:
See TracChangeset
for help on using the changeset viewer.
