\documentstyle[11pt,reduce]{article}
\title{{\tt TRIGSIMP}\\
A REDUCE Package for the Simplification and Factorization of Trigonometric
and Hyperbolic Functions}
\date{}
\author{Wolfram Koepf\\
Andreas Bernig\\
Herbert Melenk\\
ZIB Berlin \\
email: {\tt Koepf@ZIB-Berlin.de}}
\begin{document}
\maketitle
\section{Introduction}
The REDUCE package TRIGSIMP is a useful tool for all kinds of trigonometric and
hyperbolic simplification and factorization. There are three
procedures included in TRIGSIMP: trigsimp, trigfactorize and triggcd.
The first is for finding simplifications of trigonometric or
hyperbolic expressions with many options, the second for factorizing
them and the third
for finding the greatest common divisor of two trigonometric or
hyperbolic polynomials.
To start the package it must be loaded by:
{\small
\begin{verbatim}
1: load trigsimp;
\end{verbatim}
}\noindent
\section{\REDUCE{} operator {\tt trigsimp}}
As there is no normal form for trigonometric and hyperbolic functions, the same
function can convert in many different directions, e.g.
$\sin(2x) \leftrightarrow 2\sin(x)\cos(x)$.
The user has the possibility to give several parameters to the
procedure {\tt trigsimp} in order to influence the direction of transformations.
The decision whether a rational expression in trigonometric
and hyperbolic functions vanishes or not is possible.
To simplify a function {\tt f}, one uses {\tt trigsimp(f[,options])}. Example:
{\small
\begin{verbatim}
2: trigsimp(sin(x)^2+cos(x)^2);
1
\end{verbatim}
}\noindent
Possible options are (* denotes the default):
\begin{enumerate}
\item {\tt sin} (*) or {\tt cos}
\item {\tt sinh} (*) or {\tt cosh}
\item {\tt expand} (*) or {\tt combine} or {\tt compact}
\item {\tt hyp} or {\tt trig} or {\tt expon}
\item {\tt keepalltrig}
\end{enumerate}
From each group one can use at most one option, otherwise an error
message will occur. The first group fixes the preference used while
transforming a trigonometric expression:
{\small
\begin{verbatim}
3: trigsimp(sin(x)^2);
2
sin(x)
4: trigsimp(sin(x)^2,cos);
2
- cos(x) + 1
\end{verbatim}
}\noindent
The second group is the equivalent for the hyperbolic functions.
The third group determines the type of transformations. With
the default {\tt expand}, an expression is written in a form only using
single arguments and no sums of arguments:
{\small
\begin{verbatim}
5: trigsimp(sin(2x+y));
2
2*cos(x)*cos(y)*sin(x) - 2*sin(x) *sin(y) + sin(y)
\end{verbatim}
}\noindent
With {\tt combine}, products of trigonometric functions are transformed to
trigonometric functions involving sums of arguments:
{\small
\begin{verbatim}
6: trigsimp(sin(x)*cos(y),combine);
sin(x - y) + sin(x + y)
-------------------------
2
\end{verbatim}
}\noindent
With {\tt compact}, the REDUCE operator {\tt compact} \cite{hearns}
is applied to {\tt f}.
This leads often to a simple form, but in contrast to {\tt expand} one
doesn't get a normal form. Example for {\tt compact}:
{\small
\begin{verbatim}
7: trigsimp((1-sin(x)**2)**20*(1-cos(x)**2)**20,compact);
40 40
cos(x) *sin(x)
\end{verbatim}
}\noindent
With the fourth group each expression is transformed to a
trigonometric, hyperbolic or exponential form:
{\small
\begin{verbatim}
8: trigsimp(sin(x),hyp);
- sinh(i*x)*i
9: trigsimp(sinh(x),expon);
2*x
e - 1
----------
x
2*e
10: trigsimp(e^x,trig);
x x
cos(---) + sin(---)*i
i i
\end{verbatim}
}\noindent
Usually, {\tt tan}, {\tt cot}, {\tt sec}, {\tt csc} are expressed in terms of
{\tt sin} and {\tt cos}. It can
be sometimes useful to avoid this, which is handled by the option
{\tt keepalltrig}:
{\small
\begin{verbatim}
11: trigsimp(tan(x+y),keepalltrig);
- (tan(x) + tan(y))
----------------------
tan(x)*tan(y) - 1
\end{verbatim}
}\noindent
It is possible to use the options of different groups simultaneously:
{\small
\begin{verbatim}
12: trigsimp(sin(x)**4,cos,combine);
cos(4*x) - 4*cos(2*x) + 3
---------------------------
8
\end{verbatim}
}\noindent
Sometimes, it is necessary to handle an expression in different steps:
{\small
\begin{verbatim}
13: trigsimp((sinh(x)+cosh(x))**n+(cosh(x)-sinh(x))**n,expon);
2*n*x
e + 1
------------
n*x
e
14: trigsimp(ws,hyp);
2*cosh(n*x)
15: trigsimp((cosh(a*n)*sinh(a)*sinh(p)+cosh(a)*sinh(a*n)*sinh(p)+
sinh(a - p)*sinh(a*n))/sinh(a));
cosh(a*n)*sinh(p) + cosh(p)*sinh(a*n)
16: trigsimp(ws,combine);
sinh(a*n + p)
\end{verbatim}
}\noindent
\section{\REDUCE{} operator {\tt trigfactorize}}
With {\tt trigfactorize(p,x)} one can factorize the trigonometric or
hyperbolic polynomial {\tt p} with respect to the argument x. Example:
{\small
\begin{verbatim}
17: trigfactorize(sin(x),x/2);
x x
{2,cos(---),sin(---)}
2 2
\end{verbatim}
}\noindent
If the polynomial is not coordinated or balanced \cite{art},
the output will equal the input.
In this case, changing the value for x can help to find a factorization:
{\small
\begin{verbatim}
18: trigfactorize(1+cos(x),x);
{cos(x) + 1}
19: trigfactorize(1+cos(x),x/2);
x x
{2,cos(---),cos(---)}
2 2
\end{verbatim}
}\noindent
The polynomial can consist of both trigonometric and hyperbolic functions:
{\small
\begin{verbatim}
20: trigfactorize(sin(2x)*sinh(2x),x);
{4, cos(x), sin(x), cosh(x), sinh(x)}
\end{verbatim}
}\noindent
\section{\REDUCE{} operator {\tt triggcd}}
The operator {\tt triggcd} is an application of {\tt trigfactorize}.
With its help the user can find the greatest common divisor of two
trigonometric or hyperbolic polynomials. It uses the method described
in \cite{art}. The syntax is: {\tt triggcd(p,q,x)}, where p and q
are the polynomials and x is the smallest unit to use. Example:
{\small
\begin{verbatim}
21: triggcd(sin(x),1+cos(x),x/2);
x
cos(---)
2
22: triggcd(sin(x),1+cos(x),x);
1
\end{verbatim}
}\noindent
The polynomials p and q can consist of both trigonometric and hyperbolic
functions:
{\small
\begin{verbatim}
23: triggcd(sin(2x)*sinh(2x),(1-cos(2x))*(1+cosh(2x)),x);
cosh(x)*sin(x)
\end{verbatim}
}\noindent
\section{Further Examples}
With the help of the package the user can create identities:
{\small
\begin{verbatim}
24: trigsimp(tan(x)*tan(y));
sin(x)*sin(y)
---------------
cos(x)*cos(y)
25: trigsimp(ws,combine);
cos(x - y) - cos(x + y)
-------------------------
cos(x - y) + cos(x + y)
26: trigsimp((sin(x-a)+sin(x+a))/(cos(x-a)+cos(x+a)));
sin(x)
--------
cos(x)
27: trigsimp(cosh(n*acosh(x))-cos(n*acos(x)),trig);
0
28: trigsimp(sec(a-b),keepalltrig);
csc(a)*csc(b)*sec(a)*sec(b)
-------------------------------
csc(a)*csc(b) + sec(a)*sec(b)
29: trigsimp(tan(a+b),keepalltrig);
- (tan(a) + tan(b))
----------------------
tan(a)*tan(b) - 1
30: trigsimp(ws,keepalltrig,combine);
tan(a + b)
\end{verbatim}
}\noindent
Some difficult expressions can be simplified:
{\small
\begin{verbatim}
31: df(sqrt(1+cos(x)),x,4);
4 2 2 2
(sqrt(cos(x) + 1)*( - 4*cos(x) - 20*cos(x) *sin(x) + 12*cos(x)
2 4 2
- 4*cos(x)*sin(x) + 8*cos(x) - 15*sin(x) + 16*sin(x) ))/(16
4 3 2
*(cos(x) + 4*cos(x) + 6*cos(x) + 4*cos(x) + 1))
32: trigsimp(ws);
sqrt(cos(x) + 1)
------------------
16
33: load taylor;
34: taylor(sin(x+a)*cos(x+b),x,0,4);
cos(b)*sin(a) + (cos(a)*cos(b) - sin(a)*sin(b))*x
2
- (cos(a)*sin(b) + cos(b)*sin(a))*x
2*( - cos(a)*cos(b) + sin(a)*sin(b)) 3
+ --------------------------------------*x
3
cos(a)*sin(b) + cos(b)*sin(a) 4 5
+ -------------------------------*x + O(x )
3
35: trigsimp(ws,combine);
sin(a - b) + sin(a + b) 2 2*cos(a + b) 3
------------------------- + cos(a + b)*x - sin(a + b)*x - --------------*x
2 3
sin(a + b) 4 5
+ ------------*x + O(x )
3
\end{verbatim}
}\noindent
Certain integrals whose calculation was not possible in REDUCE
(without preprocessing), are now computable:
{\small
\begin{verbatim}
36: int(trigsimp(sin(x+y)*cos(x-y)*tan(x)),x);
2 2
cos(x) *x - cos(x)*sin(x) - 2*cos(y)*log(cos(x))*sin(y) + sin(x) *x
---------------------------------------------------------------------
2
37: int(trigsimp(sin(x+y)*cos(x-y)/tan(x)),x);
x 2
(cos(x)*sin(x) - 2*cos(y)*log(tan(---) + 1)*sin(y)
2
x
+ 2*cos(y)*log(tan(---))*sin(y) + x)/2
2
\end{verbatim}
}\noindent
Without the package, the integration fails, in the second case one doesn't
receive an answer for many hours.
{\small
\begin{verbatim}
38: trigfactorize(sin(2x)*cos(y)**2,y/2);
{2*cos(x)*sin(x),
y y
cos(---) + sin(---),
2 2
y y
cos(---) + sin(---),
2 2
y y
cos(---) - sin(---),
2 2
y y
cos(---) - sin(---)}
2 2
39: trigfactorize(sin(y)**4-x**2,y);
2 2
{ - sin(y) + x, - (sin(y) + x)}
40: trigfactorize(sin(x)*sinh(x),x/2);
x x x x
{4,cos(---),sin(---),cosh(---),sinh(---)}
2 2 2 2
41: triggcd(-5+cos(2x)-6sin(x),-7+cos(2x)-8sin(x),x/2);
x x
2*cos(---)*sin(---) + 1
2 2
42: triggcd(1-2cosh(x)+cosh(2x),1+2cosh(x)+cosh(2x),x/2);
x 2
2*sinh(---) + 1
2
\end{verbatim}
}
\begin{thebibliography}{99}
\bibitem{art}
Roach, Kelly: Difficulties with Trigonometrics. Notes of a talk.
\bibitem{hearns}
Hearn, A.C.: COMPACT User Manual.
\end{thebibliography}
\end{document}