USMAN IQBAL

Reg ID 150072
MS-17, IAA
Difference Equation :
In mathematics, a recurrence relation / difference equation is an equation that
recursively defines a sequence or multidimensional array of values, once one or more
initial terms are given: each further term of the sequence or array is defined as a
function of the preceding terms.
The term difference equation sometimes (and for the purposes of this article) refers to
a specific type of recurrence relation. However, "difference equation" is frequently used
to refer to any recurrence relation.
Classification:
A dierence equation is said linear when each term of the sequence is dened as a
linear function of the preceding terms and non linear otherwise.
The order of a linear recurrence relation is the number of preceeding terms required by
the denition. Thus, the relation x n = 2xn2 is of order two because at least two
preceeding terms to compute any term xn (whether they are both used or not). The
general form of a linear recurrence relation of order p is as follows:
Xn = an1Xn1 + an2Xn2 + ... + anpXnp + a0
If the coecients ai does not depend on n, then the recurrence relation is said to have
constant coecients. In addition, if a0 = 0, the recurrence relation is said to be
homogeneous. Solving a dierence equation means to nd an explicit relation between
Xn and the initial conditions. The method to solve a dierence equation depends on the
type of equation we have.
yn+1 = yn + d (1st order)
yn+1 = A yn (1st order)
yn+2 = yn+1 + yn (2nd order)
Other examples of linear dierence equations are
yn+2 + 4yn+1 3yn = n2 (2nd order)
yn+1 + yn = n 3n (1st order)
The key point is that for a dierence equation to be classied as linear the terms of the
sequence {yn} arise only to power 1 or, more precisely, the highest subscript term is

obtainable as a linear combination of the lower ones. All the examples cited above are
consequently linear. Note carefully that the term n2 in one example does not imply nonlinearity since linearity is determined by the y n terms.
Examples of non-linear dierence equations are
yn+1 = pyn + y 2 n+1 + 2 yn
3 yn+1 cos(yn+1) = yn
The ve linear equations listed above also have constant coecients; for example:
yn+2 + 4yn+1 3yn = n2 has the constant coecients 1, 4, 3. The (linear) dierence
equation n yn+2 yn+1 + yn = 0 has one variable coecient viz n and so is not classied
as a constant coecient dierence equation.
Linear Constant Coefficients Difference Equation (LCCDE)
An order d linear homogeneous recurrence relation with constant coefficients is
an equation of the form

where the d coefficients ci (for all i) are constants.

More precisely, this is an infinite list of simultaneous linear equations, one for each
n>d1. A sequence that satisfies a relation of this form is called a linear recurrence
sequence or LRS. There are d degrees of freedom for LRS, i.e., the initial values
can be taken to be any values but then the linear recurrence determines
the sequence uniquely.
Filtering (FIR and IIR)
In digital signal processing, recurrence relations can model feedback in a system, where
outputs at one time become inputs for future time. They thus arise in infinite impulse
response (IIR) digital filters.
The difference equation is a formula for computing an output sample at time based on
past and present input samples and past output samples in the time domain. We may
write the general, causal, LTI difference equation as follows:

where

x is the

input
,

signal, y is the

output signal, and

the

constants

are called the coefficients

As a specific example, the difference equation

specifies a digital filtering operation, and the coefficient sets

fully characterize the filter. In this example, we have

and

When the coefficients are real numbers, as in the above example, the filter is said to be
real. Otherwise, it may be complex.
Notice that a filter of the form of Eq. (5.1) can use ``past'' output samples (such as
) in the calculation of the ``present'' output

. This use of past output

samples is called feedback. Any filter having one or more feedback paths (
called recursive.

) is

More specifically, the

coefficients are called the feedforward coefficients and the
coefficients are called the feedback coefficients.
A filter is said to be recursive if and only if
for some
. Recursive filters are
also called infinite-impulse-response (IIR) filters. When there is no feedback
(
digital filter.

), the filter is said to be a nonrecursive or finite-impulse-response (FIR)

When used for discrete-time physical modeling, the difference equation may be referred
to as an explicit finite difference scheme.
Showing that a recursive filter is LTI is easy by considering its impulse-response
representation. For example, the recursive filter

has impulse response

Canonical Forms
In mathematics and computer science, a canonical, normal, or standard form of a
mathematical object is a standard way of presenting that object as a mathematical
expression. The distinction between "canonical" and "normal" forms varies by subfield.
In most fields, a canonical form specifies a unique representation for every object, while
a normal form simply specifies its form, without the requirement of uniqueness.
Canonical form can also mean a differential form that is defined in a natural (canonical)
way.
Partial Differential Equation
A second order linear PDE in two independent variables (x,y) can be written as
A(x,y)2u /x2+ B(x,y)2u /xy+ C(x,y)2u/y2+ D(x,y)u /x+ E(x,y)u /y+ F(x,y) u = G(x,y)
the same in short form can be written as
A(x,y) uxx + B(x,y) uxy + C(x,y) uyy + D(x,y) ux + E(x,y) uy + F(x,y) u = G(x,y)
where the subscript(s) represents the partial dierentiation with respect to the given
index (indices). Inspired by the classication of the quadratic equations as elliptic,
parabolic and hyperbolic, the second order PDE is also classied as elliptic, parabolic or
hyperbolic, at any point (x,y), depending on the value of the discriminant.
B2 4 AC
which is less than, equal to or greater than zero, respectively. Observe that the
coecients of second order partial derivatives only decide the classication. Three well
known examples for Poisson equation (Elliptic), one-dimensional unsteady diusion
equation (Parabolic) and one-dimensional wave equation (Hyperbolic) are given by
2u /x2+2u /xy= f(x,y) (Elliptic)
u /t= K22u/ x2(Parabolic)
2u /t2= a22u/ x2(Hyperbolic)
For hyperbolic equations, there exist two real directions, called characteristic directions
given by
dy/dx=B +(B2 4AC )/2A and
dy/dx=B (B2 4AC )/2A

Along these directions the partial dierential equation takes a simple form called Normal
or Canonical form. These curves are called characteristic curves.
Consider the hyperbolic equation 2u/ t2= a22u/ x2 for which the characteristic curves
can be obtained using
dx/dt=(4a2)/ 2
and
dx/dt= (4a2)/ 2
xat = C1 and x + at = C2
Therefore, the characteristic curves for are given by = x + at and = xat
Transforming in to and , using
/x=1 , / x=1 , /t= a, /t= a
Results in 2u / given equation. In general, the normal form of any hyperbolic
equation is
u = (,,u,u,u) or u u = (,,u,u,u)
Since for the parabolic equations, B24AC = 0, therefore, there exists only one real
characteristic direction (curve) given by
dy/dx=B/2A
Along the curves parabolic equations, in general, take the form
u = (,,u,u,u) or u = (,,u,u,u)
Finally, since B2 4AC < 0 for elliptic equations, there are no real characteristics for
these equations and hence the normal form for these equations will remain as
u + u = (,,u,u,u)
Solving Difference Equation
There are three main method for solving differential equations, these are:
(a) find complimentary & particular parts of solution
(b) sequential procedure
(c) z transform
We will focus on z-transform method only.

Solving with z-transforms

Certain difference equations - in particular, linear constant coefficient difference
equations - can be solved using z-transforms. The z-transforms are a class of integral
transforms that lead to more convenient algebraic manipulations and more
straightforward solutions. There are cases in which obtaining a direct solution would be
all but impossible, yet solving the problem via a thoughtfully chosen integral transform is
straightforward.
A key aspect is the inversion of the z-transform as well as demonstrating the use of
partial fractions for this purpose we show an alternative, often easier, method using
what are known as residues. z-transform is one of the mathematical tools used for the
analysis and design of discrete-time control systems. The role of the z-transform in
digital control systems is analogous to that of the Laplace transform in the continuoustime control systems. By the z-transformation, a linear discrete-time system may be
represented by a transfer function called the pulse transfer function. The z-transform of
the output signal can then be expressed as the product between the system's pulse
transfer function and the z-transform of the input signal.
Solution of difference equations using z-transforms provides a useful method of solving
certain types of dierence equation. In particular linear constant coecient dierence
equations are amenable to the z-transform technique although certain other types can
also be tackled. We shall not consider the problem of solving non-linear dierence
equations.
Solution of rst order linear constant coefcient difference equations Consider the rst
order dierence equation
yn+1 3yn = 4 ; n = 0,1,2,...
The equation could be solved in a step-by-step or recursive manner, provided that y 0 is
known because
y1 = 4 + 3y0 ;
y2 = 4 + 3y1 ;
y3 = 4 + 3y2 and so on. This process will certainly produce the terms of the solution
sequence {yn} but the general term yn may not be obvious.
Now consider
yn+1 3yn = 4 ; n = 0,1,2,...

with initial condition y0 = 1. Now we take z-transform on both side and it results in
following equation
z Y(z)zy0 3Y(z) =4z/(z1)

The equation is no longer a dierence equation instead it is an algebraic equation

where the unknown, Y(z), is the z-transform of the solution sequence {y n}. We now
insert the initial condition y0 = 1 and solve for Y(z):
(z3)Y(z)z =4z/ (z1)
Y(z) =z2 + 3z /(z1)(z3)
The nal step consists of obtaining the sequence {yn} of which Y(z) is the z-transform.
As Y(z) is not recognizable as any of the standard transforms that we have obtained.
Consequently, one method of inverting is to use a partial fraction expansion.
Y(z) = 2z/(z1) + 3z /(z3)
Now taking the inverse z-transform:
yn = 2 + 33n = 2 + 3n+1 for n = 0,1,2,...
From this solution we obtain y0 = 2 + 3 = 1 ; y1 = 2 + 32 = 7; y2 = 2 + 33 = 25 ; y3 =
2 + 34 = 7
Which agree with those obtained by recursive solution of the given problem i.e.
y1 = 4 + 3y0 = 7; y2 = 4 + 3y1 = 25 ; y3 = 4 + 3y2 = 79