Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
345 views

Recurrene Solved Examples

There are two types of linear recurrence relations: homogeneous and non-homogeneous. Homogeneous relations can be solved by finding the characteristic roots of the characteristic equation. The general solution is a linear combination of terms with each characteristic root raised to successive powers. Non-homogeneous relations involve an additional term F(n); their solutions have a general part based on the characteristic roots plus a particular part for the non-homogeneous term. In contrast, nonlinear recurrence relations generally cannot be solved analytically and are highly sensitive to initial conditions.

Uploaded by

compiler&automata
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
345 views

Recurrene Solved Examples

There are two types of linear recurrence relations: homogeneous and non-homogeneous. Homogeneous relations can be solved by finding the characteristic roots of the characteristic equation. The general solution is a linear combination of terms with each characteristic root raised to successive powers. Non-homogeneous relations involve an additional term F(n); their solutions have a general part based on the characteristic roots plus a particular part for the non-homogeneous term. In contrast, nonlinear recurrence relations generally cannot be solved analytically and are highly sensitive to initial conditions.

Uploaded by

compiler&automata
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 5

Note on Linear and Non Linear Recurrence Relations

Linear and Non-linear Recurrence Relations

Linear Recurrence Relations

Linear recurrence relations are the relations of the terms of the sequence which is in the linear form. Linear

recurrence relations are of two types: Homogeneous and Non-homogeneous. The methods to solve both

types of recurrence relations are discussed below.

Solving Linear Recurrence Relations

Homogeneous recurrence relation

A recurrence relation of the form an = c1an-1 + c2an-2 + + ckan-k is called a linear homogeneous recurrence

relation of degree k where c1, c2, , ck are constant coefficient and ck 0.

Let an = c1an-1 + c2an-2 + + ckan-k be a recurrence relation. Now, an = rn be the solution of recurrence relation

if and only if it satisfies the given recurrence relation i.e.

rn = c1rn-1 + c2rn-2 + + ckrn-k ---(i)

Dividing the equation (i) by rn-k, we get

rk = c1rk-1 + c2rk-2 + + ck

This equation is called the characteristic equation and the root of this equation is called the characteristic

root. In solving the recurrence relation of the type above, the approach is to look for the solution of the form

an = rn, where r is a constant. an = rn is a solution of a recurrence relation an = c1an-1 + c2an-2 + + ckan-k if and

only if rn = c1rn-1 + c2rn-2 + + ckrn-k. When we divide both sides by rn-k and transpose the right hand side we

have,

rk c1rk-1 - c2rk-2 - - ck = 0. Here, we can say an = rn is a solution if and only if r is the

solution if the equation rk - c1rk-1 - c2rk-2 - - ck = 0 (characteristic equation of the

recurrence relation) and solutions to this equations are called characteristic roots of the

recurrence relation.

For example: Find the characteristic root of an = an-1 + an-2

Solution:

Let an = rn

i.e. rn = rn-1 + 2rn-2 ---(i)


Dividing equation (i) by rn-2,

r2 = r+ 2

or, r2 r 2 = 0

or, r2 2r + r 2 = 0

or, r(r 2) + 1(r 2) = 0

or, (r + 1)(r + 2) = 0

r = -1, 2

r1 = -1 and r2 = 2

Theorem: Let r2 - c1r + c2 = 0 be the characteristic equation, then sequence {an} is the solution of recurrence

relation, then

an = c1an-1 + c2an-2 iff

an = 1r1n + 2r2n

where c1, c2, 1, 2 are constants and r1 and r2 are distinct roots of the characteristic equation. If the roots are

equal, then solution is of the form an = 1rn + n2rn

For example, find the solution for the given recurrence relation: an = an-1 + 2an-2 where a0 = 1 and a1= 4.

Solution:

Let an = rn

Then, rn = rn-1 + 2rn-2 ---(i)

Dividing equation (i) by rn-2,

r2 = r + 2

or, r2 r 2 = 0

or, r2 2r + r 2 = 0

or, r(r 2) + 1(r 2) = 0

or, (r 2)(r 1) = 0

r1 = 2, r2 = -1

Now, the solution of the recurrence relation is:

an = 1r1n + 2r2n

or, an = 1(2)n + 2(-1)n ---(ii)


Put n = 0,

a 0 = 1 + 2

or, 1 = 1 + 2 ---(iii)

Put n = 1,

a1 = 1(2)1 + 2(-1)1

or, 4 = 21 - 2 ---(iv)

Adding (iii) and (iv), we get

1 = 5/3

Substituting the value of 1 in equation (iii),

1 = 1 + 2

or, 1 5/3 = 2

or, 2 = -2/3

an = (5/3)(2)n (2/3)(-1)n

Non-homogeneous Recurrence Relation

A recurrence relation of the form an = c1an-1 + c2an-2 + + ckan-k + F(n) is called a linear non-homogeneous

recurrence relation where c1, c2, , ck are constants and ck 0 and F(n) is any functional equation. an = 2an-1 +

3an-2 + 2n is a linear non-homogeneous recurrence relation.

For example: Find the solution for the given non-homogeneous recurrence relation.

an = 3an-1 + 2n with a0 = 2.

Solution:

First, we find the solution associated with the homogeneous part. So,

Let an = rn

Then, rn = 3rn-1

Dividing both sides by rn-1,

r=3

The general solution of the equation is

an = c1rn

an = c13n
Now, we find the solution associated with the non-homogeneous part, which we call the particular solution.

Let the solution of non-homogeneous part be

an = c22n ---(i)

put n = n-1 in equation (i)

an-1 = c22n-1 ---(ii)

We have,

an = 3an-1 + 2n

or, c22n = 3c22n-1 + 2n

or, c2(2n 3.2n-1) = 2n

or, c22n(1 3.(1/2)) = 2n

or, c2(-1/2) = 1

or, c2 = -2

The particular solution is

an = -2.2n

Now, the total solution is

T.S. = G.S. + P.S. [G.S. = general solution and P.S. = Particular solution]

an = c13n + (-2).2n ---(iii)

put n = 0,

a0 = c1 + (-2)

or, 2 = c1 2

or, c1 = 4

The solution is

an = 4.3n 2.2n

Solving Non-Linear Recurrence Relations

By now, only linear recurrence relations can be solved and the analytic solution of even the simplest form of

the non-linear recurrence relations are impossible. Only selected non-recurrence relations can be solved. In

most cases, the behaviour after many iterations is extremely sensitive to even tiny variations in the initial
conditions and as a result, the formula representing the relationship will grow very large. So, there is no

definitive solution for the non-linear recurrence relations as of now.

You might also like