Scribd
Upload
19 views3 pages

Bessel Function Integral Representation

The document presents the derivation of a solution to a difference equation using the Laplace transform and Bessel functions. It begins with the difference equation and its solution via Laplace transform. It then introduces Bessel functions and represents them as contour integrals involving the gamma function. After substituting the gamma function representation and changing variables, it arrives at an integral representation of Bessel functions with limits. Comparing this to the original solution, it identifies the solution in terms of Bessel functions.

Uploaded by

Mefis
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
19 views3 pages

Bessel Function Integral Representation

The document presents the derivation of a solution to a difference equation using the Laplace transform and Bessel functions. It begins with the difference equation and its solution via Laplace transform. It then introduces Bessel functions and represents them as contour integrals involving the gamma function. After substituting the gamma function representation and changing variables, it arrives at an integral representation of Bessel functions with limits. Comparing this to the original solution, it identifies the solution in terms of Bessel functions.

Uploaded by

Mefis
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

The difference equation is:

Y(z + 2) − [A(z + 1) + B]Y(z + 1) + Y(z) = 0 (1)

The solution for (1) using the Laplace transformation [1] is:
Z ∞
1 1
t A +z t−1 e− A (t− t ) dt
B
Y(z) = (2)
0

Now we will proceed to simplify the expression of the solution found.


The Bessel’s function of order ν, when ν is a positive integer, may be defined
like[2]:
∞
 x ν X (−1)k  x 2k
Jν (x) = (3)
2 k=0 Γ(k + 1)Γ(ν + k + 1) 2
We use [3] the representation of the gamma function [4]:
Z
1 i
= (−u)−n e−u du (4)
Γ(n) 2π C

where C is any simple curve enclosing the point t = 0. In this case C is some
contour from ∞, turning around 0 and going back to ∞. Introducing (4)
into (3), with n = ν + k + 1, we get:
∞
(−1)k (−u)−ν−k−1  x 2k −u
Z X
i  x ν
Jν (x) = e du (5)
2π 2 C k=0 Γ(k + 1) 2

 2 k
i  x ν
Z X ∞ (−1)k x
4
(−u)−k
Jν (x) = (−u)−ν−1 e−u du (6)
2π 2 C k=0 k!

From equation (6) the term:


 2 k
∞
X (−1)k x
4
(−u)−k x2
= e 4u (7)
k=0
k!

Therefore, the equation (6) may be written like:


Z
i  x ν x2
 
−ν−1 − u− 4u
Jν (x) = (−u) e du (8)
2π 2 C

1
x

Let u = − 2
t, substituting in (8), then we get [5]:
Z Z 0
1 x 1 1 x 1
Jν (x) = t−ν−1
e( 2 )(t− t ) dt = t−ν−1 e( 2 )(t− t ) dt (9)
2πi C 2πi ∞

This is an integral representation to Bessel’s function, with ν ∈ Z.

(In the reference [4] the Bessel’s function is shown like a integral with limits,
unlike other representations where it is given with a contour integral, which
are shown in Schaum or [2]. I understand that the limits of the reference [4]
come from another way of expressing the gamma function. My question is:
Is it valid to move from a contour integral to one with limits in the equation
(9)?)

May be written like:


Z 0
x 1
2πiJν (x) = t−ν−1 e( 2 )(t− t ) dt (10)
∞

Comparing with our solution obtained (2), if:


1 1 2 1 x 1
e− A (t− t ) = e− 2A (t− t ) = e 2 (t− t ) (11)

where x = − A2 . From (2) if, BA


+ z = −ν, and inverting the limits of the
integral we get:
Z 0
x 1
Y(z) = − t−ν−1 e( 2 )(t− t ) dt = −2πiJν (x) (12)
∞

Therefore:  
2
Y(z) = −2πiJ(−( B +z)) − (13)
A A
Is the solution in form Bessel’s function.

2
References
[1] H. Levy and F. Lessman, “Finite Difference Equations”, Ed. Dover Pub-
lications.

[2] Saber Elaydi, “An introduction to difference equations”, Third Edition,


Ed. Springer.

[3] [Link]

[4] I. S. Gradshteyn and I. M. Ryzhik, “Table of integrals, series and prod-


ucts”, Sixth Edition, Ed. Academic Press. Pag. 885.

[5] I. S. Gradshteyn and I. M. Ryzhik, “Table of integrals, series and prod-


ucts”, Sixth Edition, Ed. Academic Press. Pag. 903.

You might also like