See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/325145765

Legendre Wavelet for Solving Linear System of Fredholm And Volterra

Integral Equations

Article · January 2013

CITATIONS READS

2 71

2 authors, including:

Mohamed Torky
Port Said University
6 PUBLICATIONS   31 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Solving fredholm and volterra integral equations View project

All content following this page was uploaded by Mohamed Torky on 15 May 2018.

The user has requested enhancement of the downloaded file.

International Journal of Research in Engineering and Science (IJRES)
ISSN (Online): 2320-9364, ISSN (Print): 2320-9356
www.ijres.org Volume 1 Issue 7 ǁ Nov. 2013 ǁ PP.14-22

Legendre Wavelet for Solving Linear System of Fredholm And

Volterra Integral Equations
M. A. Mohamed1, M. Sh. Torky1
1
Faculty of Science, Suez Canal University Ismailia, Egypt,
2
The High Institute of Administration and Computer, port Said University, Egypt

Abstract
In this work, we employ Legendre wavelet method to find numerical solution of system of linear Fredholm and
Volterra integral equations, which uses zeros of Legendre wavelets for collocation points is introduced and used
to reduce this type of system of integral equations to a system of algebraic equations.
Keywords: system of linear Fredholm integral equations, system of linear Volterra integral equations,
Legendre wavelets, linear algebraic equations.

I. Introduction
Consider the system of linear Fredholm and Volterra integral equations

and

Where

In system (1), (2) the known kernel K(x, t) is continuous, the function F (x) is given, and U(x) is the
solution to be determined.
The system (1) and (2) have been solved by several methods. For instance, homotopy analysis method
(HAM) [1], the Adomian decomposition method (AMD) [2], Taylor-series expansion method [3], Modified
homotopy perturbation method (MHPM) [4], delta functions (DFs) [5], hat basis functions [6].
In recent years, wavelets have found their way into many different fields of science and engineering. The uses
of wavelets method for solving integral equations has considered by many authors: Legendre and Chebyshev
wavelets method [7, 8], Wavelet Galerkin method [9].
In this study, we want to solve system (1) by using Legendre wavelets. The method consist of reducing
the solution of (1) to a system of linear algebraic equations and also a collocation method is used for solving
system (2).The method consist of expanding the solution by Legendre wavelets with unknown coefficients. For
a suitable collocation points we choose zeros of Legendre wavelets. The properties of Legendre wavelets
together with zeros of Legendre wavelets are then utilized to evaluate the unknown coefficients and to find an
approximate solution to system (2).

II. Properties of Legendre wavelets

2.1 Wavelets and Legendre wavelets
Wavelets constitute a family of functions constructed from dilation and translation of a single function
called the mother wavelet. When the dilation parameter a and the translation parameter b vary continuously we
have the following family of continuous wavelets [10]

If we restrict the parameters a and b to discrete values as and

we have the following family of discrete wavelets:

www.ijres.org 14 | Page
 Legendre wavelet for solving linear system of Fredholm and Volterra integral equations

Where form a wavelet basis for . In particular, when and then forms an
orthonormal basis [10]. Legendre wavelets have four arguments;
can assume any positive integer, m is the order for Legendre polynomials and t is the
normalized time. They are defined on the interval [0, 1):

where The coefficient is for orthonormality,the

dilation parameter is and translation parameter is .Here, Lm (t) are the well-known Legendre
polynomials of order m, which are orthogonal with respect to the weight function w (t) = 1 and satisfy the
following recursive formula:

2.2 Function approximation

A function defined over [0, 1) may be expanded as

Where in which (.,.) denotes the inner product. If the infinite series in Eq.(8) is
truncated, then Eq.(8) can be written as In the infinite series in eq.(4) is truncated , then eq.(4) can be written as :

Where and are matrices given by

Similarly, a function may be approximated as:

where matrix, with

The integration of the product of two Legendre wavelets vector function is obtained as:

where I is an identity matrix.

III. System of linear integral equations

In this section, we use Legendre wavelets method for solving system of linear Fredholm and Volterra
integral equations.

3.1 Legendre wavelets method for solving System of linear Fredholm integral equations:
Consider the system of linear Fredholm integral equations as follows:

Where

In Eq. (15), the functions K and F are given, and U is the vector function of the solution of system (15)
that will be determined. Consider the ith equation of (15)

www.ijres.org 15 | Page
 Legendre wavelet for solving linear system of Fredholm and Volterra integral equations

Where and is an unknown function.We approximate by

(6)-(9) as follows:

By substituting the above relation in (17), we have:

Then we have the following linear system:

3.2 Legendre wavelets method for solving System of linear Volterra integral equations:
For solving the system of linear Volterra integral equations (2), we consider the ith equation of (17):

where and is an unknown function.In order to use Legendre wavelets,

we first approximate as

where C and are defined similarly to Eq.(10) and (11). In view of Eqs. (22) and (23) we have

Now we collocate Eq.(24) at 2 k-1M points.

Suitable collocation points are zeros of Chebyshev polynomial [11]

However, in this method we choose zeros of Legendre wavelets for collocation points. For this purpose, we put
k = 1 in Eq. (6), so the Legendre wavelet of order m is computed as follows

where are Legendre wavelet defined on the interval [0, 1) and Lm(t) are the well-known
Legendre polynomials. Eq. (25) gives (n2k-1 M) linear equations which can be solved for the vector
in Eq.(23).

IV. Illustrative examples

To demonstrate the effectiveness of the method, here we consider some system of linear Fredholm and
Volterra integral equations. The Legendre wavelets are defined only for t ϵ [0, 1], we take a = 0, b = 1. The
computations associated with the examples were performed using Mathematica and Maple.

Example 4.1[3]: Consider the following system of Fredholm integral equations

Taylor-series expansion method has been applied to solve the system (27) by K. Maleknejad, N. Aghazadeh and
M. Rabbani en [3]. The better result accomplished with m=10 and the maximum absolute errors for u(x)
is0.01656 and for v(x) is 0.1105 (see table 1 in [3]). We apply the presented method in this section and solve
Eq. (27) with k = 1 and M = 5. Using Eq. (6), (7) and (13) we obtain

www.ijres.org 16 | Page
 Legendre wavelet for solving linear system of Fredholm and Volterra integral equations

where

Using Eq.(20) and Eq.(27) we have

where

Substituting from (29)-(31) into (20) and solving the resulting system of algibric equations we have

From Eq. (9) we have

Which is exact solution and provide that our method is more accurate than the methods in [3]

Example 4.2([4]): Consider the following system of Fredholm integral equations

With exact solution . M. Javidi en [4] solved the system (34)

by modified homotopy perturbation method when N=5 the corresponding absolute error for u(x) and v(x) are
and respectively, and when N=10 the corresponding absolute error for u(x) and
v(x) are and respectively (see table1 in [4]). Now, we apply the technique
described in example 4.1 with k = 1, M = 10 to obtain the approximate solution of (34) as

www.ijres.org 17 | Page
 Legendre wavelet for solving linear system of Fredholm and Volterra integral equations

Fig.1 shows the solution of obtained by the present method and the exact solution and the error
computed in table 1, which is more accurate, again than the solution obtained in [4].

Fig.1 exact and Legendre wavelet solution of u(x) and v(x),

Table 1. The error of u(x) and v(x) of example 4.2

x Error u Error v
0
0.2
0.4
0.6
0.8
1.0

Example 4.3 ([12]): Consider the following system of Fredholm integral equations

With exact solution . K. Maleknejad and F. Mirzaee [12] solved (2.72) with
k= 16 and 32. The best results of (2.72) obtained at k=32 and the maximum absolute error evaluated in [12] of
u(x), v(x) are 0.01566 and 0.01321.We apply the technique described in example 4.1and solved Eq.(30).In
addition, Fig.2 show the solution of obtained by the present method with k = 1, M = 5 and the exact
solution and the error computed in table 2.Through the results, obtained using our method (see table 2) prove it
more accurate than the results that have been obtained in [12].

www.ijres.org 18 | Page
 Legendre wavelet for solving linear system of Fredholm and Volterra integral equations

Table 2. The error of u(x) and v(x) of example 4.3

x Error u Error v
0 1.10145×10-11 .43682×10-5
0.2 1.64643×10-12 2.63044×10-6
-12
0.4 8.10324×10 4.94216×10-6
-11
0.6 1.82401×10 4.7315×10-6
-11
0.8 2.87543×10 2.18088×10-6
-11
1.0 3.96211×10 1.69812×10-5

Fig.2 exact and Legendre wavelet solution of u(x) and v(x),

Example 4.4 ([15, 16]): Consider the following system of Volterra integral equations

With exact solution . The system (31) solved by M.I Berenguer and D.
Gámez [15] by Biorthogonal systems and also the solution of (31) founded by J. Biazar and H. Ghazvini with
homotopy perturbation method [16]. The results obtained by M.I Berenguer and D. Gámez [15] with j= 9, 17, 33
and the best results is calculating when j=33 which the absolute maximum error of u(x) and v(x) are 2.81E-2
and 6.71E-2 respectively (see table 3 in [15]). In addition, the maximum error obtained by J. Biazar and H.
Ghazvini [16] with 5 terms for u (x), v (x) are 1.0E-3 and 1.4E-3 respectively (see table 1 in [16]). We apply
the Legendre wavelets approach (22) - (25) and solved Eq.(31). Fig.3 show the solution of obtained
by the present method with k = 1, M = 5 and the exact solution and the error computed in table 3.

Fig.3 exact and Legendre wavelet solution of u(x) and v(x),

Table 3. The error of u(x) and v(x) of example 4.4.

www.ijres.org 19 | Page
 Legendre wavelet for solving linear system of Fredholm and Volterra integral equations

x u exact – u Lw v exact – v Lw
0.2 1.2906×10-5 1.2448×10-5
0.4 8.8772×10-6 1.4751×10-5
0.6 2.6478×10-7 4.0975×10-6
0.8 4.0576×10-6 1.0665×10-5
1 3.7849×10-5 8.2414×10-5

Example 4.5 ([6]): Consider the following system of Volterra integral equations

With exact solution . E. Babolian M. Mordad [6] used hat basis functions for the
approximate solution of Eq. (2.73) with n=64 and they obtain the best results E. Babolian M. Mordad [6] by
doing large steps to obtain this solution. The maximum absolute error obtained is 0.00539 for u(x) and 0.00717
for v(x) (see table 5 in [6]).

We apply the Legendre wavelets approach (22) – (25) and solved Eq.(32), and Fig.4 show the solution of
obtained by the present method with k = 1, M = 5 and the exact solution and the error computed in
table 4 .

Table 4. The error of u(x) and v(x) of example 4.5

x u exact – u Lw v exact – v Lw
0 3.10862×10-15 1.93352×10-15
-16
0.2 1.31839×10 4.99600×10-16
-16
0.4 1.11022×10 3.33067×10-16
-16
0.6 1.11022×10 2.22045×10-16
-16
0.8 3.33067×10 0.
1 3.33067×10-16 2.22045×10-15

Fig.4 exact and Legendre wavelet solution of u(x) and v(x),

Example 4.6([15, 17, 18]): Consider the following system of Volterra integral equations

www.ijres.org 20 | Page
 Legendre wavelet for solving linear system of Fredholm and Volterra integral equations

With exact solution .

The system (33) solved by three method, first by block-by-block method by R. Kataniand and S.
Shahmorad [18], second, an expansion method by M. Rabbania and K. Maleknejad [17], third, Biorthogonal
systems M.I. Berenguer and D. Gámez [15]. The best results is obtained with h=0.05, m=1 and j=33 by these
three methods respectively. The maximum absolute errors are 1.469E-6, 3.84E-2, 3.48E-2 for u(x) and 1.128E-
6, 3.32E-2, 5.01E-4 for v(x) by using method 1(see table 5 in [18]), method 2 (see table 1in [17]) and method 3
(see table 1in [15]) respectively. We apply the Legendre wavelets approach (22) – (25) and solved Eq.(32).
Fig.5 show the solution of obtained by the present method with k = 1, M = 5 and the exact solution
and the error computed in table 5. Our method more accurate (see table 5) then these methods and get this best
results with a few steps.

Table 5. The error of u(x) and v(x) of example 4.6

x u exact – u Lw v exact – v Lw
0 2.88658×10-15 8.21565×10-15
-16
0.2 2.22045×10 6.66134×10-16
-16
0.4 2.22045×10 0.
-16
0.6 6.66134×10 4.44089×10-16
0.8 0. 2.22045×10-16
-16
1 8.88178×10 3.55271×10-15

Fig.5 exact and Legendre wavelet solution of u(x) and v(x),

V. Conclusions
In this paper, Legendre wavelets method was aplied for solving system of linear integral equations. The
present method reduced the system of integral equations to a system of linear algebraic equations. We have
provided some illustrative examples to illustrate the feasibility of and validity of this method.The results showed
that the method is very accurate and simple.

Reference
[1] A.Shidfar, A.Molabahrami, Solving a system of integral equations by an analytic method, Math. andComp. Modelling 54
(2011)828–835.
[2] E. Babolian , J. Biazar , A.R. Vahidi, On the decomposition method for system of linear equations and system of linear Volterra
integral equations, Appl. Math.and Comput. 147 (2004) 19–27.
[3] K. Maleknejad, N. Aghazadeh, M. Rabbani, Numerical solution of second kind Fredholm integral equations system by using a
Taylor-series expansion method, Appl. Math. and Comput. 175 (2006) 1229–1234.
[4] Javidi M., Modified homotopy perturbation method for solving system of linear Fredholm integral equations, Math. Comput.
Modelling 50 (2009) 159–165.

www.ijres.org 21 | Page
 Legendre wavelet for solving linear system of Fredholm and Volterra integral equations

[5] M.Roodaki, H.Almasieh, Deltabasisfunctionsandtheirapplicationstosystemsofintegralequations, Comput. Math. Appl.

63(2012)100–109.
[6] E.Babolian, M. Mordad, A numerical method for solving systems of linear and nonlinear integral equations of the second kind by
hat basis functions, Comput. Math. Appl. 62(2011)187–198.
[7] S. Yousefi, M. Razzaghi, Legendre wavelets method for the nonlinear Volterra-Fredholm integral equations, Mathematics and
Computers in Simulation, 70 (2005) 1-8.
[8] Babolian, F. Fattahzadeh, Numerical computation method in solving integral equations by using Chebyshev wavelet operational
matrix of in-tegration, Appl. Math.and Comput, 188(2007), 1016-1022.
[9] Y. Mahmoudi, Wavelet Galerkin method for numerical solution of nonlinear integral equation, Appl. Math. and Comput.,
167(2005), 1119-1129.
[10] J.S.Gu, W.S. Jiang, The Haar wavelets operational matrix of integration, International Journal of Systems Science, 27(1996),
623-628.
[11] E. Constantinides, Applied Numerical Method with Personal Computers, McGraw-Hill, New York 1987.
[14] M aleknejad K., Mirzaee F., Numerical solution of linear Fredholm integral equations system by rationalized Haar
functions method, Int. J. Comput. Math. 80 (11) (2003) 1397–1405.
[15] Berenguer M. I., Gámez D., Garralda-Guillem A. I., Ruiz Galán M., M.C. Serrano Pérez, Biorthogonal systems for solving
Volterra integral equation systems of the second kind, Journal of Computational and Appl. Math. 235 (2011)1875–1883.
[16] Biazar J , Ghazvini H , He’s homotopy perturbation method for solving systems of Volterra integral equations of the second kind,
Chaos Solitions Fractals 39 (2009) 770–777.
[17] Rabbani M., Maleknejad K., Aghazadeh N., Numerical computational solution of the Volterra integral equations system of the
second kind by using an expansion method, Appl. Math. Comput. 187 (2007) 1143–1146.
[18] Katani R., Shahmorad S., Block by block method for the systems of nonlinear Volterra integral equations, Applied Mathematical
Modelling 34 (2010) 400–406.

www.ijres.org 22 | Page

View publication stats