Forward r-difference Operator and Finding Solution of Nonhomogeneous Difference Equations

Similar documents
Review for Exam 2. Review for Exam 2.

1 Series Solutions Near Regular Singular Points

Equations with regular-singular points (Sect. 5.5).

Sequences and Series, Induction. Review

LINEAR DIFFERENCE EQUATIONS SIGMUNDUR GUDMUNDSSON. [ November 2017 ]

Second Order Linear Equations

M.SC. PHYSICS - II YEAR

Series Solution of Linear Ordinary Differential Equations

Research Article On the Stability of Cubic Mappings and Quadratic Mappings in Random Normed Spaces

Linear differential equations with constant coefficients Method of undetermined coefficients

Series Solutions of Differential Equations

ODE. Philippe Rukimbira. Department of Mathematics Florida International University PR (FIU) MAP / 92

Prime and irreducible elements of the ring of integers modulo n

Section 4.7: Variable-Coefficient Equations

FRACTIONAL HYPERGEOMETRIC ZETA FUNCTIONS

Course 311: Michaelmas Term 2005 Part III: Topics in Commutative Algebra

The Method of Frobenius

Example: This theorem is the easiest way to test an ideal (or an element) is prime. Z[x] (x)

Numerical Analysis Solution of Algebraic Equation (non-linear equation) 1- Trial and Error. 2- Fixed point

Research Article He s Variational Iteration Method for Solving Fractional Riccati Differential Equation

Chapter 8: Taylor s theorem and L Hospital s rule

Logic and Discrete Mathematics. Section 6.7 Recurrence Relations and Their Solution

Bessel s Equation. MATH 365 Ordinary Differential Equations. J. Robert Buchanan. Fall Department of Mathematics

Using Abel's Theorem to Explain Repeated Roots of the Characteristic Equation

A Computational Approach to Study a Logistic Equation

Power series solutions for 2nd order linear ODE s (not necessarily with constant coefficients) a n z n. n=0

VISVESVARAYA TECHNOLOGICAL UNIVERSITY, BELAGAVI B.E. SYLLABUS FOR

The primitive root theorem

Introduction to Series and Sequences Math 121 Calculus II Spring 2015

Module 10: Finite Difference Methods for Boundary Value Problems Lecture 42: Special Boundary Value Problems. The Lecture Contains:

4 Power Series Solutions: Frobenius Method

LINEAR DIFFERENTIAL EQUATIONS. Theorem 1 (Existence and Uniqueness). [1, NSS, Section 6.1, Theorem 1] 1 Suppose. y(x)

On the mean values of an analytic function

Ordinary Differential Equations (ODEs)

Characterizations of Prime and Weakly Prime Subhypermodules

Basic Theory of Linear Differential Equations

Polynomial Solutions of Nth Order Nonhomogeneous Differential Equations

MATHEMATICAL METHODS INTERPOLATION

Series Solutions of ODEs. Special Functions

PARTIAL DIFFERENTIAL EQUATIONS

2 Series Solutions near a Regular Singular Point

2nd-Order Linear Equations

SECOND ORDER ODE S. 1. A second order differential equation is an equation of the form. F (x, y, y, y ) = 0.

Polynomial Interpolation

Series Solutions Near a Regular Singular Point

Solution of Constant Coefficients ODE

Additional material: Linear Differential Equations

MATH 117 LECTURE NOTES

FUNCTIONAL SEQUENTIAL AND TRIGONOMETRIC SUMMABILITY OF REAL AND COMPLEX FUNCTIONS

Homogeneous Linear ODEs of Second Order Notes for Math 331

MATH 326: RINGS AND MODULES STEFAN GILLE

Math Assignment 11

Partition of Integers into Distinct Summands with Upper Bounds. Partition of Integers into Even Summands. An Example

Math 117: Continuity of Functions

5.2 - Euler s Method

Engineering Mathematics through Applications

Two Step Method for Fuzzy Differential Equations

ODEs Cathal Ormond 1

Two Step Hybrid Block Method with Two Generalized Off-step Points for Solving Second Ordinary Order Differential Equations Directly

Huang method for solving fully fuzzy linear system of equations

Second Order and Higher Order Equations Introduction

Distribution Solutions of Some PDEs Related to the Wave Equation and the Diamond Operator

ODE Homework Series Solutions Near an Ordinary Point, Part I 1. Seek power series solution of the equation. n(n 1)a n x n 2 = n=0

Math 256: Applied Differential Equations: Final Review

Theory of Higher-Order Linear Differential Equations

Discrete Orthogonal Polynomials on Equidistant Nodes

Section 1.4: Second-Order and Higher-Order Equations. Consider a second-order, linear, homogeneous equation with constant coefficients

Chapter 4: Higher-Order Differential Equations Part 1

ENGINEERING MATHEMATICS I. CODE: 10 MAT 11 IA Marks: 25 Hrs/Week: 04 Exam Hrs: 03 PART-A

The Free Central Limit Theorem: A Combinatorial Approach

Math 3313: Differential Equations Second-order ordinary differential equations

Differential Equations Practice: 2nd Order Linear: Nonhomogeneous Equations: Undetermined Coefficients Page 1

Engg. Math. II (Unit-IV) Numerical Analysis

A Brief Review of Elementary Ordinary Differential Equations

NEW ITERATIVE METHODS BASED ON SPLINE FUNCTIONS FOR SOLVING NONLINEAR EQUATIONS

Solving Nonhomogeneous PDEs (Eigenfunction Expansions)

Notation Nodes are data points at which functional values are available or at which you wish to compute functional values At the nodes fx i

Physics 307. Mathematical Physics. Luis Anchordoqui. Wednesday, August 31, 16

A Novel Sparse Orthogonal Matrix Construction over the Fields of Characteristic 2

dx n a 1(x) dy

On asymptotic behavior of a finite Markov chain

Complex Numbers. Basic algebra. Definitions. part of the complex number a+ib. ffl Addition: Notation: We i write for 1; that is we

A Comparison between Solving Two Dimensional Integral Equations by the Traditional Collocation Method and Radial Basis Functions

A Novel Computational Technique for Finding Simple Roots of Nonlinear Equations

Adomian Decomposition Method with Laguerre Polynomials for Solving Ordinary Differential Equation

On the Computational Procedure of Solving Boundary Value Problems of Class M Using the Power Series Method

Different Approaches to the Solution of Damped Forced Oscillator Problem by Decomposition Method

MEAN SQUARE SOLUTIONS OF SECOND-ORDER RANDOM DIFFERENTIAL EQUATIONS BY USING HOMOTOPY ANALYSIS METHOD

Homework 3 Solutions Math 309, Fall 2015

Linear Algebra: Lecture notes from Kolman and Hill 9th edition.

SECOND-ORDER LINEAR DIFFERENTIAL EQUATIONS

Four Point Gauss Quadrature Runge Kuta Method Of Order 8 For Ordinary Differential Equations

Infinite elementary divisor structure-preserving transformations for polynomial matrices

Notes on Continued Fractions for Math 4400

Ring Theory Problems. A σ

The Corrected Trial Solution in the Method of Undetermined Coefficients

VEER NARMAD SOUTH GUJARAT UNIVERSITY, SURAT

Understand the existence and uniqueness theorems and what they tell you about solutions to initial value problems.

Integration of Ordinary Differential Equations

Impulse Response Sequences and Construction of Number Sequence Identities

Transcription:

International Mathematical Forum, 2, 2007, no. 40, 957-968 Forward r-difference Operator and Finding Solution of Nonhomogeneous Difference Equations Hassan Hosseinzadeh and G. A. Afrouzi Islamic Azad University, Ghaemshahr Branch P.O. Box 6, Ghaemshahr, Iran Abstract In this paper we introduce a new operator that we call it the forward r-difference operator Δ r which is defined in the following: Δ r y n = y n+ ry n. Then, we investigate some properties of this new operator, we find a shift exponential formula and use it for finding solution of the nonhomogeneous difference equations with constant coefficients, may be written in the following form m ( Δ ri )y n = f n. i= Keywords: Forward difference operator Δ; Forward r-difference operator Δ r ; Difference equation; Shift exponential operator; Particular solution Preliminaries.. In Numerical Analysis, we use some linear operators, those are shift exponential operator E, Ef j = f j+, forward difference operator Δ, Δf j = f j+ f j and backward difference, f j = f j f j. These operators are used in some topics of Numerical Analysis, particularly in interpolation, quadratures, difference equations, and so forth see [], [2], [4]. Since E, Δ and are linear operator, in addition to every order of them, their inversion E,Δ, and every polynomial of them are linear too. We know that the difference equation appears in numerical solution of ODE, PDE, IE, IDE,. Corresponding author E-mail: afrouzi@umz.ac.ir also both authors are in : Department of Mathematics, Faculty of Basic Science, Mazandaran University, Babolsar, Iran.

958 H. Hosseinzadeh and G. A. Afrouzi In this paper we find particular solution of nonhomogeneous difference equations with constant coefficients. The difference equations are written in one of the following forms: Under the shift operator E P (E) y n =0 (homogeneous) () P (E) y n = f n (nonhomogeneous) (2) Under the forward difference operator Δ P (Δ) y n =0 (homogeneous) () P (Δ) y n = f n (nonhomogeneous) (4) Under the forward difference operator P ( ) y n =0 (homogeneous) (5) P ( ) y n = f n (nonhomogeneous) (6) where as P is a polynomial. For solving difference equations and finding general solution, we use the following theorems see [], [2], [4], [5]. Theorem. (superposition principle) Suppose that y n,y n2,,y nm are the (fundamental) solutions of the homogeneous difference equation P (Δ) y n =0, then each arbitrary linear combination of them is a solution for it too. Theorem 2. Suppose that the complex-valued function y n = y n + iy n2 is a solution of the homogeneous equation P (Δ) y n = 0, then each of functions y n,y n2 also are solutions for it. Theorem. Let y h be a solution for the homogeneous P (Δ) y n = 0 and y p be a solution for the nonhomogeneous P (Δ) y n = f n, then y c = y h + y p is a solution for P (Δ) y n = fn too.

Forward r-difference operator 959 2 Solution of the difference equations 2.. In this section, we discuss the difference equations with constant coefficients. Let y n = r n be a solution for equations (), () and (5), then r n must be satisfy in each of them, therefore we have, respectively P (r) = 0 (7) P (r ) = 0 (8) P ( r ) = 0 (9) Where (7) is the corresponding characteristic equation to equation (), and (8) is the corresponding characteristic equation to (), and finally (9) is the corresponding characteristic equation to (5). Remark. All roots of the characteristic equations may be distinct real values, either some of them are equal or some of them are conjugate complex number. (i) If r,r 2,,r k are distinct real roots to the characteristic equations, then r n,rn 2,,rn k will be solutions of homogeneous equations, these functions are linearly independent []. These functions are said fundamental solutions of homogeneous equation. (ii) If r = r 2 =... = r m = r, are the iterated root of the characteristic equation, then the corresponding solutions of homogeneous equation are: r n,nr n,n 2 r n,,n m r n that are linearly independent []. (iii) If r 2 = αiβ, are two conjugate complex roots, in this case the corresponding solution of homogeneous equation are, y n =(α 2 + β 2 ) n 2 cos nϕ, y n2 = (α 2 + β 2 ) n 2 sin nϕ, where ϕ = tan β []. α Example. Find the fundamental solutions of the homogeneous difference equation (E 4 4E +7E 2 6E +2)y n =0. Solution. It s characteristic equation is r 4 4r +7r 2 6r +2=0. This polynomial equation has one double root r = and two complex conjugate roots r = i, therefore the fundamental solutions may be written as follows: y n =, y n2 = n, y n =2 n 2 cos nπ 4, y n4 =2 n 2 sin nπ 4.

960 H. Hosseinzadeh and G. A. Afrouzi Example 2. Find the fundamental solutions of the following homogeneous difference equation (2Δ 2 8Δ + ) y n =0. Solution. We have 2(r ) 2 8(r ) + = 0 which yields, r = 7 6. r 2 = 2 and y n =( 7 6 )n, y n2 =( 2 )n. Example. solutions Solve the following difference equation and find the fundamental (5Δ + 6) (2Δ 2 + 56Δ + 25) y n =0. Solution. The roots of the corresponding characteristic equation are r =,r 5 2, = ( ± i) in the result the fundamental solutions will be written as 8 follows y n =( 5 )n, y n2 =2 5 nπ 2 cos, y n =2 5 nπ 2 sin 4 4. Example4. Evaluate the fundamental solutions of the following DE ( 6 6E 2 4 +9E 4 2 4E 6 ) y n =0. Solution. The roots of the characteristic equation are r = r 2 =2, r =, r 4 = r 5 =, r 6 = 0 so the fundamental solutions may be written as y n =2n, y n2 = n2 n,y n =( ) n,y n4 = n,y n5 = n n,y n6 =0. Lemma. Prove the accuracy of the following equalities. n Δ f j = f n. (0) j= Δ f n n = f j. () j= Proof. Proof is easy, consider n Δ f j = j= n n f j f j = f n. j= j=

Forward r-difference operator 96 Identity () is the inversion of (0). Remark 2. Each of the above two identities are used for finding particular solution of the nonhomogeneous difference equations with constant coefficients, therefore we can solve each of the following equations Δy n = f n, Δ m y n = f n. Example 5. Find the particular solution of the following difference equation Δy n = n +2n 2 +n +. Solution. We can write y p = n Δ (n +2n 2 +n +) = (j +2j 2 +j +) = 2 n(n +2n 2 +9n 2). j= Example 6. Find the particular solution of the following difference equation Δ y n = 20n +60. Solution. By division operation we can write y p = Δ (20n + 60) = Δ 2 ( Δ (20n + 60)) = Δ ( n (20j + 60)) = 2 Δ ( Δ (60n2 ) j= = n Δ ( 60j 2 )= Δ j= (0n(n )(2n )) n =0 (2j +j 2 + j) =5n(n ) 2 (n 2). j=

962 H. Hosseinzadeh and G. A. Afrouzi Main Results.. forward r-difference operators and solution of nonhomogeneous difference equations Definition. We define the forward r-difference operator Δ r as follows Δ r y n = y n+ ry n =(E r)y n, where y n is the approximation of function y(x) at point x n [x 0, x m ], then two operators Δ r and E r are equivalent. Corollary. Δ r is linear operator and Δ =Δ=E. Example 7. r (r n y n )=r n y n rr n y n = r n (y n y n )=r n y n 4 (4 n cos nπ 2 )=4n cos nπ 2 =4n (cos nπ 2 Example 7. cos nπ 2 ). (i) Δ 5 (5 n (n 2 6n + 0)) = 5 n [(n 2 6n + 0) ((n ) 2 6(n ) + 0)] =5 2n ( 4n 2 4n + 45). (ii) Δ ( n cos nπ )=n+ cos (n+)π n cos nπ (cos nπ + sin nπ ). = n+ 2 Four principal operations in vector space of the operator r: we define (i) Δ r +Δ s 2Δ r+s. 2 (ii) Δ r Δ s s r. (iii) Δ r Δ s E 2 (r + s)e + rs Δ r+s 2 Δ (iv) r Δ s Δ r Δ s Δ s Δ r. Particular type. Suppose that s = r, then (r+s)2 4. Δ r Δ r Δ 2 r,..., Δ r Δ m r Δ m+. In the sequel, we define order and inversion for the backward r-difference operator. Δ r s.t f n = g n Δ r g n = f n, Δ r Δ r Δ 2 r Δ r(δ r ),..., Δ m+ r (Δ m r ).

Forward r-difference operator 96 Remark. Addition operation and multiplication operation have properties such as commutative and associative, namely (Δ r +Δ s )+Δ t Δ r +(Δ s +Δ t ) Δ r+s+t, Δ r (Δ s Δ t ) (Δ r Δ s )Δ t. Theorem 4. The forward r-difference is linear operator, in addition to every order of it and every polynomial of Δ r and inversion Δ r are linear too. Proof. It is easy and left to the readers. Lemma2. Prove that n Δ r ( r n j y j = y n. (2) j=0 Δ r y n =( n r n j y j. () j=0 Proof. Is easy, above two formulas are used in finding solution of NDE with constant coefficients. Remark 4. Under the forward r-difference operator Δ r, the nonhomogeneous difference equation is may be written as follows m ( Δ rj )y n = f n. (4) j= Whereas r j,j=, 2,..., m can be real distinct, iterated or complex number. Some useful results (i) Δ r (a n )=a n (a r) Δ r a n = an a r In general (ii) Δ k r (an )=a n (a r) k Δ k r a n = an. (a r) k (iii) Δ r (na n )=(n +)a n+ rna n = na n (a r)+a n+ Δ r (na n )= n(a r) a (a r) 2. (iv) P (Δ r )a n = P (a r)a n P (Δ r) an = Lemma. Let m be integer and a r then an. P (a r)

964 H. Hosseinzadeh and G. A. Afrouzi Δ m r (an y n )=a n (aδ r a )m y n. (5) Δ m r (a n y n )=a n y Δ m n. (6) r a Proof. Equality (5) is proved by the mathematical induction, (6) is the inversion of (5). Let a = r, then last equalities will be as follows, respec- Particular type 2. tively: Δ m r (r n y n )=a n (rδ) m y n. (7) Δ m r (r n y n )=r n (rδ) m y n. (8) Example 8. Evaluate Δ 2 (2 n n). Solution. Δ 2 (2 n n)=2 n Δ (n) =2 n Δ 2 n j=0 j =2n Δ ( n j=0 (j2 + j)) = 2n 4 n j=0 (j +j 2 +2j) = 2n 6 n(n )(n 2)(n ). Example9. Find the particular solution of Δ 2 y n = n sin( nπ ). Solution.. Write y p = Δ 2 ( n sin nπ ) and use (8), thus y p = n 2 Δ 2 (sin nπ )=n 2 Δ ( 2 cos (2n )π 6 n = n 2 ( 2 n cos j=0 (2j )π 6 = n 2 ( n)+n 2 2 + 2 2 (2n )π sin. 6 Theorem 5 (shift exponential). Let P be a polynomial, then

Forward r-difference operator 965 P (Δ r )(a n y n )=a n P (aδ r a )y n. (9) P (Δ r ) (an y n )=a n P (aδ r )y n. (20) a Proof. By using Lemma 2, the proof is easy. Particular type. Suppose that a=r, then the former equalities may be written in the following forms P (Δ r )(r n y n )=r n P (rδ)y n. (2) P (Δ r ) (rn y n )=r n P (rδ) y n. (22) Example 0. Find the particular solution of the following DE (E 4 0E +5E 2 50E + 24)y n =2 n (8n + 2). Solution. This equation may be written as follows : (E )(E 2)(E )(E 4)y n = E 4 ΔΔ 2 Δ Δ 4 y n =(+Δ 2 )Δ 2 ( 2+ 2 )( + 2 ) y n =(8n + 2)2 n. Now after dividing two sides of equality equality by coefficient of y n, using the formula (22), we have y p = ( + Δ 2 )Δ 2 ( +Δ 2 )( 2+ 2 ) (n +4)2n =2 n ( + 2 )2 ( +2 )( 2+2 ) (8n + 2) =2 n 4 (Δ + 2 )Δ(Δ 2 )(Δ )(8n + 2) = 2n (n +4).

966 H. Hosseinzadeh and G. A. Afrouzi.2. Solution of NDE with constant coefficients We know that every nonhomogeneous difference equation with order m can be written in the form (4). Therefore each of the following forms may be written in the form in (4). (4) is written as follows. P (E)y n = f n, P(Δ)y n = f n, P( )y n = f n. y p = m j= Δ f n = ( (... f n...)). (2) r j Δ rm Δ rm Δ r Example. Find the particular solution of the following NDE Δ 2 Δ 4 y n = n 2 2 n. Solution. Write y p = (n 2 2 n )=2 n 2 Δ 2 Δ 4 Δ 2 Δ (n2 )=2 n 2 n(n + )(2n +) Δ 2 6 == n 2n 2 n j (2j +j 2 +j) = 2n (2n +9n 2 +25n+...). j=0 Remark 5. By using identity (2), we may use iterative divisions, in addition to, we can use the decomposition fraction, consider r Δ r Δ r2 ( ) r r 2 Δ r Δ r2 Δ r Δ r2 Δ r (r 2 r )(r 2 r ) (r r 2 )(r r ) Δ r2 + Δ r + (r r )(r r 2 ) Δ r. In general where, m j= Δr j m j= A j Δr j

Forward r-difference operator 967 m A j =,A j = j= m j i= (r j r i ). (24) Example 2. Find the particular solution of the following NDE (6 6E +E 2 E ) y n =6 n (6n2 48n + 80). Solution. Write (6E 6E 2 +E ) y n =6 n (6n 2 2n 0). y p = (2E )(E )(6E ) (6n (6n 2 2n 0)) = 6Δ 2 Δ Δ 6 (6 n (6n 2 2n 0))6 n+ ΔΔ 2 Δ (6n 2 2n 0) 2 =6 n+ ( Δ + Δ 2 2 Δ )(6n 2 2n 0) = 6 n+ n. Remark 6. If r = r 2 =... = r m = r, then we can use only iterative divisions. Remark 7. Let P (Δ r ) y n = a n f n, then by change of variable y n = Y n a n and using the shift exponential formula, we obtain the following NDE. P (aδ r a )Y n = f n. (25) Remark 8. (i) If P (a r) 0, then Let P (Δ r )y n = a n (a ), y p = P (a r) an (ii) If P (a r) =P (a r) =... = P (k ) (a r) =0,P (k) (a r) 0, then y p = P ( k)(a r) (nk a n k ).

968 H. Hosseinzadeh and G. A. Afrouzi 4 Concluding. The forward r-difference operator method is a new method in finding solution of the nonhomogeneous difference equations with constant coefficients and we can use it in finding solution of the following equation P (Δ) y n = f n. where a j s are constants. P (Δ) m a j Δ j, j=0 5 References [] Hildel brand, F.B., Introduction to Numerical Analysis, Mc. Graw- Hill, New.York, 956. [2] Isaacson, E. and H.B., Analysis of Numerical Methods, John Wiley and Sons,.New York, 966. [] Lambert, J.D. Computation methods in ordinary Differential Equations John.Wiley and Sons, New York, 97. [4] Phillips, G.M., Taylor P.J. Theory and Applications of Numberical Analysis, Fifth. Edition, Academic press, 980. [5] Ralston, A. A first course in Numerical Analysis Mc. Graw-Hill, New York. Received: November 8, 2006