Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống
1
/ 22 trang
THÔNG TIN TÀI LIỆU
Thông tin cơ bản
Định dạng
Số trang
22
Dung lượng
561,52 KB
Nội dung
Hindawi Publishing Corporation Advances in Difference Equations Volume 2008, Article ID 932831, 22 pages doi:10.1155/2008/932831 Research ArticleAsymptoticRepresentationoftheSolutionsofLinearVolterraDifference Equations Istv ´ an Gy ˝ ori and L ´ aszl ´ oHorv ´ ath Department of Mathematics and Computing, University of Pannonia, 8200 Veszpr ´ em, Egyetem u. 10, Hungary Correspondence should be addressed to Istv ´ an Gy ˝ ori, gyori@almos.vein.hu Received 26 February 2008; Accepted 4 April 2008 Recommended by Elena Braverman This article analyses theasymptotic behaviour ofsolutionsoflinearVolterra difference equations. Some sufficient conditions are presented under which thesolutions to a general linear equation converge to limits, which are given by a limit formula. This result is then used to obtain the exact asymptoticrepresentationofthesolutionsof a class of convolution scalar difference equations, which have real characteristic roots. We give examples showing the accuracy of our results. Copyright q 2008 I. Gy ˝ ori and L. Horv ´ ath. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 1. Introduction The literature on theasymptotic theory ofthesolutionsofVolterra difference equations is extensive, and application of this theory is rapidly increasing to various fields. For the basic theory of difference equations, we choose to refer to the books by Agarwal 1, Elaydi 2,and Kelley and Peterson 3. Recent contribution to theasymptotic theory of difference equations is given in the papers by Kolmanovskii et al. 4,Medina5,MedinaandGil6, and Song and Baker 7; see 8–19 for related results. The results obtained in this paper are motivated by the results of two papers by Applelby et al. 20, and Philos and Purnaras 21. This paper studies theasymptotic constancy ofthe solution ofthe system of nonconvolution Volterra difference equation zn 1 n i0 Hn, izihn,n∈ Z , 1.1 2 Advances in Difference Equations with the initial condition z0z 0 , 1.2 where z 0 ∈ R d , Hn, i 0≤i≤n and hn n≥0 are sequences with elements in R d×d and R d , respectively. Under appropriate assumptions, it is proved that the solution converges to a finite limit which obeys a limit formula. Our paper develops further the recent w ork 20. The distinction between the works is explained as follows. For large enough n,infactn ≥ 2m 2, the sum in 1.1 can be split into three terms m i0 Hn, izi n−m−1 im1 Hn, izi m j0 Hn, n − jzn − j, 1.3 since n in−m Hn, izi m j0 Hn, n − jzn − j. 1.4 In 20, Theorem 3.1 the middle sum in 1.3 contributed nothing to the limit lim n→∞ zn, since it was assumed that lim m→∞ lim sup n→∞ n−m im Hn, i O. 1.5 In our case, we split the sum in 1.1 only into two terms, and the condition 1.5 is not assumed. In fact, we show an example in Section 4,where1.5 does not hold and hence in 20, Theorem 3.1 is not applicable. At the same time our main theorem gives a limit formula. It is also interesting to note that our proof is simpler than it was applied in 20. Once our main result for, the general equation, 1.1 has been proven, we may use it for the scalar convolution Volterra difference equation with infinite delay, ΔxnAxn n j−∞ Kn − jxjgn,n∈ Z , 1.6 with the initial condition, xnϕn,n∈ Z − , 1.7 where A ∈ R,andKn n≥0 , gn n≥0 and ϕn n≤0 are real sequences. Here, Δ denotes the forward difference operator to be defined as usual, that is, Δxn : xn 1 − xn,n∈ Z . If we look for a solution xnλ n 0 ,n∈ Z λ 0 ∈ R \{0} ofthe homogeneous equation associated with 1.6, we see that λ 0 is a root ofthe characteristic equation λ 0 1 A ∞ i0 Kiλ −i 0 . 1.8 I. Gy ˝ ori and L. Horv ´ ath 3 We immediately observe that λ 0 ∈ R is a simple root if λ 0 −1 ∞ i1 i Ki λ 0 −i < 1. 1.9 In the paper 21see also 22,itisshownthatifλ 0 > 0satisfies1.8 and 1.9,and the initial sequence ϕn n≤0 is suitable, then for the solution x of 1.6 and 1.7 the sequence zn : λ −n 0 xn, n ∈ Z is bounded. Furthermore, some extra conditions guarantee that the limit z∞ : lim n→∞ zn is finite and satisfies a limit formula. In our paper, we improve considerably the result in 21. First, we give explicit necessary and sufficient conditions for the existence of a λ 0 ∈ R\{0} for which 1.8 and 1.9 are satisfied. Second, we prove the existence ofthe limit z∞ and give a limit formula for z∞ under the condition o nly λ 0 / 0. These two statements are formulated in our second main theorem stated in Section 3. The proof ofthe existence of z∞ is based on our first main result. Thearticle is organized as follows. In Section 2, we briefly explain some notation and definitions which are used to state and to prove our results. In Section 3, we state our two main results, whose proofs are relegated to Section 5. Our theory is illustrated by examples in Section 4, including an interesting nonconvolu- tion equation. This example shows the significance ofthe middle sum in 1.3, since only this term contributes to the limit ofthe solution of 1.1 in this case. 2. Mathematical preliminaries In this section, we briefly explain some notation and well-known mathematical facts which are used in this paper. Let Z be the set of integers, Z : {n ∈ Z | n ≥ 0} and Z − : {n ∈ Z | n ≤ 0}. R d stands for the set of all d-dimensional column vectors with real components and R d×d is the space of all d by d real matrices. The zero matrix in R d×d is denoted by O, and the identity matrix by I.LetE be the matrix in R d×d whose elements are all 1. The absolute value ofthe vector x x 1 , ,x d T ∈ R d and the matrix A A ij 1≤i,j≤d ∈ R d×d is defined by |x| :|x 1 |, ,|x d | T and |A| :|A ij | 1≤i,j≤d , respectively. The vector x and the matrix A is nonnegative if x i ≥ 0and A ij ≥ 0, 1 ≤ i, j ≤ d, respectively. In this case, we write x ≥ 0andA ≥ O. R d can be endowed with any norms, but they are equivalent. A vector norm is denoted by · and the norm of a matrix in R d×d induced by this vector norm is also denoted by ·. The spectral radius ofthe matrix A ∈ R d×d is given by ρA : lim n→∞ A n 1/n , which is independent ofthe norm employed to calculate it. A partial ordering is defined on R d R d×d by letting x ≤ yA ≤ B ifandonlyify − x ≥ 0B − A ≥ O. The partial ordering enables us to define the sup, inf, lim sup, lim inf, and so forth for the sequences of vectors and matrices, which can also be determined componentwise and elementwise, respectively. It is known that ρA ≤ ρ|A| for A ∈ R d×d ,andρA ≤ ρB if A, B ∈ R d×d and O ≤ A ≤ B. 3. The main results First, consider the nonconvolutional linearVolterra difference equation zn 1 n i0 Hn, izihn,n∈ Z , 3.1 4 Advances in Difference Equations with initial condition z0z 0 . 3.2 Here, we assume H 1 z 0 ∈ R d ,H:Hn, i 0≤i≤n and h :hn n≥0 are sequences with elements in R d×d and R d , respectively; H 2 for any fixed i ≥ 0 the limit H ∞ i : lim n→∞ Hn, i is finite and ∞ i0 |H ∞ i| < ∞; H 3 the matrix V : lim N→∞ lim n→∞ n jN Hn, j 3.3 is finite; H 4 the matrix W : lim N→∞ lim sup n→∞ n jN Hn, j 3.4 is finite and ρW < 1; H 5 the limit h∞ : lim n→∞ hn is finite. By a solution of 3.1, we mean a sequence z :zn n≥0 in R d satisfying 3.1 for any n ∈ Z . It is clear that 3.1 with initial condition 3.2 has a unique solution. Now, we are in a position to state our first main result. Theorem 3.1. Assume (H 1 )–(H 5 ) are satisfied. Then for any z 0 ∈ R d the unique solution z : zn n≥0 of 3.1 and 3.2 has a finite limit at ∞ and it satisfies lim n→∞ znI − V −1 ∞ i0 H ∞ izih∞ . 3.5 Under conditions H 3 and H 4 |V | lim N→∞ lim n→∞ n jN Hn, j ≤ W, 3.6 and hence ρW < 1 yields ρV < 1, thus I − V is invertible. On the other hand under our conditions the unique solution z :zn n≥0 of 3.1 and 3.2 is a bounded sequence, therefore ∞ i0 H ∞ izi is finite, and 3.5 makes sense. The second main result is dealing with the scalar Volterra difference equation ΔxnAxn n j−∞ Kn − jxjgn,n∈ Z , 3.7 I. Gy ˝ ori and L. Horv ´ ath 5 with the initial condition xnϕn,n∈ Z − , 3.8 where A ∈ R,K: Z →R,g: Z →R,andϕ : Z − →R are given. By a solution oftheVolterra difference equation 3.7 we mean a sequence x : Z→R satisfies 3.7 for any n ∈ Z . In what follows, by S we will denote the set of all initial sequence ϕ : Z − →R such that for each n ∈ Z −1 j−∞ Kn − jϕj3.9 exists. It can be easily seen that for any initial sequence ϕ ∈ S, 3.7 has exactly one solution satisfying 3.8.Thisuniquesolutionisdenotedbyxϕ : Z→R and it is called the solution ofthe initial value problem 3.7, 3.8. Theasymptoticrepresentationofthesolutionsof 3.7 is given under the next condition. A There exists a λ 0 ∈ R \{0} for which λ 0 1 A ∞ i0 Kiλ −i 0 , 3.10 λ 0 −1 ∞ i1 i Ki λ 0 −i < 1. 3.11 From the theory ofthe infinite series, one can easily see that condition A yields r : lim sup n→∞ Kn 1/n 3.12 is finite. Moreover, the mapping G : r, ∞→0, ∞, defined by Gμ : ⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩ ∞ i1 i Ki μ −i1 if μ>r, ∞ i1 i Ki r −i1 if μ r>0, ∞ if μ r 0, 3.13 is real valued on r, ∞. It is also clear see Section 5 that if there is an n 0 ≥ 1 such that Kn 0 / 0, and if Gr ≥ 1 then the equation Gμ1 3.14 has a unique solution, say μ 1 . Now we formulate the following more explicit condition: B either Kn0, n ≥ 1, and 1 A K0 / 0, 3.15 6 Advances in Difference Equations or there is an n 0 ≥ 1withKn 0 / 0, and i r defined in 3.12 is finite, ii if Gr ≥ 1, then the constant A satisfies either A>μ 1 − 1 − ∞ i0 Kiμ −i 1 3.16 or A<−μ 1 − 1 − ∞ i0 −1 i Kiμ −i 1 , 3.17 iii if Gr < 1, then the constant A satisfies either A ≥ r − 1 − ∞ i0 Kir −i 3.18 or A ≤−r − 1 − ∞ i0 −1 i Kir −i . 3.19 Remark 3.2. Let K : Z →R be a sequence such that Kn 0 / 0 for some n 0 ≥ 1. It will be proved in Lemma 5.7 that there is at most one λ 0 ∈ C \{0} satisfying 3.10 and 3.11.Itisaneasy consequence of this statement that if λ 0 ∈ C \{0} satisfies 3.10 and 3.11,andλ 1 ∈ C \{0,λ 0 } is a solution of 3.10,then|λ 1 | < |λ 0 |,thusλ 0 is the leading root of 3.10.Really,fromthe condition |λ 1 |≥|λ 0 | we have ∞ i1 i Ki λ 1 −i1 ≤ ∞ i1 i Ki λ 0 −i1 < 1, 3.20 that is 3.11 holds for λ 1 instead of λ 0 , and this contradicts the uniqueness of λ 0 . Now, we are ready to state our second result which will be proved in Section 5.This result shows that the implicit condition A and the explicit condition B are equivalent and thesolutionsof 3.7 can be asymptotically characterized by λ n 0 as n→∞. Theorem 3.3. Let A ∈ R,K: Z →R,g: Z →R,andϕ ∈ S be given. Then α Condition (A) holds if and only if condition (B) is satisfied. β If condition (A) or equivalently condition (B) holds, moreover ∞ i0 λ −i 0 −1 j−∞ Ki − jϕjgi 3.21 is finite, then for the solution xϕ of 3.7, 3.8 the limit cϕ : lim n→∞ λ −n 0 xϕn is finite and it obeys cϕ λ 0 − ∞ i1 iKiλ −i 0 −1 ∞ i0 λ −i 0 −1 j−∞ Ki − jϕjgi λ 0 ϕ0 . 3.22 I. Gy ˝ ori and L. Horv ´ ath 7 4. Examples and the discussion ofthe results In this section, we illustrate our results by examples and the interested reader could also find some discussions. Example 4.1. Our Theorem 3.1 is given for system of equations, however the next example shows that this result is also new even in scalar case. Let us consider the scalar nonconvolution Volterra difference equation zn 1 n j1 n − j α−1 j β−1 n αβ−1 zjhn,n≥ 1, 4.1 with the initial condition z1z 1 , 4.2 where a 1 α, β > 1 and real, and hn n≥1 is a real sequence such that its limit h∞ : lim n→∞ hn is finite. Now, let the values z 0 ,h0 and the sequence H :Hn, i 0≤i≤n be defined by z 0 : 0,h0 : z 1 , Hn, i : ⎧ ⎪ ⎨ ⎪ ⎩ n − i α−1 i β−1 n αβ−1 , 1 ≤ i ≤ n, 0, 0 i ≤ n. 4.3 Then, it can be easily seen that problem 4.1, 4.2 is equivalent to problem 3.1, 3.2. We find that H ∞ i : lim n→∞ Hn, i0 for any fixed i ≥ 0. It is known that lim n→∞ n j1 Hn, jlim n→∞ n j1 n − j α−1 j β−1 n αβ−1 Bα, β, 4.4 where Bα, β is the well-known Beta function at α, β defined by Bα, β : 1 0 t α−1 1 − t β−1 dt < 1. 4.5 Using the nonnegativity of Hn, i, 0 ≤ i ≤ n,andLemma 5.2 we have that O ≤ V W lim N→∞ lim n→∞ n jN Hn, j lim n→∞ n j0 Hn, jBα, β < 1. 4.6 Now, one can easily see that for the sequences h :hn n≥0 and H :Hn, i 0≤i≤n all ofthe conditions of Theorem 3.1 are satisfied. Thus, by Theorem 3.1 wegetthatthesolution z :zn n≥1 ofthe initial value problem 4.1, 4.2 satisfies lim n→∞ zn h∞ 1 − Bα, β . 4.7 On the other hand, w e know see 23 that lim N→∞ lim n→∞ n−N jN Hn, j lim N→∞ lim n→∞ n−N jN n − j α−1 j β−1 n αβ−1 Bα, β / 0, 4.8 and hence in 20, Theorem 3.1 is not applicable. 8 Advances in Difference Equations Example 4.2. Let m ≥ 1and0 <τ 1 < ··· <τ m be given integers, and assume Ki / 0ifi ∈ {τ 1 , ,τ m } and Ki0ifi ∈ Z \{τ 1 , ,τ m }. Then, n i−∞ Kn − ixi ∞ i0 Kixn − i m k1 Kτ k xn − τ k ,n∈ Z . 4.9 Thus, 3.7 reduces to the delay difference equation ΔxnAxn m k1 Kτ k xn − τ k gn,n∈ Z , 4.10 and for any sequence ϕ : Z − →R,ϕ∈ S holds. Since Kn0 for any large enough n, r : lim n→∞ |Kn| 1/n 0, moreover the function G : 0, ∞→0, ∞ defined in 3.13 satisfies Gμ ⎧ ⎪ ⎨ ⎪ ⎩ m k1 τ k Kτ k μ −τ k 1 , if μ>0, ∞, if μ 0. 4.11 Let μ 1 > 0 be the unique value satisfying Gμ 1 m k1 τ k Kτ k μ −τ k 1 1 1. 4.12 Now, statementα of Theorem 3.3 is applicable and so the next statement is valid. Proposition 4.3. For an A ∈ R thereisaλ 0 ∈ R \{0} such that λ 0 A 1 m k1 K τ k λ −τ k 0 , λ 0 −1 m k1 τ k K τ k λ 0 −τ k < 1 4.13 hold if and only if either A>μ 1 − 1 − m k1 K τ k μ −τ k 1 4.14 or A<−μ 1 − 1 − m k1 −1 τ k Kτ k μ −τ k 1 4.15 is satisfied. I. Gy ˝ ori and L. Horv ´ ath 9 Now, let m 1,τ 1 : l ∈{1, } and Kl / 0. Then μ 1 l|Kl| 1/l1 , moreover 4.14 and 4.15 reduce to A> l Kl 1/l1 − 1 − Kl l Kl −l/l1 , A<− l Kl 1/l1 − 1 − −1 l Kl l Kl −l/l1 . 4.16 If especially l 1andK1 ≥ 0, then μ 1 K1, moreover 4.14 and 4.15 are equivalent to the condition A / K1 − 1 − K1− K1 − 1 K1−1. 4.17 Example 4.4. Let q ∈ R \{0} and Ki : q i ,i∈ Z . Then, 3.7 has the following form: ΔxnAxn n j−∞ q n−j xjgn,n∈ Z . 4.18 It is clear that r : lim n→∞ |q n | 1/n |q|, and the function G defined in 3.13 is given by Gμ ⎧ ⎪ ⎨ ⎪ ⎩ |q| μ −|q| 2 , if μ>|q|, ∞, if μ |q|, 4.19 moreover μ 1 |q| |q| is the unique positive root of Gμ1. Thus statement α in Theorem 3.3 is applicable and as a corollary of it we obtain the following. Proposition 4.5. Thereisaλ 0 ∈ R \{0} such that 3.10 and 3.11 hold with the sequence Kiq i , i ∈ Z , if and only if either A>|q| |q|−1 − |q| |q| |q| |q|−q 4.20 or A<− |q| |q| − 1 − |q| |q| |q| |q| q . 4.21 Example 4.6. Let c ∈ 1, ∞ \ Z and let Ki : c i , i ∈ Z .Here c i is the extended binomial coefficient, that is c i : cc − 1 ··· c − i − 1 i! ,i∈ Z . 4.22 In this case, r lim n→∞ c n 1/n 1 and by using the well-known properties ofthe binomial series, we find Gμ ∞ i1 i c i μ −i1 1 μ 2 c 1 1 μ c−1 , if μ ≥ 1. 4.23 Thus, GrG1c2 c−1 > 1, therefore by statement α of Theorem 3.3 we get the following. 10 Advances in Difference Equations Proposition 4.7. Thereisaλ 0 ∈ R \{0} such that 3.10 and 3.11 hold with the sequence Ki c i ,i∈ Z , if and only if either A>μ 1 − 1 − 1 1 μ 1 c 4.24 or A<−μ 1 − 1 − 1 − 1 μ 1 c , 4.25 where μ 1 is the unique positive solution ofthe equation 1 μ 2 c 1 1 μ c−1 1. 4.26 Example 4.8. Let α>3andKi : 1/2i α ,i≥ 1, and K0 : 0. Then, 3.7 reduces to the special form ΔxnAxn n−1 j−∞ 1 2n − j α xjgn,n∈ Z . 4.27 It is not difficult to see that r 1/2n α 1/n 1, Gμ ∞ i1 1 2i α−1 μ −i1 ,μ≥ 1, 4.28 and G1 < 1. From statement α of Theorem 3.3 we have the following. Proposition 4.9. Thereisaλ 0 ∈ R \{0} such that 3.10 and 3.11 hold with the sequence Ki 1/2i α ,i≥ 1, if and only if either A> ∞ i1 1 2i α 1 2 ςα4.29 or A<−2 − ∞ i1 −1 i 2i α −2 − 1 2 α−1 − 1 ∞ i1 1 2i α −2 − 1 2 α−1 − 1 ςα, 4.30 where ς is the well-known Riemann function. 5. Proofs ofthe main theorems 5.1. Proof of Theorem 3.1 To prove Theorem 3.1 we need the next result from 20. [...]... Moreover, the existence of a λ0 ∈ R \ {0} satisfying 3.11 is equivalent to either |λ0 | > μ1 in case G r ≥ 1 or |λ0 | ≥ r in case G r < 1 Now, the result follows from the properties ofthe functions Fi i 1, 2 The proof of a is complete b In virtue of Lemma 5.5 the proof ofthe theorem will be complete if we show that the sequences H and h satisfy the conditions H2 – H5 in Section 3 Since the series... , 5.42 and hence the proof of Theorem 3.1 is complete 5.2 Proof of Theorem 3.3 Theorem 3.3 will be proved after some preparatory lemmas In the next lemma, we show that 3.7 can be transformed into an equation ofthe form 3.1 by using the transformation λ−n x ϕ n , 0 z n n∈Z 5.43 Lemma 5.5 Under the conditions of Theorem 3.3, the sequence z : Z →R defined by 5.43 satisfies 3.1 , where the sequences H :... finite, then by 5.19 n V : lim n→∞ ∞ H n, j − j 0 H∞ j 5.26 j 0 satisfies 5.11 ρ V < 1 follows from H4 The proof is now complete Lemma 5.3 The hypotheses of Theorem 3.1 imply that ∞ −1 I −V cz : H∞ j z j h∞ 5.27 h∞ 5.28 j 0 is the only vector satisfying the equation ∞ cz H∞ j z j V cz j 0 Proof Since ρ V ≤ ρ W < 1 the matrix I − V is invertible, which shows the uniqueness part ofthe lemma On the other... for existence of nontrivial invariant cones of nonnegative initial values of o difference equations,” Applied Mathematics and Computation, vol 36, no 2, pp 89–111, 1990 17 V Kolmanovskii and L Shaikhet, “Some conditions for boundedness ofsolutionsof difference Volterra equations,” Applied Mathematics Letters, vol 16, no 6, pp 857–862, 2003 18 Z H Li, Theasymptotic estimates ofsolutionsof difference... hypotheses of Theorem 3.1 imply that the hypotheses of Theorem A are satisfied, and hence the solution z : z n n≥0 of 3.1 , 3.2 is bounded Proof Let > 0 be such that ρ W is an M0 ≥ 0 for which E < 1 This can be satisfied because ρ W < 1 Then, there n H n, j lim sup n→∞ 0 5.58 i 1 is convergent on r, ∞ and it is divergent on 0, r b limμ→∞ G μ 0 c limμ→r G μ Gr d G is strictly decreasing e If G r ≥ 1, then the equation G μ 1 has a unique solution Proof a The root test can be applied b The series of functions 5.58 is uniformly . provided the original work is properly cited. 1. Introduction The literature on the asymptotic theory of the solutions of Volterra difference equations is extensive, and application of this theory. : Z→R and it is called the solution of the initial value problem 3.7, 3.8. The asymptotic representation of the solutions of 3.7 is given under the next condition. A There exists a λ 0 ∈. Difference Equations Volume 2008, Article ID 932831, 22 pages doi:10.1155/2008/932831 Research Article Asymptotic Representation of the Solutions of Linear Volterra Difference Equations Istv ´ an