Show Summary Details
More options …

# Open Mathematics

### formerly Central European Journal of Mathematics

Editor-in-Chief: Gianazza, Ugo / Vespri, Vincenzo

IMPACT FACTOR 2018: 0.726
5-year IMPACT FACTOR: 0.869

CiteScore 2018: 0.90

SCImago Journal Rank (SJR) 2018: 0.323
Source Normalized Impact per Paper (SNIP) 2018: 0.821

Mathematical Citation Quotient (MCQ) 2017: 0.32

ICV 2017: 161.82

Open Access
Online
ISSN
2391-5455
See all formats and pricing
More options …
Volume 16, Issue 1

# Periodic and subharmonic solutions for a 2nth-order p-Laplacian difference equation containing both advances and retardations

Peng Mei
/ Zhan Zhou
Published Online: 2018-12-29 | DOI: https://doi.org/10.1515/math-2018-0123

## Abstract

We consider a 2nth-order nonlinear difference equation containing both many advances and retardations with p-Laplacian. Using the critical point theory, we obtain some new explicit criteria for the existence and multiplicity of periodic and subharmonic solutions. Our results generalize and improve some known related ones.

MSC 2010: 39A23

## 1 Introduction

Let N, Z and R denote the sets of all natural numbers, integers and real numbers, respectively. For abZ, define Z(a) = {a, a + 1, ⋯}, Z(a, b) = {a, a + 1, ⋯, b}.

Consider the following 2nth-order nonlinear difference equation

$Δnrk−nφpΔnuk−n=(−1)nf(k,uk+τ,⋯,uk+1,uk,uk−1,⋯,uk−τ),k∈Z,$(1.1)

where τ, nZ(1), Δ is the forward difference operator defined by Δuk = uk+1-uk, Δ2 uk = Δ(Δ uk), rk > 0 is real valued for each kZ, rk and f(k, vτ, ⋯, vτ) are T-periodic in k for a given positive integer T, and fC(Z × R2τ+1, R). φp(s) is the p-Laplacian operator given by φp(s) = |s|p–2s (1 < p < ∞). For any integer m ⩾ 2, a solution to Eq. (1.1) is called a mth-order subharmonic solution if it is a mT-periodic solution.

Earlier, the main methods were all kinds of fixed point theorems in cones for the study of periodic solutions and boundary value problems of difference equations. It was not until 2003 that the critical point theory was used to establish sufficient conditions for the existence of periodic solutions. Guo and Yu [1, 2] first established sufficient conditions on the existence of periodic solutions of second-order nonlinear difference equations by using the critical point theory.

In 2007, Cai and Yu [3] obtained some criteria for the existence of periodic solutions of a 2nth-order difference equation

$Δnrk−nΔnuk−n+f(k,uk)=0,n∈Z(3),k∈Z,$(1.2)

where f grows superlinearly at both 0 and ∞. Using Linking Theorem and Saddle Point Theorem, Zhou [4] improved the results of [3] and extended f in Eq. (1.2) into sublinear or asymptotically linear.

In fact, there are some papers which studied the periodic solutions of difference equations involving both advances and retardations which have important background and meaning in the field of cybernetics and biological mathematics. Chen and Fang [5] in 2007 considered the following second-order nonlinear difference equation containing both advance and retardation with p-Laplacian

$ΔφpΔxn−1+f(n,xn+1,xn,xn−1)=0,n∈Z.$(1.3)

In 2014, Lin and Zhou [6] obtained some new sufficient conditions on the existence and multiplicity of periodic solutions of the following ϕ-Laplacian difference equation

$Δnrk−nϕΔnuk−1=(−1)nf(k,uk+1,uk,uk−1),k∈Z.$(1.4)

In the past literature, when it comes to a special Eq. (1.1) that τ = 1, many excellent works have been done (e.g. see [5, 6, 7, 8, 9]). Using the critical point theory, they obtained some sufficient conditions on the existence and multiplicity of periodic solutions in the special case of Eq. (1.1).

Nevertheless, to the best of our knowledge, the results on periodic solutions of nonlinear difference equations containing both many advances and retardations with p-Laplacian are very scare. To fill this gap, this paper gives some sufficient conditions for the existence and multiplicity of periodic and subharmonic solutions to Eq. (1.1).

We mention that, in recent years, the critical point theory is also used on the study of homoclinic solutions [10, 11, 12, 13, 14, 15, 16, 17] and boundary value problems [18, 19, 20, 21, 22, 23] for difference equations.

Let

$r_=mink∈Z(1,T){rk},r¯=maxk∈Z(1,T){rk}.$

Our main result is as follows.

#### Theorem 1.1

Assume that the following hypotheses hold:

• (T1)

there exists a functional F(k, v0, v1, ⋯, vτ) ∈ C1(Z × Rτ+1, R) with F(k, v0, v1, ⋯, vτ)⩾0 and it satisfies

$F(k+T,v0,v1,⋯,vτ)=F(k,v0,v1,⋯,vτ),k∈Z,$

$∑j=kk+τ∂F(j,vj,vj−1,⋯,vj−τ)∂vk=f(k,vk+τ,⋯,vk+1,vk,vk−1,⋯,vk−τ);$

• (T2)

there exists a constant $\begin{array}{}\alpha \in \left(0,\frac{\underset{_}{r}}{\left(\tau +1{\right)}^{p/2}p}\left(mT{\right)}^{\frac{-|2-p|}{2}}\left(2\mathrm{sin}\frac{\pi }{mT}{\right)}^{np}\right)\end{array}$ such that

$lim supδ1→0F(k,v0,v1,⋯,vτ)δ1p⩽α,fork∈Zandδ1=∑j=0τvj2;$

• (T3)

there exists a constant $\begin{array}{}\beta \in \left(\frac{\overline{r}}{\left(\tau +1{\right)}^{p/2}p}\left(mT{\right)}^{\frac{|2-p|}{2}}\left(2\mathrm{cos}\frac{1-\left(-1{\right)}^{mT}}{2mT}\pi \right){\right)}^{np},+\mathrm{\infty }\right)\end{array}$ such that

$lim infδ1→∞F(k,v0,v1,⋯,vτ)δ1p⩾β,fork∈Zandδ1=∑j=0τvj2.$

Then for any given positive integer m, Eq. (1.1) possesses at least two mT-periodic nontrivial solutions.

It is worth pointing out that our sufficient conditions are based on the limit superior and limit inferior which are more applicable. Moreover, we also extend the conclusions to a more general form. As far as we know, in most of the previous results (e.g. see [7]), the values of c1 and c2 cannot be determined, that is not operable, but we present their specific values. In fact, if τ = 1, the assumptions in Theorem 1.1 are more explicit and easier to verify than those in Theorem 1.1 in [7]. For the sake of clarity, we put the remaining results at the end of the article. Our results complement the existing ones. See Remarks 4.6 and 4.7 for details.

The outline of this paper is as follows. In Section 2 we establish the variational framework associated with Eq. (1.1) and transfer the problem of the existence of periodic solutions of Eq. (1.1) into that of the existence of critical points of the corresponding functional. In Section 3, some related fundamental results are recalled for convenience, and some lemmas are proven. Then, we complete the proof of our main result by using Linking Theorem in Section 4. Finally, in Section 5, we illustrate our results with an example.

## 2 Variational structure

This section is to establish the corresponding variational framework for Eq. (1.1) and cite some basic conclusions for the forthcoming discussion.

Let S be the set of all two-side sequences, that is

$S={{uk}|uk∈R,k∈Z}.$

For any u, vS, a, bR, au + bv is defined by

$au+bv={auk+bvk}k=−∞+∞.$

Then S is a vector space. For any given positive integers m and T, we define the subspace EmT of S as

$EmT={u∈S|uk+mT=uk,k∈Z}.$

It is trivial to show that, EmT is isomorphic to RmT and can be endowed with the inner product

$u,v=∑j=1mTujvj,∀u,v∈EmT,$

and corresponding norm

$||u||=∑j=1mTuj212,|∀|u∈EmT.$

On the other hand, we define the norm ‖⋅‖p on EmT as follows:

$||u||p=∑j=1mT|uj|p1p,$

for all uEmT. Similarly to the derivation of [6], by Hölder inequality and Jensen inequality, we have

$||u||⩽||u||p⩽(mT)2−p2p||u||,1

Let

$c1(p)=1,1

then

$c1(p)||u||⩽||u||p⩽c2(p)||u||,∀u∈EmT,$(2.1)

and

$c1(p)c2(p)=(mT)−|2−p|2p.$

Define the functional J on EmT as

$J(u)=1p∑k=1mTrk−1Δnuk−1p−∑k=1mTF(k,uk,uk−1,⋯,uk−τ),u∈EmT.$

Clearly, JC1(EmT, R) and by the fact that u0 = umT, u1 = umT+1, after a careful computation, we can find

$∂J∂uk=(−1)nΔnrk−nφpΔnuk−n−f(k,uk+τ,⋯,uk+1,uk,uk−1,⋯,uk−τ),∀k∈Z(1,mT).$

Thus, u is a critical point of J on EmT if and only if Eq. (1.1) holds.

So, we reduce the existence of periodic solutions of Eq. (1.1) to that of the critical points of the functional J on EmT. Indeed, uEmT can be identified with u = (u1, u2, ⋯, umT), where denotes the transpose of the vector.

Let P be the corresponding mT × mT matrix to the quadratic form $\begin{array}{}\frac{\mathrm{\partial }J}{\mathrm{\partial }{u}_{k}}=\left(-1{\right)}^{n}{\mathit{\Delta }}^{n}\left({r}_{k-n}{\phi }_{p}\left({\mathit{\Delta }}^{n}{u}_{k-n}\right)\right)-f\left(k,{u}_{k+\tau },\cdots ,{u}_{k+1},{u}_{k},{u}_{k-1},\cdots ,{u}_{k-\tau }\right),\phantom{\rule{thinmathspace}{0ex}}\mathrm{\forall }\phantom{\rule{thinmathspace}{0ex}}k\in \mathbf{Z}\left(1,mT\right).\end{array}$ with uk+mT = uk for kZ, which is defined by

$P=2−10⋯0−1−12−1⋯000−12⋯00⋯⋯⋯⋯⋯⋯000⋯2−1−100⋯−12.$

By the matrix theory, we obtain that the eigenvalues of P are

$λj=4sin2⁡jπmT,j=0,1,2,⋯,mT−1.$

This implies λ0 = 0, λ1 > 0, λ2 > 0, ⋯, λmT–1 > 0. Therefore,

$λ_=min{λ1,λ2,⋯,λmT−1}=4sin2⁡πmT,λ¯=max{λ1,λ2,⋯,λmT−1}=4cos2⁡1−(−1)mT2mTπ.$

Let

$W=ker⁡P={u∈EmT|Pu=0,u∈RmT}.$

Then

$W={u∈EmT|u={c},c∈R}.$

Let V be the direct orthogonal complement of EmT to W, i.e., EmT = VW.

## 3 Some results and lemmas

For the reader’s convenience, we give some basic notations and some known results about the critical point theory.

#### Definition 3.1

Let E be a real Banach space, JC1(E, R), i.e., J is a continuously Fréchet-differentiable functional defined on E. If any sequence {u(i)E for which {J(u(i)) is bounded and J (u(i)) → 0 (i → ∞) possesses a convergent subsequence, then we say J satisfies the Palais-Smale condition (P.S. condition for short).

Let Bρ denote the open ball in E about 0 of radius ρ and let ∂ Bρ denote its boundary.

#### Lemma 3.2

(Linking Theorem [24]). Let E be a real Banach space, E = E1E2, where E1 is finite dimensional. Suppose that JC1(E, R}) satisfies the P.S. condition and

• (J1)

There exist constants a > 0 and ρ > 0 such that J| BρE2a;

• (J2)

(J2) There exists an e B1E2 and a constant R0ρ such that J| Q ⩽ 0, where Q = (R0E1)⊕{se|0 < s < R0}.

Then J possesses a critical value ca, where

$c=infh∈Γsupu∈QJ(h(u)),$

and Γ = {hC(}, E)∣ h| Q = id}, where id denotes the identity operator.

Then we prove some lemmas which are useful in the proof of Theorem 1.1. First, similarly to the derivation of [3], we can find the following lemma.

#### Lemma 3.3

Let x = (Δn–1 u1, Δn–1 u2, ⋯, Δn–1 umT). For any uEmT, one has

$λ_(n−1)p2||u||p⩽||x||p⩽λ¯(n−1)p2||u||p.$

Let $\begin{array}{}G\left(u\right)=\frac{1}{p}\sum _{k=1}^{mT}{r}_{k-1}{\left|{\mathit{\Delta }}^{n}{u}_{k-1}\right|}^{p},\end{array}$ it follows from Lemma 3.2 that

$G(u)⩽r¯p(∑k=1mTΔnuk−1p)1pp⩽r¯pc2(p)(∑k=1mTΔnuk−12)12p⩽r¯pc2p(p)λ¯p2||x||p⩽r¯pc2p(p)λ¯np2||u||p$

and

$G(u)⩾r_p(∑k=1mTΔnuk−1p)1pp⩾r_pc1(p)(∑k=1mTΔnuk−12)12p⩾r_pc1p(p)λ_p2||x||p⩾r_pc1p(p)λ_np2||u||p.$

#### Lemma 3.4

Assume that (T1) and (T3) hold. Then the functional J is bounded from above on EmT.

#### Proof

By (T3), there exist constants ζ > 0 and $\begin{array}{}{\beta }^{\prime }\in \left(\frac{\overline{r}}{\left(\tau +1{\right)}^{p/2}p}\left(mT{\right)}^{\frac{|2-p|}{2}}\left(2\mathrm{cos}\frac{1-\left(-1{\right)}^{mT}}{2mT}\pi \right){\right)}^{np},\beta \right)\end{array}$ such that

$F(k,v0,v1,⋯,vτ)⩾β′∑j=0τvj2p−ζ,∀(k,v0,v1,⋯,vτ)∈Z×Rτ+1.$

For any uEmT, by Lemma 3.3 we have

$J(u)=1p∑k=1mTrk−1Δnuk−1p−∑k=1mTF(k,uk,uk−1,⋯,uk−τ) ⩽r¯pc2p(p)λ¯np2||u||p−∑k=1mTβ′∑i=−τ0uk+i2p−ζ ⩽r¯pc2p(p)λ¯np2||u||p−β′c1p(p)∑k=1mT∑i=−τ0uk+i2p2+mTζ =r¯pc2p(p)λ¯np2||u||p−(τ+1)p2β′c1p(p)||u||p+mTζ =r¯pc2p(p)λ¯np2−(τ+1)p2β′c1p(p)||u||p+mTζ ⩽mTζ.$

The proof of Lemma 3.4 is complete. □

#### Remark 3.5

The case mT = 1 is trivial. For the case mT = 2, P has a different form, namely,

$P=2−2−22.$

However, in this special case, the argument need not to be changed and we omit it.

#### Lemma 3.6

Assume that (T1) and (T3) hold. Then the functional J satisfies the P.S. condition.

#### Proof

Let {J(u(i)}) be a bounded sequence from the lower bound, i.e., there exists a positive constant M such that

$−M⩽Ju(i),∀i∈N.$

By the proof of Lemma 3.4, it is easy to see that

$−M⩽Ju(i)⩽r¯pc2p(p)λ¯np2−(τ+1)p2β′c1p(p)|u(i)|p+mTζ,∀i∈N.$

Therefore,

$(τ+1)p2β′c1p(p)−r¯pc2p(p)λ¯np2|u(i)|p⩽M+mTζ.$

Since $\begin{array}{}{\beta }^{\prime }>\frac{\overline{r}}{\left(\tau +1{\right)}^{p/2}p}{\left(\frac{{c}_{2}\left(p\right)}{{c}_{1}\left(p\right)}\right)}^{p}{\overline{\lambda }}^{\frac{np}{2}},\end{array}$ it is not difficult to know that {u(i)} is a bounded sequence on EmT. As a consequence, {u(i)} possesses a convergence subsequence and J satisfies the P.S. condition. □

## 4 Proof of the main result

In this section, we shall prove our main results by using Linking Theorem.

Assumptions (T1) and (T2) imply that F(k, 0, ⋯, 0) = 0 and f(k, 0, ⋯, 0) = 0 for kZ. Then u = 0 is a trivial mT-periodic solution of Eq. (1.1). It suffices to prove that J has at least two nontrivial critical points on EmT.

Firstly, we show the existence of one nontrivial critical point. By Lemma 3.4, J is bounded from above on EmT. The proof of it implies $\begin{array}{}\underset{\phantom{\rule{thinmathspace}{0ex}}|u\phantom{\rule{thinmathspace}{0ex}}|\to +\mathrm{\infty }}{lim}J\left(u\right)=-\mathrm{\infty }\end{array}$. This means that –J(u) is coercive. Let we define $\begin{array}{}{c}_{0}=\underset{u\in {E}_{mT}}{sup}J\left(u\right).\end{array}$

There exists ūEmT such that J(ū) = c0 by the continuity of J(u). Clearly, ū is a critical point of J.

We claim that c0 > 0, which implies that ū is a nontrivial crtical point of J. Indeed, by (T2), there exist two positive constants δ and α′ ∈ $\begin{array}{}\left(\alpha ,\frac{\underset{_}{r}}{\left(\tau +1{\right)}^{p/2}p}\left(mT{\right)}^{\frac{-|2-p|}{2}}\left(2\mathrm{sin}\frac{\pi }{mT}{\right)}^{np}\right)\end{array}$ such that $\begin{array}{}F\left(k,{u}_{0},{u}_{1},\cdots ,{u}_{\tau }\right)⩽{\alpha }^{\prime }{\left(\sqrt{\sum _{j=0}^{\tau }{u}_{j}^{2}}\right)}^{p},\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}\mathrm{f}\mathrm{o}\mathrm{r}\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}k\in \phantom{\rule{thinmathspace}{0ex}}\mathrm{Z}\phantom{\rule{thinmathspace}{0ex}}\mathrm{a}\mathrm{n}\mathrm{d}\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}\sum _{j=0}^{\tau }{u}_{j}^{2}⩽{\delta }^{2}.\end{array}$

For any uV with ‖u‖ ⩽ δ, we have

$J(u)⩾r_pc1p(p)λ_np2−(τ+1)p2α′c2p(p)∥u∥p.$(4.1)

Taking $\begin{array}{}\sigma =\left(\frac{\underset{_}{r}}{p}{c}_{1}^{p}\left(p\right){\underset{_}{\lambda }}^{\frac{np}{2}}-\left(\tau +1{\right)}^{\frac{p}{2}}{\alpha }^{\prime }{c}_{2}^{p}\left(p\right)\right){\delta }^{p},\end{array}$ then by Eq. (4.1),

$J(u)⩾σ,∀u∈V∩∂Bδ.$

This implies that $\begin{array}{}{c}_{0}=\underset{u\in {E}_{mT}}{sup}J\left(u\right)⩾\sigma >0.\end{array}$ Hence the claim is proved. At the same time, we have proved that there exist constants σ > 0 and δ > 0 such that J|BδVσ. In the other word, J satisfies the condition (J1) of Linking Theorem.

In the remaining of the proof, we shall use Lemma 3.2 to obtain another nontrivial critical point. We have known that J satisfies the P.S. condition on EmT. In the following, we shall verify the condition (J2).

Take e B1V. For any zW and sR, we denote u = se + z. Then we have

$J(u)=1p∑k=1mTrk−1Δnuk−1p−∑k=1mTF(k,uk,uk−1,⋯,uk−τ) ⩽r¯psp∑k=1mTΔnekp−∑k=1mTF(k,sek+zk,sek−1+zk−1,⋯,sek−τ+zk−τ) ⩽r¯pspc2p(p)∑k=1mTΔnek2p2−β′c1p(p)∑k=1mT∑i=−τ0(sek+i+zk+i)2p2+mTζ ⩽r¯pspc2p(p)λ¯np2−β′c1p(p)(τ+1)∑k=1mTsek+zk2p2+mTζ ⩽r¯pc2p(p)λ¯np2sp−(τ+1)p2β′c1p(p)s2+∥z∥2p2+mTζ.$

Noting that for all $\begin{array}{}u\in W,\sum _{k=1}^{mT}{r}_{k-1}{\left|{\mathit{\Delta }}^{n}{u}_{k-1}\right|}^{p}=0,\end{array}$ we have

$J(u)=1p∑k=1mTrk−1Δnuk−1p−∑k=1mTF(k,uk,uk−1,⋯,uk−τ)=−∑k=1mTF(k,uk,uk−1,⋯,uk−τ)⩽0.$

Thus, there exists a positive constant R1 > δ such that J(u) ⩽ 0 for any u Q, where Q = (R1W)⊕{se|0 < s < R1}. This verifies the condition (J2) of Lemma 3.2. Therefore, J possesses a critical value cσ > 0 with

$c=infh∈Γsupu∈QJ(h(u))andΓ={h∈C(Q¯,EmT)∣h|∂Q=id}.$

Let ũEmT be a critical point associated to the critical value c of J, i.e., J(ũ) = c. If ũū, then we are done. Otherwise, if ũ = ū, it follows that

$c0=supu∈EmTJ(u)=infh∈Γsupu∈QJ(h(u)).$

In particular, choosing h = id, we have $\begin{array}{}\underset{u\in Q}{sup}J\left(u\right)={c}_{0}.\end{array}$ Since the choice of e B1V is arbitrary, we can take –e B1V. Similarly, there exists a positive number R2 > δ, so that J(u) ⩽ 0 for any u Q1, where Q1 = (R2W) ⊕ {–se|0 < s < R2}.

Using Linking Theorem again, J possesses a critical value c′⩾ σ > 0, where

$c′=infh∈Γ1supu∈Q1J(h(u))andΓ1={h∈C(Q¯1,EmT)∣h|∂Q1=id}.$

We claim that c′ ≠ c0. Otherwise, suppose that c′ = c0, then $\begin{array}{}\underset{u\in {Q}_{1}}{sup}J\left(u\right)={c}_{0}.\end{array}$ Due to the facts J|Q ⩽ 0 and J|Q1 ⩽ 0, J attains its maximum at some points in the interior of sets Q and Q1. However, QQ1W and J(u) ⩽0 for any uW. This implies that c0 ⩽0, which contradicts c0 > 0. This proves the claim and hence the proof is complete.

According to Theorem 1.1, it is easy to obtain the following theorems and corollaries.

#### Theorem 4.1

Assume that (T1) and the following conditions hold:

• (T4)

there exist constants $\begin{array}{}{\rho }_{1}>0,\phantom{\rule{thinmathspace}{0ex}}\alpha \in \left(0,\frac{\underset{_}{r}}{\left(\tau +1{\right)}^{p/2}p}\left(mT{\right)}^{\frac{-|2-p|}{2}}\left(2\mathrm{sin}\frac{\pi }{mT}{\right)}^{np}\right)\end{array}$such that

$F(k,v0,v1,⋯,vτ)⩽α∑j=0τvj2p,fork∈Zand∑j=0τvj2⩽ρ12;$

• T5)

there exist constants $\begin{array}{}{\rho }_{2}>0,\phantom{\rule{thinmathspace}{0ex}}{\zeta }_{1}>0,\phantom{\rule{thinmathspace}{0ex}}\beta \in \left(\frac{\overline{r}}{\left(\tau +1{\right)}^{p/2}p}\left(mT{\right)}^{\frac{|2-p|}{2}}\left(2\mathrm{cos}\frac{1-\left(-1{\right)}^{mT}}{2mT}\pi \right){\right)}^{np},+\mathrm{\infty }\right)\end{array}$such that

$F(k,v0,v1,⋯,vτ)⩾β∑j=0τvj2p−ζ1,fork∈Zand∑j=0τvj2⩾ρ22.$

Then for any given positive integer m, Eq. (1.1) has at least two mT-periodic nontrivial solutions.

#### Theorem 4.2

Assume that (T1) iand the following conditions hold:

• (T6)

$limδ1→0F(k,v0,v1,⋯,vτ)δ1p=0,δ1=∑j=0τvj2,∀(k,v0,v1,⋯,vτ)∈Z×Rτ+1;$

• (T7)

there exist constants ρ3 > 0 and p′ > p such that for kZ and $\begin{array}{}\sum _{j=0}^{\tau }{v}_{j}^{2}⩾{\rho }_{3}^{2},\end{array}$

$0

Then for any given positive integer m, Eq. (1.1) has at least two mT-periodic nontrivial solutions.

If τ = 0 and f(k, uk) = qkg (uk), Eq. (1.1) reduces to the following 2nth-order nonlinear equation with p-Laplacian,

$Δnrk−nφpΔnuk−n=(−1)nqkguk,k∈Z,$(4.2)

where gC( R, R), qk+T} = qk > 0, for all kZ. Then, we have the following results.

#### Corollary 4.3

Assume that the following hypotheses hold:

• (G1)

there exists a functional G(v) ∈ C1(R, R) with G(v)⩾0 and it satisfies

$G′(v)=g(v);$

• (G2)

there exists a constant $\begin{array}{}\alpha \in \left(0,\frac{\underset{_}{r}}{\left(\tau +1{\right)}^{p/2}p}\left(mT{\right)}^{\frac{-|2-p|}{2}}\left(2\mathrm{sin}\frac{\pi }{mT}{\right)}^{np}\right)\end{array}$ such that

$lim sup|v|→0G(v)|v|p⩽α;$

• (G3)

there exists a constant $\begin{array}{}\beta \in \left(\frac{\overline{r}}{\left(\tau +1{\right)}^{p/2}p}\left(mT{\right)}^{\frac{|2-p|}{2}}\left(2\mathrm{cos}\frac{1-\left(-1{\right)}^{mT}}{2mT}\pi \right){\right)}^{np},+\mathrm{\infty }\right)\end{array}$ such that

$lim inf|v|→∞G(v)|v|p⩾β.$

Then for any given positive integer m, Eq. (4.2) has at least two mT-periodic nontrivial solutions.

#### Corollary 4.4

Assume that (G1) and the following conditions hold:

• (G4)

there exist constants $\begin{array}{}{\rho }_{1}^{\prime }>0,\alpha \in \left(0,\frac{\underset{_}{r}}{p}\left(mT{\right)}^{\frac{-|2-p|}{2}}\left(2\mathrm{sin}\frac{\pi }{mT}{\right)}^{np}\right)\end{array}$ such that

$G(v)⩽α|v|p,for|v|⩽ρ1′;$

• (G5)

there exist constants $\begin{array}{}{\rho }_{2}^{\prime }>0,\phantom{\rule{thinmathspace}{0ex}}{\zeta }_{1}^{\prime }>0,\phantom{\rule{thinmathspace}{0ex}}\beta \in \left(\frac{\overline{r}}{p}\left(mT{\right)}^{\frac{|2-p|}{2}}\left(2\mathrm{cos}\frac{1-\left(-1{\right)}^{mT}}{2mT}\pi \right){\right)}^{np},+\mathrm{\infty }\right)\end{array}$ such that

$G(v)⩾β|v|p−ζ1′,for|v|⩾ρ2′.$

Then for any given positive integer m, Eq. (4.2) has at least two mT-periodic nontrivial solutions.

#### Corollary 4.5

Assume that (G1) and the following conditions hold:

• (G6)

$lim|v|→0G(v)|v|p=0,∀v∈R;$

• (G7)

there exist constants $\begin{array}{}{\rho }_{3}^{\prime }\end{array}$ > 0 and ′ > p such that for kZ and |v|⩾ $\begin{array}{}{\rho }_{3}^{\prime }\end{array}$,

$0

Then for any given positive integer m, Eq. (4.2) has at least two mT-periodic nontrivial solutions.

#### Remark 4.6

If τ = 1, rk ≡ 1 and n = 1, Theorem 3.1 reduces to Theorem 3.1 in [5].

#### Remark 4.7

If τ = 0 and p = 2, Theorem 4.2 reduces to Theorem 1.1, Corollary 1.1 reduces to Corollary 1.1 in [3].

## 5 Example

Finally, as an application of Theorem 1.1, we give an example to illustrate our main results.

#### Example 5.1

For given nZ(1), consider the following difference equation

$Δnrk−nφpΔnuk−n=(−1)nμuk∑j=kk+τ2+sinjπT(∑i=−τ0ui+j2)μ2−1,k∈Z$(5.1)

where {rk}kZ is a real sequence and rk+T = rk > 0, 1 < p < ∞, μ > p, T is a given positive integer. Here

$f(k,vk+τ,⋯,vk+1,vk,vk−1,⋯,vk−τ)=μvk∑j=kk+τ2+sinjπT∑i=−τ0vi+j2μ2−1$

and

$F(k,v0,v−1,⋯,v−τ)=2+sinkπT∑i=−τ0vi2μ2.$

Then

$∑j=kk+τ∂F(j,vj,vj−1,⋯,vj−τ)∂vk=μvk∑j=kk+τ2+sinjπT∑i=−τ0vi+j2μ2−1.$

It is easy to verify all the assumptions of Theorem 1.1 are satisfied. Consequently, for any given positive integer m, Eq. (5.1) has at least two mT-periodic nontrivial solutions.

## Acknowledgement

We are grateful to the anonymous referee for his/her valuable comments and suggestions. This work is supported by the National Natural Science Foundation of China (Grant No. 11571084) and the Program for Changjiang Scholars and Innovative Research Team in University (Grant No. IRT_16R16).

## References

• [1]

Guo Z. M., Yu J. S., Existence of periodic and subharmonic solutions for second-order superlinear difference equations, Sci. China Math., 2003, 46, 506–515

• [2]

Guo Z. M., Yu J. S., The existence of periodic and subharmonic solutions of subquadratic second order difference equations, J. London Math. Soc., 2003, 68, 419–430

• [3]

Cai X. C., Yu J. S., Existence of periodic solutions for a 2nth-order nonlinear difference equation, J. Math. Anal. Appl., 2007, 329, 870–878

• [4]

Zhou Z., Yu J. S., Chen Y. M., Periodic solutions for a 2nth-order nonlinear difference equation, Sci. China Math., 2010, 53, 41–50

• [5]

Chen P., Fang H., Existence of periodic and subharmonic solutions for second-order p-Laplacian difference equations, Adv. Difference Equ., 2007, 2007, 9 pages

• [6]

Lin G. H., Zhou Z., Periodic and subharmonic solutions for a 2nth-order difference equation containing both advance and retardation with ϕ-Laplacian, Adv. Difference Equ., 2014, 74

• [7]

Deng X. Q., Liu X., Zhang Y. B., Shi H. P., Periodic and subharmonic solutions for a 2nth-order difference equation involving p-Laplacian, Indag. Math., 2013, 24, 613–625

• [8]

Liu X., Zhang Y. B., Shi H. P., Deng X. Q., Periodic and subharmonic solutions for fourth-order nonlinear difference equations, Appl. Math. Comput., 2014, 236, 613–620

• [9]

Shi H. P., Zhang Y. B., Existence of periodic solutions for a 2nth-order nonlinear difference equation, Taiwanese J. Math., 2016, 20, 143–160

• [10]

Zhou Z., Yu J. S., On the existence of homoclinic solutions of a class of discrete nonlinear periodic systems, J. Differential Equations, 2010, 249, 1199–1212

• [11]

Zhou Z., Yu J. S., Chen Y. M., Homoclinic solutions in periodic difference equations with saturable nonlinearity, Sci. China Math., 2011, 54, 83–93

• [12]

Chen P., Tang X. H., Existence of homoclinic orbits for 2nth-order nonlinear difference equations containing both many advances and retardations, J. Math. Anal. Appl., 2011, 381, 485–505

• [13]

Mai A., Zhou Z., Discrete solitons for periodic discrete nonlinear Schrödinger equations, Appl. Math. Comput., 2013, 222, 34–41 Google Scholar

• [14]

Zhou Z., Ma D. F., Multiplicity results of breathers for the discrete nonlinear Schrödinger equations with unbounded potentials, Sci. China Math., 2015, 58, 781–790

• [15]

Zhang Q. Q., Homoclinic orbits for discrete Hamiltonian systems with indefinite linear part, Comm. Pure Appl. Anal., 2017, 14, 1929–1940 Google Scholar

• [16]

Lin G. H., Zhou Z., Homoclinic solutions of discrete ϕ-Laplacian equations with mixed nonlinearities, Comm. Pure Appl. Anal., 2018, 17, 1723–1747

• [17]

Zhang Q. Q., Homoclinic orbits for discrete Hamiltonian systems with local super-quadratic conditions, Comm. Pure Appl. Anal., 2019, 18, 1, 425–434

• [18]

Yu J. S., Guo Z. M., On boundary value problems for a discrete generalized Emden-Fowler equation, J. Differential Equations, 2006, 231, 18–31

• [19]

Zhang Q. Q., Boundary value problems for forth order nonlinear p-Laplacian difference equations, J. Appl. Math., 2014, 2014, Article ID 343129, 6 pages Google Scholar

• [20]

Bonanno G., Candito P., D’Agui G., Variational methods on finite dimensional Banach spaces and discrete problems, Adv. Nonlinear Stud., 2014, 14, 4, 915–939 Google Scholar

• [21]

Zhou Z., Su M. T., Boundary value problems for 2nth-order ϕc-Laplacian difference equations containing both advance and retardation, Appl. Math. lett., 2015, 41, 7–11

• [22]

D’Agui G., Mawhin J., Sciammetta A., Positive solutions for a discrete two point nonlinear boundary value problem with p–Laplacian, J. Math. Anal. Appl., 2017, 447, 1, 383–397

• [23]

Balanov Z., Garcia-Azpeitiay C., Krawcewicz W., On variational and topological methods in nonlinear difference equations, Commun. Pure Appl. Anal., 2018, 17, 6, 2813–2844

• [24]

Rabinowitz P. H., Minimax Methods in Critical Point Theory with Applications to Differential Equations, CBMS Reg. Conf. Ser. Math., 1986, 65 Google Scholar

Accepted: 2018-11-08

Published Online: 2018-12-29

Citation Information: Open Mathematics, Volume 16, Issue 1, Pages 1435–1444, ISSN (Online) 2391-5455,

Export Citation