Abstract
We study the initial-boundary vaule problem of the long and short wave equations posed on the half line with initial datas \((u_{0}, n_{0})\in H^{s_{0}}({\mathbb {R}}^{+})\times H^{s_{1}}({\mathbb {R}}^{+})\) and boundary datas \((h, f)\in H^{\frac{2s_{0}+1}{4}}({\mathbb {R}}^{+})\times H^{s_{1}}({\mathbb {R}}^{+})\). We show the local well-posedness by giving the bilinear estimates of the coupling terms for the equations in suitable spaces of Bourgain type. Moreover, we consider results concerning ill-posedness for the system. Finally, the system is proved to be globally well-posed in Sobolev spaces \(H^{1}({\mathbb {R}}^{+})\times L^{2}({\mathbb {R}}^{+})\).
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
Consideration is given to the equations where u is an unknown complex function and n is an unknown real function:
where \(\delta \) is a given positive constant, the initial conditions \((u_{0}, n_{0})\in H^{s_{0}}({\mathbb {R}}^{+})\times H^{s_{1}}({\mathbb {R}}^{+})\) and the boundary conditions \((h, f)\in H^{\frac{2s_{0}+1}{4}}({\mathbb {R}}^{+})\times H^{s_{1}}({\mathbb {R}}^{+})\).
The long-short (LS) type equations depict the resonance interaction between the short wave and long wave, which were first deduced by Djordjevic and Redekopp [14]. The system has been well-studied in the past, where the short wave is often described by Schrödinger equation and the long wave is described by some sort of dispersive wave equation such as KdV equation, Benjamin-Ono equation so that the system came in various forms. The well-posedness of the Cauchy problem
was studied by Corcho and Matheus [12], which is called Schrödinger-Debye system. To apply contraction mapping principle to handle this problem, the authors started by decoupling the system to obtain the integral formulation:
where \(U(t)=e^{it\Delta /2}\) is the Schrödinger linear unitary group. When M was the real line \({\mathbb {R}}\), this system was proved to be local well-posed in \(H^{k}({\mathbb {R}})\times H^{s}({\mathbb {R}})\) for \(|k|-\frac{1}{2}\le s<\min \{k+\frac{1}{2}, 2k+\frac{1}{2}\}\) and \(k>-\frac{1}{4}\). When M was the torus T, this system was proved to be local well-posed in \(H^{k}({\mathbb {T}})\times H^{s}({\mathbb {T}})\) for \(0\le s\le 2k\) and \(|s-k|<1\). Moreover, the authors proved the global well-posedness in \(H^{s}({\mathbb {R}})\times H^{s}({\mathbb {R}})\) for \(-\frac{3}{14}<s<0\). Corcho, Oliveira and Silva in [13] considered this system with initial data in the classic Sobolev spaces \(H^{k}({\mathbb {M}})\times H^{s}({\mathbb {M}})\), with k and s satisfying \(\max (0,k-1)\le s\le \min (2k,k+1)\) when \(M=R^{n}(n=2,3)\). In [1], Arbieto and Matheus slao considered the system when \(M=T^{n}\). They proved that the system was locally and globally well-posed in \(H^{s}(T)\times H^{s}(T)\) for \(s\ge 0\) when \(n=1\).
There are also many results regarding other types of long and short wave equations. Tsutsumi and Hatano [37] studied the Benney system:
the authors proved the well-posedness for initial data \((u_{0},n_{0})\in H^{k}({\mathbb {R}})\times H^{l}({\mathbb {R}})\) with \(\frac{1}{2}\le k<1\), \(l>k-\frac{1}{2}\) in the general case \(\beta \ne 0\), and \((u_{0},n_{0})\in H^{k}({\mathbb {R}})\times H^{\frac{1}{k}}({\mathbb {R}})\) with \(0<k<\frac{1}{2}\) in the special case \(\beta =0\). Guo and Wang [18] considered the following nonlinear equations with periodic boundary conditions:
the authors showed the existence and uniqueness of the global solution of the generalized LS type equations. Huo [22] proved the Cauchy problem of the LS type equations where the long wave is described by Benjamin-Ono equation is locally well-posed in \(H^{s}({\mathbb {R}})\times H^{s-\frac{1}{2}}({\mathbb {R}})\) for \(s>0\). In [34], Pecher investigated the Cauchy problem for Schrödinger-Benjamin-Ono system and showed global well-posedness for data with infinite energy where \(s>\frac{1}{3}\). Bekiranov, Ogawa and Ponce [2] proved that the Cauchy problem of the coupled Schrödinger-KdV system is locally well-posed in \(H^{s}({\mathbb {R}})\times H^{s-\frac{1}{2}}({\mathbb {R}})\) for \(s>0\). In [38], Wu improved the local results for the Schrödinger-KdV system and obtained some ill-posedness results at low regularity and global well-posedness results in \(H^{s}({\mathbb {R}})\times H^{s}({\mathbb {R}})\) for \(s>\frac{1}{2}\). The LS type equations were also studied in many other works( [3, 31,32,33, 39]), including (2+1)-dimensional LS type equations which were worth learning for researchers.
This paper is devoted to the local well-posedness of (1) in low regularity sapces and the global well-posedness in Sobolev spaces. Bourgain space was introduced by Bourgain [6] and [7] to research the low regularity for some equations, which are to dispersive equations as Sobolev spaces are to elliptic equations. The Bourgain method is also called the Fourier restriction norm method. In [27,28,29], Kenig, Ponce and Vega developed a bilinear estimate and applied it to these equations in Bourgain space, which simplifies the Fourier restriction norm method. This method was initially applied to Schrödinger equation and KdV equation and has improved significantly compared to previous results, see [20, 35]. To study the global well-posedness in low regularity, [k; Z]-multiplier norm method and the I-method were introduced in [10, 36]. So far, the Bourgain method has been applied to many other equations such as Boussinesq equation, Hirota equation, Ostrovsky equation and the fifth order shallow water equation, see [9, 11, 17, 19, 21, 23,24,25, 30].
Although the Cauchy problem for the LS type equations have been studied by many reseachers, the initial-boundary value problem(IBVP) of the LS type equations is discussed by few authors. How do we deal with the system with nonhomogeneous boundary data on the half line in low regularity? Erdoǧan and Tzirakis’s work in [15] is the motivation for our paper. In this work we investigate the local and global regularities of the LS type equations with initial conditions \((u_{0}, n_{0})\in H^{s_{0}}({\mathbb {R}}^{+})\times H^{s_{1}}({\mathbb {R}}^{+})\) and boundary conditions \((h, f)\in H^{\frac{2s_{0}+1}{4}}({\mathbb {R}}^{+})\times H^{s_{1}}({\mathbb {R}}^{+})\). Recall that the Fourier restricted norm method has been used to many other equations and has got success, therefore we try to use this method to obtain the local well-posedness in low regularity. However, it seems that the Bourgain type spaces were always applied to dispersive equations. To find admissible \(s_{0}\) and \(s_{1}\), we try to apply the Bourgain space method for the system by special techniques where the second equation is not dispersive in this article. Then the existence and uniqueness of the solutions is obtained by the contraction mapping principle. For the completeness of the research, the ill-posedness is proved accordingly.
Moreover, the global well-posedness for the nonhomogeneous boundary-value problem is considered. Regarding the well-posedness, there is a difficulty in our case because of the presence of the boundary conditions. Keeping the difficulty in mind, we apply the work of Carroll [8] to overcome this, where the unknown boundary conditions are seen as the given undetermined coefficient.
In this paper, we say \((s_{0}, s_{1})\) is admissible if \(s_{0}, s_{1}\) satisfy
Definition 1
The system (1) is locally well-posed in \(H^{s_{0}}({\mathbb {R}}^{+})\times H^{s_{1}}({\mathbb {R}}^{+})\), if for any \(u_{0}\in H^{s_{0}}({\mathbb {R}}^{+}), h\in H^{\frac{2s_{0}+1}{4}}(R^{+}), n_{0}\in H^{s_{1}}({\mathbb {R}}^{+}), \) and \( f\in H^{s_{1}}({\mathbb {R}}^{+})\), the integral equation (4) below has a unique solution in
where \(b, b_{1}<\frac{1}{2}\) and T is sufficiently small.
Theorem 1
For any admissible pair \((s_{0}, s_{1})\), if \((u_{0},n_{0})\in H^{s_{0}}({\mathbb {R}}^{+})\times H^{s_{1}}({\mathbb {R}}^{+}), (h,f)\in H^{\frac{2s_{0}+1}{4}}(R^{+})\times H^{s_{1}}({\mathbb {R}}^{+})\), the equation (1) is locally well-posed in
The introduction of \(X^{s_{0}, b}\) and \(Y^{s_{1}, b_{1}}\) is presented in the next section.
In the next theorem, an ill-posedness result is stated for the LS type equations.
Theorem 2
If \(2s_{0}<s_{1}+\frac{1}{2}\), the associated map data-solution \(H^{s_{0}}({\mathbb {R}}^{+})\times H^{s_{1}}({\mathbb {R}}^{+})\rightarrow C_{t}^{0}([0, T];H^{s_{0}}({\mathbb {R}}^{+})\times H^{s_{1}}({\mathbb {R}}^{+}))\) is not \(C^{2}\) at zero.
Theorem 3
If \((u_{0},n_{0})\in H^{1}({\mathbb {R}}^{+})\times L^{2}({\mathbb {R}}^{+}), (h,f)\in H^{1}(R^{+})\times L^{2}({\mathbb {R}}^{+})\), the equation (1) is globally well-posed in \(H^{1}({\mathbb {R}}^{+})\times L^{2}({\mathbb {R}}^{+})\).
The paper is organized as follows: In Sect. 2, we give several notations and outline important results prepared for the later study. We use the Bourgain space method to present the bilinear estimates in Sect. 3. Section 4 is committed to the local results of the equations. In Sect. 5, the ill-posedness result is stated for the system. In Sect. 6, we obtain a priori bounds and show the global well-posedness of the LS type equations in Sobolev spaces.
2 Preliminaries
Now, we give several basic notations and important results in preparation for subsequent proof.
2.1 Notations and Function Spaces
Throughout the paper, C will signify an arbitrarily positive constant. In all this paper, for \(x, y\in {\mathbb {R}}^{+}\), we denote \(x\lesssim y\) by \(x\le Cy\). In addition, we denote \(x\sim y\) by \(x\lesssim y\) and \(y\lesssim x\). Furthermore, we use \(x\ll y\) to denote the statement \(x\le C^{-1}y\). The notation \(x+\) means \(x+\epsilon \), where \(\epsilon \) is small enough. We denote \(x-\) similarly.
We set
\(D_{0}f(t)=f(0, t)\), for a space time function f.
We fix a cut-off function \(\phi (t):=\phi \), \(\phi \in C_{0}^{\infty }({\mathbb {R}})\) such that \(\phi \equiv 1\) on [0, 1], \(\phi \equiv 0\) for \(|t|>2\).
Let \(s>-\frac{1}{2}\), we define
by the norm
where \(\chi \) is the usual characteristic function.
We denote the linear Schrödinger group by
where \(f\in L^{2}({\mathbb {R}})\) and \(p(\xi )=-\xi ^{2}\). Similarly, we use the notation \(e^{-t\delta }f(x)={\mathcal {F}}^{-1}(e^{it\phi (\cdot )}{\widehat{f}}(\cdot ))(x)\) for the second linear equation. It should be pointed out that \({\widehat{f}}\) represnts the Fourier transform of f.
We give a defineition of the function space \(X^{s, b}\):
We also give a defineition of the space \(Y^{s, b}\):
2.2 Notion of a Solution
To begin with, it is necessary to consider the linear problem on \({\mathbb {R}}^{+}\) to explore the solutions of the IBVP (1). Considering the extension of functions, the following lemma need to be stated.
Lemma 1
([16]) For \(g\in H^{s}({\mathbb {R}}^{+})\).
-
(1)
Suppose \(-\frac{1}{2}<s<\frac{1}{2}\), it holds that \(\Vert \chi _{(0, \infty )}g\Vert _{H^{s}({\mathbb {R}})}\lesssim \Vert g\Vert _{H^{s}({\mathbb {R}}^{+})}\).
-
(2)
Suppose \(\frac{1}{2}<s<\frac{3}{2}\) and \(g(0)=0\), it holds that \(\Vert \chi _{(0, \infty )}g\Vert _{H^{s}({\mathbb {R}})}\lesssim \Vert g\Vert _{H^{s}({\mathbb {R}}^{+})}\).
Then the idea is to write the solutions of two linear equations respectively. Concerning the linear Schrödinger equation, the solution
can be written as
where \(p(t)=\phi (t)D_{0}[W_{{\mathbb {R}}}(t){{\tilde{u}}}_{0}]\), which is in \(H^{\frac{2s+1}{4}}({\mathbb {R}}^{+})\) and \({{\tilde{u}}}_{0}\) is the extension of \(u_{0}\). Note that \(W_{0}^{t}(0, h)\) stands for the boundary operator which has been given in [4].
The porperties of the second linear problem is also considered:
where \(f(t)=e^{-\delta t}g(0)\) and g(x) is a arbitrary function with respect to x. Observe that when \(n_{0}=0, \)
on \({\mathbb {R}}^{+}\) is the solution of (3).
Let \({{\tilde{n}}}_{0}\) be extension of \(n_{0}\) and \(V_{0}^{t}({\tilde{n}}_{0}, f)\) be written as
where g(x) is taken as \(n_{0}(x)\) and \(r(t)=\phi (t)D_{0}(e^{-\delta t}{{\tilde{n}}}_{0})=\phi (t)[e^{-\delta t}{{\tilde{n}}}_{0}]|_{x=0}\). Note that the restrictionof \(V_{0}^{t}({\tilde{n}}_{0}, f)\) on \({\mathbb {R}}^{+}\times [0, 1]\) is the solution of (3).
The LS type equations (1) can be rewritten as the equivalent integral formulations:
where \(F(u, n)=\phi (t/T)(nu-|u|^{2}u), q(t)=\phi (t)[\int _{0}^{t}W_{{\mathbb {R}}}(t-\tau )F(u, n)d\tau ]|_{x=0}\) and \(G(u, n)=-\phi (t/T)|u|^{2}_{x}, z(t)=\phi (t)[\int _{0}^{t}e^{-\delta (t-\tau )}G(u, n)d\tau ]|_{x=0}.\)
For the dispersive equation, we collect the following basic results that are classical in previous studies. (see [15]).
For any s and b, we obtain
For \(s\in {\mathbb {R}}, 0\le b_{1}<\frac{1}{2}\) and \(0\le b_{2}<1-b_{1}\), we obtain
What’s more, for \(T<1\) and \(-\frac{1}{2}<b_{1}<b_{2}<\frac{1}{2}\), we obtain
There have been extensive results for the dispersive equations in Bourgain spaces. One of the features of this work is that we give similiar estimates in Bourgain type spaces for the second equation which is not dispersive.
Proposition 4
For any s and b, it holds that
Proof
According to the introduction of the space \(Y^{s, b}\), we can see that
This completes the proof. \(\hfill\square \)
Proposition 5
Let \(0\le b_{1}<\frac{1}{2}, b_{2}\ge 0\), with \(b_{1}+b_{2}\le 1, s\in {\mathbb {R}}\). Define \(k_{\xi }(t)\) by
Then, there exists
Proof
First, the estimates are given for some integrations. If \(|\tau |\ge 1\), there exists
If \(|\tau |\le 1\), we have the estimates as follows:
Then, we separate the integrand as follows:
where
and
For \(k_{1}\), we obtain
For \(k_{2}\), we obtain
For \(k_{3}\), we obtain
For \(k_{4}\), we obtain
The proof is completed. \(\hfill\square \)
Proposition 6
Let \(0\le b_{1}, b_{2}<\frac{1}{2}, s\in {\mathbb {R}}\), then
Proof
According to the introduction of the space \(Y^{s, b}\), we obtain the estimate as follows:
The proposition is proved, where the last inequality can be obtained by Proposition 2.3. \(\hfill\square \)
Finally, we get the lemmas [16] below which we will use throughout the paper.
Lemma 2
If \(b_{1}\ge b_{2}\ge 0\) and \(b_{1}+b_{2}>1\), then
where
Lemma 3
If \(b_{0}, b_{1}, b_{2}\ge 0\) and \(b_{0}+b_{1}+b_{2}>1\). Let \(l:=\max (1, b_{0}, b_{1}, b_{2})\). Then
the term \(\langle a\rangle ^{0+}\) can be discarded unless \(\max (b_{0}, b_{1}, b_{2})=1\).
Lemma 4
For fixed \(\beta \in (\frac{1}{2}, 1)\), we have
3 A Priori Estimates
Our destination of this section is to establish a priori estimates of the system (1) which allows us to prove the local results. We start with the linear estimates on two equations separately. Then we prove the crucial bilinear estimates in detail.
3.1 Linear Estimates
The corresponding linear equations are posed first:
For this system, we have the following lemmas.
Lemma 5
For \(s\ge 0\). For any \(u_{0}\in H^{s}({\mathbb {R}})\), it holds that \(\phi (t)W_{{\mathbb {R}}}u_{0}\in C_{x}^{0}H_{t}^{\frac{2\,s+1}{4}}({\mathbb {R}}\times {\mathbb {R}})\), and we have
Proof
Because the argument is well-known, we shall omit the details here. (see [16]) \(\hfill\square \)
Lemma 6
For \(s\in {\mathbb {R}}\). For any \(n_{0}\in H^{s}({\mathbb {R}})\), it holds that \(\phi (t)e^{-\delta t}n_{0}\in C_{x}^{0}H_{t}^{s}({\mathbb {R}}\times {\mathbb {R}})\), and we have
Proof
It is not hard to achieve the lemma since the operator is continuous. \(\hfill\square \)
Lemma 7
For \(s\ge 0\). Suppose h with \(\chi _{(0, \infty )}h\in H^{\frac{2\,s+1}{4}}({\mathbb {R}})\), it holds that \(W_{0}^{t}(0, h)\in C_{t}^{0}H_{x}^{s}({\mathbb {R}}\times {\mathbb {R}})\), and \(\phi (t)W_{0}^{t}(0, h)\in C_{x}^{0}H_{t}^{\frac{2\,s+1}{4}}({\mathbb {R}}\times {\mathbb {R}})\).
Proposition 7
For \(b\le \frac{1}{2}\) and \(s\ge 0\). Suppose h with \(\chi _{(0, \infty )}h\in H^{\frac{2\,s+1}{4}}({\mathbb {R}})\), it holds that
Proof
It can be found in [15]. \(\hfill\square \)
Lemma 8
For \(s\in {\mathbb {R}}\). Suppose f with \(\chi _{(0, \infty )}f\in H^{s}({\mathbb {R}})\), it holds that \(V_{0}^{t}(0, f)\in C_{t}^{0}H_{x}^{s}({\mathbb {R}}\times {\mathbb {R}})\), and \(\phi (t)V_{0}^{t}(0, f)\in C_{x}^{0}H_{t}^{s}({\mathbb {R}}\times {\mathbb {R}})\).
Proposition 8
For \(b\in {\mathbb {R}}\) and \(s>-\frac{1}{2}\). Suppose f with \(\chi _{(0, \infty )}f\in H^{s}({\mathbb {R}})\), it holds that
Proof
The proof is not particularly difficult but will not be reproduced here. \(\hfill\square \)
3.2 Nonlinear Estimates
Let us estimate the forced IBVP:
Proposition 9
For any \(b<\frac{1}{2}\), it holds that
Proof
The proof can be found in [15]. \(\hfill\square \)
For the second equation, we have a similar proposition:
Proposition 10
For any \(b^{\prime }<\frac{1}{2}\), it holds that
Proof
Now that \(Y^{s, b}\) norm is independent of space translation, it turns to estimate the bound above for \(\phi D_{0}(\int _{0}^{t}e^{-\delta (t-\tau )}Gd\tau )\). At \(x=0\), we get
Using
and
we obtain
Let us write
where
and \(\psi \) is defined as a smooth cut-off function in \([-1, 1]\), and \(\psi ^{c}=1-\psi \). According to Taylor expansion, we obtain
Using an inequality that \(\Vert uv\Vert _{H^{s}}\le \Vert u\Vert _{H^{1}}\Vert v\Vert _{H^{s}}\), we have
where we use that \(\Vert \phi t^{k}\Vert _{H^{1}}\le k\) and \(\sum _{k=1}^{\infty }\frac{1}{(k-1)!}\) is convergent. This completes the proof for \(s\in {\mathbb {R}}\).
For \(s>\frac{1}{2}\), by Cauchy–Schwartz inequality in \(\xi \), it is easy to find that
We can bound it by \(\Vert G\Vert _{Y^{s, -b^{\prime }}}\) since \(\int \langle \xi \rangle ^{-2s}d\xi \) is convergent.
Next we prove the second term \(S_{2}\),
For \(-\frac{1}{2}<s<\frac{1}{2}\), the proof is completed.
For \(s>\frac{1}{2}\), we see that
where we used the inequality \(\langle \lambda \rangle \lesssim \langle \lambda -\delta \rangle +\delta \). Applying Cauchy–Schwarz inquality, the second summand is bounded as follows
To compute \(\Vert S_{3}\Vert _{H^{s}({\mathbb {R}})}\), it can be completed by the method analogous to that used above. \(\hfill\square \)
3.3 Bilinear Estimate
In this section, our goal is to give bilinear estimates which is of great use to obtain local well-posedness. Due to the presence of the boundary conditions, we shall deal with more bilinear estimates than the Cauchy problem of LS type equations.
Proposition 11
For any admissible \(s_{0}, s_{1}\) and for any \(a_{0}<\min (\frac{1}{2}, s_{1}+\frac{1}{2}, s_{1}-s_{0}+1)\), for \(\frac{1}{2}-\epsilon<b, b_{1}<\frac{1}{2}\) where \(\epsilon \) is sufficient small, it holds that
Proof
According to the relation of the Fourier transform and its convolution, it is obvious to get
Therefore,
We define
and
Then it turns to prove that
Applying Cauchy–Schwarz in the \(\xi _{1}\tau _{1}\) integral and Hölder’s inequality, the norm above can be estimated by
Thus it turns to prove the proposition in the case that the supremum above is finite. Applying the Lemma 2.5, it yields the bound
With the help of the inequality \(\langle a-b\rangle \lesssim \langle \tau -a\rangle \langle \tau -b\rangle \), the supremum is reduced to
Observe that when \(|\xi |\lesssim 1\), it is not hard to obtain the bound
provided \(s_{1}+s_{0}>-\frac{1}{2}\).
When \(|\xi |\gg 1\), we take the following cases into consideration: \(i)\ |\xi _{1}|<1\), \(ii)\ |\xi _{1}-2\xi |<1\) and \(iii)\ |\xi _{1}|>1\) and \(|\xi _{1}-2\xi |>1\).
In the first case, the transformation \(\eta =\xi _{1}^{2}-2\xi \xi _{1}, d\eta =2(\xi _{1}-\xi )d\xi _{1}\) contributes to the following:
provided that \(a_{0}<\frac{1}{2}\).
In the second case, after the same variables substitution, the integral is estimated by
provided that \(a_{0}<s_{1}+\frac{1}{2}\).
In the third case, which implies that \(\langle \delta -\xi _{1}^{2}+2\xi \xi _{1}\rangle \sim \langle \xi _{1}^{2}-2\xi \xi _{1}\rangle \), it is not difficult to estimate the integral (after the variable substitution \(\xi _{1}\rightarrow \xi _{1}+\xi \))
Combining Lemma 2.5, we bound this by
which is bounded for
\(\hfill\square \)
Proposition 12
For any admissible \(s_{0}, s_{1}\) and for any \(\frac{1}{2}-s_{0}<a_{0}<\min (\frac{1}{2}, s_{1}+\frac{1}{2}, s_{1}-s_{0}+1)\), for \(\frac{1}{2}-\epsilon<b, b_{1}<\frac{1}{2}\) where \(\epsilon \) is sufficient small, it holds that
Proof
This proposition can be shown in a similar way as before. First, we have
Using Cauchy–Schwarz inequality in \(\xi _{1}, \lambda _{1}, \xi \) variables, it turns to verify that
Using Lemma 2.5 in the \(\lambda _{1}\) integral, we obtain the bound
First we consider the case \(\frac{3}{2}\le s_{0}+a_{0}\).
Using \(\langle \lambda +\xi ^{2}\rangle \lesssim \langle \lambda +(\xi -\xi _{1})^{2}-\delta \rangle \langle \xi _{1}\rangle \langle \xi _{1}-2\xi \rangle \), we bound the integral above by
For \(|\xi |\le |\xi _{1}|\), we bound this by
since \(2s_{0}>1\) and \(2s_{1}+3-2s_{0}-2a_{0}>1\).
For \(|\xi |\gg |\xi _{1}|\), we bound it by
since \(a_{0}<s_{1}+\frac{1}{2}\).
Now we consider the case \(\frac{1}{2}< s_{0}+a_{0}<\frac{3}{2}\).
Applying the inequality \(\langle a-b\rangle \lesssim \langle \tau -a\rangle \langle \tau -b\rangle \), we bound (11) by
For \(|\xi _{1}|\lesssim 1\), the integral is estimated after the variables substitution \(\eta =\xi _{1}^{2}-2\xi \xi _{1}, d\eta =2(\xi _{1}-\xi )d\xi _{1}\):
since \(s_{0}>0\) and \(s_{0}+a_{0}<1\). For \(|\xi -2\xi _{1}|\lesssim 1\), the integral is estimated after the change of variable as before \(\eta =\xi _{1}^{2}-2\xi \xi _{1}, d\eta =2(\xi _{1}-\xi )d\xi _{1}\):
since \(s_{1}+s_{0}>0\) and \(s_{0}+a_{0}<1\). For \(|\xi _{1}|\gtrsim 1\) and \(|\xi -2\xi _{1}|\gtrsim 1\), which implies that \(\langle \delta -\xi _{1}^{2}+2\xi \xi _{1}\rangle \sim \langle \xi _{1}^{2}-2\xi \xi _{1}\rangle \), we bound the integral by
Using Lemma 2.6 in the \(\xi _{1}\) integral, it is not difficult to get the bound
Under the assumption of the proposition, the integral is finite. \(\hfill\square \)
Proposition 13
For any admissible \(s_{0}, s_{1}\) and for any \(a_{0}<\min (2s_{0}+\frac{1}{2}, 1)\), for \(\frac{1}{2}-\epsilon<b, b_{1}<\frac{1}{2}\) where \(\epsilon \) is sufficient small, it holds that
Proof
The proof can be found in [16], so we omit it here. \(\hfill\square \)
Proposition 14
For any admissible \(s_{0}, s_{1}\) and for any \(\frac{1}{2}-s_{0}<a_{0}<min(1, s_{0}+\frac{1}{2}, 4s_{0})\), for \(\frac{1}{2}-\epsilon<b, b_{1}<\frac{1}{2}\) where \(\epsilon \) is sufficient small, it holds that
Proof
Similarly to Proposition 3.9. It is straightforward to show that
With aid of Cauchy–Schwarz inequality in the \(\xi _{1}, \xi _{2}, \lambda _{1}, \lambda _{2}, \xi \) variables (after the change of variable \(\xi _{2}\rightarrow -\xi _{2}, \lambda _{2}\rightarrow -\lambda _{2}\)), it turns to prove that
Using Lemma 2.5 in the \(\lambda _{1}\) and \(\lambda _{2}\) integral, the supremum is bounded by
First we consider the case \(\frac{3}{2}\le s_{0}+a_{0}\).
Note that \(\langle \lambda +\xi ^{2}\rangle \lesssim \langle \lambda +\xi _{1}^{2}-\xi _{2}^{2}+(\xi -\xi _{1}+\xi _{2})^{2}\rangle \langle \xi -\xi _{1}\rangle \langle \xi _{1}-\xi _{2}\rangle \), we bound (12) by
After the variables substitution \(\xi _{2}\rightarrow \xi _{1}+\xi _{2}, \xi _{1}\rightarrow \xi +\xi _{1}\), we bound it by
Considering the subcases \(|\xi +\xi _{1}+\xi _{2}|\gtrsim |\xi |\) and \(|\xi +\xi _{1}|\gtrsim |\xi |\), we can bound it by
Notice that \(\langle \xi +\xi _{1}\rangle \langle \xi \rangle \gtrsim \langle \xi _{1}\rangle \), we have
provided \(3-2a_{0}>1\) and \(s_{0}-a_{0}+\frac{3}{2}>1\). Using Lemma 2.5 and \(\langle \xi +\xi _{2}\rangle \langle \xi +\xi _{1}+\xi _{2}\rangle \gtrsim \langle \xi _{1}\rangle \) and \(\langle \xi +\xi _{2}\rangle \langle \xi \rangle \gtrsim \langle \xi _{2}\rangle \), we get
provided \(3-2a_{0}>1\) and \(s_{0}-a_{0}+\frac{3}{2}>1\).
Now we consider the case \(\frac{1}{2}< s_{0}+a_{0}<\frac{3}{2}\). For \(s_{0}\ge \frac{1}{2}\), applying the relation \(\langle a-b\rangle \lesssim \langle \tau -a\rangle \langle \tau -b\rangle \), we bound (12) by
provided \(s_{0}-a_{0}+\frac{3}{2}>1\).
For \(s>\frac{1}{2}\), we take the following cases into consideration:\(i)\ |\xi _{1}-\xi |<1\) or \(|\xi _{1}-\xi _{2}|<1\), \(ii)\ |\xi _{1}-\xi |\ge 1\) and \(|\xi _{1}-\xi _{2}|\ge 1\).
In the former case we find that \(\langle \xi _{1}\rangle \langle \xi -\xi _{1}+\xi _{2}\rangle \sim \langle \xi _{2}\rangle \langle \xi \rangle \), which yields the bound
Subsitituing \(\eta =(\xi _{1}-\xi )(\xi _{1}-\xi _{2})\) in the \(\xi _{1}\) integral, we obtain
provided \(a_{0}<\min (1, 4s_{0})\). In the latter case, after the variables substitution \(\xi _{2}\rightarrow \xi _{1}+\xi _{2}, \xi _{1}\rightarrow \xi +\xi _{1}\), the integral is bounded by
By symmetry, we obtain two inequalities: \(|\xi +\xi _{1}+\xi _{2}|\gtrsim |\xi |\) and \(|\xi +\xi _{1}|\gtrsim |\xi |\), leading to the bound
provided \(a_{0}<\min (1, s_{0}+\frac{1}{2})\). \(\hfill\square \)
Proposition 15
For any admissible \(s_{0}, s_{1}\) and for any \(a_{1}<\min (s_{0}-s_{1}, 2s_{0}-s_{1}-\frac{1}{2})\), for \(\frac{1}{2}-\epsilon<b, b_{1}<\frac{1}{2}\) where \(\epsilon \) is sufficient small, it holds that
Proof
Similarly to Proposition 3.9. Notice that
Therefore,
Let
and
Then we estimate the supremum by
When \(|\xi |\lesssim 1\), we bound the supermum by
When \(|\xi |\gg 1\), we estimate the supremum by
Then we consider the following two cases: when \(|\xi -2\xi _{1}|<1\), we bound the integral by
which is bounded by \(a_{1}<2s_{0}-s_{1}-\frac{1}{2}.\)
when \(|\xi -2\xi _{1}|>1\), we bound the integral by
this holds since \(a_{1}<\min (s_{0}-s_{1}, 2s_{0}-s_{1}-\frac{1}{2})\). \(\hfill\square \)
Proposition 16
For any admissible \(s_{0}, s_{1}\) and for any \(\frac{1}{2}-s_{1}<a_{1}<1-s_{1}\), for \(\frac{1}{2}-\epsilon<b, b_{1}<\frac{1}{2}\) where \(\epsilon \) is sufficient small, it holds that
Proof
Similarly to Proposition 3.9, we continue to show that
With the help of Lemma 2.5 in the \(\lambda _{1}\) integral and the variable substitution \(\eta =\xi (\xi -2\xi _{1})\) in the integral, the supremun is bounded as
Without loss of generality \(|\xi -\frac{\eta }{\xi }|\gtrsim |\xi |\), we get
Using \(s_{0}>\frac{1}{2}\), then integrating in \(\eta \), we have the bound in a similar way as before,
When \(|\xi |\lesssim 1\), we have the bound
provided \(a_{1}<\frac{3}{2}-s_{1}\).
When \(|\xi |\gg 1\), we have the bound
provided \(a_{1}<1-s_{1}\). \(\hfill\square \)
Proposition 17
For any admissible \(s_{0}, s_{1}\) and for any \(-\frac{1}{2}-s_{1}<a_{1}<\min (s_{0}-s_{1}+1, 1-s_{1})\), for \(\frac{1}{2}-\epsilon<b, b_{1}<\frac{1}{2}\) where \(\epsilon \) is sufficient small, it holds that
Proof
Similarly to Proposition 3.9, the goal is to prove that
Using Lemma 2.5 in the \(\lambda _{1}\) integral and the variable substitution \(\eta =\xi (\xi -2\xi _{1})\), the integral is estimated as
Without loss of generality \(|\xi -\frac{\eta }{\xi }|\gtrsim |\xi |\), it can be inferred that
When \(|\xi |\lesssim 1\), we have the bound
provided that \(a_{1}<\min (s_{0}-s_{1}+1, \frac{3}{2}-s_{1}).\)
When \(|\xi |\gg 1\), we have the bound
provided that \(a_{1}<1-s_{1}.\) \(\hfill\square \)
3.4 Counter-Examples
We are devoted to give two counter-examples, which shows that the restrictions on \(s_{0}\) and \(s_{1}\) are of necessity.
Theorem 18
For any \(b, b_{1}\in {\mathbb {R}}\), the estimate \(\Vert nu\Vert _{X^{s_{0}+a_{0}, -b}}\lesssim \Vert n\Vert _{Y^{s_{1}, b_{1}}}\Vert u\Vert _{X^{s_{0}, b}}\) holds only if \(s_{0}-s_{1}<1\).
Proof
Choose \(N\in Z^{+}\) a large integer. Let
Put \(\widehat{f}:=\chi _{A_{1}}\) and \(\widehat{g}:=\chi _{B_{1}}\). A straightforward computation gives that
\(\Vert f\Vert _{X^{s_{0}, b}}\sim N^{-\frac{1}{2}}\) and \(\Vert g\Vert _{Y^{s_{1}, b_{1}}}\sim N^{s_{1}-\frac{1}{2}}\).
Thus, \(\Vert fg\Vert _{X^{s_{0}, -\frac{1}{2}}}\lesssim \Vert f\Vert _{X^{s_{0}, b}}\Vert g\Vert _{Y^{s_{1}, b_{1}}}\) implies that \(s_{0}-s_{1}<1\), which completes the proof. \(\hfill\square \)
Theorem 19
For any \(b, b_{1}\in {\mathbb {R}}\), the estimate \(\Vert \partial _{x}|u|^{2}\Vert _{Y^{s_{1}+a_{1}, -b_{1}}}\lesssim \Vert u\Vert ^{2}_{X^{s_{0}, b}}\) holds only if \(2s_{0}-s_{1}\ge \frac{1}{2}\).
Proof
Choose \(N\in Z^{+}\) a large integer. Let
Put \(\widehat{f}:=\chi _{A_{2}}\). A straightforward computation gives that
Thus, \(\Vert \partial _{x}f^{2}\Vert _{Y^{s_{1}, -\frac{1}{2}}}\lesssim \Vert f\Vert _{X^{s_{0}, b}}^{2}\) implies that \(2s_{0}-s_{1}\ge \frac{1}{2}\). The proof is completed. \(\hfill\square \)
4 Local Theory
Now, we consider the IBVP:
We write the IBVP(13) in its integral equation form:
Let
A proof will be given that \(\Gamma =(\Gamma _{1}(u, n)(t), \Gamma _{2}(u, n)(t))\) is a contraction map(see [5]) from \(B_{r}\) to \(B_{r}\) for appropriate r and T. Applying (6), (7), Proposition 3.9 and Proposition 3.11, it yields that
Notice that
By (5), we get
Applying Lemma 2.1 and Proposition 3.4, it can be inferred that
By Lemma 3.1, we can obtain
According to (7), Propositions 3.7, 3.9, 3.10, 3.11 and 3.12, we see that
In view of these estimates, it shows that
Then, applying Propositions 2.4, 3.8 and 3.13, it yields that
Notice that
Using Lemma 2.1, Propositions 2.2 and 3.5, it follows that
By Lemma 3.2, we see that
Finally, according to (7), Propositions 3.8, 3.14 and 3.15, it is obvious to find
With aid of these estimates, it yields that
Hence, the main point of this problem is the following inequality. For any \(u, n\in B_{r}\),
Choosing \(r>0\) so that
Then
Therefore, with such a choice of r, \((\Gamma _{1}(u, n), \Gamma _{2}(u, n))\) maps \(B_{r}\) into \(B_{r}\). The same inequalities permit one to infer that for r as in (5),
for any \((u_{1}, n_{1}), (u_{2}, n_{2})\in B_{r}\). In other words, the map \((\Gamma _{1}(u, n), \Gamma _{2}(u, n))\) is a contraction mapping of \(B_{r}\).
Hence, we get the local well-posedness in \(X^{s_{0}, b}\times Y^{s_{1}, b_{1}}\). Now we check that \(u\in C_{t}^{0}H_{x}^{s_{0}}([0, T]\times {\mathbb {R}})\). Applying Lemma 3.1, we observe that the Schrödinger group operator \(W_{{\mathbb {R}}}{{\tilde{u}}}_{0}\) and the boundary term \(W_{0}^{t}(0, f-p+iq)\) is continuous, which means that the first and the third part of \(\Gamma _{1}\) is continuous in \(H^{s_{0}}\). We also obtain the same result for the second term owing to the embedding \(X^{s, b}\subset C_{t}^{0}H_{x}^{s}\) for \(b>1/2\) and (6) together with Propositions 3.9 and 3.11. The result that \(u\in C_{x}^{0}H_{t}^{\frac{2s_{0}+1}{4}}({\mathbb {R}}\times [0, T])\) is shown from Lemmas 3.1, 3.3 and Proposition 3.7. In addition, the statements for n are proved in a similar way. The well-posedness depending on the initial and boundary conditions follows from the a priori estimates and the above contraction mapping argument.
5 Ill-Posedness
Here, it is natural to be concerned about the ill-posedness of the LS type equation for the completeness of the research. Therefore, we shall give a proof of Theorem 1.3.
Proof
Suppose for the contradiction that the LS type equations (1) are locally well-posed on [0, T] for \(T\in (0, 1)\), and the solution map \((u_{0}, n_{0})\mapsto (u, n)\) is \(C^{2}\) from \(H^{s_{0}}({\mathbb {R}})\times H^{s_{1}}({\mathbb {R}})\) to \(C_{t}^{0}([0, T];H^{s_{0}}({\mathbb {R}})\times H^{s_{1}}({\mathbb {R}}))\). Then. by Picard iterative scheme, so is the operator \(A=(A_{1}, A_{2}):H^{s_{0}}({\mathbb {R}})\times H^{s_{1}}({\mathbb {R}})\rightarrow C_{t}^{0}([0, T];H^{s_{0}}({\mathbb {R}})\times H^{s_{1}}({\mathbb {R}}))\) defined as
Suppose N is large enough, we consider the initial data \((u_{0}, n_{0})\) such that
then \(\Vert u_{0}\Vert _{H^{s_{0}}}, \Vert n_{0}\Vert _{H^{s_{1}}}\sim \epsilon _{0}. \) We may set T=1 by choosing \(\epsilon _{0}\) small enough. Further, \(\Vert A_{2}\Vert _{C_{t}^{0}([0, 1];H^{s_{1}}}\) is equal to
This term has a lower bound of
Note that N, set \(t:=\frac{1}{100N^{4}}\) and localize to the region where \(-1\le \xi \le 1\), one can confirm that
where \(0\le \tau \le t\).
Thus, we obtain
Therefore, by choosing N large enough, we have
Since \(A_{2}\) is \(C^{2}-\)differentiable, we must have
but it fails to hold when \(s_{1}-2s_{0}+\frac{1}{2}>0\). This completes the proof. \(\hfill\square \)
6 Global Well-Posedness
We shall investigate global results in energy space \(H^{1}\times L^{2}\). To begin with, we recall the equations which need to be solved:
To obtain the global solution, we establish several a priori estimates and give the following identities. A few computational techniques are needed for the desired results.
Multiplying (16) by u, it is simple to yield
Multiplying (16) by \(2u_{t}+u\), integrating and taking the real part, it holds that
From (16) we infer that
By substituting (19) into (18), we see that
Multiplying (16) by \(u_{x}\), integrating and taking the real part, it holds that
From (16) we infer that
By substituting (22) into (21), we see that
Multipying (16) by n, we obtain
From (16) we find
Inserting (25) into (24), we get
Let
Then we have
Let
By integrating (17) with respect to t, it infers that
Therefore, we obtain
Using Gagliardo-Nirenberg inequality, we see that
In addition, note that
where C(A) represents a constant related to A. Then, we conclude that
where \(\rho \le 1, 2\delta -1+\delta C(\rho )-4\delta +\epsilon ^{-\frac{1}{7}}(1-\delta )\le \delta \). Therefore, we infer that
Using Gagliardo-Nirenberg inequality, we also see that
Then combining (23), we get that \(A\lesssim E(t)+1.\) Substituting this result into (28) and using Gronwall’s inequality, we deduce \(\Vert u\Vert _{H^{1}}+\Vert n\Vert _{L^{2}}\) remains bounded.
Data availability
The authors confirm that the data supporting the findings of this study are available within article.
References
Arbieto, A., Matheus, C.: On the periodic Schrödinger–Debye equation. Commun. Pure Appl. Anal. 7, 699–713 (2008)
Bekiranov, D., Ogawa, T., Ponce, G.: Weak solvability and well-posedness of a coupled Schrödinger–Korteweg–de Vries equation for capillary-gravity wave interactions. Proc. Am. Math. Soc. 125, 2907–2919 (1997)
Bekiranov, D., Ogawa, T., Ponce, G.: Interaction equations for short and long dispersive waves. J. Funct. Anal. 158, 357–388 (1998)
Bona, J.L., Sun, S.M., Zhang, B.Y.: Nonhomogeneous boundary-value problems for one-dimensional nonlinear Schrödinger equations. J. Math. Pures Appl. 109, 1–66 (2018)
Bota, M., Petrusel, A., Petrusel, G., Samet, B.: Coupled fixed point theorems for single-valued operators in b-metric spaces. Fixed Point Theory Appl. 2015, 1–15 (2015)
Bourgain, J.: Fourier transform restriction phenomena for certain lattice subsets and applications to nonlinear evolution equations, part i: Schrödinger equations. Geom. Funct. Anal. 3, 107–156 (1993)
Bourgain, J.: Fourier transform restriction phenomena for certain lattice subsets and applications to nonlinear evolution equations, part ii: The KdV equation. Geom. Funct. Anal. 3, 209–262 (1993)
Carroll, R., Bu, Q.: Solution of the forced nonlinear Schrödinger (nls) equation using pde techniques. Appl. Anal. 41, 33–51 (1991)
Chen, W.G.: Bourgain space \(X^{s, b}\) with applications to the KdV type equations. Adv. Math. 6, 641–654 (2011)
Colliander, J., Keel, M., Staffilani, G., Takaoka, H.: Global well-posedness for KdV in sobolev spaces of negative index. Electron. J. Differ. Equ. 2001, 1–7 (2001)
Compaan, E., Tzirakis, N.: Well-posedness and nonlinear smoothing for the “good’’ Boussinesq equation on the half-line. J. Differ. Equ. 262, 5824–5859 (2017)
Corcho, A., Matheus, C.: Sharp bilinear estimates and well-posedness for the 1-d Schrödinger–Debye system. Differ. Integr. Equ. 22, 357–391 (2008)
Corcho, A., Oliveira, F., Silva, J.: Local and global well-posedness for the critical Schrödinger–Debye system. Proc. Am. Math. Soc. 141, 3485–3499 (2013)
Djordjevic, V.D., Redekopp, L.G.: On two-dimensional packets of capillary-gravity waves. J. Fluid Mech. 79, 703–714 (1977)
Erdoğan, M.B., Tzirakis, N.: Regularity properties of the Zakharov system on the half line. Commun. Partial Differ. Equ. 42, 1121–1149 (2016)
Erdoğan, M.B., Tzirakis, N.: Regularity properties of the cubic nonlinear Schrödinger equation on the half line. Ann. Funct. Anal. 271, 2539–2568 (2015)
Gan, Z.H., Han, L.J., Zhang, J.J., Guo, B.L.: Cauchy problem for the Zakharov system arising from hot plasma with low regularity data. Commun. Math. Sci. 11, 403–420 (2012)
Guo, B.L., Wang, B.X.: The global solution and its long time behavior for a class of generalized LS type equations. Adv. Math. 6, 533–546 (1995)
Guo, B.L., Wu, J.: Well-posedness of the initial-boundary value problem for the Hirota equation on the half line. J. Math. Anal. Appl. 504, 1–25 (2021)
Holmer, J.: The initial-boundary-value problem for the 1d nonlinear Schrödinger equation on the half-line. Differ. Integr. Equ. 18, 647–668 (2005)
Huo, Z.H.: The Cauchy problem for the fifth-order shallow water equation. Acta Math. Appl. Sin-e. 21, 441–454 (2005)
Huo, Z.H., Guo, B.L.: Local well-posedness of interaction equations for short and long deipersive waves. Math. Basel. 17, 137–151 (2004)
Huo, Z.H., Jia, Y.L.: Well-posedness for the Cauchy problem to the Hirota equation in sobolev spaces of negative indices. Chin. Ann. Math. B 1, 75–88 (2005)
Isaza, P., Jorge, M.: Cauchy problem for the Ostrovsky equation in spaces of low regularity. J. Differ. Equ. 230, 661–681 (2006)
Jia, Y.L., Huo, Z.H.: Well-posedness for the fifth-order shallow water equations. J. Differ. Equ. 246, 2448–2467 (2009)
Kato, T.: Global well-posedness for the Kawahara equation with low regularity data. Commun. Pure Appl. Anal. 12, 1321–1339 (2013)
Kenig, C.E., Ponce, G., Vega, L.: Oscillatory integrals and regularity of dispersive equations. Indiana Univ. Math. J. 40, 33–69 (1991)
Kenig, C.E., Ponce, G., Vega, L.: A bilinear estimate with applications to the KdV equation. J. Am. Math. Soc. 9, 573–603 (1996)
Kenig, C.E., Vega, L.: Quadratic forms for the 1-d semilinear Schrdinger equation. Trans. Am. Math. Soc. 348, 3323–3353 (1996)
Li, S., Chen, M., Zhang, B.Y.: Low regularity solutions of non-homogeneous boundary value problems of a higher order Boussinesq equation in a quarter plane. J. Math. Anal. Appl. 492, 124406 (2020)
Liu, R.R., Liu, H., Xin, J.: Attractor for the non-autonomous long wave-short wave resonance interaction equation with damping. J. Appl. Anal. Comput. 10, 1149–1169 (2020)
Lu, H., Xin, J.: The existence of a global attractor for the (2+1)-dimensional long wave-short wave resonance interaction equation. Nonlinear Anal. 73, 3767–3778 (2010)
Oikawa, M., Okamura, M., Funakoshi, M.: Two-dimensional resonant interaction between long and short waves. J. Phys. Soc. Jpn. 58, 4416–4430 (2007)
Pecher, H.: Rough solutions of a Schrödinger–Benjamin–Ono system. Differ. Integr. Equ. 19, 517–535 (2006)
Takaok, H.: Global well-posedness for Schrödinger equations with derivative in a nonlinear term and data in low-order Sobolev spaces. Electron. J. Differ. Equ. 2001, 1–23 (2001)
Tao, T.: Multilinear weighted convolution of \(l^2\) functions, and applications to non-linear dispersive equations. Am. J. Math. 123, 839–908 (2000)
Tsutsumi, M., Hatano, S.: Well-posedness of the Cauchy problem for the long wave-short wave resonance equations. Nonlinear Anal-Theor. 22, 155–171 (1994)
Wu, Y.F.: The Cauchy problem of the Schrödinger–Korteweg–de Vries system. Differ. Integr. Equ. 23, 569–600 (2009)
Zhang, R.F.: Existence of global attractor for LS type equations. J. Math. Res. Expo. 26, 708–714 (2006)
Acknowledgements
This work is supported by National Natural Science Foundation of China (Grant 11771444, 12261053), Fundamental Research Funds for the Central Universities (No. 2024JCCXNY04) and Open Fund of State Key Labortaory for Fine Exploration and Intelligent Development of Coal Resources (Grant No. SKLCRSM24KFA01).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
Authors have no conflict of interest to declare.
Additional information
Communicated by Hongjun Gao.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Guo, C., Wang, Y., Xu, M. et al. Low Regularity for LS Type Equations on the Half Line. Bull. Malays. Math. Sci. Soc. 47, 144 (2024). https://doi.org/10.1007/s40840-024-01728-x
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s40840-024-01728-x
Keywords
- Long wave-short wave equations
- The Fourier restriction norm
- Initial-boundary value problem
- Bilinear estimates
- Local well-posedness