Strichartz estimates and low-regularity solutions to multiplicative stochastic NLS

Immanuel Zachhuber

2024-04-29

Abstract:We study Strichartz estimates with very rough potentials, the spatial white noise on the 2 dimensional torus being of particular interest. Applications include solving the multiplicative stochastic NLS with general integer powers in a low-regularity setting, moreover we obtain global well-posedness in both the energy and strong regime.

Analysis of PDEs,Probability

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is about the Strichartz estimates of the multiplicative stochastic nonlinear Schrödinger equation (NLS) with very rough potential fields (especially spatial white noise), and the existence and uniqueness of its low - regularity solutions. Specifically, the goals of the paper include:

1. **Prove Strichartz estimates**: The author studied the Strichartz estimates on the two - dimensional torus with very rough potential fields (such as spatial white noise). These estimates are crucial for dealing with the problems of low - regularity solutions.

2. **Low - regularity local well - posedness**: Through Strichartz estimates, the author proved the local well - posedness of the multiplicative stochastic NLS in the low - regularity space \(H^s\), where \(s\in\left(1-\frac{1}{n}, 1\right)\).

3. **Global well - posedness in the energy space**: The author further proved the global well - posedness of the multiplicative stochastic NLS in the energy space \(D(\sqrt{-H})\). Here, the energy space is defined as \(v\in L^2\) and \(\|\Gamma^{-1}v\|_{H^1}\sim\|v\|_{D(\sqrt{-H})} = |(-u, Hu)|^{1/2}<\infty\).

4. **Global well - posedness of strong solutions**: The author also proved the global well - posedness of the multiplicative stochastic NLS in the domain \(D(H)\) of the operator \(H\).

### Overview of Main Results

- **Theorem 1.1**: On the two - dimensional torus, for \(r\geq4\), there exists a constant \(\delta > 0\) such that for any \(\delta > 0\),
\[
\|e^{-itH}u\|_{L^r_t([0, 1]; L^r)}\lesssim\|u\|_{H^{(r - 3)(1+\delta)/r}}
\]

- **Theorem 1.2**: For \(n\in\mathbb{N}\), the multiplicative stochastic NLS
\[
(i\partial_t - H)u=-u|u|^{2n}
\]
is locally well - posed in the \(H^s\) space, where \(s\in\left(1-\frac{1}{n}, 1\right)\).

- **Theorem 1.3**: For \(n\in\mathbb{N}\), the multiplicative stochastic NLS is globally well - posed in the energy space \(D(\sqrt{-H})\).

- **Theorem 1.4**: For \(n\in\mathbb{N}\), the multiplicative stochastic NLS is globally well - posed in the domain \(D(H)\) of the operator \(H\).

### Methods and Techniques

- **Strichartz estimates**: Utilize classical Strichartz estimates and improved Strichartz estimates, especially the estimates on the two - dimensional torus.
- **Paracontrolled distribution theory**: Used to handle the definition and properties of the Anderson Hamiltonian with spatial white noise.
- **Fixed - point argument**: Prove local well - posedness by setting up a fixed - point problem in an appropriate function space.
- **Energy conservation**: Use energy conservation to control the \(L^\infty\) norm of the solution, thereby proving global well - posedness.

### Conclusion

This paper successfully solved the local and global well - posedness problems of the multiplicative stochastic NLS with very rough potential fields in the low - regularity space by introducing and proving new Strichartz estimates. These results are not only theoretically significant but also provide a powerful tool for further research on related problems.