一维三次散焦色散方程的全局解，第五部分：低正则性NLS
Global solutions for 1D cubic defocusing dispersive equations, Part V: low regularity NLS
浏览论文内容
中文总结 AI 辅助
本文针对一维含广泛非线性项（含三次NLS）的散焦色散方程，利用密度-通量恒等式等方法，将全局适定性与色散界推广到负Sobolev空间$s > -1/2$的小数据情况，还可通过标度推广到大数据。
中文摘要 AI 辅助
本文的研究动机源于第一作者与最后一作者在前期工作中提出的一个广泛猜想：具有小初值的一维三次散焦色散流存在全局色散解。该猜想最初针对一类半线性薛定谔型模型（包括经典三次NLS）在$L^2$正则性下得到证明。在互补方向上，Harrop-Griffiths、Killip和Vişan近期利用完全可积结构证明，三次NLS在$-\frac{1}{2} < s < 0$的所有$H^s$空间中是全局适定的。本文的目标是将一维三次NLS问题的全局适定性猜想推广到负 Sobolev 空间中小数据的情况，并证明全局色散界在该空间中依然成立。本文针对包含三次NLS在内的广泛非线性项完成了上述工作，这类非线性项通常产生非完全可积的流。本文的方法具有鲁棒性，依赖于密度-通量恒等式、相互作用Morawetz估计和隐式范式变换，而非可积性，且该方法一直延伸到标度临界阈值，即$s > -\frac{1}{2}$。与前期工作类似，本文得到的全局界包含$L^6_{t,x}$ Strichartz估计和双线性$L^2_{t,x}$估计；这些估计即使对于经典散焦三次NLS在负Sobolev正则性下也是新的。此外，通过标度变换，本文的色散界还可推广到大数据情况。
英文摘要
This article is motivated by a broad conjecture, formulated by the first and last authors in earlier work, asserting that one-dimensional cubic defocusing dispersive flows with small initial data have global, dispersive solutions. The conjecture was first established for a class of semilinear Schrödinger-type models at $L^2$ regularity, the classical cubic NLS among them. In a complementary direction, Harrop-Griffiths, Killip and Vişan have recently shown, using the completely integrable structure, that the cubic NLS is globally well-posed in $H^s$ for every $-\tfrac12 < s < 0$. Our aim here is to extend the reach of the global well-posedness conjecture for one dimensional cubic NLS problems to data which is small in negative Sobolev spaces, and to show that global dispersive bounds persist there. We do so for a broad class of nonlinearities which includes the cubic NLS but which in general generates flows that are not completely integrable. Our method is correspondingly robust, resting on density-flux identities, interaction Morawetz estimates and an implicit normal form transformation rather than on integrability, and it reaches all the way to the scaling-critical threshold, namely $s > -\tfrac12$. As in the earlier work, the global bounds we obtain include both $L^6_{t,x}$ Strichartz estimates and bilinear $L^2_{t,x}$ estimates; these are new even for the classical defocusing cubic NLS at negative Sobolev regularity. There, by scaling, our dispersive bounds also extend to the large data case.
