The stochastic nonlinear Schrödinger equations driven by pure jump noise (Q6165365)

The authors consider the stochastic nonlinear Schrödinger equations driven by pure jump noise and written as: \(dX(t)=i[\Delta X(t)-\lambda \left\vert X(t)\right\vert ^{\alpha -1}X(t)]dt+dL(t)\), \(X(0)=x\in L^{2}( \mathbb{R}^{d})\), where \(\lambda \in \{-1,1\}\), \(1<\alpha <1+4/d\) and \( L=(L(t))_{t\geq 0}\) is a \(L^{2}(\mathbb{R}^{d})\)-valued pure jump Lévy process defined as \(L(t)=\int_{0}^{t}\int_{\mathcal{B}}z\widetilde{N}(dz,dt)\) , where \(\mathcal{B}=\{z\in L^{2}(\mathbb{R}^{d}):0<\left\Vert z\right\Vert _{L^{2}(\mathbb{R}^{d})}\leq 1\}\), \(N\) represents a time homogeneous Poisson random measure over \((L^{2}(\mathbb{R}^{d}),\mathcal{B}(L^{2}(\mathbb{R} ^{d})))\) with \(\sigma \)-finite intensity measure \(\nu \) satisfying \( \int_{L^{2}(\mathbb{R}^{d})\setminus \{0\}}\left\Vert z\right\Vert _{L^{2}( \mathbb{R}^{d})}^{2}\wedge 1\nu (dz)<\infty \), and \(\widetilde{N} (dz,dt)=N(dz,dt)-\nu (dz)dt\) denotes the corresponding compensated Poisson random measure. The authors define a local mild solution to the above stochastic nonlinear Schrödinger equations is an \(\mathbb{F}\)-progressively measurable process \(X(t)\), \(t\in \lbrack 0,\tau )\), where \( \tau \) is an accessible stopping time with an approximating sequence \((\tau _{n})_{n\in \mathbb{N}}\) of stopping times such that for every \(n\in \mathbb{ N}\), \((X(t))_{t\in \lbrack 0,\tau _{n}]}\in D(0,\tau _{n};L^{2}(\mathbb{R} ^{d}))\), \(\mathbb{P}\)-a.s., \((X(t))_{t\in \lbrack 0,\tau _{n}]}\) belongs to \( M_{\mathbb{F}}^{p}(Y_{\tau _{n}})\), for every \(t\in \lbrack 0,T]\), and which satisfies: \(X(t\wedge \tau _{n})=S_{t\wedge \tau _{n}}x-i\lambda \int_{0}^{t\wedge \tau _{n}}S_{t\wedge \tau _{n}-s}(\left\vert X(s)\right\vert ^{\alpha -1}X(s))ds+\int_{0}^{t\wedge \tau _{n}}\int_{B}1_{[0,\tau _{n}]}(s)S_{t\wedge \tau _{n}-s}z\widetilde{N}(dz,ds) \), \(\mathbb{P}\)-a.s. They also define the notion of maximal local mild solution for this problem. The main result of the paper proves that if \( p\geq 2\), \(r=\alpha +1\) such that \((p,r)\) is an admissible pair, and for any \(x\in L^{2}(\mathbb{R}^{d})\), there exists a unique global mild solution \( X^{x}=(X^{x}(t)\), \(t\in \lbrack 0,\infty ))\), to the above stochastic nonlinear Schrödinger equations such that \(X^{x}\in D([0,\infty );L^{2}( \mathbb{R}^{d}))\cap L_{\mathrm{loc}}^{p}([0,\infty );L^{r}(\mathbb{R}^{d}))\), \( \mathbb{P}\)-a.s. \(\omega \in \Omega \). Moreover, for all \(T\in \lbrack 0,\infty )\) and any \(q\geq 2\), \(\mathbb{E}sup_{t\in \lbrack 0,T]}\left\Vert X(t)\right\Vert _{L^{2}(\mathbb{R}^{d})}^{q}\leq C(q,T,\left\Vert x\right\Vert _{L^{2}(\mathbb{R}^{d})})\). For the proof, the authors introduce a truncation function in the preceding equality for the mild solution and they quote from the literature tools which prove the existence of a global solution \(Z^{R}\) to the associated truncated equation. To this global solution, they associate a stopping time \(\tau _{R}\). They define \( X(t)=Z^{R}(t)\) for \(t\in \lbrack 0,\tau _{R}]\) and \(\tau _{\infty }=\lim_{R\uparrow \infty }\tau _{R}\). They observe that \((X(t))_{t\in \lbrack 0,\tau _{\infty })}\) is a maximal local mild solution of the stochastic nonlinear Schrödinger equations and they prove a \(L^{2}\)-estimate on this global solution, using deterministic and stochastic Strichartz estimates.