On the law of the index of Brownian loops related to the Hopf and anti-de Sitter fibrations

Fabrice Baudoin,Teije Kuijper
2024-10-14
Abstract:We give explicit formulas and asymptotics for the distribution of the index of the Brownian loop in the following geometrical settings: the complex projective line from which two points have been removed; the complex hyperbolic line from which one point has been removed; the odd dimensional spheres from which a great hypersphere has been removed; and the complex anti-de Sitter spaces. Our analysis is based on the geometry of the Hopf and anti-de Sitter fibrations, and on the relationship between winding and area forms.
Probability,Differential Geometry
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the distribution problem of the index of Brownian loops in several specific geometric settings. Specifically, the authors provide explicit formulas and asymptotic behaviors of the Brownian loop index in the following geometric settings: 1. **Complex projective line**: Two points are removed from the complex projective line $\mathbb{CP}^1$; 2. **Complex hyperbola**: One point is removed from the complex hyperbola $\mathbb{CH}^1$; 3. **Odd - dimensional sphere**: One large hypersphere is removed from the $(2n + 1)$ - dimensional sphere $S^{2n+1}$; 4. **Complex anti - de Sitter space**: The $(2n + 1)$ - dimensional complex anti - de Sitter space $\text{AdS}^{2n+1}$. These analyses are based on the geometric properties of Hopf and anti - de Sitter fibrations, as well as the relationship between the winding form and the area form. Through these studies, the authors hope to accurately describe the index distribution of Brownian loops on different Riemannian manifolds and reveal the underlying mathematical structure and physical significance. ### Specific problem description In these geometric settings, the index of a Brownian motion loop is a discrete random variable, representing the number of times the path winds around a reference point or a reference region. For example, for a Brownian loop in the plane, the index can be written in the form of a random line integral: \[ \theta(L)-\theta(0)=\int_{X_{z_0}[0,L]}\alpha \] where $\alpha=\frac{x \, dy - y \, dx}{x^2 + y^2}$ is a closed but not exact one - form, called the winding form, which is intrinsic on $\mathbb{C}\setminus\{0\}$. ### Main objective The main objective of the paper is to calculate the exact distribution of the Brownian loop index on the above - mentioned different Riemannian manifolds $M$. For a specific example, such as the Lie group $\text{SL}(2,\mathbb{R})$, the authors prove the following theorem: \[ P(\text{Ind}(X[0,L]) = k)=C e^{-\frac{\pi^2 k^2}{L}}\int_{-\infty}^{+\infty}\cos\left(\frac{\pi k y}{L}\right)\frac{y}{\sinh(y)}e^{-\frac{y^2}{4L}}\, dy \] where $C>0$ is a normalization constant. In particular, when $L\rightarrow+\infty$, we have: \[ \frac{2\pi \text{Ind}(X[0,L])}{\sqrt{2L}}\rightarrow N(0,1) \] Here $N(0,1)$ represents a normal random variable with mean zero and variance one. Through these results, the authors not only provide specific mathematical formulas, but also reveal the asymptotic behavior of the Brownian loop index distribution and its relationship with classical probability theory results (such as the Cauchy tail).