Abstract:We consider the model of branching Brownian motion with a single catalytic point at the origin and binary branching. We establish some fine results for the asymptotic behaviour of the numbers of particles travelling at different speeds and give an explicit characterisation of the spatial distribution of particles travelling at the critical speed.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the asymptotic behavior of the number of particles with different velocities and the spatial distribution characteristics of particles at the critical velocity in the branching Brownian motion model with a single - point catalytic source. Specifically, the author focuses on:
1. **The number of particles with different velocities**: For particles moving at different velocities, the author explores the asymptotic behavior of the number of these particles. In particular, for particles with velocity \( \lambda < \frac{\beta}{2} \), their number grows exponentially; while for particles with velocity \( \lambda > \frac{\beta}{2} \), their number eventually tends to zero.
2. **Spatial distribution at the critical velocity**: The author studies in detail the spatial distribution of particles moving at the critical velocity \( \lambda = \frac{\beta}{2} \) and gives an explicit characterization of this distribution. Specifically, the author proves that at the critical velocity, the particle distribution converges to a mixed Poisson point process.
### Main Results
The main results of the paper are concentrated in the following aspects:
1. **Theorem 1.1**: This theorem shows that at the critical velocity \( \frac{\beta}{2} \), the number of particles distributed in certain specific regions converges to a mixed Poisson point process. Specifically, for any initial position \( x_0 \), integers \( n, m \), non - negative integers \( k_1,\ldots,k_n, l_1,\ldots,l_m \), and mutually disjoint Borel sets \( A_1,\ldots,A_n, B_1,\ldots,B_m \), we have:
\[
\lim_{t\rightarrow\infty}P_{x_0}\left(\bigcap_{i = 1}^n\left\{\left|N_{A_i+\frac{\beta}{2}t}^t\right| = k_i\right\}\cap\bigcap_{j = 1}^m\left\{\left|N_{- B_j-\frac{\beta}{2}t}^t\right| = l_j\right\}\right)
= E_{x_0}\left[\prod_{i = 1}^n\left(\frac{(\mu(A_i)M_\infty)^{k_i}}{k_i!}e^{-\mu(A_i)M_\infty}\right)\prod_{j = 1}^m\left(\frac{(\mu(B_j)M_\infty)^{l_j}}{l_j!}e^{-\mu(B_j)M_\infty}\right)\right]
\]
where \( \mu(D)=\int_D\beta e^{-\beta y}dy \).
2. **Propositions 1.5 and 1.7**: These two propositions discuss the asymptotic behavior of the number of particles at the sub - critical velocity \( \lambda\in(0,\frac{\beta}{2}) \) and the super - critical velocity \( \lambda\in(\frac{\beta}{2},\beta) \) respectively. Specifically:
- For the sub - critical velocity \( \lambda\in(0,\frac{\beta}{2}) \), we have:
\[
e^{-\Delta_\lambda t}\left|N_{A+\lambda t}^t\right|\to\mu(A)M_\infty\quad\text{in }P_{x_0}\text{- probability sense}
\]
- For the super - critical velocity \( \lambda\in(\frac{\beta}{2},\beta) \)