Phase transition in a double branching annihilating random walk
This paper investigates the long-time behavior of double branching annihilating random walkers with nearest-neighbor dependent rates. The system consists of even number of particles which can execute nearest-neighbor random walk and they can as well give birth in a parity conserving manner to two other particles with rates $1$ and $b$, respectively, until they meet. Upon meeting, each of the adjacent particles can branch with rate $p\cdot b$ while it can annihilate, i.e. hop on, the other particle with rate $p$ for some $0< p\leq 1$. This process first appeared in the article by D. ben Avraham, F. Leyvraz and S. Redner (Propagation and extinction in branching annihilating random walks, Phys. Rev. E 50(3), 1994) and can be considered as the extension of Sudbury's model (The branching annihilating process: an interacting particle system, Ann. Probab., 18(2), 1990). We prove that in some region of the parameters $(p,b)$, the process survives with positive probability. Combining with Sudbury's extinction result it shows a phase transition phenomenon for this model. In some sense our result also shows the sharpness of the assumptions of the article by M. Balázs and A. L. Nagy (Dependent double branching annihilating random walk, Electron. J. Probab., 20(84), 2015). We use similar arguments that was developed by M. Bramson and L. Gray in their article titled "The survival of branching annihilating random walk" (Z. Wahrsch. Verw. Gebiete, 68(4), 1985).
