English

Percolation of Partially Interdependent Scale-free Networks

Data Analysis, Statistics and Probability 2013-05-30 v2 Physics and Society

Abstract

We study the percolation behavior of two interdependent scale-free (SF) networks under random failure of 1-pp fraction of nodes. Our results are based on numerical solutions of analytical expressions and simulations. We find that as the coupling strength between the two networks qq reduces from 1 (fully coupled) to 0 (no coupling), there exist two critical coupling strengths q1q_1 and q2q_2, which separate three different regions with different behavior of the giant component as a function of pp. (i) For qq1q \geq q_1, an abrupt collapse transition occurs at p=pcp=p_c. (ii) For q2<q<q1q_2<q<q_1, the giant component has a hybrid transition combined of both, abrupt decrease at a certain p=pcjumpp=p^{jump}_c followed by a smooth decrease to zero for p<pcjumpp < p^{jump}_c as pp decreases to zero. (iii) For qq2q \leq q_2, the giant component has a continuous second-order transition (at p=pcp=p_c). We find that (a)(a) for λ3\lambda \leq 3, q11q_1 \equiv 1; and for λ>3\lambda > 3, q1q_1 decreases with increasing λ\lambda. (b)(b) In the hybrid transition, at the q2<q<q1q_2 < q < q_1 region, the mutual giant component PP_{\infty} jumps discontinuously at p=pcjumpp=p^{jump}_c to a very small but non-zero value, and when reducing pp, PP_{\infty} continuously approaches to 0 at pc=0p_c = 0 for λ<3\lambda < 3 and at pc>0p_c > 0 for λ>3\lambda > 3. Thus, the known theoretical pc=0p_c=0 for a single network with λ3\lambda \leqslant 3 is expected to be valid also for strictly partial interdependent networks.

Keywords

Cite

@article{arxiv.1206.2427,
  title  = {Percolation of Partially Interdependent Scale-free Networks},
  author = {Di Zhou and Jianxi Gao and H. Eugene Stanley and Shlomo Havlin},
  journal= {arXiv preprint arXiv:1206.2427},
  year   = {2013}
}

Comments

20 pages, 17 figures