DocumentCode :
17208
Title :
Tracking a Markov-Modulated Stationary Degree Distribution of a Dynamic Random Graph
Author :
Hamdi, Mohamed ; Krishnamurthy, Vikram ; Yin, George
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of British Columbia, Vancouver, BC, Canada
Volume :
60
Issue :
10
fYear :
2014
fDate :
Oct. 2014
Firstpage :
6609
Lastpage :
6625
Abstract :
This paper considers a Markov-modulated duplication-deletion random graph where at each time instant, one node can either join or leave the network; the probabilities of joining or leaving evolve according to the realization of a finite state Markov chain. Two results are presented. First, motivated by social network applications, the asymptotic behavior of the degree distribution is analyzed. Second, a stochastic approximation algorithm is presented to track empirical degree distribution as it evolves over time. The tracking performance of the algorithm is analyzed in terms of mean square error and a functional central limit theorem is presented for the asymptotic tracking error. Also, a Hilbert-space-valued stochastic approximation algorithm that tracks a Markov-modulated probability mass function with support on the set of nonnegative integers is analyzed.
Keywords :
Markov processes; mean square error methods; network theory (graphs); Hilbert space valued stochastic approximation algorithm; Markov modulated duplication deletion random graph; Markov modulated probability mass function; Markov-modulated stationary degree distribution tracking; asymptotic tracking error; dynamic random graph; finite state Markov chain; functional central limit theorem; mean square error method; nonnegative integer set; social network application; Algorithm design and analysis; Approximation algorithms; Approximation methods; Heuristic algorithms; Markov processes; Social network services; Adaptive algorithms; Markov-modulated random graphs; complex networks; degree distribution; power law; social networks; stochastic approximation algorithms;
fLanguage :
English
Journal_Title :
Information Theory, IEEE Transactions on
Publisher :
ieee
ISSN :
0018-9448
Type :
jour
DOI :
10.1109/TIT.2014.2346183
Filename :
6873273
Link To Document :
بازگشت