Dynamic Chinese Restaurant Game: Theory and Application to Cognitive Radio Networks

• Published in: IEEE transactions on wireless communications Vol. 13; no. 4; pp. 1960 - 1973
• Main Authors: Chunxiao Jiang, Yan Chen, Yu-Han Yang, Chih-Yu Wang, Liu, K. J. Ray
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.04.2014; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Applied sciences; Bayes methods; Bayesian learning; Chinese restaurant game; Cognitive radio; Decision making; Dynamic tests; Dynamics; Exact sciences and technology; game theory; Games; Learning; Markov analysis; Markov decision process; Markov processes; Mathematical model; Networks; Radiocommunication specific techniques; Radiocommunications; Restaurants; Social network services; Social networks; Telecommunications; Telecommunications and information theory; Performance evaluation; Markov process; Markov decision; Decision support system; Bayesian learning; Social network; Controller area networks; Game theory; Learning; Optimal decision; Credal approach; cognitive radio; Simulation; Chinese; Markov decision process; Bayes methods; Chinese restaurant game; Radio communication; Software radio
• ISSN: 1536-1276; 1558-2248
• DOI: 10.1109/TWC.2014.030314.130632

Users in a social network are usually confronted with decision making under uncertain network state. While there are some works in the social learning literature on how to construct belief on an uncertain network state, few study has been made on integrating learning with decision making for the scenario where users are uncertain about the network state and their decisions influence with each other. Moreover, the population in a social network can be dynamic since users may arrive at or leave the network at any time, which makes the problem even more challenging. In this paper, we propose a Dynamic Chinese Restaurant Game to study how a user in a dynamic social network learns the uncertain network state and make optimal decision by taking into account not only the immediate utility but also subsequent users' negative influence. We introduce a Bayesian learning based method for users to learn the network state, and propose a Multi-dimensional Markov Decision Process based approach for users to achieve the optimal decisions. Finally, we apply the Dynamic Chinese Restaurant Game to cognitive radio networks and demonstrate from simulations to verify the effectiveness and efficiency of the proposed scheme.