Joint mode selection and link allocation are crucial to achieve the advantage of Device-to-Device(D2 D) communications in improving spectral efficiency. In practice, cellular users tend to not be totally altruistic or...Joint mode selection and link allocation are crucial to achieve the advantage of Device-to-Device(D2 D) communications in improving spectral efficiency. In practice, cellular users tend to not be totally altruistic or absolutely selfish. How to stimulate them to devote their links and how to allocate their links to D2 D pair candidates efficiently are two main challenges. In this paper, we encourage cellular users through the variable payment with regard to the social tie strength between cellular users and D2 D pair candidates. In particular, the social tie strength is inferred through a graph inference model and its impact on the payment is quantified as a negative exponential function. Then, we propose a resource scheduling optimization model based on the non-transferable utility coalition formation game, and a distributed coalition formation algorithm based on the Pareto preference and merge-and-split rule. From them, the final coalition structure is obtained, which reflects the strategy of mode selection and link allocation. Numerical results are presented to verify the effectiveness of our proposed scheme.展开更多
基金supported by Natural Science Foundations of China (No. 61671474)Jiangsu Provincial Natural Science Foundation for Excellent Young Scholars (No. BK20170089)
文摘Joint mode selection and link allocation are crucial to achieve the advantage of Device-to-Device(D2 D) communications in improving spectral efficiency. In practice, cellular users tend to not be totally altruistic or absolutely selfish. How to stimulate them to devote their links and how to allocate their links to D2 D pair candidates efficiently are two main challenges. In this paper, we encourage cellular users through the variable payment with regard to the social tie strength between cellular users and D2 D pair candidates. In particular, the social tie strength is inferred through a graph inference model and its impact on the payment is quantified as a negative exponential function. Then, we propose a resource scheduling optimization model based on the non-transferable utility coalition formation game, and a distributed coalition formation algorithm based on the Pareto preference and merge-and-split rule. From them, the final coalition structure is obtained, which reflects the strategy of mode selection and link allocation. Numerical results are presented to verify the effectiveness of our proposed scheme.
