论文部分内容阅读
针对无线自主网络中的信息扩散,提出了一种高效的资源分配机制.首先,使用一个传输队列来描述扩散信息的动态到达和离开过程.针对用户信息数据队列和无线传输信道的时变特性,将动态的信息扩散描述为多用户的马尔可夫决策过程,并将多用户的马尔可夫决策过程进行分解.为了降低算法的复杂度,提出了一种基于模型的在线学习方法.在用户信息数据到达率和无线信道变化的情况下,用户通过在线学习,仅需1次迭代就可确定自身具有预见性的行为决策.
Aiming at the information diffusion in wireless autonomous networks, an efficient resource allocation mechanism is proposed.First, a transmission queue is used to describe the dynamic arrival and departure of diffusion information.For the time-varying characteristics of user information data queues and wireless transmission channels, The dynamic information diffusion is described as a multi-user Markov decision process, and the multi-user Markov decision process is decomposed. In order to reduce the complexity of the algorithm, a model-based online learning method is proposed. With data arrival rates and wireless channel changes, users can determine their own predictive behavior decisions with just one iteration of online learning.