Content Placement Learning for Success Probability Maximization in Wireless Edge Caching Networks
Content Placement Learning for Success Probability Maximization in Wireless Edge Caching Networks
复制标题
DOI：
10.1109/icassp.2019.8682841
复制
发表时间：
2019-04
期刊：
影响因子：
--
通讯作者：
N. Garg;M. Sellathurai;T. Ratnarajah
中科院分区：
文献类型：
--
作者：
N. Garg;M. Sellathurai;T. Ratnarajah
为了满足无线多媒体通信日益增长的需求，重要内容的提前缓存是关键解决方案之一。最佳缓存取决于内容在未来的流行程度，这是事先未知的。本文将内容流行度建模为有限状态马尔可夫链，采用强化Q学习方法学习均匀泊松点过程（PPP）分布式缓存网络中的最优内容放置策略。给定一组可用的放置策略，模拟表明，所提出的框架成功地学习，并提供最佳的内容放置，以最大限度地提高平均成功概率。
To meet increasing demands of wireless multimedia communications, caching of important contents in advance is one of the key solutions. Optimal caching depends on content popularity in future which is unknown in advance. In this paper, modeling content popularity as a finite state Markov chain, reinforcement Q-learning is employed to learn optimal content placement strategy in homogeneous Poisson point process (PPP) distributed caching network. Given a set of available placement strategies, simulations show that the presented framework successfully learns and provides the best content placement to maximize the average success probability.