Real-world traffic analysis and joint caching and scheduling for in-RAN caching networks
Real-world traffic analysis and joint caching and scheduling for in-RAN caching networks
复制标题
DOI：
10.1007/s11432-016-0391-2
复制
发表时间：
2017-02
期刊：
影响因子：
--
通讯作者：
Zejue Wang;Hongjia Li;Zhen Xu
中科院分区：
文献类型：
--
作者：
Zejue Wang;Hongjia Li;Zhen Xu
本文分析了中国现有LTE网络的流量，研究了无线接入网络（RAN）缓存中内容对象缓存和调度的联合优化。协作缓存已经被公认为释放in-RAN缓存的最终潜力的一种方式，但其可行性仍未被探索。此外，内容对象缓存和调度是缓存部署的两个关键问题，通常需要共同考虑和解决。然而，它们是由不同时间粒度的不同事件触发的。因此，在真实数据集的基础上，从网络拓扑结构、小基站间流量负载差异以及不同基站请求内容对象的相关性分析等方面证明了in-RAN协同缓存的可行性。然后，验证了在进行内容对象缓存和调度决策时应考虑不同的时间尺度。为了利用局域网内的协同缓存，同时满足缓存和调度决策的时间尺度要求，构造了一个考虑无线和回程实际传输约束的优化问题。在此基础上，提出了一种基于拉格朗日松弛分解、加速分支和界的联合缓存、无线和回程调度算法。基于实际数据集对该算法的性能进行了评估。结果描述了缓存容量、SBSs数量、SBSs连接概率与目标性能之间的关系，表明与现有算法相比，所提算法可以获得更好的性能。
This paper analyzes the traffic of a current LTE network in China and investigates the joint optimization of content object caching and scheduling for in-radio access network (RAN) caches. Cooperative caching has been well recognized as a way of unleashing the ultimate potential of in-RAN caches, yet its feasibility is still unexplored. Moreover, content object caching and scheduling are two key issues for cache deployment, which are usually jointly considered and resolved. However, they are triggered by different events with different time granularities. Therefore, on the basis of the real-world dataset, the feasibility of in-RAN cooperative caching is proved from aspects of network topology, traffic load difference among small base stations (SBSs) and correlation analysis of content objects requested at different SBSs. Then, it is verified that different time scales should be considered in making content object caching and scheduling decisions. To exploit in-RAN cooperative caching while meeting the time scale requirement in making caching and scheduling decisions, an optimization problem is constructed considering practical transmission constraints in wireless and backhaul. It is proved to be a quadratic assignment problem, and then, a joint caching, and wireless and backhaul scheduling algorithm is proposed based on Lagrangian relaxation and decomposition, and hastening branch and bound. The performance of the proposed algorithm is evaluated based on the real-world dataset. Results depict the relationship among the cache capacity, the number of SBSs, the connection probability of SBSs and the objective performance, and show that the proposed algorithm can achieve better performance, compared with the existing algorithms.