论文部分内容阅读
随着智能手机、平板电脑等智能终端设备的快速普及,无线网络流量呈爆炸式增长,其中占主导地位的视频流量的增长尤为显著,根据思科的预测,从2014年到2019年,移动视频的复合年增长率(Compound annual growth rate,CAGR)为66%。在无线网络中部署缓存被认为是应对流量爆炸式增长的一种有效解决方案。虽然已经有很多论文关注蜂窝网络中的内容缓存问题,但这些论文基本上都集中在内容缓存的性能优化和能量有效,而忽略了多个服务提供商(Service provider servers,SPSs)之间的缓存资源共享问题。然而从SPS的角度,在基站缓存流行的内容,不仅可以改善用户体验,还可以减少对于回程网带宽的需求以节约成本,因此SPS必须要考虑最佳的缓存空间需求量以获得最大的收益。本文我们主要考虑这一问题,即在基站部署缓存的假设前提下,多个SPSs如何有效的共享缓存资源。本文的创新点主要有以下几方面:l·本文的场景为一个基站和多个SPSs,系统被建模为寡头垄断市场,其中基站是产品(缓存空间)的提供方,以一定的价格(通过价格函数定义)向产品的需求方(SPSs)收取费用,SPSs共享基站的缓存空间。l·我们将SPSs对于缓存空间的竞争建模为一个动态的非合作博弈的古诺模型,并通过基于Newton-Raphson方法的迭代算法来获得最佳的缓存空间需求量(古诺模型的纳什均衡解)。l·仿真部分详细分析了不同参数下的这种动态缓存资源分配机制的性能和稳定性特征。
With the rapid popularization of intelligent terminal devices such as smartphones and tablets, the wireless network traffic has exploded. Among them, the predominance of the increase of video traffic is remarkable. According to Cisco’s forecast, from 2014 to 2019, mobile video Compound annual growth rate (CAGR) is 66%. Deploying caching in wireless networks is considered an effective solution to the explosive growth of traffic. Although there are a number of papers focused on content caching in cellular networks, these papers have mostly focused on performance optimization and energy efficiency of content caching while ignoring the caches between service provider servers (SPSs) Resource sharing problem. However, from the SPS perspective, caching popular content at the base station not only improves the user experience but also reduces the need for backhaul bandwidth to save costs, so the SPS has to consider the best buffer space requirements for maximum benefit. In this paper, we mainly consider the issue of how multiple SPSs share cache resources efficiently under the premise of base station deployment caching. The innovation of this paper mainly includes the following aspects: 1. The scenario in this paper is a base station and multiple SPSs. The system is modeled as an oligopolistic market, where the base station is the provider of the product (cache space) Price Function Definition) Charges to the product’s demand side (SPSs), which share the base station’s cache space. We model the competition for cache space as a dynamic non-cooperative game Cournot model and obtain the best buffer space requirement through the iterative algorithm based on the Newton-Raphson method (the Nash equilibrium of the Cournot model solution). l The simulation part analyzes in detail the performance and stability characteristics of the dynamic cache resource allocation mechanism under different parameters.