With the ever-increasing mobile video traffic, edge caching enabled passive optical networks (PONs), where storages are deployed on optical network units (ONUs), are becoming a desirable back-end for mobile backhaul (MBH). By caching videos on ONUs, video requests generated in MBH can be efficiently accommodated. Thus, many previous works focus on leveraging the caching capabilities of ONUs to maximize the video cache-hit ratio. However, it is largely ignored how to stimulate individual ONUs to contribute their storages for caching videos. In this paper, we propose a game-based incentive mechanism through which the content provider (CP) rewards ONUs for their storage-contributions. We firstly formulate a two-level Stackelberg game to capture the interaction between the CP and ONUs. At the first level, the CP determines rewarding policy to maximize its utility. At the second level, ONUs react with the amount of provided storages with the aim to cooperatively maximize the social utility or competitively maximize their respective utilities, and thereby two sub-games are further modeled to depict the cooperation and competition among ONUs. We then give the specific strategies of cooperative ONUs and competitive ONUs, respectively. We also show the existence of an equilibrium in the formulated Stackelberg game. The performance of the proposed incentive scheme is finally evaluated by simulations.

随着移动视频流量的不断增长，启用边缘缓存的无源光网络（PON）将存储部署在光网络单元（ONU）上，正成为移动回程（MBH）的理想后端。通过在ONU上缓存视频，可以有效地容纳MBH中生成的视频请求。因此，许多先前的工作集中在利用ONU的缓存功能来最大化视频缓存命中率。但是，如何激发各个ONU贡献其存储以缓存视频，在很大程度上被忽略了。在本文中，我们提出了一种基于游戏的激励机制，内容提供商（CP）通过该机制来奖励ONU的存储贡献。我们首先制定了一个两级Stackelberg游戏来捕获CP和ONU之间的交互。在第一层 CP确定奖励政策以最大程度地发挥其效用。在第二级，ONU与所提供的存储量进行反应，以合作地最大化社会效用或竞争性地最大化其各自的效用，从而进一步建模了两个子游戏来描述ONU之间的合作与竞争。然后，我们分别给出了合作ONU和竞争ONU的具体策略。我们还证明了制定的Stackelberg博弈中存在均衡。最后通过仿真评估了所提出的激励方案的性能。从而进一步对两个子游戏进行建模，以描述ONU之间的合作与竞争。然后，我们分别给出了合作ONU和竞争ONU的具体策略。我们还证明了制定的Stackelberg博弈中存在均衡。最后通过仿真评估了所提出的激励方案的性能。从而进一步对两个子游戏进行建模，以描述ONU之间的合作与竞争。然后，我们分别给出了合作ONU和竞争ONU的具体策略。我们还证明了制定的Stackelberg博弈中存在均衡。最后通过仿真评估了所提出的激励方案的性能。