本课题主要研究如何建立Femtocell分布式自组织资源分配模型，并提出复杂度低、易实施的自组织资源分配算法，为IMT-A系统的射频参数的自配置自优化提供理论依据，实现在保证用户服务质量的同时网络能效最大化。基于以上研究目标，（1）针对Macro-Femto双层网络建立基于凸定价效用的斯塔博克分层博弈的分布式能效模型，将资源自组织分配问题转化为单个基站效用优化问题，实现能效提升的同时降低跨层干扰；（2）针对致密部署区域的Femtocell网络层，建立基于联盟博弈的能效模型，将资源自组织管理问题转化为Femtocell网络的效用优化问题，实现全网总体能效的提升；（3）通过引入Q学习理论，提出建立femtocell网络的继承型学习模型，实现降低网络信令开销，提高系统的能效；（4）通过感知技术检测空闲或干扰小的子信道，建立干扰测量矩阵，实现子信道动态分配，在保证用户QoS的同时实现能效最优。

The main subject of our project is IMT-Advanced femtocell green self-organizing technology, which focused on solving the problem on how to build the model of the allocation of organizational distributed or hybrid (centralized and distributed) resources (sub-channels, power, etc.) of femtocell, and then propose a self-organizing resource allocation algorithm, low-complexity and easy to implement, to maximize the energy efficiency of the system and to ensure the user's quality of service. To implement the above objectives, we have to fulfill the following research: Firstly, we will build the hierarchical energy-efficient Stackelberg Game model for the two-tier network in IMT-Advanced systems, then by using of the distributed processing method to transform the problem of self-organization distribution of the resources into the utility optimization problem for a single base station to reduce the cross-layer interference and increase the energy efficiency; Secondly, we will establish the regional alliance game model based on energy efficiency for the dense deployment Scenarios, through converting the management of resources self-organization to the optimization of the effectiveness of the femtocell network to promote the whole network's energy efficiency; Thirdly, by means of the inherited type Q-learning theory, we will set up the femtocell network learning model to reduce the network signaling overhead and improve the energy efficiency of the whole system; Finally, we can build the interference measurements matrix to achieve the dynamic allocation of the sub-channel by adopting the perception technology to detect the idle or low-interfere sub-channel, to achieve the optimal energy efficiency goals and ensure the user's quality of service.