On Uplink Interference Management With Resource Allocation And Association Control

With the explosive growth of the mobile data traffic, operators deploy lots of femto base stations(FBSs) in the traditional cellular network, which calls heterogeneous networks(Het Nets). The deployment of FBSs can not only improve the utilization ratio of the frequency and the network coverage, but also alleviate the load of macro base station(MBS). The deployment of FBSs also introduces lots of challenges. It not only produces the serious cross-tier interference, but also leads the asymmetry of uplink and downlink: in Het Nets that MBS and FBSs are co-channel, the macro users can only receive the partial channel state information of interfered FBSs, while cause interference to all FBSs with the same frequency. So with the partial channel state information, the interference management in uplink networks is a worthy problem to study.The existing resources allocation algorithm has disadvantage of high overhead of information interaction and the existing association control algorithm has disadvantage of slow convergence speed. So, in this paper, we propose an improved resources allocation and association control algorithm with partial channel state information in uplink Het Nets. Firstly, by considering users' rate and fairness, we formulate objective function as maximizing the sum of the user's logarithm rate with joint considering the resources allocation and association control. And we decompose the objective function into two sub-problems: resources allocation and association control sub-problems. Secondly, we define the reference base station(RBS), and only use the channel state information of RBS and associated base station(BS) to allocate resources. The proposed resources allocation algorithm not only considers the rate of users in associated BS, but also takes the RBS's interference into consideration. And it realizes cross-tier interference coordination. Thirdly, lagrangian dual decomposition approach is proposed to model association control sub-problem. And history sub-gradient descent method(HSDM) is proposed to solve association control sub-problem. The proposed association control algorithm realizes interference transfer. Finally, simulations and analytical results show the proposed algorithm improves the system throughput. The resources allocation algorithm with RBS reduces the overhead of information measurement and interaction and the association control algorithm with HSDM improves the convergence speed of association control.