Interference Mitigation In Heterogeneous Wireless Networks

In recent years,the amount of data and the number of devices for mobile communications have explosively increased.In order to meet the ever growing demands of wireless communication,academia and standardization organizations have proposed a variety of cuttingedge technologies to increase data rate and network capacity,such as dense heterogeneous networks(HetNets),massive multiple input multiple output(MIMO),and cognitive small cells,etc.The HetNets consisting of a massive enabled macrocell and plenty of small cells are regarded as one of the main technologies to achieve the performance requirements of future mobile networks.The macrocell can provide wide area coverage and mobility coverage.Small cells can provide higher data rate for hotspot areas,and densely deployed small cells can reuse the same spectrum resources in different regions,which will significantly improve spectrum efficiency.However,since the macrocell and small cells utilize the same band,the cross-tier interference between the macrocell and small cells,and the co-tier interference between small cells are becoming main challenges which restrict the deployment of dense HetNets.Interference in traditional HetNets can be avoided by inter-cell interference coordination(ICIC),enhanced inter-cell interference coordination(eICIC),further enhance inter-cell interference coordination(FeICIC),and cognitive radio(CR),etc.However,massive MIMO enabled HetNets have created some new features.First,the massive MIMO enabled macrocell base station can concentrate transmitting signals to intended users,and the cross-tier interference from the macrocell to small cells is restricted around scheduled users.Secondly,with the networks becoming denser,a macrocell user may be interfered by multiple small cell base stations,and the cross-tier interference from small cells to macrocell users will sharply decline the overall network performance.Thirdly,densely-deployed small cells will intensify the co-tier interference,especially in the case of small cell range expansion.Based on reasons above,this thesis focuses on interference mitigation schemes from three aspects.(1)In order to avoid the interference from small cell to macrocell users,we assume that small cell base stations are empowered with cognitive capabilities,which can perceive the channel occupation of the macrocell and access the channel opportunistically.Thus,exclusion zones(EZs)are formed around scheduled macrocell users.However,the cross-tier interference avoidance is at the expense of the performance of small cells in EZs.Therefore,there is a performance tradeoff between the macrocell and small cells in EZs.We propose a multi-user scheduling scheme of the macrocell,which considers not only the gains of the macrocell,but also the performance loss of small cells in EZs.The multi-user scheduling scheme is formulated as a weighted sum rate maximization problem.Since EZs formed by different scheduled users may intersected with each other,the optimization variables may be coupled.Therefore,we design an interfering small cell cluster segmentation approach to decompose the original problem into several subproblems based on each EZ.Finally,a greedy iterative user selection algorithm is designed to figured out scheduled users of the macrocell.Simulation results reveal that the proposed scheme can improve the performance of conventional cognitive HetNets.(2)In dense HetNets,user scheduling,precoder design or power allocation of small cells are crucial to optimize the network utility and to avoid inter-cell interference.Under the constraint of cross-tier interference,this thesis designs user scheduling,precoders and power allocations of small cells in a joint manner.The proposed scheme can not only constrain the cross-tier interference from small cell base stations to macrocell users,but also mitigate the co-tier interference between small cells by using interference cost matrix.The design of theses parameters is formulated as a multiple variables optimization problem.However,the problem is intractable to solve for two reasons.1)It contains both discrete and continuous variables;2)It is a global optimization problem,i.e.,the optimization targets are the parameters of all small cells.Therefore,we design an auxiliary variable(which is called the transmitting parameter)to characterize user scheduling,precoders and power allocation simultaneously.Thus,the original optimization problem is equivalent to an univariate version.Then,the equivalent global problem is decomposed into sub-problems based on each small cell according to the approach of interference cost matrix.Simulation results reveal that the proposed scheme outperforms conventional linear precoding schemes and nonlinear precoding schemes.Furthermore,it can achieve the performance close to coordinated multiple points(CoMP)with reduced overhead.(3)Interference coordination is an important means to mitigate interference in HetNets.With the excellent spatial focusing and resolution of massive MIMO,we extend the managing resources of eICIC from time domain to spacial domain,and constrain the application regions of eICIC into EZs.Then,we propose the exclusion zone assisted enhanced inter-cell interference coordination(EZ-eICIC)scheme.Moreover,we designed the configuration strategy of EZ-eICIC which arranges a customized eICIC configuration for each EZ independently.Then the EZ-eICIC configuration strategy is formulated as an optimization problem.By solving the problem,we derive the eICIC configuration criterion in each independent EZ.Finally,an iterative algorithm is proposed to solve the eICIC configurations of the entire networks.Simulation results reveal that the proposed EZ-eICIC scheme outperforms current eICIC scheme and FeICIC scheme in dense HetNets.