| By deploying small cells in conventional cellular networks, heterogeneous networks (HetNets) can meet the explosively increasing traffic demand in wireless networks, improve the indoor and hotspots coverage, and increase the network capacity. However, though small cells can improve the spectral efficiency, severe interference is also brought by the deployment of small cellsThe interference in HetNets can be categorized into two categories, i.e., the cross-tier interference between macro cells and small cells and the co-tier interference among small cells. These two kinds of interference significantly hinder the spatial reuse gain brought by small cells. However, the special structure and the complex interference situation of HetNets make interference management in HetNets a challenging problem. As a advanced interference processing technique, interference alignment (IA) s proved its ability in managing the interference in wireless networks. However, due to the characteristics of HetNets, applying IA to manage the interference in HetNets is more challenging than conventional cellular networks. Therefore, we choose to investigate the IA technique in HetNets. By exploiting the heterogeneity and partial connectivity of the HetNets, we propose a new IA scheme for HetNets. Meanwhile, we propose joint interference management schemes by incorporating IA with interference avoidance and interference cancellation so as to solve the complex interference problem induced by the random deployment of small cells. The main achievement and results of this dissertation are summarized as follows:1. We proposed an IA scheme for downlink MIMO HetNets so as to effectively manage the interference in HetNets and increase the network degrees of freedom (DoF). By exploiting the heterogeneity (the differences between macro base stations (BSs) and small cell BSs in transmission power, the number of equipped antenna and etc.) and the partial connectivity (the property that faraway small cells have weak mutual interference) of HetNets, we proposed a two-stage IA scheme for HetNets. In the first stage, according the heterogeneity of the HetNets, we designed IA scheme for the macro cell and a group of fully-connected small cells; In the second stage, we designed IA scheme for a group of partially-connected small cells so that they can transmit simultaneously with the cells in the first stage, thus effectively increase the DoF of HetNets. Due to the existence of cross-tier interference, the number of users served by the macro BS has significant influence on the network DoF, i.e., increasing the number of users served by the macro BS can increase the DoF of the macro cell, but will cause interference to more signal dimensions of pico cell links. Therefore, based on the proposed two-stage IA scheme, we analyzed the influence of the number of users served by the macro BS on network DoF. Particularly, we derived the condition under which serving one macro user achieves more DoF than serving multiple macro users, and based on the condition we designed an algorithm to find the optimal number of served macro users so as to maximize the system DoF. The simulation results show that the proposed IA scheme can significantly improve the system DoF.2. We proposed a joint interference alignment and avoidance for downlink HetNets to solve the complex interference induced by the random deployment of small cells. Focusing on the problem that small cells in HetNets are randomly deployed and IA is not achievable for the whole network due the limited antennas equipped by BSs and users, we proposed a joint interference alignment and avoidance scheme for HetNets. In the scheme, according the feasibility conditions of two-stage IA, we proposed a small cells scheduling scheme, which is responsible for scheduling the small cells that satisfy the feasibility conditions of the two-stage IA transmit in the same slot and scheduling the small cells that can not transmit simultaneously into different time slots for interference avoidance. The interference avoidance scheme is designed with the aim to minimize the number of time slots needed to schedule all links so as to guarantee the fairness among small cell users, and then maximize the number of scheduled links in these time slot to improve the spectrum efficiency. The simulation results show that the proposed scheme can significantly increase the sum rate of HetNets.3. We proposed a interference alignment and cancellation (IAC) scheme for the uplink of HetNets, which effectively addressed the uplink interference of HetNets. Focusing on the problem that the optimal interference cancellation order and the optimal IA scheme are hard to determine due to the random deployment of small cells. We developed a set of constraints for users, BSs, and interference cancellation, and formulated the optimal IAC scheme as an optimization problem based on the developed constraints. By giving the precoder design scheme, we proved the feasibility of the IAC scheme. The simulation results show that the proposed IAC scheme can significantly increase the number of data streams transmitted in the network. Then, in order to reduce the feedback overhead, we proposed an opportunistic interference alignment and cancellation (OIAC) scheme for HetNets. In OIAC scheme, by exploiting the multi-user diversity gain, we select the users that can align the interference in the interference subspace of BSs for data transmission. This reduces the feedback overhead significantly. In the user selection process, according to the characteristics of interference cancellation, we proposed a new user scheduling metric and a sequential user selection scheme so as to reduce the complexity of user selection. Meanwhile, we guarantee a target signal-to-interference-plus-noise ratio (SINR) for each selected user by performing user selection, decoder design and power control jointly in the user selection process. The simulation results show that the OIAC scheme can improve the achievable sum rate significantly with low feedback overhead.
|
PDF Full Text Request