Research On Joint Interference Alignment And Power Allocation Algorithm In Cellular Network

Due to the limited frequency resources and interference in the network,it is very important to effectively improve the spectral efficiency.Interference alignment(IA)is an effective interference management technology and has been shown to be a powerful technique for achieving the optimal degrees of freedom of interference channel.Many IA algorithms in MIMO IC have been well studied,but they do not take the full advantage of the system's transmit power.This thesis mainly focuses on the joint design of interference alignment and power allocation in multi-cell cellular network scenarios.The research details and main contributions are as follows:(1)In symmetric MIMO IC,a joint interference alignment algorithm that allocates power in a modified water-filling manner is proposed,which takes the maximization of the power received in the interference-free subspaces as the objective function.Then the objective function is optimized on the Grassmann manifold to reduce the complexity.This algorithm takes into account both inter-link interference and inter-stream interference.Simulation results show that the proposed algorithm can achieve higher sum-rate than the minimization of interference leakage algorithm(Min-IL)with equal power allocation.(2)In order to reduce the channel state information(CSI)in the system,a central unit(CU)is introduced into the K-user MIMO IC and a centralized iterative algorithm is proposed.The optimality criterion is developed on the basis of the system sum-rate.By using the alternating optimization strategy,two iterative algorithms can be uesd to design the precoding and decoding matrix.Then the power allocation matrix is calculated by utilizing a fixed point algorithm.Simulation results show that the proposed algorithm can provide higher system sum rate than conventional IA algorithms at high SNRs.(3)In the scenario of MIMO-NOMA downlink network,a method of interference alignment based on beamforming and power allocation is proposed for the two-cell case.The method first uses beamforming-based interference alignment to eliminate inter-cell interference and inter-cluster interference and then uses a SIC technique to remove intra-cluster interference.Then a cognitive-radio power allocation method is used to ensure the QoS of edge users and thus to improve user fairness.Then the method is extended to the three cells.Simulation results show that this method can not only support more users,but also increase the capacity of the system.