Research On Interference Mitigation Via Collaborative Beamforming In UAV Networks

The unmanned aerial vehicle(UAV)has gained significant attention from domestic and international scholars as a platform for providing aerial wireless networks and communications.However,air-ground communication dominated by line of sight(Lo S)channel can cause stronger interference to terrestrial networks,leading to a decrease in communication performance between terrestrial devices.Conventional interference mitigation methods rely on power control and trajectory design,which may decrease the transmission rate and increase the energy consumption of UAVs.Thus,in this paper,we aim to study a novel interference mitigation method via collaborative beamforming(CB)in a UAV-enabled data collection network and propose corresponding algorithms for interference problems in two scenarios: the single base station Lo S channel model and the multiple base stations probabilistic Lo S channel model.On the one hand,we model the interference problem in the single base station UAV data collection scenario under the Lo S channel as a simple interference mitigation multi-objective optimization problem(SIMMOP).The SIMMOP simultaneously improves data transmission efficiency,reduces interference impact,and increases UAV network lifetime by jointly optimizing the selection of the target base station,hovering positions,and excitation current weights of UAVs.Accordingly,we propose a chaotic multi-objective multi-verse optimizer(CMOMVO)that can effectively solve the SIMMOP.CMOMVO introduces a solution update operator based on Sine chaotic mapping to achieve the update of discrete solution dimension.Furthermore,it introduces chaotic initialization and chaos-based parameter update operators to enhance the diversity of initial solutions and balance the exploration and exploitation abilities of the algorithm,thus improving its solution performance.The experimental results indicate that the proposed CMOMVO performs better than several other comparative algorithms.On the other hand,we model the interference problem in the multiple base stations UAV data collection scenario under the probabilistic Lo S channel as a complex interference mitigation multi-objective optimization problem(CIMMOP)and prove that it is NP-hard.To address the complexity and NP-hardness of the formulated CIMMOP,we propose a chaotic non-dominated sorting genetic algorithm-II(CNSGA-II).CNSGA-II introduces chaotic initialization,chaos-based hybrid solution crossover,and chaos-based hybrid solution mutation operators to improve the performance of initial solutions as well as the global and local exploration ability of the conventional NSGA-II algorithm.Moreover,we propose an improved strategy based on an elimination mechanism to exclude nonsensical solutions and enhance convergence accuracy.The experimental results indicate that the proposed CNSGA-II algorithm can significantly reduce interference in UAV networks.