Effectiveness evaluation of multi-aircraft collaborative networking based on Mahalanobis distance
Hierarchical network communication is essential for multi-aircraft collaboration,and perfor-mance evaluation is critical for making optimization decisions.To address the limitations of traditional perfor-mance evaluation methods,which often overlook the variance in subsystem performance and use the average Euclidean distance as a benchmark,a novel multi-aircraft hierarchical network performance evaluation system grounded in Mahalanobis distance is introduced.Building on the establishment of an unmanned aerial vehicle(UAV)communication network performance evaluation system,Mahalanobis distance is leveraged to compre-hensively capture the random distribution of each node's performance within the communication network sys-tem.The proposed approach is instrumental in assessing the communication performance of multiple aircraft un-der complex scenarios by using the Mahalanobis distance performance value as the input for a backpropagation(BP)neural network.The BP neural network is further enhanced by incorporating a particle swarm optimization(PSO)algorithm,which considers the heterogeneous information within the hierarchical network.A Mahalanobis distance based network performance evaluation system is developed to assess the hierarchical network's perfor-mance.The results are employed to optimize the performance indices,distribution,and reliability of each node,thereby validating the effectiveness of the proposed method and achieving accurate enhancement of the communi-cation efficiency of the multi-aircraft network.