基于多维矩阵分解的 QSNP-LS算法的无线信号接收机

Wireless Signal Receiver Based on Multi-dimensional Matrix Decomposition QSNP-LS Algorithm

针对无人机群双跳模型下的通信系统,文中提出了一种基于多维矩阵的快启动嵌套并行最小二乘法(Quick Start Nested Parallel Least Square,QSNP-LS)接收机。该方法由多维矩阵建模、信号估计方法和可辨识性条件三部分组成。在发送端处,所提方法将经过Khatri-Rao空时编码的信号发送至无人机群,无人机群对接收的信号进行放大转发至基站,利用多维矩阵结构,在基站处形成嵌套多维矩阵模型,基于此模型实现了符号和信道的联合估计,中继处无需对信号进行处理,减轻了中继处的负担。仿真结果表明,与一些竞争算法相比,所提方法在误码率(Bit Error Rate,BER)和归一化均方误差(Normalized Mean Square Error,NMSE)方面具有显著的优越性。

For the communication system under the dual-hop model of unmanned aircraft cluster,this paper proposes a Quick Start Nested Parallel Least Square(QSNP-LS)receiver based on multidimensional matrix.The method is composed of multidimensional matrix modeling,signal estimation method and identifiability condition.At the transmitter,the proposed method sends the signals encoded by Khatri-Rao space-time code to the UAV group.The UAV group amplifies and forwards the received signals to the base station.Using the multidimensional matrix structure,a nested multidimensional matrix model is formed at the base station.Based on this model,the joint estimation of symbol and channel is realized.The relay does not need to process the signals,which reduces the burden of the relay.Simulation results show that the proposed method has significant advantages in Bit Error Rate(BER)and Normalized Mean Square Error(NMSE)compared with some competitive algorithms.