Research on Mathematical Modeling of Adaptive Matrix Decomposition Technology for Wireless Communication Networks

This project takes robust adaptive beamforming (RAB) as the research object to study the main factors of its performance degradation and even failure in the presence of coherent interference. The received codirectional quantities are decomposed into the left part of the array manifold matrix and the conjugate right multiplicative matrix P. When there is a coherent disturbance, the non-diagonal element of P characterizes the correlation between the signal and the interference, causing the RAB method to fail to work properly. The basic design and simulation of the two methods are carried out on the Xilinx ISE platform. The system is designed on Kintex-7 series XC7K325T. When the bandwidth of the input signal is equal, the calculation results of the two methods are not different. Finite fast beats can be regarded as a special coherent interference case. We construct P diagonal matrix algorithm. The RAB algorithm against non-smooth correlation interference is proposed based on the fitted covariance matrix. Simulation results show that this algorithm can still achieve the same effect without correlation, and has better convergence than the literature algorithm.