Research On Adaptive Filtering Theory And Its Application To Self-interference Cancellation Of Underwater Acoustic Communication

Adaptive filtering algorithm is one of the important technologies of digital signal processing,which is widely used in many fields,including image processing,speech processing and radar signal processing.The performance of the algorithm directly affects the filtering effect of the filter.Therefore,a series of adaptive filtering algorithms with good stability and fast convergence are proposed and applied.The adaptive filter algorithm can be used for self-interference cancellation.It dynamically adjusts the filter coefficient according to the actual situation,which can cancel the self-interference signal to the maximum extent,improve the quality of the transmitted signal and the performance of the communication system.However,due to the fixed step size of the traditional adaptive filtering algorithm,the stability and convergence of the algorithm cannot be taken into account at the same time,so that the performance of adaptive filtering algorithm in self-interference cancellation will be greatly reduced.To solve the above problems,this thesis aims to study the variable step size adaptive filtering algorithm with excellent performance,and apply it to the self-interference cancellation of underwater acoustic communication system to remove the self-interference.Therefore,the Least Mean p-Power(LMP)algorithm and the Maximum Correntropy Criterion(MCC)algorithm with good robustness to impulse noise are studied in this thesis.Specific research work is as follows:Firstly,to solve the problem of poor performance of traditional adaptive filtering algorithm due to constant step size,considering that softsign function and tanh function have similar properties,but softsign function does not contain exponential terms,and the calculation is easier.This thesis presents an Improved Variable Step-Size Least Mean p-Power(IVS-LMP)algorithm based on improved softsign Function.The idea is that in the initial stage of the algorithm,a larger step size is generated in order to accelerate the algorithm convergence speed,and then a smaller step size is generated as the algorithm convergence progresses,so as to ensure the low steady-state error of the algorithm while the algorithm convergence speed slows down.Under the disturbance of ocean pulse noise,the simulation results show that the proposed IVS-LMP algorithm can take into account both fast convergence rate and low steady-state error.Secondly,in view of the influence of kernel width and step size factor on the performance of the adaptive filtering algorithm based on the maximum correntropy criterion,this thesis selects the variable step size problem to improve the MCC algorithm.In this thesis,the Improved Variable Step-Size Maximum Correntropy Criterion Based on Improved Variable Kernel Width(IVSS-IVKWMCC)algorithm is proposed based on a high-performing Improved Variable Kernel Width Maximum Correntropy Criterion(IVKW-MCC)algorithm.The proposed algorithm constructs variable step size function by simplifying Gaussian kernel function and improving the variable kernel width function of IVKW-MCC algorithm,and updates the step size by moving weighted average method.In addition,an improved hyperbolic tangent function is introduced to accelerate the convergence of the algorithm.Simulation results show that the proposed IVSS-IVKWMCC algorithm can further accelerate the convergence rate while ensuring the low steady-state error under the disturbance of ocean pulse noise,and has excellent performance.Finally,based on the advantages of Orthogonal Frequency Division Multiplexing(OFDM)technology,such as high frequency band utilization rate,high spectral efficiency and strong anti-interference,it is used as the modulation mode of underwater acoustic communication in this thesis.At the same time,the proposed IVS-LMP algorithm and IVSS-IVKWMCC algorithm are applied to the self-interference cancellation of underwater acoustic full duplex communication system based on OFDM.The estimation performance of the two algorithms in underwater acoustic self-interference channels and the self-interference cancellation performance of the OFDM underwater acoustic full duplex communication system are analyzed.Simulation results show that the two improved variable step size algorithms can achieve better self-interference cancellation effect in the system and reduce the bit error rate of the system.