A design method for second-order parallel adaptive Volterra filter

In recent years, the adaptive Volterra filter (AVF) based on discrete Volterra series has been used to identify a nonlinear system, remove nonlinear distortion, etc. However, when memory length becomes long, or when filtering high-order nonlinear characteristics, AVF has a problem that the calculation work becomes extremely large. This paper presents a type of parallel adaptive Volterra filter (PAVF) to alleviate the computational complexity. Designed PAVF is divided from AVF and connected in parallel, with parallel processing and the small size of the matrices and vectors used in PAVF, the amount of calculation can be cut down. Finally, to illustrate the effectiveness of the proposed method, computer simulation results are presented.