This paper studies the parameter estimation algorithms of a finite impulse response system with colored noise. To suppress the negative effects of the colored noises, a novel gradient-based algorithm is developed by means of the cost function of the c...
This paper studies the parameter estimation algorithms of a finite impulse response system with colored noise. To suppress the negative effects of the colored noises, a novel gradient-based algorithm is developed by means of the cost function of the continuous mixed p-norm (CMPN). It combines the p-norms for 1 6 p 6 2, which control the proportions of the error norms and generate an adjustable gain to adapt the data quality. Moreover, to improve the convergence rate, a CMPN multi-innovation gradient recursive algorithm is derived through expanding the innovation scalar to the innovation vector. Finally, two examples are given to demonstrate the validity of the proposed algorithms.