An interference-robust stochastic gradient algorithm with a gradient-adaptive step-size

An algorithm for adaptive FIR (finite impulse response) filters is proposed. This algorithm provides both fast convergence and small final misadjustment with an adaptive step size even under an interference to the error. It pays special attention to the interference which contaminates the error. To enhance robustness to the interference, it imposes a special limit on the increment/decrement of the step size. The limit itself is also varied according to the step size. The algorithm is extended for application to nonstationary signals. Simulation results with white signals show that the final misadjustment is reduced by as much as 20 dB under severe observation noise with a negligible expense of the convergence speed. An echo canceler simulation with a real speech signal exhibits its potential for a nonstationary signal.<>