A lattice predictor based adaptive Volterra filter and a synchronized learning algorithm

This paper proposes a lattice predictor based adaptive Volterra filter and a synchronized learning algorithm. In the adaptive Volterra filter (AVF), the eigenvalue spread of a correlation matrix is extremely amplified, and its convergence is very slow for gradient methods. A lattice predictor is employed for whitening the input signal. Its convergence property is analyzed. Convergence is highly dependent on a time constant parameter used in updating the reflection coefficients. Furthermore, a synchronized learning algorithm is proposed, by which fast convergence and a small residual error can be achieved. Computer simulations, using colored signals and real speech signals, demonstrate that the proposed method is 10 times as fast as the DCT based AVF.