An Improved All-Pass Filter Design Using Closed-Form Toeplitz-plus-Hankel Matrix
Yue‐Dar Jou, Fu‐Kun Chen · Journal of Computer Science Technology Updates · 2015
The least-squares design of infinite impulse response all-pass filter can be formulated as to solve a system of linear equations without directly computing a matrix inversion. The set of linear equations associated matrix is further expressed as a Toeplitz-plus-Hankel matrix such that the optimal filter coefficients are efficiently solved by employing a robust Cholesky decomposition or the split Levinson technique. This paper proposes closed-form expressions for efficiently computing Toeplitz-plus-Hankel matrix based on trigonometric identities. The closed-form expressions of the Toeplitz-plus-Hankel matrix can be directly evaluated as the passband edges are specified without sampling the frequency band as that of the previous computing-efficiency algorithm. The proposed new and simpler closed-form expressions are indicated from simulation results to accurately improve the design performance as well as achieve computational efficiency.