Fixed-pole second-order parallel filters provide an efficient way of implementing IIR filters with a logarithmic frequency resolution. However, the fine frequency resolution needed at low frequencies can only be achieved by poles near the unit circle. This may lead to large roundoff noise at low frequencies when the filters are implemented using bit-depths of 24 bits or lower in fixed-point arithmetic. This paper investigates the performance improvement when the parallel second-order sections are implemented as warped IIR filters. In addition, an analytical expression is given for computing the warping parameter as a function of the pole location of the original second-order section so that the quantization noise power is minimized.
(See document for exact affiliation information.)
https://aes2.org/publications/elibrary-page/?id=18713
(230KB)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
