Fractional delay optimization method applicable to multiple Nyquist regions and implementation structure thereof

Abstract

A fractional delay optimization method applicable to multiple Nyquist regions and an implementation structure thereof belong to the field of digital signal processing. First, fractional delay processing is carried out on an input data sequence to get a first output data sequence, and multiplication operation and Hilbert transformation are carried out on the input data sequence to get a second output data sequence. Then, the second output data sequence is subtracted from the first output data sequence to get an output data sequence after fractional delay applicable to multiple Nyquist regions. The multiplication operation is to multiply the input data sequence by a zoom factor A=(-1)<NZ>*ceil[(NZ-1) / 2]*2 Pai D, wherein NZ represents a Nyquist region where the frequency f(in) of the input data sequence is located relative to the sampling frequency f(s), ceil refers to rounding up (NZ-1) / 2 to an integer, and D represents the delay of a fractional delay filter. The application band range of the traditional fractional delay filter can be extended to a high Nyquist region to realize the fractional delay of multiple Nyquist regions. The structure is simple, and the method is easy to implement.

Description

technical field [0001] The invention relates to the field of digital signal processing, in particular to a fractional delay optimization method suitable for multiple Nyquist zones and the design of a fractional delay filter suitable for multiple Nyquist zones. Background technique [0002] Fractional delay filter, also known as continuously variable digital delay unit, is a digital filter with a delay of non-integer samples, widely used in speech coding and synthesis, digital communication, delay estimation, and sampling rate conversion and other fields. [0003] There are relatively mature design methods for fractional delay filters, such as the design method based on polynomials (Farrow structure), the design method based on Lagrangian interpolation method, and the design method based on least squares method. However, the disadvantage of the fractional delay filter designed by the traditional method is that its application frequency band is in the first Nyquist zone (0 ~ ...

AI Technical Summary

Problems solved by technology

However, the disadvantage of the fractional delay filter designed by the traditional method is that its application frequency band is in the first Nyquist zone (0 ~ fs / 2, fs represents the sampling rate of the digital signal), and the application frequency band refers to the filter The input signal frequency (fin) range that the filter can handle, so when fin exceeds fs / 2, the fractional delay filter cannot work normally, which greatly limits the application range of the fractional delay filter

Images