Method for designing coefficients of digital reconstruction filter with random sampling rate conversion

Abstract

The invention relates to a method for designing coefficients of a digital reconstruction filter with random sampling rate conversion. According to the method, the coefficients of the digital reconstruction filter are designed based on a reconstruction theory and an FIR filter window function design method, a window parameter and an intercepting parameter are selected through the minimum attenuation of an attenuation band, the multiphase branch number of the reconstruction filter is determined by the maximum sampling rate conversion factor, and an ideal reconstruction function is subjected to windowing and discrete sampling according to the parameters, so that the coefficients of the digital reconstruction filter are obtained and are quantized. The reconstruction filter with sampling rate conversion designed through the method provided by the invention is small in pass-band fluctuation and can better inhabit a mirror image, FPGA realization is easy, the design complexity is greatly reduced, and in addition, the filter coefficients are stored by reading a multiphase structure of a meter, random sampling rate conversion with high performance is realized with smaller resource consumption, the variable range of the sampling rate is large, and the realizing difficulty is small.

Description

technical field [0001] The invention relates to a method for designing reconstruction filter coefficients, in particular to a method for designing digital reconstruction filter coefficients suitable for arbitrary sampling rate conversion, and belongs to the technical field of digital signal processing. Background technique [0002] The symbol rate of modern software radio communication systems is mostly variable or even continuously variable. Both the modulation end and the demodulation end need to perform sampling rate conversion to save resources and reduce power consumption. For the variable or even continuously variable symbol rate, it is necessary to design an arbitrary sampling rate conversion reconstruction filter, and the performance of the filter determines the performance of the signal sampling rate conversion processing. [0003] The signal sampling rate conversion can be realized by a single-stage digital reconstruction filter. There are two ways to implement a ...

AI Technical Summary

Problems solved by technology

Polynomial approximation has Farrow structure based on Lagrangian interpolation and B-spline approximation: Farrow structure only needs to store fewer coefficients and does not need to calculate power values, but real-time calculation of coefficients will cause long delay; and the length is L B Each item of the B-spline interpolation is L B power calculation, the approximation is smoother at the edges

Although the polynomial approximation has less storage capacity, the real-time calculation of the coefficients makes the power consumption larger

Images