Method for calculating filter coefficient and filtering method

Abstract

The invention relates to a method for calculating a filter coefficient and a filtering method. By adopting the technologies of spectral decomposition and cepstrum, the invention directly decomposes a particular function in an inverse frequency domain to obtain a denominator function coefficient of the filter. In other words, by adopting the non-interative arithmetic, the embodiment not only can reduce the complexity of the coefficient calculation, but also can avoid the problem of coefficient convergence. In addition, a compensation function is comprised in the particular function and enablesthe Fourier transform with significantly reduced length to be applied when the spectral decomposition is carried out, so that the calculation amount of a receiving terminal is reduced and at the sametime the receiving terminal is kept in good effectiveness.

Description

technical field [0001] The present invention relates to a filter technology, and in particular to a method for calculating filter coefficients. Background technique [0002] In the communication system, the equalizer in the receiving end can be used to eliminate the interference of the channel to the received signal, and the equalizer type can be divided into infinite impulse response (Infinite Impulse Response, IIR) and finite impulse response (Finite Impulse Response , FIR). In terms of performance, the IIR equalizer can achieve the Minimum Mean-Square Error (MMSE), and its performance is obviously better than that of the FIR equalizer. However, when calculating the IIR equalizer coefficients, spectral factorization or other iterative algorithms, such as the Kalman algorithm, must be performed. Among them, if a non-iterative spectral decomposition technique is used to calculate the IIR equalizer coefficients, such as the cepstrum method (cepstrum method), it is necessary...

AI Technical Summary

Problems solved by technology

Among them, if a non-iterative spectral decomposition technique is used to calculate the IIR equalizer coefficients, such as the cepstrum method, a Fourier transform with a large length must be used, and a sufficiently high spectral sampling rate is used to ensure the accuracy of the decomposition. The resulting huge additional computational complexity has caused serious limitations in the application of the cepstrum method

If an iterative algorithm is used, the problems of time-consuming and coefficient convergence must be considered

Therefore, the IIR equalizer must spend a huge amount of calculation or time to calculate its coefficients; and, in terms of circuit design, the complexity of the IIR equalizer is relatively increased

In addition, the transfer function of the IIR equalizer is divided into a numerator function and a denominator function. When the pole of the denominator function of the IIR equalizer is too close to the unit circle on the plane, the IIR equalizer will easily suffer from instability.

Images