US 2004/0107078 A1Published application

Optimized filter parameters design for digital IF programmable downconverter

AssigneeChunghwa Telecom Co., Ltd.

InventorsChe-Sheng Chiu, Kai-An Tang, Su-Chin Hsieh, Chiung-Jang Chen

Life of this patent

Estimating the term…

Loading patent…

Abstract

The present invention discloses a method for finding an optimized filter parameters design to meet input specifications and hardware constraints in accordance with a typical single channel digital IF programmable downconverter. Said typical single channel digital IF programmable downconverter comprises four stages, including a high speed down-sampling stage, a spectral shaping stage, a rate matching stage and an oversampling stage. According to input specifications, said method firstly determines a number of Halfband interpolation filters used in said oversampling stage, secondly obtains optimized parameters for a CIC decimation filter and a Halfband decimation filter in said high speed down-sampling stage, using as many Halfband decimation filters as possible. Then determine if it is necessary to use re-sampling FIR filter for rate matching. Finally, use various function window to design programmable FIR filter for spectral shaping, obtains an overall synthesized spectral response which meets input specifications.

Claims13

Dependent claims

A method for finding an optimized filter parameters design for a digital IF programmable downconverter, comprises the following steps:

(1) receiving an input specification, said input specification comprising an input signal sampling rate, an input data rate, an oversampling factor R, an available number of Halfband decimation filters, a passband frequency, a stopband frequency, a passband ripple, a stopband attenuation and a optimal filter parameter satisfying said input specification;
(2) determining if said oversampling factor R is equal to 1;
(3) obtaining an largest available number nHB of Halfband decimation filters and setting an initial value of nHB equal to a total of Halfband decimation filters;
(4) obtaining an largest available first number of Halfband decimation filters satisfying said input specification and determining if re-sampling FIR filter is necessary;
(5) obtaining an optimized combination of Halfband decimation filters satisfying said input specification;
(6) obtaining a total n of window function for designing filters;
(7) obtaining a window function index j and setting said windows function index to 1;
and
(8) obtaining a programmable FIR filter satisfying said input specification.

6 more elements

Dependencies are inferred from the claim text.