CN104539261B - A kind of interpolation filtering processing method of any sample rate conversion - Google Patents
A kind of interpolation filtering processing method of any sample rate conversion Download PDFInfo
- Publication number
- CN104539261B CN104539261B CN201410745446.7A CN201410745446A CN104539261B CN 104539261 B CN104539261 B CN 104539261B CN 201410745446 A CN201410745446 A CN 201410745446A CN 104539261 B CN104539261 B CN 104539261B
- Authority
- CN
- China
- Prior art keywords
- fir filter
- signal
- filter
- coefficient
- clock
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Complex Calculations (AREA)
Abstract
The present invention relates to a kind of interpolation filtering processing method of any sample rate conversion, this method step includes：(1) according to the bandwidth requirement of input signal and filtering accuracy coefficient, FIR filter tap coefficient table is produced；(2) 1 haplotype data clock signal is produced with NCO, and obtains the clock phase Phase of the clock signal；(3) signal reception is carried out using 1 haplotype data clock signal；(4) corresponding filter factor is extracted in FIR filter coefficient table according to clock phase Phase；(5) obtained filter factor will be extracted applied to FIR filter, the docking collection of letters number is filtered processing, you can the sample rate of input signal is transformed into system operating frequency fs, realizes the interpolation filtering processing of sample rate conversion；The interpolation filter used in this method is made up of a horizontal FIR filter and a bookbinding the ROM tables of filter coefficient, realizes that simply, consuming hardware resource is few, practical.
Description
Technical field
The present invention relates to digital filtering technique field, more particularly to a kind of interpolation filtering processing side of any sample rate conversion
Method.
Background technology
In the design of existing digital interpolation filter, signal sampling rate is changed into system work clock, in order to adapt to data
A kind of requirement of rate-compatible, method is the speed according to transmission data, changes DA work using clock chip is reconfigured
Clock, this method is limited by hardware platform；Another method is to be carried out for specific several speed using multiple filter
Integral multiple filtering interpolation, this method filter order is more, complicated, it is huge to expend resource, and data rate can not connect
Continuous variable, practicality is not strong.
The content of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide a kind of interpolation filtering of any sample rate conversion
The interpolation filter used in processing method, this method has filter coefficient by a horizontal FIR filter and a bookbinding
ROM tables are constituted, within each system work clock fs cycles, and interpolation filter is searched according to the clock phase phase after quantization
Coefficient table, real-time update FIR filter coefficient simultaneously carries out filtering operation, obtains the signal after the interpolation that sample rate is fs, this is interior
Insert filtering method and realize that simply, consuming hardware resource is few, practical.
What the above-mentioned purpose of the present invention was mainly achieved by following technical solution：
A kind of interpolation filtering processing method of any sample rate conversion, comprises the following steps：
1st, a kind of interpolation filtering processing method of any sample rate conversion, it is characterised in that comprise the following steps：
(1), according to the bandwidth B w of input signal and the signal resolution parameter K of setting, K group FIR filter coefficients are generated,
And the K groups filter coefficient is stored in FIR filter tap coefficient table；Wherein, the signal resolution parameter K=
2L, L is positive integer；
(2), according to the code check Rb and system clock frequency fs of input signal, 1 haplotype data clock signal is produced using NCO,
The frequency Rs=Rb of the data clock signal, the clock phase Phase of the data clock signal is equal to the NCO accumulated values
High L interception values, i.e., described clock phase Phase span is 0~K-1；Wherein, the bit wide of the NCO is M and M
≥L；
(3), the 1 haplotype data clock signal produced according to step (2) is received to input signal, obtains receiving signal；
(4) the clock phase Phase of clock value, is received according to signal, the FIR filter produced in step (1) is taken out
The tap coefficient of FIR filter is extracted in head coefficient table, wherein, if the clock phase Phase=k-1, in the FIR
In filter tap coefficients table extract obtain kth group low-pass FIR filter coefficient h 'k(n), wherein, k=1,2 ..., K, n=1,
2 ..., N, N are the tap number of the FIR filter；
(5) step (4), is extracted to obtained FIR filter coefficient to be applied in FIR filter, to obtaining in step (3)
Reception signal carry out interpolation filtering processing, obtain signal after interpolation filtering.
The interpolation filtering processing method of above-mentioned any sample rate conversion, in step (1), according to the bandwidth of input signal
The Bw and signal resolution parameter K of setting, generates K group FIR filter coefficients, and the K groups filter coefficient is stored in into FIR
In filter tap coefficients table, process is implemented as follows：
(1a), according to signal bandwidth Bw, one group of generation meets taking out for the FIR filter of filtering requirements in the signal bandwidth
Head coefficient h (n), wherein, n=1,2 ..., N, N is the tap number of the FIR filter；
(1b), K proportion function obtained according to the signal resolution parameter K of setting, wherein, proportion function described in k-th
ForK=1,2 ..., K, n=1,2 ..., N；
(1c), the FIR filter tap coefficient h (n) obtained to step (1a) carry out leaf transformation in N point discrete Fouriers, obtain
FIR filter receptance function H (jw)=DFT (h (n)), wherein, DFT () represents discrete Fourier transform；The receptance function H
(jw) N number of value is respectively h1,h2,…,hn,…,hN；
The receptance function H (jw) that (1d), the K proportion function obtained according to step (1b) and step (1c) are obtained, obtains K
Individual new receptance function, wherein, new receptance function described in k-th is H 'k(jw), H 'k(jw) N number of value is respectively h1′,
h2′,…,hn′,…,hN', wherein, hn'=hn×fk(n), k=1,2 ..., K, n=1,2 ..., N；
(1e), the K obtained to step (1d) new receptance functions carry out inverse Fourier transform and obtain K group filter coefficients,
Wherein, filter coefficient described in kth group is h 'k(n)=IDFT (H 'k(jw)), IDFT represents inverse discrete Fourier transform, k=1,
2,…,K；
(1f), K group filter coefficients will be obtained in step (1e) and be sequentially stored into tables of data, obtain FIR filter tap
Coefficient table.
The interpolation filtering processing method of above-mentioned any sample rate conversion, FIR filter tap coefficient table is bound in ROM tables
In.
The advantage of the present invention compared with prior art is：
(1), the ROM that the present invention has filter coefficient by a horizontal FIR filter and a bookbinding constitutes fraction and inserted again
Value filter, can realize the sample rate of different rates signal by signal frequency to system work by the fraction double interpolation wave filter
The sample rate conversion of working frequency, realizes simple in construction, and can meet the continuous variable requirement of input signal, practical.
(2) fraction double interpolation wave filter of the invention has filter coefficient by a horizontal FIR filter and a bookbinding
ROM tables composition, within each system work clock fs cycles, internal filter is searched according to the clock phase phase after quantization
Device coefficient table, real-time update FIR filter coefficient simultaneously carries out filtering operation, obtains the signal after the interpolation that sample rate is fs.Should
Wave filter implementation method is simple, and hardware resource used is few, can effectively overcome prior art to be limited by hardware platform, present invention side
Interpolation filter in method is simple in construction, operand is small；
(3) the ROM table sizes of the bookbinding filter coefficient used in the present invention can be according to the requirement of input signal resolution ratio
It is adjusted flexibly, performance requirement can be met, resource can be saved again, flexibility is strong.
Brief description of the drawings
Fig. 1 is the process flow block diagram of the interpolation filtering processing method of the present invention.
Embodiment
The present invention is described in further detail with specific embodiment below in conjunction with the accompanying drawings：
Shown in process flow block diagram of the present invention as shown in Figure 1, at the interpolation filtering that any sample rate of the invention is changed
Reason method, comprises the following steps：
(1), according to the bandwidth B w of input signal and the signal resolution parameter K of setting, K group FIR filter coefficients are generated,
And the K groups filter coefficient is stored in FIR filter tap coefficient table；Wherein, the signal resolution parameter K=
2L, L is positive integer；Implement step as follows：
(1a), according to signal bandwidth Bw, one group of generation meets taking out for the FIR filter of filtering requirements in the signal bandwidth
Head coefficient h (n), wherein, n=1,2 ..., N, N is the tap number of the FIR filter；
(1b), K proportion function obtained according to the quality coefficient K of setting, wherein, proportion function is described in k-thK=1,2 ..., K, n=1,2 ..., N；
(1c), the FIR filter tap coefficient h (n) obtained to step (1a) carry out leaf transformation in N point discrete Fouriers, obtain
FIR filter receptance function H (jw)=DFT (h (n)), wherein, DFT () represents discrete Fourier transform；The receptance function H
(jw) N number of value is respectively h1,h2,…,hn,…,hN；
The receptance function H (jw) that (1d), the K proportion function obtained according to step (1b) and step (1c) are obtained, obtains K
Individual new receptance function, wherein, new receptance function described in k-th is H 'k(jw), H 'k(jw) N number of value is respectively h1′,
h2′,…,hn′,…,hN', wherein, hn'=hn×fk(n), k=1,2 ..., K, n=1,2 ..., N；
(1e), the K obtained to step (1d) new receptance functions carry out inverse Fourier transform and obtain K group filter coefficients,
Wherein, filter coefficient described in kth group is h 'k(n)=IDFT (H 'k(jw)), IDFT represents inverse discrete Fourier transform, k=1,
2,…,K；
(1f), K group filter coefficients will be obtained in step (1e) and be sequentially stored into tables of data, obtain FIR filter tap
Coefficient table, the FIR filter tap coefficient table can be bound in ROM tables, be easy to hardware to realize, can be with horizontal FIR filter
It is used in combination, realizes the filtering interpolation processing of variable sampling rate.
In the present embodiment, FIR filter is set as the transversal filter of 32 ranks, i.e., with 32 delay cells, common N=
33 tap coefficients；And the signal resolution parameter K=256, i.e. positive integer L=8 of setting；
(2), according to the code check Rb and system clock frequency fs of input signal, 1 haplotype data clock signal is produced using NCO,
The frequency Rs=Rb of the data clock signal, the clock phase Phase of the data clock signal is equal to the NCO accumulated values
High L interception values, i.e., described clock phase Phase span is 0~K-1；Wherein, the bit wide of the NCO is M and M
≥L；The working frequency of the NCO is fs；
In the present embodiment, NCO bit wide is M=32, in the input signal resolution parameter K=256 of setting, clock phase
Position phase is the most-significant byte interception value of NCO accumulated values, that is to say, that the value of clock phase is 0~255；In the present invention,
It is the extraction sequence number that FIR filter coefficient is obtained using the clock phase, realizes that the coefficient of filtering interpolation is extracted.
(3), the 1 haplotype data clock signal produced according to step (2) is received to input signal, obtains receiving signal；
I.e. the sample rate of the reception signal is Rs, and the purpose of the present invention is exactly that the sample rate of the signal is uniformly converted into systematic sampling
Rate, and ensure that signal is undistorted；
(4) the phase Phase of clock value, the FIR filter tap system produced in step (1), are received according to signal
The tap coefficient of FIR filter is extracted in number table, wherein, if the clock phase Phase=k-1, in FIR filtering
In device tap coefficient table extract obtain kth group low-pass FIR filter coefficient h 'k(n), wherein, k=1,2 ..., K, n=1,
2 ..., N, N are the tap number of the FIR filter；
(5) step (4), is extracted to obtained FIR filter coefficient to be applied in FIR filter, to obtaining in step (3)
Reception signal carry out interpolation filtering processing, obtain signal after interpolation filtering, the sample rate of the signal is fs, i.e., more than at
The Interpolation of signals that sample rate is Rs by reason is filtered into the signal that sample rate is fs, so as to realize the conversion of sample rate.
In the present embodiment, interpolation filter of the invention has filter by the horizontal FIR filter of 32 ranks and a bookbinding
The ROM tables composition of ripple device coefficient.Within each system work clock fs cycles, according to the clock phase phase after quantization,
Filter coefficient is searched in ROM tables, real-time update FIR filter coefficient simultaneously carries out filtering operation, obtain sample rate inserting for fs
Signal after value.If quantified to each filter factor by 16, clock phase phase is quantified by 8, then is bound
The amount of storage of the ROM tables of FIR filter coefficient is 33*16*28=135168bit.
The present invention realizes that different rates signal interpolation is filtered, whole work(using the fraction double interpolation wave filter of variable coefficient
It is able to can be realized using FPGA.
It is described above, it is only the optimal embodiment of the present invention, but protection scope of the present invention is not limited thereto,
Any one skilled in the art the invention discloses technical scope in, the change or replacement that can be readily occurred in,
It should all be included within the scope of the present invention.
The content not being described in detail in description of the invention belongs to professional and technical personnel in the field's known technology.
Claims (2)
1. a kind of interpolation filtering processing method of any sample rate conversion, it is characterised in that comprise the following steps：
(1), according to the bandwidth B w of input signal and the signal resolution parameter K of setting, K group FIR filter coefficients are generated, and will
The K groups filter coefficient is stored in FIR filter tap coefficient table；Wherein, the signal resolution parameter K=2L, L is
Positive integer；Implement process as follows：
(1a), according to signal bandwidth Bw, one group of generation meets the tap system of the FIR filter of filtering requirements in the signal bandwidth
Number h (n), wherein, n=1,2 ..., N, N is the tap number of the FIR filter；
(1b), K proportion function obtained according to the signal resolution parameter K of setting, wherein, proportion function is described in k-th
(1c), the FIR filter tap coefficient h (n) obtained to step (1a) carry out leaf transformation in N point discrete Fouriers, obtain FIR filters
Ripple device receptance function H (jw)=DFT (h (n)), wherein, DFT () represents discrete Fourier transform；The receptance function H (jw)
N number of value be respectively h1,h2,…,hn,…,hN；
The receptance function H (jw) that (1d), the K proportion function obtained according to step (1b) and step (1c) are obtained, obtains K newly
Receptance function, wherein, new receptance function described in k-th be H 'k(jw), Hk' (jw) N number of value is respectively h '1,h′2,…,
hn′,…,hN', wherein, hn'=hn×fk(n), k=1,2 ..., K, n=1,2 ..., N；
(1e), the K obtained to step (1d) new receptance functions carry out inverse Fourier transform and obtain K group filter coefficients, wherein,
Filter coefficient described in kth group is h 'k(n)=IDFT (H 'k(jw)), IDFT represents inverse discrete Fourier transform, k=1, and 2 ...,
K；
(1f), K group filter coefficients will be obtained in step (1e) and be sequentially stored into tables of data, obtain FIR filter tap coefficient
Table；
(2), according to the code check Rb and system clock frequency fs of input signal, 1 haplotype data clock signal is produced using NCO, it is described
The frequency Rs=Rb of data clock signal, the clock phase Phase of the data clock signal are equal to high L of NCO accumulated values
Interception value, i.e., described clock phase Phase span is 0~K-1；Wherein, the bit wide of the NCO is M and M >=L；
(3), the 1 haplotype data clock signal produced according to step (2) is received to input signal, obtains receiving signal；
(4) the clock phase Phase of clock value, the FIR filter tap system produced in step (1), are received according to signal
The tap coefficient of FIR filter is extracted in number table, wherein, if the clock phase Phase=k-1, in FIR filtering
In device tap coefficient table extract obtain kth group low-pass FIR filter coefficient h 'k(n), wherein, k=1,2 ..., K, n=1,
2 ..., N, N are the tap number of the FIR filter；
(5) step (4), is extracted to obtained FIR filter coefficient to be applied in FIR filter, to connecing for obtaining in step (3)
The collection of letters number carries out an interpolation filtering processing, obtains signal after interpolation filtering.
2. a kind of interpolation filtering processing method of any sample rate conversion according to claim 1, it is characterised in that：FIR
Filter tap coefficients table is bound in ROM tables.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410745446.7A CN104539261B (en) | 2014-12-08 | 2014-12-08 | A kind of interpolation filtering processing method of any sample rate conversion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410745446.7A CN104539261B (en) | 2014-12-08 | 2014-12-08 | A kind of interpolation filtering processing method of any sample rate conversion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104539261A CN104539261A (en) | 2015-04-22 |
| CN104539261B true CN104539261B (en) | 2017-08-29 |
Family
ID=52854748
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410745446.7A Active CN104539261B (en) | 2014-12-08 | 2014-12-08 | A kind of interpolation filtering processing method of any sample rate conversion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104539261B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105306008A (en) * | 2015-11-13 | 2016-02-03 | 北京遥测技术研究所 | Interpolation filtering processing method of arbitrary sampling rate conversion |
| CN108871385B (en) * | 2017-05-12 | 2021-09-07 | 西门子公司 | Encoder, motor, encoder data processing method, and storage medium |
| CN108366320A (en) * | 2018-01-08 | 2018-08-03 | 联创汽车电子有限公司 | Vehicle-mounted feedforward active noise reduction system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1192609A (en) * | 1997-01-10 | 1998-09-09 | 哈里公司 | Re-sampling circuit and associated method |
| CN1226888C (en) * | 2001-09-07 | 2005-11-09 | 三星Thales株式会社 | Block interpolating filter structure using seeking table |
| CN103650445A (en) * | 2011-03-09 | 2014-03-19 | 联邦科学技术研究组织 | Arbitrary sample rate conversion for communication systems |
-
2014
- 2014-12-08 CN CN201410745446.7A patent/CN104539261B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1192609A (en) * | 1997-01-10 | 1998-09-09 | 哈里公司 | Re-sampling circuit and associated method |
| CN1226888C (en) * | 2001-09-07 | 2005-11-09 | 三星Thales株式会社 | Block interpolating filter structure using seeking table |
| CN103650445A (en) * | 2011-03-09 | 2014-03-19 | 联邦科学技术研究组织 | Arbitrary sample rate conversion for communication systems |
Non-Patent Citations (3)
| Title |
|---|
| "一种用于符号同步的奈奎斯特滤波器优化设计";叶双应 等;《电子器件》;20060630;第29卷(第2期);第493-496页 * |
| "基于FPGA的EDSL系统QAM解调器的研究与实现";杨霖;《中国优秀硕士学位论文全文数据库 信息科技辑》;20091115(第11期);第四章第4.2-4.3节 * |
| "基于Gardner算法的位定时同步研究";朱雪阳;《中国优秀硕士学位论文全文数据库 信息科技辑》;20100915(第09期);第3章 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104539261A (en) | 2015-04-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104539262B (en) | A kind of digital fabrication filter processing method of continuous variable speed | |
| CN104539261B (en) | A kind of interpolation filtering processing method of any sample rate conversion | |
| CN101567701B (en) | High efficient multi-path digital down converter system | |
| CN103873016B (en) | A kind of arbitrarily sampling rate conversion digital reconstruction filter coefficient method for designing | |
| CN104467739B (en) | The adjustable digital filter of a kind of bandwidth, center frequency point and its implementation | |
| CN101257482A (en) | Method and device for realizing digital baseband variable velocity to convert modulating system | |
| CN104506164B (en) | Method for optimally designing graph filter banks on basis of two-step process | |
| CN105337587B (en) | A kind of non-maximum extraction system synthesis filter group building method based on DFT | |
| CN103268308B (en) | Support calculating device and the method for mixed base DFT | |
| CN102064797B (en) | Parallel implementation method and device for fractional sampling rate transformation | |
| CN104393854A (en) | FPGA-based time division multiplexing cascaded integrator-comb decimation filter and realization method thereof | |
| CN115473512B (en) | Parallel timing synchronization method based on polyphase filter bank structure | |
| CN110365351B (en) | Method and device for realizing channelized reception based on filter multiplexing architecture | |
| CN104883157B (en) | A kind of variable subband digital filter | |
| CN104301085B (en) | A kind of Perfect Reconstruction Substrip analysis device of resource multiplex | |
| CN204290909U (en) | The digital filter that a kind of bandwidth, center frequency point are adjustable | |
| CN104753534A (en) | Device and method for expanding the bandwidth of ADC sampling | |
| CN106130581A (en) | A kind of multiphase filtering wideband digital channel receiver improves system | |
| CN109474356B (en) | Broadband multi-channel signal energy detection system and method | |
| CN105306008A (en) | Interpolation filtering processing method of arbitrary sampling rate conversion | |
| CN103001586A (en) | Broadband two-channel digital down converter | |
| Zhang et al. | The design of digital down converter based on FPGA | |
| Salgado et al. | Power and area efficient comb-based decimator for sigma-delta ADCs with high decimation factors | |
| CN209267558U (en) | A kind of matched filter multiplexer in multiple signals digital receiver | |
| CN110690909B (en) | Low-complexity dynamic non-uniform channelized user separation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |