CN106197480A - A kind of processing system of Low SNR signal - Google Patents
A kind of processing system of Low SNR signal Download PDFInfo
- Publication number
- CN106197480A CN106197480A CN201610496936.7A CN201610496936A CN106197480A CN 106197480 A CN106197480 A CN 106197480A CN 201610496936 A CN201610496936 A CN 201610496936A CN 106197480 A CN106197480 A CN 106197480A
- Authority
- CN
- China
- Prior art keywords
- noise
- data
- signal
- parameter
- unit
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D3/00—Indicating or recording apparatus with provision for the special purposes referred to in the subgroups
- G01D3/028—Indicating or recording apparatus with provision for the special purposes referred to in the subgroups mitigating undesired influences, e.g. temperature, pressure
- G01D3/032—Indicating or recording apparatus with provision for the special purposes referred to in the subgroups mitigating undesired influences, e.g. temperature, pressure affecting incoming signal, e.g. by averaging; gating undesired signals
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
- Indication And Recording Devices For Special Purposes And Tariff Metering Devices (AREA)
Abstract
The present invention provides the processing system of a kind of Low SNR signal, it is characterised in that including: sensor unit, for physical signalling is converted to the signal of telecommunication；Analogue signal amplification/conditioning unit, for the analogue signal 1 being converted to through sensor unit being converted, provides for rear class ADC sampling unit and treats ADC converting analogue signals 2；ADC sampling unit, for the circuit unit with ADC modulus conversion chip as core, for by amplifying through analogue signal/conditioning unit converts the analogue signal 2 obtained and completes the analogue signal conversion to digital signal；PLD logical device, for programmable logic device (CPLD) or FPGA, the digital signal being converted to through ADC sampling unit is delivered to PLD logical device through digital signal channel 1 and is processed；Processor, for DSP or high-performance CPU, has been used for computational algorithm, and has completed the function that physical parameter calculates.This processing system is can to meet requirement of real-time, calculated performance particularly noise reduction capability, the Noise Signals Processing System of flexible structure simultaneously.
Description
Technical field
The invention belongs to transducing signal process field, relate to the processing system of a kind of Low SNR signal, specifically, originally
Invention can be used for the applications such as radar, voice, image, biomedicine, earthquake, and can process Low SNR signal in real time and obtain
Obtain high-quality result.
Background technology
At present, in the application practice that transducing signal processes, signal is generally mixed with noise, thus can be at extensive time multiplexed signal
Cover some important information.Relating to radar, sound, communication, earthquake, biomedical signal processing field, no matter front end is treated
Measurement is optics, acoustics, electromagnetic signal, signal of video signal, all can run into such situation.To this end, be developed many special
Algorithm for enhancing signal signal to noise ratio tackles this application scenarios, such as Kalman filtering, adaptive mesh filtering, small echo
The methods such as conversion, time-frequency method, these methods are better than the filter of traditional filtering algorithm such as medium filtering, average in performance
Ripple, fir filtering etc., but often calculation procedure is complicated, it is many to consume calculating resource, and this is for using embedded system in practice
Realize processing in real time of signal and propose challenge.
It was noticed that the signal to noise ratio that a kind of common scene is measured signal is not constant in applying at the scene,
Due to the multiformity of noise source, noise signal often changes along with test environment and the change of noise coupling, now, uses
Specific filtering algorithm parameter can not meet application requirement；And the application of physical quantity reflected signal is actively sent in some detections
Occasion, such as radar and sensory field of optic fibre, the signal to noise ratio principle of pending signal sequence is exactly time-varying, is i.e. in detection
The Signal-to-Noise of " far-end " substantially can be less than " near-end "；Due to the electronic component sensitivity to temperature, at different temperatures bar
Under part, the signal to noise ratio of measured signal as electronic component or the temperature-sensing property of LNA circuit and different.
More than Zong He, the real time processing system of reply time-varying Low SNR signal need to possess following feature, meets application real
The requirement of time property, the signal of low signal-to-noise ratio can be strengthened, Signal-to-Noise calculates dynamic when changing and follows the tracks of and obtain reason
The result thought.
Summary of the invention
It is an object of the invention to: improve the signal processing mode of band noise data, it is provided that one can meet real simultaneously
The requirement of time property, calculated performance particularly noise reduction capability, the Noise Signals Processing System of flexible structure, it is provided that corresponding hardware solution
Certainly scheme.
In order to achieve the above object, the technical solution used in the present invention is: the processing system of a kind of Low SNR signal, its
It is characterised by, including:
Sensor unit, for being converted to the signal of telecommunication by physical signalling；
Analogue signal amplification/conditioning unit, for carrying out level bias change by the analogue signal 1 being converted to through sensor unit
Change, differential transformation, coupling transform or voltage x current range conversion, provide for rear class ADC sampling unit and treat that ADC converting analogue is believed
Numbers 2；
ADC sampling unit, for the circuit unit with ADC modulus conversion chip as core, is used for amplify through analogue signal/nursing one's health
The analogue signal 2 that unit conversion obtains completes the analogue signal conversion to digital signal；
PLD logical device, for programmable logic device (CPLD) or FPGA, through the digital signal that ADC sampling unit is converted to
Deliver to PLD logical device through digital signal channel 1 process；
Processor, for DSP or high-performance CPU, has been used for computational algorithm, and has completed the function that physical parameter calculates；PLD logic
Bidirectional data path between device and processor is digital signal channel 2, and its implementation is high speed serialization line group, standard
The high speed parallel bus pci bus of high-speed serial bus PCIE or standard.
The processing system of Low SNR signal as above, it is characterised in that PLD logical device farther includes ADC
Interface circuit, prefilter, noise parameter extractor, processor data interface unit, its data stream packets includes original sampling data
Stream, pre-filtered data sequence；Digital signal that PLD logical device carries out processing is delivered to by ADC interface electricity through digital signal channel 1
Road resolves to original sampling data stream, and original sampling data stream is the numeral expression of the sampling of the physical quantity after digitized；Former
Beginning sampled data stream is processed as pre-filtered data sequence via prefilter, and prefilter uses can be former with streamline formal layout
The wave filter of beginning sampled data stream；Pre-filtered data sequence after prefilter processes delivers to noise parameter extractor, makes an uproar
Sound parameter extractor is that the software extracting noise data algorithm in real time data realizes；Processor data interface unit has been used for
Become the communication between PLD logical device and processor, functionally complete to send original sampling data stream, pre-filtered data sequence,
Extract noise parameter set from noise parameter extractor and manage device everywhere, and accept to come from two class data of processor: wave filter
Parameter configuration set and noise rating device configuration parameter set, the former is for configuring the calculating parameter of prefilter；The latter is used for
Parameter configuration in noise parameter extractor.
The processing system of Low SNR signal as above, it is characterised in that processor farther includes noise range and distinguishes
Other device, data slicer device, high-quality denoiser, data splicer, physical parameter computer；Its data stream packets includes noise characteristic ginseng
Manifold is closed, is extracted noise parameter set, noise range data acquisition system, noise data sheet, noise reduction data slice, noise reduction data sequence；Make an uproar
Acoustic feature parameter set pointer is to application-specific scene, special sensor, the typical characteristic of noise signal of special matched circuit
Set；Extract the noise ginseng extracted from noise parameter extractor that noise parameter set is in described PLD logical device
Manifold is closed；Noise range discriminator is for comparing noise characteristic parameter sets and extraction noise parameter set, it is judged that data
In which section belong to noise range, the evaluation of estimate of noise intensity is how many, and result is collected for noise range data acquisition system；Data are cut
Sheet device is for intercepting noise data sheet according to noise range data acquisition system in pre-filtered data sequence；High-quality denoiser is used for will
The noise filtering of noise data sheet, obtains noise reduction data slice；Data splicer is for substituting pre-filtered data by noise reduction data slice
Part corresponding in sequence, obtains noise reduction data sequence；Physical parameter computer is that sensor-based system realizes parameter sensing calculating
Unit, the result of calculation of usual physical parameter is shown to user by man machine interface.
The processing system of Low SNR signal as above, it is characterised in that prefilter be linear FIR filter,
One in medium filtering wave filter, mean filter wave filter.
The processing system of Low SNR signal as above, it is characterised in that noise parameter extractor uses extreme value to search
Rope algorithm.
The processing system of Low SNR signal as above, it is characterised in that processor data interface unit uses
Data-interface form is the LINKPORT interface of ADI TIGERSHARC series DSP.
The processing system of Low SNR signal as above, it is characterised in that noise range discriminator uses extreme value merger
With noise range discrimination algorithm, the extreme value density principle determined in foundation noise parameter set, i.e. extreme value in certain sampling interval
Point number is noise range more than theoretical value A, obtains the number of the size of noise range noise according to the numerical computations of extreme point density
Value is evaluated.
The processing system of Low SNR signal as above, it is characterised in that high-quality denoiser uses time-frequency peak value
Filtering algorithm.
The invention has the beneficial effects as follows: the present invention solve in the following manner in background technology propose some problem:
1) use special noise parameter extractor and noise range discriminator to quantify and the size of discernible signal signal to noise ratio, utilize
The noise range data acquisition system of gained is to data slicer, the most only to needing low signal-noise ratio data district to be processed to carry out at noise reduction
Reason.On the one hand the method alleviates the calculating resource pressure of high-quality denoiser unit, helps avoid as all Noise reducing of data bands
The risk that the process of calculating can not complete in real time, the most dynamically configuration denoiser performs parameter, improves high-quality denoiser
The specific aim calculated, it is to avoid use preset parameter to be calculated distortion or raw information distorted result.In a word, design more satisfied
System real time requires and the system of algorithm prescription.
2) executive component selecting high-performance processor to be high-quality denoiser, for the Real-time of high-quality denoiser
High-speed high-performance platform is now provided, and uses high-level language rapid and convenient to complete the exploitation of datatron, improve product development/
The production efficiency updated.
3) use prefilter to complete the filtering to primary signal, generally use the calculation that can realize with streamline
Such as medium filtering, FIR filter etc., use PLD device to realize this calculating and can guarantee that real-time performance, and without special storage
Resource consumption.
4) using prefilter to complete the filtering to primary signal, its major function is the routine for original sampled signal
Characteristic filtering, certain depth or smaller layers degree improve Signal-to-Noise, are not directed to the special disposal of Low SNR signal, retain
Such design cell can retain the process part in traditional design person to non-Low SNR signal, makes system remain to very well and adapts to
Application scenario under the conditions of this.
5) using prefilter to complete the filtering to primary signal, the filter parameter calculating gained according to subsequent cell is joined
Put set, dynamically adjust wave filter and perform parameter, adjust the filtering degree of depth, make prefilter can and can provide for high-quality denoiser
Preferably pre-processed results.
6) use noise parameter extractor to complete the low level to noise parameter to extract, conventional, use can be with streamline
The extreme value judgement that structure realizes calculates, and uses PLD device to realize this calculating and can guarantee that real-time performance, and without special storage
Resource consumption.
7) use noise parameter extractor to complete the low level to noise parameter to extract, calculate gained according to subsequent cell
Noise rating device configuration parameter set, dynamically adjusts extractor and performs parameter, make calculating gained extract noise parameter set dynamic
Updating, the most real-time serves subsequent calculations unit.
Accompanying drawing explanation
Fig. 1 is system hardware module annexation figure in application.
Fig. 2 is the PLD i.e. data flow diagram of digital logic device.
Fig. 3 be noise parameter extractor with "+" typical pattern after labelling.
Fig. 4 is the calculating data flow diagram of processor unit.
Fig. 5 is the PLD i.e. data flow diagram of digital logic device in case study on implementation.
Fig. 6 is the calculating data flow diagram of processor unit in case study on implementation.
Detailed description of the invention
In order to be more fully understood that the present invention, it is further elucidated with present disclosure below in conjunction with embodiment, but the present invention
Content is not limited solely to the following examples.The present invention can be made various changes or modifications by those skilled in the art, these
The equivalent form of value is equally within the scope of claims listed by the application limit.
The invention provides one utilizes FPGA (PLD) device and high performance signal processor to be combined into core
Digital signal processing hardware combined system, more importantly, with regard to characteristic allocation and the different calculations of combination of noise data, emphasizes
Algorithms of different calculates feature and the matching of device self character, makes the design property taken into account on the basis of appropriate design system sequence
The each side factors such as energy, function, cost.
A kind of processing system of Low SNR signal as shown in Figure 1, wherein:
Sensor unit is the functional unit being converted to the signal of telecommunication by other physical quantitys, and other physical signallings common can be shake
The physical quantitys such as dynamic signal, acoustical signal, optical signalling.
Analogue signal 1 is the conversion gained signal of telecommunication.
The function of analogue signal amplification/conditioning unit is mainly level bias conversion, differential transformation, coupling transform, voltage
Current range conversion etc., primarily to rear class ADC provide appropriate bandwidth, interior low noise, suitable voltage current range treat ADC
Converting analogue signals.
Analogue signal 2 is ADC sample objects；
ADC sampling unit is with ADC(i.e. modulus conversion chip) circuit unit as core, complete analogue signal to digital signal
Conversion.
The digital signal that digital signal channel 1 is converted to for ADC, according to manufacturer and the difference of chip model, signal can
For parallel signal, serial signal, differential serial signals group etc..
PLD logical device is PLD, usually CPLD or FPGA, corresponding to high speed and high amount of calculation
Application scenario, being generally selected FPGA is executive component.
Digital signal channel 2 is PLD device and the bidirectional data path of microprocessor (DSP/ high-performance CPU), and it realizes
Mode can be high speed serialization line group, the high-speed serial bus (such as PCIE) of standard, standard high speed parallel bus (as PCI is total
Line) etc., owing to present invention focuses on noise signal processing mode and corresponding Computational frame, the application specific details of this interface as in
Disconnected/inquiry waits and is not among the subjects to be discussed.
DSP/ high-performance CPU i.e. high-performance microprocessor has been mainly used to the high-quality computational algorithm of complexity, and complete
Become the parameters such as the function that physical parameter calculates, the such as heart rate/P ripple time of ECG signal process, the edge of optical fiber sensing equipment
Temperature/pressure distributed constant of optical fiber etc..
For calculation, PLD calculates data flow diagram as shown in Figure 2, wherein:
ADC interface circuit completes function typically has specific adc data format analysis to obtain real-time sampling data stream, concrete DIGEN
Serioparallel exchange, data encoding format transcoding, sequential coupling electricity, synchronous circuit etc. may be had according to ADC chip definition.
Original sampling data stream is the numeral expression of the sampling of the physical quantity after digitized.
Prefilter similar linear FIR filter, medium filtering wave filter, mean filter wave filter etc. can be with rule
Then form presses the wave filter of streamline formal layout pending data stream (herein for original sampling data stream), uses PLD device
The resource consumption cost realizing such wave filter is little, and the time consumption cost of pipeline processes is little, real-time.
Pre-filtering sequence is original sampling data stream result after prefilter processes.
Noise parameter extractor is that the software extracting noise data algorithm in real time data realizes, based on peakvalue's checking and
The algorithm of statistics, the effective calculation for one is directly perceived.Such as, utilize the data sequence of the distance of adjacent extreme point,
Can be with labelling noise parameter, its step is after maximum/minimizing coordinate, to be sequentially placed in sequence in obtaining data, then
Extrapolate the sequence at extrema elimination interval according to sample rate, its Computing Principle is described as, it is assumed that sampling non-uniform time is spaced apart T,
Coordinate sequence is [X1,X2,X3,X4…Xn], then noise parameter can be defined as the time interval of extreme value, T* [X2-X1,X3-X2,
X4-X3,…,Xn-Xn-1], data instance is, the extreme coordinates sequence of 100Mhz sample rate be [10,20,40,55,85,135,
147], then noise parameter is [100ns, 200ns, 150ns, 300ns, 500ns, 120ns ...], with "+" allusion quotation after labelling
Type pattern is as shown in Figure 3.
Processor data interface unit major function has been the communication between PLD device and processor, functionally completes
Transmission original sampling data stream, pre-filtered data stream, extraction noise parameter set are managed device everywhere, and are accepted to come from processor
Two class data, filter parameter configuration set and noise rating device configuration parameter set, the former is by configuring based on prefilter
Calculating coefficient or the exponent number of median filter of parameter, such as FIR filter, the latter's parameter in noise rating device unit is joined
Put, such as the minimum range of adjacent extreme value in the algorithm of peakvalue's checking and statistics.
For calculation, processor calculates data flow diagram as shown in Figure 4, wherein:
Noise characteristic parameter sets pointer is to application-specific scene, special sensor, the allusion quotation of noise signal of special matched circuit
Type feature, its essence be on the basis of sensing objects physical characteristic or combine physical circuit feature data describe, be right
In noise characteristic parameter such as the signal of certain type optical fiber sensor acquisition, the continuous print periodic peaks more than 3Mhz is
Typical noise, because normal physical process will not the superposition frequency signal higher than 3Mhz.3Mhz is noise parameter set
In a certain specific targets.
Extract noise parameter set and be in aforementioned PLD data stream same connotation.
Noise range discriminator refers to according to noise characteristic parameter sets and the comparative result of extraction noise parameter set, it is judged that number
According to, which section belongs to noise range, and the evaluation of estimate of noise intensity is how many accordingly, that is noise range data acquisition system.In principle,
After time series signal intercepts a segment signal according to both fixed step sizes, extract the maximum minimum point that noise parameter set shows
Sequence, extracts noise parameter set and is defined as extreme point time interval, for T* [X2-X1,X3-X2,X4-X3,…,Xn-Xn-1], and
In noise characteristic parameter sets, the upper limiting frequency correspondence time interval of F is that 1/F, i.e. extreme point are spaced apart T, the average time of sequence
It is spaced apart VT=(Xn-X1)/(n-1), meansigma methods V of definition time interval sequenceTPut off the time with noise characteristic parameter sets
The normalization difference (T-V of interval TT)/T is noise rating number, and this value span is real number field.This value is the biggest, and district is described
The intensity of territory noise is the biggest.This value is negative, and the least, illustrates that noise intensity is the least.
Such as, extract the maximum minimum point sequence that noise parameter set shows, extract noise parameter set and be defined as
Extreme point time interval, for [100ns, 200ns, 150ns, 300ns, 500ns, 120ns ...], and in noise characteristic parameter sets
The upper limiting frequency correspondence time interval of 1Mhz is that 1000ns, i.e. extreme point are spaced apart 500ns, and the average time interval of sequence is VT
=228, hence it is evident that less than 500ns, meansigma methods VT and the noise characteristic parameter sets of definition time interval sequence put off between the time
Normalization difference (T-V every TT)/T=0.54 is noise rating number, and this value span is real number field.
Data slicer device is the function intercepting noise data sheet according to noise range data acquisition system in pre-filtered data sequence
Unit, for microprocessor program, its execution form is deposit pre-filtered data from memorizer corresponding containing noise
Taking-up noise range, the contiguous memory address data of data field scope, the number that during generally this is pre-filtered data sequence, noise ratio is bigger
According to region.
High-quality denoiser, for filtering noise in noise data sheet, obtains the computing unit of noise reduction data slice, generally uses
Complex high-quality noise reduction algorithm can be only achieved preferable noise reduction, but simultaneously, calculate resource cost many, and calculate
Structure and process are complicated, are unfavorable for using PLD device pipeline computing to realize, and utilize the spy that microprocessor dominant frequency is high, internal memory is big
Point realizes this kind of algorithm, can obtain preferable effect.
Data splicer uses noise reduction data slice to substitute part corresponding in pre-filtered data sequence, obtains noise reduction data sequence
Row.
Physical parameter computer is the unit that sensor-based system realizes that parameter sensing calculates, according to the difference of application, in calculating
Hold difference, such as, be applied to the Temperature Distribution along optical fiber of Fibre Optical Sensor, for another example position etc. of target in Radar Signal Processing.Logical
Chang Di, the result of calculation of these physical parameters can be displayed to the user that by man machine interface.
For calculation, example PLD calculates data flow diagram as shown in Figure 5, wherein:
LVDS turns the specific implementation that parallel interface is ADC interface circuit, for turning through over level from ADC chip data bag
Change, obtain described original sampling data stream after serioparallel exchange.
FIR filter is prefilter specific implementation, can press streamline formal layout pending data stream with rule format
The wave filter of (being herein original sampling data stream), the resource consumption cost that use PLD device realizes such wave filter is little, and
The time consumption cost of pipeline processes is little, real-time.Acquired results is described pre-filtering sequence.
Extremum seeking algorithm is noise parameter extractor specific implementation, the effective calculation for one is directly perceived.Use
The resource consumption cost that PLD device realizes such wave filter is little, and the time consumption cost of pipeline processes is little, real-time.
Processor data interface unit major function has been the communication between PLD device and processor, data-interface shape
Formula is the LINKPORT interface of ADI TIGERSHARC series DSP.
For calculation, example processor calculates data flow diagram as shown in Figure 6, wherein:
Extreme value merger and the specific implementation that noise range discrimination algorithm is noise range discriminator, according to true in noise parameter set
Fixed extreme value density principle, i.e. in certain sampling interval, extreme point number is noise range, according to extreme point more than theoretical value A
The numerical value of density can be calculated the numerical Evaluation of the size of noise range noise.
Time-frequency method device is the realization of high-quality denoiser, time-frequency peak filtering, theoretical based on time frequency analysis
Cutting down white Gaussian noise, ultimate principle is that through frequency modulation(PFM), the signal amplitude fallen into oblivion under noise is become a normal amplitude
The instantaneous frequency of FM signal, goes out instantaneous frequency with the peak estimation of Wigner-Ville distribution, i.e. the amplitude of original signal, thus
Realizing signal to strengthen, it is computationally intensive, uses high-speed dsp to realize this algorithm and can meet requirement of real-time.
The content not being described in detail in this specification belongs to prior art known to professional and technical personnel in the field.
Claims (8)
1. the processing system of a Low SNR signal, it is characterised in that including:
Sensor unit, for being converted to the signal of telecommunication by physical signalling；
Analogue signal amplification/conditioning unit, for carrying out level bias change by the analogue signal 1 being converted to through sensor unit
Change, differential transformation, coupling transform or voltage x current range conversion, provide for rear class ADC sampling unit and treat that ADC converting analogue is believed
Numbers 2；
ADC sampling unit, for the circuit unit with ADC modulus conversion chip as core, is used for amplify through analogue signal/nursing one's health
The analogue signal 2 that unit conversion obtains completes the analogue signal conversion to digital signal；
PLD logical device, for programmable logic device (CPLD) or FPGA, through the digital signal that ADC sampling unit is converted to
Deliver to PLD logical device through digital signal channel 1 process；
Processor, for DSP or high-performance CPU, has been used for computational algorithm, and has completed the function that physical parameter calculates；PLD logic
Bidirectional data path between device and processor is digital signal channel 2, and its implementation is high speed serialization line group, standard
The high speed parallel bus pci bus of high-speed serial bus PCIE or standard.
The processing system of Low SNR signal the most according to claim 1, it is characterised in that PLD logical device is further
Including ADC interface circuit, prefilter, noise parameter extractor, processor data interface unit, its data stream packets includes original adopting
Sample data stream, pre-filtered data sequence；Digital signal that PLD logical device carries out processing is delivered to by ADC through digital signal channel 1
Interface circuit resolves to original sampling data stream, and original sampling data stream is the digital table of the sampling of the physical quantity after digitized
Reach；Original sampling data flows through and is processed as pre-filtered data sequence by prefilter, and prefilter uses can be with streamline form
Process the wave filter of original sampling data stream；Pre-filtered data sequence after prefilter processes is delivered to noise parameter and is extracted
Device, noise parameter extractor is that the software extracting noise data algorithm in real time data realizes；Processor data interface unit
For completing the communication between PLD logical device and processor, functionally complete to send original sampling data stream, pre-filtered data
Sequence, extract noise parameter set from noise parameter extractor and manage device everywhere, and accept to come from two class data of processor: filter
Ripple device parameter configuration set and noise rating device configuration parameter set, the former is for configuring the calculating parameter of prefilter；The latter
Parameter configuration in noise parameter extractor.
The processing system of Low SNR signal the most according to claim 2, it is characterised in that processor farther includes to make an uproar
Sound area discriminator, data slicer device, high-quality denoiser, data splicer, physical parameter computer；Its data stream packets includes noise
Set of characteristic parameters, extraction noise parameter set, noise range data acquisition system, noise data sheet, noise reduction data slice, noise reduction data sequence
Row；Noise characteristic parameter sets pointer is to application-specific scene, special sensor, the typical case of noise signal of special matched circuit
The set of feature；Extract noise parameter set be in described PLD logical device from making an uproar that noise parameter extractor extracts
Sound parameter sets；Noise range discriminator is for comparing noise characteristic parameter sets and extraction noise parameter set, it is judged that
In data, which section belongs to noise range, and the evaluation of estimate of noise intensity is how many, and result is collected for noise range data acquisition system；Number
According to food slicer for intercepting noise data sheet in pre-filtered data sequence according to noise range data acquisition system；High-quality denoiser is used
In the noise filtering by noise data sheet, obtain noise reduction data slice；Data splicer is for substituting pre-filtering by noise reduction data slice
Part corresponding in data sequence, obtains noise reduction data sequence；Physical parameter computer is that sensor-based system realizes parameter sensing meter
The unit calculated, the result of calculation of usual physical parameter is shown to user by man machine interface.
The processing system of Low SNR signal the most according to claim 2, it is characterised in that prefilter is linear FIR
One in wave filter, medium filtering wave filter, mean filter wave filter.
The processing system of Low SNR signal the most according to claim 2, it is characterised in that noise parameter extractor uses
Extremum seeking algorithm.
The processing system of Low SNR signal the most according to claim 2, it is characterised in that processor data interface unit
The LINKPORT interface that data-interface form is ADI TIGERSHARC series DSP used.
The processing system of Low SNR signal the most according to claim 3, it is characterised in that noise range discriminator uses pole
Value merger and noise range discrimination algorithm, according to the extreme value density principle determined in noise parameter set, i.e. at certain sampling interval
Interior extreme point number is noise range more than theoretical value A, obtains the big of noise range noise according to the numerical computations of extreme point density
Little numerical Evaluation.
The processing system of Low SNR signal the most according to claim 3, it is characterised in that when high-quality denoiser uses
Frequently peak filtering algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610496936.7A CN106197480B (en) | 2016-06-30 | 2016-06-30 | A kind of processing system of Low SNR signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610496936.7A CN106197480B (en) | 2016-06-30 | 2016-06-30 | A kind of processing system of Low SNR signal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106197480A true CN106197480A (en) | 2016-12-07 |
| CN106197480B CN106197480B (en) | 2019-01-29 |
Family
ID=57463282
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610496936.7A Expired - Fee Related CN106197480B (en) | 2016-06-30 | 2016-06-30 | A kind of processing system of Low SNR signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106197480B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108922561A (en) * | 2018-06-04 | 2018-11-30 | 平安科技（深圳）有限公司 | Speech differentiation method, apparatus, computer equipment and storage medium |
| CN110618464A (en) * | 2019-06-28 | 2019-12-27 | 中国地质大学（武汉） | System and method for improving Larmor precession signal-to-noise ratio of Overhauser magnetic sensor |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07131721A (en) * | 1993-11-01 | 1995-05-19 | Konica Corp | Digital still camera |
| KR20020086786A (en) * | 2001-05-11 | 2002-11-20 | 엘지이노텍 주식회사 | Radar system for preventing impact of vehicle |
| CN104181418A (en) * | 2014-08-19 | 2014-12-03 | 杭州亿恒科技有限公司 | Electric device noise testing device based on DSP and testing method thereof |
| CN104236911A (en) * | 2014-09-28 | 2014-12-24 | 江苏润仪仪表有限公司 | Train bogie bearing service process monitoring and fault diagnosis system and method |
| CN204288513U (en) * | 2014-10-27 | 2015-04-22 | 浙江省环境保护科学设计研究院 | A kind of intelligent environment noise and vibration monitoring device |
-
2016
- 2016-06-30 CN CN201610496936.7A patent/CN106197480B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07131721A (en) * | 1993-11-01 | 1995-05-19 | Konica Corp | Digital still camera |
| KR20020086786A (en) * | 2001-05-11 | 2002-11-20 | 엘지이노텍 주식회사 | Radar system for preventing impact of vehicle |
| CN104181418A (en) * | 2014-08-19 | 2014-12-03 | 杭州亿恒科技有限公司 | Electric device noise testing device based on DSP and testing method thereof |
| CN104236911A (en) * | 2014-09-28 | 2014-12-24 | 江苏润仪仪表有限公司 | Train bogie bearing service process monitoring and fault diagnosis system and method |
| CN204288513U (en) * | 2014-10-27 | 2015-04-22 | 浙江省环境保护科学设计研究院 | A kind of intelligent environment noise and vibration monitoring device |
Non-Patent Citations (1)
| Title |
|---|
| 许延丽 等: "基于DSP和FPGA技术的低信噪比雷达信号检测", 《现代电子技术》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108922561A (en) * | 2018-06-04 | 2018-11-30 | 平安科技（深圳）有限公司 | Speech differentiation method, apparatus, computer equipment and storage medium |
| CN110618464A (en) * | 2019-06-28 | 2019-12-27 | 中国地质大学（武汉） | System and method for improving Larmor precession signal-to-noise ratio of Overhauser magnetic sensor |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106197480B (en) | 2019-01-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105078505B (en) | Physiological single processing method and processing unit | |
| WO2021184805A1 (en) | Blood pressure prediction method and device using multiple data sources | |
| CN104305992B (en) | A kind of interactive fast automatic extracting method of Fetal ECG | |
| CN107184187B (en) | Pulse Wave Signal Denoising processing method based on DTCWT-Spline | |
| CN102663695B (en) | DR image denoising method based on wavelet transformation and system thereof | |
| CN104143035B (en) | A kind of method splitting breast lesion | |
| Zhao et al. | Parameter optimal determination for canny edge detection | |
| CN106815819B (en) | More strategy grain worm visible detection methods | |
| CN100493445C (en) | Automatic testing method for traditional Chinese medical pulse manifestation characteristics parameter | |
| CN117033911B (en) | Step counting analysis method based on intelligent glasses data | |
| WO2019041772A1 (en) | Electroencephalogram signal-based anesthesia depth monitoring method and system | |
| CN104182946A (en) | Spectral denoising method based on index soft and hard averaging threshold function | |
| CN107133938B (en) | Robust image fusion method based on wavelet coefficient statistical model | |
| CN105741305A (en) | Method and system for filtering electromyographical interference based on stationary wavelet transformation | |
| CN106197480A (en) | A kind of processing system of Low SNR signal | |
| CN105303538B (en) | A kind of Gaussian noise variance method of estimation based on NSCT and PCA | |
| WO2019075913A1 (en) | Signal processing system applied to remove otdr noise | |
| CN109949897A (en) | Display methods, device and the terminal device of monitoring curve | |
| Patil et al. | New approach of threshold estimation for denoising ECG signal using wavelet transform | |
| CN107392172B (en) | fetal heart sound processing method and processing device thereof | |
| CN115293219B (en) | Pulse signal denoising method integrating wavelet and kurtosis | |
| CN111134634A (en) | Photoelectric volume pulse wave analysis processing method based on cluster analysis | |
| CN110008881A (en) | The recognition methods of the milk cow behavior of multiple mobile object and device | |
| Malhotra et al. | Electrocardiogram signals denoising using improved variational mode decomposition | |
| Sreeja et al. | Detection of tumours in digital mammograms using wavelet based adaptive windowing 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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190129 Termination date: 20210630 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |