CN101632587A - Tread signal extracting method based on wavelet transformation - Google Patents

Tread signal extracting method based on wavelet transformation Download PDF

Info

Publication number
CN101632587A
CN101632587A CN200910183045A CN200910183045A CN101632587A CN 101632587 A CN101632587 A CN 101632587A CN 200910183045 A CN200910183045 A CN 200910183045A CN 200910183045 A CN200910183045 A CN 200910183045A CN 101632587 A CN101632587 A CN 101632587A
Authority
CN
China
Prior art keywords
signal
tread
wavelet
acceleration
acceleration signal
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.)
Pending
Application number
CN200910183045A
Other languages
Chinese (zh)
Inventor
庄建军
符懋敬
侯凤贞
展庆波
邵毅
宁新宝
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University
Original Assignee
Nanjing University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing University filed Critical Nanjing University
Priority to CN200910183045A priority Critical patent/CN101632587A/en
Publication of CN101632587A publication Critical patent/CN101632587A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

The invention relates to a tread signal extracting method based on wavelet transformation, comprising the following steps: recording an acceleration signal generated when a human body walks; dimensionally analyzing the acceleration signal by using the disperse wavelet transformation; reasonably selecting a wavelet basis and an analysis dimension; determining the suitable wavelet basis and a characteristic dimension; using a threshold value method to detect peak values on the characteristic dimension; calculating a time interval between any two alternate peak values; and finally forming left foot tread sequences and right foot tread sequences when the human body walks, namely tread signals. Compared with the prior art that the threshold value method is simply used for directly detecting the peak values of the acceleration signals, the tread signal extracting method greatly increases the detectable rate of peak value signals and further reduces the false detection and the leak detection of the peak values, can also ensure the accuracy of the extracted tread sequences even more serious noise interference exists in the original signals, and has an important meaning for the subsequent analysis of the tread sequences and a larger value in the theory modeling and the practical application of the tread sequences.

Description

A kind of tread signal extracting method based on wavelet transformation
Technical field
The present invention relates to the medical signals process field, is a kind of new technique of gait signal extraction, can be used as a pretreatment module and is applied in gait signal processing and the analytical system, is a kind of tread signal extracting method based on wavelet transformation.
Background technology
The current step of same foot and the formed sequence of the interval between next step when gait sequence is meant human body walking.As a class important physical signal of human body output, implying the information of a large amount of relevant human motion neuromodulation abilities in the gait sequence.By analysis, not only can extract the characteristic parameter of relevant Human Physiology and pathology, the control ability of all right appraiser's somatic motor nerve system to gait sequence.Studies show that along with the appearance of the aging and nervous system disease at age, tangible change can take place the rhythm and pace of moving things of body gait signal.
Usually, gait sequence can obtain by following manner: the acceleration signal that produces when the dynamic acquisition human body is normally walked, then it being carried out peak value detects, this is because each foot lands Shi Douhui and produces the maximum of an acceleration in vertical direction during human body walking, these peaked positions are decided successively, then calculate the interval between any two alternate peak values, and the left and right sides step attitude sequence that is produced just can constitute human body walking by these intervals the time.This shows that it is the key link that can gait sequence accurately obtain that the acceleration signal peak value detects, especially all the more so when interference is comparatively serious.
Traditional signal analysis technology is to be based upon on the basis of Fourier transformation.Fourier transformation uses multiple SIN function as basic function, though this function has positioning function preferably at frequency domain, can't describe the character of signal in time domain simultaneously.And wavelet transformation is as a kind of analytical method of multiple dimensioned, multiresolution, can be according to the width of the requirement of resolution being regulated automatically time window and frequency window.This adaptivity makes wavelet transformation have higher frequency resolution in the signal low frequency part, has higher temporal resolution at the signal HFS, still is that frequency domain all has the ability of describing the signal local feature in time domain.
Wavelet analysis is frontier that develops rapidly in current application mathematics and the engineering discipline, and through nearly 10 years exploratory development, important mathematical form system is set up, and theoretical basis is more sturdy.Compare with Fourier transform, wavelet transformation is the partial transformation of space (time) and frequency, thereby can information extraction from signal effectively.
1. continuous wavelet transform
The continuous wavelet transform of signal f (t) carries out the flexible and displacement of yardstick by basic small echo ψ (t) to signal itself and obtains, and wherein a and b are respectively scale factor and shift factor.If a and b constantly change, then can get one group of function ψ A, b(t):
ψ a , b ( t ) = 1 a ψ ( t - b a ) - - - ( 1 )
By ψ A, b(t) continuous wavelet transform that can get signal f (t) is defined as:
W f ( a , b ) = ∫ f ( t ) ψ a , b * ( t ) dt - - - ( 2 )
" * " expression conjugation in the following formula, W fThe wavelet transformation of expression signal.As basic small echo ψ (t) when satisfying admissible condition, can get inverse wavelet transform and be:
f(t)=∫∫W f(a,b)ψ a,b(t)db?da (3)
2. wavelet transform
Generally adopt the wavelet transform processing signals on computers, scale factor a and shift factor b are carried out discretization, make a = a 0 m , b = n b 0 a 0 m (m, n ∈ Z), then Li San basic small echo is defined as:
ψ m , n ( t ) = a 0 - 0.5 m ψ ( a 0 - m t - n b 0 ) - - - ( 4 )
Work as a 0=2, b 0=1, and a=2 jThe time (j ∈ Z), can get two of signal f (t) and advance wavelet transform and be defined as:
W 2 j f ( t ) = 1 2 j ∫ - ∞ + ∞ f ( t ) ψ ( τ - t 2 j ) dt - - - ( 5 )
The Mallat algorithm can realize that two of discrete sample signals f (n) advances wavelet transform, and this algorithm carries out multistage decomposition with f (n) according to different frequency channels, and process is as follows: ψ M, n(t) bank of filters that constitutes with H and G represents that wherein H is a low pass filter, and H={h j, j ∈ Z, G are high pass filter, and G={g j, then signal can be decomposed into:
A ( j ) ( n ) = Σ j ∈ Z h j A ( j - 1 ) ( n - 2 j - 1 l ) - - - ( 6 )
D ( j ) ( n ) = Σ j ∈ Z g j A ( j - 1 ) ( n - 2 j - 1 l ) - - - ( 7 )
A in the following formula (0)(n) expression discrete sample signals f (n), n ∈ Z.The catabolic process of signal as shown in Figure 1, D at different levels (j)(n) be the discrete detail signal of signal under yardstick j, A at different levels (j)(n) be smoothly approaching signal under the yardstick j.
The frequency band range of supposing primary signal f (n) is 0~f iHz, then h jAnd g jThe signal of each frequency range all can be decomposed into the low-frequency band SPACE V jWith the high frequency band space W jFig. 2 be in the Mallat algorithm signal at the sketch map of the shared bandwidth of each yardstick.
By above-mentioned derivation as can be known, the wavelet transformation of signal f (t) is actually and uses different basic functions on different yardsticks this signal to be similar to.Therefore, wavelet transformation can equivalence be one group of band filter, and the mid frequency of wave filter constantly moves to the low-frequency band space along with the increase of change of scale.After the band filter group was finished the step by step decomposition descending to signal band, the result of wavelet transformation can the variation characteristic of shows signal on special frequency band.This feature of wavelet transformation has great advantage it in the time frequency analysis of non-stationary, nonlinear properties.How wavelet transformation is applied in the extraction of gait signal, obtains gait information accurately, researching value is very arranged.
Summary of the invention
The problem to be solved in the present invention is: the signal processing technology of seeking a kind of advanced person, extracting left and right sides step attitude sequence exactly the acceleration signal that produces when making from the primary human body walking that contains much noise becomes possibility, for the extraction of the subsequent analysis processing of gait sequence and physiology, pathological parameter provides reliable quality assurance.
Technical scheme of the present invention is: a kind of tread signal extracting method based on wavelet transformation, and the acceleration signal that produces when utilizing three dimension acceleration sensor record human body walking carries out 0.2~35Hz bandpass filtering treatment, saves as document form; Read out the acceleration signal of preserving in the file, utilizing wavelet transform that original acceleration signal is carried out yardstick decomposes: wavelet basis and decomposition scale are chosen, determine suitable wavelet basis and comprise the characteristic dimension of the main energy of the gait signal rhythm and pace of moving things, then on characteristic dimension, utilize threshold method to carry out peak detection, after detecting all peak values, calculate the interval between any two alternate peak values, the left and right sides step attitude sequence that produces when finally forming human body walking, i.e. gait signal.
Wavelet transform carries out the yardstick decomposition to original acceleration signal and is specially:
1), selecting the db9 small echo is wavelet basis;
2), determine that characteristic dimension is 2 4
3), on the basis of having determined wavelet basis and characteristic dimension, original acceleration signal is carried out wavelet decomposition;
4), utilizing threshold method to carry out acceleration peak value detects, after all acceleration peak values all detect, calculate the interval that any alternate peak value occurs, draw current step and the interval between next step, the sequence that is made of all these intervals is required left foot or right crus of diaphragm gait signal.
The present invention writes analysis software with Matlab, extracts the gait signal from the acceleration signal file.
Though the pretreatment of original acceleration signal process bandpass filtering, but still comprise much noise.And with respect to noise frequency, the frequency of the acceleration signal that produces in the human body walking is lower.Therefore, the low frequency part of primary signal is only the useful signal that we need extract.From the angle of wavelet analysis, the low-frequency component correspondence in the primary signal higher decomposition level.Along with the progressively increase of decomposing level, the high-frequency noise that contains in the primary signal reduces gradually, makes acceleration signal correctly to separate, and then obtains accurate gait sequence.
In the identification of existing gait signal, the application of pair wavelet transformation is also arranged, the present invention and its technology contents have the difference of essence, the existing gait signal analysis and processing of using wavelet transformation to as if contain the image of human walking motion, utilize image processing techniques, seek suitable gait feature and sorting technique thereby extract means such as motion outline; The acceleration signal of the object of analyzing and processing for writing down among the human walking procedure among the present invention do not need photographic equipment record moving image, and the convenience that gait detects improves greatly.The inventive method is tested on one 20 people's small sample data base, and directly from original acceleration signal, survey peak value and compare with being beneficial to threshold method, the inventive method can bring up to 98.84 ± 1.17% from 90.72 ± 2.05% (means standard deviation) with the recall rate of peak signal.
In the processing of the acceleration signal that the present invention produces when wavelet transformation is applied to human body walking, utilize the multiple dimensioned of wavelet transform, the multiresolution characteristic is carried out yardstick to original acceleration signal and is decomposed, by to wavelet basis and decompose after the number of plies rationally chooses, thereby decomposite the characteristic dimension that comprises the main energy of the gait signal rhythm and pace of moving things exactly, then on characteristic dimension, utilize threshold method to carry out peak detection, after detecting all peak values, calculate the interval between any two alternate peak values, the left and right sides step attitude sequence that produces when finally forming human body walking.The inventive method with utilize threshold method directly the peak value of acceleration signal to be surveyed merely to compare, greatly improved the recall rate of peak signal, reduced the flase drop and the omission of peak value to the full extent; Even in primary signal, exist when comparatively severe noise is disturbed, this method also can guarantee the accuracy of the gait sequence that extracted, this subsequent analysis for gait sequence has great important, all has very big value in the theoretical modeling of gait sequence and practical application.
Description of drawings
Fig. 1 Mallat algorithm decomposed signal sketch map.
Each decomposition scale Seize ACK message bandwidth sketch map of Fig. 2 Mallat.
Fig. 3 is implementing procedure figure of the present invention.
Fig. 4 decomposes the acceleration signal design sketch for using different wavelet basiss.
Fig. 5 is the frequency content sketch map of the acceleration signal under the different decomposition yardstick.
Fig. 6 extracts the gait sequence design sketch on the characteristic dimension after the wavelet decomposition.
Fig. 7 directly extracts the gait sequence design sketch for utilizing threshold method from acceleration signal.
Fig. 8 is different db small echo form sketch map.
The specific embodiment
A kind of tread signal extracting method of the present invention based on wavelet transformation, specifically implementing procedure is as shown in Figure 3: the acceleration signal that produces when utilizing three dimension acceleration sensor record human body walking, and carry out 0.2~35Hz bandpass filtering treatment, save as document form.Read out the acceleration signal of preserving in the file, isolate with the close vertical direction of gait signal relation on acceleration signal.
Select suitable wavelet basis, so-called wavelet basis is exactly basic small echo, claims female small echo again, or is called for short small echo, and it has limited duration on the time domain, and average is zero characteristics.Fourier analysis is the sinusoidal signal that signal decomposition is become a series of different frequencies, and wavelet transformation is to be signal decomposition a series of through translation and flexible basic small echo.Basic small echo commonly used has Haar, Daubechies, Biorthogonal, Coiflets, Symlets, Morlets, Mexican Hat, Meyer small echo etc.Selecting for use which type of wavelet basis relevant with the characteristics and the form of signal itself during concrete decomposed signal, often is rule of thumb to determine with a large amount of repetition tests.Db small echo full name is the Daubechies small echo, is a kind of small echo that is most widely used, and it has tight support, quadrature and asymmetrical characteristic, thereby can well be applied in the occasion that discrete wavelet is analyzed.Daubechies is gang's base small echo, and naming rule is dbN, and wherein N represents exponent number, and wherein db1 is exactly the Haar small echo, and different db small echo forms as shown in Figure 8.According to experiment repeatedly, the present invention selects the db9 small echo, the contrast effect of db9 wavelet basis and other wavelet basis is seen Fig. 4, why the db9 wavelet basis can have good positioning accuracy to the peak value of acceleration signal, be because the Lipschitz index of its systematicness coefficient and acceleration signal about equally, makes both waveforms can coincide to greatest extent at the singular point place of signal.
Determine characteristic dimension, characteristic dimension is exactly to be hidden in the relevant best decomposition scale of the gait rhythm and pace of moving things in the acceleration signal with what we were concerned about, and it is corresponding with the frequency band of the actual walking pattern signal rhythm and pace of moving things.On the correlated basis of experiment, analyze and determine that characteristic dimension is 2 4, the contrast effect of different decomposition yardstick is seen Fig. 5, decomposition scale 3 expressions is 2 among the figure 4, it is good that the periodicity of signal peak keeps, and the peak value of hiding in the primary signal can reveal clearly, determines that it is characteristic dimension of the present invention.
On the basis of having determined wavelet basis and best features yardstick, original acceleration signal is carried out wavelet decomposition.On characteristic dimension 24, utilize threshold method to carry out acceleration peak value and detect, testing result as shown in Figure 6, wherein " ▲ " marks is the acceleration signal peak value that left foot forms when landing, what " ● " marked is right crus of diaphragm acceleration signal peak value.After all acceleration peak values all detect, then calculate the interval that any alternate peak value occurs: the sampling time * interval counts, just can draw current step and the interval between next step, the sequence that is made of all these intervals is exactly required left foot or right crus of diaphragm gait signal.
In order to contrast, threshold method is directly used in the peak detection of original acceleration signal, concrete effect as shown in Figure 7, " ▲ " marks the real acceleration signal peak value that threshold method extracts, " X " marks the signal peak of this method flase drop or omission.As can be seen from the figure, directly the threshold application method has to noisy data sensitive poor robustness, the shortcoming that error is big from original acceleration signal extraction gait sequence.And elder generation adopts the wavelet decomposition original acceleration signal, then the method for the peak value of detectable signal then can improve the recall rate of peak signal greatly on characteristic dimension, the left and right foot sequence accuracy of Huo Deing is higher, more genuine and believable in this way, can be used for subsequent analysis and processing reliably.

Claims (3)

1, a kind of tread signal extracting method based on wavelet transformation, the acceleration signal that produces when it is characterized in that utilizing three dimension acceleration sensor record human body walking carries out 0.2~35Hz bandpass filtering treatment, saves as document form; Read out the acceleration signal of preserving in the file, utilizing wavelet transform that original acceleration signal is carried out yardstick decomposes: wavelet basis and decomposition scale are chosen, determine suitable wavelet basis and comprise the characteristic dimension of the main energy of the gait signal rhythm and pace of moving things, then on characteristic dimension, utilize threshold method to carry out peak detection, after detecting all peak values, calculate the interval between any two alternate peak values, the left and right sides step attitude sequence that produces when finally forming human body walking, i.e. gait signal.
2, a kind of tread signal extracting method based on wavelet transformation according to claim 1 is characterized in that wavelet transform carries out the yardstick decomposition to original acceleration signal and is specially:
1), selecting the db9 small echo is wavelet basis;
2), determine that characteristic dimension is 2 4
3), on the basis of having determined wavelet basis and characteristic dimension, original acceleration signal is carried out wavelet decomposition;
4), utilizing threshold method to carry out acceleration peak value detects, after all acceleration peak values all detect, calculate the interval that any alternate peak value occurs, draw current step and the interval between next step, the sequence that is made of all these intervals is required left foot or right crus of diaphragm gait signal.
3, a kind of tread signal extracting method based on wavelet transformation according to claim 1 and 2 is characterized in that writing analysis software with Matlab, extracts the gait signal from the acceleration signal file.
CN200910183045A 2009-08-05 2009-08-05 Tread signal extracting method based on wavelet transformation Pending CN101632587A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN200910183045A CN101632587A (en) 2009-08-05 2009-08-05 Tread signal extracting method based on wavelet transformation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN200910183045A CN101632587A (en) 2009-08-05 2009-08-05 Tread signal extracting method based on wavelet transformation

Publications (1)

Publication Number Publication Date
CN101632587A true CN101632587A (en) 2010-01-27

Family

ID=41592043

Family Applications (1)

Application Number Title Priority Date Filing Date
CN200910183045A Pending CN101632587A (en) 2009-08-05 2009-08-05 Tread signal extracting method based on wavelet transformation

Country Status (1)

Country Link
CN (1) CN101632587A (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101828917A (en) * 2010-05-07 2010-09-15 深圳大学 Method and system for extracting electrocardiosignal characteristic
CN105378432A (en) * 2013-03-15 2016-03-02 谷歌公司 System and method for attitude correction
CN106529433A (en) * 2016-10-25 2017-03-22 天津大学 Queue pace uniformity evaluation method based on sound signals
CN106923839A (en) * 2013-09-19 2017-07-07 卡西欧计算机株式会社 Exercise assist device, exercising support method and recording medium
WO2017193595A1 (en) * 2016-05-13 2017-11-16 广州视源电子科技股份有限公司 Hypnotic state electroencephalogram signal extraction method and system
CN109561854A (en) * 2016-08-02 2019-04-02 美敦力公司 It is detected using the paces of accelerometer axis
CN110661529A (en) * 2019-11-06 2020-01-07 杭州姿感科技有限公司 Method and device for generating step amplitude sequence
CN111160185A (en) * 2019-12-20 2020-05-15 中国农业大学 Multi-scale time sequence remote sensing image trend and breakpoint detection method
CN111780780A (en) * 2020-06-16 2020-10-16 贵州省人民医院 Step counting method and device based on filter bank
CN111832332A (en) * 2019-03-29 2020-10-27 中国石油天然气集团有限公司 Mud pulse signal processing method and device
CN113673424A (en) * 2021-08-19 2021-11-19 华南理工大学 Method for judging walking gait cycle phase
CN115171886A (en) * 2022-07-25 2022-10-11 北京戴来科技有限公司 Frozen gait detection method and device based on random forest algorithm and storage medium

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101828917A (en) * 2010-05-07 2010-09-15 深圳大学 Method and system for extracting electrocardiosignal characteristic
CN105378432A (en) * 2013-03-15 2016-03-02 谷歌公司 System and method for attitude correction
CN105378432B (en) * 2013-03-15 2019-06-18 谷歌有限责任公司 System and method for attitude updating
CN106923839A (en) * 2013-09-19 2017-07-07 卡西欧计算机株式会社 Exercise assist device, exercising support method and recording medium
WO2017193595A1 (en) * 2016-05-13 2017-11-16 广州视源电子科技股份有限公司 Hypnotic state electroencephalogram signal extraction method and system
CN109561854A (en) * 2016-08-02 2019-04-02 美敦力公司 It is detected using the paces of accelerometer axis
CN109561854B (en) * 2016-08-02 2022-01-04 美敦力公司 Step detection using accelerometer axes
CN106529433A (en) * 2016-10-25 2017-03-22 天津大学 Queue pace uniformity evaluation method based on sound signals
CN106529433B (en) * 2016-10-25 2019-07-16 天津大学 Queue march in step degree evaluation method based on voice signal
CN111832332A (en) * 2019-03-29 2020-10-27 中国石油天然气集团有限公司 Mud pulse signal processing method and device
CN110661529A (en) * 2019-11-06 2020-01-07 杭州姿感科技有限公司 Method and device for generating step amplitude sequence
CN110661529B (en) * 2019-11-06 2023-05-30 杭州姿感科技有限公司 Method and device for generating step amplitude sequence
CN111160185A (en) * 2019-12-20 2020-05-15 中国农业大学 Multi-scale time sequence remote sensing image trend and breakpoint detection method
CN111160185B (en) * 2019-12-20 2023-11-10 中国农业大学 Multi-scale time sequence remote sensing image trend and breakpoint detection method
CN111780780A (en) * 2020-06-16 2020-10-16 贵州省人民医院 Step counting method and device based on filter bank
CN113673424A (en) * 2021-08-19 2021-11-19 华南理工大学 Method for judging walking gait cycle phase
CN113673424B (en) * 2021-08-19 2023-09-08 华南理工大学 Judgment method for walking gait cycle stage
CN115171886A (en) * 2022-07-25 2022-10-11 北京戴来科技有限公司 Frozen gait detection method and device based on random forest algorithm and storage medium

Similar Documents

Publication Publication Date Title
CN101632587A (en) Tread signal extracting method based on wavelet transformation
CN107817527B (en) Seismic exploration in desert stochastic noise suppression method based on the sparse compressed sensing of block
Feng et al. Denoising deep learning network based on singular spectrum analysis—DAS seismic data denoising with multichannel SVDDCNN
US6952649B2 (en) Petroleum exploration and prediction apparatus and method
CN106886044B (en) A kind of microseism first break pickup method based on shearing wave and Akaike's Information Criterion
CN104266894B (en) A kind of mine microquake signal preliminary wave moment extracting method based on correlation analysis
CN100493445C (en) Automatic testing method for traditional Chinese medical pulse manifestation characteristics parameter
CN109063741A (en) A kind of energy spectrum analysis method based on hibert curve transformation and deep learning
CN103049741A (en) Foot-to-ground acting force-based gait feature extraction method and gait identification system
CN103306671A (en) Four-quadrant reservoir type identification method and system
CN104360382A (en) Method for detecting oil and gas by aid of stacked seismic data
CN109613610B (en) Automatic picking method for microseism signal arrival time difference
CN104794440A (en) False fingerprint detection method based on multi-block multi-scale LBP
Yao et al. An adaptive seismic signal denoising method based on variational mode decomposition
CN106842324A (en) A kind of optical cable coupled noise drawing method based on the sparse optimization of time-frequency domain
Molino-Minero-Re et al. Application of a time-scale local Hurst exponent analysis to time series
CN112989481B (en) Method for processing stable visual image data of complex geological tunnel construction surrounding rock
CN104880730A (en) Seismic data time-frequency analysis and attenuation estimation method based on Synchrosqueezing transform
CN107436450A (en) A kind of seismic signal bandwidth broadning method based on continuous wavelet transform
CN106908855A (en) A kind of method that geochemistry element combinations are selected based on GIS spatial analysis
CN113687440A (en) Milnaciaceae vicky cycle-based ancient water depth quantitative recovery method and storage medium
Coffin Spatial frequency analysis of block letters does not predict experimental confusions
CN102880858B (en) Automatic identification method for coal petrography image
CN102866429A (en) Method for determining groundwater occurrence
Li et al. Magnetotelluric signal-noise separation method based on SVM–CEEMDWT

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C12 Rejection of a patent application after its publication
RJ01 Rejection of invention patent application after publication

Open date: 20100127