CN114259225B - Identity recognition method and system based on millimeter wave radar - Google Patents
Identity recognition method and system based on millimeter wave radar Download PDFInfo
- Publication number
- CN114259225B CN114259225B CN202111532892.6A CN202111532892A CN114259225B CN 114259225 B CN114259225 B CN 114259225B CN 202111532892 A CN202111532892 A CN 202111532892A CN 114259225 B CN114259225 B CN 114259225B
- Authority
- CN
- China
- Prior art keywords
- template
- heart beat
- heart
- peak
- radar
- 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
- 238000000034 method Methods 0.000 title claims abstract description 24
- 230000000694 effects Effects 0.000 claims abstract description 18
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 22
- 238000005070 sampling Methods 0.000 claims description 16
- 238000005311 autocorrelation function Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 7
- 238000012706 support-vector machine Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 210000001765 aortic valve Anatomy 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000012935 Averaging Methods 0.000 claims description 2
- 230000002708 enhancing effect Effects 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 3
- 239000003814 drug Substances 0.000 description 3
- 230000008451 emotion Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000000747 cardiac effect Effects 0.000 description 2
- 210000004115 mitral valve Anatomy 0.000 description 2
- 238000010146 3D printing Methods 0.000 description 1
- 208000035473 Communicable disease Diseases 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000003205 diastolic effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000001727 in vivo Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000005693 optoelectronics Effects 0.000 description 1
- 230000037380 skin damage Effects 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 210000003371 toe Anatomy 0.000 description 1
Abstract
The invention provides an identity authentication and identification method and system based on millimeter wave radar, comprising the following steps: acquiring radar echo data of a plurality of different users by utilizing a millimeter wave radar; decomposing radar heart mechanical activity waveform data RCG from the acquired radar echo data in a preprocessing mode; performing heart rate estimation and heart beat positioning on the radar heart mechanical activity waveform data RCG; performing heart beat cutting and heart beat alignment according to the positioned heart beat position, and finally generating a heart beat template of the current user; normalizing the current user heart beat template; and matching and identifying the normalized heart beat template of the current user with the template in the template library. The invention can realize non-contact identity recognition with high safety coefficient by utilizing the uniqueness and the difficult imitation characteristic of the heart beat template of the user.
Description
[ field of technology ]
The invention relates to the technical field of mode identification, in particular to an identity identification method and system based on millimeter wave radar.
[ background Art ]
The identification has huge application prospect in the information security fields such as security monitoring, financial security and the like, however, the traditional identification mode has the problems of easy loss and easy counterfeiting, such as passwords, tokens, fingerprints, faces, voices and the like. The biological recognition technology makes use of the characteristic that the inherent physiological signals of the human body are difficult to replicate, and becomes a novel identity recognition mode which is widely focused. The chest vibration signal (heart mechanical activity signal) caused by heartbeat is a medical physiological signal, the heartbeat mode depends on biological characteristics and geometric structures of the heart, strong individual variability exists, the necessary characteristics of identity recognition are met, and secondly, the monitoring of the heart mechanical activity signal has the property of living body detection, so that the imitation problems of finger model, photo, 3D printing technology and the like can be effectively avoided.
Millimeter wave radar can monitor mechanical activity of the heart by monitoring chest micro-motion caused by heart motion. A complete cycle of heart beat motion is observed by the waveform of the mechanical heart motion consisting of seven phases, isovolumetric Contraction (IC), mitral valve closure (MC), aortic valve opening (AO), isovolumetric relaxation (IM), aortic valve closure (AC), mitral valve opening (MO) and diastolic Rapid Filling (RF). The chest vibration mode in one heart cycle has obvious differences in biological characteristics and geometric structures of the heart, can be used as a unique identity characteristic of a monitored object, and is difficult to steal and imitate. In addition, compared with the existing identity authentication modes such as fingerprint, electrocardio and heart sound, the identity recognition process based on monitoring the heart mechanical activity by the millimeter wave radar has the characteristics of no contact, long distance and small influence of environmental noise. Chinese patent CN108647961a discloses a digital money wallet based on electrocardiographic signal identity authentication. Comprises an electrocardiograph lock hardware and an electrocardiograph lock wallet APP. When a user opens the electrocardiograph lock wallet APP, the electrocardiograph lock hardware is automatically awakened, the electrocardiograph lock hardware collects electrocardiograph signals of the user, the electrocardiograph signals are extracted in characteristics, and electrocardiograph identification is carried out based on the extracted characteristics. The electrocardio-lock hardware is an electrocardio-lock bracelet, and an optoelectronic continuous electrocardio-sensor is adopted for acquiring the original electrocardio-signal. The electrocardiographic signals are characterized by a QRS wave duration, an R-R interval, and an average R-R interval of the cardiac beat. Chinese patent CN101421744B discloses an electro-biological identity recognition method and apparatus. Comprising two biological signatures, the first biological signature being the difference between the manifestation of a particular individual's heartbeat pattern and a stored common characteristic of a plurality of individuals' heartbeat patterns; the second biosignal is the difference between the representation of the selected individual's heartbeat pattern and the stored common characteristics of the plurality of individual's heartbeat patterns, wherein the heartbeat pattern is an electrocardiographic signal (ECG signal) waveform. The ECG signal acquisition mode is that electrodes are placed at acquisition positions fixed by a user to acquire signals, wherein the positions comprise arms and legs (including fingers and toes), the fingers of the left hand touch a first sensor, the fingers of the right hand touch another sensor in the use process of the user, and the touched contact is usually conductive media such as metal. However, CN101421744B needs to collect electrophysiological signals at a specific location by using an electrode patch, CN108647961a needs to collect electrocardiographic signals by using an electrocardiograph lock bracelet, and when a user uses an electrocardiograph lock wallet APP, the user needs to wear signal collection equipment all the time, and this contact type equipment is not suitable for users who have skin damage, infectious diseases or are not suitable to wear signal collection equipment. Secondly, the collected electrocardiographic signals described in the prior art CN108647961a are characterized by QRS wave duration, R-R interval and average R-R interval, and the characteristics about duration are affected by the current exercise state, heart rate, tension emotion and taking medicine of the user, and there may be large fluctuation, which finally results in that the user is refused to use the electrocardiograph wallet APP due to exercise, tension emotion, taking medicine and the like.
It can be seen that in the above solutions of the prior art, there are problems that the biometric features used for identity recognition are easily stolen, forged, and difficult to obtain without contact.
[ invention ]
In order to overcome the defects of the prior art, the invention provides an identity recognition method and system based on millimeter wave radar by utilizing the uniqueness of a heart mechanical motion mode and the characteristic of difficult counterfeiting.
In one aspect, the invention provides an identity recognition method based on millimeter wave radar, which comprises the following steps:
step 1: acquiring radar echo data of a plurality of different users by utilizing a millimeter wave radar;
step 2: decomposing radar heart mechanical activity waveform data RCG from the acquired radar echo data in a preprocessing mode;
step 3: performing heart rate estimation and heart beat positioning on the radar heart mechanical activity waveform data RCG;
step 4: performing heart beat cutting and heart beat alignment according to the positioned heart beat position, and finally generating a heart beat template of the current user;
step 5: normalizing the current user heart beat template;
step 6: and matching and identifying the normalized heart beat template of the current user with the template in the template library.
Further, on the basis of the above technical solution, the step 1 further includes:
the number of the collected users is more than 30, and the collected users cover people with different sexes, different age groups and different heart rates;
the shortest duration of data acquisition is 20 seconds, and the acquisition mode is that a user stands just opposite to the signal acquisition device, and the distance is between 20cm and 50 cm.
Further, on the basis of the above technical solution, the preprocessing in the step 2 includes the following processing modes:
beamforming, phase information extraction, phase unwrapping, and bandpass filter filtering.
Further, based on the above technical solution, the step 3 further includes:
1) Drawing an autocorrelation ACF curve of the RCG signal obtained in the step 2 by adopting an autocorrelation function ACF mode, and estimating a heart rate HR, wherein an abscissa corresponding to the highest peak in the ACF curve is defined as the heart rate HR of the RCG signal;
2) The upper and lower cut-off frequencies of Butterworth band-pass filters in the Pan-Tompkins algorithm are respectively set to 8Hz and 20Hz, the aopeak of the aortic open in the RCG signal is identified, an AO peak sequence is generated, the element in the AO peak sequence is marked as AOI [ i ], wherein i is the sequence number in the AO peak sequence, 1.ltoreq.i.ltoreq.M, M is the maximum value of the sequence number, the interval between adjacent AO peaks is calculated, the interval number is the larger one of the sequence numbers of the adjacent AO peaks, and the larger one is marked as diff_AOI [ j ], wherein 2.ltoreq.j.ltoreq.M, for example: diff_AOI [ k ] is the interval between AOI [ k ] and AOI [ k-1 ];
if the average mean_diff_AOI of adjacent AO peak intervals is in the range of [60 Xfs/HR-10, 60 Xfs/HR+10 ], judging that the AO peak sequence calculated by the current RCG signal is valid, and if the current AO peak sequence is invalid, prompting a user to re-acquire data until the AO peak sequence meets the valid requirement, wherein each element of the valid AO peak sequence is marked as AO_loc [ i ], and fs is the sampling frequency.
Further, based on the technical scheme, fs is 100Hz.
Further, on the basis of the above technical solution, the step 4 further includes:
1) The RCG signal in step 2 is high-pass filtered, and a bat Wo Sigao pass filter is adopted, wherein parameters are selected as follows: the normalized cut-off frequency Wn is 0.08, is used for strengthening AO peak, look for the maximum value in the interval range of [ AO_loc [ i ] -10, AO_loc [ i+10 ], carry on AO peak position fine tuning, use said AO peak as the datum point all the time while cutting heart beat subsequently, carry on the alignment of heart beat;
2) Calculating duration len_beat of the heart beat according to the heart rate HR obtained in the step 3, wherein a calculation formula is as follows:
len_beat＝60 X fs/HR，
taking the AO peak sequence obtained in the step 3 as a reference point, taking m sampling points forwards from a point AO_loc [ i ], taking n sampling points backwards, and performing heartbeat cutting, wherein m is an integer value obtained by rounding off a calculation result of 0.35 Xlen_bat, and n is an integer value obtained by rounding off a calculation result of 0.65 Xlen_bat;
3) And averaging the cut and aligned heart beats to generate a final heart beat template template_A.
Further, on the basis of the above technical solution, the step 5 further includes:
1) The user faces the millimeter wave radar at different angles and distances, and the template template_A of the heart beat obtained in the step 4 is normalized by 0-1 of amplitude;
2) If the length of the heart beat template template_A obtained in the step 4 is greater than 120 sampling points, adopting a downsampling mode, and if the length of the template_A is less than 120 sampling points, adopting an upsampling mode to process the current heart beat template, so that the heart beat duration time is fixed to 120 sampling points;
3) And finally obtaining a normalized heart beat Template of the current user, and marking the normalized heart beat Template as template_norm.
Further, on the basis of the above technical solution, the step 6 further includes:
1) Calculating Euclidean distance between the current user normalization Template template_Norm and the Template in the user database in the step 5, summing to obtain a final result, and marking the final result as D [ n ];
2) Performing cross-correlation calculation on the current user normalization Template template_norm in the step 5 and a Template in a user database respectively, and summing to obtain a calculation result, and marking the calculation result as Corr [ n ];
3) Obtaining the difference between the amplitude of AO peak of the current user normalized Template template_Norm and the amplitude of AO peak in the user database in step 5, taking absolute value, summing to obtain calculation result, and recording as AMP [ n ];
4) Constructing a Support Vector Machine (SVM) classifier, selecting a radial basis function as a kernel function, determining parameters of the radial basis function by using grid search, taking D [ n ], corr [ n ] and AMP [ n ] as input features of the classifier, and dividing a heart beat template into a positive class and a negative class, wherein the positive class represents authorized users, and the negative class represents unauthorized users;
where n is the number of templates in the user database.
On the other hand, the invention also provides an identity recognition system based on millimeter wave radar, which is characterized by comprising the following steps: a processor, a memory and a controller, said controller controlling millimeter radar wave acquisition data, said memory storing a medium of program code, said device being capable of performing the method of any one of claims 1-8 when said processor system reads said medium stored program code.
Based on the inventive concept of the present invention, the present invention can obtain the following beneficial technical effects:
1. non-contact characteristics: the millimeter wave radar can monitor the mechanical activity of the heart by monitoring the chest micro-motion caused by the heart motion in a non-contact way, a user only needs to stand just opposite to the millimeter wave radar, the RCG signal of the user can be acquired under the non-contact condition, and the heart beat template of the user is analyzed for identity authentication.
2. Characteristics of difficult counterfeiting: the chest vibration signal (heart mechanical activity signal) caused by heartbeat is a medical physiological signal, the heartbeat mode depends on biological characteristics and geometric structures of the heart, strong individual variability exists, the necessary characteristics of identity recognition are met, and secondly, the monitoring of the heart mechanical activity signal has the property of living body detection, and an in-vivo secret key is input, so that the heart mechanical activity signal is difficult to imitate.
3. Adaptive characteristics: and normalizing the calculated heart beat template in amplitude and duration, and eliminating the influence of the user facing the millimeter wave radar at different angles and distances and the conditions of the user such as the current exercise state, heart rate, tension emotion, medicine taking and the like on the heart beat template.
[ description of the drawings ]
FIG. 1 is a flow chart of a preferred embodiment of the present invention.
[ detailed description ] of the invention
For easy understanding, the present embodiment is a preferred embodiment of the millimeter wave radar-based identification method and system according to the present invention, so as to describe the structure and the invention in detail, but the invention is not limited by the scope of the claims.
Referring to fig. 1, a flow chart of a preferred embodiment includes:
step S01: radar echo data is collected. Specifically, a millimeter wave radar is utilized to collect radar echo data of a plurality of different users. The number of the collected users exceeds 30 people, and people with different sexes, different age groups and different heart rates are required to be covered. The shortest duration of data acquisition is 20s, the acquisition mode is that a user stands just opposite to the signal acquisition device, the distance is between 20cm and 50cm, and the signal acquisition device is a millimeter wave radar. And taking the data as experimental data of millimeter wave radar identity recognition.
Step S02: and (5) pretreatment. Specifically, the radar echo data acquired in the step S01 is decomposed into radar heart mechanical activity waveform data (RCG) by a preprocessing mode. The preprocessing part comprises beam forming, phase information extraction, phase unwrapping and band-pass filter filtering.
The beam forming comprises weighting echo signals of all radar antennas according to arrangement conditions of the radar antennas; phase information extraction, including extracting a signal whose phase changes with time at a target position; phase unwrapping, including such that phase w does not jump at pi; the band-pass filter filters, including using a Butterworth band-pass filter to filter the phase signal to decompose the mechanical vibration waveform of the heart, with upper and lower cut-off frequencies of the band-pass filter set to 0.8Hz and 45Hz, respectively.
Step S03: and (5) positioning the heart beat. Heart rate estimation and beat localization are performed on the cardiac mechanical activity waveform data (RCG) of step S02. The method comprises the following steps:
1) And (3) drawing an autocorrelation ACF curve of the RCG signal obtained in the step S02 by adopting an autocorrelation function ACF mode, and estimating the heart rate HR. The abscissa corresponding to the highest peak in the ACF curve is defined as the heart rate HR of the RCG signal.
2) The Pan-Tompkins algorithm is adopted to detect the AO peak position, the Pan-Tompkins aiming at the R wave detection of the Electrocardiosignal (ECG) is optimized and improved, the upper limit cut-off frequency and the lower limit cut-off frequency of the Butterworth band-pass filter in the Pan-Tompkins algorithm are changed from the traditional 5Hz and 11Hz to 8Hz and 20Hz, so that the method is more suitable for RCG signals acquired by the millimeter wave radar aiming at the embodiment, and then the AO peaks in the RCG signals are identified. The spacing of adjacent AO peaks is calculated and noted diff_aoi. If the mean_diff_aoi of the adjacent AO peak intervals is in the range of [60 x fs/HR-10,60 x fs/hr+10], where fs is the sampling frequency, in this embodiment, fs may be preferably 100Hz, it is determined that the AO peak sequence calculated by the current RCG signal is valid, and denoted as ao_loc.
Step S04: and (5) extracting a heart beat template. And (3) specifically, performing heart beat cutting and heart beat alignment according to the heart beat position positioned in the step (S03), and finally generating a heart beat template of the current user. The method comprises the following steps:
1) The RCG signal in the step S02 is subjected to high-pass filtering, the normalized cut-off frequency is 0.08, the normalized cut-off frequency is used for enhancing AO peaks, local maximum values are found in the interval range of [ AO_loc [ i ] -10, AO_loc [ i+10 ], the position of the AO peak is finely adjusted, and the reference point of the AO peak is always used for the subsequent heart beat cutting, so that the heart beat alignment is facilitated.
2) The duration of the heart beat len_beat is calculated from the heart rate HR obtained in step S03, and the heart beat is cut by taking the AO peak sequence ao_loc obtained in step S03 as a reference point, taking forward 0.35×len_beat (the calculation result is rounded) from the point ao_loc [ i ], taking backward 0.65×len_beat (the calculation result is rounded) from the point ao_loc [ i ], and taking backward 0.65×len_beat (the calculation result is rounded) from the point ao_loc [ i ].
3) The cut and aligned beats are averaged to generate a final beat template template_A to reduce occasional non-uniform periods and noise effects for motion.
Step S05: normalizing the heart beat template. And (3) carrying out normalization processing on the current user heart beat template obtained in the step S04. The method comprises the following steps:
1) The user faces the millimeter wave radar at different angles and distances, the amplitude of the acquired signal has small change, and in order to enable the heart beat templates acquired under different conditions to have comparable amplitudes, the heart beat template template_A acquired in the step S04 is normalized by [0,1] of the amplitude.
2) In order to exclude the influence of the intrinsic heart rate variability and different motion states contained in the heartbeat mode on the heart rate, namely the influence on the heart beat duration, the fixed heart beat duration is 120 sampling points, if the length of the heart beat template template_A obtained in the step 4 is greater than 120 sampling points, a downsampling mode is adopted, if the length of the template_A is less than 120 sampling points, the current heart beat template is processed in an upsampling mode, and the duration of the heart beat template is unified on the premise of not changing the waveform form.
3) And finally obtaining a normalized heart beat Template of the current user, and marking the normalized heart beat Template as template_norm.
Step S06: judging whether the heart beat belongs to an authorized user. And (3) matching and identifying the current user normalized heart beat template obtained in the step S05 with a template in a template library. The method comprises the following steps:
1) And (3) calculating Euclidean distance between the current user normalization Template template_Norm obtained in the step S05 and the Template in the user database respectively, and summing to obtain a final result, which is recorded as D [ n ].
2) And (3) carrying out cross-correlation calculation on the current user normalization Template template_norm obtained in the step S05 and templates in a user database respectively, and summing to obtain a calculation result, and recording the calculation result as Corr [ n ].
3) And (3) obtaining the difference between the amplitude of the AO peak of the current user normalized Template template_Norm obtained in the step S05 and the amplitude of the AO peak in the Template in the user database, taking the absolute value, and summing to obtain a calculation result, and recording the calculation result as AMP [ n ].
4) And constructing an SVM classifier, selecting a radial basis function as a kernel function, and determining parameters of the radial basis function by using grid search. D [ n ], corr [ n ] and AMP [ n ] are used as input features of the classifier, the heart beat templates are divided into positive classes (authorized users) and negative classes (unauthorized users), the classifier can calculate the possibility that the heart beat templates of the current user belong to the positive classes, and if the heart beat features of the current user have low adaptability to the authorized users, the heart beat templates are judged to be the negative classes.
The present invention is not limited to the above-mentioned embodiments, but all simple changes to the technical features of the present invention can be made, and equivalent changes or modifications of the construction, features and principles described in the claims of the present invention will fall within the scope of the present invention.
Claims (6)
1. An identity recognition method based on millimeter wave radar is characterized by comprising the following steps:
step 1: acquiring radar echo data of a plurality of different users by utilizing a millimeter wave radar;
step 2: decomposing radar heart mechanical activity waveform data RCG from the acquired radar echo data in a preprocessing mode;
step 3: performing heart rate estimation and heart beat positioning on the radar heart mechanical activity waveform data RCG;
step 4: performing heart beat cutting and heart beat alignment according to the positioned heart beat position, and finally generating a heart beat template of the current user;
the step 4 specifically includes:
1) The RCG signal in step 2 is high-pass filtered, and a bat Wo Sigao pass filter is adopted, wherein parameters are selected as follows: the normalized cut-off frequency Wn is 0.08, which is used for enhancing AO peak, searching the maximum value in the interval range of [ AO_loc [ i ] -10, AO_loc [ i+10 ], finely adjusting the position of AO peak, and aligning heart beat by taking the AO peak as the datum point all the time when cutting heart beat later;
2) Calculating duration len_beat of the heart beat according to the heart rate HR obtained in the step 3, wherein a calculation formula is as follows:
len_beat＝60X fs/HR，
taking the AO peak sequence obtained in the step 3 as a reference point, taking m sampling points forwards from a point AO_loc [ i ], taking n sampling points backwards, and performing heartbeat cutting, wherein m is an integer value obtained by rounding off a calculation result of 0.35 Xlen_bat, and n is an integer value obtained by rounding off a calculation result of 0.65 Xlen_bat;
3) Averaging the cut and aligned heart beats to generate a final heart beat template template_A;
step 5: normalizing the current user heart beat template;
the step 5 specifically includes:
1) The user faces the millimeter wave radar at different angles and distances, and the template template_A of the heart beat obtained in the step 4 is normalized by 0-1 of amplitude;
2) If the length of the heart beat template template_A obtained in the step 4 is greater than 120 sampling points, adopting a downsampling mode, and if the length of the template_A is less than 120 sampling points, adopting an upsampling mode to process the current heart beat template, so that the heart beat duration time is fixed to 120 sampling points;
3) Finally, a normalized heart beat Template of the current user is obtained and is recorded as a template_norm;
step 6: matching and identifying the normalized heart beat template of the current user with templates in a template library;
the step 6 specifically includes:
1) Calculating Euclidean distance between the current user normalization Template template_Norm and the Template in the user database in the step 5, summing to obtain a final result, and marking the final result as D [ n ];
2) Performing cross-correlation calculation on the current user normalization Template template_norm in the step 5 and a Template in a user database respectively, and summing to obtain a calculation result, and marking the calculation result as Corr [ n ];
3) Obtaining the difference between the amplitude of AO peak of the current user normalized Template template_Norm and the amplitude of AO peak in the user database in step 5, taking absolute value, summing to obtain calculation result, and recording as AMP [ n ];
4) Constructing a Support Vector Machine (SVM) classifier, selecting a radial basis function as a kernel function, determining parameters of the radial basis function by using grid search, taking D [ n ], corr [ n ] and AMP [ n ] as input features of the classifier, and dividing a heart beat template into a positive class and a negative class, wherein the positive class represents authorized users, and the negative class represents unauthorized users;
where n is the number of templates in the user database.
2. The identification method according to claim 1, wherein the step 1 further comprises:
the number of the collected users is more than 30, and the collected users cover people with different sexes, different age groups and different heart rates;
the shortest duration of data acquisition is 20 seconds, and the acquisition mode is that a user stands just opposite to the signal acquisition device, and the distance is between 20cm and 50 cm.
3. The identification method according to claim 1, wherein the preprocessing in step 2 includes the following processing modes:
beamforming, phase information extraction, phase unwrapping, and bandpass filter filtering.
4. A method of identity recognition according to any one of claims 1 to 3, wherein step 3 further comprises:
1) Drawing an autocorrelation ACF curve of the RCG signal obtained in the step 2 by adopting an autocorrelation function ACF mode, and estimating a heart rate HR, wherein an abscissa corresponding to the highest peak in the ACF curve is defined as the heart rate HR of the RCG signal;
2) The upper and lower limit cut-off frequencies of Butterworth band-pass filters in a Pan-Tompkins algorithm are respectively set to 8Hz and 20Hz, aortic valve opening AO peaks in RCG signals are identified, AO peak sequences are generated, each element in the AO peak sequences is marked as AOI [ i ], wherein i is the sequence number in the AO peak sequences, i is more than or equal to 1 and less than or equal to M, M is the maximum value of the sequence numbers, the interval between adjacent AO peaks is calculated, the interval number is the larger one of the sequence numbers of the adjacent AO peaks and is marked as diff_AOI [ j ], and j is more than or equal to 2 and less than or equal to M;
if the mean_diff_AOI of adjacent AO peak intervals is in the range of [60 Xfs/HR-10, 60 Xfs/HR+10 ], the AO peak sequence calculated by the current RCG signal is judged to be valid, and each element is marked as AO_loc [ i ], wherein fs is the sampling frequency.
5. The identification method as claimed in claim 4, wherein fs is 100Hz.
6. An identity recognition system based on millimeter wave radar, which is characterized by comprising: a processor, a memory and a controller, said controller controlling millimeter radar wave acquisition data, said memory storing a medium of program code, said system being capable of performing the method of any one of claims 1-5 when said processor system reads said medium stored program code.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111532892.6A CN114259225B (en) | 2021-12-15 | 2021-12-15 | Identity recognition method and system based on millimeter wave radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111532892.6A CN114259225B (en) | 2021-12-15 | 2021-12-15 | Identity recognition method and system based on millimeter wave radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114259225A CN114259225A (en) | 2022-04-01 |
| CN114259225B true CN114259225B (en) | 2024-03-15 |
Family
ID=80827225
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111532892.6A Active CN114259225B (en) | 2021-12-15 | 2021-12-15 | Identity recognition method and system based on millimeter wave radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114259225B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115177267B (en) * | 2022-09-13 | 2023-01-03 | 合肥心之声健康科技有限公司 | Heart beat artifact identification method and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008102291A2 (en) * | 2007-02-20 | 2008-08-28 | Philips Intellectual Property & Standards Gmbh | Method and system for identifying a subject |
| CN107257653A (en) * | 2015-02-26 | 2017-10-17 | 皇家飞利浦有限公司 | Processing equipment, system and method for handling the accelerometer signal being used in being monitored to the vital sign of object |
| CN109310371A (en) * | 2016-06-16 | 2019-02-05 | 阿克瑞克公司 | Quantitative seismocardiography |
| CN110491500A (en) * | 2019-08-07 | 2019-11-22 | 王满 | A kind of identification system and method based on cardiac function dynamic monitoring and analysis |
| CN111770722A (en) * | 2018-02-27 | 2020-10-13 | 罗伯特·博世有限公司 | Wearable health device system with automatic reference of cardiac vibrographic signals |
| CN112971771A (en) * | 2021-02-23 | 2021-06-18 | 浙江大学计算机创新技术研究院 | Method for reconstructing electrocardiogram based on millimeter waves |
| CN113035345A (en) * | 2021-03-25 | 2021-06-25 | 苏州沪应通智能科技有限公司 | Mattress with vital sign monitoring function and monitoring system thereof |
| CN113100725A (en) * | 2021-04-28 | 2021-07-13 | 中物慧芯信息科技(苏州)有限公司 | Health state rapid detection device and method based on optical fiber sensor |
| CN113468989A (en) * | 2021-06-18 | 2021-10-01 | 南京润楠医疗电子研究院有限公司 | Non-contact personnel identification method using heart radar signals |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006132865A2 (en) * | 2005-06-03 | 2006-12-14 | Acceleron Medical Systems, Inc. | Improved scg point pick process device and method |
| US8463361B2 (en) * | 2007-05-24 | 2013-06-11 | Lifewave, Inc. | System and method for non-invasive instantaneous and continuous measurement of cardiac chamber volume |
| EP3840639A4 (en) * | 2018-08-20 | 2022-04-06 | MacDonald, Dettwiler and Associates Inc. | Method and apparatus for deriving biometric information using multiple-axis seismocardiography |
-
2021
- 2021-12-15 CN CN202111532892.6A patent/CN114259225B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008102291A2 (en) * | 2007-02-20 | 2008-08-28 | Philips Intellectual Property & Standards Gmbh | Method and system for identifying a subject |
| CN107257653A (en) * | 2015-02-26 | 2017-10-17 | 皇家飞利浦有限公司 | Processing equipment, system and method for handling the accelerometer signal being used in being monitored to the vital sign of object |
| CN109310371A (en) * | 2016-06-16 | 2019-02-05 | 阿克瑞克公司 | Quantitative seismocardiography |
| CN111770722A (en) * | 2018-02-27 | 2020-10-13 | 罗伯特·博世有限公司 | Wearable health device system with automatic reference of cardiac vibrographic signals |
| CN110491500A (en) * | 2019-08-07 | 2019-11-22 | 王满 | A kind of identification system and method based on cardiac function dynamic monitoring and analysis |
| CN112971771A (en) * | 2021-02-23 | 2021-06-18 | 浙江大学计算机创新技术研究院 | Method for reconstructing electrocardiogram based on millimeter waves |
| CN113035345A (en) * | 2021-03-25 | 2021-06-25 | 苏州沪应通智能科技有限公司 | Mattress with vital sign monitoring function and monitoring system thereof |
| CN113100725A (en) * | 2021-04-28 | 2021-07-13 | 中物慧芯信息科技(苏州)有限公司 | Health state rapid detection device and method based on optical fiber sensor |
| CN113468989A (en) * | 2021-06-18 | 2021-10-01 | 南京润楠医疗电子研究院有限公司 | Non-contact personnel identification method using heart radar signals |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114259225A (en) | 2022-04-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pinto et al. | Evolution, current challenges, and future possibilities in ECG biometrics | |
| US8232866B2 (en) | Systems and methods for remote long standoff biometric identification using microwave cardiac signals | |
| Agrafioti et al. | Heart biometrics: Theory, methods and applications | |
| Singh et al. | Bioelectrical signals as emerging biometrics: Issues and challenges | |
| Abo-Zahhad et al. | Biometric authentication based on PCG and ECG signals: present status and future directions | |
| Labati et al. | ECG biometric recognition: Permanence analysis of QRS signals for 24 hours continuous authentication | |
| Chamatidis et al. | Using deep learning neural networks for ECG based authentication | |
| Abdeldayem et al. | ECG-based human authentication using high-level spectro-temporal signal features | |
| Fang et al. | QRS detection-free electrocardiogram biometrics in the reconstructed phase space | |
| Silva et al. | ECG biometrics: Principles and applications | |
| Pal et al. | ECG biometric recognition | |
| JP2009525070A (en) | Identity verification system using physiological biometrics to determine physiological fingerprint and identification method | |
| Teodoro et al. | Feature selection for biometric recognition based on electrocardiogram signals | |
| Wu et al. | ECG biometric recognition: unlinkability, irreversibility, and security | |
| CN110598625A (en) | Identity recognition technology based on pulse wave non-reference characteristics | |
| Chin et al. | Bayesian real-time QRS complex detector for healthcare system | |
| Islam et al. | Selection of heart-biometric templates for fusion | |
| CN114259225B (en) | Identity recognition method and system based on millimeter wave radar | |
| Bastos et al. | Double authentication model based on ppg and ecg signals | |
| CN112804937A (en) | High frequency QRS in biometric identification | |
| Vuksanovic et al. | AR-based method for ECG classification and patient recognition | |
| Kuila et al. | Feature extraction of electrocardiogram signal using machine learning classification | |
| Alotaiby et al. | A nonfiducial PPG-based subject Authentication Approach using the statistical features of DWT-based filtered signals | |
| Karegar et al. | ECG based human authentication with using Generalized Hurst Exponent | |
| Canento et al. | Review and comparison of real time electrocardiogram segmentation algorithms for biometric applications |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |