CN106908095B - A kind of extraction and appraisal procedure of sensing data alignment features - Google Patents
A kind of extraction and appraisal procedure of sensing data alignment features Download PDFInfo
- Publication number
- CN106908095B CN106908095B CN201710013933.8A CN201710013933A CN106908095B CN 106908095 B CN106908095 B CN 106908095B CN 201710013933 A CN201710013933 A CN 201710013933A CN 106908095 B CN106908095 B CN 106908095B
- Authority
- CN
- China
- Prior art keywords
- data
- speed
- appraisal procedure
- sensor
- sensors
- 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
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
- G01D18/00—Testing or calibrating apparatus or arrangements provided for in groups G01D1/00 - G01D15/00
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The extraction and appraisal procedure of sensing data alignment features, step are as follows: 1., for static scene, build multisensor cross-beta platform, and multiple sensors need to be calibrated the corresponding high-precision measuring instrument of sensor.For mobile context, the sensing node of integrated design multiple sensors and communication module is deployed in the move vehicle of the environment, and real-time data collection simultaneously passes cloud back.2. the velocity estimation under mobile context merges the measured value of different sensors module by Kalman filtering, to obtain optimal velocity estimation value.3. a pair data pre-process, need to be calibrated sensor for every kind, consistent data sample on space-time is generated, with correlation technique to feature importance appraisal procedure using the data of corresponding high precision instrument as label using the data of other sensors as feature.Linear appraisal procedure and nonlinear appraisal procedure.
Description
Technical field
The present invention relates to the extraction and appraisal procedure of a kind of sensing data alignment features, especially static and dynamic environment
Deployment, seek using data fusion the importance appraisal procedure of speed and linear character and nonlinear characteristic.
Background technique
City perception based on extensive sensing network provides the data letter of fine granularity and various dimensions for smart city
Breath, and it is increasingly becoming the main source of city big data.It is of new generation with the rise of sensor technology, it is inexpensive, it is portable
Formula sensor has gradually come into people's lives.These sensors are based on the sensor of electrochemistry, gradually by industry and
Art circle has been used in the city perception of large-scale wireless network.However due to by time migration, environmental factor, mobile context
Etc. influence, these sensors have low precision, measured value offset and sensor reading and the weak dependence of true value etc.
Problem.These characteristics show even if the calibration for having factory, in order to ensure regularly recalibration is necessary for the accuracys of data.
In the tasks such as urban air-quality monitoring, reliable data can help people preferably decision and planning, therefore sensor
Data are calibrated to for an extremely important problem.We need to calibrate the data of sensor, the work of calibration
Be exactly in order to minimize deviation.
“Pre-Deployment Testing,Augmentation and Calibration of Cross-
Point out that there is correlation between the data of different sensors in Sensitive Sensors ", the test for having built multisensor is flat
Platform verifies the correlation between different sensors data, including temperature and humidity, ozone, nitrogen dioxide, the sensors such as PM2.5.It is real
It tests the result shows that the data and temperature and humidity of most of gas sensor have correlation, the sensor and ozone of nitrogen dioxide sense
The reading of device has apparent correlation.Based on data dependence, author is proposed based on polynary least square method (MLS)
Calibration method is calibrated to merge the data of multisensor.The cross-beta platform wherein built is not directed to mobile context
Under, also alignment features are not extracted and assessed.
For " Pre-Deployment Testing, Augmentation and Calibration of Cross-
Disadvantage in Sensitive Sensors ", this paper presents for static environment and mobile environment, deployment and is commented feature extraction
Estimate method.
Summary of the invention
The present invention will overcome the disadvantages mentioned above of the prior art, provide a kind of extraction of sensing data alignment features and comment
Estimate method.
In order to achieve the above object, the technical solution used in the present invention is: a kind of extraction of sensing data alignment features
And appraisal procedure, comprising the following steps:
Step 1, cross-beta platform is built, comprising:
(1.1) select equipment, the corresponding high precision measuring instrument of sensor array and each for needing to be calibrated with
And other relevant sensor arrays.
(1.2) deployed with devices, main includes static deployment and Dynamical Deployment.
(1.2.1) static deployment, in static scene lower sensor array, together with high-precision reference instrument and other
Sensor deployment is in a static environment.
(1.2.1) Dynamical Deployment, for being integrated in mobile context lower sensor array and together with other sensors,
It is deployed on mobile vehicle together with corresponding high-precision reference instrument.
(1.3) data acquire, and pass the data that sensor and reference instrument acquire back cloud by corresponding communication module.
Step 2, the velocity characteristic estimation under mobile context, comprising:
(2.1) motion sensor, including accelerometer module and GPS module are integrated.
(2.2) equation of accelerometer and GPS module on estimating speed is established respectively.
(2.2.1) seeks speed by acceleration, and speed is obtained by the integral of acceleration in time.
(2.2.2) seeks speed by GPS, and speed can be displaced derivation in time by GPS and obtain
(2.3) speed is asked by data fusion, the estimation of two speed is melted using data fusion method
It closes to obtain optimal estimated value.
Step 3, feature extraction and assessment, comprising:
(3.1) data prediction, main includes the standardization and normalization of data, denoising.
(3.2) space-time consistency data sample is generated, for each sensor for needing to be calibrated, by other sensors
For data as feature, sensor corresponds to the test of high precision instrument as label, generates consistent data sample on space-time.
(3.3) it for every group of data sample, is assessed with importance of the relevant method to feature.Including linear special
Levy appraisal procedure and nonlinear characteristic appraisal procedure.
The assessment of (3.2.1) linear character, the importance of linear character, which is assessed, acquires Monomial coefficient using least square
It is assessed with Pearson's coefficient.
The importance assessment of the assessment of (3.2.2) nonlinear characteristic, nonlinear characteristic is carried out using the method for categorised decision tree
Assessment.
The beneficial effects of the present invention are: this method is directed to static situation and mobile environment, the portion of relevant device is carried out respectively
Administration.To the velocity characteristic under mobile context, accurate estimation has been carried out.Linear character and nonlinear characteristic difference for calibration
Carry out importance assessment.
Detailed description of the invention
Fig. 1 is the work flow diagram of the method for the present invention.
Specific embodiment
The invention will be further described below in conjunction with the accompanying drawings.A specific embodiment of the invention is as follows:
Step 1, cross-beta platform is built, comprising:
(1.1) select equipment, the corresponding high precision measuring instrument of sensor array and each for needing to be calibrated with
And other relevant sensor arrays.Related sensor includes temperature, and humidity, air pressure, illumination, pressure and sound etc. observe phenomenon
Sensor.
(1.2) deployed with devices, main includes static deployment and Dynamical Deployment.Under mobile context, integrated GPS sensor is needed
And acceleration transducer.
(1.2.1) static deployment, in static scene lower sensor array, together with high-precision reference instrument and other
Sensor deployment is in a static environment.
(1.2.1) Dynamical Deployment, for being integrated in mobile context lower sensor array and together with other sensors,
It is deployed on mobile vehicle together with corresponding high-precision reference instrument.
(1.3) data acquire, and pass the data that sensor and reference instrument acquire back cloud by corresponding communication module.
Data transfer mode is transmitted by GPRS module.
Step 2, the velocity characteristic estimation under mobile context, comprising:
(2.1) motion sensor, including accelerometer module and GPS module are integrated.
(2.2) equation of accelerometer and GPS module on estimating speed is established respectively.
(2.2.1) seeks speed by acceleration, and speed is obtained by the integral of acceleration in time.It is as follows:
V [t] is the speed of t moment, and A indicates acceleration, and u [t] is process caused by having acceleration transducer measurement error
Noise.
(2.2.2) seeks speed by GPS, and speed can be displaced derivation in time by GPS and obtain, as follows:
V [t]=X [t]+v [t] formula (2)
X [t] is the estimated value of the speed obtained by GPS, and v [t] is that have GPS measurement error to lead to observation noise.
(2.3) speed is asked by data fusion, the estimation of two speed is carried out using data fusion correlation technique
Fusion is to obtain optimal estimated value.
Step 3, feature extraction and assessment, comprising:
(3.1) data prediction, main includes the standardization and normalization of data, denoising.Signal weight is taken for denoising
Actual signal is considered as x ∈ R by the method builtn(length of n expression signal), it is assumed that disturbance of the signal by additional noise v, xcor
=x+v, wherein be the original signal uploaded by sensor, it is simple herein to assume that noise v is to be considered as unknown, smaller and rapid change
Change with xcorMachine variable.Goal is in known xcorIn the case where, obtain the estimated value of x
Signal reconstruction[42]The problem of following standard being turned to by form
It minimizes
WhereinIt is to need optimised target, xcorIt is the parameter of optimization problem.Function phi: Rn→ R is a convergence letter
Number, it is smooth objective function here, its value can be withThe measurement of degree of roughness.Optimization problem is attempted to look for close to original
Signal (xcor), while again it is smooth (It is smaller)
(3.2) space-time consistency data sample is generated, for each sensor for needing to be calibrated, by other sensors
For data as feature, sensor corresponds to the test of high precision instrument as label, generates consistent data sample on space-time.
(3.3) it for every group of data sample, is assessed with importance of the relevant method to feature.Including linear special
Levy appraisal procedure and nonlinear characteristic appraisal procedure.
The assessment of (3.2.1) linear character, the importance of linear character, which is assessed, acquires Monomial coefficient using least square
It is assessed with Pearson's coefficient.
The importance assessment of the assessment of (3.2.2) nonlinear characteristic, nonlinear characteristic is carried out using the method for categorised decision tree
Assessment.
Content described in this specification embodiment is only enumerating to the way of realization of inventive concept, protection of the invention
Range should not be construed as being limited to the specific forms stated in the embodiments, and protection scope of the present invention is also and in art technology
Personnel conceive according to the present invention it is conceivable that equivalent technologies mean.
Claims (2)
1. a kind of extraction and appraisal procedure of sensing data alignment features, it is characterised in that following steps:
Step 1, cross-beta platform is built, comprising:
(1.1) select equipment, the corresponding high precision measuring instrument of sensor array and each for needing to be calibrated and its
His relevant sensor array;
(1.2) deployed with devices, main includes static deployment and Dynamical Deployment;
(1.2.1) static deployment, in static scene lower sensor array, together with high precision measuring instrument and other sensors
Deployment is in a static environment;
(1.2.1) Dynamical Deployment, for being integrated in mobile context lower sensor array and together with other sensors, together with
Corresponding high precision measuring instrument is deployed on mobile vehicle;
(1.3) data acquire, the data for being acquired sensor and corresponding high precision measuring instrument by corresponding communication module
Pass cloud back;
Step 2, the velocity characteristic estimation under mobile context, comprising:
(2.1) motion sensor, including accelerometer module and GPS module are integrated;
(2.2) equation of accelerometer module and GPS module on estimating speed is established respectively;
(2.2.1) seeks speed by acceleration, and speed is obtained by the integral of acceleration in time;
(2.2.2) seeks speed by GPS, and speed can be displaced derivation in time by GPS and obtain;
(2.3) speed is asked by data fusion, the estimation of two speed is merged using data fusion method to obtain most
Excellent estimated value;
Step 3, feature extraction and assessment, comprising:
(3.1) data prediction, main includes the standardization and normalization of data, denoising;
(3.2) space-time consistency data sample is generated, for each sensor for needing to be calibrated, by the data of other sensors
As feature, sensor corresponds to the test of high precision measuring instrument as label, generates consistent data sample on space-time;
(3.3) it for every group of data sample, is assessed with importance of the relevant method to feature, including linear character is commented
Estimate method and nonlinear characteristic appraisal procedure;
The assessment of (3.2.1) linear character, Monomial coefficient that the importance assessment of linear character is acquired using least square method and
Pearson's coefficient is assessed;
The importance assessment of the assessment of (3.2.2) nonlinear characteristic, nonlinear characteristic is assessed using the method for categorised decision tree.
2. the extraction and appraisal procedure of a kind of sensing data alignment features according to claim 1, it is characterised in that
It in the speed estimation method of the step (2.3), is minimized by the measured value of fusion accelerometer module and GPS module
Speed estimation error.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710013933.8A CN106908095B (en) | 2017-01-09 | 2017-01-09 | A kind of extraction and appraisal procedure of sensing data alignment features |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710013933.8A CN106908095B (en) | 2017-01-09 | 2017-01-09 | A kind of extraction and appraisal procedure of sensing data alignment features |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106908095A CN106908095A (en) | 2017-06-30 |
CN106908095B true CN106908095B (en) | 2019-04-26 |
Family
ID=59207448
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710013933.8A Active CN106908095B (en) | 2017-01-09 | 2017-01-09 | A kind of extraction and appraisal procedure of sensing data alignment features |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106908095B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109429194B (en) * | 2017-08-17 | 2020-12-11 | 浙江大学 | Method and device for determining position of reference node in mobile sensing network |
CN108801322A (en) * | 2018-07-06 | 2018-11-13 | 哈尔滨工业大学 | Reliability estimation method for UAV Flight Control System MEMS sensor |
CN111114463B (en) * | 2018-10-30 | 2023-01-17 | 百度在线网络技术(北京)有限公司 | Method and device for acquiring blind area noise |
CN109631973B (en) * | 2018-11-30 | 2021-04-16 | 苏州数言信息技术有限公司 | Automatic calibration method and system for sensor |
CN109900309B (en) * | 2019-03-08 | 2021-03-16 | 重庆邮电大学 | Sensor data blind correction method based on mixed state space model |
CN113391040B (en) * | 2021-07-12 | 2023-09-15 | 北京清环宜境技术有限公司 | Automatic calibration method for data artificial intelligence of atmospheric micro station |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102252677A (en) * | 2011-04-18 | 2011-11-23 | 哈尔滨工程大学 | Time series analysis-based variable proportion self-adaptive federal filtering method |
CN103630136A (en) * | 2013-12-05 | 2014-03-12 | 中国航空无线电电子研究所 | Optimum navigational parameter fusion method based on three-level filtering under redundant sensor configuration |
CN104880722B (en) * | 2015-03-25 | 2017-08-11 | 清华大学 | The GPS velocity and position detection rejecting outliers method of unmanned plane |
CN104864869B (en) * | 2015-06-05 | 2017-11-21 | 中国电子科技集团公司第二十六研究所 | A kind of initial dynamic attitude determination method of carrier |
CN104977006A (en) * | 2015-08-11 | 2015-10-14 | 北京纳尔信通科技有限公司 | Indoor positioning method based on fuzzy theory and multi-sensor fusion |
-
2017
- 2017-01-09 CN CN201710013933.8A patent/CN106908095B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN106908095A (en) | 2017-06-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106908095B (en) | A kind of extraction and appraisal procedure of sensing data alignment features | |
Zhang et al. | Allan variance analysis on error characters of MEMS inertial sensors for an FPGA-based GPS/INS system | |
US11181402B2 (en) | System and method for the assisted calibration of sensors distributed across different devices | |
CN109188026B (en) | Automatic calibration deep learning method suitable for MEMS accelerometer | |
KR102044041B1 (en) | Apparatus for measuring earthquake intensity and method for the same | |
CN105823907A (en) | Built-in self test circuit and self test method for parallel plate capacitive MEMS (Micro-Electro-Mechanical-System) accelerometer | |
Wei et al. | A research on calibration of low-precision MEMS inertial sensors | |
CN108478222B (en) | Neural network-based joint angle error compensation experimental device and method | |
EP3591354A1 (en) | Axle-load measuring apparatus and axle-load measuring method | |
CN104597273B (en) | A kind of test method and equipment of movement velocity | |
EP2951956A1 (en) | Determining response similarity neighborhoods | |
Havaran et al. | Extracting structural dynamic properties utilizing close photogrammetry method | |
Marinov et al. | Analysis of sensors noise performance using Allan deviation | |
CN109579832A (en) | A kind of personnel's height autonomous positioning algorithm | |
Tran et al. | Development of a rainfall-triggered landslide system using wireless accelerometer network | |
Lo Iacono et al. | Structural monitoring of “Himera” viaduct by low-cost MEMS sensors: characterization and preliminary results | |
US8566057B2 (en) | Method for self-adjustment of a triaxial acceleration sensor during operation, and sensor system having a three-dimensional acceleration sensor | |
CN103499354B (en) | Neyman-Pearson criterion-based zero speed detection method | |
EP2988095B1 (en) | Altitude detecting unit and altitude detecting method | |
Aggarwal et al. | Cost-effective testing and calibration of low cost MEMS sensors for integrated positioning, navigation and mapping systems | |
JP3164632U (en) | Device for determining the flow rate of a bi-directional unsteady fluid flow | |
Muttillo et al. | Structural health continuous monitoring of buildings–A modal parameters identification system | |
CN109884985A (en) | The measurement method of numerically-controlled machine tool complete machine machining state dynamic characteristic | |
KR101434393B1 (en) | A displacement vibration measuring method using camera image | |
RU2682080C1 (en) | Method for measuring temperature, pressure and density section in liquid |
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 |