CN110909873A - Method and device for denoising passive RFID (radio frequency identification) mobile ranging data and computer equipment - Google Patents
Method and device for denoising passive RFID (radio frequency identification) mobile ranging data and computer equipment Download PDFInfo
- Publication number
- CN110909873A CN110909873A CN201910950182.1A CN201910950182A CN110909873A CN 110909873 A CN110909873 A CN 110909873A CN 201910950182 A CN201910950182 A CN 201910950182A CN 110909873 A CN110909873 A CN 110909873A
- Authority
- CN
- China
- Prior art keywords
- label signal
- value
- signal intensity
- signal strength
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S11/00—Systems for determining distance or velocity not using reflection or reradiation
- G01S11/02—Systems for determining distance or velocity not using reflection or reradiation using radio waves
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
Abstract
The application discloses a passive RFID mobile ranging data noise reduction method, a passive RFID mobile ranging data noise reduction device and computer equipment, and can solve the problem that errors exist in measured and calculated distances easily due to environmental interference when a passive electronic tag is used for positioning. The method comprises the following steps: configuring an electronic tag for an object to be detected; acquiring the label signal intensity corresponding to the object to be detected according to a preset route in a preset period; inputting the label signal intensity into an optimized denoising model successfully trained to obtain an optimal estimation value of the label signal intensity; and the data processing center calculates the distance between the object to be measured and the inductor according to the label signal intensity optimal estimation value. The method and the device are suitable for denoising the label signal strength in the passive RFID mobile ranging process.
Description
Technical Field
The application relates to the technical field of computers, in particular to a passive RFID mobile ranging data noise reduction method and device and computer equipment.
Background
In order to use and manage urban components, storage materials and the like more intelligently, finely and efficiently, the technology of the internet of things is widely applied. For the problems that the variety of managed assets such as urban parts, storage materials and the like is various, the updating speed is high, the per capita management area is continuously increased, and the supervision difficulty is continuously increased, more and more managers select and apply an electronic tag (also called a radio frequency identification tag or an RFID tag) positioning technology to realize the appearance of a platform for monitoring the assets, and the current situations that manual inspection is needed, time and labor are consumed, and the efficiency is low are solved.
The RFID electronic tags are mainly divided into active tags and passive tags. Compared with an active tag, the passive tag is low in cost, small in size, manufactured in a card mode and high in applicability, and therefore when the electronic tag is used for positioning, the passive tag is high in selection rate.
However, when the passive electronic tag is used for positioning, the signal strength is greatly influenced by the environment, and the RFID device is susceptible to receiving the influenced signal strength due to the system error caused by the factors such as the antenna and the like, so that an incorrect distance value is obtained, and the judgment of management and maintenance personnel is influenced.
Disclosure of Invention
In view of this, the present application provides a method, an apparatus, and a computer device for reducing noise of passive RFID mobile ranging data, and mainly aims to solve the problem that an error is easily caused in distance measurement due to environmental interference when a passive electronic tag is used for positioning.
According to one aspect of the application, a method for passive RFID mobile ranging data noise reduction is provided, the method comprising:
configuring an electronic tag for an object to be detected;
acquiring the label signal intensity corresponding to the object to be detected according to a preset route in a preset period;
inputting the label signal intensity into an optimized denoising model successfully trained to obtain an optimal estimation value of the label signal intensity;
and the data processing center calculates the distance between the object to be measured and the inductor according to the label signal intensity optimal estimation value.
Optionally, before inputting the label signal strength into an optimized denoising model successfully trained and obtaining an optimal estimation value of the label signal strength, the method further includes:
creating an optimized denoising model, wherein the optimized denoising model comprises a Kalman filtering calculation module and a BP neural network model;
and carrying out model training according to the BP neural network model so as to enable the optimized denoising model to meet a preset standard.
Optionally, the performing model training according to the BP neural network model to make the optimized denoising model meet a preset standard includes:
acquiring first label signal intensity corresponding to the object to be detected, wherein the first label signal intensity corresponds to an average value of label signal intensities corresponding to each first acquisition point in a preset route;
importing the first label signal strength into a Kalman filtering calculation module of the optimized denoising model to obtain first input data comprising three parameters, wherein the three parameters comprise a difference between a label signal strength predicted value and a label signal strength optimal estimated value, Kalman filtering gain and innovation;
calculating a first distance between the object to be measured and the sensor according to the data acquisition position recorded by the GPS receiver in each first acquisition point and the position coordinate of the object to be measured, wherein the first distance corresponds to an average value of the distances between each first acquisition point and the object to be measured in a preset route;
substituting the first label signal strength and the first distance into a path loss model, and calculating to obtain a label signal strength theoretical value;
determining the difference between the theoretical value of the label signal strength and the predicted value of the label signal strength as first output data;
training the BP neural network model by using the first input data and the first output data;
re-acquiring second label signal intensity corresponding to the object to be detected according to the preset route, wherein the second label signal intensity corresponds to the average value of the label signal intensity corresponding to each second acquisition point in the preset route;
importing the second label signal intensity into the optimized denoising model to obtain second output data;
calculating a label signal intensity predicted value corresponding to the second label signal intensity according to a Kalman filtering algorithm;
calculating a label signal strength theoretical value corresponding to the second label signal strength according to a path loss model;
calculating a difference value between the label signal strength theoretical value and the second output data and the label signal strength predicted value, if the difference value is judged to be equal to zero or smaller than a preset threshold value, determining that the optimized denoising model passes training, and determining Kalman filtering gain and innovation in the Kalman filtering calculation module as target parameters;
and if the difference value is judged to be larger than or equal to the preset threshold value, adjusting Kalman filtering gain and innovation in the Kalman filtering calculation module until an output value meets an error requirement so as to further determine the target parameter.
Optionally, the inputting the label signal strength into an optimized denoising model successfully trained to obtain an optimal estimate value of the label signal strength includes:
acquiring target Kalman filtering gain and target information corresponding to the target parameters;
and substituting the label signal intensity, the target Kalman filtering gain and the target information into a Kalman filtering calculation module of the optimized denoising model, and calculating by using an optimal estimation equation of the Kalman filtering algorithm to obtain an optimal estimation value of the label signal intensity.
Optionally, the optimal estimation equation corresponds to a formula:
wherein, Xk+1|kThe intensity of the tag signal detected at time K +1, Xk+1|k+1Is the optimal estimated value of the label signal strength at the moment K +1, Kk+1In order to obtain the gain of the kalman filter,
is new.
Optionally, the calculating, by the data processing center, a distance between the object to be measured and the sensor according to the optimal estimation value of the tag signal strength includes:
and substituting the optimal estimation value of the label signal intensity into a path loss model stored in a data processing center, and calculating a distance value between the object to be measured and the inductor. Optionally, the equation corresponding to the number path loss model is:
wherein, RSSI is the label signal strength, d is the distance value between the object to be measured and the inductor, n is the path attenuation factor, d0 is the reference distance, a is the signal strength at the position d0 away from the receiving node, and is gaussian random noise with zero mean and variance.
According to another aspect of the present application, there is provided a passive RFID mobile ranging data noise reduction apparatus, including:
the configuration module is used for configuring the electronic tag for the object to be detected;
the acquisition module is used for acquiring the label signal intensity corresponding to the object to be detected at a preset period according to a preset route;
the input module is used for inputting the label signal strength into an optimized denoising model which is successfully trained to obtain an optimal estimation value of the label signal strength;
and the resolving module is used for resolving the distance between the object to be measured and the inductor by the data processing center according to the optimal estimated value of the label signal intensity.
Optionally, the apparatus further comprises:
the device comprises a creating module, a denoising module and a denoising module, wherein the creating module is used for creating an optimized denoising model which comprises a Kalman filtering calculation module and a BP neural network model;
and the training module is used for carrying out model training according to the BP neural network model so as to enable the optimized denoising model to meet a preset standard.
Optionally, the training module is further configured to:
acquiring first label signal intensity corresponding to the object to be detected, wherein the first label signal intensity corresponds to an average value of label signal intensities corresponding to each first acquisition point in a preset route;
importing the first label signal strength into a Kalman filtering calculation module of the optimized denoising model to obtain first input data comprising three parameters, wherein the three parameters comprise a difference between a label signal strength predicted value and a label signal strength optimal estimated value, Kalman filtering gain and innovation;
calculating a first distance between the object to be measured and the sensor according to the data acquisition position recorded by the GPS receiver in each first acquisition point and the position coordinate of the object to be measured, wherein the first distance corresponds to an average value of the distances between each first acquisition point and the object to be measured in a preset route;
substituting the first label signal strength and the first distance into a path loss model, and calculating to obtain a label signal strength theoretical value;
determining the difference between the theoretical value of the label signal strength and the predicted value of the label signal strength as first output data;
training the BP neural network model by using the first input data and the first output data;
re-acquiring second label signal intensity corresponding to the object to be detected according to the preset route, wherein the second label signal intensity corresponds to the average value of the label signal intensity corresponding to each second acquisition point in the preset route;
importing the second label signal intensity into the optimized denoising model to obtain second output data;
calculating a label signal intensity predicted value corresponding to the second label signal intensity according to a Kalman filtering algorithm;
calculating a label signal strength theoretical value corresponding to the second label signal strength according to a path loss model;
calculating a difference value between the label signal strength theoretical value and the second output data and the label signal strength predicted value, if the difference value is judged to be equal to zero or smaller than a preset threshold value, determining that the optimized denoising model passes training, and determining Kalman filtering gain and innovation in the Kalman filtering calculation module as target parameters;
and if the difference value is judged to be larger than or equal to the preset threshold value, adjusting Kalman filtering gain and innovation in the Kalman filtering calculation module until an output value meets an error requirement so as to further determine the target parameter.
Optionally, the input module is further configured to:
acquiring target Kalman filtering gain and target information corresponding to the target parameters;
and substituting the label signal intensity, the target Kalman filtering gain and the target information into a Kalman filtering calculation module of the optimized denoising model, and calculating by using an optimal estimation equation of the Kalman filtering algorithm to obtain an optimal estimation value of the label signal intensity.
Optionally, the optimal estimation equation applied in the input module corresponds to a formula:
wherein, Xk+1|kThe intensity of the tag signal detected at time K +1, Xk+1|k+1Is the optimal estimated value of the label signal strength at the moment K +1, Kk+1Is CarrThe gain of the man-filtering is,
is new.
Optionally, the calculating module is further configured to: and substituting the optimal estimation value of the label signal intensity into a path loss model stored in a data processing center, and calculating a distance value between the object to be measured and the inductor. Optionally, the equation corresponding to the number path loss model applied in the calculation module is as follows:wherein, RSSI is the label signal intensity, d is the distance value between the object to be measured and the inductor, n is the path attenuation factor, d0For reference distance, A is the signal strength at d0 from the receiving node, XσIs gaussian random noise with zero mean and variance.
According to yet another aspect of the application, a non-transitory readable storage medium is provided, on which a computer program is stored, which program, when executed by a processor, implements the above method of passive RFID mobile ranging data noise reduction.
According to yet another aspect of the application, a computer device is provided, comprising a non-volatile readable storage medium, a processor and a computer program stored on the non-volatile readable storage medium and executable on the processor, the processor implementing the above passive RFID mobile ranging data noise reduction method when executing the program.
By the technical scheme, the method, the device and the computer equipment for reducing the noise of the passive RFID mobile ranging data are provided, compared with the traditional Kalman filtering algorithm calculation mode, the method creates an optimized denoising model for eliminating data interference, inputs the acquired label signal intensity corresponding to the object to be measured into the trained optimized denoising model, can obtain the optimal estimated value of the label signal intensity, thereby improving the detection precision of the label signal intensity, then calculating more accurate distance data by utilizing the optimal estimated value of the label signal intensity and the path loss model stored in the data processing center, realizing the real-time accurate monitoring of the position of the component or the material, and then can improve the work efficiency of management and maintainer, effectively reduce maintainer's work load and administrative cost.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application to the disclosed embodiment. In the drawings:
FIG. 1 is a schematic diagram illustrating a flow chart of passive RFID mobile ranging data noise reduction according to an embodiment of the present application;
FIG. 2 is a schematic flow chart illustrating another passive RFID mobile ranging data noise reduction provided by an embodiment of the present application;
FIG. 3 is a schematic diagram illustrating an operating scenario of noise reduction of passive RFID mobile ranging data according to an embodiment of the present application;
FIG. 4 is a schematic diagram illustrating a training structure of an optimized denoising model provided by an embodiment of the present application;
FIG. 5 is a schematic structural diagram illustrating an apparatus for passive RFID mobile ranging data noise reduction according to an embodiment of the present disclosure;
fig. 6 shows a schematic structural diagram of another passive RFID mobile ranging data noise reduction provided in an embodiment of the present application.
Detailed Description
The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
Aiming at the problem that errors exist in measured and calculated distances easily due to environmental interference when a passive electronic tag is used for positioning at present, the embodiment of the application provides a passive RFID mobile ranging data noise reduction method, as shown in FIG. 1, the method comprises the following steps:
101. and configuring an electronic tag for the object to be detected.
The electronic tag is configured at the object to be detected, so that the passive RFID can be used for detecting the tag signal intensity corresponding to the electronic tag, and further the specific position of the object to be detected is determined.
102. And acquiring the label signal intensity corresponding to the object to be detected according to a preset route in a preset period.
The preset route can be set according to a specific application scene, and a specific route starting point, a specific route ending point and each data acquisition point can be set; the preset period corresponds to the acquisition interval of the signal intensity of the adjacent tags, and the value setting can be specifically carried out according to the specific application scene.
For this embodiment, in a specific application scenario, after the tag signal strength corresponding to each data acquisition point is acquired, the acquisition position and the acquisition time of the tag signal strength may be used as a tag, and each tag signal strength is stored in the database.
103. And inputting the label signal intensity into the successfully trained optimized denoising model to obtain an optimal estimation value of the label signal intensity.
The purpose of optimizing the denoising model is to filter the influence of environmental data on the label signal intensity, and the optimal estimation value of the label signal intensity is obtained by utilizing the optimizing denoising model.
104. And the data processing center calculates the distance between the object to be measured and the inductor according to the optimal estimation value of the label signal intensity.
The data processing center comprises a path loss model, in a specific application scene, the distance between the object to be detected and the inductor can be calculated by using the path loss model, after the calculation of the distance is completed, the data processing center can also carry out effectiveness detection on the calculated distance value, and when the effective value of the value is judged, the distance between the object to be detected and the inductor is further output as a detection result.
By the method for reducing the noise of the passive RFID mobile ranging data, an optimized denoising model for eliminating data interference can be created in advance, the acquired label signal strength corresponding to the object to be measured is input into the trained optimized denoising model, the optimal estimated value of the label signal strength can be obtained, the detection precision of the label signal strength can be improved, accurate distance data can be obtained by calculating the optimal estimated value of the label signal strength and the path loss model stored in the data processing center, the position of a component or material can be accurately monitored in real time, the working efficiency of management and maintenance personnel can be improved, and the workload and the management cost of the maintenance personnel can be effectively reduced.
Further, as a refinement and an extension of the specific implementation of the foregoing embodiment, in order to fully illustrate the specific implementation process in this embodiment, another passive RFID mobile ranging data noise reduction method is provided, as shown in fig. 2, and the method includes:
201. and configuring an electronic tag for the object to be detected.
In a specific application scene, an RFID passive electronic tag can be configured for an object to be measured, and then the specific position of the object to be measured is determined by detecting the tag signal intensity corresponding to the passive RFID detection electronic tag.
202. And acquiring the label signal intensity corresponding to the object to be detected according to a preset route in a preset period.
In a specific application scenario, as shown in fig. 3, the RFID sensor may be placed on a distance measuring cart, and connected to a notebook computer in a network cable port manner, and the tag signal strength value obtained by the RFID sensor is stored in a designed database. Designing a moving line of the distance measuring vehicle, and adjusting proper RFID data reading and recording frequency; meanwhile, a high-precision GPS receiver is arranged at the position of the sensor and is connected with a notebook computer in a serial port mode, and GPS data are stored in a database; the distance measuring vehicle executes a fixed moving line to keep running at a constant speed, the distance measuring vehicle approaches a target to be measured by about 10 meters, and after the received tag signal intensity data are stable, the received tag signal intensity data and the GPS position data start to be stored in a database until the planned moving line of the distance measuring vehicle is completed.
203. And creating an optimized denoising model, wherein the optimized denoising model comprises a Kalman filtering calculation module and a BP neural network model.
The Kalman filtering calculation module in the optimized denoising model is mainly used for calculating a predicted value of the label signal strength; the BP neural network model is mainly used for correcting parameters in a Kalman filtering calculation module, the BP neural network is a multi-layer feedforward network trained according to error back propagation (error back propagation for short), the algorithm is called as BP algorithm, the basic idea is a gradient descent method, and the error mean square error of an actual output value and an expected output value of the network is minimized by using a gradient search technology. In the scheme, the method and the device are mainly used for learning and correcting the parameter values in the Kalman filtering calculation module, so that the error between the output predicted value of the label signal strength and the theoretical value of the label signal strength is small.
204. And carrying out model training according to the BP neural network model so as to enable the optimized denoising model to meet a preset standard.
For the present embodiment, in a specific application scenario, the step 204 of the embodiment may specifically include: acquiring first label signal intensity corresponding to a distance object to be detected, wherein the first label signal intensity corresponds to an average value of label signal intensities corresponding to each first acquisition point in a preset route; importing the first label signal strength into a Kalman filtering calculation module of an optimized denoising model to obtain first input data comprising three parameters, wherein the three parameters comprise the difference between a label signal strength predicted value and a label signal strength optimal estimated value, Kalman filtering gain and innovation; calculating a first distance between the object to be measured and the sensor according to the data acquisition position recorded by the GPS receiver in each first acquisition point and the position coordinate of the object to be measured, wherein the first distance corresponds to an average value of the distances between each first acquisition point and the object to be measured in a preset route; substituting the first label signal strength and the first distance into a path loss model, and calculating to obtain a label signal strength theoretical value; determining the difference between the theoretical value of the label signal intensity and the predicted value of the label signal intensity as first output data; training a BP neural network model by utilizing first input data and first output data; re-acquiring second label signal intensity corresponding to the object to be detected according to the preset route, wherein the second label signal intensity corresponds to the average value of the label signal intensity corresponding to each second acquisition point in the preset route; importing the second label signal intensity into an optimized denoising model to obtain second output data; calculating a label signal intensity predicted value corresponding to the second label signal intensity according to a Kalman filtering algorithm; calculating a label signal intensity theoretical value corresponding to the second label signal intensity according to the path loss model; calculating a difference value between the label signal intensity theoretical value and the second output data and the label signal intensity predicted value, if the difference value is judged to be equal to zero or smaller than a preset threshold value, determining that an optimized denoising model passes training, and determining Kalman filtering gain and innovation in a Kalman filtering calculation module as target parameters; and if the difference value is judged to be larger than or equal to the preset threshold value, adjusting Kalman filtering gain and innovation in the Kalman filtering calculation module until the output value meets the error requirement so as to further determine the target parameter.
In a specific application scenario, the first acquisition point and the second acquisition point may correspond to the same position of the preset line or may be different positions, in order to ensure the training precision of the optimized denoising model, in this embodiment, different positions are preferred, and the specific positions may be preset when the preset line is planned.
In a specific application scenario, the training principle of optimizing the denoising model is shown in fig. 4, the difference between the predicted value of the label signal strength and the optimal estimated value of the label signal strength, the kalman filter gain and the innovation are extracted by using the RSSI optimal estimation of the kalman filter in advance, the three variable parameters are used as training input data of the BP neural network model, the predicted value of the label signal strength and the true value of the label signal strength are calculated separately, the difference between the predicted value of the label signal strength and the true value of the label signal strength is used as training output data of the BP neural network model, and the parameters in the training input data are adjusted by using the training output data, so that the optimal estimated value of the label signal strength output by the kalman filter calculation module is approximately equal to the true value of the label signal strength. Acquiring the label signal intensity according to the same preset route, inputting the label signal intensity into an optimized denoising model formed by a Kalman filtering calculation module and a BP neural network model, obtaining output data, judging whether the sum of a label signal intensity predicted value and the output data which are calculated independently is equal to a label signal intensity theoretical value corresponding to the label signal intensity calculated according to a path loss model, wherein the label signal intensity predicted value corresponds to the difference between a label signal intensity predicted value and a label signal intensity optimal estimated value after the optimized denoising model is optimized; if the addition result is judged to be approximately equal to the label signal strength theoretical value, the parameter in the current Kalman filtering calculation module can be judged to be the target parameter, the target parameter can be applied to the calculation of the label signal strength optimal estimation value of the object to be detected in a specific scene, and if the difference value between the label signal strength theoretical value and the second output data and the label signal strength predicted value is judged to be larger than or equal to a preset threshold value, Kalman filtering gain and innovation in the Kalman filtering calculation module are adjusted until the output value meets the error requirement, so that the target parameter can be further determined.
205. And acquiring target Kalman filtering gain and target information corresponding to the target parameters.
For this embodiment, the target kalman filter gain and the target information acquired in step 204 of the embodiment may be determined as calculation parameters in the kalman filter calculation module, and are used to participate in calculating the subsequent optimal estimation value of the tag signal strength and ensure that the optimal estimation value of the tag signal strength is approximately equal to the theoretical value of the tag signal strength.
206. And substituting the label signal intensity, the target Kalman filtering gain and the target information into a Kalman filtering calculation module of the optimized denoising model, and calculating by using an optimal estimation equation of a Kalman filtering algorithm to obtain an optimal estimation value of the label signal intensity.
For this embodiment, in a specific application scenario, the formula corresponding to the optimal estimation equation is as follows:
wherein, Xk+1|kThe intensity of the tag signal detected at time K + 1, Xk+1|k+1Is the optimal estimated value of the label signal strength at the moment K +1, Kk+1In order to obtain the gain of the kalman filter,
is new.
207. And substituting the optimal estimation value of the label signal intensity into a path loss model stored in a data processing center to calculate a distance value between the object to be measured and the inductor.
For this embodiment, in a specific application scenario, the formula corresponding to the number path loss model is as follows:
wherein, RSSI is the label signal intensity, d is the distance value between the object to be measured and the inductor, n is the path attenuation factor, d0For reference distance, A is the signal strength at d0 from the receiving node, XσIs gaussian random noise with zero mean and variance.
By the passive RFID mobile ranging data noise reduction method, a multi-station acquisition mode with different distances can be adopted for the same object to be measured according to a certain route, the RFID sensor acquires the label signal intensity at each moment, and the GPS receiver is used for recording the mobile position. Substituting the signal intensity into a preset optimization denoising program for training to obtain an optimization denoising model meeting a preset standard; in the training process, corresponding sample input and output values in the database are read, parameter training is carried out by using a BP neural network model, relevant parameters are adjusted step by step until the optimal estimated value of the label signal intensity is judged to be approximately equal to the theoretical value of the label signal intensity, and the optimal denoising model can be judged to pass through training, so that target parameters are extracted. The position detection of the object to be measured can be realized based on the target parameters, the label signal intensity corresponding to the object to be measured is led into the optimized denoising model, and the optimal estimated value of the label signal intensity is obtained. And then substituting the optimal estimated value of the label signal intensity into a path loss model stored in a data processing center, so that the distance value between the object to be measured and the inductor can be calculated. Compared with the traditional Kalman filtering algorithm, the method and the device can effectively reduce the influence of environmental factors, enhance the stability of data calculation, further enable the finally calculated distance value between the object to be measured and the sensor to be more real and effective, and reduce the detection error.
Further, as a concrete embodiment of the method shown in fig. 1 and fig. 2, an embodiment of the present application provides a device for reducing noise of passive RFID mobile ranging data, as shown in fig. 5, the device includes: the device comprises a configuration module 31, an acquisition module 32, an input module 33 and a resolving module 34.
The configuration module 31 is configured to configure an electronic tag for an object to be measured;
the acquisition module 32 is configured to acquire the tag signal strength corresponding to the object to be detected at a predetermined period according to a preset route;
the input module 33 is configured to input the label signal strength into the successfully trained optimized denoising model, and obtain an optimal estimation value of the label signal strength;
and the calculating module 34 is used for calculating the distance between the object to be measured and the inductor by the data processing center according to the optimal estimated value of the label signal intensity.
In a specific application scenario, in order to train and obtain an optimized denoising model for obtaining an optimal estimation value of the label signal strength, as shown in fig. 6, the apparatus further includes: a creation module 35 and a training module 36.
The creating module 35 is configured to create an optimized denoising model, where the optimized denoising model includes a kalman filter calculation module and a BP neural network model;
and the training module 36 is configured to perform model training according to the BP neural network model, so that the optimized denoising model meets a preset standard.
In a specific application scenario, in order to perform model training according to the BP neural network model so that the optimized denoising model meets a preset standard, the training module 36 is specifically configured to obtain a first tag signal intensity corresponding to an object to be detected, where the first tag signal intensity corresponds to an average value of tag signal intensities corresponding to each first acquisition point in a preset route; importing the first label signal strength into a Kalman filtering calculation module of an optimized denoising model to obtain first input data comprising three parameters, wherein the three parameters comprise the difference between a label signal strength predicted value and a label signal strength optimal estimated value, Kalman filtering gain and innovation; calculating a first distance between the object to be measured and the sensor according to the data acquisition position recorded by the GPS receiver in each first acquisition point and the position coordinate of the object to be measured, wherein the first distance corresponds to an average value of the distances between each first acquisition point and the object to be measured in a preset route; substituting the first label signal strength and the first distance into a path loss model, and calculating to obtain a label signal strength theoretical value; determining the difference between the theoretical value of the label signal intensity and the predicted value of the label signal intensity as first output data; training a BP neural network model by utilizing first input data and first output data; re-acquiring second label signal intensity corresponding to the object to be detected according to the preset route, wherein the second label signal intensity corresponds to the average value of the label signal intensity corresponding to each second acquisition point in the preset route; importing the second label signal intensity into an optimized denoising model to obtain second output data; calculating a label signal intensity predicted value corresponding to the second label signal intensity according to a Kalman filtering algorithm; calculating a label signal intensity theoretical value corresponding to the second label signal intensity according to the path loss model; calculating a difference value between the label signal strength theoretical value and the second output data and the label signal strength predicted value, if the difference value is judged to be equal to zero or smaller than a preset threshold value, determining that the optimized denoising model passes training, and determining Kalman filtering gain and innovation in the Kalman filtering calculation module as target parameters; and if the difference value is judged to be larger than or equal to the preset threshold value, adjusting Kalman filtering gain and innovation in the Kalman filtering calculation module until the output value meets the error requirement so as to further determine the target parameter.
Correspondingly, in order to obtain an optimal estimation value of the label signal strength, the input module 33 is specifically configured to obtain a target kalman filter gain and a target information corresponding to the target parameter; and substituting the label signal intensity, the target Kalman filtering gain and the target information into a Kalman filtering calculation module of the optimized denoising model, and calculating by using an optimal estimation equation of a Kalman filtering algorithm to obtain an optimal estimation value of the label signal intensity.
Optionally, the optimal estimation equation applied in the input module 33 corresponds to the formula:
wherein, Xk+1|kThe intensity of the tag signal detected at time K + 1, Xk+1|k+1Is the optimal estimated value of the label signal strength at the moment K +1, Kk+1In order to obtain the gain of the kalman filter,
is new.
In a specific application scenario, in order to calculate the distance between the object to be measured and the sensor according to the optimal estimation value of the tag signal strength by using the data processing center, the calculating module 34 is specifically configured to: and substituting the optimal estimation value of the label signal intensity into a path loss model stored in a data processing center to calculate a distance value between the object to be measured and the inductor.
Optionally, the equation corresponding to the number path loss model applied in the calculation module 34 is:
wherein, RSSI is the label signal intensity, d is the distance value between the object to be measured and the inductor, n is the path attenuation factor, d0For reference distance, A is the signal strength at d0 from the receiving node, XσIs gaussian random noise with zero mean and variance.
It should be noted that other corresponding descriptions of the functional units related to the passive RFID mobile ranging data noise reduction device provided in this embodiment may refer to the corresponding descriptions in fig. 1 to fig. 2, and are not described herein again.
Based on the methods shown in fig. 1 and fig. 2, correspondingly, the embodiment of the present application further provides a storage medium, on which a computer program is stored, and the program, when executed by a processor, implements the method for denoising passive RFID mobile ranging data shown in fig. 1 and fig. 2.
Based on such understanding, the technical solution of the present application may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.), and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method of the embodiments of the present application.
Based on the method shown in fig. 1 and fig. 2 and the virtual device embodiment shown in fig. 5 and fig. 6, in order to achieve the above object, an embodiment of the present application further provides a computer device, which may specifically be a personal computer, a server, a network device, and the like, where the entity device includes a storage medium and a processor; a storage medium for storing a computer program; a processor for executing a computer program to implement the above-described method for passive RFID mobile ranging data noise reduction as shown in fig. 1 and 2.
Optionally, the computer device may also include a user interface, a network interface, a camera, Radio Frequency (RF) circuitry, sensors, audio circuitry, a WI-FI module, and so forth. The user interface may include a Display screen (Display), an input unit such as a keypad (Keyboard), etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (e.g., a bluetooth interface, WI-FI interface), etc.
It will be understood by those skilled in the art that the computer device structure provided in the present embodiment is not limited to the physical device, and may include more or less components, or combine some components, or arrange different components.
The nonvolatile readable storage medium can also comprise an operating system and a network communication module. The operating system is a program of hardware and software resources of the physical device for passive RFID mobile ranging data noise reduction, and supports the running of an information processing program and other software and/or programs. The network communication module is used for realizing communication among components in the nonvolatile readable storage medium and communication with other hardware and software in the entity device.
Through the above description of the embodiments, those skilled in the art will clearly understand that the present application can be implemented by software plus a necessary general hardware platform, and can also be implemented by hardware. Through the technical scheme who uses this application, compare with current prior art, this application can be to the same object of waiting to find a distance, according to a certain route, adopts different distance multistation collection modes, and the RFID inductor acquires the label signal strength at each moment, utilizes GPS receiver record mobile position. Substituting the signal intensity into a preset optimization denoising program for training to obtain an optimization denoising model meeting a preset standard; in the training process, corresponding sample input and output values in the database are read, parameter training is carried out by using a BP neural network model, relevant parameters are adjusted step by step until the optimal estimated value of the label signal intensity is judged to be approximately equal to the theoretical value of the label signal intensity, and the optimal denoising model can be judged to pass through training, so that target parameters are extracted. The position detection of the object to be measured can be realized based on the target parameters, the label signal intensity corresponding to the object to be measured is led into the optimized denoising model, and the optimal estimated value of the label signal intensity is obtained. And then substituting the optimal estimated value of the label signal intensity into a path loss model stored in a data processing center, so that the distance value between the object to be measured and the inductor can be calculated. Compared with the traditional Kalman filtering algorithm, the method and the device can effectively reduce the influence of environmental factors, enhance the stability of data calculation, further enable the finally calculated distance value between the object to be measured and the sensor to be more real and effective, and reduce the detection error.
Those skilled in the art will appreciate that the figures are merely schematic representations of one preferred implementation scenario and that the blocks or flow diagrams in the figures are not necessarily required to practice the present application. Those skilled in the art will appreciate that the modules in the devices in the implementation scenario may be distributed in the devices in the implementation scenario according to the description of the implementation scenario, or may be located in one or more devices different from the present implementation scenario with corresponding changes. The modules of the implementation scenario may be combined into one module, or may be further split into a plurality of sub-modules.
The above application serial numbers are for description purposes only and do not represent the superiority or inferiority of the implementation scenarios. The above disclosure is only a few specific implementation scenarios of the present application, but the present application is not limited thereto, and any variations that can be made by those skilled in the art are intended to fall within the scope of the present application.
Claims (10)
1. A method for denoising passive RFID mobile ranging data is characterized by comprising the following steps:
configuring an electronic tag for an object to be detected;
acquiring the label signal intensity corresponding to the object to be detected according to a preset route in a preset period;
inputting the label signal intensity into an optimized denoising model successfully trained to obtain an optimal estimation value of the label signal intensity;
and the data processing center calculates the distance between the object to be measured and the inductor according to the label signal intensity optimal estimation value.
2. The method as claimed in claim 1, wherein before inputting the label signal strength into the successfully trained optimized denoising model and obtaining an optimal estimate of the label signal strength, the method further comprises:
creating an optimized denoising model, wherein the optimized denoising model comprises a Kalman filtering calculation module and a BP neural network model;
and carrying out model training according to the BP neural network model so as to enable the optimized denoising model to meet a preset standard.
3. The method according to claim 2, wherein the model training according to the BP neural network model to make the optimized denoising model meet a preset standard specifically comprises:
acquiring first label signal intensity corresponding to the object to be detected, wherein the first label signal intensity corresponds to an average value of label signal intensities corresponding to each first acquisition point in a preset route;
importing the first label signal strength into a Kalman filtering calculation module of the optimized denoising model to obtain first input data comprising three parameters, wherein the three parameters comprise a difference between a label signal strength predicted value and a label signal strength optimal estimated value, Kalman filtering gain and innovation;
calculating a first distance between the object to be measured and the sensor according to the data acquisition position recorded by the GPS receiver in each first acquisition point and the position coordinate of the object to be measured, wherein the first distance corresponds to an average value of the distances between each first acquisition point and the object to be measured in a preset route;
substituting the first label signal strength and the first distance into a path loss model, and calculating to obtain a label signal strength theoretical value;
determining the difference between the theoretical value of the label signal strength and the predicted value of the label signal strength as first output data;
training the BP neural network model by using the first input data and the first output data;
re-acquiring second label signal intensity corresponding to the object to be detected according to the preset route, wherein the second label signal intensity corresponds to the average value of the label signal intensity corresponding to each second acquisition point in the preset route;
importing the second label signal intensity into the optimized denoising model to obtain second output data;
calculating a label signal intensity predicted value corresponding to the second label signal intensity according to a Kalman filtering algorithm;
calculating a label signal strength theoretical value corresponding to the second label signal strength according to a path loss model;
calculating a difference value between the label signal strength theoretical value and the second output data and the label signal strength predicted value, if the difference value is judged to be equal to zero or smaller than a preset threshold value, determining that the optimized denoising model passes training, and determining Kalman filtering gain and innovation in the Kalman filtering calculation module as target parameters;
and if the difference value is judged to be larger than or equal to the preset threshold value, adjusting Kalman filtering gain and innovation in the Kalman filtering calculation module until an output value meets an error requirement so as to further determine the target parameter.
4. The method according to claim 3, wherein the inputting the label signal strength into an optimized denoising model successfully trained to obtain an optimal estimate value of the label signal strength comprises:
acquiring target Kalman filtering gain and target information corresponding to the target parameters;
and substituting the label signal intensity, the target Kalman filtering gain and the target information into a Kalman filtering calculation module of the optimized denoising model, and calculating by using an optimal estimation equation of the Kalman filtering algorithm to obtain an optimal estimation value of the label signal intensity.
5. The method of claim 4, wherein the optimal estimation equation corresponds to the formula:
wherein, Xk+1|kThe intensity of the tag signal detected at time K +1, Xk+1|k+1Is the optimal estimated value of the label signal strength at the moment K +1, Kk+1In order to obtain the gain of the kalman filter,
is new.
6. The method according to claim 4, wherein the data processing center calculates the distance between the object to be measured and the sensor according to the optimal estimation value of the tag signal strength, and specifically comprises:
and substituting the optimal estimation value of the label signal intensity into a path loss model stored in a data processing center, and calculating a distance value between the object to be measured and the inductor.
7. The method of claim 6, wherein the fractional path loss model corresponds to the formula:
wherein, RSSI is the label signal intensity, d is the distance value between the object to be measured and the inductor, n is the path attenuation factor, d0For reference distance, A is the signal strength at d0 from the receiving node, XσIs gaussian random noise with zero mean and variance.
8. A device for reducing noise of passive RFID mobile ranging data is characterized by comprising:
the configuration module is used for configuring the electronic tag for the object to be detected;
the acquisition module is used for acquiring the label signal intensity corresponding to the object to be detected at a preset period according to a preset route;
the input module is used for inputting the label signal strength into an optimized denoising model which is successfully trained to obtain an optimal estimation value of the label signal strength;
and the resolving module is used for resolving the distance between the object to be measured and the inductor by the data processing center according to the optimal estimated value of the label signal intensity.
9. A non-transitory readable storage medium having stored thereon a computer program, wherein the program when executed by a processor implements the method of passive RFID mobile ranging data noise reduction of any of claims 1 to 7.
10. A computer device comprising a non-volatile readable storage medium, a processor, and a computer program stored on the non-volatile readable storage medium and executable on the processor, wherein the processor when executing the program implements the method of passive RFID mobile ranging data noise reduction of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910950182.1A CN110909873B (en) | 2019-10-08 | 2019-10-08 | Method and device for denoising passive RFID (radio frequency identification) mobile ranging data and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910950182.1A CN110909873B (en) | 2019-10-08 | 2019-10-08 | Method and device for denoising passive RFID (radio frequency identification) mobile ranging data and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110909873A true CN110909873A (en) | 2020-03-24 |
| CN110909873B CN110909873B (en) | 2022-07-22 |
Family
ID=69814964
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910950182.1A Active CN110909873B (en) | 2019-10-08 | 2019-10-08 | Method and device for denoising passive RFID (radio frequency identification) mobile ranging data and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110909873B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111693139A (en) * | 2020-06-19 | 2020-09-22 | 浙江讯飞智能科技有限公司 | Sound intensity measuring method, device, equipment and storage medium |
| CN112733975A (en) * | 2020-12-17 | 2021-04-30 | 深圳市信心智能标签技术有限公司 | Positioning method, device, equipment and medium based on RFID (radio frequency identification) label |
| CN113466841A (en) * | 2021-07-16 | 2021-10-01 | Oppo广东移动通信有限公司 | Distance measuring method, distance measuring device, electronic device, and storage medium |
| CN113466841B (en) * | 2021-07-16 | 2024-05-03 | Oppo广东移动通信有限公司 | Distance measuring method, device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101782652A (en) * | 2010-02-08 | 2010-07-21 | 上海和为科技有限公司 | Indoor positioning system based on RFID technology |
| CN105512722A (en) * | 2015-11-26 | 2016-04-20 | 中国石油大学(华东) | BP neural network-based Kalman gain correction method |
| CN106714302A (en) * | 2017-01-23 | 2017-05-24 | 吉林大学 | Indoor positioning device based on BP-Landmarc neural network and control method |
| CN108682023A (en) * | 2018-03-21 | 2018-10-19 | 南京理工大学 | Close coupling Unscented kalman tracking filter algorithm based on Elman neural networks |
| CN109444813A (en) * | 2018-10-26 | 2019-03-08 | 南京邮电大学 | A kind of RFID indoor orientation method based on BP and DNN amphineura network |
-
2019
- 2019-10-08 CN CN201910950182.1A patent/CN110909873B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101782652A (en) * | 2010-02-08 | 2010-07-21 | 上海和为科技有限公司 | Indoor positioning system based on RFID technology |
| CN105512722A (en) * | 2015-11-26 | 2016-04-20 | 中国石油大学(华东) | BP neural network-based Kalman gain correction method |
| CN106714302A (en) * | 2017-01-23 | 2017-05-24 | 吉林大学 | Indoor positioning device based on BP-Landmarc neural network and control method |
| CN108682023A (en) * | 2018-03-21 | 2018-10-19 | 南京理工大学 | Close coupling Unscented kalman tracking filter algorithm based on Elman neural networks |
| CN109444813A (en) * | 2018-10-26 | 2019-03-08 | 南京邮电大学 | A kind of RFID indoor orientation method based on BP and DNN amphineura network |
Non-Patent Citations (2)
| Title |
|---|
| JING, CF (JING, CHANGFENG) ET AL.: "A Robust Noise Mitigation Method for the Mobile RFID Location in Built Environment", 《SENSORS》 * |
| 周蕾蕾等: "粒子群神经网络算法的RFID定位应用研究", 《中国医疗设备》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111693139A (en) * | 2020-06-19 | 2020-09-22 | 浙江讯飞智能科技有限公司 | Sound intensity measuring method, device, equipment and storage medium |
| CN111693139B (en) * | 2020-06-19 | 2022-04-22 | 浙江讯飞智能科技有限公司 | Sound intensity measuring method, device, equipment and storage medium |
| CN112733975A (en) * | 2020-12-17 | 2021-04-30 | 深圳市信心智能标签技术有限公司 | Positioning method, device, equipment and medium based on RFID (radio frequency identification) label |
| CN112733975B (en) * | 2020-12-17 | 2023-05-23 | 深圳市信心智能标签技术有限公司 | Positioning method, device, equipment and medium based on RFID (radio frequency identification) tag |
| CN113466841A (en) * | 2021-07-16 | 2021-10-01 | Oppo广东移动通信有限公司 | Distance measuring method, distance measuring device, electronic device, and storage medium |
| CN113466841B (en) * | 2021-07-16 | 2024-05-03 | Oppo广东移动通信有限公司 | Distance measuring method, device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110909873B (en) | 2022-07-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109444813B (en) | RFID indoor positioning method based on BP and DNN double neural networks | |
| CN109275095B (en) | Bluetooth-based indoor positioning system, positioning equipment and positioning method | |
| WO2016165459A1 (en) | Method and device for indoor positioning | |
| CN105424026A (en) | Indoor navigation and localization method and system based on point cloud tracks | |
| CN107517446A (en) | Indoor orientation method and device based on Wi Fi focuses | |
| WO2017121168A1 (en) | Cluster-based magnetic positioning method, device and system | |
| CN106525031A (en) | Combined indoor positioning method | |
| CN110909873B (en) | Method and device for denoising passive RFID (radio frequency identification) mobile ranging data and computer equipment | |
| WO2016103499A1 (en) | Location determination system | |
| US20140025334A1 (en) | Terminal and server for modifying magnetic field and method thereof | |
| US9749873B1 (en) | Estimation devices and methods for estimating communication quality of wireless network and method for installing meters thereof | |
| CN106470478B (en) | Positioning data processing method, device and system | |
| CN111148030A (en) | Fingerprint database updating method and device, server and storage medium | |
| JPWO2016103498A1 (en) | Location determination system and location determination program | |
| CN107607122A (en) | Towards the location fingerprint storehouse structure and dynamic updating method of indoor positioning | |
| Chen et al. | MeshMap: A magnetic field-based indoor navigation system with crowdsourcing support | |
| CN114631365A (en) | Electronic device for detecting position and method thereof | |
| Wei et al. | Efficient Wi-Fi fingerprint crowdsourcing for indoor localization | |
| Vy et al. | A precise tracking algorithm using PDR and Wi-Fi/iBeacon corrections for smartphones | |
| Wei et al. | MM-Loc: Cross-sensor indoor smartphone location tracking using multimodal deep neural networks | |
| CN110856100B (en) | Method and device for terminal positioning and positioning model construction based on 5G signal | |
| CN106717083A (en) | Method for position detection by mobile computing device, and mobile computing device performing same | |
| Gunther et al. | A UHF RFID positioning system for use in warehouse navigation by employees with cognitive disability | |
| CN110517063A (en) | Method, apparatus and server are determined into shop consumption person-time | |
| Huang et al. | A Directional Particle Filter-Based Multi-Floor Indoor Positioning System |
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 |