CN114531729A - Positioning method, system, storage medium and device based on channel state information - Google Patents
Positioning method, system, storage medium and device based on channel state information Download PDFInfo
- Publication number
- CN114531729A CN114531729A CN202210432885.7A CN202210432885A CN114531729A CN 114531729 A CN114531729 A CN 114531729A CN 202210432885 A CN202210432885 A CN 202210432885A CN 114531729 A CN114531729 A CN 114531729A
- Authority
- CN
- China
- Prior art keywords
- phase data
- training
- subcarrier phase
- neural network
- network 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
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012549 training Methods 0.000 claims abstract description 144
- 238000003062 neural network model Methods 0.000 claims abstract description 53
- 239000011159 matrix material Substances 0.000 claims description 38
- 230000006870 function Effects 0.000 claims description 31
- 238000004422 calculation algorithm Methods 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 13
- 210000002569 neuron Anatomy 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 9
- 230000009466 transformation Effects 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 abstract description 44
- 238000001514 detection method Methods 0.000 abstract description 27
- 230000008569 process Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 230000007613 environmental effect Effects 0.000 description 6
- 230000008859 change Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 239000000969 carrier Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000011478 gradient descent method Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- BULVZWIRKLYCBC-UHFFFAOYSA-N phorate Chemical compound CCOP(=S)(OCC)SCSCC BULVZWIRKLYCBC-UHFFFAOYSA-N 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
- H04W64/006—Locating users or terminals or network equipment for network management purposes, e.g. mobility management with additional information processing, e.g. for direction or speed determination
-
- 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
- G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
- G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
- G01S5/0252—Radio frequency fingerprinting
- G01S5/02521—Radio frequency fingerprinting using a radio-map
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a positioning method, a system, a storage medium and equipment based on channel state information, which comprises the steps of obtaining a plurality of groups of subcarrier phase data which are in one-to-one correspondence with a plurality of positions, and constructing a training set according to the calibrated subcarrier phase data; inputting the training set into a deep neural network model for training to obtain an optimal deep neural network model and fingerprint data corresponding to a target position, and establishing a fingerprint database according to the fingerprint data; and matching the subcarrier phase data of the moving target with a fingerprint database to determine the position of the moving target. According to the positioning method, the positioning system, the storage medium and the positioning equipment based on the channel state information, the influence of the subcarrier phase data on factors such as environment in a low signal-to-noise ratio environment is reduced through deep neural network training, and the accuracy and the reliability of the subcarrier phase data for target position detection are improved.
Description
Technical Field
The present invention relates to the field of wireless signal and information processing technologies, and in particular, to a positioning method, system, storage medium, and device based on channel state information.
Background
With the development of environmental detection technology research, CSI (channel state information) becomes a typical application of environmental detection. CSI is a set of information describing the state of a wireless channel, including information such as subcarrier signal strength, amplitude, phase, and signal delay. Compared with the prior environment positioning and ranging technology based on RSSI (received signal Strength indicator), the adopted CSI has higher fine granularity, namely, the target environment can be described more finely, meanwhile, the CSI-based positioning and ranging technology has better stability for the interference of external environment factors, the problem of low accuracy and the like caused by multipath effect is solved, and the positioning and ranging technology has better positioning and ranging effects and service capability.
When the CSI is applied to environment detection, the received CSI cannot be directly used due to the problems that the frequency cannot be accurately synchronized when equipment receives the CSI in the CSI transmitting and receiving process, the signal error of the received CSI can be eliminated by reasonably converting and processing the CSI, and higher detection precision can be obtained in the applications of environment detection and the like.
In practical application, however, receiving errors are eliminated, and signals are affected by factors such as blocking, interference, noise and the like in the transmission process in a multi-obstacle and multi-path complex environment, so that certain deviation exists between CSI transmission and CSI reception, and certain influence is brought to subsequent phase calibration, establishment of a fingerprint library and fingerprint matching. Therefore, the reliability and accuracy of the position detection of the moving object may be deteriorated in a complicated environment by establishing the fingerprint database using the conventional method of directly measuring and directly using the received data.
Disclosure of Invention
Based on this, the present invention provides a positioning method, system, storage medium and device based on channel state information, which solves the problem in the background art that the reliability and accuracy of the position detection of a moving target may be deteriorated in a complex environment.
One aspect of the present invention provides a positioning method based on channel state information, the method comprising:
pre-dividing a target position into a plurality of positions, acquiring a plurality of groups of subcarrier phase data which correspond to the plurality of positions one by one, calibrating the plurality of groups of subcarrier phase data, and constructing a training set according to the calibrated subcarrier phase data;
constructing a deep neural network model, inputting a training set into the deep neural network model for training to obtain an optimal deep neural network model and fingerprint data corresponding to a target position;
and establishing a fingerprint database according to the fingerprint data, acquiring the subcarrier phase data of the moving target, matching the subcarrier phase data of the moving target with the fingerprint database, and determining the position of the moving target according to the matching result.
The positioning method based on the channel state information of the invention constructs a deep neural network training model and pre-divided subcarrier phase data of a plurality of positions, obtaining an optimal deep neural network model through deep neural network training, reducing the influence of subcarrier phase data due to factors such as environment and the like in a low signal-to-noise ratio environment through the deep neural network training, establishing a fingerprint library corresponding to a target position according to the subcarrier phase data calibrated after the training, comparing the subcarrier data of a moving target with the fingerprint library to obtain the position information of the moving target, the accuracy and the reliability of the received subcarrier phase data are improved through deep neural network training, therefore, the reliability and the accuracy of the position detection of the moving target are improved, and the problem that the reliability and the accuracy of the position detection of the moving target are poor in the complex environment in the background technology is solved.
In addition, the above positioning method based on channel state information according to the present invention may further have the following additional technical features:
further, the deep neural network model comprises an input layer, a plurality of hidden layers and an output layer, wherein the input layer is used for inputting an input matrix established by a plurality of groups of subcarrier phase data, the output layer is used for outputting an output matrix obtained by training the plurality of groups of subcarrier phase data through the plurality of hidden layers, and the output matrix comprises the optimal deep neural network model and fingerprint data corresponding to a target position.
Further, the steps of constructing a deep neural network model, inputting the training set into the deep neural network model for training to obtain the optimal deep neural network model and the fingerprint data corresponding to the target position comprise:
and constructing a model loss function, training the training set based on the deep neural network model of the model loss function, and outputting the optimal model parameters of the deep neural network model and the output fingerprint data of the corresponding target position when the output value of the model loss function reaches the network model parameters of a preset value, namely the optimal model parameters of the deep neural network model.
Further, the step of training the training set based on the deep neural network model of the model loss function includes:
inputting a plurality of groups of subcarrier phase data into the hidden layers after passing through the input layers, estimating the training result of each hidden layer of the subcarrier phase data according to a contrast divergence algorithm, and estimating network model parameters between different hidden layers according to the training result of each hidden layer.
Further, the hidden layers comprise a first hidden layer, a second hidden layer and a third hidden layer; the method comprises the following steps of estimating the training result of each hidden layer of subcarrier phase data according to a contrast divergence algorithm, and estimating network model parameters between different hidden layers according to the training result of each hidden layer, wherein the steps comprise:
performing joint probability distribution multiplication on the input of the input layer and the output state of each neuron of the first hidden layer to estimate the output of the first hidden layer;
then, the output of the first hidden layer and the output state of each neuron of the second hidden layer are subjected to joint probability distribution multiplication to estimate the output of the second hidden layer;
performing joint probability distribution multiplication on the output of the second hidden layer and the output state of each neuron of the third hidden layer to estimate the output of the third hidden layer;
and reversely calculating the input of each hidden layer according to the output of each hidden layer, and further obtaining network model parameters among different hidden layers according to the input of each hidden layer.
Further, the step of estimating the training result of each hidden layer of the subcarrier phase data according to the contrast divergence algorithm further comprises:
and finally, obtaining the optimal network model parameters by comparing the errors between the input and the output of each hidden layer and continuously adjusting the network model parameters between different hidden layers by using an error back propagation algorithm.
Further, the step of calibrating the plurality of sets of subcarrier phase data comprises:
and performing linear transformation processing on the plurality of groups of subcarrier phase data to obtain subcarrier phase data which correspond to the plurality of positions one by one after calibration.
Another aspect of the present invention provides a positioning system based on channel state information, the system comprising:
the training set construction module is used for pre-dividing a target position into a plurality of positions, acquiring a plurality of groups of subcarrier phase data which correspond to the plurality of positions one by one, calibrating the plurality of groups of subcarrier phase data, and constructing a training set according to the calibrated subcarrier phase data;
the model training module is used for constructing a deep neural network model, inputting a training set into the deep neural network model for training to obtain an optimal deep neural network model and fingerprint data corresponding to a target position;
and the position calculation module is used for establishing a fingerprint database according to the fingerprint data, acquiring the subcarrier phase data of the moving target, matching the subcarrier phase data of the moving target with the fingerprint database, and determining the position of the moving target according to the matching result.
Another aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the channel state information-based positioning method according to any one of the above.
Another aspect of the present invention also provides a data processing apparatus, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any one of the above positioning methods based on channel state information.
Drawings
Fig. 1 is a flowchart of a positioning method based on channel state information according to a first embodiment of the present invention;
FIG. 2 is a flow chart of a positioning method based on channel state information according to a second embodiment of the present invention;
fig. 3 is a system block diagram of a positioning system based on channel state information according to a fourth embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating a process of creating a fingerprint database after training through a deep neural network according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of layer-to-layer relationships when a contrast divergence algorithm is used for training a model in an embodiment of the present invention;
FIG. 6 is a schematic flow chart of model adjustment using an error back propagation algorithm according to an embodiment of the present invention;
FIG. 7 shows an embodiment of the present invention25m×20mThe schematic flow chart of detecting a moving target in the target environment of (1);
fig. 8 is a schematic view of an application scenario of the positioning method based on channel state information according to the embodiment of the present invention;
the following detailed description will further illustrate the invention in conjunction with the above-described figures.
Detailed Description
To facilitate an understanding of the invention, the invention will now be described more fully with reference to the accompanying drawings. Several embodiments of the invention are shown in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
The present invention is mainly applied to the situation of environment detection based on wireless signals, and as shown in fig. 8, an application scenario diagram of the present invention is shown, which includes a target environment, a network device and a positioning device. The method comprises the steps that a plurality of different positions are pre-marked in a target environment, network signals sent by network equipment cover the whole target environment, the network equipment receives subcarrier signals of different positions in the target environment and sends the subcarrier signals to a positioning device, and the positioning device carries out matching positioning according to the subcarrier signals.
The invention firstly decomposes the intelligent environment detection problem into two stages of training and on-line real-time calculation, utilizes the deep neural network with three hidden layers to correct the phase data of the sub-carrier wave in the off-line training stage, and firstly decomposes the intelligent environment detection problem into two stages of training and on-line real-time calculation, utilizes the deep neural network with three hidden layers to correct the phase data of the sub-carrier wave in the off-line training stage
The method comprises the steps of carrying out intelligent processing, reducing influences caused by factors such as environment and the like in a low signal-to-noise ratio environment, then establishing a CSI fingerprint base under the environment according to an output result obtained by deep neural network processing, carrying out fingerprint matching on a newly received and processed subcarrier signal and different fingerprint data in the CSI fingerprint base in an online real-time calculation stage when a moving target moves in the environment in real time, accumulating and comparing for many times, and finally outputting a fingerprint matching result, thereby determining the position information of the moving target in a detection environment, reflecting the position change of the moving target in the fingerprint network and realizing the motion state detection of the moving target under the environment.
Example one
Referring to fig. 1, a positioning method based on channel state information according to a first embodiment of the present invention is shown, including steps S11-S13.
And S11, pre-dividing the target position into a plurality of positions, acquiring a plurality of groups of subcarrier phase data corresponding to the positions one by one, calibrating the plurality of groups of subcarrier phase data, and constructing a training set according to the calibrated subcarrier phase data.
Constructing a training set: the training set being collected and calibrated by linear transformationnGroup ofiSubcarrier phase data from calibratednGroup ofiInput matrix establishment for subcarrier phase datah 0 Input ofh 0 Is thatn×iA matrix of dimensions is formed by a matrix of dimensions,nindicating the total number of set positions where fingerprint information is created,iindicating input at a certain positioniThe output result of the calibrated subcarrier phase data is the samen×iFingerprint output matrix of dimension, represented innOutput in different positionsiInputting the fingerprint information data obtained after model training into matrixh 0 Each row of (a) corresponds to each row of the fingerprint output matrix.
S12, constructing a deep neural network model, inputting the training set into the deep neural network model for training, and obtaining the optimal deep neural network model and the fingerprint data of the corresponding target position.
Constructing a deep neural network: the deep neural network comprises an input layer, three hidden layers and an output layer.h 0 Representing an input matrix, i.e.nGroup ofiThe phase data of each of the calibrated sub-carriers,W 1 、W 2 、W 3 and representing network model parameters between the input layer and the first hidden layer, between the first hidden layer and the second hidden layer, between the second hidden layer and the third hidden layer.h 0 And the data is sent into the deep neural network model through the input layer, and the results are output by the output layer after the training between layers.
Wherein, when the collected subcarrier data is input into the deep neural network for training,ngroup ofiAnd inputting the subcarrier phase data serving as an input matrix into an input layer in the neural network, further sequentially training in the three hidden layers, and obtaining an output matrix established by the trained subcarrier phase data in an output layer.
In the training process, in order to better acquire the optimal network model parameters and train the model to be more optimal, a model loss function is constructed, wherein the model loss function is as follows:
wherein,h k-1 、h k are respectively shown atK-1Layer and the firstKThe output of the layer(s) is,h 0 which represents the input data, is,h 1 、h 2 、h 3 respectively representing input datah 0 At the first layer passing through the model hidden layerK 1 Second layer ofK 2 Third layerK 3 Output result of (2)，b k-1 、b k Respectively expressed in the modelK-1Layer and the firstKThe deviation of the layer or layers is,b 0 、b 1 、b 2 、b 3 respectively expressed in the model input layer and the first layer of the hidden layerK 1 Second layerK 2 Third layerK 3 The deviation of (2). Network model parameters for making output value of loss function reach preset valueW 1 、W 2 、W 3 I.e. the optimal model parameters obtained by the model training.
Input matrixh 0 After the deep neural network training with the optimal model parameters, the output is outputn×iOutput matrices of dimensions, i.e. correspondencesnFingerprint data at different positions
。
S13, establishing a fingerprint database according to the fingerprint data, acquiring the subcarrier phase data of the moving target, matching the subcarrier phase data of the moving target with the fingerprint database, and determining the position of the moving target according to the matching result.
As shown in fig. 4, the fingerprint data obtained after model training is used
Errors due to environmental factors and the like have been eliminated, and thus fingerprint data is obtained
Forming subcarrier phase data corresponding to different locations in an environment
Is/are as followsCSIAnd (4) fingerprint database. Through the fingerprint database established after deep training, the fingerprint data in the fingerprint database is subjected to deep neural network training, so that the influence of the phase data of the subcarriers on factors such as environment in a low signal-to-noise ratio environment is reduced, the accuracy and effectiveness of the data are improved, and the accuracy and reliability of subsequent phase calibration and fingerprint matching can be improved.
In the fingerprint matching process, because the channel is in the same transceiving end and common channel environment, the number of the subcarrier phase information carried by the CSI isXAnd its inter-subcarrier frequency spacing
Are determined and the same, under the condition, the time interval data caused by propagation environment error, time delay error and the like in the calibrated subcarrier phase data
Becomes a key factor for detecting moving targets, and the time interval data generated under different target environments is known
Is linear, and thus
The position of the moving target is reflected to a certain extent.
Therefore, in the actual fingerprint matching operation, only the calibrated subcarrier phase data needs to be compared
And fingerprint information at each location in the fingerprint library
The fingerprint matching operation can be completed at the time interval.
It can be known that, after subcarrier information carried by CSI received from a moving target at a certain position is calibrated, there is calibrated subcarrier phase data
：
And the fingerprint data at each position in the established fingerprint library:
the two are compared and matched:
when the above formula is observed, it can be found from the obtained results that
The closer the two are, i.e. moving the calibrated subcarrier phase data of the target
And fingerprint data at a certain position in the fingerprint database
The closer the distance is to each other,
the closer to 1 the value of (b) is, and vice versa, the closer to 0.
Obtaining the sub-carrier phase data of the moving target, carrying out linear transformation calibration, and receiving and processing the obtained sub-carrier phase data after the moving target calibration
By means of fingerprint data at different positions from those in the fingerprint library
By performing comparison and matching, different calculation results can be obtained, and the calculation results can reflect
And
to eliminate the interference of the white environmental noise to the CSI receiving and matching results, the matching degree of (1) is to be determined
And with
Multiple fingerprint matching is carried out, and multiple matching results are accumulated and
the represented position information corresponds to each other one by one, and the fingerprint matching result under the position information can be output:
in the above formula:Pshow to proceedPThe secondary fingerprint is matched with the primary fingerprint,Nrepresenting the interference of white noise on the CSI reception and fingerprint matching.
In the output fingerprint matching result, the phase data of the subcarriers after the CSI of the moving target is calibrated can be obtained
Fingerprint data at different positions from the fingerprint database
And performing fingerprint matching for multiple times, so that the position information of the moving object in the detection environment can be obtained and the position of the moving object in the fingerprint library can be estimated. Meanwhile, time interval data such as propagation environment error, time delay error and the like of down-sampling in different target environments
The influence of (2) is linear, i.e. the target position is paired
The influence of the method has a certain linear rule, so that the position change of the moving target in the environment can be estimated by comparing and comparing fingerprint matching results output by different groups, thereby determining the moving direction, distance and speed of the moving target and realizing the detection of the moving state of the moving target in the target environment.
In summary, in the positioning method based on channel state information in the above embodiments of the present invention, a deep neural network training model and pre-divided subcarrier phase data at multiple positions are constructed, an optimal deep neural network model is obtained by training the deep neural network, the influence of subcarrier phase data in a low signal-to-noise ratio environment due to factors such as environment can be reduced by deep neural network training, a fingerprint library corresponding to a target position is established according to the calibrated subcarrier phase data after training, the subcarrier data of a moving target is compared with the fingerprint library to obtain position information of the moving target, the accuracy and reliability of the received subcarrier phase data are improved by deep neural network training, the reliability and accuracy of detecting the position of the moving target are improved, and the problem that in a complex environment in the background art is solved, the reliability and accuracy of the position detection of the moving object may be deteriorated.
Example two
Referring to fig. 2, a positioning method based on channel state information according to a second embodiment of the present invention is shown, including steps S21-S24.
And S21, pre-dividing the target position into a plurality of positions, acquiring a plurality of groups of subcarrier phase data corresponding to the positions one by one, calibrating the plurality of groups of subcarrier phase data, and constructing a training set according to the calibrated subcarrier phase data.
Constructing a training set: the training set being collected and calibrated by linear transformationnGroup ofiSubcarrier phase data from calibratednGroup ofiInput matrix establishment for subcarrier phase datah 0 Input ofh 0 Is thatn×iA matrix of dimensions is formed by a matrix of dimensions,nindicating the total number of set positions where fingerprint information is created,iindicating input at a certain positioniThe output result of the calibrated subcarrier phase data is the samen×iFingerprint output matrix of dimension, represented innOutput in different positionsiInputting the fingerprint information data obtained after model training into matrixh 0 Each row of (a) corresponds to each row of the fingerprint output matrix.
Subcarrier phase data calibration includes calibrating for differences in the environmentnAt the position belowCollected
Sub-carrier phase data
Performing linear conversion to eliminate the influence of subcarrier frequency offset and ADC sampling frequency offset, and calibratingnGroup ofiCalibrated subcarrier phase data
The calibration formula is as follows:
establishing an input matrix according to the calibrated n groups of i subcarrier phase datah 0 Inputting ofh 0 Is thatn×iA matrix of dimensions, n representing the total number of set positions where fingerprint information is created,iindicating input at a certain positioniThe output result of the calibrated subcarrier phase data is the samen×iFingerprint output matrix of dimension, represented innOutput in different positionsiInputting the fingerprint information data obtained after model training into matrixh 0 Each row of (a) corresponds to each row of the fingerprint output matrix.
S22, constructing a deep neural network model, inputting the training set into the deep neural network model, and training the model by adopting a contrast divergence algorithm.
Constructing a deep neural network: the deep neural network comprises an input layer, three hidden layers and an output layer.h 0 Representing an input matrix, i.e.nGroup ofiThe phase data of each of the calibrated sub-carriers,W 1 、W 2 、W 3 representing network modes between the input layer and the first hidden layer, the first hidden layer and the second hidden layer, the second hidden layer and the third hidden layerAnd (4) a type parameter.h 0 And the data is sent into the deep neural network model through the input layer, and the results are output by the output layer after the training between layers.
Wherein, when the collected subcarrier data is input into the deep neural network for training,ngroup ofiAnd inputting the subcarrier phase data serving as an input matrix into an input layer in the neural network, further sequentially training in the three hidden layers, and obtaining an output matrix established by the trained subcarrier phase data in an output layer.
wherein,h k-1 、h k are respectively shown atK-1Layer and the firstKThe output of the layer(s) is,h 0 which represents the input data, is,h 1 、h 2 、h 3 respectively representing input datah 0 At the first layer passing through the model hidden layerK 1 Second layerK 2 Third layerK 3 Output result of (2)，b k-1 、b k Respectively expressed in the modelK-1Layer and the firstKThe deviation of the layer or layers is,b 0 、b 1 、b 2 、b 3 respectively expressed in the model input layer and the first layer of the hidden layerK 1 Second layerK 2 Third layer ofK 3 The deviation of (2). Network model parameters for making output value of loss function reach preset valueW 1 、W 2 、W 3 I.e. the optimal model parameters obtained by the model training.
In the prior art, random gradient descent method is usually adopted to update the netThe model is trained by the parameters of the complex model, but the random gradient descent method causes the problems of overhigh model training complexity, too slow model training, overlow training efficiency and the like, so that the contrast divergence algorithm is used in the first model in the implementationK-1Layer and the firstKTraining between layers, as shown in FIG. 5K-1The input of the layer is transmitted to the first layer through network trainingKIn the layer, the states of all neurons are independent, and the training result is only the neuron transmission between layers, so K can be used-1Input of layersh k-1 After model network model parameter training, the model is compared with the modelKLayer of each neuron
Multiplying the joint probability distribution of the output states toKThe final output of the layer is approximated:
wherein,
is the firstK-1The input of the layer passes throughK-1Layer and the firstKTraining and bringing in of network of layerssigmoidThe approximate result of the function is that of a function,Sigmoidthe function is a sigmoid function, also known as a sigmoidal growth curve, which is common in biology. In the information science, due to the properties of single increment, single increment of an inverse function and the like,Sigmoidthe function is often used as a threshold function for neural networks, mapping variables between 0 and 1, bySigmoidThe formula for fitting the function to the training process is as follows:
and S23, adjusting the estimated network model parameters after the warping training according to an error back propagation algorithm to obtain the optimal deep neural network model and the fingerprint data corresponding to the target position.
As shown in fig. 6, inputh 0 After passing through the model input layer and the first layerK 1 Then, based on the initial network model parameters obtained by fittingb 0 、b 1 、W 1 To calculate the input layerK 0 And a first layerK 1 In (1)
Can obtainK 1 Probability distribution of all neurons in the layer is obtained after multiplication
Thereby obtaining a first layer of the hidden layerK 1 Output value of (2)h 1 The first layer of the hidden layer can be finally obtained by calculating in each layer according to the method in turn according to the formulaK 1 Second layerK 2 Third layerK 2 Output value ofh 1 、h 2 、h 3 While distributively multiplying network model parameters between corresponding layers and substituting into sigmoid function for approximate inference
And
can be calculated reversely to obtainh 2 ，h 1 ，h 0 。
Output from each training
And the input of the starth 0 The error between the different layers of the network is adjusted by using an error back propagation algorithm, the training is repeated in the way, and when the error is trained to a preset value, the model parameter can be finally obtainedW 1 、W 2 、W 3 The optimum value of (c).
Wherein ∂ is a learning rate preset before model training.
Input matrixh 0 After the deep neural network training with the optimal model parameters, the output is outputn×iOutput matrices of dimensions, i.e. correspondencesnFingerprint data at different positions
。
S24, establishing a fingerprint database according to the fingerprint data, obtaining the subcarrier phase data of the moving target, matching the subcarrier phase data of the moving target with the fingerprint database, and determining the position of the moving target according to the matching result.
As shown in fig. 4, the fingerprint data obtained after model training is used
The errors caused by the influence of various factors in the low signal-to-noise ratio environment are eliminated, and the fingerprint data are obtained according to the errors
Forming subcarrier phase data corresponding to different locations in an environment
The CSI fingerprint library of (1). Through the fingerprint database established after deep training, the fingerprint data in the fingerprint database has higher accuracy, and the accuracy and the reliability of subsequent phase calibration and fingerprint matching can be improved.
In the fingerprint matching process, because the channel is in the same transceiving end and common channel environment, the number of the subcarrier phase information carried by the CSI isXAnd its inter-subcarrier frequency spacing
Are determined and the same, under the condition, the time interval data caused by propagation environment error, time delay error and the like in the calibrated subcarrier phase data
Becomes a key factor for detecting a moving target, and known time interval data such as propagation environment errors, delay errors and the like under different target environments
Is linear, and thus
The position of the moving target is reflected to a certain extent.
Therefore, in the actual fingerprint matching operation, only the calibrated subcarrier phase data needs to be compared
And fingerprint information at each location in the fingerprint library
The fingerprint matching operation can be completed at the time interval.
It can be known that, after subcarrier information carried by CSI received from a moving target at a certain position is calibrated, there is calibrated subcarrier phase data
：
With fingerprint data at each location in the established fingerprint library
：
The two are compared and matched:
when the above formula is observed, it can be found from the obtained results that
The closer the two are, i.e. moving the calibrated subcarrier phase data of the target
And fingerprint data at a certain position in the fingerprint database
The closer the distance is to each other,
the closer to 1 the value of (b) is, and vice versa, the closer to 0.
Obtaining the sub-carrier phase data of the moving target, carrying out linear transformation calibration, and receiving and processing the obtained sub-carrier phase data after the moving target calibration
By data of fingerprints at different locations from the fingerprint database
By performing comparison and matching, different calculation results can be obtained, and the calculation results can reflect
And
to eliminate the interference of the white environmental noise to the CSI receiving and matching results, the matching degree of (1) is to be determined
And
multiple fingerprint matching is carried out, and multiple matching results are accumulated and
the represented position information corresponds to each other one by one, and the fingerprint matching result under the position information can be output:
in the above formula:Pshow to proceedPThe secondary fingerprint is matched with the fingerprint,Nrepresenting the interference of white noise on the CSI reception and fingerprint matching.
In the output fingerprint matching result, the phase data of the subcarriers after the CSI of the moving target is calibrated can be obtained
Fingerprint data at different positions from the fingerprint database
And performing fingerprint matching for multiple times, so that the position information of the moving object in the detection environment can be obtained and the position of the moving object in the fingerprint library can be estimated. Meanwhile, time interval data such as propagation environment error, time delay error and the like of down-sampling in different target environments
The influence of (2) is linear, i.e. the target position is paired
The influence of the method has a certain linear rule, so that the position change of the moving target in the environment can be estimated by comparing and comparing fingerprint matching results output by different groups, thereby determining the moving direction, distance and speed of the moving target and realizing the running state detection of the moving target in the target environment.
In summary, in the positioning method based on channel state information in the above embodiments of the present invention, a deep neural network training model and pre-divided subcarrier phase data at multiple positions are constructed, an optimal deep neural network model is obtained by training the deep neural network, the influence of subcarrier phase data in a low signal-to-noise ratio environment due to factors such as environment can be reduced by deep neural network training, a fingerprint library corresponding to a target position is established according to the calibrated subcarrier phase data after training, the subcarrier phase data of a moving target is compared with the fingerprint library to obtain the position information of the moving target, the accuracy and reliability of the received subcarrier phase data are improved by deep neural network training, the reliability and accuracy of detecting the position of the moving target are improved, and the problem that in a complex environment in the background art is solved, the reliability and accuracy of the position detection of the moving object may be deteriorated.
EXAMPLE III
As shown in FIG. 7, the embodiment further provides a method pair according to the above steps S21-S2425m×20mA moving object detection method in an object environment.
S31: in that25m×20mIn the target environment of (2), n =2000 different training positions are determined0.5m×0.5mAnd collecting at each position by means of the apparatusi=90Sub-carrier phase data
After linear processing, the calibrated at each position can be obtainedi=90Calibrated subcarrier phase data
N =2000 groupsi=90Calibrated subcarrier phase data
：
S32: it is understood that each location can be grouped into a groupi=90Calibrated subcarrier phase data
Establishing fingerprint library, and constructing input with training set of 2000 × 90 dimensionsh 0 Each row corresponds to a different positioni=90Calibrated subcarrier phase data
After the constructed deep neural network training, the output is outputn=2000The fingerprint output matrix of dimension, each row corresponds to the fingerprint data under different positions, and inputs simultaneouslyh 0 Each row of (a) corresponds to each row of the fingerprint output matrix.
S33: by use of an input layerk 0With 90 input nodes in a first hidden layerK 1 With 60 nodes in a second hidden layerK 2 With 30 nodes in a third hidden layerK 3 The deep neural network with 15 nodes and 6 output nodes on an output layer correspondingly inputs the data of each line of input data into each input node, the input data correspondingly outputs different data when passing through different layers of the model, and each time complete training is carried out, the data can be input according to the input datah 0 And output
Using an error back-propagation algorithm to adjust the layer andnetwork model parameters between layers, setting learning rate before training
After repeated training, the error reaches the minimum or the final network model parameter can be obtained within the precision range, 6 output nodes of the output layer can output the fingerprint data corresponding to each training position, and when the training is repeated, the fingerprint data corresponding to each training position can be outputn=2000After the fingerprint data under each training position are generated, the fingerprint data are output2000×90And outputting a fingerprint output matrix of the dimension, and forming a fingerprint library under the environment according to the fingerprint output matrix.
S34: when the moving target moves in the established fingerprint database, the receiving device can acquire the subcarrier information carried by the CSI of the receiving device
And the fingerprint information is combined with the fingerprint information at each position in the fingerprint database
And (3) carrying out comparison and matching:
to eliminate the environmental white noiseNThe influence on the CSI receiving and matching results is that a group of subcarrier information carried by the mobile object CSI obtained by the receiving equipment in the same period of time
Can be matched with fingerprint information on various positions of fingerprint library
Performing multiple comparison matching, will
With each position in the fingerprint library
P =20 fingerprint matches were made and the comparison match results with the same location were accumulated:
finally outputting fingerprint matching results after 20 times of accumulation at 2000 different positions
S35: every Δ T =1s, the receiving device may obtain subcarrier information carried by a set of CSI of the moving target
The fingerprint information is compared with the fingerprint information at different positions in the fingerprint database
The comparison and the matching are carried out,
over a period of timeT(T＞1)After comparison of multiple groups of matching results, if the fingerprint matching result is not obviously changed, the moving target is static in the time, namely, the moving target does not move in the target environment; if the fingerprint matching result changes between two adjacent groups of matching, the motion direction and the track of the moving target in the time can be determined according to the position change in the fingerprint library, and then the motion speed of the moving target is calculated, so that the motion detection of the moving target in the environment is realized.
In summary, in the positioning method based on channel state information in the above embodiments of the present invention, a deep neural network training model and pre-divided subcarrier phase data at multiple positions are constructed, an optimal deep neural network model is obtained by training the deep neural network, the influence of subcarrier phase data in a low signal-to-noise ratio environment due to factors such as environment can be reduced by deep neural network training, a fingerprint library corresponding to a target position is established according to the calibrated subcarrier phase data after training, the subcarrier data of a moving target is compared with the fingerprint library to obtain position information of the moving target, the accuracy and reliability of the received subcarrier phase data are improved by deep neural network training, the reliability and accuracy of detecting the position of the moving target are improved, and the problem that in a complex environment in the background art is solved, the reliability and accuracy of the position detection of the moving object may be deteriorated.
Example four
In another aspect, the present invention further provides a positioning system based on channel state information, referring to fig. 3, which illustrates a positioning system based on channel state information in a third embodiment of the present invention, including:
the training set construction module is used for pre-dividing a target position into a plurality of positions, acquiring a plurality of groups of subcarrier phase data which correspond to the plurality of positions one by one, calibrating the plurality of groups of subcarrier phase data, and constructing a training set according to the calibrated subcarrier phase data;
the model training module is used for constructing a deep neural network model, inputting the training set into the deep neural network model for training to obtain an optimal deep neural network model and fingerprint data corresponding to a target position;
and the position calculation module is used for establishing a fingerprint database according to the fingerprint data, acquiring subcarrier phase data of a moving target, matching the subcarrier phase data of the moving target with the fingerprint database, and determining the position of the moving target according to a matching result.
Further, in some other optional embodiments, the model building module comprises:
the deep neural network unit comprises an input layer, a plurality of hidden layers and an output layer, wherein the input layer is used for inputting an input matrix established by the plurality of groups of subcarrier phase data, the output layer is used for outputting an output matrix obtained after the plurality of groups of subcarrier phase data are trained by the plurality of hidden layers, and the output matrix comprises an optimal deep neural network model and fingerprint data corresponding to a target position.
Further, in some other optional embodiments, the model training module comprises:
and the loss function construction unit is used for training the training set based on the deep neural network model of the model loss function, and outputting the optimal model parameters of the deep neural network model and the output fingerprint data corresponding to the target position when the output value of the model loss function reaches the network model parameters of a preset value, namely the optimal model parameters of the deep neural network model.
Further, in some other optional embodiments, the loss function constructing unit includes:
and the contrast divergence algorithm subunit is used for inputting the plurality of groups of subcarrier phase data into the hidden layers after passing through the input layer, estimating the training result of each hidden layer of the subcarrier phase data according to the contrast divergence algorithm, and estimating the network model parameters between different hidden layers according to the training result of each hidden layer.
Further, in some other optional embodiments, the hidden layers include a first hidden layer, a second hidden layer, and a third hidden layer; the contrast divergence algorithm subunit includes:
a joint probability distribution multiplication subunit, configured to perform joint probability distribution multiplication on the input of the input layer and the output states of the neurons of the first hidden layer to estimate an output of the first hidden layer;
then, the output of the first hidden layer is multiplied by the output state of each neuron of the second hidden layer through joint probability distribution so as to estimate the output of the second hidden layer;
performing joint probability distribution multiplication on the output of the second hidden layer and the output state of each neuron of the third hidden layer to estimate the output of the third hidden layer;
and reversely calculating the input of each hidden layer according to the output of each hidden layer, and further obtaining network model parameters among different hidden layers according to the input of each hidden layer.
Further, in some other optional embodiments, the loss function constructing unit further includes:
and the error back propagation algorithm subunit is used for continuously adjusting network model parameters between different hidden layers by comparing errors between the input and the output of each hidden layer and utilizing an error back propagation algorithm to finally obtain the optimal network model parameters.
Further, in some other optional embodiments, the training set constructing module includes:
and the linear transformation unit is used for performing linear transformation processing on the multiple groups of subcarrier phase data to obtain subcarrier phase data which correspond to the multiple positions one by one after calibration.
The functions or operation steps implemented by the modules and units when executed are substantially the same as those of the method embodiments, and are not described herein again.
In summary, in the positioning system based on channel state information in the above embodiments of the present invention, a deep neural network training model and pre-divided subcarrier phase data at multiple positions are constructed, an optimal deep neural network model is obtained by training the deep neural network, the influence of subcarrier phase data in a low signal-to-noise ratio environment due to factors such as environment can be reduced by deep neural network training, a fingerprint library corresponding to a target position is established according to the calibrated subcarrier phase data after training, the subcarrier data of a moving target is compared with the fingerprint library to obtain position information of the moving target, the accuracy and reliability of the received subcarrier phase data are improved by deep neural network training, the reliability and accuracy of detecting the position of the moving target are improved, and the problem that in a complex environment in the background art is solved, the reliability and accuracy of the position detection of the moving object may be deteriorated.
An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the positioning method based on channel state information in the foregoing embodiments.
EXAMPLE five
In another aspect, the present invention further provides an apparatus, where the system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the positioning method based on channel state information in the foregoing embodiments. In some embodiments, the processor may be an Electronic Control Unit (ECU), a Central Processing Unit (CPU), a controller, a microcontroller, a microprocessor or other data processing chip, and is configured to run program codes stored in the memory or process data, such as executing an access restriction program.
Wherein the memory includes at least one type of readable storage medium including flash memory, hard disks, multimedia cards, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disks, optical disks, etc. The memory may in some embodiments be an internal storage unit of the device, for example a hard disk of the device. The memory may also be an external storage device of the device in other embodiments, such as a plug-in hard disk provided on the device, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash card (FlashCard), and the like. Further, the memory may also include both internal storage units of the device and external storage. The memory can be used not only to store application software and various types of data installed in the device, but also to temporarily store data that has been output or is to be output.
In summary, the device in the above embodiments of the present invention, by constructing a deep neural network training model and pre-partitioned subcarrier phase data at multiple positions, an optimal deep neural network model is obtained by training a deep neural network, the influence of subcarrier phase data caused by factors such as environment and the like in a low signal-to-noise ratio environment can be reduced by training the deep neural network, a fingerprint library corresponding to a target position is established according to the subcarrier phase data calibrated after training, the subcarrier data of a moving target is compared with the fingerprint library to obtain the position information of the moving target, the accuracy and the reliability of the received subcarrier phase data are improved through deep neural network training, therefore, the reliability and the accuracy of the position detection of the moving target are improved, and the problem that the reliability and the accuracy of the position detection of the moving target are poor in the complex environment in the background technology is solved.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is specific and detailed, but not to be understood as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A positioning method based on channel state information, the method comprising:
pre-dividing a target position into a plurality of positions, acquiring a plurality of groups of subcarrier phase data which correspond to the plurality of positions one by one, calibrating the plurality of groups of subcarrier phase data, and constructing a training set according to the calibrated subcarrier phase data;
constructing a deep neural network model, inputting the training set into the deep neural network model for training to obtain an optimal deep neural network model and fingerprint data corresponding to a target position;
and establishing a fingerprint database according to the fingerprint data, acquiring subcarrier phase data of a moving target, matching the subcarrier phase data of the moving target with the fingerprint database, and determining the position of the moving target according to a matching result.
2. The channel state information-based positioning method according to claim 1, wherein the deep neural network model includes an input layer, a plurality of hidden layers, and an output layer, the input layer is configured to input an input matrix created by the plurality of sets of subcarrier phase data, the output layer is configured to output an output matrix obtained by training the plurality of sets of subcarrier phase data by the plurality of hidden layers, and the output matrix includes an optimal deep neural network model and fingerprint data corresponding to a target position.
3. The channel state information-based positioning method according to claim 2, wherein the step of constructing a deep neural network model, inputting the training set into the deep neural network model for training to obtain an optimal deep neural network model and fingerprint data corresponding to the target position comprises:
and constructing a model loss function, training the training set based on the deep neural network model of the model loss function, and outputting the optimal model parameters of the deep neural network model and the output fingerprint data corresponding to the target position when the output value of the model loss function reaches the network model parameters of a preset value, namely the optimal model parameters of the deep neural network model.
4. The channel state information-based positioning method according to claim 3, wherein the step of training the training set based on the deep neural network model of the model loss function comprises:
inputting the multiple groups of subcarrier phase data into the hidden layers after passing through the input layers, estimating the training result of each hidden layer of the subcarrier phase data according to a contrast divergence algorithm, and estimating network model parameters between different hidden layers according to the training result of each hidden layer.
5. The channel state information-based positioning method according to claim 4, wherein the hidden layers comprise a first hidden layer, a second hidden layer and a third hidden layer; the step of estimating the training result of each hidden layer of the subcarrier phase data according to a contrast divergence algorithm and estimating the network model parameters between different hidden layers according to the training result of each hidden layer comprises the following steps:
performing joint probability distribution multiplication on the input of the input layer and the output states of the neurons of the first hidden layer to estimate the output of the first hidden layer;
then, the output of the first hidden layer is multiplied by the output state of each neuron of the second hidden layer through joint probability distribution so as to estimate the output of the second hidden layer;
performing joint probability distribution multiplication on the output of the second hidden layer and the output state of each neuron of the third hidden layer to estimate the output of the third hidden layer;
and reversely calculating the input of each hidden layer according to the output of each hidden layer, and further obtaining network model parameters among different hidden layers according to the input of each hidden layer.
6. The channel state information-based positioning method according to claim 5, wherein the step of estimating the training result of each hidden layer of the subcarrier phase data according to the contrast divergence algorithm further comprises:
and finally, obtaining the optimal network model parameters by comparing the errors between the input and the output of each hidden layer and continuously adjusting the network model parameters between different hidden layers by using an error back propagation algorithm.
7. The channel state information-based positioning method according to claim 1, wherein the step of calibrating the plurality of sets of subcarrier phase data comprises:
and performing linear transformation processing on the multiple groups of subcarrier phase data to obtain subcarrier phase data which correspond to the multiple positions one by one after calibration.
8. A positioning system based on channel state information, the system comprising:
the training set construction module is used for pre-dividing a target position into a plurality of positions, acquiring a plurality of groups of subcarrier phase data which correspond to the plurality of positions one by one, calibrating the plurality of groups of subcarrier phase data, and constructing a training set according to the calibrated subcarrier phase data;
the model training module is used for constructing a deep neural network model, inputting the training set into the deep neural network model for training to obtain an optimal deep neural network model and fingerprint data corresponding to a target position;
and the position calculation module is used for establishing a fingerprint database according to the fingerprint data, acquiring subcarrier phase data of a moving target, matching the subcarrier phase data of the moving target with the fingerprint database, and determining the position of the moving target according to a matching result.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a channel state information based positioning method according to any one of claims 1 to 7.
10. A data processing device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the positioning method based on channel state information according to any one of claims 1 to 7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210432885.7A CN114531729B (en) | 2022-04-24 | 2022-04-24 | Positioning method, system, storage medium and device based on channel state information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210432885.7A CN114531729B (en) | 2022-04-24 | 2022-04-24 | Positioning method, system, storage medium and device based on channel state information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114531729A true CN114531729A (en) | 2022-05-24 |
| CN114531729B CN114531729B (en) | 2022-08-09 |
Family
ID=81628121
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210432885.7A Active CN114531729B (en) | 2022-04-24 | 2022-04-24 | Positioning method, system, storage medium and device based on channel state information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114531729B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116343966A (en) * | 2023-03-27 | 2023-06-27 | 山东大学 | Probability multiplication accumulation structural damage imaging positioning method and system based on delay factors |
| CN116527462A (en) * | 2023-05-25 | 2023-08-01 | 兰州交通大学 | Wireless positioning method based on channel state change CSI value |
| CN117062001A (en) * | 2023-08-01 | 2023-11-14 | 哈尔滨工程大学 | 5G NR indoor positioning method and system based on interpretable deep learning |
| WO2024138739A1 (en) * | 2022-12-30 | 2024-07-04 | Oppo广东移动通信有限公司 | Method and apparatus for acquiring label information, and device and storage medium |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105929419A (en) * | 2016-04-13 | 2016-09-07 | 南京理工大学 | BP artificial neural network-based GPS carrier tracking method |
| CN106772219A (en) * | 2017-03-08 | 2017-05-31 | 南京大学 | Indoor orientation method based on CSI signals |
| WO2017100706A1 (en) * | 2015-12-09 | 2017-06-15 | Origin Wireless, Inc. | Method, apparatus, and systems for wireless event detection and monitoring |
| CN109151731A (en) * | 2018-06-14 | 2019-01-04 | 西北师范大学 | A kind of indoor orientation method based on CSI |
| US20190049551A1 (en) * | 2017-08-09 | 2019-02-14 | SWFL, Inc., d/b/a "Filament" | Systems and methods for coherence based positioning |
| CN110222776A (en) * | 2019-06-10 | 2019-09-10 | 安徽磐众信息科技有限公司 | A kind of indoor Passive Location based on CSI finger print information |
| CN111556431A (en) * | 2020-04-21 | 2020-08-18 | 西安交通大学 | Fingerprint indoor positioning method based on signal subspace |
| CN111770436A (en) * | 2020-05-07 | 2020-10-13 | 普联国际有限公司 | Indoor Wi-Fi positioning method and device based on CSI and storage medium |
| CN111866713A (en) * | 2020-07-09 | 2020-10-30 | 北京邮电大学 | Positioning method and device and electronic equipment |
| CN112040400A (en) * | 2020-08-25 | 2020-12-04 | 西安交通大学 | Single-site indoor fingerprint positioning method based on MIMO-CSI, storage medium and equipment |
| CN112147573A (en) * | 2020-09-14 | 2020-12-29 | 山东科技大学 | Passive positioning method based on amplitude and phase information of CSI (channel State information) |
| CN112954791A (en) * | 2021-01-26 | 2021-06-11 | 电子科技大学 | Channel State Information (CSI) positioning method based on subcarrier screening |
| CN113852908A (en) * | 2021-07-29 | 2021-12-28 | 上海大学 | Indoor positioning method based on antenna extension |
-
2022
- 2022-04-24 CN CN202210432885.7A patent/CN114531729B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017100706A1 (en) * | 2015-12-09 | 2017-06-15 | Origin Wireless, Inc. | Method, apparatus, and systems for wireless event detection and monitoring |
| CN105929419A (en) * | 2016-04-13 | 2016-09-07 | 南京理工大学 | BP artificial neural network-based GPS carrier tracking method |
| CN106772219A (en) * | 2017-03-08 | 2017-05-31 | 南京大学 | Indoor orientation method based on CSI signals |
| US20190049551A1 (en) * | 2017-08-09 | 2019-02-14 | SWFL, Inc., d/b/a "Filament" | Systems and methods for coherence based positioning |
| CN109151731A (en) * | 2018-06-14 | 2019-01-04 | 西北师范大学 | A kind of indoor orientation method based on CSI |
| CN110222776A (en) * | 2019-06-10 | 2019-09-10 | 安徽磐众信息科技有限公司 | A kind of indoor Passive Location based on CSI finger print information |
| CN111556431A (en) * | 2020-04-21 | 2020-08-18 | 西安交通大学 | Fingerprint indoor positioning method based on signal subspace |
| CN111770436A (en) * | 2020-05-07 | 2020-10-13 | 普联国际有限公司 | Indoor Wi-Fi positioning method and device based on CSI and storage medium |
| CN111866713A (en) * | 2020-07-09 | 2020-10-30 | 北京邮电大学 | Positioning method and device and electronic equipment |
| CN112040400A (en) * | 2020-08-25 | 2020-12-04 | 西安交通大学 | Single-site indoor fingerprint positioning method based on MIMO-CSI, storage medium and equipment |
| CN112147573A (en) * | 2020-09-14 | 2020-12-29 | 山东科技大学 | Passive positioning method based on amplitude and phase information of CSI (channel State information) |
| CN112954791A (en) * | 2021-01-26 | 2021-06-11 | 电子科技大学 | Channel State Information (CSI) positioning method based on subcarrier screening |
| CN113852908A (en) * | 2021-07-29 | 2021-12-28 | 上海大学 | Indoor positioning method based on antenna extension |
Non-Patent Citations (5)
| Title |
|---|
| JINGJING WANG等: "first_page settings", 《SENSORS》 * |
| XIANDI LI等: "DeFe: indoor localization based on channel state information feature using deep learning", 《JOURNAL OF PHYSICS: CONFERENCE SERIES 1303》 * |
| 刘帅等: "一种基于卷积神经网络的CSI指纹室内定位方法", 《工程科学学报》 * |
| 江小平等: "基于信道状态信息幅值-相位的被动式室内指纹定位", 《电子与信息学报》 * |
| 韩卓宜等: "基于CSI指纹序列匹配的LTE室内定位方法", 《无线电工程》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024138739A1 (en) * | 2022-12-30 | 2024-07-04 | Oppo广东移动通信有限公司 | Method and apparatus for acquiring label information, and device and storage medium |
| CN116343966A (en) * | 2023-03-27 | 2023-06-27 | 山东大学 | Probability multiplication accumulation structural damage imaging positioning method and system based on delay factors |
| CN116343966B (en) * | 2023-03-27 | 2023-11-17 | 山东大学 | Probability multiplication accumulation structural damage imaging positioning method and system based on delay factors |
| US12062167B1 (en) | 2023-03-27 | 2024-08-13 | Shandong University | Probability multiply-sum structural damage imaging positioning method and system based on delay factor |
| CN116527462A (en) * | 2023-05-25 | 2023-08-01 | 兰州交通大学 | Wireless positioning method based on channel state change CSI value |
| CN116527462B (en) * | 2023-05-25 | 2024-02-02 | 兰州交通大学 | Wireless positioning method based on channel state change CSI value |
| CN117062001A (en) * | 2023-08-01 | 2023-11-14 | 哈尔滨工程大学 | 5G NR indoor positioning method and system based on interpretable deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114531729B (en) | 2022-08-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114531729B (en) | Positioning method, system, storage medium and device based on channel state information | |
| CN110632572B (en) | Radar radiation source individual identification method and device based on unintentional phase modulation characteristics | |
| CN110222748B (en) | OFDM radar signal identification method based on 1D-CNN multi-domain feature fusion | |
| US9182473B2 (en) | System, method and product for locating vehicle key using neural networks | |
| CN107703418B (en) | Shelf depreciation location error compensation method based on more radial base neural nets | |
| CN114449452A (en) | Indoor positioning algorithm for heterogeneous equipment | |
| CN114201987B (en) | Active interference identification method based on self-adaptive identification network | |
| CN110308432A (en) | A kind of radar self-adaption waveform selection Activity recognition method neural network based | |
| CN114114166A (en) | Radar pulse de-interlacing method based on DTM algorithm | |
| Efimov et al. | Artificial neural network based angle-of-arrival estimator | |
| CN116047427B (en) | Small sample radar active interference identification method | |
| CN114117912A (en) | Sea clutter modeling and inhibiting method under data model dual drive | |
| CN113518307A (en) | Position fingerprint positioning method based on CSI and crowdsourcing migration self-calibration updating | |
| KR102407835B1 (en) | Method and apparatus for classifying pulse radar signal properties based on machine learning | |
| KR102407834B1 (en) | Method and apparatus for property-based classification of long-pulse radar signals | |
| CN108631817B (en) | Method for predicting frequency hopping signal frequency band based on time-frequency analysis and radial neural network | |
| Milanese et al. | Filter design from data: direct vs. two-step approaches | |
| CN114004250A (en) | Method and system for identifying open set of modulation signals of deep neural network | |
| Guven et al. | Classifying LPI radar waveforms with time-frequency transformations using multi-stage CNN system | |
| Pei et al. | Multi-Level Feature Extraction and Autoregressive Prediction Based Wi-Fi Indoor Fingerprint Localization | |
| CN106257302A (en) | A kind of cognitive radar target tracking method under clutter environment | |
| CN117368877A (en) | Radar image clutter suppression and target detection method based on generation countermeasure learning | |
| CN115494466A (en) | Self-calibration method for distributed radar | |
| CN115598714A (en) | Time-space coupling neural network-based ground penetrating radar electromagnetic wave impedance inversion method | |
| CN117911852B (en) | Underwater target distance prediction method based on self-adaption in part of unsupervised field |
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 |