CN111263295A - WLAN indoor positioning method and device - Google Patents

WLAN indoor positioning method and device Download PDF

Info

Publication number
CN111263295A
CN111263295A CN202010043140.2A CN202010043140A CN111263295A CN 111263295 A CN111263295 A CN 111263295A CN 202010043140 A CN202010043140 A CN 202010043140A CN 111263295 A CN111263295 A CN 111263295A
Authority
CN
China
Prior art keywords
information data
target
reference point
signal strength
environmental information
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
Application number
CN202010043140.2A
Other languages
Chinese (zh)
Other versions
CN111263295B (en
Inventor
谭宇亨
谢侃
谢胜利
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong University of Technology
Original Assignee
Guangdong University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangdong University of Technology filed Critical Guangdong University of Technology
Priority to CN202010043140.2A priority Critical patent/CN111263295B/en
Publication of CN111263295A publication Critical patent/CN111263295A/en
Application granted granted Critical
Publication of CN111263295B publication Critical patent/CN111263295B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/023—Services making use of location information using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds
    • 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
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/04—Inference or reasoning models
    • G06N5/048—Fuzzy inferencing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/33—Services specially adapted for particular environments, situations or purposes for indoor environments, e.g. buildings

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biophysics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Automation & Control Theory (AREA)
  • Fuzzy Systems (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)

Abstract

The application discloses a WLAN indoor positioning method and a device, which are used for carrying out principal component analysis on acquired first environmental information data in a target room and selecting the first environmental information data corresponding to the accumulated contribution rate meeting preset conditions as second environmental information data; constructing a target indoor model based on the second environmental information data; arranging APs and setting reference points in the target indoor model, and acquiring a signal strength RSSI value of each reference point; based on the signal strength RSSI value of each reference point, carrying out region division on the environment in the target room according to a fuzzy clustering algorithm to obtain a plurality of sub-regions; the obtained signal strength RSSI value of the test point in the subregion is input into a preset convolutional neural network model corresponding to the subregion, and the positioning result of the test point is output, so that the technical problems of low positioning speed and low positioning precision existing in the existing indoor positioning method for searching and positioning by adopting Euclidean distance are solved.

Description

WLAN indoor positioning method and device
Technical Field
The present application relates to the field of indoor positioning technologies, and in particular, to a WLAN indoor positioning method and apparatus.
Background
With the popularization of wireless networks, location-based services are receiving more and more attention from people. The existing global positioning system estimates the position by measuring the time difference of arrival of a satellite signal through a receiver, and can analyze and obtain position information with higher precision. However, in cities with dense indoor and high buildings, the positioning function is difficult to realize indoors due to the serious multipath and non-line-of-sight interference of satellite signals.
At present, Wireless Local Area Networks (WLANs) are widely distributed in domestic airports, train stations, libraries, government office buildings and shopping malls, and in a WLAN environment, corresponding location information can be obtained by measuring a Signal Strength Indicator (RSSI) from an Access Point (AP). However, the existing indoor positioning method mostly adopts the Euclidean distance to search and position, and has the problems of low speed and low positioning accuracy.
Disclosure of Invention
The application provides a WLAN indoor positioning method and device, which are used for solving the technical problems of low positioning speed and low positioning accuracy in the existing indoor positioning method which adopts Euclidean distance for searching and positioning.
In view of the above, a first aspect of the present application provides a WLAN indoor positioning method, including:
acquiring first environmental information data in a target room, wherein the first environmental information data comprises spatial three-dimensional point cloud data, position coordinates of a target object, an azimuth angle and a pitch angle of the target object;
performing principal component analysis on the first environmental information data, and selecting the first environmental information data corresponding to the accumulated contribution rate meeting preset conditions as second environmental information data;
constructing the environment in the target room based on the second environment information data to obtain a target indoor model;
arranging APs and setting reference points in the target indoor model, and acquiring a signal strength RSSI value of each reference point in the target indoor model;
based on the signal strength RSSI value of each reference point, carrying out region division on the environment in the target room according to a fuzzy clustering algorithm to obtain a plurality of sub-regions;
and inputting the obtained signal strength RSSI value of the test point in the subregion into a preset convolutional neural network model corresponding to the subregion, and outputting a positioning result of the test point.
Preferably, the performing principal component analysis on the first environmental information data, and selecting the first environmental information data corresponding to the cumulative contribution rate meeting the preset condition as the second environmental information data includes:
calculating a correlation coefficient based on the first environment information data to generate a correlation coefficient matrix;
solving a characteristic equation constructed based on the correlation coefficient matrix to obtain a characteristic value;
calculating a contribution rate based on the feature values;
and selecting the first environment information data corresponding to the characteristic value with the minimum quantity corresponding to the accumulated contribution rate of more than 85% as the second environment information data.
Preferably, the arranging an AP and setting a reference point in the target indoor model, and acquiring a signal strength RSSI value of each reference point in the target indoor model further include:
and carrying out coordinate processing on the target indoor model.
Preferably, the arranging an AP and setting a reference point in the target indoor model, and acquiring a signal strength RSSI value of each reference point in the target indoor model includes:
arranging a plurality of APs and setting a plurality of reference points in the target indoor model, wherein the reference points are uniformly distributed according to a preset distance;
and acquiring the RSSI value of each reference point from each AP.
Preferably, the arranging of a plurality of APs and the setting of a plurality of reference points in the target indoor model further comprise:
selecting a target reference point from the plurality of reference points, establishing a three-dimensional coordinate system by taking the target reference point as an origin, and obtaining the position coordinate of each reference point based on the three-dimensional coordinate system.
Preferably, the inputting the obtained signal strength RSSI values of the test points in the sub-region into a preset convolutional neural network model corresponding to the sub-region and outputting the positioning results of the test points further includes:
taking the position coordinates of the reference point of each sub-area and the signal strength RSSI value of the reference point as a training set;
and each training set trains a convolutional neural network model, when the convolutional neural network model reaches a convergence condition, a plurality of trained convolutional neural network models are obtained, and the trained convolutional neural network models are used as the preset convolutional neural network models.
A second aspect of the present application provides a WLAN indoor positioning apparatus, including:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring first environmental information data in a target room, and the first environmental information data comprises spatial three-dimensional point cloud data, position coordinates of a target object, an azimuth angle and a pitch angle of the target object;
the principal component analysis module is used for carrying out principal component analysis on the first environmental information data and selecting the first environmental information data corresponding to the accumulated contribution rate meeting the preset conditions as second environmental information data;
the construction module is used for constructing the environment in the target room based on the second environment information data to obtain a target indoor model;
the second acquisition module is used for arranging APs and setting reference points in the target indoor model and acquiring the signal strength RSSI value of each reference point in the target indoor model;
the dividing module is used for carrying out region division on the environment in the target room according to a fuzzy clustering algorithm based on the signal strength RSSI value of each reference point to obtain a plurality of sub-regions;
and the positioning module is used for inputting the acquired signal strength RSSI value of the test point in the subregion into a preset convolutional neural network model corresponding to the subregion and outputting a positioning result of the test point.
Preferably, the method further comprises the following steps:
and the preprocessing module is used for carrying out coordinate processing on the target indoor model.
Preferably, the second obtaining module includes:
the arrangement submodule is used for arranging a plurality of APs in the target indoor model and setting a plurality of reference points, and the reference points are uniformly distributed according to a preset distance;
and the acquisition sub-module is used for acquiring the signal strength RSSI value of each reference point from each AP.
Preferably, the second obtaining module further comprises:
and the selection submodule is used for selecting a target reference point from the plurality of reference points, establishing a three-dimensional coordinate system by taking the target reference point as an origin, and obtaining the position coordinate of each reference point based on the three-dimensional coordinate system.
According to the technical scheme, the method has the following advantages:
the application provides a WLAN indoor positioning method, which comprises the following steps: acquiring first environmental information data in a target room, wherein the first environmental information data comprises spatial three-dimensional point cloud data, position coordinates of a target object, an azimuth angle and a pitch angle of the target object; performing principal component analysis on the first environment information data, and selecting the first environment information data corresponding to the accumulated contribution rate meeting the preset conditions as second environment information data; constructing an environment in the target room based on the second environment information data to obtain a target indoor model; arranging APs and setting reference points in the target indoor model, and acquiring a signal strength RSSI value of each reference point in the target indoor model; based on the signal strength RSSI value of each reference point, carrying out region division on the environment in the target room according to a fuzzy clustering algorithm to obtain a plurality of sub-regions; and inputting the obtained signal strength RSSI value of the test point in the subregion into a preset convolutional neural network model corresponding to the subregion, and outputting a positioning result of the test point.
According to the WLAN indoor positioning method, the obtained first environmental information data in the target room are subjected to principal component analysis, the first environmental information data corresponding to the accumulated contribution rate meeting the preset conditions are selected as the second environmental information data, the first environmental information data which has a large influence on the construction of the target indoor model are reserved, the first environmental information data which has a small influence on the target indoor model are removed, a certain data dimension reduction effect is achieved, and the positioning speed is improved to a certain extent; the method comprises the steps of constructing a target indoor model based on second environment information data, reasonably arranging APs and setting reference points in the target indoor model, obtaining signal strength RSSI values of the reference points from the APs, carrying out region division on the environment in the target indoor through a clustering algorithm to obtain a plurality of sub-regions, inputting the signal strength RSSI values of test points in the sub-regions into preset convolutional neural network models corresponding to the sub-regions to obtain positioning results of the test points, and accurately positioning small regions on the basis of dividing the sub-regions.
Drawings
Fig. 1 is a schematic flowchart of a WLAN indoor positioning method according to an embodiment of the present application;
fig. 2 is another schematic flow chart of a WLAN indoor positioning method according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a WLAN indoor positioning apparatus according to an embodiment of the present disclosure.
Detailed Description
In order to make the technical solutions of the present application better understood, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
For easy understanding, referring to fig. 1, an embodiment of a WLAN indoor positioning method provided in the present application includes:
step 101, acquiring first environmental information data in a target room.
It should be noted that the first environment information data includes spatial three-dimensional point cloud data, position coordinates of the target object, an azimuth angle of the target object, and a pitch angle. In the embodiment of the application, first environmental information data in a target room can be acquired through a plurality of cameras and a plurality of radars, an environmental image in the target room is acquired through the cameras, target objects such as tables and chairs in the target room can be calibrated through the acquired image, time difference between a transmission pulse and an echo pulse is measured through the radars, and the accurate distance of the target object can be converted due to propagation of electromagnetic waves at the speed of light.
And 102, performing principal component analysis on the first environment information data, and selecting the first environment information data corresponding to the accumulated contribution rate meeting the preset condition as second environment information data.
It should be noted that, the principal component analysis is performed on the first environment information data, the contribution rate corresponding to each first environment information data is calculated, the contribution rate represents the size of the influence of each first environment information data on the modeling of the target indoor environment, the larger the contribution rate is, the larger the influence of the corresponding first environment information data on the modeling of the target indoor environment is, the first environment information data with larger influence is obtained by screening the contribution rate and is used as the second environment information data for constructing the target indoor model, and the influence of redundant data on the modeling is avoided while the data achieves a certain degree of dimensionality reduction.
And 103, constructing the target indoor environment based on the second environment information data to obtain a target indoor model.
It should be noted that the process of constructing the target indoor environment based on the second environment information data to obtain the target indoor model belongs to the prior art, and the specific process of modeling is not described herein again.
And 104, arranging APs and setting reference points in the target indoor model, and acquiring the signal strength RSSI value of each reference point in the target indoor model.
It should be noted that, after the APs and the reference points are arranged in the target indoor model, a signal receiver may be set at each reference point, and the RSSI value of the signal strength sent by each AP is recorded by the signal receiver, so as to obtain the RSSI value of each reference point.
And 105, carrying out region division on the environment in the target room according to a fuzzy clustering algorithm based on the signal strength RSSI value of each reference point to obtain a plurality of sub-regions.
It should be noted that, a part of reference points are marked as known class information of the fuzzy clustering, the target indoor environment is divided into a plurality of sub-regions by the fuzzy clustering algorithm, and the class information of the sub-region to which each reference point belongs is marked, so as to realize the division of the region.
And 106, inputting the acquired signal strength RSSI values of the test points in the subareas into a preset convolutional neural network model corresponding to the subareas, and outputting the positioning results of the test points.
It should be noted that each sub-region corresponds to one preset convolutional neural network model, the preset convolutional neural network model may be a trained BP neural network model or other trained convolutional neural network models, and when a certain test point needs to be located, a signal receiver may be arranged at the test point to record a signal strength RSSI value sent by each AP, and the obtained signal strength RSSI values from different APs are input into the preset convolutional neural network model corresponding to the sub-region where the test point is located, so as to output location information of the test point.
According to the WLAN indoor positioning method in the embodiment of the application, the acquired first environmental information data in the target room are subjected to principal component analysis, the first environmental information data corresponding to the accumulated contribution rate meeting the preset conditions are selected as the second environmental information data, the first environmental information data which has a large influence on the construction of the target indoor model is reserved, the first environmental information data which has a small influence on the target indoor model is removed, a certain data dimension reduction effect is achieved, and the positioning speed is improved to a certain extent; the method comprises the steps of constructing a target indoor model based on second environment information data, reasonably arranging APs and setting reference points in the target indoor model, obtaining signal strength RSSI values of the reference points from the APs, carrying out region division on the environment in the target indoor through a clustering algorithm to obtain a plurality of sub-regions, inputting the signal strength RSSI values of test points in the sub-regions into preset convolutional neural network models corresponding to the sub-regions to obtain positioning results of the test points, and accurately positioning small regions on the basis of dividing the sub-regions.
For easy understanding, referring to fig. 2, another embodiment of a WLAN indoor positioning method provided in the present application includes:
step 201, acquiring first environmental information data in a target room.
It should be noted that the first environment information data includes spatial three-dimensional point cloud data, position coordinates of the target object, an azimuth angle of the target object, and a pitch angle. In the embodiment of the application, first environmental information data in a target room can be acquired through a plurality of cameras and a plurality of radars, an environmental image in the target room is acquired through the cameras, target objects such as tables and chairs in the target room can be calibrated through the acquired image, time difference between a transmission pulse and an echo pulse is measured through the radars, and the accurate distance of the target object can be converted due to propagation of electromagnetic waves at the speed of light.
Step 202, performing principal component analysis on the first environmental information data, and selecting the first environmental information data corresponding to the accumulated contribution rate meeting the preset condition as the second environmental information data.
It should be noted that, the first environment information data acquired by the different cameras and radars are respectively denoted as O1,O2,…Oi,…,Oj,…,OnN is an integer greater than 0, and a correlation coefficient r is calculated based on the first environment information dataijA correlation coefficient matrix R is generated, namely:
Figure BDA0002368452640000071
wherein r isijIs the first environment information data OiAnd first environment information data OjCoefficient of correlation between, rij＝rji，
Figure BDA0002368452640000072
Are each OiAnd OjWhen all correlation coefficients are calculated, the generated correlation coefficient matrix is
Figure BDA0002368452640000073
Solving a characteristic equation lambda I-R0 constructed based on the correlation coefficient matrix to obtain a characteristic value lambdalAnd l is 1,2, …, n, which can be calculated by a tool such as MATLAB.
Calculating the contribution rate based on the characteristic value, wherein the calculation formula of the contribution rate is as follows:
Figure BDA0002368452640000081
selecting the first environment information data corresponding to the least number of characteristic values corresponding to the accumulated contribution rate greater than 85% as the second environment information data, wherein the calculation formula of the accumulated contribution rate is as follows:
Figure BDA0002368452640000082
the contribution rate represents the influence of each first environment information data on the modeling of the target indoor environment, the larger the contribution rate is, the larger the influence of the corresponding first environment information data on the modeling of the target indoor environment is, the first environment information data with larger influence is obtained by screening the contribution rate and is used as the second environment information data for constructing the target indoor model, and the influence of redundant data on the modeling is avoided while the dimensionality reduction effect on the data is achieved to a certain degree.
And 203, constructing the target indoor environment based on the second environment information data to obtain a target indoor model.
It should be noted that the process of constructing the target indoor environment based on the second environment information data to obtain the target indoor model belongs to the prior art, and the specific process of modeling is not described herein again.
And step 204, carrying out coordinate processing on the target indoor model.
It should be noted that performing the coordinate processing on the model in the target room simulates an environment in the target room into a spatial three-dimensional coordinate system, and if an origin is set in the coordinate system, a complete three-dimensional coordinate system capable of obtaining coordinate values is formed, which provides a basis for subsequently obtaining position coordinates of a reference point, and performing the coordinate processing on the model belongs to the prior art, and a specific process of the coordinate processing on the model is not repeated here. The coordinated target indoor model can be stored in the memory.
Step 205, arranging APs and setting reference points in the target indoor model, and acquiring the position coordinates of each reference point and the signal strength RSSI value of each reference point in the target indoor model.
It should be noted that, a plurality of APs are arranged in the target indoor model, when the APs are arranged, it is ensured that any point in the target indoor model is covered by signals sent by two or more APs, when a plurality of reference points are set in the target indoor model, each reference point is uniformly arranged according to a preset distance, and specific values of the preset distance can be selected according to actual situations. Selecting one reference point from all the reference points as a target reference point, establishing a three-dimensional coordinate system by taking the target reference point as an origin, and obtaining the position coordinate of each reference point based on the three-dimensional coordinate system; a signal receiver may be provided at each reference point position, and the RSSI value of each AP is recorded by the signal receiver, so as to obtain the RSSI value of each reference point.
And step 206, based on the signal strength RSSI value of each reference point, performing region division on the environment in the target room according to a fuzzy clustering algorithm to obtain a plurality of sub-regions.
It should be noted that, the original data matrix a is obtained according to the RSSI value of each reference point:
A＝(xij)n×m,j＝1,2,…,m；i＝1,2,…,n；
where n is the number of APs, m is the number of reference points, xijAnd the RSSI value of the signal strength sent by the ith AP and received by the jth reference point.
Establishing a fuzzy similar matrix through the lattice closeness, and enabling:
Figure BDA0002368452640000091
wherein r isijIn order to be a similarity coefficient of the image data,
Figure BDA0002368452640000092
aj、bj、aiand biIs an approximate value of the membership degree of the RSSI value of the signal strength;
the fuzzy similarity matrix is:
R＝(rij)m×m；
calculating a fuzzy equivalent matrix according to the fuzzy similar matrix:
t(R)＝R*＝R2；
is provided with
t(R)＝(rij')m×m；
t(R)λ＝(rij'(λ))m×m；
Figure BDA0002368452640000093
Constructing a fuzzy equivalent matrix R based on the obtained fuzzy similar matrix R*The propagation closure t (R) of R can be obtained by using a flat method, and then a group of lambda epsilon [0,1 ] is taken from large to small]Where λ is 0.998, if rijIf' λ ═ 1, it is considered that the arrangement position of the ith AP and the position of the jth reference point may be regarded as being in one sub-area, thereby achieving the division of the target indoor environment into a plurality of sub-areas.
Step 207, using the position coordinates of the reference point of each sub-area and the signal strength RSSI value of the reference point as a training set.
It should be noted that the position coordinates of the reference point of each sub-region and the RSSI value of the reference point are used as a training set, each sub-region corresponds to one training set, and there are as many training sets as there are sub-regions.
And 208, training a convolutional neural network model by each training set, obtaining a plurality of trained convolutional neural network models when the convolutional neural network models reach the convergence condition, and taking the trained convolutional neural network models as preset convolutional neural network models.
It should be noted that, in the embodiment of the present application, each training set trains one convolutional neural network model, and how many training sets correspond to how many convolutional neural network models, and the number h of hidden layer nodes of the convolutional neural network model in the training process is determined by an empirical formula, that is:
Figure BDA0002368452640000101
wherein o is the number of nodes of the input layer, p is the number of nodes of the output layer, and q is an adjusting constant between 1 and 10.
According to the input vector, the connection weight w of the input layer and the hidden layerijAnd a threshold value ajAnd calculating hidden layer output H:
Figure BDA0002368452640000102
where f (·) is an activation function, and the activation function in this embodiment is f (x) 1/(1+ e)-x)。
According to the hidden layer output H and the connection layer weight wjkAnd a threshold value bkComputing output layer output Ok：
Figure BDA0002368452640000103
Wherein m is the number of output layer nodes.
According to the output value O of the convolutional neural networkkAnd the desired output yk(i.e., the location coordinates of the reference point), the model identification error E is calculated:
Figure BDA0002368452640000104
updating the network connection weight w according to the model identification error EijAnd wjk：
Figure BDA0002368452640000105
Figure BDA0002368452640000106
δjk＝(yk-Ok)·Ok·(1-Ok)；
Wherein the content of the first and second substances,
Figure BDA0002368452640000107
for learning rate, according to realityThe situation is set.
Updating the threshold value a of the network node according to the model identification error Ej、bk：
Figure BDA0002368452640000108
Figure BDA0002368452640000109
And when the identification error of the convolutional neural network model is smaller than a preset threshold value, finishing iteration to obtain a trained convolutional neural network model, and taking the trained convolutional neural network model as the preset convolutional neural network model.
And 209, inputting the acquired signal strength RSSI values of the test points in the subareas into a preset convolutional neural network model corresponding to the subareas, and outputting the positioning results of the test points.
It should be noted that each sub-region corresponds to one preset convolutional neural network model, when a certain test point needs to be located, a signal receiver may be arranged at the test point to record a signal strength RSSI value sent by each AP, and the obtained signal strength RSSI values from different APs are input into the preset convolutional neural network model corresponding to the sub-region where the test point is located, so as to output a position coordinate of the test point and obtain location information of the test point, where the sub-region where the test point is located may be determined by the aforementioned clustering algorithm. Comparing the error of the WLAN indoor positioning in the embodiment of the present application with the error of the conventional positioning method, as shown in table 1, the WLAN indoor positioning method in the embodiment of the present application has smaller error and higher accuracy.
TABLE 1 error comparison table
Test point Positioning error of traditional scheme Positioning error of the present application
Test point 1 5.3％ 3.2％
Test point 2 6.1％ 4.1％
Test point 3 10.4％ 5.3％
Test point 4 7.2％ 3.1％
For easy understanding, referring to fig. 3, an embodiment of a WLAN indoor positioning apparatus provided in the present application includes:
the first obtaining module 301 is configured to obtain first environmental information data in a target room, where the first environmental information data includes spatial three-dimensional point cloud data, a position coordinate of a target object, an azimuth angle of the target object, and a pitch angle.
The principal component analysis module 302 is configured to perform principal component analysis on the first environmental information data, and select the first environmental information data corresponding to the cumulative contribution rate meeting the preset condition as the second environmental information data.
The building module 303 is configured to build an environment in the target room based on the second environment information data, so as to obtain a target indoor model.
A second obtaining module 304, configured to arrange APs and set reference points in the target indoor model, and obtain a signal strength RSSI value of each reference point in the target indoor model.
The dividing module 305 is configured to perform region division on the target indoor environment according to a fuzzy clustering algorithm based on the RSSI value of each reference point to obtain a plurality of sub-regions.
And the positioning module 306 is configured to input the obtained signal strength RSSI value of the test point in the sub-region to a preset convolutional neural network model corresponding to the sub-region, and output a positioning result of the test point.
Further, still include:
and the preprocessing module 307 is configured to perform coordinate processing on the target indoor model.
Further, the second obtaining module 304 includes:
an arrangement submodule 3041 for arranging a plurality of APs and setting a plurality of reference points in the target indoor model, the reference points being uniformly distributed according to the preset distance.
An acquisition sub-module 3042 for acquiring the RSSI value of each reference point from each AP.
Further, the second obtaining module 304 further includes:
the selecting submodule 3043 is configured to select a target reference point from the plurality of reference points, establish a three-dimensional coordinate system with the target reference point as an origin, and obtain a position coordinate of each reference point based on the three-dimensional coordinate system.
Further, still include:
a training set obtaining module 308, configured to use the position coordinates of the reference point of each sub-region and the signal strength RSSI value of the reference point as a training set;
the training module 309 is configured to train a convolutional neural network model in each training set, obtain a plurality of trained convolutional neural network models when the convolutional neural network models reach a convergence condition, and use the trained convolutional neural network models as preset convolutional neural network models.
Further, the principal component analysis module 302 is specifically configured to:
calculating a correlation coefficient based on the first environment information data to generate a correlation coefficient matrix;
solving a characteristic equation constructed based on the correlation coefficient matrix to obtain a characteristic value;
calculating a contribution rate based on the feature values;
and selecting the first environment information data corresponding to the least number of characteristic values corresponding to the accumulated contribution rate of more than 85% as the second environment information data.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for executing all or part of the steps of the method described in the embodiments of the present application through a computer device (which may be a personal computer, a server, or a network device). And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.

Claims (10)

1. A WLAN indoor positioning method, comprising:
acquiring first environmental information data in a target room, wherein the first environmental information data comprises spatial three-dimensional point cloud data, position coordinates of a target object, an azimuth angle and a pitch angle of the target object;
performing principal component analysis on the first environmental information data, and selecting the first environmental information data corresponding to the accumulated contribution rate meeting preset conditions as second environmental information data;
constructing the environment in the target room based on the second environment information data to obtain a target indoor model;
arranging APs and setting reference points in the target indoor model, and acquiring a signal strength RSSI value of each reference point in the target indoor model;
based on the signal strength RSSI value of each reference point, carrying out region division on the environment in the target room according to a fuzzy clustering algorithm to obtain a plurality of sub-regions;
and inputting the obtained signal strength RSSI value of the test point in the subregion into a preset convolutional neural network model corresponding to the subregion, and outputting a positioning result of the test point.
2. The WLAN indoor positioning method according to claim 1, wherein the performing principal component analysis on the first environment information data and selecting the first environment information data corresponding to an accumulated contribution rate satisfying a preset condition as the second environment information data includes:
calculating a correlation coefficient based on the first environment information data to generate a correlation coefficient matrix;
solving a characteristic equation constructed based on the correlation coefficient matrix to obtain a characteristic value;
calculating a contribution rate based on the feature values;
and selecting the first environment information data corresponding to the characteristic value with the minimum quantity corresponding to the accumulated contribution rate of more than 85% as the second environment information data.
3. The WLAN indoor positioning method according to claim 1, wherein the step of arranging APs and setting reference points in the target indoor model, and obtaining RSSI (signal strength indicator) values of each reference point in the target indoor model further comprises:
and carrying out coordinate processing on the target indoor model.
4. The WLAN indoor positioning method according to claim 3, wherein the arranging APs and setting reference points in the target indoor model, and obtaining a signal strength RSSI value of each reference point in the target indoor model comprises:
arranging a plurality of APs and setting a plurality of reference points in the target indoor model, wherein the reference points are uniformly distributed according to a preset distance;
and acquiring the RSSI value of each reference point from each AP.
5. The WLAN indoor positioning method according to claim 4, wherein the arranging of a plurality of APs and the setting of a plurality of reference points in the target indoor model further comprise:
selecting a target reference point from the plurality of reference points, establishing a three-dimensional coordinate system by taking the target reference point as an origin, and obtaining the position coordinate of each reference point based on the three-dimensional coordinate system.
6. The WLAN indoor positioning method according to claim 5, wherein the step of inputting the obtained RSSI values of the test points in the sub-area to a preset convolutional neural network model corresponding to the sub-area and outputting the positioning results of the test points further comprises:
taking the position coordinates of the reference point of each sub-area and the signal strength RSSI value of the reference point as a training set;
and each training set trains a convolutional neural network model, when the convolutional neural network model reaches a convergence condition, a plurality of trained convolutional neural network models are obtained, and the trained convolutional neural network models are used as the preset convolutional neural network models.
7. A WLAN indoor positioning apparatus, comprising:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring first environmental information data in a target room, and the first environmental information data comprises spatial three-dimensional point cloud data, position coordinates of a target object, an azimuth angle and a pitch angle of the target object;
the principal component analysis module is used for carrying out principal component analysis on the first environmental information data and selecting the first environmental information data corresponding to the accumulated contribution rate meeting the preset conditions as second environmental information data;
the construction module is used for constructing the environment in the target room based on the second environment information data to obtain a target indoor model;
the second acquisition module is used for arranging APs and setting reference points in the target indoor model and acquiring the signal strength RSSI value of each reference point in the target indoor model;
the dividing module is used for carrying out region division on the environment in the target room according to a fuzzy clustering algorithm based on the signal strength RSSI value of each reference point to obtain a plurality of sub-regions;
and the positioning module is used for inputting the acquired signal strength RSSI value of the test point in the subregion into a preset convolutional neural network model corresponding to the subregion and outputting a positioning result of the test point.
8. The WLAN indoor positioning apparatus of claim 7, further comprising:
and the preprocessing module is used for carrying out coordinate processing on the target indoor model.
9. The WLAN indoor positioning apparatus of claim 8, wherein the second acquisition module comprises:
the arrangement submodule is used for arranging a plurality of APs in the target indoor model and setting a plurality of reference points, and the reference points are uniformly distributed according to a preset distance;
and the acquisition sub-module is used for acquiring the signal strength RSSI value of each reference point from each AP.
10. The WLAN indoor positioning apparatus of claim 9, wherein the second acquisition module further comprises:
and the selection submodule is used for selecting a target reference point from the plurality of reference points, establishing a three-dimensional coordinate system by taking the target reference point as an origin, and obtaining the position coordinate of each reference point based on the three-dimensional coordinate system.
CN202010043140.2A 2020-01-15 2020-01-15 WLAN indoor positioning method and device Active CN111263295B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010043140.2A CN111263295B (en) 2020-01-15 2020-01-15 WLAN indoor positioning method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010043140.2A CN111263295B (en) 2020-01-15 2020-01-15 WLAN indoor positioning method and device

Publications (2)

Publication Number Publication Date
CN111263295A true CN111263295A (en) 2020-06-09
CN111263295B CN111263295B (en) 2021-08-13

Family

ID=70952116

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010043140.2A Active CN111263295B (en) 2020-01-15 2020-01-15 WLAN indoor positioning method and device

Country Status (1)

Country Link
CN (1) CN111263295B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113206774A (en) * 2021-03-19 2021-08-03 武汉特斯联智能工程有限公司 Control method and device of intelligent household equipment based on indoor positioning information
CN115988634A (en) * 2022-11-23 2023-04-18 香港中文大学（深圳） Indoor positioning method and subspace feature extraction method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1182897A1 (en) * 2000-07-10 2002-02-27 ScoreBoard, Inc. Collection and analysis of signal propagation data in a wireless system
CN104093203A (en) * 2014-07-07 2014-10-08 浙江师范大学 Access point selection algorithm used for wireless indoor positioning
CN104574386A (en) * 2014-12-26 2015-04-29 速感科技（北京）有限公司 Indoor positioning method based on three-dimensional environment model matching
CN105101408A (en) * 2015-07-23 2015-11-25 常熟理工学院 Indoor positioning method based on distributed AP selection strategy
CN106131959A (en) * 2016-08-11 2016-11-16 电子科技大学 A kind of dual-positioning method divided based on Wi Fi signal space

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1182897A1 (en) * 2000-07-10 2002-02-27 ScoreBoard, Inc. Collection and analysis of signal propagation data in a wireless system
CN104093203A (en) * 2014-07-07 2014-10-08 浙江师范大学 Access point selection algorithm used for wireless indoor positioning
CN104574386A (en) * 2014-12-26 2015-04-29 速感科技（北京）有限公司 Indoor positioning method based on three-dimensional environment model matching
CN105101408A (en) * 2015-07-23 2015-11-25 常熟理工学院 Indoor positioning method based on distributed AP selection strategy
CN106131959A (en) * 2016-08-11 2016-11-16 电子科技大学 A kind of dual-positioning method divided based on Wi Fi signal space

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113206774A (en) * 2021-03-19 2021-08-03 武汉特斯联智能工程有限公司 Control method and device of intelligent household equipment based on indoor positioning information
CN115988634A (en) * 2022-11-23 2023-04-18 香港中文大学（深圳） Indoor positioning method and subspace feature extraction method
CN115988634B (en) * 2022-11-23 2023-08-18 香港中文大学（深圳） Indoor positioning method and subspace feature extraction method

Also Published As

Publication number Publication date
CN111263295B (en) 2021-08-13

Similar Documents

Publication Publication Date Title
CN110012428B (en) Indoor positioning method based on WiFi
CN107071743B (en) Rapid KNN indoor WiFi positioning method based on random forest
Zhang et al. A comprehensive study of bluetooth fingerprinting-based algorithms for localization
CN110933599B (en) Self-adaptive positioning method fusing UWB and WIFI fingerprints
CN106851571B (en) Decision tree-based rapid KNN indoor WiFi positioning method
CN104507050B (en) Probabilistic type finger print matching method in a kind of WiFi indoor positionings
CN108919177B (en) Positioning map construction method based on virtual information source estimation and track correction
WO2014106363A1 (en) Mobile device positioning system and method
CN111263295B (en) WLAN indoor positioning method and device
CN112748397A (en) UWB positioning method based on self-adaptive BP neural network under non-line-of-sight condition
CN109379711B (en) positioning method
CN109348416B (en) Fingerprint indoor positioning method based on binary k-means
WO2016112758A1 (en) Method and apparatus for locating terminal
CN103945531B (en) Method for WLAN indoor positioning Radio Map updating based on information entropy
CN111711923A (en) Wireless sensor network node positioning method based on UAV
CN114449452A (en) Indoor positioning algorithm for heterogeneous equipment
CN112367614A (en) LSTM-based Wi-Fi and geomagnetic field fusion indoor positioning algorithm
CN112533136A (en) WLAN fingerprint positioning method based on deep learning
WO2022242018A1 (en) Indoor target positioning method based on improved cnn model
Bal et al. Regression of large-scale path loss parameters using deep neural networks
CN108680897B (en) Indoor positioning method and device, electronic equipment and storage medium
CN110596668B (en) Target external radiation source passive positioning method based on reciprocal deep neural network
CN108491730B (en) Inter-track correlation privacy protection method based on Lagrange optimization
CN113194401B (en) Millimeter wave indoor positioning method and system based on generative countermeasure network
CN113645565B (en) Indoor positioning method based on hexagonal closest packing structure

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