CN103533647A - Radio frequency map self-adaption positioning method based on clustering mechanism and robust regression - Google Patents

Radio frequency map self-adaption positioning method based on clustering mechanism and robust regression Download PDF

Info

Publication number
CN103533647A
CN103533647A CN201310505864.4A CN201310505864A CN103533647A CN 103533647 A CN103533647 A CN 103533647A CN 201310505864 A CN201310505864 A CN 201310505864A CN 103533647 A CN103533647 A CN 103533647A
Authority
CN
China
Prior art keywords
radio frequency
rssi
point
frequency map
check
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
CN201310505864.4A
Other languages
Chinese (zh)
Other versions
CN103533647B (en
Inventor
叶阿勇
杨小亮
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.)
Fujian Normal University
Original Assignee
Fujian Normal University
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 Fujian Normal University filed Critical Fujian Normal University
Priority to CN201310505864.4A priority Critical patent/CN103533647B/en
Publication of CN103533647A publication Critical patent/CN103533647A/en
Application granted granted Critical
Publication of CN103533647B publication Critical patent/CN103533647B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Position Fixing By Use Of Radio Waves (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention relates to a radio frequency map self-adaption positioning method based on a clustering mechanism and robust regression. According to the method, in the offline training stage, signal features of multiple reference points and check nodes of one scene are collected firstly, and a static radio frequency map is established; then, a path loss parameter of each reference point is calculated according to the static radio frequency map of the scene; the reference points are clustered according to the path loss parameters, and a positioning area is divided into a plurality of subareas; finally, RSSI (received signal strength identification) fingerprints of a reference point in each subarea and RSSI fingerprints of a check point in the subarea are subjected to linear robust regression; and in the online positioning stage, the radio frequency map is updated by check point RSSI vectors and robust regression parameters which are acquired at regular time, and then, the radio frequency map is used in a weighted K neighbor algorithm, so that the positioning is realized. The method is simple, easy to implement and high in positioning accuracy, influences of outdating of the radio frequency map on positioning calculation can be effectively reduced, and the outdating of the radio frequency map is caused by factors such as RSSI random jittering, interference of walking of indoor workers and the like.

Description

Radio frequency map self-adaptive positioning method based on clustering mechanism and robust regression
Technical Field
The invention relates to the technical field of indoor wireless positioning, in particular to a radio frequency map self-adaptive positioning method based on a clustering mechanism and robust regression.
Background
With the development of mobile communication and wireless technology, location-based services are receiving more and more attention. People's demand for indoor positioning information increases day by day, and large-scale indoor places such as underground parking area, logistics storage, mine, hospital, prison, archaeological scene, exhibition room, museum all need carry out real-time location to personnel or article, can realize functions such as navigation, control and intelligent management. The indoor positioning method based on the radio frequency map is one of the most important indoor wireless positioning methods in recent years due to high positioning accuracy and simple calculation method.
The indoor positioning method based on the radio frequency map mainly obtains the position of an unknown node by matching and calculating the unknown node in the online positioning stage and the signal fingerprint in the radio frequency map of the reference point established in the offline stage. However, because the indoor environment is influenced by factors such as wall shielding and people walking, the radio frequency map cannot accurately represent the real-time environment signal fingerprint, and thus the positioning accuracy is not high. In order to improve the accuracy of indoor positioning, the radio frequency map established previously needs to be updated regularly. In the past, the indoor positioning method aiming at radio frequency map updating simply adds check nodes, and then determines a regression coefficient through a general multiple linear regression method. The algorithms do not consider the actual loss characteristics of the RSSI signal and the update function is influenced by the abnormal value of the RSSI, so the obtained result is often greatly influenced by indoor interference factors.
Therefore, it is a technical issue to be urgently solved by those skilled in the art to provide a method for reducing the influence of indoor interference on RSSI positioning accuracy, aiming at the accuracy problem of an indoor positioning method based on a radio frequency map in indoor wireless network positioning.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a radio frequency map self-adaptive positioning method based on a clustering mechanism and robust regression, which is favorable for reducing the positioning error of RSSI (received signal strength indicator) caused by interference factors such as indoor multipath, diffraction, obstacles, personnel walking and the like, thereby improving the indoor positioning precision.
In order to achieve the purpose, the technical scheme of the invention is as follows: a radio frequency map self-adaptive positioning method based on a clustering mechanism and robust regression comprises the following steps:
(1) dividing a target area into uniform grids, placing a proper amount of check nodes and beacon nodes in the target area, measuring the received RSSI of each beacon node at the center point of each grid and the check nodes respectively, and constructing a static RSSI radio frequency map;
(2) calculating a path loss parameter of each reference point, namely a grid central point according to the static RSSI radio frequency map;
(3) clustering the reference points according to the path loss parameters, and dividing the positioning area into a plurality of sub-areas;
(4) performing robust linear regression on the RSSI fingerprint of the reference point in each sub-region and the RSSI fingerprint of the check point in the sub-region;
(5) and updating the radio frequency map by using the regularly acquired check point RSSI and the robust regression coefficient, and using the radio frequency map in a weighted K neighbor algorithm to estimate the position of the unknown node.
Further, in the step (1), the target area is divided into uniform grids, the check nodes placed in the target area are uniformly distributed in the target area, one beacon node is placed at each of four corners of the target area, and the beacon nodes communicate with the base station through the mobile nodes.
Further, the static RSSI RF map is at t0The time is obtained in the following way: acquiring RSSI values at the center point of each grid and the check node for n times continuously, and averaging the received RSSI values of each beacon node, namely:
Figure 2013105058644100002DEST_PATH_IMAGE002
Figure 2013105058644100002DEST_PATH_IMAGE004
wherein,u ij (t 0) Is composed oft 0At the first momentiN numbers of received from reference pointsjThe average of the RSSI values of the individual beacons,r kj (t 0) Is composed oft 0At the first momentkN numbers of check nodesjThe average of the RSSI values of the individual beacons,i=(1,…,N)，Nfor reference purposesThe number of the points is such that,k=(1,…,M)，Min order to check the number of nodes,j=(1,…,P)，Pn is a natural number.
Further, in the step (2), the method for calculating the path loss parameter of each reference point includes: using indoor signal propagation feature models
Figure 2013105058644100002DEST_PATH_IMAGE006
Obtained at the reference pointl iMeasure the beacon node APjPath loss parameter ofWhereind ij Is as followsiA reference point and the firstjThe distance between the individual beacon nodes is,αis a constant in this environment.
Further, in the step (3), the method for clustering the reference points according to the path loss parameters is as follows:
reference pointl iVector for path loss parameters corresponding to each beacon node
Figure 2013105058644100002DEST_PATH_IMAGE010
It is shown that the path loss parameter similarity is defined as:
Figure 2013105058644100002DEST_PATH_IMAGE012
，
the above formula represents
Figure 2013105058644100002DEST_PATH_IMAGE014
And
Figure 2013105058644100002DEST_PATH_IMAGE016
the similarity degree of (1) is less than or equal tou≤M，1≤v≤M；
Dividing each reference point into K clusters by using a K mean clustering algorithm, and expressing asC={c 1,c 2,…,c k For a certain fixed K, maximize the following:
Figure 2013105058644100002DEST_PATH_IMAGE018
wherein
Figure 2013105058644100002DEST_PATH_IMAGE020
Is a
Figure 2013105058644100002DEST_PATH_IMAGE022
Average value of (1) representsyThe cluster center of the class is the center of the cluster,is of the class
Figure 499641DEST_PATH_IMAGE022
The number of samples.
Further, in the step (4), the method for performing robust linear regression on the RSSI fingerprint of the reference point in each sub-region and the RSSI fingerprint of the check point in the sub-region includes: within each sub-region according to t0In time-of-day static RSSI radio frequency mapsu ij (t 0) Andr kj (t 0) Determining t using robust multiple linear regression0The relation between the RSS of each reference point and the RSS of the check node at the moment is as follows:
Figure 2013105058644100002DEST_PATH_IMAGE026
estimating regression coefficient by using iterative weighted least square, and determining weight of each point according to regression residual errorw iTo achieve the purpose of robustness, the optimized objective function is:
Figure 2013105058644100002DEST_PATH_IMAGE028
further, in the step (5), a check point RSSI vector obtained at a fixed time is used
Figure 2013105058644100002DEST_PATH_IMAGE030
And a firstiRobust regression coefficient for each reference position point
Figure 2013105058644100002DEST_PATH_IMAGE032
Updating the radio frequency map, wherein the robust regression coefficients aref ijComprises the following steps:
，1≤i≤N，1≤j≤M
in the positioning stage, for each reference position pointl iUsing the positionf ijThe sum check point receives the RSS value of the beacon node r k=(r k1, …,r kj,…,r kM) To the position pointl iPeriodically updating the radio frequency fingerprints corresponding to the beacon nodes; supposing a time of daytAt the reference position pointl iBy passingf ijThe function obtains the radio frequency fingerprint corresponding to each beacon node as
Figure 2013105058644100002DEST_PATH_IMAGE036
And then the radio frequency map is used for a weighted K neighbor algorithm to estimate the position of the unknown node.
Compared with the prior art, the invention has the beneficial effects that: the invention makes the reference points with the same or similar radio frequency signal propagation characteristics into the same cluster by clustering the reference points, and obtains the relationship between the RSSI of each reference point in the cluster and the RSSI of the check node in the same cluster, thereby ensuring that the fingerprint map updated by using the relationship has higher reliability. In addition, the invention performs robust linear regression on the reference point in each sub-area and the RSSI fingerprint of the check point in the sub-area, weakens the influence of the check point on acquiring the abnormal RSSI value of each beacon node, and further improves the positioning accuracy. The radio frequency map self-adaptive positioning method based on the clustering mechanism and the robust regression has the advantages of no large amount of calculation, simple realization and suitability for application scenes of wireless networks with limited energy, limited computing capacity, limited storage resources and limited communication capacity.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention.
Fig. 2 is a schematic diagram of a wireless network node and a mesh point deployment according to an embodiment of the present invention.
Fig. 3 is a characteristic diagram of the path loss parameter η of signal propagation in different regions according to an embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and the embodiments.
The invention relates to a radio frequency map self-adaptive positioning method based on a clustering mechanism and robust regression, which comprises the following steps:
(1) dividing a target area into uniform grids, placing a proper amount of check nodes and beacon nodes in the target area, measuring the received RSSI of each beacon node at the center point of each grid and the check nodes respectively, and constructing a static RSSI radio frequency map;
(2) calculating a path loss parameter of each reference point, namely a grid central point according to the static RSSI radio frequency map;
(3) clustering the reference points according to the path loss parameters, and dividing the positioning area into a plurality of sub-areas;
(4) performing robust linear regression on the RSSI fingerprint of the reference point in each sub-region and the RSSI fingerprint of the check point in the sub-region;
(5) and updating the radio frequency map by using the regularly acquired check point RSSI and the robust regression coefficient, and using the radio frequency map in a weighted K neighbor algorithm to estimate the position of the unknown node.
In an embodiment of the present invention, in the step (1), the target area is divided into uniform grids, an interval between center points of the grids is 1.5m, the check nodes placed in the target area are uniformly distributed in the target area, one beacon node is placed at each of four corners of the target area, and the beacon nodes communicate with the base station through the mobile node.
In an embodiment of the invention, the static RSSI radio frequency map is at t0The time is obtained in the following way: acquiring RSSI values at the center point of each grid and the check node for n times continuously, and averaging the received RSSI values of each beacon node, namely:
Figure 404012DEST_PATH_IMAGE002
wherein,u ij (t 0) Is composed oft 0At the first momentiN numbers of received from reference pointsjThe average of the RSSI values of the individual beacons,r kj (t 0) Is composed oft 0At the first momentkN numbers of check nodesjThe average of the RSSI values of the individual beacons,i=(1,…,N)，Nfor the number of reference points it is,k=(1,…,M)，Min order to check the number of nodes,j=(1,…,P)，Pn is a natural number.
In an embodiment of the present invention, in the step (2), the method for calculating the path loss parameter of each reference point includes: using indoor signal propagation feature models
Figure 800544DEST_PATH_IMAGE006
Obtained at the reference pointl iMeasure the beacon node APjPath loss parameter of
Figure 623007DEST_PATH_IMAGE008
Whereind ij Is as followsiA reference point and the firstjThe distance between the individual beacon nodes is,αis a constant in this environment.
In an embodiment of the present invention, in the step (3), the method for clustering the reference points according to the path loss parameters includes:
reference pointl iVector for path loss parameters corresponding to each beacon node
Figure 890040DEST_PATH_IMAGE010
To cluster all the reference points, the path loss parameter similarity is defined as:
Figure 26623DEST_PATH_IMAGE012
，
the above formula represents
Figure 695502DEST_PATH_IMAGE014
And
Figure 688866DEST_PATH_IMAGE016
is likeDegree of 1 is less than or equal tou≤M，1≤v≤M；
Dividing each reference point into K clusters by using a K mean clustering algorithm, and expressing asC={c 1,c 2,…,c k For a certain fixed K, maximize the following:
Figure 443195DEST_PATH_IMAGE018
whereinIs a
Figure 890543DEST_PATH_IMAGE022
Average value of (1) representsyThe cluster center of the class is the center of the cluster,
Figure 320387DEST_PATH_IMAGE024
is of the class
Figure 499696DEST_PATH_IMAGE022
The number of samples.
In an embodiment of the present invention, in the step (4), the method of performing robust linear regression on the RSSI fingerprint of the reference point in each sub-region and the RSSI fingerprint of the check point in the sub-region includes: within each sub-region according to t0In time-of-day static RSSI radio frequency mapsu ij (t 0) Andr kj (t 0) Determining t using robust multiple linear regression0The relation between the RSS of each reference point and the RSS of the check node at the moment is as follows:estimating regression coefficient by using iterative weighted least square, and determining weight of each point according to regression residual errorw iTo achieve the purpose of robustness, optimization thereofThe standard function is:
Figure 418290DEST_PATH_IMAGE028
in an embodiment of the present invention, in the step (5), the RSSI vector of the check point obtained at regular time is utilized
Figure 753457DEST_PATH_IMAGE030
And a firstiRobust regression coefficient for each reference position point
Figure 683977DEST_PATH_IMAGE032
Updating the radio frequency map, wherein the robust regression coefficients aref ijComprises the following steps:
，1≤i≤N，1≤j≤M
in the positioning stage, for each reference position pointl iUsing the positionf ijThe sum check point receives the RSS value of the beacon node r k=(r k1, …,r kj,…,r kM) To the position pointl iPeriodically updating the radio frequency fingerprints corresponding to the beacon nodes; supposing a time of daytAt the reference position pointl iBy passingf ijThe function obtains the radio frequency fingerprint corresponding to each beacon node as
Figure 260769DEST_PATH_IMAGE036
And then the radio frequency map is used for a weighted K neighbor algorithm to estimate the position of the unknown node.
Specifically, as shown in fig. 1, the method comprises the following steps:
step 1: in this embodiment, the wireless network nodes are deployed as shown in fig. 2, a target area is divided into uniform grids, the interval between the center points of each grid is 1.5m, a proper amount of check nodes are placed in the target area, the check nodes are uniformly distributed in the target area, and a beacon node is placed at each of four corners of the target area, and the beacon node communicates with a base station through a mobile node. Acquiring RSSI values at the central point of the grid to be acquired and the check node for n times continuously, and averaging the received RSSI values of each beacon node, namely:
Figure 32416DEST_PATH_IMAGE002
，
Figure 920738DEST_PATH_IMAGE004
，
Figure DEST_PATH_IMAGE038
,for the number of reference points it is,
Figure DEST_PATH_IMAGE042
，Min order to check the number of nodes,，
Figure DEST_PATH_IMAGE046
n is a natural number.
Step 2: calculating the path loss parameter of each reference point according to the static RSSI radio frequency map of the scene by utilizing an indoor signal propagation characteristic model
Figure DEST_PATH_IMAGE048
Is obtained atl iMeasured beacon APjPath loss parameter ofWherein
Figure DEST_PATH_IMAGE052
Is as followsiA reference point and the firstjThe distance between the individual beacon nodes is,
Figure DEST_PATH_IMAGE054
is a constant in this environment.
And step 3: clustering the reference points according to the path loss parameter pairs obtained in the step 2, wherein the clustering basis is as follows: as shown in FIG. 3, in different regions, the signal propagation path loss parameterηIs different in value, but in a certain areaηThe same or similar indicates that the radio frequency signal propagation characteristics are similar in this region. Therefore we can be based onηThe areas where the reference points are located are clustered according to the value approximation degree, and according to the analysis, the radio frequency signal propagation characteristics of the reference point positions in the same cluster are the same or similar. The reason for the clustering is that the reference points RSSI and the check points far away from the reference points RSSI are not closely related, so that the reference points with the same or similar radio frequency signal propagation characteristics can be classified into the same cluster, and t is calculated0And the relationship between the RSSI of each reference point in the time cluster and the RSSI of the check node in the same cluster, so that the fingerprint map updated by using the relationship has higher reliability.
The clustering method comprises the following steps:
reference pointl iAvailable vector of path loss parameters corresponding to each beacon node
Figure 927877DEST_PATH_IMAGE010
. To cluster all reference points, the path loss parameter similarity is defined as
Figure DEST_PATH_IMAGE056
It representsAnd
Figure 409860DEST_PATH_IMAGE016
to the extent of the similarity in the direction of the line,
Figure DEST_PATH_IMAGE058
,
Figure DEST_PATH_IMAGE060
。the greater the value of (a) is,and
Figure DEST_PATH_IMAGE066
the more similar.
And dividing each reference point into K clusters by using a K mean clustering algorithm, wherein for a certain fixed K, a heuristic algorithm is adopted, namely the following formula is maximized:
Figure DEST_PATH_IMAGE068
wherein
Figure 972428DEST_PATH_IMAGE020
Is a
Figure 393045DEST_PATH_IMAGE022
Average value of (1) representsyThe cluster center of the class is the center of the cluster,
Figure 865615DEST_PATH_IMAGE024
is of the class
Figure 916747DEST_PATH_IMAGE022
The number of samples.
And 4, step 4: and performing robust linear regression on the RSSI fingerprint of the reference point in each sub-region and the RSSI fingerprint of the check point in the sub-region. Since the indoor wireless signal is susceptible to interference factors such as indoor multipath, diffraction, article displacement, personnel walking and the like, the RSSI value of each beacon node acquired by the check point is easy to generate an abnormal value. In order to reduce the effect of the abnormal point, different weights can be applied to different points, namely, a point with small residual error is given a larger weight, a point with larger residual error is given a smaller weight, the weight is determined according to the size of the residual error, weighted least square estimation is established according to the weight, and iteration is repeated to improve the weight coefficient until the change of the weight coefficient is smaller than a certain allowable error. And the abnormal RSS value interfered indoors can be effectively eliminated by adopting the robust regression estimation and utilizing the robust characteristic of the robust regression estimation on the error term distribution. Especially when the error is not normal, it is much better than LSE.
Within each sub-region according to t0In time-of-day static RSSI radio frequency maps
Figure DEST_PATH_IMAGE070
And
Figure DEST_PATH_IMAGE072
，
Figure DEST_PATH_IMAGE074
，
Figure 31859DEST_PATH_IMAGE042
，
Figure 256166DEST_PATH_IMAGE044
determining t using robust multiple linear regression0The relation between the RSS of each reference point and the RSS of the check node at the moment is as follows:estimating regression coefficient by using iterative weighted least square, and determining weight of each point according to regression residual errorw iTo achieve the purpose of robustness, the optimized objective function is:
Figure 539697DEST_PATH_IMAGE028
and 5: according to check points obtained at fixed time
Figure 952224DEST_PATH_IMAGE030
And a firstiRobust regression coefficient for each reference position point
Figure 980223DEST_PATH_IMAGE032
Updating the radio frequency map, wherein the robust regression coefficients aref ijComprises the following steps:
，,
Figure DEST_PATH_IMAGE080
a positioning stage of positioning each reference position pointl iUsing the positionf ijThe sum check point receives the RSS value of the beacon node r k=(r k1, …,r kj,…,r Mk) To the position pointl iAnd periodically updating the radio frequency fingerprints corresponding to the beacon nodes.
Supposing a time of daytAt the reference position pointl iBy passingf ijAnd (4) performing function to obtain the radio frequency fingerprints corresponding to the beacon nodes, and then using the radio frequency map for a weighted K neighbor algorithm to estimate the positions of the unknown nodes.
In summary, the reference points are clustered according to the path loss parameters, the positioning area is divided into a plurality of sub-areas, and then the RSSI fingerprint of the reference point in each sub-area and the RSSI fingerprint of the check point in the sub-area are subjected to robust linear regression; in the on-line positioning stage, the radio frequency map is updated by using the regularly acquired check point RSSI and the steady regression coefficient, and then the radio frequency map is used for a weighted K neighbor algorithm, so that the positioning is realized. Therefore, the influence of outdated radio frequency maps on positioning calculation caused by RSSI random jitter, indoor personnel walking interference and other factors can be effectively reduced, and indoor positioning is accurately realized.
The above are preferred embodiments of the present invention, and all changes made according to the technical scheme of the present invention that produce functional effects do not exceed the scope of the technical scheme of the present invention belong to the protection scope of the present invention.

Claims (7)

1. A radio frequency map self-adaptive positioning method based on a clustering mechanism and robust regression is characterized by comprising the following steps:
(1) dividing a target area into uniform grids, placing a proper amount of check nodes and beacon nodes in the target area, measuring the received RSSI of each beacon node at the center point of each grid and the check nodes respectively, and constructing a static RSSI radio frequency map;
(2) calculating a path loss parameter of each reference point, namely a grid central point according to the static RSSI radio frequency map;
(3) clustering the reference points according to the path loss parameters, and dividing the target area into a plurality of sub-areas;
(4) performing robust linear regression on the RSSI fingerprint of the reference point in each sub-region and the RSSI fingerprint of the check point in the sub-region;
(5) and updating the radio frequency map by using the regularly acquired check point RSSI and the robust regression coefficient, and using the radio frequency map in a weighted K neighbor algorithm to estimate the position of the unknown node.
2. The method according to claim 1, wherein in step (1), the target area is divided into a uniform grid, the check nodes disposed in the target area are uniformly distributed in the target area, one beacon node is disposed at each of four corners of the target area, and the beacon nodes communicate with the base station through the mobile nodes.
3. The method as claimed in claim 2, wherein the static RSSI RF map is at t0The time is obtained in the following way: acquiring RSSI values at the center point of each grid and the check node for n times continuously, and averaging the received RSSI values of each beacon node, namely:
Figure 2013105058644100001DEST_PATH_IMAGE004
wherein,u ij (t 0) Is composed oft 0At the first momentiN numbers of received from reference pointsjThe average of the RSSI values of the individual beacons,r kj (t 0) Is composed oft 0At the first momentkN numbers of check nodesjThe average of the RSSI values of the individual beacons,i=(1,…,N)，Nfor the number of reference points it is,k=(1,…,M)，Min order to check the number of nodes,j=(1,…,P)，Pn is a natural number.
4. The method according to claim 1, wherein the step (2) of calculating the path loss parameter of each reference point comprises: using indoor signal propagation feature models
Figure 2013105058644100001DEST_PATH_IMAGE006
Obtained at the reference pointl iMeasure the beacon node APjPath loss parameter of
Figure 2013105058644100001DEST_PATH_IMAGE008
Whereind ij Is as followsiA reference point and the firstjThe distance between the individual beacon nodes is,αis a constant in this environment.
5. The method for adaptive positioning of radio frequency map based on clustering mechanism and robust regression as claimed in claim 1, wherein in step (3), the method for clustering the reference points according to the path loss parameters is as follows:
reference pointl iVector for path loss parameters corresponding to each beacon node
Figure 2013105058644100001DEST_PATH_IMAGE010
It is shown that the path loss parameter similarity is defined as:
Figure 2013105058644100001DEST_PATH_IMAGE012
，
the above formula representsAnd
Figure 2013105058644100001DEST_PATH_IMAGE016
the similarity degree of (1) is less than or equal tou≤M，1≤v≤M；
Dividing each reference point into K clusters by using a K mean clustering algorithm, and expressing asC={c 1,c 2,…,c k For a certain fixed K, maximize the following:
wherein
Figure 2013105058644100001DEST_PATH_IMAGE020
Is a
Figure 2013105058644100001DEST_PATH_IMAGE022
Average value of (1) representsyThe cluster center of the class is the center of the cluster,
Figure 2013105058644100001DEST_PATH_IMAGE024
is of the class
Figure 916531DEST_PATH_IMAGE022
The number of samples.
6. The method as claimed in claim 1, wherein in step (4), the RSSI fingerprint of the reference point in each sub-region and the RSSI fingerprint of the check point in the sub-region are subjected to robust linear positioningThe regression method comprises the following steps: within each sub-region according to t0In time-of-day static RSSI radio frequency mapsu ij (t 0) Andr kj (t 0) Determining t using robust multiple linear regression0The relation between the RSS of each reference point and the RSS of the check node at the moment is as follows:estimating regression coefficient by using iterative weighted least square, and determining weight of each point according to regression residual errorw iTo achieve the purpose of robustness, the optimized objective function is:
Figure 2013105058644100001DEST_PATH_IMAGE028
。
7. the method as claimed in claim 6, wherein in step (5), the timing-derived checkpoint RSSI vector is used to perform the adaptive positioning method
Figure 2013105058644100001DEST_PATH_IMAGE030
And a firstiRobust regression coefficient for each reference position pointUpdating the radio frequency map, wherein the robust regression coefficients aref ijComprises the following steps:
Figure 2013105058644100001DEST_PATH_IMAGE034
，1≤i≤N，1≤j≤M
in the positioning stage, for each reference position pointl iUsing the positionf ijReceiving beacon section by checksum check pointRSS value of a point r k=(r k1, …,r kj,…,r kM) To the position pointl iPeriodically updating the radio frequency fingerprints corresponding to the beacon nodes; supposing a time of daytAt the reference position pointl iBy passingf ijThe function obtains the radio frequency fingerprint corresponding to each beacon node as
Figure DEST_PATH_IMAGE036
And then the radio frequency map is used for a weighted K neighbor algorithm to estimate the position of the unknown node.
CN201310505864.4A 2013-10-24 2013-10-24 A kind of radio frequency map adaptive location method based on sub-clustering mechanism and robustness regression Expired - Fee Related CN103533647B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310505864.4A CN103533647B (en) 2013-10-24 2013-10-24 A kind of radio frequency map adaptive location method based on sub-clustering mechanism and robustness regression

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310505864.4A CN103533647B (en) 2013-10-24 2013-10-24 A kind of radio frequency map adaptive location method based on sub-clustering mechanism and robustness regression

Publications (2)

Publication Number Publication Date
CN103533647A true CN103533647A (en) 2014-01-22
CN103533647B CN103533647B (en) 2016-05-18

Family

ID=49935221

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310505864.4A Expired - Fee Related CN103533647B (en) 2013-10-24 2013-10-24 A kind of radio frequency map adaptive location method based on sub-clustering mechanism and robustness regression

Country Status (1)

Country Link
CN (1) CN103533647B (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104185275A (en) * 2014-09-10 2014-12-03 北京航空航天大学 Indoor positioning method based on WLAN
CN104198988A (en) * 2014-06-27 2014-12-10 武汉理工大学 Indoor three-dimensional positioning method based on RSSI
CN105163385A (en) * 2015-08-25 2015-12-16 华南理工大学 Localization algorithm based on sector overlapping area of clustering analysis
WO2016078574A1 (en) * 2014-11-17 2016-05-26 上海鸿研物流技术有限公司 Sorting method for recyclable logistics implements provided with wireless tags
CN106548197A (en) * 2016-10-20 2017-03-29 天津大学 Improved multipath cluster clustering method
CN106804059A (en) * 2017-01-18 2017-06-06 南京邮电大学 Terminal difference removing method in indoor positioning technologies based on signal intensity
CN106950540A (en) * 2017-03-23 2017-07-14 北京理工大学 A kind of indoor orientation method based on residual weighted
CN107360548A (en) * 2017-08-21 2017-11-17 山东师范大学 Indoor article localization method and system based on RFID
CN107635207A (en) * 2017-09-29 2018-01-26 华南师范大学 Mobile node positioning method under a kind of obstacle environment
CN108834047A (en) * 2018-06-08 2018-11-16 武汉理工大学 A kind of AP selection indoor orientation method of path loss model
CN110087179A (en) * 2019-03-26 2019-08-02 深圳先进技术研究院 A kind of indoor positioning control method, system and electronic equipment
CN112698112A (en) * 2021-03-23 2021-04-23 中国人民解放军国防科技大学 Electromagnetic spectrum map construction method and device, computer equipment and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110306354A1 (en) * 2009-12-10 2011-12-15 Massachusetts Institute Of Technology Method and Apparatus for Constructing a User-Generated Geological System
CN102821465A (en) * 2012-09-07 2012-12-12 哈尔滨工业大学 WLAN (Wireless Local Area Network) indoor positioning method based on subregion information entropy gain
CN102932738A (en) * 2012-10-31 2013-02-13 北京交通大学 Improved positioning method of indoor fingerprint based on clustering neural network
US20130225209A1 (en) * 2012-02-24 2013-08-29 Broadcom Corporation Wireless Communication Device Capable of Performing Enhanced Fingerprint Mapping and Location Identification

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110306354A1 (en) * 2009-12-10 2011-12-15 Massachusetts Institute Of Technology Method and Apparatus for Constructing a User-Generated Geological System
US20130225209A1 (en) * 2012-02-24 2013-08-29 Broadcom Corporation Wireless Communication Device Capable of Performing Enhanced Fingerprint Mapping and Location Identification
CN102821465A (en) * 2012-09-07 2012-12-12 哈尔滨工业大学 WLAN (Wireless Local Area Network) indoor positioning method based on subregion information entropy gain
CN102932738A (en) * 2012-10-31 2013-02-13 北京交通大学 Improved positioning method of indoor fingerprint based on clustering neural network

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104198988A (en) * 2014-06-27 2014-12-10 武汉理工大学 Indoor three-dimensional positioning method based on RSSI
CN104185275B (en) * 2014-09-10 2017-11-17 北京航空航天大学 A kind of indoor orientation method based on WLAN
CN104185275A (en) * 2014-09-10 2014-12-03 北京航空航天大学 Indoor positioning method based on WLAN
WO2016078574A1 (en) * 2014-11-17 2016-05-26 上海鸿研物流技术有限公司 Sorting method for recyclable logistics implements provided with wireless tags
CN105163385B (en) * 2015-08-25 2019-01-29 华南理工大学 A kind of localization method based on fan-shaped overlapping region clustering
CN105163385A (en) * 2015-08-25 2015-12-16 华南理工大学 Localization algorithm based on sector overlapping area of clustering analysis
CN106548197A (en) * 2016-10-20 2017-03-29 天津大学 Improved multipath cluster clustering method
CN106804059A (en) * 2017-01-18 2017-06-06 南京邮电大学 Terminal difference removing method in indoor positioning technologies based on signal intensity
CN106950540A (en) * 2017-03-23 2017-07-14 北京理工大学 A kind of indoor orientation method based on residual weighted
CN107360548A (en) * 2017-08-21 2017-11-17 山东师范大学 Indoor article localization method and system based on RFID
CN107635207A (en) * 2017-09-29 2018-01-26 华南师范大学 Mobile node positioning method under a kind of obstacle environment
CN107635207B (en) * 2017-09-29 2020-01-14 华南师范大学 Mobile node positioning method in obstacle environment
CN108834047A (en) * 2018-06-08 2018-11-16 武汉理工大学 A kind of AP selection indoor orientation method of path loss model
CN108834047B (en) * 2018-06-08 2020-09-15 武汉理工大学 AP selection indoor positioning method of path loss model
CN110087179A (en) * 2019-03-26 2019-08-02 深圳先进技术研究院 A kind of indoor positioning control method, system and electronic equipment
CN112698112A (en) * 2021-03-23 2021-04-23 中国人民解放军国防科技大学 Electromagnetic spectrum map construction method and device, computer equipment and storage medium
CN112698112B (en) * 2021-03-23 2021-06-01 中国人民解放军国防科技大学 Electromagnetic spectrum map construction method and device, computer equipment and storage medium

Also Published As

Publication number Publication date
CN103533647B (en) 2016-05-18

Similar Documents

Publication Publication Date Title
CN103533647A (en) Radio frequency map self-adaption positioning method based on clustering mechanism and robust regression
KR102116824B1 (en) Positioning system based on deep learnin and construction method thereof
CN104853317B (en) The structure and update method of fingerprint base in a kind of WiFi indoor positionings
CN105704652B (en) Fingerprint base acquisition and optimization method in a kind of positioning of WLAN/ bluetooth
CN108919177B (en) Positioning map construction method based on virtual information source estimation and track correction
CN108709557B (en) Indoor map generation method based on multi-user track fitting
CN103747419B (en) A kind of indoor orientation method based on signal strength difference and dynamic linear interpolation
CN104581943B (en) Node positioning method for Distributed Wireless Sensor Networks
CN109379711B (en) positioning method
Zhou et al. Indoor fingerprint localization based on fuzzy c-means clustering
CN104507159A (en) A method for hybrid indoor positioning based on WiFi (Wireless Fidelity) received signal strength
Ning et al. Outdoor location estimation using received signal strength-based fingerprinting
CN102665277A (en) Method for locating nodes in wireless sensor network (WSN)
CN110213003A (en) A kind of wireless channel large-scale fading modeling method and device
CN105101090B (en) A kind of node positioning method of environmental monitoring wireless sense network
CN103929717A (en) Wireless sensor network positioning method based on weight Voronoi diagrams
CN109842935A (en) A kind of weighting DV-HOP localization method based on mixing SMPSO optimization
CN111757258A (en) Self-adaptive positioning fingerprint database construction method under complex indoor signal environment
CN117241215A (en) Wireless sensor network distributed node cooperative positioning method based on graph neural network
Sadhukhan et al. A novel weighted fusion based efficient clustering for improved wi-fi fingerprint indoor positioning
Biswas et al. New RSSI-fingerprinting-based smartphone localization system for indoor environments
Labinghisa et al. Improved indoor localization system based on virtual access points in a Wi-Fi environment by filtering schemes
Zhang et al. A node localization algorithm for wireless sensor network based on improved particle swarm optimization
CN105960011B (en) Indoor objects localization method based on Sensor Network and bayes method
CN109640253B (en) Mobile robot positioning method

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20160518

Termination date: 20181024

CF01 Termination of patent right due to non-payment of annual fee