CN107820314B - Dwknn position fingerprint positioning method based on AP selection - Google Patents

Dwknn position fingerprint positioning method based on AP selection Download PDF

Info

Publication number
CN107820314B
CN107820314B CN201711122277.1A CN201711122277A CN107820314B CN 107820314 B CN107820314 B CN 107820314B CN 201711122277 A CN201711122277 A CN 201711122277A CN 107820314 B CN107820314 B CN 107820314B
Authority
CN
China
Prior art keywords
positioning
aps
rss
similarity
value
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201711122277.1A
Other languages
Chinese (zh)
Other versions
CN107820314A (en
Inventor
卢先领
施涛涛
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.)
WUXI HUAYIDE CONSTANT TEMPERATURE FITTINGS Co.,Ltd.
Original Assignee
Jiangnan 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 Jiangnan University filed Critical Jiangnan University
Priority to CN201711122277.1A priority Critical patent/CN107820314B/en
Publication of CN107820314A publication Critical patent/CN107820314A/en
Application granted granted Critical
Publication of CN107820314B publication Critical patent/CN107820314B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
    • H04W64/006—Locating users or terminals or network equipment for network management purposes, e.g. mobility management with additional information processing, e.g. for direction or speed determination
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S11/00—Systems for determining distance or velocity not using reflection or reradiation
    • G01S11/02—Systems for determining distance or velocity not using reflection or reradiation using radio waves
    • G01S11/06—Systems for determining distance or velocity not using reflection or reradiation using radio waves using intensity measurements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations

Abstract

The invention provides a Dwknn position fingerprint positioning algorithm based on AP selection, which comprises the following steps: step one, excessive APs in the positioning process can cause the increase of positioning time and the reduction of positioning accuracy; selecting the APs by using information entropy and aroma theorem according to different expressive forces of the APs in the positioning areas, and rejecting the APs with less information content and greater similarity; secondly, the Wknn algorithm uses the similarity of k reference points as weight to realize accurate positioning, but the k value is fixed and unchanged, so that the similarity of a far reference point is easy to be used as the weight, and errors are easy to cause; a dynamic k value weighting positioning algorithm is provided, and experimental results show that the influence of a far reference point on positioning can be weakened. The invention reduces the influence of interference signals, and simulation results show that the provided algorithm can effectively improve the positioning efficiency and the positioning precision.

Description

Dwknn position fingerprint positioning method based on AP selection
Technical Field
The invention relates to the technical field of indoor fingerprint positioning, in particular to a Dwknn position fingerprint positioning method based on AP selection.
Background
With the development of the internet of things technology, location services have become one of the key requirements. On the geographical location division, location services are divided into outdoor positioning and indoor positioning technologies. Global Positioning System (GPS) and the beidou System are widely used in outdoor Positioning, and have good effects, but the performance of the GPS and the beidou System is seriously affected by the complicated and changeable indoor environment. However, the importance of indoor positioning is becoming apparent in particular places. Therefore, the development of the indoor positioning technology has wide application prospect and important significance.
Indoor positioning algorithms can be divided into three categories: triangulation location, proximity location, location fingerprint location. In the case of Line of Sight (LOS), triangulation location methods have higher location accuracy than proximity measurement location methods, but are sensitive to the heterogeneity and synchronicity of hardware. The installation process is complex, is easily limited by field conditions, and the positioning accuracy is related to the position and the number of the equipment. Therefore, from the technical and application perspectives, researchers propose a position fingerprint positioning method with low price and high precision aiming at the defects of the first two types of positioning methods. The location fingerprinting algorithm is divided into two parts: an off-line phase and a positioning phase. In the off-line stage, Signal Strength values are collected in a positioning area mainly by using handheld terminal equipment, and a database of Received Signal Strength (RSS) feature vectors and position relations is established. In the positioning stage, positioning is mainly performed through a matching algorithm, and typical matching algorithms include a nearest neighbor method, a K neighbor method, a weighted K neighbor method and the like. However, the fingerprint library it creates contains a large amount of garbage.
In contrast, M dash selects APs in consideration of the entropy and stability of AP information, which can remove some redundancy, but does not consider the similarity between APs, and is easy to store similar APs. The L Elina selects the AP by using the lowest similarity among the APs, and the established RSS database has low similarity of the APs, but the L Elina may delete the APs containing a large amount of useful information, so that the L Elina is not beneficial to positioning at the present stage. H Zou comprehensively considers the similarity between the information content of each AP and the AP, but the H Zou uses a fixed k value weighted neighbor algorithm, so that a remote RP point is easy to use, and the positioning accuracy is reduced.
The terms referred to herein are:
AP: access Point, refers to a base station;
RP: a reference point;
RSS: the received signal strength.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a Dwknn position fingerprint positioning method based on AP selection, which is called APs-Dwknn algorithm for short. The technical scheme adopted by the invention is as follows:
a Dwknn position fingerprint positioning method based on AP selection comprises the following steps:
step one, selecting the APs by using information entropy and aroma theorem according to different expressive forces of the APs in the positioning areas, and rejecting the APs with less information content and greater similarity;
step two, adopting a dynamic k value weighting positioning algorithm to weaken the influence of a far reference point on positioning:
the first step specifically comprises the following steps:
step S1, establishing fingerprint database, mainly realizing the uniqueness of matching relation with RSS and position; setting a positioning area to have D RP points, and acquiring n APs one by one at each RP point to obtain the RSS value of the ith RP point as follows:
Figure GDA0002411716740000021
wherein
Figure GDA0002411716740000022
Representing that the RSS average value is acquired for the jth AP m times at the i RP points;
the two-dimensional spatial coordinates of the ith RP point can be expressed as:
Gi＝(xi,yi) (2)
step S2, storing the acquired offline location information into RSS fingerprint database, which is shown below
Figure GDA0002411716740000023
The corresponding two-dimensional space coordinates are as follows:
Figure GDA0002411716740000024
step S3, excessive APs in the positioning process may cause an increase in positioning time and a decrease in positioning accuracy; selecting the APs by using information entropy and aroma theorem according to different expressive forces of the APs in the positioning areas, and rejecting the APs with less information content and greater similarity;
Figure GDA0002411716740000025
p(Gg|APj＝v)＝p(APj＝v|Gg)p(Gg)/p(APj＝v) (6)
Figure GDA0002411716740000026
IG(APi)＝H(G)-H(G|APj) (8)
wherein n istIndicating the number of reference points, G, contained in the tth positioning areagDenotes the g grid, v denotes the received RSS value of the AP, APjLine j, H (G | AP) representing the RSS fingerprint libraryi) Is APiConditional information entropy of (1), p (G)g|APjV) at APjConditional information entropy of v, h (g) is position information entropy, IG (AP)i) The information gain carried by the ith AP; to ensure that the similarity between APs is low, the similarity calculation is performed using the fragrance concentration theorem as follows:
Figure GDA0002411716740000031
Figure GDA0002411716740000032
wherein, diffi,jRepresenting the similarity between the ith AP and the jth AP;
Figure GDA0002411716740000033
representing the RSS mean of the ith AP; siRepresents the sum of similarity values of the ith AP and all the other APs;
MAX(W(AP)＝S+IG) (11)
wherein, W (AP) represents that the information content contained in each AP and the similarity value with other APs are comprehensively considered, and the APs with the lowest similarity and the largest information entropy are selected to be combined into a new fingerprint database;
the second step specifically comprises;
a dynamic k-value weighted positioning algorithm, as shown below;
sim(i)＝RSS_test-RSS(i) (12)
Figure GDA0002411716740000034
RSS_test_choose＝(RSS_test_sum*a)/D (14)
wherein sim represents a set of similarity values between the terminal to be positioned and all RPs, and similarity of k RP points with sim < RSS _ test _ choose is selected as a weight value for positioning, as shown in the following formula:
Figure GDA0002411716740000035
wherein RSS _ test represents that the terminal to be positioned receives RSS values of all APs (where RSS _ test is vector representation), RSS (i) represents the RSS value received by the ith reference point in the new fingerprint library, a is a coefficient,
Figure GDA0002411716740000036
and indicating the final positioning result of the terminal to be positioned.
The invention has the advantages that: in the APs-Dwknn algorithm, in an off-line stage, the algorithm uses an information entropy and aroma similarity measurement algorithm to select the AP from an acquired database, so that redundant APs are removed; in the positioning stage, the algorithm uses the dynamic k value to eliminate the influence of the long-distance RP on the positioning. The algorithm achieves the positioning effect that the average error is only 0.997m, the maximum error is 2.762m and the minimum error is 0.054m in a 12 m-7.8 m positioning area.
Drawings
FIG. 1 is a flow chart of APs-Dwknn positioning according to the present invention.
Fig. 2 is a schematic diagram of AP distribution according to the present invention.
FIG. 3 is a diagram of positioning errors for different AP numbers according to the present invention.
FIG. 4 is a diagram of the test point positioning error of the present invention.
Detailed Description
The invention is further illustrated by the following specific figures and examples.
The invention provides an AP selection-based Dwknn position fingerprint positioning method (Access Point selection Dynamic weighted K-near Neighbor), which is abbreviated as APs-Dwknn;
in an off-line stage, firstly carrying out equalization processing on the strength of the collected received signals by the algorithm, then carrying out grid method division, carrying out learning on AP signals by using information entropy and Shannon theorem, and finally selecting an optimal AP set with the least redundancy as a new fingerprint library; in the positioning stage, the similarity between the reference point and the point to be positioned is calculated, and k reference points with the maximum similarity are dynamically searched by utilizing a Dwknn algorithm to realize positioning. Compared with the existing positioning algorithm without deleting redundant information, the APs-Dwknn improves the data processing efficiency and reduces the influence of interference signals, and simulation results show that the provided algorithm can effectively improve the positioning efficiency and the positioning precision.
The algorithm mainly comprises two steps:
step one, excessive APs in the positioning process can cause the increase of positioning time and the reduction of positioning accuracy; selecting the APs by using information entropy and aroma theorem according to different expressive forces of the APs in the positioning areas, and rejecting the APs with less information content and greater similarity;
secondly, the Wknn algorithm uses the similarity of k reference points as weight to realize accurate positioning, but the k value is fixed and unchanged, so that the similarity of a far reference point is easy to be used as the weight, and errors are easy to cause; the dynamic k value weighting positioning algorithm is provided, and experimental results show that the influence of a far reference point on positioning can be weakened;
the first step specifically comprises the following steps:
step S1, establishing fingerprint database, mainly realizing the uniqueness of matching relation with RSS and position; setting a positioning area to have D RP points, and acquiring n APs one by one at each RP point to obtain the RSS value of the ith RP point as follows:
Figure GDA0002411716740000041
wherein
Figure GDA0002411716740000042
Representing that the RSS average value is acquired for the jth AP m times at the i RP points;
the two-dimensional spatial coordinates of the ith RP point can be expressed as:
Gi＝(xi,yi) (2)
step S2, storing the acquired offline location information into RSS fingerprint database, which is shown below
Figure GDA0002411716740000043
The corresponding two-dimensional space coordinates are as follows:
Figure GDA0002411716740000044
step S3, excessive APs in the positioning process may cause an increase in positioning time and a decrease in positioning accuracy; selecting the APs by using information entropy and aroma theorem according to different expressive forces of the APs in the positioning areas, and rejecting the APs with less information content and greater similarity;
Figure GDA0002411716740000051
p(Gg|APj＝v)＝p(APj＝v|Gg)p(Gg)/p(APj＝v) (6)
Figure GDA0002411716740000052
IG(APi)＝H(G)-H(G|APj) (8)
wherein n istIndicating the number of reference points, G, contained in the tth positioning areagDenotes the g grid, v denotes the received RSS value of the AP, APjDenotes the jth AP, H (G | AP)i) Is APiConditions of (2)Entropy of information, p (G)g|APjV) at APjConditional information entropy of v, h (g) is position information entropy, IG (AP)i) The information gain carried by the ith AP; to ensure that the similarity between APs is low, the similarity calculation is performed using the fragrance concentration theorem as follows:
Figure GDA0002411716740000053
Figure GDA0002411716740000058
wherein, diffi,jRepresenting the similarity between the ith AP and the jth AP;
Figure GDA0002411716740000054
representing the RSS mean of the ith AP; siRepresents the sum of similarity values of the ith AP and all the other APs;
MAX(W(AP)＝S+IG) (11)
wherein, W (AP) represents that the information content contained in each AP and the similarity value with other APs are comprehensively considered, and the APs with the lowest similarity and the largest information entropy are selected to be combined into a new fingerprint database.
The second step specifically comprises:
the Wknn algorithm uses the similarity of k reference points as weight to realize accurate positioning, but the k value is fixed and unchanged, so that the similarity of a far reference point is easy to be used as the weight, and errors are easy to cause; therefore, a dynamic k value weighting positioning algorithm is provided, and the influence of a far reference point on positioning is reduced to a certain extent;
sim(i)＝RSS_test-RSS(i) (12)
Figure GDA0002411716740000055
RSS_test_choose＝(RSS_test_sum*a)/D (14)
wherein sim represents a set of similarity values between the terminal to be positioned and all RPs, and similarity of k RP points with sim < RSS _ test _ choose is selected as a weight value for positioning, as shown in the following formula:
Figure GDA0002411716740000056
wherein, RSS _ test represents that the terminal to be positioned receives RSS values of all APs (the RSS _ test is represented by a vector), RSS (i) represents an RSS value received by the ith reference point in a new fingerprint database, a value is a coefficient, and the a value is obtained by calculating through an off-line fingerprint database by adopting an empirical value,
Figure GDA0002411716740000057
and indicating the final positioning result of the terminal to be positioned.
The experimental site is selected in a classroom of Internet of things engineering college C529 of south Jiangnan university, 9 AP points are arranged in the classroom, one AP is placed in the classroom C529 for more truly restoring the positioning environment, the rest APs are placed in other rooms, and obstacles such as corridors, walls, glass, iron doors and the like are arranged in the middle of the classroom. The data acquisition uses open source signal receiving software inSSIDer, which has good information acquisition capacity and meets the requirements of experiments. Where 60 reference points, 11 test points, were collected.
The algorithm compared to the AP-selected reconstructed fingerprint library proposed herein is: test one (similarity and stability between APs), test two (logarithmic dissimilarity), test three (entropy), Max RSS, mean maximum RSS, and random selection (this is the unpopulated location fingerprint library to ensure the normality of the experiment). When an AP is tested for selection, the information entropy weight is 0.3, and the stability weight is 0.7, the effect is optimal. The results of the experiment are shown in FIG. 3.
As can be seen from fig. 3, the positioning accuracy is improved to some extent as the number of reference points AP increases, which indicates that the number of APs is one of the factors affecting the positioning accuracy. However, when the number of APs reaches a certain number, the positioning accuracy thereof does not change much. The similarity between the APs is not considered in the second test and the third test, so that the position fingerprint database established by the second test and the third test is low in positioning accuracy and large in fluctuation. It can be seen that the algorithm proposed herein has an error after the 5 th AP which is already close to the positioning error of the fingerprint library without eliminating the AP, and even the positioning accuracy of the algorithm proposed herein is 1.23m when 6 APs are included, while the (randomly selected) accuracy of the fingerprint positioning method without eliminating the AP is 1.28 m. Illustrating the performance of the AP selection algorithm herein over other algorithms.
To verify the performance of the weighted dynamic k-nearest neighbor algorithm herein, verification of its performance is performed as follows. To verify overall performance, the comparison algorithm is divided herein into Wknn and Dwknn without AP selection and Wknn and Dwknn with AP selection. Through experiments, in the case of AP selection, a is 0.57; in the case where no AP selection is made, a is selected to be 0.5 herein.
The following table shows different performance comparisons for different positioning algorithms
Figure GDA0002411716740000061
In FIG. 4 and the above table, Wknn and Dwknn indicate the use of databases without AP selection for localization, and APs-Wknn and APs-Dwknn indicate the use of databases after AP selection for localization. Through performance comparison, the average error, the maximum error, the minimum error and the percentage of the error smaller than 1m of the algorithm provided by the method are all superior to those of the other three methods. From locating the AP-selected database and the non-AP-selected database using Wknn and Dwknn, the AP-selected data location accuracy is improved. The Dwknn proposed herein has a higher positioning accuracy when used with a database selected for AP or a database not selected for AP.
For the problem that the off-line database contains a large number of interference signals, the APs-Dwknn algorithm is proposed herein. In an off-line stage, the algorithm uses an information entropy and aroma similarity algorithm to select the AP from the collected database, and redundant APs are removed. In the positioning stage, the algorithm uses the dynamic k value to eliminate the influence of the long-distance RP on the positioning. The algorithm in the text is used for achieving the positioning effect that the average error is only 0.997m, the maximum error is 2.762m, and the minimum error is 0.054m in a 12 m-7.8 m positioning area.

Claims (1)

1. A Dwknn position fingerprint positioning method based on AP selection is characterized by comprising the following steps:
step one, selecting the APs by using information entropy and aroma theorem according to different expressive forces of the APs in the positioning areas, and rejecting the APs with less information content and greater similarity;
step two, adopting a dynamic k value weighting positioning algorithm to weaken the influence of a far reference point on positioning:
the first step specifically comprises the following steps:
step S1, establishing fingerprint database, mainly realizing the uniqueness of matching relation with RSS and position; setting a positioning area to have D RP points, and acquiring n APs one by one at each RP point to obtain the RSS value of the ith RP point as follows:
Figure FDA0002411716730000011
wherein
Figure FDA0002411716730000012
Representing that the RSS average value is acquired for the jth AP m times at the i RP points;
the two-dimensional spatial coordinates of the ith RP point can be expressed as:
Gi＝(xi,yi) (2)
step S2, storing the acquired offline location information into RSS fingerprint database, which is shown below
Figure FDA0002411716730000013
The corresponding two-dimensional space coordinates are as follows:
Figure FDA0002411716730000014
step S3, excessive APs in the positioning process may cause an increase in positioning time and a decrease in positioning accuracy; selecting the APs by using information entropy and aroma theorem according to different expressive forces of the APs in the positioning areas, and rejecting the APs with less information content and greater similarity;
Figure FDA0002411716730000015
p(Gg|APj＝v)＝p(APj＝v|Gg)p(Gg)/p(APj＝v) (6)
Figure FDA0002411716730000016
IG(APi)＝H(G)-H(G|APj) (8)
wherein n istIndicating the number of reference points, G, contained in the tth positioning areagDenotes the g grid, v denotes the received RSS value of the AP, APjLine j, H (G | AP) representing the RSS fingerprint libraryi) Is APiConditional information entropy of (1), p (G)g|APjV) at APjConditional information entropy of v, h (g) is position information entropy, IG (AP)i) The information gain carried by the ith AP; to ensure that the similarity between APs is low, the similarity calculation is performed using the fragrance concentration theorem as follows:
Figure FDA0002411716730000021
wherein, diffi,jRepresenting the similarity between the ith AP and the jth AP;
Figure FDA0002411716730000023
representing the RSS mean of the ith AP; siRepresents the sum of similarity values of the ith AP and all the other APs;
MAX(W(AP)＝S+IG) (11)
wherein, W (AP) represents that the information content contained in each AP and the similarity value with other APs are comprehensively considered, and the APs with the lowest similarity and the largest information entropy are selected to be combined into a new fingerprint database;
the second step specifically comprises;
a dynamic k-value weighted positioning algorithm, as shown below;
sim(i)＝RSS_test-RSS(i) (12)
Figure FDA0002411716730000024
RSS_test_choose＝(RSS_test_sum*a)/D (14)
wherein sim represents a set of similarity values between the terminal to be positioned and all RPs, and similarity of k RP points with sim < RSS _ test _ choose is selected as a weight value for positioning, as shown in the following formula:
Figure FDA0002411716730000025
wherein RSS _ test represents that the terminal to be positioned receives RSS values of all APs (where RSS _ test is vector representation), RSS (i) represents the RSS value received by the ith reference point in the new fingerprint library, a is a coefficient,
Figure FDA0002411716730000026
and indicating the final positioning result of the terminal to be positioned.
CN201711122277.1A 2017-11-14 2017-11-14 Dwknn position fingerprint positioning method based on AP selection Active CN107820314B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711122277.1A CN107820314B (en) 2017-11-14 2017-11-14 Dwknn position fingerprint positioning method based on AP selection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711122277.1A CN107820314B (en) 2017-11-14 2017-11-14 Dwknn position fingerprint positioning method based on AP selection

Publications (2)

Publication Number Publication Date
CN107820314A CN107820314A (en) 2018-03-20
CN107820314B true CN107820314B (en) 2020-05-05

Family

ID=61609027

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711122277.1A Active CN107820314B (en) 2017-11-14 2017-11-14 Dwknn position fingerprint positioning method based on AP selection

Country Status (1)

Country Link
CN (1) CN107820314B (en)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108519579B (en) * 2018-03-29 2021-11-16 吴帮 WiFi fingerprint positioning method for analyzing optimal AP based on interval overlapping degree
CN108712714B (en) * 2018-04-02 2020-05-22 北京邮电大学 Method and device for selecting AP (access point) in indoor WLAN (wireless local area network) fingerprint positioning
CN110493867B (en) * 2019-06-27 2020-12-22 湖南大学 Wireless indoor positioning method for signal selection and position correction
CN111757249B (en) * 2020-05-21 2022-05-03 重庆邮电大学 Multi-feature fuzzy mapping access point optimization method based on information gain ratio
CN111612767B (en) * 2020-05-21 2022-10-28 西南交通大学 Vision-based precision structural part identification method
CN113055814B (en) * 2021-02-05 2023-03-21 江苏大学 Indoor positioning method for improving WKNN
CN114353787A (en) * 2021-12-06 2022-04-15 理大产学研基地(深圳)有限公司 Multi-source fusion positioning method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101980550A (en) * 2004-10-29 2011-02-23 探空气球无线公司 Method for locating user device and server system
CN103476115A (en) * 2013-09-22 2013-12-25 中国地质大学（武汉） Method for WiFi fingerprint positioning based on AP set similarity
WO2016086994A1 (en) * 2014-12-04 2016-06-09 Here Global B.V. Supporting positioning quality assurance
CN106792522A (en) * 2016-12-09 2017-05-31 北京羲和科技有限公司 A kind of fingerprint base localization method and system based on access point AP

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101980550A (en) * 2004-10-29 2011-02-23 探空气球无线公司 Method for locating user device and server system
CN103476115A (en) * 2013-09-22 2013-12-25 中国地质大学（武汉） Method for WiFi fingerprint positioning based on AP set similarity
WO2016086994A1 (en) * 2014-12-04 2016-06-09 Here Global B.V. Supporting positioning quality assurance
CN106792522A (en) * 2016-12-09 2017-05-31 北京羲和科技有限公司 A kind of fingerprint base localization method and system based on access point AP

Also Published As

Publication number Publication date
CN107820314A (en) 2018-03-20

Similar Documents

Publication Publication Date Title
CN107820314B (en) Dwknn position fingerprint positioning method based on AP selection
Zuo et al. Multi-phase fingerprint map based on interpolation for indoor localization using iBeacons
CN107318084B (en) Fingerprint positioning method and device based on optimal similarity
Shin et al. Enhanced weighted K-nearest neighbor algorithm for indoor Wi-Fi positioning systems
CN104796984B (en) Base station positioning method and device
CN104703143A (en) Indoor positioning method based on WIFI signal strength
CN109195110B (en) Indoor positioning method based on hierarchical clustering technology and online extreme learning machine
CN106658704A (en) Positioning method and system of starting point of indoor positioning
Fang et al. Application of an improved K nearest neighbor algorithm in WiFi indoor positioning
CN109348416B (en) Fingerprint indoor positioning method based on binary k-means
Ding et al. AP weighted multiple matching nearest neighbors approach for fingerprint-based indoor localization
CN107677989A (en) A kind of indoor location localization method that RSSI removal noises are carried out based on RSSI maximums
CN109145225B (en) Data processing method and device
CN110850363A (en) Method for carrying out dynamic filtering optimization based on real-time positioning track data
US20170004400A1 (en) A genetic method for the acquisition and tracking of time varying signals
Dong et al. Dealing with insufficient location fingerprints in Wi-Fi based indoor location fingerprinting
CN115100376A (en) Electromagnetic spectrum map implementation method based on improved inverse distance interpolation method
CN110011838B (en) Real-time tracking method for PageRank value of dynamic network
CN112526572B (en) Network switching method and positioning system for indoor and outdoor seamless navigation
CN104023392B (en) The method and apparatus for determining the position of WAP
CN109561384B (en) Wireless sensor network node positioning method under composite noise condition
Xia et al. A navigation satellite selection method based on Tabu search artificial bee colony algorithm
US10191928B2 (en) Planar graph generation device and method
CN113970762A (en) Method and system for positioning multistage interference source
Elbakly et al. A calibration-free RF localization system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20210901

Address after: 214161 No. 18, Tianzhu Road, Hudai Town, Binhu District, Wuxi City, Jiangsu Province

Patentee after: WUXI HUAYIDE CONSTANT TEMPERATURE FITTINGS Co.,Ltd.

Address before: No. 1800 road 214122 Jiangsu Lihu Binhu District City of Wuxi Province

Patentee before: Jiangnan University

TR01 Transfer of patent right