CN112188385A - Indoor positioning method for narrowing positioning area based on AP sequence - Google Patents
Indoor positioning method for narrowing positioning area based on AP sequence Download PDFInfo
- Publication number
- CN112188385A CN112188385A CN201910587280.3A CN201910587280A CN112188385A CN 112188385 A CN112188385 A CN 112188385A CN 201910587280 A CN201910587280 A CN 201910587280A CN 112188385 A CN112188385 A CN 112188385A
- Authority
- CN
- China
- Prior art keywords
- positioning
- sequence
- aps
- sub
- fingerprint
- 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.)
- Pending
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
- 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- 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
-
- 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
Abstract
The invention relates to an indoor positioning method for reducing a positioning area based on an AP (signal access point) sequence, which belongs to the field of wireless communication and is realized in an indoor positioning system based on WiFi (wireless fidelity) signal strength. The method comprises the steps of processing an off-line stage data set, selecting an on-line fingerprint database and carrying out a soft-constraint WKNN positioning algorithm. Firstly, constructing a stable AP sequence by utilizing signal intensity sequencing, and establishing a plurality of off-line sub-fingerprint databases; and in the positioning stage, signal strength information sent by a user is constructed into an AP sequence by utilizing the proposed maximum signal difference principle, a sub-fingerprint library is matched, a positioning area is reduced in the sub-fingerprint library, and finally a positioning algorithm estimates the position. By reducing the positioning area, the positioning algorithm only needs to traverse reference points in the area, so that the problem of errors caused by inconsistency of signal space distance and actual distance in the traditional positioning based on signal intensity is effectively solved, and the positioning speed and precision are obviously improved. Meanwhile, the AP sequence is utilized to construct a fingerprint database, and the anti-interference capability of the positioning system is improved.
Description
Technical Field
The invention relates to an indoor positioning method for reducing a positioning area based on an AP (signal access point) sequence, which processes a large amount of data, extracts effective data characteristics and reduces the positioning area, belongs to the field of wireless communication and the technical field of data mining classification, and is applied to the field of indoor positioning.
Background
In recent years, with the development of radio communication technology, multiple information fusion technology, and internet of things technology, commercial marketing, medical emergency, industrial personnel and warehouse management, virtual games, and the like require precise location services, and thus the demand for precise location in an indoor environment is sharply increasing. Indoor positioning based on WiFi is easy to deploy and wide in coverage range, and is an ideal method in indoor positioning. However, WiFi signals are susceptible to multipath and environmental interference during propagation, different orientations, blocking and movement of human body when receiving signals, and interference from other devices, so the positioning system has poor interference resistance; when determining the final position of the point to be located, a common method is a KNN (K nearest neighbor) positioning algorithm, but the similarity of signal strength (RSSI) is only considered during positioning, and the actual spatial position is not considered, so that a large error is caused during position estimation, and meanwhile, the calculation of the euclidean distance requires traversing all reference points in a database, so that the positioning speed is seriously reduced.
The fluctuation of a general stable Wi-Fi signal source does not exceed 10dB, so that the magnitude sequence of RSSI values from a plurality of APs measured by different devices at the same position tends to be stable, and data with the magnitude sequence is an AP sequence. A soft-constrained WKNN algorithm (SRL-KNNs) is a KNN algorithm related to a current estimated position and a last estimated position, a penalty function of a reference point is added on the basis of a traditional calculation formula for calculating Euclidean distance, a possible position of a to-be-positioned point is limited in a circular area with the last positioned position as a circle center and the maximum moving distance as a radius in proportion, and the positioning precision is greatly improved.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an indoor positioning method for narrowing a positioning area based on an AP sequence.
The technical scheme of the invention comprises the following steps:
step 1: taking k (k < n) APs from the detected n different APs to form a C (n, k) sub-fingerprint database;
step 2: after receiving real-time data sent by a user, firstly selecting k representative APs by using a maximum difference principle, and firstly selecting a sub-fingerprint database;
and step 3: and selecting points which are the same as the AP sequence to be positioned from the sub-fingerprint database X, reducing the positioning area where the unknown position is located, completing coarse positioning, and completing accurate positioning and determining the final position by utilizing SRL-KNNs in the reduced area.
Compared with the prior art, the invention has the beneficial effects that:
the AP sequence is utilized to construct a fingerprint database, and the magnitude sequence of the signal intensity is utilized to replace the unstable signal intensity, so that the errors caused by signal fluctuation and equipment diversity are reduced; by reducing the positioning area, the error caused by the inconsistency of the signal space distance and the actual distance is effectively solved, and meanwhile, the Euclidean distance is calculated only by traversing the reference point in the reduced area; and the final position is determined by using the SRL-KNNs, and the accuracy is higher when the positioning point is related to the last positioning point in the track positioning.
The overall positioning precision is higher, the positioning speed is higher, and the positioning system can be more stable.
Drawings
FIG. 1 is an overall flow chart of the invention;
FIG. 2 is a flow chart of a maximum difference algorithm;
FIG. 3 is a flowchart of a reduced positioning area process;
FIG. 4 is a CDF comparison chart of positioning with different numbers of selected APs;
FIG. 5 is a comparison graph of the selection of 4 APs versus no AP selection at a single point location;
FIG. 6 is a CDF graph comparing different positioning algorithms.
Detailed Description
The present invention will be described in further detail with reference to the following detailed description, wherein the preferred embodiments are only for illustrating the technical solutions of the present invention, and the present invention is not limited thereto.
The general framework schematic diagram of the invention is shown in fig. 1, firstly, the obtained WiFi signal intensity is processed through Gaussian filtering, and a plurality of sub-fingerprint databases are established by an original fingerprint database through an AP selection algorithm; and then after fingerprint data of the user is acquired, selecting a representative AP sequence by using the maximum signal intensity difference, matching the AP sequence with the AP sequence in the fingerprint library, determining the region where the point to be positioned is located, further reducing the range of the positioning region, completing the first-step positioning, and finally completing the second-step positioning in the selected region by using a KNN algorithm (SRL-KNNs) with soft constraints. The following describes a specific implementation process of the technical solution of the present invention with reference to the accompanying drawings.
1. Data acquisition
A laboratory in an indoor environment is selected as a positioning area, the test area is 20m × 16m, a total of 6 TP-LINK TL-WDR6500 wireless routers are used as Wi-Fi hot spots AP, and an indoor plan view is shown in figure 2. A Lenovo IdeaPad computer is adopted, a C # console program self-compiled by VS2013 is utilized to collect data, and people move around in a laboratory in the collection process. 180 offline test points are arranged, one reference point is arranged at intervals of 1.2m under the condition that no barrier exists, signal intensity data are acquired at each offline reference point for 100 times, the acquired signal intensity data are automatically stored in Excel after being acquired, and the acquired signal intensity data are directly imported into MATLAB for preliminary processing such as filtering, so that the time for sorting the offline data is reduced.
And in the online positioning process, 57 points are selected as to-be-positioned points, each positioning point is acquired for 1s, and each point data is acquired for 10 times. Note that when processing the initial data, it may happen that some AP signals are not received for 1-2 times among 10 times of acquisition, but the signal values received for several times are not low, and then the data cannot be directly rejected, but the signal values not received are set to-100 dB.
2. Establishment of off-line fingerprint database
The method comprises the steps of firstly selecting APs, selecting 4 APs with reliable received signal strength from 6 APs, forming a sub-fingerprint library of which the number is C (6, 4) ═ 15, naming each fingerprint library by using serial numbers of 1-15, sequencing the signal strength of the APs in the sub-fingerprint library, but not directly positioning the sub-fingerprint library as final data at the time, removing the AP value with the signal strength less than-75 dB and the point with the difference between two adjacent signal values in the AP sequence less than +/-5 dB due to the fact that the APs of each reference point are fully combined and each reference point has the best combination mode, and correspondingly updating the geographical location library of the offline reference point. Such a sub-fingerprint library may be used for online localization.
3. On-line positioning
And receiving fingerprint information sent by a user, filtering out APs with RSSI less than-80 dB, and selecting 4 APs from 6 APs so as to make the signal strength difference of the 4 APs large. The RSSI of 6 APs is sorted firstly, the RSSI of adjacent APs is subtracted, the obtained 5 difference values are sorted, the difference values with the minimum difference value are combined into one class, the process is repeated until the difference values are combined into 4 classes, and one representative AP is selected from each class and is arranged into an AP sequence. The algorithm flow chart is shown in fig. 2. It is worth mentioning that the clustering using the K-means algorithm herein is not ideal, and the difference between neighboring APs in the clustering result is not the maximum.
The first step of positioning is the reduction of a positioning area, firstly, a sub-fingerprint library is selected according to the selected AP, reference points in the same AP sequence are selected in the sub-fingerprint library, the area formed by the reference points is the reduced positioning area, if the sub-fingerprint library does not have the same AP sequence, a representative AP of each type is reselected, if all the possibilities of combination are selected, and if no AP sequence which can be matched exists, the position positioning is failed, and an algorithm flow chart is shown in figure 3; and when the final position is determined by the positioning in the second step by using the SRL-KNNs algorithm, assuming that the first position is known, and the value of sigma in the formula is 2m, only traversing the reference point in the reduced area when calculating the Euclidean distance. The SRL-KNNs algorithm is generally described herein.
Assuming that the number of available APs is N, the Euclidean distance formula of the SRL-KNNs algorithm is as follows.
Wherein, theoretically, M is the total number of all reference points in the fingerprint database, here, the number of reference points in the first step coarse positioning area,
i.e. a penalty function for the reference point i, (x)pre，ypre) σ is the maximum distance that the user can move within the sampling time interval Δ t from the last position to l for the last position of the point to be located l. The penalty function obeys a mean value of (x)pre，ypre) The standard deviation is a Gaussian distribution of σ, so the function
The action mode of (1) actually limits the possible positions of the to-be-positioned points in proportion to the circular area which takes the positioning position of the last time as the center of a circle and takes sigma as the radius, and the areas outside the area are not directly removed, but the weight is much smaller than that in the area. The final position is a weight of the nearest K data.
Wherein
Is the modified euclidean distance in equation (2).
4. Performance testing
And (3) selecting the AP, not selecting the AP from the 6 APs with reliable received signal strength, directly positioning by using the original fingerprint, and comparing the positioning by selecting 2-6 APs, wherein the positioning algorithm of the second step adopts a classical WKNN algorithm to determine the final position. The CDF comparison for positioning with different numbers of APs is shown in fig. 4, and the CDF comparison for positioning with 4 APs and a single point without AP selection is shown in fig. 5, and the time consumption for positioning all the points to be positioned is shown in table 1 below.
TABLE 1 time consumption
After 4 APs are selected to process the online data, two algorithms of Bayes and WKNN are respectively adopted during positioning by a positioning algorithm, and the comparison result of the SRL-KNNs algorithm and the Bayes and WKNN algorithms is shown in FIG. 6.
Through comparison between the real coordinates of the points to be positioned and the positioning coordinates of the embodiment, the indoor positioning method for reducing the positioning area based on the AP sequence is proved to realize high-precision indoor positioning, realize an average positioning error of 0.74m, test data with an error of 75% less than 1m and higher positioning speed.
Claims (4)
1. An indoor positioning method for narrowing down a positioning area based on an AP sequence, the method comprising the following steps:
(1) in the off-line fingerprint database processing stage, a plurality of sub-fingerprint databases are established by utilizing an AP selection algorithm and are numbered;
(2) in the online stage, the signal intensity of a user is received, representative APs are selected to form a sequence, and a corresponding sub-fingerprint database is selected;
(3) and selecting fingerprints with the same AP sequence in the sub-fingerprint library, and estimating the final position by using a soft-constrained WKNN algorithm.
2. The method of claim 1, wherein in step (1), the AP selection algorithm selects N APs from all the detected N APs by permutation and combination to construct
A sub-fingerprint library.
3. The method of claim 1, wherein in step (2), the sequence of n representative APs is selected by using the principle of maximum difference, m (m > -n) APs are detected and sorted according to signal strength, signal strength values of adjacent APs are subtracted, and the obtained differences are sorted again, and the differences with minimum difference are combined into one type until the differences are combined into n types.
4. The method of claim 1, wherein in step (3), the final position is estimated within a reduced location area consisting of reference points selected according to the same AP sequence order, and if there is no matching reference point in the representative AP order, the representative AP is reselected from each class.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910587280.3A CN112188385A (en) | 2019-07-02 | 2019-07-02 | Indoor positioning method for narrowing positioning area based on AP sequence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910587280.3A CN112188385A (en) | 2019-07-02 | 2019-07-02 | Indoor positioning method for narrowing positioning area based on AP sequence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112188385A true CN112188385A (en) | 2021-01-05 |
Family
ID=73915469
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910587280.3A Pending CN112188385A (en) | 2019-07-02 | 2019-07-02 | Indoor positioning method for narrowing positioning area based on AP sequence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112188385A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117295158A (en) * | 2023-11-27 | 2023-12-26 | 华润数字科技有限公司 | WiFi positioning method, device, equipment and medium based on fingerprint matching |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103796163A (en) * | 2013-11-03 | 2014-05-14 | 北京工业大学 | Indoor positioning method for receiving signal strength ordering fingerprint |
| CN103901398A (en) * | 2014-04-16 | 2014-07-02 | 山东大学 | Position fingerprint positioning method based on combination ordering classification |
| CN105101406A (en) * | 2015-06-26 | 2015-11-25 | 上海汇纳信息科技股份有限公司 | Wireless intensity based indoor positioning method and system |
| CN105101408A (en) * | 2015-07-23 | 2015-11-25 | 常熟理工学院 | Indoor positioning method based on distributed AP selection strategy |
| CN105578420A (en) * | 2014-10-17 | 2016-05-11 | 深圳航天科技创新研究院 | WLAN based indoor positioning method and device |
| CN106060779A (en) * | 2016-07-18 | 2016-10-26 | 北京方位捷讯科技有限公司 | Fingerprint feature matching method and device |
| 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 |
| CN108174343A (en) * | 2017-11-29 | 2018-06-15 | 全球能源互联网研究院有限公司 | A kind of wireless location method and system towards electric power indoor communications O＆M scenarios |
| WO2018176511A1 (en) * | 2017-03-28 | 2018-10-04 | 华为技术有限公司 | Fingerprint locating method and related device |
| CN109121083A (en) * | 2018-09-25 | 2019-01-01 | 西安电子科技大学 | A kind of indoor orientation method of the fingerprint similarity based on AP sequence |
| WO2019090527A1 (en) * | 2017-11-08 | 2019-05-16 | 上海诺基亚贝尔股份有限公司 | Indoor positioning method and apparatus using reconfigurable antenna |
-
2019
- 2019-07-02 CN CN201910587280.3A patent/CN112188385A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103796163A (en) * | 2013-11-03 | 2014-05-14 | 北京工业大学 | Indoor positioning method for receiving signal strength ordering fingerprint |
| CN103901398A (en) * | 2014-04-16 | 2014-07-02 | 山东大学 | Position fingerprint positioning method based on combination ordering classification |
| CN103901398B (en) * | 2014-04-16 | 2016-08-24 | 山东大学 | A kind of location fingerprint localization method based on combination collating sort |
| CN105578420A (en) * | 2014-10-17 | 2016-05-11 | 深圳航天科技创新研究院 | WLAN based indoor positioning method and device |
| CN105101406A (en) * | 2015-06-26 | 2015-11-25 | 上海汇纳信息科技股份有限公司 | Wireless intensity based indoor positioning method and system |
| CN105101408A (en) * | 2015-07-23 | 2015-11-25 | 常熟理工学院 | Indoor positioning method based on distributed AP selection strategy |
| CN106060779A (en) * | 2016-07-18 | 2016-10-26 | 北京方位捷讯科技有限公司 | Fingerprint feature matching method and device |
| 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 |
| WO2018176511A1 (en) * | 2017-03-28 | 2018-10-04 | 华为技术有限公司 | Fingerprint locating method and related device |
| WO2019090527A1 (en) * | 2017-11-08 | 2019-05-16 | 上海诺基亚贝尔股份有限公司 | Indoor positioning method and apparatus using reconfigurable antenna |
| CN108174343A (en) * | 2017-11-29 | 2018-06-15 | 全球能源互联网研究院有限公司 | A kind of wireless location method and system towards electric power indoor communications O＆M scenarios |
| CN109121083A (en) * | 2018-09-25 | 2019-01-01 | 西安电子科技大学 | A kind of indoor orientation method of the fingerprint similarity based on AP sequence |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117295158A (en) * | 2023-11-27 | 2023-12-26 | 华润数字科技有限公司 | WiFi positioning method, device, equipment and medium based on fingerprint matching |
| CN117295158B (en) * | 2023-11-27 | 2024-02-13 | 华润数字科技有限公司 | WiFi positioning method, device, equipment and medium based on fingerprint matching |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110012428B (en) | Indoor positioning method based on WiFi | |
| CN106851573B (en) | Log path loss model-based joint weighting K nearest neighbor indoor positioning method | |
| CN109672973B (en) | Indoor positioning fusion method based on strongest AP | |
| CN108540929B (en) | Indoor fingerprint positioning method based on RSSI signal strength sequencing | |
| CN101639527B (en) | K nearest fuzzy clustering WLAN indoor locating method based on REE-P | |
| CN102821465B (en) | WLAN (Wireless Local Area Network) indoor positioning method based on subregion information entropy gain | |
| CN107453993B (en) | Target tracking method based on Fisher information matrix and SUKF | |
| CN110366244B (en) | WiFi fingerprint indoor positioning method | |
| CN106412839A (en) | Indoor positioning and tracking method based on secondary partition and gradient fingerprint match | |
| CN112135248B (en) | WIFI fingerprint positioning method based on K-means optimal estimation | |
| CN103987118B (en) | Access point k means clustering methods based on received signal strength signal ZCA albefactions | |
| CN109511085B (en) | UWB fingerprint positioning method based on MeanShift and weighted k nearest neighbor algorithm | |
| CN109348416B (en) | Fingerprint indoor positioning method based on binary k-means | |
| CN114449452B (en) | Wi-Fi indoor positioning method based on CNN-RNN | |
| CN111050282A (en) | Multi-time fuzzy inference weighted KNN positioning method | |
| CN105682048B (en) | Fingerprint positioning method in Subspace Matching room based on PCA under cellular network environment | |
| CN107290714B (en) | Positioning method based on multi-identification fingerprint positioning | |
| CN108834047B (en) | AP selection indoor positioning method of path loss model | |
| Zhong et al. | WiFi indoor localization based on K-means | |
| CN110611952B (en) | Fingerprint matching and positioning method | |
| CN109541537B (en) | Universal indoor positioning method based on ranging | |
| Badawy et al. | Decision tree approach to estimate user location in WLAN based on location fingerprinting | |
| CN108521631B (en) | Mobile AP (access point) identification method for indoor positioning | |
| CN110636437A (en) | Fingerprint pair-based auxiliary positioning method | |
| CN110958583A (en) | Indoor positioning method based on position and strength dual clustering |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210105 |