CN106412841A - Indoor location method based on DBSCAN algorithm - Google Patents
Indoor location method based on DBSCAN algorithm Download PDFInfo
- Publication number
- CN106412841A CN106412841A CN201611059276.2A CN201611059276A CN106412841A CN 106412841 A CN106412841 A CN 106412841A CN 201611059276 A CN201611059276 A CN 201611059276A CN 106412841 A CN106412841 A CN 106412841A
- Authority
- CN
- China
- Prior art keywords
- sample point
- signal strength
- sample
- represent
- signal intensity
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/22—Traffic simulation tools or models
- H04W16/225—Traffic simulation tools or models for indoor or short range network
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The present invention provides an indoor location method based on a DBSCAN algorithm, belonging to the indoor location technology field. The method comprises: installing a signal emitter and a signal collector in a building requiring location, sending the number of the sample points, obtaining the signal intensity sequence of each sample point, performing recoding and numbering of the coordinates and the signal intensity sequence of each sample point, employing the DBSCAN algorithm to build an indoor location model, obtaining the signal intensity sequences of the location points, and obtaining the location coordinates of the points to be located according to the established indoor location model. The indoor location method based on the DBSCAN algorithm employs the unsupervised-learning algorithm to perform modeling and performs classification clustering according to the wireless signal intensity value (RSSI) without artificial grid division to allow the grid division to more correspond with the reality so as to improve the stability and the precision of the indoor location.
Description
Technical field
The invention belongs to indoor positioning technologies field is and in particular to a kind of indoor orientation method based on DBSCAN algorithm.
Background technology
With the development of Internet of Things, the demand of indoor positioning is also grown with each passing day, such as in market, school, office building, doctor
It is required for indoor positioning technologies that resource, personnel are efficiently managed in the indoor environments such as institute, hotel, airport, warehouse,
VR field there is also substantial amounts of demand to position Indoor Location Information of player etc., therefore, how to better meet and increasingly increases
Plus indoor positioning demand, have become as the hot issue in current location technology.
In existing indoor positioning technologies, it is widely used based on the fingerprint positioning method of wireless signal strength (RSSI)
In various indoor locating systems, indoors in environment, the propagation due to wireless signal is easily walked about by temperature, humidity and personnel
Impact, therefore wireless signal strength (RSSI) measured value fluctuation larger;In order to improve stability and the precision of measurement, commonly use
The mode of coupling fingerprint base realizes indoor positioning, has simple shellfish using the common method that coupling fingerprint database realizes indoor positioning
Ye Sifa, k-nearest neighbor (KNN), Bayesian Estimation, neural network etc., these are all the method for supervised learning, it usually needs people
For grid division, these networks are typically divided by Averaging Area, because wireless signal strength (RSSI) is distributed not in practice
Averagely, the place that can lead to wireless signal strength (RSSI) sequence similarity is divided in different grids, multiple neighbouring grids
Signal strength signal intensity (RSSI) sequence is very close, but is finally occurred by the situation of random position to one of grid, leads to positioning knot
Really inaccurate.
Content of the invention
For solving the deficiencies in the prior art, the present invention proposes a kind of indoor orientation method based on DBSCAN algorithm, including
Following steps:
Step 1, in the building of required location, several signal projectors and several signal pickers are installed;
Step 2, the number of setting sample point, determine the coordinate of each sample point, and it are strong to obtain the signal of each sample point
Degree series, that is, each sample point is to the signal strength signal intensity of each signal projector;
Step 3, by the coordinate of each sample point and signal strength signal intensity sequence carry out record numbering;
Step 4, the coordinate according to each sample point after numbering and signal strength signal intensity sequence, set up room using DBSCAN algorithm
Interior location model, comprises the following steps:
Step 4.1, setting density selection parameter, that is, set the sample that the length of neighborhood and each kernel object include at least
The number of point;
Step 4.2, the signal strength signal intensity sequence according to each sample point after numbering and set density selection parameter set
Determine the constraints of kernel object, determine the kernel object of sample point according to constraints, further determined that by kernel object
Clustering cluster;The constraints of described kernel object is:The number of other sample points being comprised in each sample neighborhood of a point
The number of the sample point whether including at least more than or equal to each set kernel object;
Step 4.3, obtain the average signal strength of each clustering cluster, and the coordinate according to each sample point after numbering,
Obtain the mean place coordinate of each clustering cluster, that is, Indoor Locating Model is set up and completed;
Step 5, obtain the signal strength signal intensity sequence in site undetermined, and treat according to setting up the Indoor Locating Model that completes and obtaining this
The elements of a fix of anchor point, specially:Obtain the Euclidean distance in this site undetermined and the average signal strength of each clustering cluster, and
Select the elements of a fix as this site undetermined for the mean place coordinate of the clustering cluster minimum with its Euclidean distance.
The constraints of the kernel object described in step 4.2, formula is as follows:
|Nθ(xj)|≥Minpts (1)
Wherein, NθRepresent the number of samples in the θ neighborhood of certain sample point;θ represents the length of neighborhood;xjRepresent j-th sample
This point, j represents natural number;Minpts represents the number of the sample point that each kernel object includes at least;
Wherein,
Nθ(xj)={ xi∈D|d(xi,xj)≤θ} (2)
Wherein, xiRepresent i-th sample point;I represents natural number;D represent numbering after the coordinate of each sample point and letter
The data set of number sequence of intensity;d(xi,xj) represent xiWith xjEuclidean distance between two sample points;
Wherein,
Wherein, n represents the number of the wireless signal strength of each sample point record;K represents natural number;rikRepresent i-th
K-th wireless signal strength value of sample point;rjkRepresent k-th wireless signal strength value of j-th sample point.
Advantages of the present invention:
The present invention proposes a kind of indoor orientation method based on DBSCAN algorithm, is built using unsupervised-learning algorithm
Mould, carries out classification according to wireless signal strength value (RSSI) and clusters, and non-artificial carry out stress and strain model, so that stress and strain model is more met
Reality, improves stability and the accuracy of indoor positioning.
Brief description
Fig. 1 is the indoor orientation method flow chart based on DBSCAN algorithm of an embodiment of the present invention;
Fig. 2 is the schematic diagram of the indoor positioning of an embodiment of the present invention.
Specific embodiment
Below in conjunction with the accompanying drawings an embodiment of the present invention is described further.
In the embodiment of the present invention, a kind of indoor orientation method based on DBSCAN algorithm, method flow diagram as shown in figure 1,
Comprise the following steps:
Step 1, in the building of required location, several signal projectors and several signal pickers are installed;
Step 2, the number of setting sample point, determine the coordinate (Z of each sample point1,Z2), and obtain each sample point
Signal strength signal intensity sequence { r1,r2,···,rn, that is, each sample point is to the signal strength signal intensity of each signal projector;
Step 3, the coordinate of each sample point and signal strength signal intensity sequence are carried out record number as a sample data, obtain
Obtain data set D={ x1,x2,···,xm};Wherein m represents natural number, i.e. the sequence number of sample data;
Step 4, the coordinate according to each sample point after numbering and signal strength signal intensity sequence, set up room using DBSCAN algorithm
Interior location model, comprises the following steps:
Step 4.1, setting density selection parameter, that is, set the sample that length θ of neighborhood and each kernel object include at least
Number Minpts of this point;
Step 4.2, the signal strength signal intensity sequence according to each sample point after numbering and set density selection parameter set
Determine the constraints of kernel object, determine whole kernel object of sample point according to constraints, enter one by kernel object
Step determines clustering cluster;The constraints of described kernel object is:Other sample points being comprised in each sample neighborhood of a point
The number of sample point that whether includes at least more than or equal to each set kernel object of number, comprise the following steps that:
Step 4.2.1, according to numbering after the signal strength signal intensity sequence of each sample point and the length of set neighborhood set
Determine the constraints of kernel object, and whole kernel objects in sample point are determined according to constraints, constraints formula is such as
Under:
|Nθ(xj)|≥Minpts (1)
Wherein, NθRepresent the number of samples in the θ neighborhood of certain sample point;θ represents the length of neighborhood;xjRepresent j-th sample
This point, j represents natural number, i.e. the sequence number of sample point;Minpts represents the number of the sample point that each kernel object includes at least;
In the embodiment of the present invention, for xj∈ D, its θ neighborhood comprise in sample set D with xjDistance be less than or equal to θ sample
This, formula is:
Nθ(xj)={ xi∈D|d(xi,xj)≤θ} (2)
Wherein, xiRepresent i-th sample point;I represents natural number, i.e. the sequence number of sample point;D represents each sample after numbering
The coordinate of this point and the data set of signal strength signal intensity sequence;
In the embodiment of the present invention, d (xi,xj) represent xiWith xjEuclidean distance between two sample points, i.e. signal strength signal intensity sequence
Row { r1,r2,···,rnConstitute n dimension Euclidean space in xiWith xjThe distance of point-to-point transmission, formula is as follows:
Wherein, n represents the number of the wireless signal strength of each sample point record;K represents natural number, and that is, wireless signal is strong
The sequence number of angle value;rikRepresent k-th wireless signal strength value of i-th sample point;rjkRepresent k-th of j-th sample point no
Line signal strength values;
Step 4.2.2, whole clustering cluster are further determined that by kernel object;
In the embodiment of the present invention, if xjPositioned at xiThe field of θ in, and xiIt is kernel object, then claim xjBy xiDensity is gone directly;
For xiWith xjIf there is sample sequence { p1,p2,···,pn, wherein p1=xi, pn=xjEach of, and sample sequence
Sample all can be gone directly by previous sample rate, then claim xjBy xiDensity up to;Arbitrarily select one of sample set D core pair
Find out as " seed ", thus setting out its all density up to sample point, that is, determine corresponding clustering cluster;By whole core
Heart object conducts interviews, and obtains whole clustering cluster;
Step 4.3, obtain the average signal strength of each clustering cluster, and the coordinate according to each sample point after numbering,
Calculate the mean place coordinate obtaining each clustering cluster, that is, Indoor Locating Model is set up and completed;
In the embodiment of the present invention, the average signal strength of described each clustering cluster of acquisition, using below equation:
Wherein, h represents the number of sample point in each clustering cluster;ri1Represent of i-th sample point in each clustering cluster
One wireless signal strength;ri2Represent second wireless signal strength of i-th sample point in each clustering cluster;rinRepresent every
N-th wireless signal strength of i-th sample point in individual clustering cluster;
In the embodiment of the present invention, the mean place coordinate of described each clustering cluster of acquisition, using below equation:
Wherein, Zi1Represent the abscissa value of i-th sample point in each clustering cluster;Zi2Represent i-th in each clustering cluster
Sample point ordinate value;
Step 5, obtain the signal strength signal intensity sequence in site undetermined by modes such as on-line measurements, and the room being completed according to foundation
Interior location model obtains the elements of a fix in this site undetermined, specially:Obtain the average letter in this site undetermined and each clustering cluster
The Euclidean distance of number intensity, and select the mean place coordinate of the clustering cluster minimum with its Euclidean distance as this site undetermined
The elements of a fix;
In the embodiment of the present invention, as shown in Fig. 2 being simulated to this method on computers, set 14 sample points, and with
Machine sets a site undetermined, and the wireless signal strength sequence that each sample point receives is { r1,r2, using DBSCAN algorithm
Sample point be divide into two clustering cluster of C1 and C2, obtain the central point of each clustering cluster respectively, i.e. average signal strength point;Obtain
Obtain the Euclidean distance in site undetermined and each clustering cluster central point, minimum with the Euclidean distance of clustering cluster C1 here, then position
Coordinate is the mean place coordinate of clustering cluster C1.
Claims (2)
1. a kind of indoor orientation method based on DBSCAN algorithm it is characterised in that:Comprise the following steps:
Step 1, in the building of required location, several signal projectors and several signal pickers are installed;
Step 2, the number of setting sample point, determine the coordinate of each sample point, and obtain the signal strength signal intensity sequence of each sample point
Row, that is, each sample point is to the signal strength signal intensity of each signal projector;
Step 3, by the coordinate of each sample point and signal strength signal intensity sequence carry out record numbering;
Step 4, the coordinate according to each sample point after numbering and signal strength signal intensity sequence, it is indoor fixed to be set up using DBSCAN algorithm
Bit model, comprises the following steps:
Step 4.1, setting density selection parameter, the sample point that is, length of setting neighborhood includes at least with each kernel object
Number;
Step 4.2, the signal strength signal intensity sequence according to each sample point after numbering and set density selection parameter set core
The constraints of heart object, determines the kernel object of sample point, further determines that cluster by kernel object according to constraints
Cluster;The constraints of described kernel object is:Whether the number of other sample points being comprised in each sample neighborhood of a point
The number of the sample point including at least more than or equal to each set kernel object;
Step 4.3, obtain the average signal strength of each clustering cluster, and the coordinate according to each sample point after numbering, obtain
The mean place coordinate of each clustering cluster, i.e. Indoor Locating Model foundation completes;
Step 5, the signal strength signal intensity sequence in acquisition site undetermined, and it is to be positioned to obtain this according to the Indoor Locating Model that foundation completes
The elements of a fix of point, specially:Obtain the Euclidean distance in this site undetermined and the average signal strength of each clustering cluster, and select
The mean place coordinate of minimum clustering cluster is as the elements of a fix in this site undetermined with its Euclidean distance.
2. the indoor orientation method based on DBSCAN algorithm according to claim 1 it is characterised in that:Described in step 4.2
Kernel object constraints, formula is as follows:
|Nθ(xj)|≥Minpts (1)
Wherein, NθRepresent the number of samples in the θ neighborhood of certain sample point;θ represents the length of neighborhood;xjRepresent j-th sample
Point, j represents natural number;Minpts represents the number of the sample point that each kernel object includes at least;
Wherein,
Nθ(xj)={ xi∈D|d(xi,xj)≤θ} (2)
Wherein, xiRepresent i-th sample point;I represents natural number;D represent numbering after the coordinate of each sample point and signal strength signal intensity
The data set of sequence;d(xi,xj) represent xiWith xjEuclidean distance between two sample points;
Wherein,
Wherein, n represents the number of the wireless signal strength of each sample point record;K represents natural number;rikRepresent i-th sample
K-th wireless signal strength value of point;rjkRepresent k-th wireless signal strength value of j-th sample point.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611059276.2A CN106412841A (en) | 2016-11-23 | 2016-11-23 | Indoor location method based on DBSCAN algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611059276.2A CN106412841A (en) | 2016-11-23 | 2016-11-23 | Indoor location method based on DBSCAN algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106412841A true CN106412841A (en) | 2017-02-15 |
Family
ID=58082772
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611059276.2A Pending CN106412841A (en) | 2016-11-23 | 2016-11-23 | Indoor location method based on DBSCAN algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106412841A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108307308A (en) * | 2018-01-25 | 2018-07-20 | 北京小米移动软件有限公司 | Localization method, device and the storage medium of WLAN devices |
CN108574927A (en) * | 2017-03-09 | 2018-09-25 | 中国移动通信有限公司研究院 | A kind of mobile terminal locating method and device |
CN110443376A (en) * | 2019-08-30 | 2019-11-12 | 中国南方电网有限责任公司超高压输电公司贵阳局 | State analysis method and its application module based on non-supervisory machine learning algorithm |
CN110850369A (en) * | 2019-09-23 | 2020-02-28 | 广州极汇信息科技有限公司 | Indoor positioning method, system, device and storage medium |
CN111372186A (en) * | 2019-12-17 | 2020-07-03 | 广东小天才科技有限公司 | Position calculation method under non-uniform positioning scene and terminal equipment |
CN111693938A (en) * | 2020-06-10 | 2020-09-22 | 北京云迹科技有限公司 | Floor positioning method and device of robot, robot and readable storage medium |
CN113596989A (en) * | 2021-08-04 | 2021-11-02 | 电子科技大学 | Indoor positioning method and system for intelligent workshop |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130024479A1 (en) * | 2011-07-20 | 2013-01-24 | Salesforce.Com, Inc. | Methods and systems for processing large graphs using density-based processes using map-reduce |
CN103200674A (en) * | 2013-03-12 | 2013-07-10 | 华北电力大学 | Mobile object positioning and following method facing toward unattended transformer substation |
CN104267439A (en) * | 2014-08-20 | 2015-01-07 | 哈尔滨工程大学 | Unsupervised human detecting and positioning method |
CN106125045A (en) * | 2016-06-15 | 2016-11-16 | 成都信息工程大学 | A kind of ADAPTIVE MIXED indoor orientation method based on Wi Fi |
CN106131958A (en) * | 2016-08-09 | 2016-11-16 | 电子科技大学 | A kind of based on channel condition information with the indoor Passive Location of support vector machine |
-
2016
- 2016-11-23 CN CN201611059276.2A patent/CN106412841A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130024479A1 (en) * | 2011-07-20 | 2013-01-24 | Salesforce.Com, Inc. | Methods and systems for processing large graphs using density-based processes using map-reduce |
CN103200674A (en) * | 2013-03-12 | 2013-07-10 | 华北电力大学 | Mobile object positioning and following method facing toward unattended transformer substation |
CN104267439A (en) * | 2014-08-20 | 2015-01-07 | 哈尔滨工程大学 | Unsupervised human detecting and positioning method |
CN106125045A (en) * | 2016-06-15 | 2016-11-16 | 成都信息工程大学 | A kind of ADAPTIVE MIXED indoor orientation method based on Wi Fi |
CN106131958A (en) * | 2016-08-09 | 2016-11-16 | 电子科技大学 | A kind of based on channel condition information with the indoor Passive Location of support vector machine |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108574927A (en) * | 2017-03-09 | 2018-09-25 | 中国移动通信有限公司研究院 | A kind of mobile terminal locating method and device |
CN108574927B (en) * | 2017-03-09 | 2020-11-10 | 中国移动通信有限公司研究院 | Mobile terminal positioning method and device |
CN108307308A (en) * | 2018-01-25 | 2018-07-20 | 北京小米移动软件有限公司 | Localization method, device and the storage medium of WLAN devices |
CN108307308B (en) * | 2018-01-25 | 2021-09-14 | 北京小米移动软件有限公司 | Positioning method, device and storage medium for wireless local area network equipment |
CN110443376A (en) * | 2019-08-30 | 2019-11-12 | 中国南方电网有限责任公司超高压输电公司贵阳局 | State analysis method and its application module based on non-supervisory machine learning algorithm |
CN110443376B (en) * | 2019-08-30 | 2024-05-17 | 中国南方电网有限责任公司超高压输电公司贵阳局 | State analysis method based on non-supervision machine learning algorithm and application module thereof |
CN110850369A (en) * | 2019-09-23 | 2020-02-28 | 广州极汇信息科技有限公司 | Indoor positioning method, system, device and storage medium |
CN111372186A (en) * | 2019-12-17 | 2020-07-03 | 广东小天才科技有限公司 | Position calculation method under non-uniform positioning scene and terminal equipment |
CN111372186B (en) * | 2019-12-17 | 2021-08-24 | 广东小天才科技有限公司 | Position calculation method under non-uniform positioning scene and terminal equipment |
CN111693938A (en) * | 2020-06-10 | 2020-09-22 | 北京云迹科技有限公司 | Floor positioning method and device of robot, robot and readable storage medium |
CN113596989A (en) * | 2021-08-04 | 2021-11-02 | 电子科技大学 | Indoor positioning method and system for intelligent workshop |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106412841A (en) | Indoor location method based on DBSCAN algorithm | |
WO2019233054A1 (en) | Fingerprint-based positioning method and system for smart classroom | |
CN103476118B (en) | A kind of WLAN indoor location fingerprint positioning method for monitoring in real time | |
CN105223546B (en) | Indoor orientation method based on received signal strength and reference point locations double focusing class | |
KR102116824B1 (en) | Positioning system based on deep learnin and construction method thereof | |
US9842159B2 (en) | Data analysis apparatus and method | |
CN105221356B (en) | The system and method estimated for Wind turbine nacelle position recalibration and wind direction | |
CN105636201B (en) | Indoor orientation method based on sparse signal fingerprint database | |
CN104038901B (en) | Indoor positioning method for reducing fingerprint data acquisition workload | |
CN109242041A (en) | A kind of electric energy meter abnormal deviation data examination method, device, equipment and storage medium | |
CN106612495B (en) | A kind of indoor orientation method and system based on propagation loss study | |
CN106131797A (en) | A kind of water-saving irrigation monitoring network locating method based on RSSI range finding | |
CN110536257B (en) | Indoor positioning method based on depth adaptive network | |
CN109034232B (en) | Automatic output system and control method for urban planning condition verification result report | |
CN105554873A (en) | Wireless sensor network positioning algorithm based on PSO-GA-RBF-HOP | |
CN106407627A (en) | Wind speed probability distribution modeling method and system | |
CN106842118A (en) | For the time difference positioning method of Distributed Multi positioning monitoring system | |
CN105868853A (en) | Method for predicting short-term wind power combination probability | |
CN114004374B (en) | Air conditioning unit sensor fault diagnosis method based on Boltzmann machine voting | |
CN109143161B (en) | High-precision indoor positioning method based on mixed fingerprint quality evaluation model | |
CN110263998A (en) | Multi-source numerical weather forecast set bilayer modification method | |
CN103249144A (en) | C-type-based wireless sensor network node location method | |
CN106997476A (en) | Transmission system performance degradation assessment method of the multi-source without label data learning model building | |
CN109840466A (en) | Based on the comprehensive multiple measurement learning method of cluster and the overall situation/local distance | |
CN104050254B (en) | The method that feature database in 3D rooms is built using house data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into 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: 20170215 |