CN104684080B - A kind of three-dimensional WLAN indoor orientation methods - Google Patents
A kind of three-dimensional WLAN indoor orientation methods Download PDFInfo
- Publication number
- CN104684080B CN104684080B CN201510070419.9A CN201510070419A CN104684080B CN 104684080 B CN104684080 B CN 104684080B CN 201510070419 A CN201510070419 A CN 201510070419A CN 104684080 B CN104684080 B CN 104684080B
- Authority
- CN
- China
- Prior art keywords
- mrow
- msubsup
- msub
- mtd
- mtr
- 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
Links
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
- H04W64/003—Locating users or terminals or network equipment for network management purposes, e.g. mobility management locating network equipment
-
- H04W4/04—
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The present invention relates to a kind of three-dimensional WLAN indoor orientation methods, comprise the following steps：(1) scene to be studied is divided into several cube of lattice, the signal strength of its cube lattice that can be covered is recorded for each wireless access point；(2) nonzero value of cluster in the signal distributions data of each wireless access point is realized and compressed by server by fitting function；(3) user terminal senses and records the signal strength of each wireless access point in real time, uploads onto the server and carries out location Calculation with request server；(4) server according to the strength information of each wireless access point in the signal strength information of the received user terminal, return to user terminal by the estimated location for calculating user terminal.Compared with prior art, the present invention avoids conventional machines learning algorithm dimension disaster problem, and orientation range is expanded to three dimensions；Whole audience AP signal intensity profile data are recorded, and data are compressed, reduce memory space.
Description
Technical field
The present invention relates to a kind of localization method, more particularly, to a kind of three-dimensional WLAN indoor orientation methods.
Background technology
With the continuous development of mobile communication technology and being on the increase for new business demand, location-aware computing especially base
Service (Location Based Service, LBS) in position gradually causes the concern of people, and how to determine the position of user
Put be LBS key problem.Existing generally existing and widely used position service system has global positioning system
(Global Positioning System, GPS) and cellular network location system etc., under open outdoor environment, these are fixed
Position system can provide more satisfied positioning and navigation Service for people, however, in the interior that people often work and live
In environment, their positioning accuracy can not but make us receiving.Therefore many scholars and research institution begin one's study indoor positioning skill
Art.
Nowadays, various radiotechnics have been applied in the research of indoor positioning, such as infrared technique, ultra wide band
(Ultra-wide Band, UWB) technology, radio frequency identification (Radio-frequency Identification, RFID) technology
Deng Centimeter Level [1] can be reached in a small range positioning accuracy based on the alignment system of these wireless technologys.However, these are fixed
Position system be required to redeploy network with extra signal measurement hardware device, cost is higher, thus in application by
To limitation.
In recent years, WLAN (Wireless Local Area Networks, WLAN) has covered people's life
Living and work most hot zones and indoor environment, and most of laptops, smart mobile phone etc. are mobile now
Terminal also all possesses WLAN access functions.Gradually popularized with the application of WLAN, the indoor positioning technologies based on WLAN also by
Gradually grow up, attracted the extensive concern of domestic and international research institution, enterprise and scholar.In addition, the indoor positioning based on WLAN
Technology can make full use of the resource of existing WLAN, it is not necessary to extra network deployment or facility so that be based on
The indoor positioning technologies of WLAN have the low-cost advantage that other wireless location technologies can not be equal to.
At present location fingerprint positioning mode, usual base are mostly used for the indoor positioning technologies research based on WLAN both at home and abroad
In the thought of the machine learning such as nearest neighbor search, naive Bayesian statistics, BP neural network or support vector machines.These are based on position
The location technology for putting fingerprint technique more stresses to overweight the application scenarios of two dimensional surface, and related experiment is also in two dimensional surface
Completed in indoor environment.However, in practice people may require that an alignment system can in shopping mall market, library, write
The positional information provided in the places such as word building, hospital, airport hall, museum not only includes longitude and dimension or other two dimensions
The coordinate representation of plane, will also include height or place floor.
Location fingerprint positioning mode based on machine learning algorithm is there are intrinsic dimension disaster drawback, as the spy of description problem
When sign number exceedes some maximum, the effect of machine learning cannot not only improve, and degenerate on the contrary.And three-dimensional WLAN rooms
Interior orientation problem is necessarily required to substantial amounts of feature to portray the fingerprint of a certain position.
The content of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide one kind is based on whole audience position
Finger scan and the three-dimensional WLAN indoor orientation methods of data compression.
The purpose of the present invention can be achieved through the following technical solutions：
A kind of three-dimensional WLAN indoor orientation methods, it is characterised in that comprise the following steps：
(1) three dimensions in scene to be studied is divided into several isometric cube lattice, for each wireless access
Point, records the signal strength of its cube lattice that can be covered, and forms the signal distributions data set of the wireless access point and uploads to
Server；
(2) nonzero value of cluster in the signal distributions data of each wireless access point is passed through fitting by the server
Function realizes compression, and is stored in background data base；
(3) user terminal senses and records the signal strength of each wireless access point in real time, and in the information
Reach server and location Calculation is carried out with request server；
(4) intensity of each wireless access point is believed in signal strength information of the server according to the received user terminal
Breath, the estimated location for calculating user terminal concurrently return user terminal.
The volume of cube lattice is the half of body size.
The signal strength of cube lattice is specially the signal strength at cube lattice center.
Server described in the step (2) is by the non-zero of cluster in the signal distributions data of each wireless access point
Value realizes that compression is specially by fitting function：
(1) a cube lattice for the signal strength nonzero value of cluster are segmented by certain length；
(2) for each section of cube lattice obtained in step (1), the section cube lattice are intended by fitting function
Close, be denoted as
dataj={ [is1, ie1, f1, p1], [is2, ie2, f2, p2] ..., [isn, ien, fn, pn]}
Wherein, datajRepresent the signal distributions data of j-th of wireless access point, isk、iekFor the starting and ending of kth section
The label of cube lattice, fkFor fitting function type, pkFor the parameter vector of function.
The fitting function includes linear function f (x)=ax+b, chi square function f (x)=ax2+ bx+c and exponential function
F (x)=aebx。
The fitting function is according to following method choice：
(1) for each section of cube lattice, with x ∈ [isk,isk+1,isk+2,…,iek] it is independent variable, Substitute into equation below respectively for dependent variable and carry out linear function, chi square function and index letter
Several fittings；Wherein
The fitting formula of the parameter of linear function f (x)=ax+b is
Chi square function f (x)=ax2The fitting formula of the parameter of+bx+c is
Exponential function f (x)=aebxThe fitting formula of parameter be
(2) three fitting functions obtained in (1) step are calculated with minimal error quadratic sum SSE respectively, selects minimum miss
Poor quadratic sum SSE recklings are the fitting function of this cube of lattice section, and the calculation formula of minimal error quadratic sum SSE is as follows,
Wherein, wiRepresent the weight of each sample data, yiRepresent actual measured value,Represent match value.
Sample data weight w in the calculation formula of the minimal error quadratic sum SSEiFor 1.
The lattice number of the every section cube of lattice is not less than 10.
The computational methods of the user terminal estimated location are：
The intensity of each wireless access point in signal strength information of the server according to the received user terminal
Information searches the wireless access point cube lattice set equal with the signal strength in background data base；Server will be directed to every
Cube lattice set that a wireless access point is tried to achieve does intersection operation, and tries to achieve the barycenter of all cubes of lattice in the intersection as user
The estimated location of terminal.
Compared with prior art, the present invention has the following advantages：
1. conventional machines learning algorithm dimension disaster problem is avoided, so that orientation range expands to three dimensions；
2. recording whole audience AP signal intensity profile data, and data are compressed, reduce memory space；
3. server end can cache the AP signal intensity profile data often accessed, locating speed, specific caching are improved
Process is as follows：With cubic block marked as key, AP signal strengths are value, cache AP signal intensity profile data, and caching is effective
Phase, specific cache database selected memcached or redis in units of day.
Embodiment
With reference to specific embodiment, the present invention is described in detail.
Embodiment
First, the early-stage preparations stage
Positioning scene to be studied is divided into N number of cube of lattice, i-th cube of lattice are just represented with label i, and record i-th
The three dimensional space coordinate of a cube of lattice is (xi,yi,zi).Wireless access point AP number in this positioning scene is counted, is denoted as M.Jth
The signal strength that a AP is produced in i-th cube of lattice isSetFor describing the signal of j-th of AP
Intensity distribution.In fact,In have many values as 0, as long as we store those nonzero values, one
As these nonzero values be cluster, then achieve the purpose that compression storage by carrying out Function Fitting to the nonzero value of cluster.This hair
It is bright piecewise fitting to be carried out using linear function, chi square function or exponential function to the nonzero value signal strength of cluster, specifically according to
Determined according to indoor landform or environment, for example either one paths of stair as a segmentation or use more along corridor
Short segmentation, to improve degree of fitting, but section length is too small, and data will be It is not necessary to compression, directly storage, general segmentation
Length is not less than 10.Note storageData structure be dataj
dataj={ [is1,ie1,f1,p1],[is2,ie2,f2,p2],…,[isn,ien,fn,pn]}
Wherein,Have the nonzero value of n sections of clusters, for kth (1≤k≤n) section cluster nonzero value and
Speech, starting and ending label is respectively isk、iek；
Fitting function type used is fk, present invention relates solely to 3 kinds of fitting functions, be respectively f (x)=ax+b, f (x)=
ax2+ bx+c or f (x)=aebx, x value ranges are isk、isk+1、isk+2、……iek；
The parameter vector p of function usedkIf fitting function type is linear function and exponential function, pk=[ak,
bk], if chi square function, then pk=[ak,bk,ck]。
Before data are uncompressed,The data that the nonzero value of kth section cluster needs to store areIt is now to express the nonzero value number of this kth section cluster according to a kind of fitting function of selection
According to independent variable x ∈ [isk,isk+1,isk+2,…,iek], dependent variableIn this data
Linear function fit, chi square function fitting and exponential function fitting are carried out respectively.Cutting edge aligned letter is given respectively based on least square method
Number, chi square function and exponential function fitting formula, (1), (2), (3) are shown as the following formula.
Least square method is often used as one by the use of minimal error quadratic sum SSE (Sum of Squares Due to Error)
The index of the effect quality of fitting function, the calculation formula of SSE are
Wherein, wiRepresent the weight of each sample data, usually take wi=1, yiRepresent observation,Represent match value.
CompareThe SSE of upper above-mentioned 3 fitting functions, it is best fit letter to take SSE recklings
Number, and record corresponding parameter vector pk。
2nd, later stage positioning stage
(1) there is k AP when a mobile terminal senses surrounding, this mobile terminal will record information as shown in Table 1,
And the information is uploaded onto the server, and request server carries out location Calculation.
Table 1
(2) server stores the signal distributions data of each AP in positioning scene in background data base, such as the institute of table 2
Show.
Table 2
| Sequence number | MAC Address | Signal intensity profile |
| 1 | MAC1 | data1 |
| 2 | MAC2 | data2 |
| 3 | MAC3 | data3 |
| … | … | … |
| n | MACn | datan |
When server receive some mobile terminal transmission shape information as shown in Table 1, background data base will be inquired about, according to
The MAC Address MAC of i-th of APiFind the letter strength distributing information data on this APi, and lookup and signal strength wherein
RSSIiEqual cube case marker number set, since so set has k, finally takes this k intersection of sets collection, and set final friendship
M label is concentrated with, respectively k1、k2、……km, as represented shown in 3.
Table 3
The barycenter of the intersection of these small squares is taken as the location estimation of mobile terminal, then the three-dimensional coordinate of mobile terminal
ForFinally, service
The estimate of the three-dimensional coordinate of mobile terminal is returned to mobile terminal by device.
Claims (5)
1. a kind of three-dimensional WLAN indoor orientation methods, it is characterised in that comprise the following steps：
(1) three dimensions in scene to be studied is divided into several isometric cube lattice, for each wireless access point, note
The signal strength of its cube lattice that can be covered is recorded, the signal distributions data set of the wireless access point is formed and uploads to service
Device；
(2) nonzero value of cluster in the signal distributions data of each wireless access point is passed through fitting function by the server
Realize compression, and be stored in background data base；
(3) user terminal senses and records the signal strength of each wireless access point in real time, and the signal strength is believed
Breath uploads onto the server carries out location Calculation with request server；
(4) server is according to the strength information of each wireless access point in the signal strength information of the received user terminal,
The estimated location for calculating user terminal concurrently returns user terminal；
The signal strength of cube lattice is specially the signal strength at cube lattice center；
Server described in the step (2) leads to the nonzero value of cluster in the signal distributions data of each wireless access point
Over-fitting function realizes that compression is specially：
(1) cube lattice of the signal strength nonzero value of cluster are segmented by setting length；
(2) for each section of cube lattice obtained in step (1), the section cube lattice are fitted by fitting function, are remembered
For
dataj={ [is1,ie1,f1,p1],[is2,ie2,f2,p2],…,[isn,ien,fn,pn]}
Wherein, datajRepresent the signal distributions data of j-th of wireless access point, isk、iekFor the starting and ending cube of kth section
The label of lattice, fkFor fitting function type, pkFor the parameter vector of function, 1≤k≤n；
The fitting function includes linear function f (x)=ax+b, chi square function f (x)=ax2+ bx+c and exponential function f (x)
=aebx；
The fitting function is according to following method choice：
(1) for each section of cube lattice, with x ∈ [isk,isk+1,isk+2,…,iek] it is independent variable, y ∈Substitute into equation below respectively for dependent variable and carry out linear function, chi square function and exponential function
Fitting；Wherein n=ek-sk,
The fitting formula of the parameter of linear function f (x)=ax+b is
<mrow>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<mi>a</mi>
</mtd>
</mtr>
<mtr>
<mtd>
<mi>b</mi>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<mfrac>
<mrow>
<msubsup>
<mi>n&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
</mrow>
<mrow>
<msubsup>
<mi>n&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</mfrac>
</mtd>
</mtr>
<mtr>
<mtd>
<mfrac>
<mrow>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
<mo>-</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
</mrow>
<mrow>
<msubsup>
<mi>n&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</mfrac>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
Chi square function f (x)=ax2The fitting formula of the parameter of+bx+c is
<mrow>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<mi>c</mi>
</mtd>
</mtr>
<mtr>
<mtd>
<mi>b</mi>
</mtd>
</mtr>
<mtr>
<mtd>
<mi>a</mi>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>=</mo>
<msup>
<mrow>
<mo>(</mo>
<msup>
<mi>X</mi>
<mi>T</mi>
</msup>
<mi>X</mi>
<mo>)</mo>
</mrow>
<mrow>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msup>
<msup>
<mi>X</mi>
<mi>T</mi>
</msup>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<msub>
<mi>y</mi>
<mn>1</mn>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>y</mi>
<mn>2</mn>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>...</mn>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>y</mi>
<mi>n</mi>
</msub>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>,</mo>
<mi>X</mi>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<mi>n</mi>
</mtd>
<mtd>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
</mrow>
</mtd>
<mtd>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
</mrow>
</mtd>
<mtd>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
</mrow>
</mtd>
<mtd>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>3</mn>
</msubsup>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
</mrow>
</mtd>
<mtd>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>3</mn>
</msubsup>
</mrow>
</mtd>
<mtd>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>4</mn>
</msubsup>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
Exponential function f (x)=aebxThe fitting formula of parameter be
<mrow>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<mi>a</mi>
</mtd>
</mtr>
<mtr>
<mtd>
<mi>b</mi>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<mfrac>
<mrow>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mrow>
<mo>(</mo>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mi>ln</mi>
<mi> </mi>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mi>ln</mi>
<mi> </mi>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mrow>
<mo>(</mo>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mo>(</mo>
<mrow>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</mfrac>
</mtd>
</mtr>
<mtr>
<mtd>
<mfrac>
<mrow>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mi>ln</mi>
<mi> </mi>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mi>ln</mi>
<mi> </mi>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mrow>
<mo>(</mo>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msup>
<mrow>
<mo>(</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mo>(</mo>
<mrow>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</mfrac>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
(2) three fitting functions obtained in (1) step are calculated with minimal error quadratic sum SSE respectively, selects minimal error to put down
Side and SSE recklings are the fitting function of this cube of lattice section, and the calculation formula of minimal error quadratic sum SSE is as follows,
<mrow>
<mi>S</mi>
<mi>S</mi>
<mi>E</mi>
<mo>=</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>w</mi>
<mi>i</mi>
</msub>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<mover>
<msub>
<mi>y</mi>
<mi>i</mi>
</msub>
<mo>^</mo>
</mover>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
Wherein, wiRepresent the weight of each sample data, yiRepresent actual measured value,Represent match value.
A kind of 2. three-dimensional WLAN indoor orientation methods according to claim 1, it is characterised in that the body of cube lattice
Product is the half of body size.
3. a kind of three-dimensional WLAN indoor orientation methods according to claim 1, it is characterised in that the minimal error is put down
Sample data weight w in the calculation formula of side and SSEiFor 1.
4. according to any a kind of three-dimensional WLAN indoor orientation methods in claim 1 or 3, it is characterised in that described
The lattice number of every section cube of lattice is not less than 10.
5. according to any a kind of three-dimensional WLAN indoor orientation methods in claim 1 or 3, it is characterised in that the use
The computational methods of family terminal estimated location are：
The strength information of each wireless access point in signal strength information of the server according to the received user terminal
The wireless access point cube lattice set equal with the signal strength is searched in background data base；Server will be directed to each nothing
Cube lattice set that line access point is tried to achieve does intersection operation, and tries to achieve the barycenter of all cubes of lattice in the intersection as user terminal
Estimated location.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510070419.9A CN104684080B (en) | 2015-02-10 | 2015-02-10 | A kind of three-dimensional WLAN indoor orientation methods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510070419.9A CN104684080B (en) | 2015-02-10 | 2015-02-10 | A kind of three-dimensional WLAN indoor orientation methods |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104684080A CN104684080A (en) | 2015-06-03 |
| CN104684080B true CN104684080B (en) | 2018-04-27 |
Family
ID=53318507
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510070419.9A Active CN104684080B (en) | 2015-02-10 | 2015-02-10 | A kind of three-dimensional WLAN indoor orientation methods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104684080B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017015872A1 (en) * | 2015-07-28 | 2017-02-02 | 深圳市润安科技发展有限公司 | Method for positioning special patient in hospital and system for positioning special patient in hospital |
| CN108107407A (en) * | 2017-12-18 | 2018-06-01 | 黑龙江大学 | A kind of three-dimensional sound localization method based on spatially distributed microphone |
| CN110351665B (en) * | 2019-07-31 | 2020-10-30 | 中国联合网络通信集团有限公司 | Method, apparatus and computer-readable storage medium for habitual identification of a user |
| CN114585012B (en) * | 2020-12-01 | 2023-09-05 | 中移(成都)信息通信科技有限公司 | Method, device, equipment and storage medium for determining wireless signal strength |
| CN114928811B (en) * | 2022-05-05 | 2024-05-07 | 南京理工大学 | Ultra-wideband mine underground personnel accurate positioning method based on neural network |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104080166A (en) * | 2014-06-13 | 2014-10-01 | 上海交通大学 | Indoor positioning method based on intelligent wireless routers |
| CN104159293A (en) * | 2014-07-08 | 2014-11-19 | 上海交通大学 | Indoor positioning method for high-speed unmanned rotor craft |
-
2015
- 2015-02-10 CN CN201510070419.9A patent/CN104684080B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104080166A (en) * | 2014-06-13 | 2014-10-01 | 上海交通大学 | Indoor positioning method based on intelligent wireless routers |
| CN104159293A (en) * | 2014-07-08 | 2014-11-19 | 上海交通大学 | Indoor positioning method for high-speed unmanned rotor craft |
Non-Patent Citations (2)
| Title |
|---|
| A 3-D Indoor Location Tracking and Visualization System Based on Wireless Sensor Networks;Mert Bal 等;《Systems Man and Cybernetics (SMC), 2010》;20101122;全文 * |
| Properties of indoor received signal strength for WLAN location fingerprinting;K. Kaemarungsi 等;《Mobile and Ubiquitous Systems: Networking and Services, 2004》;20040903;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104684080A (en) | 2015-06-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104684080B (en) | A kind of three-dimensional WLAN indoor orientation methods | |
| CN111479231B (en) | Indoor fingerprint positioning method for millimeter wave large-scale MIMO system | |
| CN103402258B (en) | Wi-Fi (Wireless Fidelity)-based indoor positioning system and method | |
| CN102480677B (en) | A kind of determination method and apparatus of fingerprint positioning error | |
| US20130260771A1 (en) | Generation and use of coverage area models | |
| Yu et al. | 5 G wifi signal-based indoor localization system using cluster k-nearest neighbor algorithm | |
| CN104717744B (en) | A kind of indoor orientation method based on WLAN and hierarchical cluster | |
| CN104144493B (en) | Localization method, alignment system and base station location platform | |
| Ding et al. | Overview of received signal strength based fingerprinting localization in indoor wireless LAN environments | |
| CN104703128B (en) | A kind of indoor locating system and method based on WLAN wireless signal strengths | |
| CN103049772A (en) | Indoor positioning tracking system and method based on passive radiofrequency identification technology | |
| CN105592420B (en) | Environmental characteristic library generates and indoor orientation method and device based on environmental characteristic library | |
| Zandbergen | Comparison of WiFi positioning on two mobile devices | |
| CN113596989A (en) | Indoor positioning method and system for intelligent workshop | |
| CN104683953B (en) | Indoor WLAN based on SimRank Similar Composite Systems neighborhood graph structure positions network-building method | |
| CN107124455A (en) | Indoor orientation method based on high in the clouds plateform system | |
| Maaloul et al. | Bluetooth beacons based indoor positioning in a shopping malls using machine learning | |
| Yang et al. | An improved geometric algorithm for indoor localization | |
| Cheng et al. | A New Three‐Dimensional Indoor Positioning Mechanism Based on Wireless LAN | |
| Bak et al. | Characteristics of a large-scale wifi radiomap and their implications in indoor localization | |
| CN105072169A (en) | Intelligent information display system of culture exhibition hall | |
| CN110320493B (en) | Indoor positioning method, device, electronic equipment and computer storage medium | |
| Li et al. | Outdoor location estimation in changeable environments | |
| KR20140060734A (en) | Method for determining position by using weight, apparatus therefor | |
| Hongpeng et al. | A hybrid modeling for WLAN positioning system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |