CN106507476A - A kind of WiFi localization methods and server and location model construction method - Google Patents
A kind of WiFi localization methods and server and location model construction method Download PDFInfo
- Publication number
- CN106507476A CN106507476A CN201611044936.XA CN201611044936A CN106507476A CN 106507476 A CN106507476 A CN 106507476A CN 201611044936 A CN201611044936 A CN 201611044936A CN 106507476 A CN106507476 A CN 106507476A
- Authority
- CN
- China
- Prior art keywords
- training
- location
- data
- neural network
- deep neural
- 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
- 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)
- Position Fixing By Use Of Radio Waves (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a kind of WiFi localization methods, methods described includes step：In S100, acquisition detection zone, each WAP receives the signal strength data of the signal that client to be detected sends；S200, the input data layer that the signal strength data obtained in step S100 is input into location model；S300, based on training after the Internet of location model calculate the signal strength data, in the output layer outgoing position forecast set of location model, the position prediction collection includes predicted position label and corresponding probit, and the position corresponding to the predicted position label for concentrating probit maximum the position prediction is defined as the position that the client to be detected is located.Location model in the present invention is trained to deep neural network by a large amount of training sample data using the deep neural network after training, lifts Position location accuracy and precision.
Description
Technical field
The present invention relates to wireless local area network technology field, more particularly to a kind of WiFi localization methods and server and positioning mould
Type construction method.
Background technology
Location technology worldwide mainly has GPS location, Wi-Fi positioning, bluetooth positioning etc., GPS location at present
Outdoor is mainly used in, Wi-Fi, bluetooth positioning can be not only used for interior, it can also be used to outdoor.As Wi-Fi positions relative maturity,
The particular content of the present invention is introduced with Wi-Fi location technologies as background below.With the popularization of wireless router, current big portion
Point public territory all has been carried out more than ten or even tens WiFi signals are covered, and these routers are being propagated to surrounding
While WiFi signal, its physical address and the information such as signal intensity is also ceaselessly sent, as long as in its signal cover,
Even if not knowing the password of Wi-Fi, these information can be similarly obtained.
General WiFi indoor positioning technologies are the WLAN (WLAN) based on IEEE802.11b/g agreements mostly
Signal intensity location technology.It is that letter is calculated according to the intensity of the signal for receiving based on the location technology ultimate principle of signal intensity
The distance between number receptor and signal source, are largely divided into two classes：Triangle intensity algorithm and location fingerprint recognizer.Its
Intermediate cam shape intensity arithmetic accuracy is low, it is difficult to meet indoor positioning requirement；And there is receiving device in general fingerprint recognizer
Different and cause to receive the defect that signal has error.
Content of the invention
For solving above-mentioned technical problem, the present invention provides a kind of WiFi localization methods and server and location model structure side
Method, by gathering the corresponding signal strength data of each WAP, realizes that the WiFi based on deep neural network is positioned.
The technical scheme that the present invention is provided is as follows：
The invention discloses a kind of WiFi localization methods, methods described includes step：S100, obtain detection zone in each
WAP receives the signal strength data of the signal that client to be detected sends；S200, will in step S100 obtain
The signal strength data be input into the input data layer of location model；S300, based on training after location model Internet
The signal strength data is calculated, in the output layer outgoing position forecast set of location model, the position prediction collection includes predicting
Location tags and corresponding probit, the position corresponding to the predicted position label for concentrating probit maximum the position prediction
Put the position for being defined as that the client to be detected is located.
It is further preferred that also including step before step S100：S000, training in advance deep neural network, will
Deep neural network after training is used as the location model.
It is further preferred that step S000 further includes step：S001, stress and strain model is carried out to detection zone,
Multiple plane grids are obtained, and distributes the corresponding training location tags for training for each plane grid；S002, exist successively
On the default training position coordinateses, training terminal, each institute that each WAP is received in acquisition testing region are set
State the signal strength data of the signal for training terminal to send on training location tags；S003, respectively by each training position
Each described training location tags that each WAP gathered in label and step S002 is received send
The signal strength data of signal generates training dataset, and sends in deep neural network as one group of training sample data；
Signal strength data in S004, each group of training sample data that concentrates training data successively is input into the depth nerve net
The data input layer of network, exports corresponding training result through the deep neural network；S005, the training that will be exported successively
As a result the corresponding training location tags are compared, and deep neural network are trained according to comparative result, will
Deep neural network after training is used as the location model.
It is further preferred that step S005 further includes step：Calculate the corresponding institute of the comparative result
State the error of training location tags, according to the parameter of the error transfer factor deep neural network, until the comparative result and its
The error convergence of the corresponding predicted position label is in preset range.
The invention also discloses a kind of WiFi location-servers, including：Data acquisition module, for obtaining in detection zone
Each WAP receives the signal strength data of the signal that client to be detected sends；Locating module, for obtaining
The signal strength data for taking is input into the input data layer of location model, based on training after the Internet of location model calculate
The signal strength data, in the output layer outgoing position forecast set of location model, the position prediction collection includes predicted position
Label and corresponding probit, the position corresponding to the predicted position label for concentrating probit maximum the position prediction are true
It is set to the position that the client to be detected is located.
It is further preferred that also including：Training module, for training in advance deep neural network, by training after depth
Neutral net is used as the location model.
It is further preferred that the training module is further included：Stress and strain model submodule, for carrying out to detection zone
Stress and strain model, obtains multiple plane grids, and distributes the corresponding training location tags for training for each plane grid；Instruction
Practice data set generation submodule, for training terminal, acquisition testing region being arranged on the default training position coordinateses successively
The signal intensity number of the signal for training terminal to send on each described training location tags that each WAP interior is received
According to each the described training position for receiving each WAP of each described training location tags and collection respectively
The signal strength data of the signal that label sends generates training dataset, and sends into depth god as one group of training sample data
In through network；Training prediction submodule, for successively will training data concentrate each group of training sample data in signal strong
Degrees of data is input into the data input layer of the deep neural network, exports corresponding training knot through the deep neural network
Really, and successively the predicted position label corresponding for the training result of output is compared, according to comparative result to depth
Degree neutral net be trained, using training after deep neural network as the location model.
The invention also discloses a kind of location model construction method, including step：S10, detection zone is carried out grid draw
Point, multiple plane grids are obtained, and distributes the corresponding training location tags for training for each plane grid；S11, successively
On the default training position coordinateses, training terminal, each that each WAP is received in acquisition testing region are set
The signal strength data of the signal that training terminal sends on the training location tags；S12, respectively by each training position
The letter that each described training location tags that each WAP gathered in label and step S11 is received send
Number signal strength data as one group of training sample data, generate training dataset, and send in deep neural network；S13、
Signal strength data in each group of training sample data that concentrates training data successively is input into the deep neural network
Data input layer, exports corresponding training result through the deep neural network；S14, the training result that will be exported successively
The corresponding training location tags are compared, and deep neural network are trained according to comparative result, will training
Deep neural network afterwards is used as the location model.
It is further preferred that step S14 further includes step：Calculate corresponding described of the comparative result
The error of training location tags, according to the parameter of the error transfer factor deep neural network, until the comparative result right with which
The error convergence of the predicted position label that answers is in preset range.
Compared with prior art, a kind of localization method and server and positioning mould based on WiFi network that the present invention is provided
Type construction method, is trained by collecting the signal strength data input corresponding with each WAP of client to be measured
Location model, you can obtain the probability of classification belonging to client position to be measured, client to be measured determined according to probit
Position, by using containing a large amount of training sample data training data set pair deep neural network train, using depth god
Through network as location model, the lifting of positioning precision is not only lifted, while can carry in the case where locating speed is not affected
The accuracy of positioning result is risen, successfully orientation problem is dissolved in the background of big data, and effectively using the excellent of big data
Gesture is improving the performance of real-time positioning server.
Description of the drawings
Below by the way of clearly understandable, preferred implementation is described with reference to the drawings, the present invention is given furtherly
Bright.
Fig. 1 is a kind of key step schematic diagram of WiFi localization methods of the invention；
Fig. 2 is a kind of main composition schematic diagram of WiFi location-servers of the invention；
Fig. 3 is that a kind of WiFi location-servers of the invention are fully composed schematic diagram；
Fig. 4 is a kind of key step schematic diagram of location model construction method of the invention.
Reference：
100th, data acquisition module, 200, locating module, 300, training module, 301, stress and strain model submodule, 302, instruction
Practice data set generation submodule, 303, training prediction submodule.
Specific embodiment
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below by control description of the drawings
The specific embodiment of the present invention.It should be evident that drawings in the following description are only some embodiments of the present invention, for
For those of ordinary skill in the art, on the premise of not paying creative work, can be obtaining other according to these accompanying drawings
Accompanying drawing, and obtain other embodiments.
For making simplified form, in each figure, part related to the present invention is only schematically show, they do not represent
Its practical structures as product.In addition, so that simplified form is readily appreciated, there is in some figures identical structure or function
Part, only symbolically depicts one of those, or has only marked one of those.Herein, " one " is not only represented
" only this ", it is also possible to represent the situation of " more than one ".
Fig. 1 is a kind of key step schematic diagram of WiFi localization methods of the invention, as shown in figure 1, a kind of WiFi positioning sides
Method, methods described include step：In S100, acquisition detection zone, each WAP receives what client to be detected sent
The signal strength data of signal；S200, by the defeated of the signal strength data input location model obtained in step S100
Enter data Layer；S300, based on training after the Internet of location model calculate the signal strength data, in the defeated of location model
Go out a layer outgoing position forecast set, the position prediction collection includes predicted position label and corresponding probit, by the position prediction
The position corresponding to the predicted position label for concentrating probit maximum is defined as the position that the client to be detected is located.
Specifically, above-mentioned client to be detected (hereinafter referred to as STA) is with smart mobile phone, notebook computer or personal flat board
The intelligent terminals such as computer are carrier.
STA sends detection frame within a detection region in real time, and WAP obtains detection frame signal after receiving is strong
Degree, each WAP report signal intensity to home server or Cloud Server, and server is according to each WAP
The RSSI field intensity message for reporting generates signal strength data.The present invention is wireless with each by obtained each WAP
Initial data of the corresponding signal strength data of access point as input location model, for example, the form of signal strength data is<
RSSI1, RSSI2, RSSI3, RSSI4, RSSI5>, the RSSI of the STA that wherein RSSI1 is received for AP1, RSSI2 are received for AP2
The RSSI of STA, by that analogy.
Location model in the present invention adopts the deep neural network after training, by a large amount of training sample data to depth
Neutral net is trained, and lifts Position location accuracy and precision.The present invention can also be by the letter related to signal strength data
The data such as number road, the input power of AP and weather environment are together gathered, by increasing capacitance it is possible to increase setting accuracy.
Preferably, also include step before step S100：S000, training in advance deep neural network, after training
Deep neural network as the location model.
The deep neural network of present invention design is training network, for the training of network parameter.Depth in the present embodiment
Neutral net is as follows as training network structure：
Data Layer->Convolutional layer 1->Convolutional layer 2->ReLU layers->Max Pooling layers->Full articulamentum 1->Full articulamentum
2->SoftMaxLoss layers
Network ginseng after network parameter training is completed, after being updated to train by the network parameter of deep neural network
Number, while by last layer of SoftMaxLoss layer of training network more SoftMax layers, forming and implementing network, for as positioning
Model participates in actual location process.When wherein SoftMaxLoss layers are trained for deep neural network, output training is defeated
Go out the error of result and the training location tags of reality, and SoftMax layers are used for, when network is implemented in positioning, exporting to be detected
The probit of classification belonging to client position.
Training network and implement network except last layer different (training network be SoftMaxLoss layers, implement network
For), all, the network parameter obtained by training network can be used directly in enforcement network other layers.
Preferably, step S000 further includes step：S001, stress and strain model is carried out to detection zone, obtained many
Individual plane grid, and distribute the corresponding training location tags for training for each plane grid；S002, gather each successively
The signal intensity number of each WAP signal is received on the corresponding grid in detection zone of the training location tags
According to；S003, respectively by each described training location tags and its corresponding position on receive each WAP signal
Signal strength data generates training dataset, and sends in deep neural network as one group of training sample data；S004, according to
Signal strength data in secondary each group of training sample data for concentrating training data is input into the number of the deep neural network
According to input layer, corresponding training result is exported through the deep neural network；S005, successively by output training result with
Its corresponding described training location tags is compared, and deep neural network is trained according to comparative result, after training
Deep neural network as the location model.
Specifically, method of the present invention using the global parameter training for having supervision：Known corresponding with each WAP
The physical location of signal strength data belong to certain grid, the net that deep neural network is caused by constantly adjustment network parameter
The output of network layers is identical with real result.
The training process of of the present invention deep neural network is discussed in detail with reference to above-mentioned steps with instantiation below.
1st, it is that detection zone is divided first, in detection zone, gathers training sample data.
Step S001 carries out stress and strain model to detection zone, obtains multiple plane grids, and is each plane grid
The corresponding training location tags for training of distribution are specific as follows：Deep neural network by detection zone net in the present embodiment
Format and be divided into multiple plane grids, be that each plane grid distributes corresponding training location tags, such as detection zone is one
Length direction, it is assumed that a length of M, a width of N, area are M*N.According in precision characteristic the present embodiment of WIFI using 3 meters as substantially single
Position, then this inner space be divided into M/3*N/3 grid.Explain for convenience, it is assumed that M/3 and N/3 is integer, it is assumed that M
=30, N=21, then M/3=10, N/3=7, whole detection zone are divided into 70 spaces after stress and strain model, define this
70 classes of 70 spaces for deep learning neutral net, are respectively this 70 class distribution training location tags, for example, it is possible to according to from
Left-to-right order from top to bottom, is followed successively by each plane grid and is numbered, such that it is able to obtain this 70 marks from 1 to 70
Know.Mark for each plane grid distribution can be used as training location tags.For example, the mark " 34 " of the 34th plane grid
Just can be used as training location tags.
Training terminal is arranged successively on the default training position coordinateses to step S002 according to examples detailed above, is adopted
The signal for training terminal to send on each described training location tags that each WAP is received in collection detection zone
Signal strength data is illustrated, specific as follows：The 1 corresponding position of grid is designated in the detection zone of above-mentioned 70 grids
Putting and detection frame being sent by training terminal, WAP obtains the signal intensity of the detection frame after receiving, each is wireless
Access point reports signal intensity to home server or Cloud Server.Obtained by the RSSI field intensity messages of collection of server each AP
Signal strength data corresponding with each AP is obtained, meanwhile, by the training location tags binding signal intensity number of the grid for being designated 1
According to one group of training sample data of generation, it is assumed that have 4 AP in detection zone, then one group of training sample data is expressed as：(RSSI1,
RSSI2, RSSI3, RSSI4,34), are expressed as with example<- 30, -12, -14, -67,34>, wherein, the unit of front four numerical value
Can be dBm, last numerical value can be with dimensionless.The training sample data can represent that the access point that numbering is 1 is corresponding
RSSI is -30dBm, and it is -12dBm that numbering is the 2 corresponding RSSI of access point, numbering be 3 the corresponding RSSI of access point for -
14dBm, it is -67dBm that numbering is the 4 corresponding RSSI of access point, and the STA is located at and is designated in 34 plane grid.
2nd, secondly the signal strength data input deep neural network in training sample data is calculated, is finally exported
Training result and the error for training location tags.
RSSI1 to RSSI4 in training sample data is input into from data Layer, is carried out with label in SoftMaxLoss layers
Return, Loss is exported in SoftMaxLoss layers by training.
3rd, cause the Loss i.e. error of whole network minimum finally by the parameter in percentage regulation neutral net.
It should be noted that do not indicate design parameter in entire depth neutral net, because these parameters and specific
The number of space and AP is relevant, not in the range of this patent.
Preferably, step S005 further includes step：Calculate the corresponding training of the comparative result
The error of location tags, according to the parameter of the error transfer factor deep neural network, until the comparative result corresponding
The error convergence of the predicted position label is in preset range.
Specifically, in the present embodiment, training process can adopt stochastic gradient descent, batch gradient to decline and Conjugate gradient descent
Etc. method, specifically it is not construed as limiting.
Parameter in the present invention after deep neural network training can be used directly in enforcement network, will implement network conduct
Location model.Deep neural network is as follows as network structure is implemented：
Data Layer->Convolutional layer 1->Convolutional layer 2->ReLU layers->Max Pooling layers->Full articulamentum 1->Full articulamentum
2->SoftMax layers
Below with example introduction based on the deep neural network for training to residing for client position to be detected classify
It is predicted, detailed process is as follows：
STA to be detected is placed in any one position in the detection zone.The outside broadcast probe request frames of the STA, detection zone
Each AP in domain is stimulated generation RSSI field intensity message and is reported to server, the RSSI fields that server is reported according to all AP
Strong message obtains signal strength data corresponding with each AP, is represented by conjunction with examples detailed above, (RSSI1, RSSI2, RSSI3,
RSSI4).
RSSI1 to the RSSI4 of the signal strength data for detecting of STA positions to be predicted is input into from data Layer,
SoftMax layers export 70 class prediction probabilities.
Assume that 70 class probability of output are arranged as from big to small：
Label 44：0.7
Label 1：0.2
Label 21：0.1
Label 11：0.04
Labe l9：0.01
…
…
Label44 is then chosen as last prediction, that is, predicts that this STA is being designated in 44 grid.
Fig. 2 is a kind of composition schematic diagram of WiFi location-servers of the invention, as shown in Fig. 2 a kind of WiFi positioning services
Device, including：Data acquisition module 100, receives client to be detected for obtaining each WAP in detection zone
The signal strength data of the signal for going out；Locating module 200, for being input into location model by the signal strength data for obtaining
Input data layer, based on training after the Internet of location model calculate the signal strength data, in the defeated of location model
Go out a layer outgoing position forecast set, the position prediction collection includes predicted position label and corresponding probit, by the position prediction
The position corresponding to the predicted position label for concentrating probit maximum is defined as the position that the client to be detected is located.
Specifically, above-mentioned client to be detected (hereinafter referred to as STA) is with smart mobile phone, notebook computer or personal flat board
The intelligent terminals such as computer are carrier.
STA sends detection frame within a detection region in real time, and WAP obtains detection frame signal after receiving is strong
Degree, each WAP report signal intensity to home server or Cloud Server, and server is according to each WAP
The RSSI field intensity message for reporting generates signal strength data.The present invention is wireless with each by obtained each WAP
Initial data of the corresponding signal strength data of access point as input location model, for example, the form of signal strength data is<
RSSI1, RSSI2, RSSI3, RSSI4, RSSI5>, the RSSI of the STA that wherein RSSI1 is received for AP1, RSSI2 are received for AP2
The RSSI of STA, by that analogy.
Location model in the present invention adopts the deep neural network after training, by a large amount of training sample data to depth
Neutral net is trained, and lifts Position location accuracy and precision.
Fig. 3 is that a kind of WiFi location-servers of the invention are fully composed schematic diagram.Preferably, as shown in figure 3, also wrapping
Include：Training module 300, for training in advance deep neural network, using training after deep neural network as the positioning mould
Type.
The deep neural network of present invention design is training network, for the training of network parameter, when network parameter is trained
After completing, by the network parameter of deep neural network be updated to train after network parameter, while by training network last
Layer SoftMaxLoss layers more SoftMax layers, form and implement network, for participating in actual location process as location model.Its
When middle SoftMaxLoss layers are trained for deep neural network, the output result of output training is marked with the training position of reality
The error of label, and SoftMax layers are used for, when network is implemented in positioning, exporting classification belonging to client position to be detected
Probit.
Training network and implement network except last layer different (training network be SoftMaxLoss layers, implement network
For), all, the network parameter obtained by training network can be used directly in enforcement network other layers.
Preferably, the training module 300 is further included：Stress and strain model submodule 301, for carrying out to detection zone
Stress and strain model, obtains multiple plane grids, and distributes the corresponding training location tags for training for each plane grid；Instruction
Practice data set generation submodule 302, for training terminal, acquisition testing area being arranged on the default training position coordinateses successively
The signal intensity of the signal for training terminal to send on each described training location tags that each WAP is received in domain
Data, each described training position that respectively each WAP of each described training location tags and collection is received
The signal strength data of the signal that label sends is put as one group of training sample data, training dataset is generated, and is sent into depth
In neutral net；Training prediction submodule 303, for successively by the letter in each group of training sample data of training data concentration
Number intensity data is input into the data input floor of the deep neural network, exports corresponding instruction through the deep neural network
Practice result, and successively the predicted position label corresponding for the training result of output is compared, according to comparative result
Deep neural network is trained, using training after deep neural network as the location model.
Specifically, method of the present invention using the global parameter training for having supervision：Known corresponding with each WAP
The physical location of signal strength data belong to certain grid, the net that deep neural network is caused by constantly adjustment network parameter
The output of network layers is identical with real result.
It should be noted that for the training process of above-mentioned training module 300 refer to the present invention for WiFi localization methods
The explanation of middle training deep neural network, is no longer repeated herein.Information exchange, implementation procedure in book server between each module
Same design is based on etc. content and said method embodiment, particular content can be found in the narration in the inventive method embodiment, this
Place repeats no more.
Fig. 4 is a kind of key step schematic diagram of location model construction method of the invention, as shown in figure 4, a kind of positioning mould
Type construction method, including step：S10, stress and strain model is carried out to detection zone, obtain multiple plane grids, and be each plane
The corresponding training location tags for training of grid distribution；S11, successively setting training on the default training position coordinateses
Terminal, trains terminal to send on each that each WAP is received in acquisition testing region training location tags
The signal strength data of signal；S12, respectively by each described training location tags and step S11 in collection each
The signal strength data of the signal that each described training location tags that WAP is received send is used as one group of training sample
Notebook data, generates training dataset, and sends in deep neural network；S13, each group of training that successively training data is concentrated
Signal strength data in sample data is input into the data input layer of the deep neural network, through the deep neural network
The corresponding training result of output；S14, successively the training location tags corresponding for the training result of output are compared
Compared with, deep neural network is trained according to comparative result, using training after deep neural network as the location model.
Preferably, step S14 further includes step：Calculate the corresponding training position of the comparative result
The error of label is put, according to the parameter of the error transfer factor deep neural network, until the corresponding institute of the comparative result
The error convergence of predicted position label is stated in preset range.
It should be noted that with regard to depth in the construction method of location model and above-mentioned WiFi localization methods in the present embodiment
Neural network training method can be found in the above-mentioned explanation for deep neural network training method part based on same design, this
Place does not repeat.
It should be noted that above-described embodiment can independent assortment as needed.The above is only the preferred of the present invention
Embodiment, it is noted that for those skilled in the art, in the premise without departing from the principle of the invention
Under, some improvements and modifications can also be made, these improvements and modifications also should be regarded as protection scope of the present invention.
Claims (9)
1. a kind of WiFi localization methods, it is characterised in that methods described includes step：
In S100, acquisition detection zone, each WAP receives the signal intensity of the signal that client to be detected sends
Data；
S200, the input data layer that the signal strength data obtained in step S100 is input into location model；
S300, based on training after the Internet of location model calculate the signal strength data, in the output layer of location model
Outgoing position forecast set, the position prediction collection include predicted position label and corresponding probit, and the position prediction is concentrated
Position corresponding to the maximum predicted position label of probit is defined as the position that the client to be detected is located.
2. WiFi localization methods as claimed in claim 1, it is characterised in that also include step before step S100：
S000, training in advance deep neural network, using training after deep neural network as the location model.
3. WiFi localization methods as claimed in claim 2, it is characterised in that step S000 further includes step：
S001, stress and strain model is carried out to detection zone, obtain multiple plane grids, and distribute corresponding use for each plane grid
Training location tags in training；
S002, successively setting training terminal, each wireless access in acquisition testing region on the default training position coordinateses
The signal strength data of the signal for training terminal to send on each described training location tags that point is received；
S003, respectively each WAP of collection in each training location tags and step S002 is received
To the signal strength data of signal that sends of each described training location tags as one group of training sample data, generate training
Data set, and send in deep neural network；
Signal strength data in S004, each group of training sample data that concentrates training data successively is input into the depth god
Through the data input layer of network, corresponding training result is exported through the deep neural network；
S005, successively the training location tags corresponding for the training result of output are compared, according to comparative result
Deep neural network is trained, using training after deep neural network as the location model.
4. WiFi localization methods as claimed in claim 3, it is characterised in that step S005 further includes step：
The error of the corresponding training location tags of the comparative result is calculated, according to error transfer factor depth nerve
The parameter of network, until the error convergence of the corresponding predicted position label of the comparative result is in preset range.
5. a kind of WiFi location-servers, it is characterised in that include：
Data acquisition module, receives the signal that client to be detected sends for obtaining each WAP in detection zone
Signal strength data；
Locating module, the signal strength data for obtaining are input into the input data layer of location model, based on training after
The Internet of location model calculate the signal strength data, in the output layer outgoing position forecast set of location model, described
Position prediction collection includes predicted position label and corresponding probit, and the position prediction is concentrated the maximum prediction of probit
Position corresponding to location tags is defined as the position that the client to be detected is located.
6. WiFi location-servers as claimed in claim 5, it is characterised in that also include：
Training module, for training in advance deep neural network, using training after deep neural network as the location model.
7. WiFi location-servers based on deep neural network as claimed in claim 6, it is characterised in that the training mould
Block is further included：
Stress and strain model submodule, for carrying out stress and strain model to detection zone, obtains multiple plane grids, and is each plane net
The corresponding training location tags for training of lattice distribution；
Training dataset generates submodule, for arranging training terminal, collection inspection on the default training position coordinateses successively
Survey the signal of the signal for training terminal to send on each described training location tags that each WAP is received in region
Intensity data, each described instruction that respectively each WAP of each described training location tags and collection is received
Practice the signal strength data of the signal that location tags send as one group of training sample data, generate training dataset, and send into
In deep neural network；
Training prediction submodule, for successively by the signal strength data in each group of training sample data of training data concentration
The data input layer of the deep neural network is input into, and corresponding training result is exported through the deep neural network, and
Successively the predicted position label corresponding for the training result of output is compared, according to comparative result to depth nerve
Network is trained, using training after deep neural network as the location model.
8. a kind of location model construction method, it is characterised in that including step：
S10, stress and strain model is carried out to detection zone, obtain multiple plane grids, and distribute corresponding use for each plane grid
Training location tags in training；
S11, successively setting training terminal, each WAP in acquisition testing region on the default training position coordinateses
The signal strength data of the signal for training terminal to send on each the described training location tags for receiving；
S12, respectively by each described training location tags and step S11 in gather each WAP receive
The signal strength data of signal that sends of each described training location tags as one group of training sample data, generate training number
According to collection, and send in deep neural network；
Signal strength data in S13, each group of training sample data that concentrates training data successively is input into the depth god
Through the data input layer of network, corresponding training result is exported through the deep neural network；
S14, successively the training location tags corresponding for the training result of output are compared, according to comparative result pair
Deep neural network is trained, using training after deep neural network as the location model.
9. location model construction method as claimed in claim 8, it is characterised in that step S14 further includes step：
The error of the corresponding training location tags of the comparative result is calculated, according to error transfer factor depth nerve
The parameter of network, until the error convergence of the corresponding predicted position label of the comparative result is in preset range.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611044936.XA CN106507476A (en) | 2016-11-22 | 2016-11-22 | A kind of WiFi localization methods and server and location model construction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611044936.XA CN106507476A (en) | 2016-11-22 | 2016-11-22 | A kind of WiFi localization methods and server and location model construction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106507476A true CN106507476A (en) | 2017-03-15 |
Family
ID=58328373
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611044936.XA Pending CN106507476A (en) | 2016-11-22 | 2016-11-22 | A kind of WiFi localization methods and server and location model construction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106507476A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107194412A (en) * | 2017-04-20 | 2017-09-22 | 百度在线网络技术（北京）有限公司 | A kind of method of processing data, device, equipment and computer-readable storage medium |
| CN108734734A (en) * | 2018-05-18 | 2018-11-02 | 中国科学院光电研究院 | Indoor orientation method and system |
| CN108989974A (en) * | 2018-04-08 | 2018-12-11 | 深圳清创新科技有限公司 | Animal localization method, device, computer equipment and storage medium |
| CN109151995A (en) * | 2018-09-04 | 2019-01-04 | 电子科技大学 | A kind of deep learning recurrence fusion and positioning method based on signal strength |
| CN109413385A (en) * | 2018-10-23 | 2019-03-01 | 深圳如果技术有限公司 | A kind of video location monitoring method, system and Cloud Server |
| CN110225453A (en) * | 2019-06-24 | 2019-09-10 | 鲸数科技（北京）有限公司 | Mobile terminal locating method, device, electronic equipment and storage medium |
| CN111028358A (en) * | 2018-10-09 | 2020-04-17 | 香港理工大学深圳研究院 | Augmented reality display method and device for indoor environment and terminal equipment |
| CN111044974A (en) * | 2020-03-18 | 2020-04-21 | 腾讯科技（深圳）有限公司 | Indoor positioning method and device based on WiFi signal and storage medium |
| CN111885699A (en) * | 2020-05-21 | 2020-11-03 | 北京嘀嘀无限科技发展有限公司 | Method and system for determining current position of user |
| CN112533137A (en) * | 2020-11-26 | 2021-03-19 | 北京爱笔科技有限公司 | Device positioning method and device, electronic device and computer storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101778399A (en) * | 2010-01-18 | 2010-07-14 | 哈尔滨工业大学 | Method for optimizing WLAN (Wireless Local Area Network) indoor ANN (Artificial Neural Network) positioning based on FCM (fuzzy C-mean) and least-squares curve surface fitting methods |
| CN103476118A (en) * | 2013-09-29 | 2013-12-25 | 哈尔滨工业大学 | WLAN indoor location fingerprint positioning method used for real-time monitoring |
| CN103945533A (en) * | 2014-05-15 | 2014-07-23 | 济南嘉科电子技术有限公司 | Big data based wireless real-time position positioning method |
| CN104427612A (en) * | 2013-08-29 | 2015-03-18 | 北京优锐科技有限公司 | WiFi positioning system based on probability and support vector machine computation and method thereof |
-
2016
- 2016-11-22 CN CN201611044936.XA patent/CN106507476A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101778399A (en) * | 2010-01-18 | 2010-07-14 | 哈尔滨工业大学 | Method for optimizing WLAN (Wireless Local Area Network) indoor ANN (Artificial Neural Network) positioning based on FCM (fuzzy C-mean) and least-squares curve surface fitting methods |
| CN104427612A (en) * | 2013-08-29 | 2015-03-18 | 北京优锐科技有限公司 | WiFi positioning system based on probability and support vector machine computation and method thereof |
| CN103476118A (en) * | 2013-09-29 | 2013-12-25 | 哈尔滨工业大学 | WLAN indoor location fingerprint positioning method used for real-time monitoring |
| CN103945533A (en) * | 2014-05-15 | 2014-07-23 | 济南嘉科电子技术有限公司 | Big data based wireless real-time position positioning method |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107194412A (en) * | 2017-04-20 | 2017-09-22 | 百度在线网络技术（北京）有限公司 | A kind of method of processing data, device, equipment and computer-readable storage medium |
| CN108989974A (en) * | 2018-04-08 | 2018-12-11 | 深圳清创新科技有限公司 | Animal localization method, device, computer equipment and storage medium |
| CN108989974B (en) * | 2018-04-08 | 2020-12-11 | 深圳一清创新科技有限公司 | Animal positioning method, animal positioning device, computer equipment and storage medium |
| CN108734734A (en) * | 2018-05-18 | 2018-11-02 | 中国科学院光电研究院 | Indoor orientation method and system |
| CN109151995A (en) * | 2018-09-04 | 2019-01-04 | 电子科技大学 | A kind of deep learning recurrence fusion and positioning method based on signal strength |
| CN109151995B (en) * | 2018-09-04 | 2020-12-01 | 电子科技大学 | Deep learning regression fusion positioning method based on signal intensity |
| CN111028358A (en) * | 2018-10-09 | 2020-04-17 | 香港理工大学深圳研究院 | Augmented reality display method and device for indoor environment and terminal equipment |
| CN111028358B (en) * | 2018-10-09 | 2024-02-27 | 香港理工大学深圳研究院 | Indoor environment augmented reality display method and device and terminal equipment |
| CN109413385A (en) * | 2018-10-23 | 2019-03-01 | 深圳如果技术有限公司 | A kind of video location monitoring method, system and Cloud Server |
| CN110225453A (en) * | 2019-06-24 | 2019-09-10 | 鲸数科技（北京）有限公司 | Mobile terminal locating method, device, electronic equipment and storage medium |
| CN111044974A (en) * | 2020-03-18 | 2020-04-21 | 腾讯科技（深圳）有限公司 | Indoor positioning method and device based on WiFi signal and storage medium |
| CN111044974B (en) * | 2020-03-18 | 2020-07-07 | 腾讯科技（深圳）有限公司 | Indoor positioning method and device based on WiFi signal and storage medium |
| CN111885699A (en) * | 2020-05-21 | 2020-11-03 | 北京嘀嘀无限科技发展有限公司 | Method and system for determining current position of user |
| CN112533137A (en) * | 2020-11-26 | 2021-03-19 | 北京爱笔科技有限公司 | Device positioning method and device, electronic device and computer storage medium |
| CN112533137B (en) * | 2020-11-26 | 2023-10-17 | 北京爱笔科技有限公司 | Positioning method and device of equipment, electronic equipment and computer storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106507476A (en) | A kind of WiFi localization methods and server and location model construction method | |
| CN106535326A (en) | WiFi locating method based on depth neural network and server | |
| CN106793067A (en) | A kind of many floor indoor orientation methods and server based on joint network | |
| CN106792769A (en) | A kind of WiFi localization methods and server and location model method for building up | |
| CN106535134A (en) | Multi-room locating method based on wifi and server | |
| CN106792553A (en) | A kind of many floor location methods and server based on wifi | |
| CN108696932B (en) | Outdoor fingerprint positioning method using CSI multipath and machine learning | |
| CN104239556B (en) | Adaptive trajectory predictions method based on Density Clustering | |
| CN104270816B (en) | The adaptive dynamic fingerprint base construction method of LED visible light indoor locating system | |
| CN106792507A (en) | A kind of WiFi localization methods and server based on network data | |
| CN107466043A (en) | A kind of azimuthal method and apparatus for determining antenna for base station | |
| CN106982436A (en) | The detection method of adjustment and device of a kind of indoor and outdoor network coverage | |
| Han et al. | Joint deployment optimization and flight trajectory planning for UAV assisted IoT data collection: A bilevel optimization approach | |
| CN106793070A (en) | A kind of WiFi localization methods and server based on reinforcement deep neural network | |
| CN106993273A (en) | Based on distance weighted and genetic optimization DV Hop localization methods | |
| CN102395194B (en) | ANFIS (Adaptive Neural Fuzzy Inference System) indoor positioning method based on improved GA(Genetic Algorithm) optimization in WLAN (Wireless Local Area Network) environment | |
| CN104039008B (en) | A kind of hybrid locating method | |
| CN106250979A (en) | A kind of particle group optimizing method based on mutual information similarity Strategy | |
| CN106357720A (en) | System and method for displaying data on basis of internet of things | |
| CN108882174A (en) | Mobile terminal locating method, device, electronic equipment and storage medium | |
| CN106792506A (en) | A kind of WiFi localization methods and server | |
| CN107736071A (en) | It is determined that the apparatus and method of the wave beam for guiding the signal that be sent to user equipment | |
| CN107968987A (en) | RSSI weighted mass center localization methods based on definite integral combining environmental parameter | |
| CN106488559A (en) | A kind of outdoor positioning method based on visibility and server | |
| CN106993296A (en) | The performance estimating method and device of terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20201029 Address after: 318015 no.2-3167, zone a, Nonggang City, no.2388, Donghuan Avenue, Hongjia street, Jiaojiang District, Taizhou City, Zhejiang Province Applicant after: Taizhou Jiji Intellectual Property Operation Co.,Ltd. Address before: 201616 Shanghai city Songjiang District Sixian Road No. 3666 Applicant before: Phicomm (Shanghai) Co.,Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170315 |