CN106992822A - A kind of localization method of the blind node of wireless sensor network - Google Patents
A kind of localization method of the blind node of wireless sensor network Download PDFInfo
- Publication number
- CN106992822A CN106992822A CN201710197628.9A CN201710197628A CN106992822A CN 106992822 A CN106992822 A CN 106992822A CN 201710197628 A CN201710197628 A CN 201710197628A CN 106992822 A CN106992822 A CN 106992822A
- Authority
- CN
- China
- Prior art keywords
- node
- blind
- blind node
- rss
- received signal
- 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
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/20—Monitoring; Testing of receivers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/023—Services making use of location information using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
Abstract
N anchor node not point-blank, n >=3 are determined the invention discloses a kind of localization method of the blind node of wireless sensor network, including step;Select some blind node at random from m blind nodes, be designated as B (xest, yest);Measure all anchor node i, j, k ... and described blind node B (xest, yest) between received signal strength valueCalculate the difference between maximum received signal strength value and each received signal strength value;Step 5) calculate blind node estimate coordinate;Final coordinate (the x of blind node is calculated by maximum-likelihood estimatorrss,yrss).The present invention improves the locating speed of the blind node of wireless sensor network on the premise of positioning precision is not reduced.
Description
Technical field
Particularly it is a kind of localization method of the blind node of wireless sensor network the present invention relates to wireless sensor network.
Background technology
Positioning is the Location-Unknown that positioning node is gone by information such as the positions of some anchor nodes (known to the position of node)
Blind node process, occupied an important position in the research of wireless sensor network.In general, the positioning of node needs
By some form of communication, such as radio or acoustic communication between anchor node and blind node.Sensor node localization mistake
Cheng Zhong, blind node is after obtaining the distance for closing on anchor node or closing on the angle between anchor node and blind node, and reselection is closed
Suitable algorithm calculates the position of oneself.According to the distance of measurement actual node whether is needed in position fixing process, location algorithm
It is divided into:Location algorithm based on distance and apart from unrelated location algorithm.The location algorithm accuracy for being typically based on distance is higher,
Generally pay the utmost attention to, its main ranging technology included includes being based on received signal strength (Received Signal
Strength, RSS) location algorithm, based on arrival time (Time of Arrival, TOA) location algorithm, based on arrival
The location algorithm of time difference (Time Difference of Arrival, TDOA) and based on angle of arrival (Angle of
Arrival, AOA) location algorithm etc..
In known localization method, the method based on RSS is most widely used, because this kind of method has low be calculated as
This, low-power consumption, low hardware complexity.In addition, in the case where nodal distance is relatively near, the localization method of received signal strength can be with
Obtain higher precision.Wherein, path loss model defines the relation between distance and received signal strength, and this also to lead to
Anchor node positioning blind node is crossed to be possibly realized.In all location algorithms, the maximum likelihood estimate based on RSS is higher than big absolutely
Other most algorithms.But it is due to need to use conjugate gradient algorithms when asking maximum-likelihood estimator, by successive ignition,
Therefore its efficiency is heavily dependent on the degree of accuracy that blind node location is estimated.Good predictive algorithm can reduce iteration
Number of times, improves the speed of positioning.
By the retrieval discovery to prior art, a kind of base is disclosed in Chinese invention patent publication number CN104678351
In the indoor locating system solution of ZigBee technology, with reference to based on signal intensity (RSS) and scene fingerprint location method, use
RSS " scene characteristic information " sets up " finger print information " database, finally utilizes arest neighbors as " finger print information " of positioning scene
Occupy matching algorithm and realize indoor positioning.Mainly include two steps:Step 1: off-line phase.Fingerprint database is set up in completion, and record is not
With the scenario parameters of position, take blind node to roam in localization region, record signal intensity that each position is subject to and other refer to
Line information, then recorded these information and current roaming status in database in pairs;Step 2: on-line stage.By obtaining
" finger print information " of blind node is obtained, " finger print information " in information and date storehouse is matched, matching degree highest position is selected,
Determine physical location.The information that this method is gathered when positioning first to blind node is not also very perfect, and is used
Be fingerprint matching algorithm, in the case where sample is few, ratio of precision maximum- likelihood estimation can be weaker.
Chinese invention patent publication number CN105093174 discloses a kind of positioning based on the 2.5G wireless network signal profits and losses
Algorithm.It is broadly divided into five steps:Step 1: obtaining the antenna that mobile terminal carries out signal transmission with base station in 2.5G wireless networks
Gain and loss;Step 2: selecting the base station that several can be with communication of mobile terminal, calculate and moved eventually with each base station communication
The estimation coordinate value at end;Step 3: passing through the estimated coordinates value of mobile terminal and the seat with each base station of communication of mobile terminal
Scale value information, calculates the average distance estimation error obtained between mobile terminal and base station;Step 4: according to antenna gain, damaging
Consumption, average distance and base station density, calculate the coordinate value for obtaining each mobile terminal;Step 5: to several mobile terminals
Coordinate is summed, then average, obtains the true coordinate value of mobile terminal.The method of this invention take into account antenna gain and signal is damaged
The factor of the influence positioning such as consumption, but the location algorithm used is mainly simple weighted mass center algorithm, obtained result accuracy
It is not high.
In all location algorithms, based on received signal strength (received signal strength, RSS) most
The maximum-likelihood estimation technique is higher than other most algorithms.But meanwhile, maximum likelihood estimate needs to use conjugate gradient algorithms, leads to
Iteration after several times obtains maximum-likelihood estimator, is positioned such that algorithm will be to serious if none of good estimate
Influence locating speed.
The content of the invention
The present invention is directed to above-mentioned the deficiencies in the prior art, it is proposed that a kind of positioning side of the blind node of wireless sensor network
Method, while using based on the maximum- likelihood estimation for receiving signal intensity, introduces the algorithm estimated, improves algorithm
Arithmetic speed, and obtain the result close to minimal error in theory.
The present invention technical solution be:
A kind of localization method of the blind node of wireless sensor network, comprises the following steps:
Step 1) the n anchor node of determination not point-blank, n >=3;
Step 2) select some blind node at random from m blind nodes, it is designated as B (xest, yest);
Step 3) measure all anchor node i, j, k ... and described blind node B (xest, yest) between received signal strength
Value
Step 4) the maximum node of the anchor node nearest from blind node B, i.e. received signal strength value is found, it is used as reference node
Point, and the received signal strength value of reference mode is made for Pmax, calculate maximum received signal strength value and each reception signal is strong
Difference between angle value, formula is as follows:
In formula, dmaxRepresent the distance between reference mode and blind node B, diRepresent between anchor node i and blind node away from
From;
Step 5) blind node estimates coordinate (xest, yest) calculation formula it is as follows:
In formula, wiFor weight,
Step 6) pass through final coordinate (x of the maximum-likelihood estimator by the described blind node B of following equation calculatingrss,
yrss):
(xrss,yrss)=arg max f (x, y) (10)
In formula:The estimated value of distance between anchor node i and blind node is represented,For reality between anchor node i and blind node
The received signal strength value of measurement, XσIt is that an average is 0, standard deviation is σdBGaussian noise, i.e. Xσ~N (0, σdB),
Argmaxf (x, y) represents the value of the x and y when f (x, y) takes maximum, and ln (a) represents to take natural logrithm to a;
First by the received signal strength value of measurementFormula 8 is substituted into obtainValue, then carried out finally with conjugate gradient algorithms
Solution, by multiple iteration, obtain the coordinate estimate (x of the maximum of formula 10, the i.e. blind noderss,yrss);
Step 7) return to step 2), calculate the coordinate estimate of other blind nodes.
Compared with prior art, the beneficial effects of the invention are as follows improve wireless sensing on the premise of positioning precision is not reduced
The locating speed of the blind node of device network.
Brief description of the drawings
Fig. 1 is the flow chart of the localization method of the blind node of wireless sensor network of the present invention;
Fig. 2 is the physical location schematic diagram of anchor node of the embodiment of the present invention and blind node;
Fig. 3 is weighted mass center algorithm schematic diagram;
The physical location of the blind nodes of Fig. 4 and calculating position.
Embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings.
The main flow of improved location algorithm proposed by the present invention is as shown in figure 1, main process includes:Initialization, in advance
Estimate and maximum- likelihood estimation implementation.Each key step includes some subprocess.
The embodiment of the present invention:
The localization method of the blind node of wireless sensor network of the present invention, comprises the following steps:
1. node location is arranged.In 50m*50m simulated environment, one has 25 nodes, wherein there is 4 anchor nodes
With 21 blind nodes, wherein 4 anchor nodes are placed in 4 corners, the position of blind section is then generated at random.Fig. 2 shows anchor section
The physical location of point and blind node.
2. just starting first to select one of those blind node at random from 21 blind nodes, blind node B is designated as, this is calculated blind
Received signal strength (received signal strength, RSS) between node and anchor node.Here road is introduced
Footpath loss model, the intensity that path loss model is based on signal is constantly weak with the increase of distance.Distance is believed with RSS
Number functional relation as shown in Equation 1.
In equation 1, di is the distance between anchor node i and the blind node, and d0 is reference distance, is usually arranged as 1m;
Pi is RSS values when distance is di;P0 is RSS values when distance is 1m;N is path loss index, relevant with specific environment,
Typically between 2 and 6.N=2, d0=1m are set, one group of RSS value can be released.
Following step 3 and step 4 are performed parallel, for being compared.
3. performing common weighted mass center algorithm, obtain blind node estimates position.Weighted mass center location algorithm anchor section
RSS values between point and blind node are as weight, in order to estimate the position of blind node, it is necessary first to obtain blind node and some anchors
RSS values between node.RSS values are bigger, illustrate that two nodes are nearer.Therefore, RSS values can reflect that each anchor node is being estimated
Calculate contribution during blind node location.The schematic diagram of weighted mass center algorithm is as shown in Figure 2.It is known by n in wireless sensor network
The anchor node of position, their coordinate is (x1, y1), (x1, y1) ..., (xn, yn).Between these anchor nodes and blind node
RSS values are P1, P2 ..., Pn.The coordinate of blind node is can be obtained by by the two conditions, as shown in Equation 2.
4. performing improved weighted mass center algorithm, obtain blind node estimates position.Common weighted mass center algorithm incision
Point is to regard RSS size as weight.The relatively directly also easily operation of this method.However, this method is not accounted for very
Real environment, the result estimated and actual result have very big difference.Therefore, present invention also proposes the improved weighting of one kind
Centroid algorithm, this method consideration simulates actual environment with path loss model, and is estimated as weight with RSS difference blind
The position of node.According to formula 1 as can be seen that the difference of the RSS values between anchor node i and anchor node j and blind node can be used
Formula 3 is represented.
In order to obtain weight, the maximum node of the anchor node nearest from blind node B, i.e. RSS values is found first, reference is used as
Node.The RSS values of reference mode are set to Pmax.Difference for some other anchor nodes i, RSS is expressed as formula 4.
According to formula 3 and 4, the expression formula 5 of weight can be obtained.
Likewise, estimating coordinate according to what formula 2 can obtain final blind node.With common weighted mass center algorithm phase
Weight is calculated with RSS relative different than, improved weighted mass center algorithm, and has used path loss model, and this is just very
Influence of the environmental factor to result is eliminated in big degree.
5. carrying out final positioning with maximum- likelihood estimation, the coordinate of blind node is obtained.Pass through estimating above
Journey, can obtain blind node estimates coordinate (xest, yest).Next blind node is just obtained by maximum- likelihood estimation
Estimated position.
In equation 1, di the distance between is anchor node i with the blind node, may also indicate that for:
The influence of environmental change is not considered loss model in data in formula 1, if it is considered that the influence of environmental change,
Path loss obeys logarithm normal distribution, and revised path loss model formula 7 is represented.
XσVariable is that an average is 0, and standard deviation is σdBGaussian noise, i.e. Xσ~N (0, σdB).Therefore, distance is estimated
Calculation valueIt can be represented with formula 8
In order to build a maximal possibility estimation equation, first according to formula 1, in the coordinate and blind section of known anchor node
Point estimate coordinate on the premise of, instead release one group of RSS value, use PiRepresent.Then with the RSS values actually measuredWith reference to rise
Come, possibility predication equation can be obtained, as shown in Equation 9
Formula 6,7 is combined, then by the logarithm of the derivation of equation 9, maximum is taken, is simplified, maximum is obtained seemingly
Right estimate.The maximal possibility estimation of blind node coordinate can be represented with formula 10, formula 11:
(xrss,yrss)=arg max f (x, y) (10)
In order to calculate the maximum of formula 11, namely required blind node coordinate.First, the RSS values of measurement are passed throughSubstituting into formula 8 can obtain in formula 11Value.Then the coordinate for estimating the blind node that process is obtained above is passed through
(xest,yest).3rd step, final solution is carried out with conjugate gradient algorithms, by multiple iteration, can obtain formula 10
Maximum, namely the blind node coordinate estimate (xrss,yrss)。
Then repeat step 2 arrives step 5, calculates the coordinate estimate of other blind nodes.
By carrying out the execution of repeatedly improved location algorithm, the position of all blind nodes, i.e. (x can be obtainedrss1,
yrss1),(xrss2,yrss2)…(xrssm,yrssm).Meanwhile, the physical location for obtaining these blind nodes can be measured, they are
(xBN1,yBN1),(xBN2,yBN2)…(xBNm,yBNm).Mean square error can be obtained by this two groups of data, as shown in Equation 12.
Square error is to evaluate the important indicator of positioning precision.
The effect of algorithm is tested by emulating.Emulation is carried out under MATLAB environment, emulates the ring set
Border is 50m*50m environment, and one has 25 nodes, wherein having 4 anchor nodes and 21 blind nodes, wherein 4 anchor sections
Point is placed in 4 corners, and the position of blind section is then generated at random.The position of blind node that Fig. 2 shows anchor node and generated at random
Put.
In simulations, different σ are setdBValue, it is therefore an objective to obtain mean square error with σdBThe curve of change.Then, respectively
The weighted mass center algorithm for performing weighted mass center algorithm and improvement is estimated, and obtain blind node estimates position.Next, can
To perform-conjugate gradient algorithms to formula 9, maximum-likelihood estimator is obtained by iteration several times, also just calculated
Blind node coordinate, while writing down the iterations of distinct methods.Finally, the coordinate and reality of the blind node obtained with calculating
The coordinate of blind node calculates mean square error in the lump.In order to reduce the error in calculating, No. 100 meters are carried out to each blind node
Calculate, average.Coordinate and actual coordinate that blind node is calculated are shown in Fig. 4.Clearly opened up in picture to allow
Show, therefore the result of first 10 times in 100 experiments is only have chosen in picture and be shown.
By emulation, the iterations and mean square error of algorithms of different can be compared, the object mainly compared includes common
Maximum- likelihood estimation, the maximum- likelihood estimation estimated by weighted mass center algorithm, and the present invention proposes
Improved location algorithm, i.e., the maximum- likelihood estimation estimated by the weighted mass center algorithm of improvement.
Table 1 is illustrated in maximum- likelihood estimation, and the iterations of conjugate gradient algorithms is with σdBThe change of change
Trend.
The difference of table 1 σdBCorresponding iterations result
It can be seen from the results above that improved location algorithm has following advantage and gain effect:
1st, the localization method based on RSS inherently has a low calculating cost, low-power consumption, the advantage such as low hardware complexity.In addition,
In the case where nodal distance is relatively near, the localization method of received signal strength can obtain higher precision.Meanwhile, maximum likelihood
Algorithm for estimating is in estimation accuracy better than the core algorithm in most of other location algorithms, therefore the present invention using estimated
RSS maximum- likelihood estimation.
2nd, improved location algorithm can largely reduce the iterations of conjugate gradient algorithms.Such as, in σdBFor
When 2dB, it is more than 23 using the iterations required for maximum- likelihood estimation again after random initializtion, and with commonly
Weighted mass center algorithm estimated after, iterations drops to 14.5 times or so, the maximal possibility estimation with random initializtion
Algorithm is compared and saved 8.5 times.And after being estimated with the weighted mass center algorithm after improvement, iterations drops to 12 times,
It is all more efficient compared with the above two.Simultaneously it can also be seen that taking different σdBValue, the iterations of algorithm saving is hardly sent out
Changing.Because σdBThere is no any influence to estimating process.
3rd, final mean square error result is approximately equal to the minimum value in the theory of mean square error, i.e. carat Metro lower bound,
This also illustrates that algorithm proposed by the present invention reduces iterations under the premise of ensureing correctly, has saved the time.
It should be noted last that, the above embodiments are merely illustrative of the technical solutions of the present invention and it is unrestricted, although ginseng
The present invention is described in detail according to preferred embodiment, it will be understood by those within the art that, can be to invention
Technical scheme is modified or equivalent substitution, and without departing from the spirit and scope of technical solution of the present invention, it all should cover
Among scope of the presently claimed invention.
Claims (1)
1. a kind of localization method of the blind node of wireless sensor network, it is characterised in that this method comprises the following steps:
Step 1) the n anchor node of determination not point-blank, n >=3;
Step 2) select some blind node at random from m blind nodes, it is designated as B (xest, yest);
Step 3) measure all anchor node i, j, k ... and described blind node B (xest, yest) between received signal strength value
Step 4) the maximum node of the anchor node nearest from blind node B, i.e. received signal strength value is found, as reference mode,
And make the received signal strength value of reference mode for Pmax, calculate maximum received signal strength value and each received signal strength value
Between difference, formula is as follows:
In formula, dmaxRepresent the distance between reference mode and blind node B, diRepresent the distance between anchor node i and blind node;
Step 5) blind node estimates coordinate (xest, yest) calculation formula it is as follows:
In formula, wiFor weight,
Step 6) pass through final coordinate (x of the maximum-likelihood estimator by the described blind node B of following equation calculatingrss,yrss):
(xrss,yrss)=arg max f (x, y) (10)
In formula:The estimated value of distance between anchor node i and blind node is represented,Actually to be measured between anchor node i and blind node
Received signal strength value, XσIt is that an average is 0, standard deviation is σdBGaussian noise, i.e. Xσ~N (0, σdB), argmaxf
(x, y) represents the value of the x and y when f (x, y) takes maximum, and ln (a) represents to take natural logrithm to a;
First by the received signal strength value of measurementFormula 8 is substituted into obtainValue, then carries out with conjugate gradient algorithms asking finally
Solution, by multiple iteration, obtains the coordinate estimate (x of the maximum of formula 10, the i.e. blind noderss,yrss);
Step 7) return to step 2), calculate the coordinate estimate of other blind nodes.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710197628.9A CN106992822A (en) | 2017-03-29 | 2017-03-29 | A kind of localization method of the blind node of wireless sensor network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710197628.9A CN106992822A (en) | 2017-03-29 | 2017-03-29 | A kind of localization method of the blind node of wireless sensor network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106992822A true CN106992822A (en) | 2017-07-28 |
Family
ID=59412360
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710197628.9A Pending CN106992822A (en) | 2017-03-29 | 2017-03-29 | A kind of localization method of the blind node of wireless sensor network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106992822A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108471596A (en) * | 2018-03-06 | 2018-08-31 | 广州杰赛科技股份有限公司 | Node positioning method, node locating device, computer equipment and storage medium |
CN109959918A (en) * | 2019-02-22 | 2019-07-02 | 西安电子科技大学 | A kind of method, apparatus and computer storage medium of solid bodies positioning |
CN110012420A (en) * | 2019-02-14 | 2019-07-12 | 中国科学院信息工程研究所 | The method and device towards unknown transmission power malice AP based on RSS |
CN110493740A (en) * | 2018-05-14 | 2019-11-22 | 中国移动通信有限公司研究院 | A kind of indoor orientation method and location-server |
CN114895242A (en) * | 2022-07-12 | 2022-08-12 | 中国人民解放军国防科技大学 | Radiation source positioning method in complex electromagnetic environment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102209382A (en) * | 2011-05-18 | 2011-10-05 | 杭州电子科技大学 | Wireless sensor network node positioning method based on received signal strength indicator (RSSI) |
JP2014534425A (en) * | 2011-10-17 | 2014-12-18 | コミサリア ア レネルジィ アトミーク エ オ ゼネ ルジイ アルテアナティーフCommissariata L’Energie Atomique Et Aux Energies Alternatives | Node positioning method for wireless network |
CN105072581A (en) * | 2015-08-26 | 2015-11-18 | 四川星网云联科技有限公司 | Indoor positioning method of path attenuation coefficient based database construction |
-
2017
- 2017-03-29 CN CN201710197628.9A patent/CN106992822A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102209382A (en) * | 2011-05-18 | 2011-10-05 | 杭州电子科技大学 | Wireless sensor network node positioning method based on received signal strength indicator (RSSI) |
JP2014534425A (en) * | 2011-10-17 | 2014-12-18 | コミサリア ア レネルジィ アトミーク エ オ ゼネ ルジイ アルテアナティーフCommissariata L’Energie Atomique Et Aux Energies Alternatives | Node positioning method for wireless network |
CN105072581A (en) * | 2015-08-26 | 2015-11-18 | 四川星网云联科技有限公司 | Indoor positioning method of path attenuation coefficient based database construction |
Non-Patent Citations (1)
Title |
---|
CHAOCHEN WANG等: "An Improved Localization Framework based on Maximum Likelihood for Blind WSN Nodes", 《2015 IEEE 17TH INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING AND COMMUNICATIONS》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108471596A (en) * | 2018-03-06 | 2018-08-31 | 广州杰赛科技股份有限公司 | Node positioning method, node locating device, computer equipment and storage medium |
CN110493740A (en) * | 2018-05-14 | 2019-11-22 | 中国移动通信有限公司研究院 | A kind of indoor orientation method and location-server |
CN110493740B (en) * | 2018-05-14 | 2021-01-15 | 中国移动通信有限公司研究院 | Indoor positioning method and positioning server |
CN110012420A (en) * | 2019-02-14 | 2019-07-12 | 中国科学院信息工程研究所 | The method and device towards unknown transmission power malice AP based on RSS |
CN110012420B (en) * | 2019-02-14 | 2020-07-07 | 中国科学院信息工程研究所 | RSS (received Signal Strength) -based method and device for malicious access point with unknown transmission power |
CN109959918A (en) * | 2019-02-22 | 2019-07-02 | 西安电子科技大学 | A kind of method, apparatus and computer storage medium of solid bodies positioning |
CN114895242A (en) * | 2022-07-12 | 2022-08-12 | 中国人民解放军国防科技大学 | Radiation source positioning method in complex electromagnetic environment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106992822A (en) | A kind of localization method of the blind node of wireless sensor network | |
CN102427603B (en) | Positioning method of WLAN (Wireless Local Area Network) indoor mobile user based on positioning error estimation | |
CN103068035B (en) | A kind of wireless network localization method, Apparatus and system | |
CN106793087B (en) | Array antenna indoor positioning method based on AOA and PDOA | |
CN106646356A (en) | Nonlinear system state estimation method based on Kalman filtering positioning | |
CN106793082A (en) | A kind of positioning of mobile equipment method in WLAN/ bluetooth heterogeneous network environments | |
CN108650628A (en) | The indoor orientation method that ranging and fingerprint based on Wi-Fi network are combined | |
Gan et al. | Deep learning for weights training and indoor positioning using multi-sensor fingerprint | |
CN106912105A (en) | 3-D positioning method based on PSO_BP neutral nets | |
CN106093854A (en) | A kind of method of air quality monitoring spot net location based on RSSI range finding | |
CN103338516A (en) | Two-step positioning method of wireless sensor network based on total least squares | |
CN104918214B (en) | A kind of secret protection quantitative analysis method in location-based service | |
CN103916954A (en) | Probability locating method and locating device based on WLAN | |
Xu et al. | Self-adapting multi-fingerprints joint indoor positioning algorithm in WLAN based on database of AP ID | |
CN110536245A (en) | A kind of indoor wireless positioning method and system based on deep learning | |
CN113438731B (en) | Positioning method and system based on signal quality control and characteristic fingerprint enhancement | |
CN106961659A (en) | The bluetooth localization method of one species fingerprint | |
CN106162559A (en) | For estimating equipment and the method for position in a wireless communication system | |
CN108737952A (en) | Based on the improved polygon weighted mass center localization method of RSSI rangings | |
Hoang et al. | A hidden Markov model for indoor user tracking based on WiFi fingerprinting and step detection | |
CN110247719A (en) | The playback of 5G time varying channel and emulation mode based on machine learning | |
Mahfouz et al. | Kernel-based localization using fingerprinting in wireless sensor networks | |
Zhang et al. | Generative adversarial networks based digital twin channel modeling for intelligent communication networks | |
CN104050254B (en) | The method that feature database in 3D rooms is built using house data | |
CN106125037A (en) | Indoor wireless focus based on WiFi signal intensity and Micro Model backtracking localization method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170728 |