CN103929810A - DV-Hop wireless sensor network node locating method based on wavelet neural network - Google Patents
DV-Hop wireless sensor network node locating method based on wavelet neural network Download PDFInfo
- Publication number
- CN103929810A CN103929810A CN201410193590.4A CN201410193590A CN103929810A CN 103929810 A CN103929810 A CN 103929810A CN 201410193590 A CN201410193590 A CN 201410193590A CN 103929810 A CN103929810 A CN 103929810A
- Authority
- CN
- China
- Prior art keywords
- node
- anchor
- jumping
- distance
- wireless sensor
- 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.)
- Granted
Links
Landscapes
- Mobile Radio Communication Systems (AREA)
Abstract
The invention provides a DV-Hop wireless sensor network node locating method based on a wavelet neural network. The method comprises the following steps that firstly, a wireless sensor network is initialized, the minimum hop number between every two anchor nodes and the minimum hop numbers between unknown nodes and the anchor nodes are obtained, and the hop distance between of each hop between every two anchor nodes is calculated through the known distance between every two anchor nodes and the obtained minimum hop numbers; secondly, the wavelet neural network is initialized to obtain the hop distance of each hop suitable for the whole wireless sensor network is obtained; thirdly, the obtained minimum hop numbers between the unknown nodes and the anchor nodes in the step one are multiplied by the obtained hop distance of each hop of the whole wireless sensor network in the step two to obtain the distance between the unknown nodes and the anchor nodes; fourthly, the coordinates of the unknown nodes are obtained through a least square method and a three-side measurement method according to the obtained distance between the unknown nodes and the anchor nodes in the step three, and therefore locating is achieved. The locating performance is good, locating errors are small, locating precision is high, and network adaptability is strong.
Description
Technical field
The present invention relates to wireless sensor network technology field, be specifically related to the DV-Hop wireless sensor network node positioning method based on small echo nerve.
Background technology
Wireless sensor network (Wireless Sensor Network, WSN) location technology is one of main support technology of wireless sensor network, the process of location is the node that utilizes a few locations known, determines the position of unknown node according to certain algorithm or certain mechanism.In these algorithms, one class is the location algorithm based on hardware range finding, this class algorithm needs distance or the angle information of point-to-point between the commercial measurement nodes such as RSSI, AOA, TOA or TDOA, and then by methods such as trilateration, triangulation or maximum likelihood estimates, comes the position of computing node; Another kind of is the location algorithm that does not need additional hardware ranging technology to support, only need to node, position according to the information such as connectedness of network, and conventional location technology has Centroid, CONVEXPROGRAMMING, APIT and DV-Hop etc.Location algorithm based on range finding, although can obtain relatively high positional accuracy, but hardware range finding need to increase extra, complicated hardware, and increased the energy consumption of sensor node self and whole sensor network, and be not suitable for low cost, low-power consumption, the wireless sensor network that software and hardware resources is limited, this just makes at energy, the location algorithm of the aspects such as expansivity and range-independence receives the general concern of industrial quarters and academia, this DV-Hop (Distance Vector Hop) algorithm that wherein people such as NICULESCU proposes is the location algorithm without additional hardware range finding of studying at present and being most widely used, DV-Hop algorithm is a kind of location algorithm without the auxiliary range finding of hardware that utilizes the principle of distance vector route and GPS positioning combination to put forward, the core of algorithm is to utilize internodal estimated distance to replace actual measurement distance, and estimated distance to be the product of jumping distance by jumping figure between unknown node and anchor node and every jumping obtain, by the coordinate of internodal estimated distance and anchor node, calculate afterwards the elements of a fix of unknown node.The advantage of DV-Hop location algorithm is that computational process is simple, hardware requirement is low, but also have larger shortcoming, by analysis and research, the source that error produces is mainly simultaneously: traditional algorithm calculate average every jumping jump apart from not being suitable for actual whole radio sensing network.In wireless sensor network, node random distribution, distance between each node is different in size, node density distribution is sparse difference also, and in DV-Hop algorithm, every jumping of jumping apart from value as whole network with every jumping that method of average estimation obtains is jumped apart from value, obviously will there is larger error with actual distance in this distance estimating, thereby cause the position error of DV-Hop algorithm larger, and positioning precision is low.
Summary of the invention
Technical problem to be solved by this invention is to provide the wireless sensor network node positioning method of the DV-Hop based on small echo nerve, and real-time is good, and position error is little, and positioning precision is high, and network applicability is strong.
For solving above-mentioned existing technical problem, the present invention adopts following scheme: the DV-Hop wireless sensor network node positioning method based on small echo nerve (being called for short WNNDV-Hop algorithm), comprises the following steps:
Step 1: wireless sensor network initialization, obtain the minimum hop count between minimum hop count, unknown node and the anchor node between every two anchor nodes, utilize every jumping that the minimum hop count of known distance between every two anchor nodes and acquisition is calculated between every two anchor nodes to jump distance;
Step 2: wavelet neural network initialization, the every jumping between every two anchor nodes in step 1 is jumped apart from carrying out data processing, obtain being applicable to every jumping jumping distance of whole wireless sensor network;
Step 3: distance is jumped in every jumping of the whole wireless sensor network that utilizes the unknown node that obtains in step 1 and the minimum hop count between anchor node to be multiplied by obtain in step 2, show that unknown node is to the distance between anchor node;
Step 4: to the distance between anchor node, utilize minimum according to the unknown node obtaining in step 3
Square law and trilateration are tried to achieve the coordinate of unknown node, thereby complete location.
As preferably, in described step 1, the initialized concrete steps of radio sensing network are: first anchor node is to the grouping of its each neighbor node broadcast self-position information, and its jumping figure is initialized as to 0, receiving node is recorded to the minimum hop count of each anchor node, ignore the grouping from the larger jumping figure of same anchor node simultaneously, then its jumping figure value is added to 1 and be transmitted to neighbor node, thereby draw the minimum hop count between minimum hop count, unknown node and the anchor node between every two anchor nodes; Then according to formula
J ≠ i, with the known distance between two anchor nodes, divided by the minimum hop count between corresponding two anchor nodes, distance, wherein (x are jumped in the every jumping obtaining between two anchor nodes
i, y
i), (x
j, y
j) be anchor node i, the coordinate of j, h
ijanchor node i, the minimum hop count between j (i ≠ j).In this step, receiving node records the minimum hop count of each anchor node, the reason of simultaneously ignoring from the grouping of the larger jumping figure of same anchor node is: radio sensing network node is for arbitrarily shedding, network node skewness, node topology structure has randomness, between network node, distance distance differs, between anchor node and unknown node, distance is far away, jumping figure is more, the possibility of the jumping section path deviation straight line between them is just larger, consequent estimated distance error is also just larger, so choose from the anchor node of receiving node minimum hop count, select the anchor node less with node jumping figure to participate in location Calculation, not only can avoid in advance the accumulation of error on the impact of location Calculation afterwards, also can reduce certain amount of calculation simultaneously, reduce the consumption of whole radio sensing network energy, extend the useful life of network, can also accelerate computational speed, improve node locating real-time.
As preferably, in described step 2, the initialized concrete steps of wavelet neural network are: the every jumping using between every two anchor nodes that obtain in step 1 is jumped apart from the input data as the neural training network of small echo, and input data are taked interval grouping and moved the mode that window formula chooses and carry out preliminary treatment, then input data and object vector are normalized; Then the transfer function using wavelet function Mallet as hidden layer is tried to achieve the output valve of hidden layer, and what output layer used is on the basis of Sigmoid function, to add a contraction-expansion factor c to try to achieve output valve; Introduce factor of momentum aerfa revises each weights coefficient, threshold value coefficient simultaneously; Finally with Err_NetOut, judge whether all samples calculate completely; Err_NetOut represents the error of the quadratic sum of output valve and desired value, uses Err_NetOut and count as the threshold value of wavelet neural network algorithmic statement, and distance is jumped in the every jumping that obtains whole wireless sensor network.In this step, data taked to interval grouping and moved the mode that window formula chooses and process, thereby can increase like this its utilance to data in the situation that data are less, promoting the adaptability of radio sensing network to environment; In this step, adopted wavelet neural network intelligent algorithm, be because wavelet neural network is inherited the advantage of wavelet analysis and neural net simultaneously: wavelet analysis is enjoyed the analyzing and processing of multiresolution by flexible or translation to signal or function, thereby signal, function are being carried out to the extraction of local message, and analysis aspect advantage is huge; Neural net has self study, self adaptation, self-organizing feature, the feature that fault-tolerant ability is strong in function approximation.It is excitation function that neural net adopts Sigmoid function, it is excitation function that wavelet analysis adopts Morlet function, the excitation function of wavelet analysis has also been introduced translation yardstick and contraction-expansion factor, fast convergence rate when this makes wavelet neural network training, thereby make node locating speed fast, network positions real-time is good; And while making the neural prediction of small echo, reached better precision of prediction, and then made the location, position of node more accurate; The generalization of wavelet neural network is better, more flexible for the foundation of training network and prediction network model, has stronger applicability.
Beneficial effect:
The present invention adopts technique scheme that the wireless sensor network node positioning method of the DV-Hop based on small echo nerve is provided, and real-time is good, and position error is little, and positioning precision is high, and network applicability is strong.
Accompanying drawing explanation
Fig. 1 is flow chart of the present invention;
Fig. 2 is unknown node and anchor node random distribution figure in the present invention;
Fig. 3 is that position error and the anchor node of average each node while adopting respectively the present invention and traditional DV-Hop algorithm under equivalent environment counted graph of relation;
Fig. 4 is that while adopting respectively three kinds of different communication radius in the present invention, position error and the anchor node of average each node counted graph of relation;
Fig. 5 is that while adopting respectively three kinds of different number of network node in the present invention, position error and the anchor node of average each node counted graph of relation.
Embodiment
As shown in Figure 1, the DV-Hop wireless sensor network node positioning method based on small echo nerve, comprises the following steps:
Step 1: radio sensing network initialization, obtain the minimum hop count between minimum hop count, unknown node and the anchor node between every two anchor nodes, utilize every jumping that the minimum hop count of known distance between every two anchor nodes and acquisition is calculated between every two anchor nodes to jump distance;
Step 2: wavelet neural network initialization, the every jumping between every two anchor nodes in step 1 is jumped and processed apart from data, obtain being applicable to every jumping jumping distance of whole wireless sensor network;
Step 3: distance is jumped in every jumping of the whole wireless sensor network that utilizes the unknown node that obtains in step 1 and the minimum hop count between anchor node to be multiplied by obtain in step 2, calculates unknown node to the distance between anchor node;
Step 4: to the distance between anchor node, utilize least square method and trilateration to try to achieve the coordinate of unknown node, thereby complete location according to the unknown node obtaining in step 3.
In described step 1, the initialized concrete steps of radio sensing network are: first anchor node is to the grouping of its each neighbor node broadcast self-position information, and its jumping figure is initialized as to 0, receiving node is recorded to the minimum hop count of each anchor node, ignore the grouping from the larger jumping figure of same anchor node simultaneously, then its jumping figure value is added to 1 and be transmitted to neighbor node, thereby draw the minimum hop count between minimum hop count, unknown node and the anchor node between every two anchor nodes; Then according to formula
J ≠ i, with the known distance between two anchor nodes, divided by the minimum hop count between two anchor nodes of correspondence, distance, wherein (x are jumped in the every jumping obtaining between two anchor nodes
i, y
i), (x
j, y
j) be anchor node i, the coordinate of j, h
ijanchor node i, the minimum hop count between j (i ≠ j).In described step 2, the initialized concrete steps of wavelet neural network are: the every jumping using between every two anchor nodes that obtain in step 1 is jumped apart from the input data as the neural training network of small echo, and input data are taked interval grouping and moved the mode that window formula chooses and process, then input data and object vector are normalized; Then the transfer function using wavelet function Mallet as hidden layer, tries to achieve the output valve of hidden layer; What output layer used is on the basis of Sigmoid function, to add a contraction-expansion factor c to try to achieve output valve; Introduce factor of momentum aerfa revises each weights coefficient, threshold value coefficient simultaneously; Finally with Err_NetOut, judge whether all samples calculate completely; Err_NetOut represents the error of the quadratic sum of output valve and desired value, uses Err_NetOut and count as the threshold value of wavelet neural network algorithmic statement, and distance is jumped in the every jumping that obtains whole wireless sensor network.
During real work, in step 2, input data taked to interval grouping and move the mode that window formula chooses and process, such as by 50 data A={a1, a2, a3 ..., a50} is divided into 8 groups, 11 every group data: X1=(a1:a11), Y1=a15; X2=(a6:a16), Y2=a20; X3=(a11:a21), Y3=a25; X8=(a36:a46), Y8=a50.The sample input that wherein X is training network, Y is sample output.Again input data and object vector are normalized, weight coefficient, threshold value and wavelet parameter are carried out to initialization.
Wavelet function Mallet (fai (x)=cos (1.75.*x) * exp (x.^2/2)), as the transfer function of hidden layer, is tried to achieve to the output valve of hidden layer:
oxhp(h,1)＝fai((ixhp(h,1)-b(h,1))/a(h,1))
What output layer used is at Sigmoid (fnn) function f (x)=1/ (1+e
-x) basis on add a contraction-expansion factor c and try to achieve output valve:
ixjp2＝c.*ixjp
oxjp(p)＝fnn(ixjp2)
Introduce factor of momentum aerfa revises each weights coefficient, threshold value coefficient simultaneously:
wjh＝wjh+(1+aerfa)*detawjh；
whi＝whi+(1+aerfa)*detawhi；
a＝a+(1+aerfa)*detaa；
b＝b+(1+aerfa)*detab；
c＝c+(1+aerfa)*detac；
Finally with Err_NetOut, judge whether all samples calculate completely, Err_NetOut represents the error of the quadratic sum of output valve and desired value, threshold value with Err_NetOut and count as wavelet neural network algorithmic statement, distance is jumped in the every jumping that obtains whole wireless sensor network.
For verifying validity of the present invention, to the present invention's (being called for short WNNDV-Hop algorithm) and DV-Hop algorithm, adopt Matlab7.8.0 (R2009a) to realize emulation testing experiment.If node is randomly dispersed in the square region that the length of side is 100m, the random coordinate that produces unknown node and anchor node, unknown node and anchor node random distribution are as shown in Figure 2, in the present invention, parameter arranges as follows: the input number of nodes H=21 of wavelet neural network, worst error Err_NetOut (end)=0.05, maximum number of run count=2000 time, contraction-expansion factor c=0.5, factor of momentum aerfa=0.965, node searching scope 0≤X≤100m, 0≤Y≤100m, excellent evaluation criterion of algorithm is the average position error of DV-Hop algorithm and each unknown node iterative computation number of times times=100 of the present invention.
As shown in Figure 3, common environment network nodes Node Sum=100, communication radius R=20m are set, under same environmental conditions, while adopting respectively the present invention and traditional DV-Hop algorithm, both position errors all reduce along with the increase of anchor node number, when anchor node is counted ASum=5, position error of the present invention only has 12%, than the node locating error of traditional DV-Hop algorithm, reduces 19%; At anchor node, count after ASum=10, all steadily decline of the position error that increases the present invention and traditional DV-Hop algorithm along with anchor node number, position error of the present invention on average reduces 15% than the position error of traditional DV-Hop algorithm, show thus, when anchor node number is fewer, the superiority of positioning performance of the present invention is more obvious, and the radio sensing network adaptability less to anchor node is stronger.
As shown in Figure 4, common number of network node Node Sum=200 is set, chooses communication radius and be respectively 20m, 35m and 50m, when communication radius R=20m, error of the present invention is minimum, and positioning performance is good; Along with the increase of node communication radius, between two anchor nodes, the estimated value error of every jumping jumping distance increases thereupon, causes position error increase, positioning precision to decrease; When communication radius R=20m, node locating error is 27% to the maximum, and minimum can reach 23.15%.
As shown in Figure 5, common communication radius R=20m is set, choose number of network node and be respectively 100,200 and 300, horizontal analysis, under three kinds of different number of network node, position error of the present invention is the steady decrease along with the increase of anchor node number all, and when anchor node number is increased to ASum=50, global error is minimum; Vertical analysis, along with the increase of network node, anchor node number is constant, and anchor node ratio reduces, and the error of node locating increases.But on the whole, when number of network node Node Sum=300, anchor node, count ASum=5 time error and be 25.4% to the maximum.
Specific embodiment described herein is only to the explanation for example of the present invention's spirit.Those skilled in the art can make various modifications or supplement or adopt similar mode to substitute described specific embodiment, but can't depart from spirit of the present invention or surmount the defined scope of appended claims.
Claims (3)
1. the DV-Hop wireless sensor network node positioning method based on small echo nerve, is characterized in that: comprise the following steps:
Step 1: radio sensing network initialization, obtain the minimum hop count between minimum hop count, unknown node and the anchor node between every two anchor nodes, utilize every jumping that the minimum hop count of known distance between every two anchor nodes and acquisition is calculated between every two anchor nodes to jump distance;
Step 2: wavelet neural network initialization, the every jumping between every two anchor nodes in step 1 is jumped apart from carrying out data processing, obtain being applicable to every jumping jumping distance of whole wireless sensor network;
Step 3: distance is jumped in every jumping of the whole wireless sensor network that utilizes the unknown node that obtains in step 1 and the minimum hop count between anchor node to be multiplied by obtain in step 2, show that unknown node is to the distance between anchor node;
Step 4: to the distance between anchor node, utilize least square method and trilateration to try to achieve the coordinate of unknown node, thereby complete location according to the unknown node obtaining in step 3.
2. the DV-Hop wireless sensor network node positioning method based on small echo nerve according to claim 1, it is characterized in that: in described step 1, the initialized concrete steps of radio sensing network are: first anchor node is to the grouping of its each neighbor node broadcast self-position information, and its jumping figure is initialized as to 0, receiving node is recorded to the minimum hop count of each anchor node, ignore the grouping from the larger jumping figure of same anchor node simultaneously, then its jumping figure value is added to 1 and be transmitted to neighbor node, thereby draw the minimum hop count between every two anchor nodes, minimum hop count between unknown node and anchor node, then according to formula
J ≠ i, with the known distance between two anchor nodes, divided by the minimum hop count between two anchor nodes of correspondence, distance, wherein (x are jumped in the every jumping obtaining between two anchor nodes
i, y
i), (x
j, y
j) be anchor node i, the coordinate of j, h
ijanchor node i, the minimum hop count between j (i ≠ j).
3. the DV-Hop wireless sensor network node positioning method based on small echo nerve according to claim 1, it is characterized in that: in described step 2, the initialized concrete steps of wavelet neural network are: the every jumping using between every two anchor nodes that obtain in step 1 is jumped apart from the input data as the neural training network of small echo, and input data are taked interval grouping and moved the mode that window formula chooses and process, then input data and object vector are normalized; Then the transfer function using wavelet function Mallet as hidden layer is tried to achieve the output valve of hidden layer, and what output layer used is on the basis of Sigmoid function, to add a contraction-expansion factor c to try to achieve output valve; Introduce factor of momentum aerfa revises each weights coefficient, threshold value coefficient simultaneously; Finally with Err_NetOut, judge whether all samples calculate completely; Err_NetOut represents the error of the quadratic sum of output valve and desired value, uses Err_NetOut and count as the threshold value of wavelet neural network algorithmic statement, and distance is jumped in the every jumping that obtains whole wireless sensor network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410193590.4A CN103929810B (en) | 2014-05-09 | 2014-05-09 | DV Hop wireless sensor network node positioning methods based on wavelet neural |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410193590.4A CN103929810B (en) | 2014-05-09 | 2014-05-09 | DV Hop wireless sensor network node positioning methods based on wavelet neural |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103929810A true CN103929810A (en) | 2014-07-16 |
| CN103929810B CN103929810B (en) | 2017-08-25 |
Family
ID=51147845
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410193590.4A Active CN103929810B (en) | 2014-05-09 | 2014-05-09 | DV Hop wireless sensor network node positioning methods based on wavelet neural |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103929810B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105554873A (en) * | 2015-11-10 | 2016-05-04 | 胡燕祝 | Wireless sensor network positioning algorithm based on PSO-GA-RBF-HOP |
| CN108366408A (en) * | 2017-04-26 | 2018-08-03 | 南京理工大学 | A kind of DV-Hop location technologies based on neural network |
| CN108966120A (en) * | 2018-06-09 | 2018-12-07 | 中国电子科技集团公司第五十四研究所 | A kind of three side localization method of combination and system for dynamic cluster network improvement |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101772013A (en) * | 2010-01-26 | 2010-07-07 | 哈尔滨工程大学 | Voting mechanism based WSN anti-attack node positioning method |
| CN103149580A (en) * | 2013-02-04 | 2013-06-12 | 东南大学 | Global position system (GPS)/inertial navigation system (INS) combined navigation method based on strong tracking kalman filter (STKF) and wavelet neural network (WNN) |
-
2014
- 2014-05-09 CN CN201410193590.4A patent/CN103929810B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101772013A (en) * | 2010-01-26 | 2010-07-07 | 哈尔滨工程大学 | Voting mechanism based WSN anti-attack node positioning method |
| CN103149580A (en) * | 2013-02-04 | 2013-06-12 | 东南大学 | Global position system (GPS)/inertial navigation system (INS) combined navigation method based on strong tracking kalman filter (STKF) and wavelet neural network (WNN) |
Non-Patent Citations (1)
| Title |
|---|
| 李辉，熊盛武等: "无线传感器网络DV-HOP定位算法的改进", 《传感技术学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105554873A (en) * | 2015-11-10 | 2016-05-04 | 胡燕祝 | Wireless sensor network positioning algorithm based on PSO-GA-RBF-HOP |
| CN105554873B (en) * | 2015-11-10 | 2019-03-15 | 胡燕祝 | A kind of Wireless Sensor Network Located Algorithm based on PSO-GA-RBF-HOP |
| CN108366408A (en) * | 2017-04-26 | 2018-08-03 | 南京理工大学 | A kind of DV-Hop location technologies based on neural network |
| CN108966120A (en) * | 2018-06-09 | 2018-12-07 | 中国电子科技集团公司第五十四研究所 | A kind of three side localization method of combination and system for dynamic cluster network improvement |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103929810B (en) | 2017-08-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shareef et al. | Localization using neural networks in wireless sensor networks | |
| Nguyen et al. | Real-time estimation of sensor node's position using particle swarm optimization with log-barrier constraint | |
| Cheng et al. | Indoor robot localization based on wireless sensor networks | |
| CN105676181B (en) | Underwater movement objective Extended Kalman filter tracking based on distributed sensor energy ratio | |
| CN103298156B (en) | Based on the passive multi-target detection tracking of wireless sensor network | |
| CN103152745B (en) | Method of locating mobile node with strong adaptivity | |
| CN104023394A (en) | WSN positioning method based on self-adaptation inertia weight | |
| CN104066179B (en) | A kind of improved adaptive iteration UKF WSN node positioning methods | |
| CN103885030A (en) | Locating method of mobile node in wireless sensor network | |
| CN103415072B (en) | Based on the localization method estimating distance in a kind of radio sensing network | |
| CN106468771A (en) | A kind of multi-target detection and tracking method under the high clutter conditions of low Observable | |
| Gallicchio et al. | User movements forecasting by reservoir computing using signal streams produced by mote-class sensors | |
| CN105445732B (en) | The targetpath initial method observed under the conditions of dense clutter using multipath | |
| CN103929810A (en) | DV-Hop wireless sensor network node locating method based on wavelet neural network | |
| Zhu et al. | Intelligent fingerprint-based localization scheme using CSI images for internet of things | |
| Shit et al. | AI-enabled fingerprinting and crowdsource-based vehicle localization for resilient and safe transportation systems | |
| Luo et al. | A node depth adjustment method with computation-efficiency based on performance bound for range-only target tracking in UWSNs | |
| Dong et al. | Data fusion of the real time positioning system based on rssi and tof | |
| Fahama et al. | An Experimental Comparison of RSSI-Based Indoor Localization Techniques Using ZigBee Technology | |
| Mozamir et al. | Indoor localization estimation techniques in wireless sensor network: A review | |
| Ren et al. | FSTNet: Learning spatial–temporal correlations from fingerprints for indoor positioning | |
| CN102752850A (en) | Range-free based device and method for screening network anchor nodes | |
| CN103079168A (en) | Distributed motion node positioning method based on hidden Markov model | |
| Li et al. | An indoor hybrid localization approach based on signal propagation model and fingerprinting | |
| Lei et al. | Robust cooperative indoor localization based on reliability evaluation |
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 |