CN107426814B - Wireless sensor network positioning method based on multi-granularity framework node selection - Google Patents
Wireless sensor network positioning method based on multi-granularity framework node selection Download PDFInfo
- Publication number
- CN107426814B CN107426814B CN201710165803.6A CN201710165803A CN107426814B CN 107426814 B CN107426814 B CN 107426814B CN 201710165803 A CN201710165803 A CN 201710165803A CN 107426814 B CN107426814 B CN 107426814B
- Authority
- CN
- China
- Prior art keywords
- nodes
- node
- layer
- representative frame
- similarity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 239000011159 matrix material Substances 0.000 claims abstract description 67
- 230000008569 process Effects 0.000 claims description 13
- 101100153586 Caenorhabditis elegans top-1 gene Proteins 0.000 claims description 7
- 101100537629 Caenorhabditis elegans top-2 gene Proteins 0.000 claims description 7
- 101100370075 Mus musculus Top1 gene Proteins 0.000 claims description 7
- 101150107801 Top2a gene Proteins 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000000354 decomposition reaction Methods 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 4
- 230000000717 retained effect Effects 0.000 claims 1
- 230000008901 benefit Effects 0.000 abstract description 3
- 238000004364 calculation method Methods 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
- H04W64/003—Locating users or terminals or network equipment for network management purposes, e.g. mobility management locating network equipment
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
- H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a wireless sensor network positioning method based on multi-granularity frame node selection, which comprises the following steps: obtaining ranging information between adjacent nodes according to the received signal strength and constructing an inter-node similarity matrix; step two: selecting upper-layer frame nodes of each area according to the similarity among all nodes at the bottom layer, establishing a multi-granularity structure from bottom to top, and transmitting the similarity of the frame nodes among a plurality of granularity layers by utilizing an interpolation matrix; step three: reversely solving and calculating a distance matrix of the top-level representative frame node by using a similarity matrix of the top-level representative frame node, and obtaining low-dimensional relative coordinate representation of the top-level representative frame node by using a multidimensional scaling analysis (MDS) algorithm; step four: interpolating the low-dimensional relative coordinates of the top-level representative frame nodes, returning to obtain the relative coordinate representation of all the nodes of the network, and obtaining the absolute coordinates of the unknown nodes by using an absolute coordinate weighting strategy; the method has the advantage of greatly reducing the operation complexity of calculating the unknown node coordinates.
Description
Technical Field
The invention belongs to the field of wireless sensor network positioning, and particularly relates to a wireless sensor network positioning method based on multi-granularity framework node selection.
Background
The wireless sensor network is a distributed sensor network, the end of the wireless sensor network is a sensor capable of sensing the outside world, the wireless sensor network can be widely applied in real life due to the variety of the sensors, and the position information is the basis for realizing many practical applications. The existing wireless sensor network positioning method based on distance measurement can obtain better positioning accuracy on the whole, but is limited in hardware cost and energy consumption, and especially when the network scale is large, namely the number of nodes in the network is large, the calculation process involved in solving a large number of unknown node coordinates becomes very complicated, so that the operation complexity is higher, hardware resources are excessively consumed, and the algorithm practicability is poor. Aiming at the technical defects, the invention provides a positioning method which is small in calculation amount and simple to realize.
Disclosure of Invention
The present invention is directed to solving the above problems of the prior art. The wireless sensor network positioning method based on multi-granularity framework node selection greatly reduces the computational complexity and is accurate in description. The technical scheme of the invention is as follows:
a wireless sensor network positioning method based on multi-granularity framework node selection comprises the following steps:
the method comprises the following steps: arranging anchor nodes in an area to be positioned, sending beacon signals to non-anchor nodes by the anchor nodes, acquiring ranging information between adjacent nodes according to the signal strength received by the non-anchor nodes and constructing similarity matrixes among all nodes of the network;
step two: according to the similarity matrix among all the nodes constructed in the step one, selecting upper-layer frame nodes of each area according to the similarity among all the nodes at the bottom layer, establishing a multi-granularity structure from bottom to top, and transmitting the similarity of the frame nodes among a plurality of granularity layers by utilizing an interpolation matrix;
step three: similarity matrix W with top-level representative frame nodes[top]Reversely solving and calculating a distance matrix of the top-level representative frame node, and obtaining low-dimensional relative coordinate representation of the top-level representative frame node by using a multi-dimensional scale analysis algorithm;
step four: interpolating the low-dimensional relative coordinates of the top-level representative frame nodes, returning to obtain the relative coordinate representation of all the nodes of the network, and obtaining the absolute coordinates of the unknown nodes by using an absolute coordinate weighting strategy;
further, in the first step, before the unknown nodes of the area to be positioned are positioned, the m anchor nodes { anchor are positionedp(xp,yp) The node with known position information is deployed in a wireless sensor network area to be positioned, p is 1,2, m, and periodically sends beacon signals to other n non-anchor nodes in the network, and distance measurement information D between nodes can be obtained according to the strength of received signals and a signal propagation modelijConstructing a neighbor graph G[0](V[0],W[0])，V[0]: set of all nodes in the network, W[0]: and (5) all the inter-node similarity matrixes.
Further, the expression of the similarity matrix between the nodes is as follows:the similarity matrix representing nodes i and j is defined as follows:
wherein Ne (i) and Ne (j) represent sets of neighbor points of nodes i and j, respectively, DijThe distance, representing nodes i and j, is a kernel function parameter.
Further, in the process of establishing a bottom-to-top multi-grain layer structure, the upper representative frame is selected each time
All anchor nodes are reserved for a node.
Further, the specific process of establishing a bottom-to-top multi-grain layer structure and transferring the similarity of the frame nodes between the multiple grain layers by using the interpolation matrix in the second step is as follows:
(1) defining a strong connectivity between nodes i and j satisfies the following condition:
Wij≥θMaxk≠i{Wik},0<θ≤1
in a network node k neighbor graph, the strong connection of the nodes i and j indicates that the similarity of the nodes i and j accounts for a greater proportion of the maximum similarity between i and other neighbor points, when the conditions are met and the degree of the node i is greater than that of the node j, the node i is selected as an upper-layer representative frame node, otherwise, the node j is selected as an upper-layer representative frame node, and in a multi-particle-layer structure from bottom to top, the selected frame node representative point of the s-th layer should meet the requirement of the s-th-layer frame node representative pointThe frame nodes which represent different areas of the network and are obtained in the way become candidate nodes of the upper layer, and meanwhile, non-representative nodes are omitted;
(2) establishing an interpolation matrix P between layers[s]The essence of the interpolation matrix is: the similarity of the selected representative frame points is preserved, and the rest of the similaritiesDefining a similarity transformation interpolation matrix P from the s-th to (s +1) -th layer calculated from the average similarity on the reachable paths[s]Comprises the following steps:
the similarity matrix of the upper layer representative points can be represented by W[s+1]＝P[s]TW[s]P[s]It is derived that, obviously, after the complete multi-granularity hierarchy is constructed, the similarity matrix of each layer can be obtained by the following method:
W[s+1]＝P[s]TP[s-1]T...P[1]TP[0]TW[0]P[0]P[1]...P[s]P[s-1]。
thus, the similarity matrix of the top-level representative frame nodes is W[top]＝P[top-1]TP[top-2]T...P[1]TP[0]TW[0]P[0]P[1]...P[top-2]P[top-1]。
Furthermore, the third step obtains the low-dimensional relative coordinate representation of the top-level representative frame node by using the corresponding distance matrix of the top-level representative frame node and the MDS algorithmThe MDS comprises the following specific steps:
calculating the mutual distance between the top-level representative frame points;
bi-centering the distance square between the representative frame points;
thirdly, decomposing the distance characteristic values after double centralization to obtain maximum dim characteristic values lambda after decomposition1,...,λdimAnd corresponding feature vector V1,...,VdimThe low-dimensional relative coordinates of the representative frame points are expressed as:
further, the mutual distance between the top-level representative frame points is approximately estimated by a logarithmic function of a similarity matrix of the top-level representative frame nodes, and the distance is defined as follows:
where d is the corresponding distance matrix of the representative frame node, alpha is a parameter of the kernel function,a similarity matrix representing the top level representative frame nodes i and j.
Further, obtaining a low-dimensional relative coordinate representation of the top-level representative frame pointAnd then, interpolating the low-dimensional relative coordinates of the top-level representative frame points back to the low-dimensional relative coordinate representation of all nodes of the bottom-level network layer by using an interpolation matrixThe layer-by-layer interpolation process of the low-dimensional relative coordinates is as follows:
the interpolation conversion matrix used for interpolation return is the same as the interpolation matrix used for establishing the multi-granularity structure from bottom to top, and the interpolation return formula is defined as follows:
wherein,is the low-dimensional relative coordinate of the upper layer representative frame point,the low-dimensional relative coordinates of all the nodes at the bottommost layer are expressed by the recursion formula for the low-dimensional relative coordinates of the representative frame points at the lower layerComprises the following steps:
further, the absolute coordinates of the unknown nodes are obtained by using an absolute coordinate weighting strategyThe absolute coordinate weighting strategy comprises the following specific steps:
(ii) a low-dimensional relative coordinate representation of all the bottommost nodes returned from interpolationExtracting the relative coordinate representation of the m anchor nodes { Ranchorp(Rxp,Ryp)},p＝1,2,...m；
Solving the relative coordinate representation of the m anchor nodes { Rancho rp(Rxp,Ryp) And the real coordinate representation { anchor }p(xp,yp) Existence of a roto-translational relationship (B, B) (which can be denoted as anchor)p＝BRanchorp+b)；
Third, the low-dimensional relative coordinate representation of all unknown nodes except the anchor nodes in the bottommost networkAbsolute coordinates of any unknown node qCan be composed ofThus obtaining the product.
The invention has the following advantages and beneficial effects:
the invention is characterized in that a multi-granularity hierarchical structure from bottom to top of all nodes in the network is established by an algebraic multi-grid method, representative frame nodes with the most representative network topology structure are selected layer by layer, all unknown node coordinates in the one-time calculation network are simplified into estimation coordinates only for calculating the top representative frame nodes, then the estimation coordinates of the top representative frame nodes are returned by a simpler interpolation mode to obtain the estimation coordinates of all nodes in the network, and the operation complexity is reduced on the premise of keeping the network topology structure as much as possible.
The wireless sensor network positioning method based on multi-granularity frame node selection can establish similarity transmission of a plurality of inter-granularity representative frame nodes, and uses fewer representative frame nodes representing a network structure to carry out manifold dimension reduction, thereby greatly reducing the computational complexity; and the grain layer number in the selection process of the frame node can be customized, when more layers are adopted, a topological structure with a thicker network is shown, many nodes which are not representative are omitted, and when less layers are adopted, the whole network can be depicted more finely. The method has high calculation efficiency, strong stability and clear flow, respects the manifold structure of the original network data, can flexibly control the number of the particle layers for selecting the representative frame nodes, and is beneficial to improving the positioning efficiency.
Drawings
FIG. 1 is a flow chart of a specific implementation of a preferred embodiment of the present invention;
FIG. 2 is a schematic diagram of a process for selecting representative frame points according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described in detail and clearly with reference to the accompanying drawings. The described embodiments are only some of the embodiments of the present invention.
The technical scheme for solving the technical problems is as follows:
the wireless sensor network positioning method based on multi-granularity frame node selection is implemented according to the following steps:
the method comprises the following steps: before coordinate positioning is carried out on unknown nodes, m anchor nodes { anchorp(xp,yp) The anchor node is a node with known position information in the network and periodically sends beacon signals to other nodes in the network, and n other non-anchor nodes in the network receive signals sent by the beacon nodes to form a received signal strength matrix with dimension of m multiplied by n. Distance measurement information D between nodes can be obtained according to the signal propagation modelijFurther constructing a neighbor graph G[0](V[0],W[0]) In which V is[0]Is the set of all nodes in the network, W[0]Is the similarity matrix between all nodes.The similarity matrix representing nodes i and j is defined as follows:
wherein Ne (i) and Ne (j) represent sets of neighbor points of nodes i and j, respectively, DijRepresenting the distance of nodes i and j, and α is a kernel function parameter.
Step two: by similarity W between all nodes at the bottom[0]Selecting an upper-layer representative frame node set V representing the structure of each area of the network[s+1]N (N represents the number of layers), a bottom-to-top multi-grain layer structure is established (due to the anchor node anchor)pAll anchor nodes are reserved each time the upper layer representative frame node is selected), and an interpolation matrix P from the s-th to the (s +1) -th layer is utilized[s]And transmitting the similarity of the frame nodes among the grain layers.
Step three: similarity matrix W using top-level representative frame nodes[top]Inverse solution (logarithmic function) to compute the top-level representative frameworkDistance matrix d of nodesijAnd obtaining the low-dimensional relative coordinate representation of the top-level representative frame node by using a classical dimension reduction distance maintenance algorithm MDS through characteristic value decomposition
Step four: by interpolating the matrix sequence P[0]P[1]...P[top-2]P[top-1]Interpolating the low-dimensional relative coordinates of the top-level representative frame nodes, and returning to obtain the relative coordinate representation of all the nodes of the networkAnd obtaining absolute coordinates of all unknown nodes by using an absolute coordinate weighting strategy
The second embodiment is as follows: the first difference between the present embodiment and the specific embodiment is: the specific process of establishing a bottom-to-top multi-grain layer structure and transmitting the similarity of the frame nodes among the grain layers by using the interpolation matrix in the second step is as follows:
(1) defining a strong connectivity between nodes i and j satisfies the following condition:
Wij≥θMaxk≠i{Wik},0<θ≤1
wherein, θ is a strong neighbor threshold value, reflecting the strength of the neighbor relation between nodes. In the neighbor graph of the network node k, the strong connection between the node i and the node j indicates that the similarity between the node i and the node j takes a larger weight in the maximum similarity between the node i and other neighbor points. And when the conditions are met and the degree of the node i is greater than that of the node j, selecting the node i as an upper-layer representative frame node, otherwise, selecting the node j as an upper-layer representative frame node. Due to anchor node anchorpThe method has the advantage that all anchor nodes are reserved each time the upper-layer representative frame node is selected, so that the top-layer representative frame node can be guaranteed to contain all anchor nodes.
It is clear that in the bottom-to-top multi-grain layer structure, the choice is madeThe representing points of the frame nodes of the s-th layer should meetThe frame nodes representing different areas of the network obtained in this way will become candidate nodes of the previous layer, while non-representative nodes will be ignored. It is possible to reduce the amount of calculation and enhance the area structure information of the network.
(2) Establishing an interpolation matrix P between layers[s]. The essence of the interpolation matrix is: the similarity of the selected representative frame points is preserved, and the remaining similarities are calculated from the average similarity on the reachable paths. Defining a similarity transformation interpolation matrix P from the s-th to the (s +1) -th layer[s]Comprises the following steps:
the similarity matrix of the upper layer representative points can be represented by W[s+1]＝P[s]TW[s]P[s]And (4) deducing. Obviously, after constructing a complete multi-granularity hierarchical structure, the similarity matrix of each layer can be obtained by the following method:
W[s+1]＝P[s]TP[s-1]T...P[1]TP[0]TW[0]P[0]P[1]...P[s]P[s-1]
thus, the similarity matrix of the top-level representative frame nodes is W[top]＝P[top-1]TP[top-2]T...P[1]TP[0]TW[0]P[0]P[1]...P[top-2]P[top-1]。
The third concrete implementation mode: the present embodiment differs from the first or second embodiment in that: in the third step, the similarity matrix of the top-level representative frame node is used for solving reversely to calculate the distance matrix, and the MDS algorithm is used for obtaining the low-dimensional relative coordinate representation of the top-level representative frame node, and the specific process comprises the following steps:
(1) using interpolation matrixer in step twoCalculating similarity matrix W of top-level representative frame nodes[top]. Since the top-level representative frame node is only a part of all the nodes in the network, recalculation of the euclidean distance or geodesic distance is inaccurate, and the distance of the representative frame node is approximated by a logarithmic function of the similarity matrix of the top-level representative frame node. This avoids the need for repeated calculations while more accurately estimating the distance between representative frame points, which is defined as follows:
where d is the corresponding distance matrix of the representative frame node, alpha is a parameter of the kernel function,a similarity matrix representing the top level representative frame nodes i and j.
(2) Obtaining the low-dimensional relative coordinate representation of the top-level representative frame node by using the corresponding distance matrix of the representative frame node and the MDS algorithmThe MDS comprises the following specific steps:
calculating the mutual distance between the top-level representative frame points;
bi-centering the distance square between the representative frame points;
thirdly, decomposing the distance characteristic values after double centralization to obtain maximum dim characteristic values lambda after decomposition1,...,λdimAnd corresponding feature vector V1,...,VdimThe low-dimensional relative coordinates of the representative frame points are expressed as:
the fourth concrete implementation mode: the difference between this embodiment mode and one of the first to third embodiment modes is: in the fourth step, the specific process of interpolating the low-dimensional relative coordinates of the top-level representative frame nodes, returning to obtain the relative coordinate representation of all the nodes in the network, and obtaining the absolute coordinates of the unknown nodes by using an absolute coordinate weighting strategy comprises the following steps:
(1) obtaining a low-dimensional relative coordinate representation of the top-level representative frame pointAnd then, interpolating the low-dimensional relative coordinates of the top-level representative frame points back to the low-dimensional relative coordinate representation of all nodes of the bottom-level network layer by using an interpolation matrixThe layer-by-layer interpolation process of the low-dimensional relative coordinates is as follows:
the interpolation conversion matrix used for interpolation return is the same as the interpolation matrix used for establishing the multi-granularity structure from bottom to top. The interpolation return formula is defined as:
wherein,is the low-dimensional relative coordinate of the upper layer representative frame point,is the low-dimensional relative coordinates of the underlying representative frame points. Through the recursion formula, the low-dimensional relative coordinates of all the nodes at the bottom layer are expressedComprises the following steps:
(2) using absolute coordinate weighting strategies to obtain unknown nodesAbsolute coordinatesThe absolute coordinate weighting strategy comprises the following specific steps:
(ii) a low-dimensional relative coordinate representation of all the bottommost nodes returned from interpolationExtracting the relative coordinate representation of the m anchor nodes { Ranchorp(Rxp,Ryp)},p＝1,2,...m；
Solving the relative coordinate representation of the m anchor nodes { Rancho rp(Rxp,Ryp) And the real coordinate representation { anchor }p(xp,yp) Existence of a roto-translational relationship (B, B) (which can be denoted as anchor)p＝BRanchorp+b)；
Third, the low-dimensional relative coordinate representation of all unknown nodes except the anchor nodes in the bottommost networkAbsolute coordinates of any unknown node qCan be composed ofThus obtaining the product.
The above examples are to be construed as merely illustrative and not limitative of the remainder of the disclosure. After reading the description of the invention, the skilled person can make various changes or modifications to the invention, and these equivalent changes and modifications also fall into the scope of the invention defined by the claims.
Claims (7)
1. A wireless sensor network positioning method based on multi-granularity frame node selection is characterized by comprising the following steps:
the method comprises the following steps: arranging anchor nodes in an area to be positioned, sending beacon signals to non-anchor nodes by the anchor nodes, acquiring ranging information between adjacent nodes according to the signal strength received by the non-anchor nodes and constructing similarity matrixes among all nodes of the network;
step two: according to the similarity matrix among all the nodes constructed in the step one, selecting upper-layer frame nodes of each area according to the similarity among all the nodes at the bottom layer, establishing a multi-granularity structure from bottom to top, and transmitting the similarity of the frame nodes among a plurality of granularity layers by utilizing an interpolation matrix;
step three: similarity matrix W with top-level representative frame nodes[top]，W[top]Is W[0]A part of (A), W[0]Representing similarity matrixes among all nodes, reducing the number of selected nodes in the selection process from bottom to top, reversely solving and calculating a distance matrix of the top-level representative frame nodes, and obtaining low-dimensional relative coordinate representation of the top-level representative frame nodes by using a multi-dimensional scale analysis algorithm;
step four: interpolating the low-dimensional relative coordinates of the top-level representative frame nodes, returning to obtain the relative coordinate representation of all the nodes of the network, and obtaining the absolute coordinates of the unknown nodes by using an absolute coordinate weighting strategy;
before positioning unknown nodes of a region to be positioned, m anchor nodes { anchor are positionedp(xp，yp) The node with known position information is deployed in a wireless sensor network area to be positioned, p is 1,2, m, and periodically sends beacon signals to other n non-anchor nodes in the network, and distance measurement information D between nodes can be obtained according to the strength of received signals and a signal propagation modelijConstructing a neighbor graph G[0](V[0]，W[0])，V[0]: all node sets in the network;
the similarity matrix expression among the nodes is as follows:the similarity matrix representing nodes i and j is defined as follows:
wherein Ne (i) and Ne (j) represent sets of neighbor points of nodes i and j, respectively, DijRepresenting the distance of nodes i and j, and α is a kernel function parameter.
2. The method of claim 1, wherein all anchor nodes are retained each time an upper representative frame node is selected in establishing a bottom-to-top multi-granular structure.
3. The method for positioning a wireless sensor network based on multi-granularity frame node selection according to claim 1, wherein the specific process of establishing a bottom-to-top multi-granularity layer structure and transferring the similarity of the frame nodes among the multiple granularity layers by using an interpolation matrix in the second step is as follows:
(1) defining a strong connectivity between nodes i and j satisfies the following condition:
Wij≥θMaxk≠i{Wik},0＜θ≤1
in a network node k neighbor graph, the strong connection of the nodes i and j indicates that the similarity of the nodes i and j accounts for a greater proportion of the maximum similarity between i and other neighbor points, when the conditions are met and the degree of the node i is greater than that of the node j, the node i is selected as an upper-layer representative frame node, otherwise, the node j is selected as an upper-layer representative frame node, and in a multi-particle-layer structure from bottom to top, the selected frame node representative point of the s-th layer should meet the requirement of the s-th-layer frame node representative pointThe frame nodes which represent different areas of the network and are obtained in the way become candidate nodes of the upper layer, and meanwhile, non-representative nodes are omitted;
(2) establishing layer-to-layer interpolationMatrix P[s]The essence of the interpolation matrix is: the similarity of the selected representative frame points is preserved, and the rest similarities are calculated by the average similarity on the reachable paths, and a similarity conversion interpolation matrix P from the s th layer to the (s +1) th layer is defined[s]Comprises the following steps:
the similarity matrix of the upper layer representative points can be represented by W[s+1]＝P[s]TW[s]P[s]It is derived that, obviously, after the complete multi-granularity hierarchy is constructed, the similarity matrix of each layer can be obtained by the following method:
W[s+1]＝P[s]TP[s-1]T...P[1]TP[0]TW[0]P[0]P[1]...P[s]P[s-1]；
thus, the similarity matrix of the top-level representative frame nodes is W[top]＝P[top-1]TP[top-2]T...P[1]TP[0]TW[0]P[0]P[1]...P[top-2]P[top-1]。
4. The method of claim 2, wherein the third step utilizes a distance matrix corresponding to the top-level representative frame node and an MDS algorithm to obtain a low-dimensional relative coordinate representation of the top-level representative frame nodeThe MDS comprises the following specific steps:
calculating the mutual distance between the top-level representative frame points;
bi-centering the distance square between the representative frame points;
thirdly, decomposing the distance characteristic values after double centralization to obtain maximum dim characteristic values lambda after decomposition1,...,λdimAnd corresponding feature vector V1,...,VdimThe low-dimensional relative coordinates of the representative frame points are expressed as:
5. the method of claim 4, wherein the mutual distance between the top-level representative frame points is approximately estimated by a logarithmic function of a similarity matrix of the top-level representative frame nodes, and the distance is defined as follows:
where d is the corresponding distance matrix of the top level representative frame node, α is a parameter of the kernel function,a similarity matrix representing the top level representative frame nodes i and j.
6. The method of claim 4, wherein obtaining the low-dimensional relative coordinate representation of the top-level representative frame point is performed by using a multi-granularity frame node selection-based wireless sensor network positioning methodAnd then, interpolating the low-dimensional relative coordinates of the top-level representative frame points back to the low-dimensional relative coordinate representation of all nodes of the bottom-level network layer by using an interpolation matrixThe layer-by-layer interpolation process of the low-dimensional relative coordinates is as follows:
the interpolation conversion matrix used for interpolation return is the same as the interpolation matrix used for establishing the multi-granularity structure from bottom to top, and the interpolation return formula is defined as follows:
wherein,is the low-dimensional relative coordinate of the upper layer representative frame point,the low-dimensional relative coordinates of all the nodes at the bottommost layer are expressed by the recursion formula for the low-dimensional relative coordinates of the representative frame points at the lower layerComprises the following steps:
7. the method of claim 6, wherein the absolute coordinates of the unknown nodes are obtained by using an absolute coordinate weighting strategyThe absolute coordinate weighting strategy comprises the following specific steps:
(ii) a low-dimensional relative coordinate representation of all the bottommost nodes returned from interpolationExtracting the relative coordinate representation of the m anchor nodes { Ranchorp(Rxp,Ryp)},p＝1,2,...m；
Solving the relative coordinate representation of the m anchor nodes { Rancho rp(Rxp,Ryp) And real coordinatesRepresents { anchorp(xp,yp) Existence of a roto-translational relationship (B, B) between them, which can be denoted as anchorp＝BRanchorp+b；
Third, the low-dimensional relative coordinate representation of all unknown nodes except the anchor nodes in the bottommost networkAbsolute coordinates of any unknown node qCan be composed ofThus obtaining the product.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710165803.6A CN107426814B (en) | 2017-03-20 | 2017-03-20 | Wireless sensor network positioning method based on multi-granularity framework node selection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710165803.6A CN107426814B (en) | 2017-03-20 | 2017-03-20 | Wireless sensor network positioning method based on multi-granularity framework node selection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107426814A CN107426814A (en) | 2017-12-01 |
| CN107426814B true CN107426814B (en) | 2019-12-31 |
Family
ID=60423492
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710165803.6A Active CN107426814B (en) | 2017-03-20 | 2017-03-20 | Wireless sensor network positioning method based on multi-granularity framework node selection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107426814B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102186242A (en) * | 2011-05-09 | 2011-09-14 | 江南大学 | Method for positioning mobile node of wireless sensor network in fixed area |
| CN102711239A (en) * | 2012-05-10 | 2012-10-03 | 电子科技大学 | RSS (received signal strength) fingerprint database based secondary fuzzy clustering indoor-positioning method |
| CN104302000A (en) * | 2014-10-15 | 2015-01-21 | 上海交通大学 | Indoor positioning method based on signal receiving strength indicator correlation |
| CN106358233A (en) * | 2016-08-24 | 2017-01-25 | 哈尔滨工业大学 | RSS data flatting method based on multi-dimension analysis algorithm |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8625490B2 (en) * | 2011-01-07 | 2014-01-07 | Apple Inc. | Multiple granularity location determination |
-
2017
- 2017-03-20 CN CN201710165803.6A patent/CN107426814B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102186242A (en) * | 2011-05-09 | 2011-09-14 | 江南大学 | Method for positioning mobile node of wireless sensor network in fixed area |
| CN102711239A (en) * | 2012-05-10 | 2012-10-03 | 电子科技大学 | RSS (received signal strength) fingerprint database based secondary fuzzy clustering indoor-positioning method |
| CN104302000A (en) * | 2014-10-15 | 2015-01-21 | 上海交通大学 | Indoor positioning method based on signal receiving strength indicator correlation |
| CN106358233A (en) * | 2016-08-24 | 2017-01-25 | 哈尔滨工业大学 | RSS data flatting method based on multi-dimension analysis algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 基于多粒度流形学习的无线传感器网络定位方法;曾宪华，唐胜枰;《传感技术学报》;20130831;论文第1-4页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107426814A (en) | 2017-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112153736B (en) | Personnel action identification and position estimation method based on channel state information | |
| CN103747419B (en) | A kind of indoor orientation method based on signal strength difference and dynamic linear interpolation | |
| CN107592671B (en) | Networked multi-agent active variable topology autonomous cooperative positioning method | |
| CN104008304B (en) | A kind of weary information multisensor neutral net entropy evaluation of uncertainty in measurement method | |
| CN109151727B (en) | WLAN fingerprint positioning database construction method based on improved DBN | |
| CN107133690A (en) | A kind of lake water systems connects engineering proposal preference ordering method | |
| CN109039720A (en) | A kind of method for parameter estimation for binary sensor network | |
| CN108765180A (en) | The overlapping community discovery method extended with seed based on influence power | |
| CN109658477A (en) | A kind of DEM generating algorithm based on LIDAR data | |
| CN104331630B (en) | A kind of multi tate observes the state estimation and data fusion method of data | |
| CN117241215A (en) | Wireless sensor network distributed node cooperative positioning method based on graph neural network | |
| CN103313386B (en) | Based on the radio sensing network method for tracking target of consistency on messaging right-value optimization | |
| CN103164487B (en) | A kind of data clustering method based on density and geological information | |
| CN107426814B (en) | Wireless sensor network positioning method based on multi-granularity framework node selection | |
| CN108287866A (en) | Community discovery method based on node density in a kind of large scale network | |
| CN110035405A (en) | A kind of efficient fusion method of Multisensor Distributed based on random set theory | |
| CN117351378A (en) | Multi-aircraft ground target association method, system, equipment and medium | |
| Rekha et al. | An advanced DV-Hop localization algorithm for random mobile nodes in wireless sensor networks | |
| CN109362049B (en) | Square root volume filtering method based on mixed information weighting consistency | |
| CN108765472B (en) | Image set registration method based on sparse directed graph | |
| Li et al. | Multi-view matrix factorization for sparse mobile crowdsensing | |
| CN102289797B (en) | Method for hierarchically segmenting image based on Bhattacharyya distance | |
| CN109561384A (en) | Wireless sensor network node positioning method under composite noise condition | |
| CN106055568B (en) | A kind of automatic group technology of friend of the social networks based on single step addition group | |
| CN104346520B (en) | A kind of Data Dimensionality Reduction system and its dimension reduction method based on neuroid |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |