Background technology
Wireless sensor network (Wireless Sensor Networks, WSN) node locating has a wide range of applications in various fields such as target following, security protection and military surveillance, modern logistics, intelligent transportation, business shopping guide, guides.
The research origin of WSN, in 20 century 70s, is applied to military field the earliest.Enter after 21 century, development along with sensor technology, embedded technology, distributed information processing, wireless communication technology and network technology, the WSN being comprised of the microsensor node in a large number with microprocessing capability has obtained further development, and its technology starts to be widely used in the non-military fields such as industrial or agricultural, service trade, cultural industry, business.Within 2004, Japan has released the national information strategy U-Japan based on Internet of Things, wishes to expedite the emergence of and lead generation information scientific and technological revolution.The target of U-Japan is by immanent Internet of Things, creates a new informationized society, and Japan is built up to a country full of vigour, makes all tongues, and particularly children and disabled person, can actively participate in various social activitieies effectively expediently.The U.S., also actively responds " the wisdom earth " concept that IBM Corporation proposes, and very soon Internet of Things construction plan is upgraded to national strategy with regard to trade-after president Obama.In the global Internet of Things meeting in November, 2009, European Union's expert introduction the Internet of Things action plan > > of < < European Union, be intended to make Europe can lead the development of world's technology of Internet of things and industry.Under such overall background, the Chinese government also pays much attention to Internet of Things industry, and is classified as one of national strategy new industry, and governments at all levels have launched to give special assistance to comprising the information technology of future generation of Internet of Things.
The object localization method of existing WSN is all to design for concrete environment and application, does not also have pervasive object localization method at present.Application for a patent for invention (application number WO2007002286) as EUROPEAN PATENT OFFICE's on April 1st, 2007 disclosed " Scable Sensor Localization for Wireless Sensor Networks " by name, has proposed a kind of method that rule-based adaptive approach carries out WSN node locating.The method is measure portion unknown node and the distance of closing on anchor (anchor) node first, adopt geometry optimization algorithm-SDP Relaxation model (Semi-Definite Programming Model, SDP) coordinate of this part unknown node is solved, then required unknown node is decided to be to anchor node; Constantly repeat said process, until finally realize the location to all unknown node coordinates in network.Patent Office of the People's Republic of China discloses the application for a patent for invention (application number 200810103124.7) of " a kind of node positioning method of wireless sensor network " by name on August 27th, 2008, first it is the local relative coordinate system of each local structure of WSN, again each local relative coordinate is merged, obtain the overall relative coordinate of all nodes, and the beaconing nodes of use location ten-four, overall relative coordinate is converted to overall absolute coordinate, finally by the method for iteration refinement, determine finish node coordinate.On March 16th, 2011, Patent Office of the People's Republic of China discloses the application for a patent for invention (application number 201010538486.6) of " wireless sensor network locating method " by name, it receives by unknown node the information that comprises self-position that mobile anchor node broadcasts sends, two-dimensional direct angle coordinate system is set up in WSN region, determine the mobile route of mobile anchor point; When unknown node receives the information of three anchor nodes, calculate the distance that each anchor node receiving arrives unknown node, finally according to trilateration, unknown node is positioned.Other WSN node positioning method also comprises centroid method, maximal possibility estimation algorithm, the distance vector location algorithm based on jumping figure (Distance Vector Hop, DV-HOP), least square localization method and multiscale analysis location (MDS-MAP) method etc.
Although academia has proposed many location algorithms for WSN node, in numerous practical applications, owing to being subject to the impact of the aspect factors such as environment, network topology, joint behavior, algorithm complex, making a lot of algorithms there is no method and obtain practical application.
Summary of the invention
The object of the invention is to overcome the deficiencies in the prior art, a kind of fixed area wireless network sensor Network Mobility node positioning method is provided, its positioning action is convenient, and antijamming capability is strong, and positioning precision is high, and adaptability is good, safe and reliable.
According to technical scheme provided by the invention, described fixed area mobile node of wireless sensor network localization method, described mobile node positioning method comprises the steps:
A, in fixed area, place M anchor point, in fixed area, each M point is provided with corresponding identifier, the coordinate of specifying one of them anchor point is the origin of coordinates, in fixed area, all the other anchor points are determined corresponding coordinate according to the position relationship of the anchor point with as the origin of coordinates, and measure the distance between two between all anchor points in fixed area;
B, mobile node is put into the fixed area of step a, by being fixed the distance value between two between all anchor points in region, stored in mobile node;
After c, mobile node place and determine, corresponding anchor point range finding in mobile node and fixed area, finds out interior N-1 the nearest anchor point of distance moving node of mobile node communication range, N>3 wherein, M>N; Measure the distance of mobile node and a selected N-1 anchor point;
D, measure mobile node with as the distance between the anchor point of the origin of coordinates, using between the distance of mobile node and anchor point spacing, mobile node and a selected N-1 anchor point as the origin of coordinates and N-1 anchor point between two apart from structure distance matrix;
E, the distance matrix that steps d is obtained convert, and obtain range conversion matrix;
F, the range conversion matrix obtaining is carried out to Eigenvalues Decomposition, according to individual features value and the restructuring of described characteristic value characteristic of correspondence vector, obtain positional matrix;
G, by the coordinate of positional matrix and a selected N-1 anchor point, calculate the coordinate of mobile node, complete the location to mobile node.
In described step g, according to three limit location algorithms and multi-dimentional scale transform method, by the coordinate of positional matrix and a selected N-1 anchor point, calculate the coordinate of mobile node.
In described step f, by singular value decomposition, obtain the characteristic value of distance matrix, choose and be greater than 0 characteristic value and corresponding characteristic vector restructuring obtains positional matrix.
In described step a, in fixed area, the coordinate of anchor point obtains by mounting and positioning device on anchor point or demarcation.
In described step c, the range measurement of mobile node and selected anchor point obtains by the method for received signal strength indicator, time of arrival (toa) or signal arrival time difference.
In described steps d, when the anchor point as the origin of coordinates is positioned at the communication range of mobile node, mobile node obtains apart from the method by received signal strength indicator, time of arrival (toa) or signal arrival time difference with the anchor point as the origin of coordinates; When the anchor point as the origin of coordinates is positioned at outside the communication range of mobile node, mobile node and anchor point distance as the origin of coordinates by mobile node and the distance of route anchor point recently and recently route anchor point with the anchor point as the origin of coordinates apart from acquisition; Recently route anchor point refers to mobile node and communicates by letter in route as the anchor point of the origin of coordinates, the anchor point of corresponding the most close mobile node.
In described step e, the element value in range conversion matrix is:
Wherein, b
ijthe element value that represents the capable j row of i in range conversion matrix, d
1ii the anchor point that represents to choose in mobile node communication range and as the distance of the anchor point of the origin of coordinates, d
1jj the anchor point that represents to choose in mobile node communication range and as the distance of the anchor point of the origin of coordinates, d
ijrepresent the distance between interior i the anchor point of mobile node communication range and j anchor point.
According to choosing corresponding characteristic value and with the positional matrix obtaining with the restructuring of described characteristic value characteristic of correspondence vector being:
wherein, V
1represent that r is greater than 0 the vectorial matrix forming of characteristic value characteristic of correspondence, D
1represent that r is greater than the diagonal matrix that 0 characteristic value forms.
The expression formula of asking for of described mobile node is
Wherein,
For front n element of the n+1 row of positional matrix, X
nn * 2 dimension the matrix forming for the coordinate of N anchor point.
Described positioner comprises global positioning system.
Advantage of the present invention: in fixed area, M anchor point is set, selects one of them anchor point as the origin of coordinates, all the other anchor points in fixed area carry out coordinate according to the anchor point relation with as the origin of coordinates and determine; After mobile node is placed in fixed area, corresponding anchor point in selected mobile node communication range, by building distance matrix using mobile node, the corresponding anchor point of choosing and as the distance relation of the anchor point of the origin of coordinates, by obtaining range conversion matrix after the matrixing operation of adjusting the distance, realize the adjustment of adjust the distance matrix interior element position and relation; The transformation matrix of adjusting the distance carries out singular value decomposition, chooses the characteristic value and the corresponding characteristic vector structure positional matrix that are greater than 0; After combining with multidigit scale transformation method by three limit location algorithms, obtain the solution formula of mobile node coordinate, realize the effective location to mobile node coordinate; During location, can not affect location because of the fault of indivedual anchor points; In actual applications, can effectively reduce the impact of barrier on positioning precision; By transformation matrix is carried out to Eigenvalues Decomposition, then select to be greater than 0 characteristic value characteristic of correspondence vector and re-construct positional matrix, can effectively remove the harmful effect of range error to positioning precision; Localization method of the present invention have physical meaning clear, with network node distribution correlation, meet zone location actual demand, the calculating relating to and communication overhead is less and positioning precision is high feature; The method has considered this physical restriction factor of node communication scope, adapts with resource-constrained features such as wireless sensor network node location Calculation, storage and energy; Positioning action is convenient, and antijamming capability is strong, and positioning precision is high, and adaptability is good, safe and reliable.
Embodiment
Below in conjunction with concrete drawings and Examples, the invention will be further described.
As shown in Figure 1: in order to carry out location effectively fast to mobile node in wireless network sensor network, described mobile node positioning method comprises the steps:
A, in fixed area, place M anchor point, in fixed area, each M point is provided with corresponding identifier, the coordinate of specifying one of them anchor point is the origin of coordinates, in fixed area, all the other anchor points are determined corresponding coordinate according to the position relationship of the anchor point with as the origin of coordinates, and measure the distance between two between all anchor points in fixed area;
Anchor point is the known node of coordinate in wireless sensor network, and in fixed area, the coordinate of anchor point can obtain by mounting and positioning device or artificial demarcation; Described positioner can position for global positioning system (GPS); In fixed area, can be set to Yi Zuo building, after setting fixed area, can not affect the location to mobile node because of other anchor point fault, can effectively reduce the impact of barrier on mobile node positioning precision;
B, mobile node is put into the fixed area of step a, by being fixed the distance value between two between all anchor points in region, stored in mobile node;
In fixed area, in M anchor point, the distance between anchor point can obtain by ranging technology between two, described ranging technology can adopt general received signal strength indicator (Received Signal Strength Indication, RSSI), time of arrival (toa) (Time of Arrival, TOA), the method such as signal arrival time difference (Time Different of Arrival) obtains;
After c, mobile node place and determine, corresponding anchor point range finding in mobile node and fixed area, finds out interior N-1 the nearest anchor point of distance moving node of mobile node communication range, N>3 wherein, M>N; Measure the distance of mobile node and a selected N-1 anchor point;
After mobile node is placed in fixed area, because mobile node has certain communication range, the anchor point that surpasses mobile node communication range can not communicate with mobile node; In order to build follow-up distance matrix, need to choose N-1 anchor point, and N>3, M>N; Because a selected N-1 anchor point is positioned at the communication range of mobile node, can adopt said method to carry out directly range finding and obtain;
D, measure mobile node with as the distance between the anchor point of the origin of coordinates, using between the distance of mobile node and anchor point spacing, mobile node and a selected N-1 anchor point as the origin of coordinates and N-1 anchor point between two apart from structure distance matrix;
Determine mobile node and as the distance between the anchor point of the origin of coordinates, need to consider the communication range of mobile node; When the anchor point as the origin of coordinates is positioned at the communication range of mobile node, mobile node obtains apart from the method by received signal strength indicator, time of arrival (toa) or signal arrival time difference with the anchor point as the origin of coordinates; When the anchor point as the origin of coordinates is positioned at outside the communication range of mobile node, mobile node and anchor point distance as the origin of coordinates by mobile node and the distance of route anchor point recently and recently route anchor point with the anchor point as the origin of coordinates apart from acquisition; Recently route anchor point refers to mobile node and communicates by letter in route as the anchor point of the origin of coordinates, the anchor point of corresponding the most close mobile node; Distance matrix is the square formation of N+1 dimension, and the distance relation between distance matrix interior element is mobile node, as the distance relation between N-1 anchor point selected in the anchor point of the origin of coordinates and mobile node communication range;
E, the distance matrix that steps d is obtained convert, and obtain range conversion matrix;
Described range conversion matrix is B
n+1, in described range conversion matrix, the expression formula of respective element is:
Wherein, b
ijthe element value that represents the capable j row of i in range conversion matrix, d
1ii the anchor point that represents to choose in mobile node communication range and as the distance of the anchor point of the origin of coordinates, d
1jj the anchor point that represents to choose in mobile node communication range and as the distance of the anchor point of the origin of coordinates, d
ijrepresent the distance between interior i the anchor point of mobile node communication range and j anchor point, d
ijinside comprised mobile node and as the distance between the anchor point of the origin of coordinates;
F, the range conversion matrix obtaining is carried out to Eigenvalues Decomposition, according to individual features value and the restructuring of described characteristic value characteristic of correspondence vector, obtain positional matrix;
By singular value decomposition, obtain the characteristic value of distance matrix, choose and be greater than 0 characteristic value and corresponding characteristic vector restructuring obtains positional matrix; Selected characteristic value is greater than 0 characteristic value and the characteristic vector of character pair value, can effectively remove the harmful effect of range error to positioning precision;
Distance matrix is carried out to Eigenvalues Decomposition, and choose after corresponding characteristic value and character pair vector, the expression formula of distance matrix can be expressed as
Defining new transformation matrix is positional matrix B'
n+1:
In above-mentioned expression formula, D
1for r is greater than the diagonal matrix that 0 characteristic value forms, V
1for this r is individual, be greater than 0 the vectorial matrix forming of characteristic value characteristic of correspondence, D
2for the diagonal matrix that all the other n+1-r characteristic values form, V
2for D
2the matrix that forms of characteristic value characteristic of correspondence vector;
for matrix V
1turn order,
for matrix V
2turn order;
G, by the coordinate of positional matrix and a selected N-1 anchor point, calculate the coordinate of mobile node, complete the location to mobile node;
According to three limit location algorithms and multi-dimentional scale transform method, by the coordinate of positional matrix and a selected N-1 anchor point, calculate the coordinate of mobile node; During calculating, by three limit location algorithms and multi-dimentional scale transform method, in conjunction with rear derivation, obtain computing formula below,
The coordinate that we can calculate mobile node B is:
Wherein,
For B'
n+1front n element of n+1 row, X
nfor the N * 2 dimension matrix that the coordinate of N anchor point forms, x
n+1coordinate for mobile node.
As shown in Figure 2: be the schematic diagram of mobile node of the present invention location.The fixed area in Fig. 2 with the box indicating setting of outside, in fixed area, there is corresponding two-dimensional coordinate value, in fixed area, there are two mobile nodes of A and B, wherein, as the anchor point of the origin of coordinates, be positioned at the communication range of mobile node A, as the anchor point of the origin of coordinates, be positioned at outside the communication range of mobile node B.
When needs position mobile node A, first to choose the anchor point in mobile node A communication range, in mobile node A, store the distance between two of anchor point in fixed area, mobile node A obtains by conventional distance measuring method with the distance of anchor point as the origin of coordinates.After above-mentioned range measurement obtains, distance structure distance matrix using mobile node A with the distance as the anchor point of the origin of coordinates, the distance between two of choosing anchor point in mobile node communication range and mobile node A and selected anchor point, distance matrix is carried out to singular value decomposition, and choose and be greater than 0 characteristic value and corresponding characteristic vector, build range conversion matrix, after being processed, range conversion matrix obtains positional matrix, after having some transform method combination by three limit location algorithms and multidimensional, obtain computing formula, can access the coordinate figure of corresponding mobile node A.
When needs position mobile node B, first to choose the anchor point in mobile node B communication range, in mobile node B, store the distance between two of anchor point in fixed area, mobile node B and distance as the anchor point of the origin of coordinates by mobile node B and the distance of route anchor point recently and recently the distance of route route anchor point obtain, select after nearest route anchor point mobile node B and be all positioned at the communication range of nearest route anchor point as the anchor point of the origin of coordinates; Route anchor point is the nearest of distance moving Node B recently; In fixed area, in order to calculate mobile node B and distance as the anchor point of the origin of coordinates, any anchor point is all likely as route anchor point recently.After above-mentioned range measurement obtains, distance structure distance matrix using mobile node B with the distance as the anchor point of the origin of coordinates, the distance between two of choosing anchor point in mobile node communication range and mobile node B and selected anchor point, distance matrix is carried out to singular value decomposition, and choose and be greater than 0 characteristic value and corresponding characteristic vector, build range conversion matrix, after being processed, range conversion matrix obtains positional matrix, by obtaining computing formula after three limit location algorithms and the combination of multi-dimentional scale transform method, can access the coordinate figure of corresponding mobile node B.
The present invention arranges M anchor point in fixed area, selects one of them anchor point as the origin of coordinates, and all the other anchor points in fixed area carry out coordinate according to the anchor point relation with as the origin of coordinates and determine; After mobile node is placed in fixed area, corresponding anchor point in selected mobile node communication range, by building distance matrix using mobile node, the corresponding anchor point of choosing and as the distance relation of the anchor point of the origin of coordinates, by obtaining range conversion matrix after the matrixing operation of adjusting the distance, realize the adjustment of adjust the distance matrix interior element position and relation; The transformation matrix of adjusting the distance carries out singular value decomposition, chooses the characteristic value and the corresponding characteristic vector structure positional matrix that are greater than 0; After combining with multidigit scale transformation method by three limit location algorithms, obtain the solution formula of mobile node coordinate, realize the effective location to mobile node coordinate; During location, can not affect location because of the fault of indivedual anchor points; In actual applications, can effectively reduce the impact of barrier on positioning precision; By transformation matrix is carried out to Eigenvalues Decomposition, then select to be greater than 0 characteristic value characteristic of correspondence vector and re-construct positional matrix, can effectively remove the harmful effect of range error to positioning precision; Localization method of the present invention have physical meaning clear, with network node distribution correlation, meet zone location actual demand, the calculating relating to and communication overhead is less and positioning precision is high feature; The method has considered this physical restriction factor of node communication scope, adapts with resource-constrained features such as wireless sensor network node location Calculation, storage and energy; Positioning action is convenient, and antijamming capability is strong, and positioning precision is high, and adaptability is good, safe and reliable.