CN110501694B - Estimation Method of Passive Velocity of Underwater Nodes Based on Doppler Frequency Shift Estimation - Google Patents

Abstract

The invention discloses a Doppler frequency shift estimation-based underwater node passive motion speed estimation method, which comprises the steps of setting up the current kth positioning, establishing a positioning model based on an arrival time difference TDOA, obtaining communication time information of an underwater sensing node and a reference buoy node in the kth positioning process, and calculating the positioning position of the sensing node in the kth positioning process; calculating a projection component of the actual motion speed vector of the sensing node on a connecting line between the position of the sensing node and the position of the reference buoy node in the k-1 positioning process according to the Doppler frequency shift measured value and the actual motion speed vector of the reference buoy node obtained in the k-1 positioning process; estimating an actual motion velocity vector of the sensing node from a kth-1 positioning process to a kth positioning process; and obtaining the actual motion speed vector predicted value of the sensing node during the kth positioning process to the kth+1th positioning process, and predicting the real-time position of the node.

Description

Passive motion velocity estimation method of underwater nodes based on Doppler frequency shift estimation

Technical Field

The invention belongs to the field of underwater sensor network positioning, and in particular relates to a method for estimating a passive motion velocity vector of an underwater node based on Doppler frequency shift estimation.

Background Art

The ocean is a huge treasure trove of resources on the earth, but human beings have only known and investigated less than 10% of the ocean. The vast ocean is waiting for human cognition, development and utilization. Marine environmental monitoring, marine resource development and marine rights protection have become major strategic needs of various countries. The huge underwater pressure and huge area are not suitable for direct long-term human operations. With the development of embedded systems, the advancement of underwater acoustic communication and signal processing technology, underwater acoustic communication sensor networks have become an important means to achieve deep-sea oil well exploration, earthquake observation, hydrological environment monitoring, underwater navigation, underwater rescue and other applications.

In underwater acoustic communication sensor networks, positioning is indispensable. On the one hand, the control center needs to know the location information of network nodes for recovery and maintenance. On the other hand, sensor data is more practical when equipped with location information. However, due to the fluidity of water, underwater nodes will passively move with the water, and the positioning process needs to be performed periodically to maintain the accuracy of the positioning position.

In order to predict the location of a node at any time between two adjacent positioning processes, it is necessary to estimate the passive motion velocity vector of the node. There are two main traditional node motion velocity vector prediction methods. The first method obtains the past node motion velocity information based on the past positioning position information of the node, and uses the Wiener filter to predict the future node motion velocity vector. The accuracy of the node motion velocity vector estimation of this method depends on the positioning accuracy. If the ranging positioning error of the positioning system increases, it will have a great impact on the node motion velocity vector estimation. The second method uses a weighted average method to estimate the motion speed of the target node based on the motion speed information of the target node's neighbor nodes. This method requires each node to broadcast a data packet containing its own speed information at least once during the positioning phase, which increases the network communication burden and node energy overhead.

In actual ocean or river water, the flow velocity of water at different depths is different, so there is relative motion between nodes at different depths. In two adjacent positionings, the signal Doppler frequency shift can be used to estimate the relative motion of the sending and receiving nodes. When the node obtains the relative motion information of multiple reference nodes, its true motion velocity vector can be further estimated. Patent CN201710657205 discloses a method for underwater sensor node positioning based on TOA ranging and Doppler effect, in which the node motion speed is obtained by sensor measurement, and the Doppler frequency shift measurement value is used as a positioning model parameter to improve positioning accuracy. The difference between the patent of the present invention and the above patent is that the patent of the present invention discloses a method for estimating the passive motion velocity vector of underwater nodes based on Doppler frequency shift estimation, which is a new node motion estimation method and is used for node position prediction.

Summary of the invention

In view of the deficiencies in the prior art, the present invention provides a method for estimating the passive motion velocity vector of an underwater node based on Doppler frequency shift estimation, aiming to solve the problem that the passive motion of the underwater node due to the influence of water flow leads to invalid positioning information.

The technical solution of the present invention provides a method for estimating the passive motion speed of an underwater node based on Doppler frequency shift estimation, comprising the following steps:

Step 1: Assume that the kth positioning is currently being performed, establish a positioning model based on the time difference of arrival (TDOA), obtain the communication time information between the underwater sensor node and the reference buoy node during the k-1th positioning process, and calculate the positioning position of the sensor node during the k-1th positioning process;

Step 2, measuring the Doppler frequency shift according to the communication time information between the sensor node and the reference buoy node obtained in step 1, and the communication time information between the sensor node and the reference buoy node in the kth positioning process;

Step 3, according to the Doppler frequency shift measurement value obtained in step 2, and the actual motion velocity vector of the reference buoy node in the k-1th positioning process and the kth positioning process, calculate the projection component of the actual motion velocity vector of the sensor node on the line connecting the sensor node position and the reference buoy node position in the k-1th positioning process;

Step 4, based on the projection component of the actual motion velocity vector between the sensor node and each reference buoy node obtained in step 3, estimate the actual motion velocity vector of the sensor node from the k-1th positioning process to the kth positioning process;

Step 5, according to the estimated value of the actual motion velocity vector of the sensor node obtained in step 4, obtain the predicted value of the actual motion velocity vector of the sensor node during the kth positioning process to the k+1th positioning process, and predict the real-time position of the node until the new positioning process starts;

Step 6: loop through steps 1-5 to continue estimating the motion speed.

Moreover, in step 1, the positioning model is composed of underwater sensor nodes and surface reference buoy nodes. The surface buoy nodes serve as positioning reference nodes to assist the underwater sensor nodes in positioning. The sensor nodes adopt active positioning method and achieve positioning through ranging method based on arrival time.

Moreover, the sensor node is positioned using the ranging method based on arrival time as follows:

Let p = 1, 2, ..., P be the sensor node number, b _n be the one-hop neighbor reference buoy node within the communication range of sensor node p, and n = 1, 2, ..., N be the reference buoy node number;

During the k-1th positioning process, the time when the sensor node p sends the positioning request information is The location to be located is The time when the reference buoy node _bn receives the positioning request information is The reference position is N is the total number of reference nodes that receive the positioning request information of sensor node p; the time when reference buoy node _bn sends the positioning response information is The time when the sensor node p receives the positioning response information of the reference buoy node _bn is

According to the three-dimensional geometric relationship, the distance between the sensor node p and the reference buoy node b _n It is expressed as:

distance Can be expressed as:

Where c is the speed of sound, formula (1) is expanded to:

in is a constant term related to the sensor node p and the reference buoy node b _n ;

When n=1 and n≠1, according to the TDOA positioning principle, the distance difference between the sensor node p and the reference buoy nodes _b1 and _bn is for:

Another expression of formula (4) is:

in n≠1 is the distance between the sensor node p and the reference buoy node _bn in the k-1th positioning process, n≠1 is the distance difference between the reference buoy nodes _b1 and _bn in the k-1th positioning process, is the distance between the sensor node p and the reference buoy node _b1 in the k-1th positioning process. Substituting equation (3) into equation (5) yields:

In formula (6), The term is written in the form of formula (3), and formula (6) is rewritten as:

in is the constant term related to the sensor node p and the reference buoy node _b1 during the k-1th positioning process, is the x,y coordinate value of the reference buoy node b ₁ ;

The distance information between the sensor node p and the reference buoy nodes b ₁ , b _n , n = 2, 3, ..., N in the k-1th positioning process is written in matrix form and expressed as:

H＝GΨ+W (8)

In formula (8), W is the Gaussian noise matrix:

in represents the Gaussian noise term associated with the reference buoy nodes _b1 and _bn , n=2,...,N in the k-1th positioning process, N(0, ^σ2 ) represents that the mean of the Gaussian noise is 0 and the variance is ^σ2 ;

In formula (8), H and G are known constant matrices, which are expressed as:

In formula (8), Ψ is the parameter to be determined, which contains the horizontal position information of the sensor node p and is expressed as:

The least squares solution of formula (8) is:

Ψ＝(G ^T G) ^-1 G ^T H (12)

Where G ^T is the transposed matrix of G, () ^-1 indicates the matrix is inverted;

The Chan algorithm and Taylor series expansion algorithm are used to solve equation (12) and the estimated value of the positioning position of the sensor node p in the k-1th positioning process is obtained:

Moreover, step 2 is implemented as follows,

Assume that during the kth positioning process, the time when sensor node p sends the positioning request information is The location to be located is The time when the reference buoy node _bn receives the positioning request information is The reference position is The reference buoy node _bn sends the positioning response information at the time The time when the sensor node p receives the positioning response information of the reference buoy node _bn is

According to step 1, the estimated value of the positioning position of the sensor node p in the kth positioning process is obtained In the kth positioning process and the k-1th positioning process, the time difference between the sensor node p sending the positioning request information is The frequency of positioning request information is The time difference of receiving the positioning request information by reference buoy node _bn is The frequency of receiving positioning request information is The Doppler frequency shift associated with the positioning request information between the sensor node p and the reference buoy node _bn is,

in, Indicates that the nodes are relatively close. Indicates that the nodes are relatively far away.

Moreover, step 3 is implemented as follows,

During the kth positioning process and the k-1th positioning process, the actual motion velocity vector of the reference buoy node b _n is for:

The reference buoy node _bn estimates the position of the sensor node p during the k-1th positioning process. and the reference position of the reference buoy node b _n Projection on the line for:

in Estimate the position of sensor node p and reference buoy node b _n reference position The angle between the line and the horizontal plane,

In the kth positioning process and the k-1th positioning process, due to the slow passive movement of the sensor node and the reference buoy node with the water flow, During the time, the passive motion distance is much smaller than the distance between the sensor node and the reference buoy node, which constitutes a far-field condition, and the propagation trajectory of the positioning request information is approximately parallel;

Location request information propagation distance difference for:

in is the estimated position of sensor node p in the k-1th positioning process and the reference position of the reference buoy node b _n The projection component on the connecting line;

Positioning request information propagation phase difference for:

in is the communication carrier wavelength, c is the speed of sound, and f is the carrier frequency. The second expression of the Doppler frequency shift in equation (13) is:

Combined with equations (17), (18), and (19), we can obtain the estimated position of the sensor node p during the k-1th positioning process by the actual motion velocity vector of the sensor node p: and the reference position of the reference buoy node b _n Projection component on the line The calculation expression is:

Moreover, step 4 is implemented as follows,

The projection component of the actual motion velocity vector of the sensor node p in step 3 is Written in vector form:

in for The absolute value of is the estimated position of sensor node p in the k-1th positioning process and the reference position of the reference buoy node b _n The unit direction vector on the connecting line is:

in are the unit vectors in the directions of the coordinate axes x, y, and z respectively;

Passing the location of sensor node p And the projection component of the actual motion velocity vector of the sensor node p The general expression for the perpendicular plane equation is:

in are the plane equation parameters, x, y, z represent the coordinate axes, and the plane equation normal vector It can be expressed as:

The projection component of the actual motion velocity vector of the sensor node p It is essentially a set of normal vectors of the plane that satisfies equation (23), let

Then the plane equation parameters for:

The general expression of the plane equation (23) is the projection component of the actual motion velocity vector passing through the sensor node p Endpoint In vector form it is:

Set the vector endpoints The coordinates represented Substituting into the general expression (23) of the plane equation, we get the plane equation parameters

When n＝1,2,...,N, multiple sets of plane equations are expressed in matrix form:

in is the plane equation parameter matrix, is the intersection of the plane equation, and is the position of the sensor node p The actual velocity vector of the starting point The end point, It is expressed as:

The nodes involved need to satisfy the node position constraint: at least three reference buoy nodes are not located on the same straight line; solve equation (29) using the least squares method to obtain the sensor node p with position is the estimated value of the actual motion velocity vector of the starting point The end point coordinates

During the k-1th positioning and the kth positioning, the actual motion velocity vector estimate of the sensor node p is for:

Furthermore, step 5 is implemented as follows,

During the kth positioning and k+1th positioning, the actual motion velocity vector prediction value of the sensor node p is for:

During the kth positioning and k+1th positioning, the positioning prediction value of sensor node p is for:

in is the estimated value of the k-th localization sensor node p, is the time when the sensor node p sends the positioning request information After the time has passed, satisfy hour, represents any time between the k+1th positioning and the kth positioning of sensor node p.

Aiming at the problem of positioning failure caused by passive movement of underwater nodes under the influence of water flow, the present invention proposes a method for estimating passive motion velocity vector of underwater nodes based on Doppler frequency shift estimation. The method does not require velocity vector information of neighboring nodes, reduces network communication overhead, and saves node energy. The passive motion velocity vector of the underwater node obtained by the method is only related to the first positioning position in two adjacent positioning processes, reduces the influence of positioning error on the velocity vector estimation value, is conducive to improving the accuracy of node positioning prediction, and is a method for estimating passive motion velocity vector of underwater nodes with good application prospects.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic diagram of a positioning scenario according to an embodiment of the present invention.

FIG. 2 is a schematic diagram of node data interaction in a single positioning process according to an embodiment of the present invention.

FIG. 3 is a schematic diagram of node position constraint conditions according to an embodiment of the present invention.

FIG. 4 is a schematic diagram of passive motion velocity vector estimation of an underwater node according to an embodiment of the present invention.

DETAILED DESCRIPTION

The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings and embodiments.

Examples of the embodiments are shown in the drawings. The embodiments described below with reference to the drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

An embodiment of the present invention provides a method for estimating a passive motion velocity vector of an underwater node based on Doppler frequency shift estimation, comprising the following steps:

Step 1: Assume that the kth positioning is currently being performed, establish a positioning model based on the time difference of arrival (TDOA), obtain the communication time information between the sensor node and the reference buoy node during the k-1th positioning process, and calculate the positioning position of the sensor node during the k-1th positioning process;

Preferably, the positioning model described in step 1 is composed of underwater sensor nodes and surface buoy nodes. All nodes are equipped with pressure sensors, which can obtain node depth information and keep the depth unchanged; the surface buoy node obtains the real-time position through the global positioning system, and serves as a positioning reference node to assist the underwater sensor node positioning, that is, the reference buoy node; the underwater sensor node adopts an active positioning method and realizes positioning through a ranging method based on arrival time. A rectangular coordinate system is established in the positioning area, and the coordinate axes are represented by x, y, and z. In the embodiment, x is a horizontal coordinate, indicating the east-west direction, and the positive direction is east; y is a horizontal coordinate, indicating the north-south direction, and the positive direction is north; z is a vertical coordinate, indicating depth, and the positive direction is vertically downward. k = 2, 3, ..., K is the positioning process number, indicating the kth positioning; p = 1, 2, ..., P is the sensor node number; b _n is a one-hop neighbor reference buoy node within the communication range of the sensor node p, and n = 1, 2, ..., N is the reference buoy node number.

Assume that in the k-1th positioning process, the time when the sensor node p sends the positioning request information is The location to be located is are the x, y, and z coordinate values of the sensor node p, p represents the sensor node number, and k-1 represents the positioning process number. The time when the reference buoy node _bn receives the positioning request information is The reference position is are the x, y, and z coordinate values of the reference buoy node _bn, respectively. n represents the buoy node number, and takes the value n＝1, 2, ..., N. N is the total number of reference nodes that receive the positioning request information of the sensor node p. The successful positioning of the node requires N≥3. The time when the reference buoy node _bn sends the positioning response information is The time when the sensor node p receives the positioning response information of the reference buoy node _bn is

According to the three-dimensional geometric relationship, the distance between the sensor node p and the reference buoy node b _n It is expressed as:

distance It can also be expressed as:

Where c is the speed of sound. Formula (1) can be expanded to obtain:

in is a constant term related to the sensor node p and the reference buoy node _bn . When n=1 and n≠1, according to the TDOA positioning principle, the distance difference between the sensor node p and the reference buoy nodes _b1 and _bn is for:

Another expression of formula (4) is:

in n≠1 is the distance between the sensor node p and the reference buoy node _bn in the k-1th positioning process, n≠1 is the distance difference between the reference buoy nodes _b1 and _bn in the k-1th positioning process, is the distance between the sensor node p and the reference buoy node _b1 in the k-1th positioning process. Substituting equation (3) into equation (5) yields:

In formula (6), The term is written in the form of formula (3), and formula (6) can be rewritten as:

in is the constant term related to the sensor node p and the reference buoy node _b1 during the k-1th positioning process, are the x,y coordinate values of the reference buoy node _b1 .

The distance information between the sensor node p and the reference buoy nodes b ₁ , b _n , n=2,3,...,N in the k-1th positioning process is expressed in matrix form as follows:

H＝GΨ+W (8)

In formula (8), W is the Gaussian noise matrix:

in represents the Gaussian noise term associated with the reference buoy nodes _b1 and _bn , n=2,...,N in the k-1th positioning process, N(0, ^σ2 ) represents that the mean of the Gaussian noise is 0 and the variance is ^σ2 ;

In formula (8), H and G are known constant matrices, which are expressed as:

In formula (8), Ψ is the parameter to be determined, which contains the horizontal position information of the sensor node p and is expressed as:

The least squares solution of formula (8) is:

Ψ＝(G ^T G) ^-1 G ^T H (12)

Where ^GT is the transposed matrix of G, and () ^-1 indicates the inversion of the matrix.

The Chan algorithm and Taylor series expansion algorithm are used to solve equation (12) and the estimated value of the positioning position of the sensor node p in the k-1th positioning process is obtained: They are x, y, and z coordinate values respectively.

Step 2: According to the communication time information between the sensor node and the reference buoy node obtained in step 1, and the communication time information between the sensor node and the reference buoy node in the kth positioning process, the Doppler frequency shift is measured;

During the kth positioning process, the time when sensor node p sends positioning request information is The location to be located is The time when the reference buoy node _bn receives the positioning request information is The reference position is are x, y, and z coordinate values respectively. n represents the buoy node number, which can be 1, 2, ..., N. N is the total number of reference nodes that receive the positioning request information of sensor node p. The successful positioning of the node requires N ≥ 3. The time when the reference buoy node b _n sends the positioning response information is The time when the sensor node p receives the positioning response information of the reference buoy node _bn is According to the positioning method described in step 1, the estimated position of the sensor node p in the kth positioning process is They are x, y, and z coordinate values respectively.

In the kth positioning process and the k-1th positioning process, the time difference between the sensor node p sending the positioning request information is The frequency of positioning request information is The time difference of receiving the positioning request information by reference buoy node _bn is The frequency of receiving positioning request information is The Doppler frequency shift between the sensor node p and the reference buoy node _bn related to the positioning request information is:

Indicates that the nodes are relatively close; Indicates that the nodes are relatively far away.

Step 3: Based on the Doppler frequency shift measurement value obtained in step 2 and the actual motion velocity vector of the reference buoy node in the k-1th positioning process and the kth positioning process, calculate the projection component of the actual motion velocity vector of the sensor node on the line connecting the sensor node position and the reference buoy node position in the k-1th positioning process;

In the embodiment, during the kth positioning process and the k-1th positioning process, the actual motion speed vector of the reference buoy node b _n is for:

The reference buoy node _bn estimates the position of the sensor node p during the k-1th positioning process. and the reference position of the reference buoy node b _n Projection on the line for:

in Estimate the position of sensor node p and reference buoy node b _n reference position The angle between the connecting line and the horizontal plane is:

In the kth positioning process and the k-1th positioning process, due to the slow passive movement of the sensor node and the reference buoy node with the water flow, During this time, the passive motion distance is much smaller than the distance between the sensor node and the reference buoy node, forming a far-field condition. The trajectory of the positioning request information propagation is approximately parallel, and the positioning request information propagation distance is different. for:

in is the estimated position of sensor node p in the k-1th positioning process and the reference position of the reference buoy node b _n The projection component on the line.

Positioning request information propagation phase difference for:

in is the communication carrier wavelength, c is the speed of sound, and f is the carrier frequency. The second expression of the Doppler frequency shift in equation (13) is:

Combined with equations (17), (18), and (19), we can obtain the estimated position of the sensor node p during the k-1th positioning process by the actual motion velocity vector of the sensor node p: and the reference position of the reference buoy node b _n Projection component on the line The calculation expression is:

Step 4: Estimate the actual motion velocity vector of the sensor node from the k-1th positioning process to the kth positioning process based on the projection component of the actual motion velocity vector between the sensor node p and each reference buoy node _bn obtained in step 3;

In the embodiment, the projection component of the actual motion velocity vector of the sensor node p in step 3 is Written in vector form:

in for The absolute value of is the estimated position of sensor node p in the k-1th positioning process and the reference position of the reference buoy node b _n The unit direction vector on the connecting line is:

in are the unit vectors in the x, y, and z directions of the coordinate axes respectively. Passing through the location of the sensor node p And the projection component of the actual motion velocity vector of the sensor node p The general expression for the perpendicular plane equation is:

in are the plane equation parameters, x, y, z represent the coordinate axes, and the plane equation normal vector It can be expressed as:

The projection component of the actual motion velocity vector of the sensor node p It is essentially a set of normal vectors of the plane that satisfies equation (23), let

Then the plane equation parameters for:

The general expression of the plane equation (23) is the projection component of the actual motion velocity vector passing through the sensor node p Endpoint In vector form it is:

Set the vector endpoints The coordinates represented Substituting into the general expression (23) of the plane equation, we get the plane equation parameters

When n＝1,2,...,N, multiple sets of plane equations are expressed in matrix form:

in is the plane equation parameter matrix, is the intersection of the plane equation, and is the position of the sensor node p The actual velocity vector of the starting point The end point, It is expressed as:

The nodes involved need to satisfy the node position constraint: at least three reference buoy nodes are not located on the same straight line. Solve equation (29) using the least squares method to obtain the sensor node p with position is the estimated value of the actual motion velocity vector of the starting point The end point coordinates

During the k-1th positioning and the kth positioning, the actual motion velocity vector estimate of the sensor node p is for:

Step 5: According to the estimated value of the actual motion velocity vector of the sensor node obtained in step 4, the predicted value of the actual motion velocity vector of the sensor node during the kth positioning process to the k+1th positioning process is obtained, and the real-time position of the node is predicted until the new positioning process starts;

In the embodiment, during the kth positioning and the k+1th positioning, the actual motion speed vector prediction value of the sensor node p is for:

During the kth positioning and k+1th positioning, the positioning prediction value of sensor node p is for:

in is the estimated value of the k-th localization sensor node p, is the time when the sensor node p sends the positioning request information After the time has passed, satisfy hour, represents any time between the k+1th positioning and the kth positioning of sensor node p.

Step 6: Return to step 1, and loop through steps 1-5 to perform the next positioning, and continue to estimate the motion speed.

In the embodiment, for k=2, 3, ..., K, let k=k+1, and repeat step 1 to step 5. In specific implementation, the execution may be continued until the system stops the process.

In specific implementation, the method provided by the present invention can realize automatic operation of the process based on software technology, and the corresponding device for executing the process should also be within the protection scope of the present invention.

Figure 1 is a schematic diagram of a positioning scenario of an embodiment of the present invention. The positioning system consists of reference buoy nodes and sensor nodes. The nodes move passively with the water flow. The reference buoy node locates itself through the global positioning system of the satellite. Its real-time position is known and serves as a positioning reference node. The sensor node actively sends positioning request information and locates according to the positioning reply information returned by the reference buoy node. The successful positioning of the sensor node requires receiving at least three or more positioning reply information from reference buoy nodes that are not in the same straight line.

FIG2 is a schematic diagram of data interaction in an embodiment of the present invention. Send positioning request information at all times and At time n＝1,2,...,N,N≥3, the reference buoy nodes b ₁ ,b ₂ ,...,b _n ,n＝1,2,...,N,N≥3 are reached. The reference buoy nodes b ₁ ,b ₂ ,...,b _n ,n＝1,2,...,N,N≥3 are respectively The positioning reply information is returned at time n＝1,2,...,N,N≥3 and reaches the sensor node p. The arrival times are n＝1,2,...,N,N≥3.

FIG3 is a schematic diagram of the relationship between the real motion vector and the projected motion vector of the sensor node according to an embodiment of the present invention, and is illustrated by taking three reference buoy nodes as an example. In the k-1th positioning process, the positioning position of the sensor node p is From the k-1th positioning process to the kth positioning process, the sensor node p is located at The end point of the real motion velocity vector starting from is During the k-1th positioning process, the reference buoy node position is The real motion velocity vector of the sensor node p is located on the line connecting the sensor node and the reference buoy node. The projected velocity vector is a directed line segment, which can be expressed as A, B, and C are projection points. The sensor node p is located at The end point of the real motion velocity vector starting from is are projection points A, B, and C respectively, and are The common intersection of three perpendicular planes.

FIG4 is a schematic diagram of the passive motion velocity vector estimation of underwater nodes according to an embodiment of the present invention. In the k-1, k, k+1th positioning process, the sensor node p sends the positioning request information at the time respectively. The estimated positions obtained by the positioning algorithm are The reference buoy node _bn receives the positioning request information at the time When receiving positioning request information, the positions are During the k-1th positioning process to the kth positioning process, the actual motion velocity vector of the reference buoy node _bn is The actual motion velocity vector prediction value of the sensor node p obtained by the method described in the embodiment of the present invention is The estimated value is During the kth positioning process to the k+1th positioning process, the actual motion velocity vector of the reference buoy node _bn is The actual motion velocity vector prediction value of the sensor node p obtained by the method described in the embodiment of the present invention is The estimated value is During the k+1th positioning process to the k+2th positioning process, the actual motion velocity vector prediction value of the sensor node p obtained by the method described in the embodiment of the present invention is The actual motion velocity vector prediction value of the sensor node p satisfies the relationship with the estimated value

It should be understood that parts not elaborated in detail in this specification belong to the prior art.

It should be understood that the above description of the preferred embodiment is relatively detailed and cannot be regarded as limiting the scope of patent protection of the present invention. Under the enlightenment of the present invention, ordinary technicians in this field can also make substitutions or modifications without departing from the scope of protection of the claims of the present invention, which all fall within the scope of protection of the present invention. The scope of protection requested for the present invention shall be based on the attached claims.

Claims (3)

1. A method for estimating the passive motion speed of an underwater node based on Doppler frequency shift estimation, characterized in that it comprises the following steps: Step 1: Assume that the kth positioning is currently being performed, establish a positioning model based on the time difference of arrival (TDOA), obtain the communication time information between the underwater sensor node and the reference buoy node during the k-1th positioning process, and calculate the positioning position of the sensor node during the k-1th positioning process; Step 2, measuring the Doppler frequency shift according to the communication time information between the sensor node and the reference buoy node obtained in step 1, and the communication time information between the sensor node and the reference buoy node in the kth positioning process; Step 2 is implemented as follows: Assume that during the kth positioning process, the time when sensor node p sends the positioning request information is The location to be located is The time when the reference buoy node _bn receives the positioning request information is The reference position is The reference buoy node _bn sends the positioning response information at the time The time when the sensor node p receives the positioning response information of the reference buoy node _bn is According to step 1, the estimated value of the positioning position of the sensor node p in the kth positioning process is obtained In the kth positioning process and the k-1th positioning process, the time difference between the sensor node p sending the positioning request information is The frequency of sending positioning request information is The time difference of receiving the positioning request information by reference buoy node _bn is The frequency of receiving positioning request information is The Doppler frequency shift associated with the positioning request information between the sensor node p and the reference buoy node _bn is, in, Indicates that the nodes are relatively close. Indicates that the relative position of the node is far away; Step 3, according to the Doppler frequency shift measurement value obtained in step 2, and the actual motion velocity vector of the reference buoy node in the k-1th positioning process and the kth positioning process, calculate the projection component of the actual motion velocity vector of the sensor node on the line connecting the sensor node position and the reference buoy node position in the k-1th positioning process; Step 3 is implemented as follows: During the kth positioning process and the k-1th positioning process, the actual motion velocity vector of the reference buoy node b _n is for: The actual motion velocity vector of the reference buoy node _bn is the estimated position of the sensor node p during the k-1th positioning process. and the reference position of the reference buoy node b _n Projection on the line for: in Estimate the position of sensor node p and reference buoy node b _n reference position The angle between the line and the horizontal plane, In the kth positioning process and the k-1th positioning process, due to the slow passive movement of the sensor node and the reference buoy node with the water flow, During the time, the passive motion distance is much smaller than the distance between the sensor node and the reference buoy node, which constitutes a far-field condition, and the propagation trajectory of the positioning request information is approximately parallel; Location request information propagation distance difference for: in is the estimated position of sensor node p during the k-1th positioning process. and the reference position of the reference buoy node b _n The projection component on the connecting line; Positioning request information propagation phase difference for: in is the communication carrier wavelength, c is the speed of sound, f is the carrier frequency, and the second expression of the Doppler frequency shift in equation (13) is: Combined with equations (17), (18), and (19), we can obtain the estimated position of the sensor node p during the k-1th positioning process by the actual motion velocity vector of the sensor node p: and the reference position of the reference buoy node b _n Projection component on the line The calculation expression is: Step 4, based on the projection component of the actual motion velocity vector between the sensor node and each reference buoy node obtained in step 3, estimate the actual motion velocity vector of the sensor node from the k-1th positioning process to the kth positioning process; Step 4 is implemented as follows: The projection component of the actual motion velocity vector of the sensor node p in step 3 is Written in vector form: in for The absolute value of is the estimated position of sensor node p in the k-1th positioning process and the reference position of the reference buoy node b _n The unit direction vector on the connecting line is: in are the unit vectors in the directions of the coordinate axes x, y, and z respectively; Passing the location of sensor node p And the projection component of the actual motion velocity vector of the sensor node p The general expression for the perpendicular plane equation is: in are the plane equation parameters, x, y, z represent the coordinate axes, and the plane equation normal vector It can be expressed as: The projection component of the actual motion velocity vector of the sensor node p It is essentially a set of normal vectors of the plane that satisfies equation (23), let Then the plane equation parameter for: The general expression of the plane equation (23) is the projection component of the actual motion velocity vector passing through the sensor node p Endpoint In vector form it is: Set the vector endpoints The coordinates represented Substituting into the general expression (23) of the plane equation, we get the plane equation parameters When n＝1,2,...,N, multiple sets of plane equations are expressed in matrix form: in is the plane equation parameter matrix, is the intersection of the plane equation, and is the position of the sensor node p The actual velocity vector of the starting point The end point, It is expressed as: The nodes involved need to satisfy the node position constraint: at least three reference buoy nodes are not located on the same straight line; solve equation (29) using the least squares method to obtain the sensor node p with position The actual velocity vector estimate of the starting point The end point coordinates Among them, Ψ is the parameter to be determined, including the horizontal position information of the sensor node p; During the k-1th positioning and the kth positioning, the actual motion velocity vector estimate of the sensor node p is for: Step 5, according to the estimated value of the actual motion velocity vector of the sensor node obtained in step 4, obtain the predicted value of the actual motion velocity vector of the sensor node during the kth positioning process to the k+1th positioning process, and predict the real-time position of the node until the new positioning process starts; Step 5 is implemented as follows: During the kth positioning and k+1th positioning, the actual motion velocity vector prediction value of the sensor node p is for: During the kth positioning and k+1th positioning, the positioning prediction value of sensor node p is for: in is the estimated value of the k-th localization sensor node p, is the time when the sensor node p sends the positioning request information After the time has passed, satisfy hour, represents any time between the k+1th positioning and the kth positioning of sensor node p; Step 6, return to step 1, loop through steps 1-5 to perform the next positioning, and continue to estimate the motion speed. 2. According to the method for estimating the passive motion speed of underwater nodes based on Doppler frequency shift estimation in claim 1, it is characterized in that: in step 1, the positioning model is composed of underwater sensor nodes and reference buoy nodes on the water surface, and the reference buoy nodes on the water surface serve as positioning reference nodes to assist the positioning of underwater sensor nodes; the sensor nodes adopt active positioning method and realize positioning through a ranging method based on arrival time. 3. The method for estimating the passive motion speed of underwater nodes based on Doppler frequency shift estimation according to claim 2 is characterized in that: the sensor nodes are positioned by using a ranging method based on arrival time as follows: Let p = 1, 2, ..., P be the sensor node number, b _n be the one-hop neighbor reference buoy node within the communication range of sensor node p, and n = 1, 2, ..., N be the reference buoy node number; During the k-1th positioning process, the time when the sensor node p sends the positioning request information is The location to be located is The time when the reference buoy node _bn receives the positioning request information is The reference position is N is the total number of reference nodes that receive the positioning request information of sensor node p; the time when reference buoy node _bn sends the positioning response information is The time when the sensor node p receives the positioning response information of the reference buoy node _bn is According to the three-dimensional geometric relationship, the distance between the sensor node p and the reference buoy node b _n It is expressed as: distance Can be expressed as: Where c is the speed of sound, formula (1) is expanded to: in is a constant term related to the sensor node p and the reference buoy node b _n ; When n=1 and n≠1, according to the TDOA positioning principle, the distance difference between the sensor node p and the reference buoy nodes _b1 and _bn is for: Another expression of formula (4) is: in is the distance between the sensor node p and the reference buoy node _bn in the k-1th positioning process, is the distance difference between the reference buoy nodes b ₁ and b _n in the k-1th positioning process, is the distance between the sensor node p and the reference buoy node _b1 in the k-1th positioning process. Substituting equation (3) into equation (5) yields: In formula (6), The term is written in the form of formula (3), and formula (6) is rewritten as: in is the constant term related to the sensor node p and the reference buoy node _b1 during the k-1th positioning process, is the x,y coordinate value of the reference buoy node b ₁ ; The distance information between the sensor node p and the reference buoy nodes b ₁ , b _n , n = 2, 3, ..., N in the k-1th positioning process is written in matrix form and expressed as: H＝GΨ+W (8) In formula (8), W is the Gaussian noise matrix: in represents the Gaussian noise term associated with the reference buoy nodes _b1 and _bn , n=2,...,N in the k-1th positioning process, N(0, ^σ2 ) represents that the mean of the Gaussian noise is 0 and the variance is ^σ2 ; In formula (8), H and G are known constant matrices, which are expressed as: In formula (8), Ψ is the parameter to be determined, which contains the horizontal position information of the sensor node p and is expressed as: The least squares solution of formula (8) is: Ψ＝(G ^T G) ^-1 G ^T H (12) Where G ^T is the transposed matrix of G, () ^-1 indicates the matrix is inverted; The Chan algorithm and Taylor series expansion algorithm are used to solve equation (12) to obtain the estimated position of the sensor node p in the k-1th positioning process: