Indoor cluster target localization method

Abstract

The present invention discloses a method for positioning indoor cluster targets, which mainly solves the problems that the existing indoor cluster positioning method cannot determine the optimal virtual base station, there are cumulative errors in multi-hop positioning, and the positioning accuracy is poor. The implementation steps are: obtaining the ranging and self-movement information of all nodes; selecting an unlocated node with the smallest number of hops from the base station as the target to be positioned; calculating the trust value of each located node; using the located node as a virtual base station, selecting a virtual base station according to the trust value and calculating the initial position of the target to be positioned; calculating the position compensation vector of the target to be positioned, searching for its position compensation coefficient, and combining the compensated position coordinates of the two to calculate the updated position compensation vector; comparing the updated position compensation vector with the set threshold to determine the position of the target to be positioned. The present invention improves the positioning accuracy of the target by selecting a virtual base station to perform position compensation on the target to be positioned, and can be used for an indoor cluster positioning system.

Description

Technical Field

[0001] The invention belongs to the field of communication technology, and in particular relates to a method for positioning indoor cluster targets, which can be used for a positioning system in a large indoor space. Background Art

[0002] Services based on location information are widely used in transportation, medical care, logistics and other fields. Traditional outdoor positioning technologies such as satellites and communication base stations are mature and widely used. However, people spend most of their time indoors, so there is a strong demand for indoor positioning. However, due to factors such as building occlusion, outdoor positioning technology is not suitable for indoors. The industry has proposed various indoor cluster positioning methods. The existing indoor cluster positioning methods usually deploy multiple base stations first, use these base stations to measure the distance to the target through wireless signals at the same time, and then calculate the position of the target by measuring the arrival time TOA and the signal arrival time difference TDOA method. This method has a high system cost and poor positioning accuracy in non-line-of-sight scenarios.

[0003] In order to solve the above problems, in 2021, Li Mingdong proposed an indoor cluster positioning method in his master's thesis. This method introduces a two-node relative positioning algorithm and uses all located nodes as virtual base stations. When the number of base stations does not meet the conditions of the TOA method, a virtual base station is arbitrarily selected to calculate the position of the target using the two-node relative positioning algorithm. Although this method can locate clusters in large indoor spaces without the need to deploy base stations in advance, if the target to be located is within the coverage range of multiple virtual base stations, this method will not be able to determine the optimal virtual base station, affecting the positioning accuracy of the system. At the same time, this method has cumulative errors when performing multi-hop positioning of long-distance targets, and the positioning accuracy is poor. Summary of the invention

[0004] The purpose of the present invention is to address the deficiencies of the above-mentioned indoor cluster positioning solution and propose a positioning solution for indoor cluster targets, so as to reduce the positioning error of large indoor scenes and improve the positioning accuracy.

[0005] To achieve the above object, the technical solution of the present invention comprises the following steps:

[0006] 1. A method for positioning indoor cluster targets, characterized by comprising the following steps:

[0007] S1) Obtain ranging information and self-movement information of all nodes in the cluster;

[0008] S2) sorting the unlocated nodes in ascending order according to the number of hops from the base station, and selecting an unlocated node with the smallest number of hops from the base station as the target to be located;

[0009] S3) Calculate the trust value u of each located node i :

[0010]

[0011] Among them, R i = {r i,1 ,r i,2 ,...,r i,j ,…r i,K} is the set of neighboring nodes of the located node i, r i,j is the jth neighbor node of the located node i, K is the number of neighbor nodes of the located node i; δ i,j is the measured distance between the located node i and the located node j; d i,j The calculated distance between the located node i and the located node j is obtained according to the positioning coordinates; m u i is the mth historical trust value of the located node i; the upper limit τ of m is 4 to 5; α is a parameter that adjusts the weight of the historical trust value in the current trust value;

[0012] S4) All located nodes are used as virtual base stations, and the base station A with the largest trust value that can cover the target to be located is selected. Based on this base station, the initial position l of the target i to be located is calculated using the two-node relative positioning algorithm. i ;

[0013] S5) Based on the initial position l of the target i to be located i , the calculated distance d' between the node to be located i and the located node j i,j , calculate the position compensation vector

[0014]

[0015] in, is the unit vector from the target i to be located to its located neighboring node j, u j is the trust value of the located node j;

[0016] S6) Use the golden section method to search for the position compensation coefficient β along the position compensation vector Direction moves the initial position of the target to be located l i , get the position coordinates after compensation The position compensation vector of the target to be located is calculated using the formula in step S5).

[0017] S7) Set the threshold thr, and compensate the position vector of the target to be located. Compared with the threshold thr:

[0018] like by Replacement position compensation vector The position after compensation Replace the initial position l i , and then return to step S6);

[0019] like Then determine the coordinates after compensation is the position of the target i to be located.

[0020] The present invention proposes a positioning solution for indoor cluster targets, which selects located nodes as virtual base stations through node trust values, solving the problem that the existing methods cannot determine the optimal virtual base station. At the same time, the positioning results are iteratively compensated based on the node trust values, reducing the accumulated error in multi-hop positioning and improving the cluster positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flow chart for implementing the technical solution of the present invention. Specific implementation plan

[0022] The embodiments of the present invention are further described in detail below with reference to the accompanying drawings.

[0023] Reference Figure 1 ,The indoor cluster target positioning method of this example is implemented as follows:

[0024] Step 1: Obtain the ranging information and self-movement information of all nodes in the cluster.

[0025] The distance information is obtained by multiplying the flight time of the wireless signal by the speed of light through multiple communications between any node i in the cluster and all neighboring nodes j within its line of sight. i,j ;

[0026] The self-movement information is the precise movement trajectory of all nodes in a short period of time measured by an inertial navigation system, including the movement direction β and the movement distance n.

[0027] Step 2: determine the target to be located.

[0028] For nodes in the cluster that are adjacent to the base station, their position coordinates can be determined using existing positioning algorithms. However, for nodes that are not adjacent to the base station, i.e., unlocated nodes, they cannot be located using existing positioning algorithms. Therefore, this example sorts the unlocated nodes from small to large according to the number of hops from the base station, and then arbitrarily selects an unlocated node with the smallest number of hops from the base station as the target to be located.

[0029] Step 3: Calculate the trust value u of each located node based on the location coordinates of the located nodes. i .

[0030] 3.1) Calculate the distance d between the located node i and its located neighboring node j according to the location coordinates i,j :

[0031]

[0032] Among them, (x i ,y i ) is the position coordinate of the located node i, (x j ,y j ) is the position coordinate of the located node j;

[0033] 3.2) Based on the calculated distance d between the located node i and its located neighboring node j i,j , calculate the temporary amount ξ:

[0034]

[0035] Among them, R i = {r i,1 ,r i,2 ,...,r i,j ,...r i,K} is the set of neighboring nodes of the located node i, r i,j is the jth neighbor node of node i, K is the number of neighbor nodes of located node i; δ i,j is the measured distance between the located node i and the located node j; m u i is the mth historical trust value of the located node i; the upper limit τ of m is 4 to 5;

[0036] 3.3) According to the temporary amount ξ, adjust the weight parameter α of the historical trust value in the current trust value:

[0037]

[0038] 3.4) Calculate the trust value of the located node i according to the weight parameter α of the historical trust value in the current trust value:

[0039]

[0040] Among them, d i,j It is the calculated distance between the located node i and the located node j obtained according to the positioning coordinates.

[0041] Step 4: Based on the two-node relative positioning algorithm, calculate the initial position l of the target i to be located i .

[0042] The existing initial position l of the target i to be located can be calculated iThe methods include: two-node relative positioning algorithm, TOA algorithm, TDOA algorithm, AOA algorithm, etc. This example uses but is not limited to the two-node relative positioning algorithm to calculate the initial position l of the target i to be located i , the specific implementation is as follows:

[0043] 4.1) All located nodes are used as virtual base stations, and base station A with the largest trust value that can cover the target to be located is selected. The position coordinates of base station A are (x A ,y A ), the time points of the three most recent ranging measurements between base station A and target i are t1, t2, and t3, and the results of the three ranging measurements are d1, d2, and d3, respectively;

[0044] 4.2) Let the self-moving direction of the target i to be located measured at time t2 be β2 and the moving distance be n2; let the self-moving direction of the target i to be located measured at time t3 be β3 and the moving distance be n3;

[0045] 4.3) According to the measured distances between the base station A and the target i to be located at time t1 and t2 in step 4.1) and the self-moving direction β2 and moving distance n2 of the target i to be located measured at time t2 in step 4.2), the following set of equations is constructed:

[0046]

[0047] Among them, (x i ,y i ) is the position coordinate of the target i to be located that needs to be solved;

[0048] 4.4) Solve the above equations and determine the initial position l of the target i to be located according to the solution results i :

[0049] If the system of equations has no solution, target i cannot be located, and the next distance measurement is performed and the process returns to step 4.1);

[0050] If the system of equations has only one set of solutions, then this set of solutions is determined as the initial position l of the target i to be located i ;

[0051] If the system of equations has two solutions, proceed to step 4.5);

[0052] 4.5) From the two solutions of the above equations, select one of the solutions as the initial position l of the node i to be located i .

[0053] 4.5.1) Let the two sets of solutions of the equations be According to the location coordinates of base station A (x A ,y A) Calculate the distance between base station A and target i at time t3

[0054]

[0055] 4.5.2) Calculate the calculated distance between base station A and the target i to be located at time t3 corresponding to the two sets of solutions The difference between the measured distance d3 and the measured distance d4 is used to select one of the solutions as the initial position l of the node i to be located according to the following rules: i :

[0056] like Then confirm is the initial position l of the target i to be located i ;

[0057] like Then confirm is the initial position l of the target i to be located i .

[0058] Step 5: Based on the initial position l of the target i to be located i , calculate the position compensation vector

[0059] 5.1) Based on the initial position l of the target i to be located i , calculate the distance d between the node to be located i and the located node j i ' ,j :

[0060]

[0061] Among them, (x i ',y i ') is the initial position l of the target i to be located i Coordinates, (x j ,y j ) is the position coordinate of the located node j;

[0062] 5.2) Based on the initial position l of the target i to be located i , calculate the unit vector from the initial position of the target i to be located to the located node j

[0063]

[0064] 5.3) According to the calculated distance d between the node to be located i and the located node j i ' ,j , the unit vector from the target i to be located to the located node j Calculate the position compensation vector

[0065]

[0066] Among them, u j is the trust value of the located node j; δ i,j is the measured distance between the node i to be located and the located node j.

[0067] Step 6: Search for the position compensation coefficient β and combine it with the position compensation vector Calculate the position coordinates of the node to be located after compensation And calculate the node to be located at Position compensation vector at

[0068] 6.1) Using the existing golden section method, in the range [0,8], along the node to be located at the initial position l i Position compensation vector at Direction, search position compensation coefficient β;

[0069] 6.2) Based on the position compensation coefficient β, the node to be located is at the initial position l i Position compensation vector at Move the initial position of the target to be located l i , get the position coordinates after compensation

[0070]

[0071] 6.3) Let the position coordinates after compensation be according to Calculate the distance between the node i to be located and the located node j after position compensation

[0072]

[0073] Among them, (x j ,y j ) is the position coordinate of the located node j;

[0074] 6.4) Let the position coordinates of the node to be located after compensation according to Calculate the position after compensation from the node i to be located Unit vector pointing to located node j

[0075]

[0076] 6.5) Based on the calculated distance between the node to be located i and the located node j Calculate the position of the node to be located after compensation Position compensation vector at

[0077]

[0078] Step 7: Based on the position of the node to be located after compensation Position compensation vector at Determine the position of the target i to be located.

[0079] Set the position compensation threshold thr∈(0,1], and convert the position compensation vector of the target to be positioned after position compensation into Compared with the threshold thr:

[0080] like Then Replacement position compensation vector The position after compensation Replace the initial position l i , and then return to step S6);

[0081] like Then determine the coordinates after compensation is the position of the target i to be located.

[0082] The above description is only a specific example of the present invention and does not constitute any limitation to the present invention. It is obvious that for professionals in this field, after understanding the content and principles of the present invention, they may make various modifications and changes in form and details without departing from the principles and structures of the present invention. However, these modifications and changes based on the ideas of the present invention are still within the scope of protection of the claims of the present invention.

Claims

1. A method for positioning indoor cluster targets, characterized by: The steps include: S1) Obtain ranging information and self-movement information of all nodes in the cluster; S2) sorting the unlocated nodes in ascending order according to the number of hops from the base station, and selecting an unlocated node with the smallest number of hops from the base station as the target to be located; S3) Calculate the trust value u of each located node i : Among them, R i = {r i,1 ,r i,2 ,...,r i,j ,...r i,K } is the set of neighboring nodes of the located node i, r i,j is the jth neighbor node of the located node i, K is the number of neighbor nodes of the located node i; δ i,j is the measured distance between the located node i and the located node j; d i,j The calculated distance between the located node i and the located node j is obtained according to the positioning coordinates; m u i is the mth historical trust value of the located node i; the upper limit τ of m is 4 to 5; α is a parameter that adjusts the weight of the historical trust value in the current trust value; S4) All located nodes are used as virtual base stations, and the base station A with the largest trust value that can cover the target to be located is selected. Based on this base station, the initial position l of the target i to be located is calculated using the two-node relative positioning algorithm. i ; S5) Based on the initial position l of the target i to be located i , the calculated distance d' between the node to be located i and the located node j i，j , calculate the position compensation vector in, is the unit vector from the target i to be located to its located neighboring node j, u j is the trust value of the located node j; S6) Use the golden section method to search for the position compensation coefficient β along the position compensation vector Direction moves the initial position of the target to be located l i , get the position coordinates after compensation l i + , using the formula in step S5) to calculate the position compensation vector of the target to be positioned S7) Set the threshold thr, and compensate the position vector of the target to be located. Compared with the threshold thr: like by Replacement position compensation vector The position after compensation Replace the initial position l i , and then return to step S6); like Then determine the coordinates after compensation is the position of the target i to be located.

2. The method according to claim 1, characterized in that In step S3), the calculated distance d between the located node i and its located neighboring node j is obtained according to the positioning coordinates. i,j , the formula is as follows: Among them, (x i ,y i ) is the position coordinate of the located node i, (x j ,y j ) is the position coordinate of the located node j.

3. The method according to claim 1, characterized in that In step S3), the parameter α of the weight of the historical trust value in the current trust value is adjusted as follows: S3a) Based on the measured distance between the located node i and the located node j, and the calculated distance d between the located node i and the neighboring node j i,j , calculate the temporary amount ξ: S3b) Adjust the weight parameter α of the historical trust value in the current trust value according to the temporary amount ξ:

4. The method according to claim 1, characterized in that In step S4), the initial position l of the target i to be located is calculated based on the selected base station A using a two-node relative positioning algorithm. i , implemented as follows: S4a) Let the location coordinates of base station A be (x A ,y A ), the time points of the three most recent ranging measurements between base station A and target i are t1, t2, and t3, and the results of the three ranging measurements are d1, d2, and d3, respectively; S4b) Let the self-moving direction of the target i to be located measured at time t2 be β2, and the moving distance be n2; let the self-moving direction of the target i to be located measured at time t3 be β3, and the moving distance be n3; S4c) constructs the following set of equations based on the measured distances between the base station A and the target i to be located at times t1 and t2 in step S4a), and the self-moving direction β2 and moving distance n2 of the target i to be located measured at time t2 in step S4b): Among them, (x i ,y i ) is the position coordinate of the target i to be located that needs to be solved; S4d) Solve the above equations to determine the initial position l of the target i to be located i : If the system of equations has no solution, target i cannot be located, and the process returns to step S4a) when the next distance measurement is performed; If the system of equations has only one set of solutions, then this set of solutions is determined as the initial position l of the target i to be located i ; If the system of equations has two solutions, execute step S4e); S4e) Let the two sets of solutions of the equations be According to the location coordinates of base station A (x A ,y A ) Calculate the distance between base station A and target i at time t3 S4f) Calculate the calculated distance between base station A and the target i to be located at time t3 corresponding to the two sets of solutions The difference between the measured distance d3 and the measured distance d4 is used to select one of the solutions as the initial position l of the node i to be located according to the following rules: i : like Then confirm is the initial position l of the target i to be located i ; like Then confirm is the initial position l of the target i to be located i .

5. The method according to claim 1, characterized in that In step S5), based on the initial position l of the target i to be located i , calculate the distance d' between the node to be located i and the located node j i,j , the formula is as follows: Among them, (x' i ,y' i ) is the initial position l of the target i to be located i Coordinates, (x j ,y j ) is the position coordinate of the located node j.

6. The method according to claim 1, characterized in that In step S6), based on the position compensation vector and position compensation coefficient β to move the initial position l of the target i to be located i , obtain the position coordinates l of the target i to be located after compensation i + , the formula is as follows: .

Images