CN109257693B - An indoor collaborative positioning method based on spatial analysis - Google Patents

Abstract

The invention discloses an indoor cooperative positioning method based on space analysis, which belongs to the field of indoor positioning. Including: improved ray tracing algorithm for positioning space analysis; indoor space area determination algorithm based on improved ray tracing technology; two-stage cooperative positioning algorithm based on space analysis. According to the relatively stable spatial structure in the indoor environment, the environment is analyzed based on ray tracing technology, different positioning areas are divided, and the number of positioning reference signals in the line-of-sight environment of each area is counted, thereby judging the direct positioning area and Collaborative positioning area. According to the spatial partition, the target nodes that need cooperative localization and those that do not need cooperative localization are determined, and an indoor cooperative localization method based on spatial analysis is proposed. This indoor cooperative positioning method effectively avoids the problem of reduced positioning accuracy caused by moving reference nodes with large errors, and improves the overall positioning accuracy.

Description

An indoor collaborative positioning method based on spatial analysis

technical field

The invention belongs to the field of indoor positioning, in particular to an indoor cooperative positioning method based on spatial analysis.

Background technique

The essence of indoor positioning technology is to estimate the position of the terminal to be positioned. The variables required for positioning in the indoor positioning algorithm mainly include signal arrival time, arrival angle, signal arrival time difference, received signal strength and many other types with different calculation methods.

The cooperative positioning technology has two main advantages: First, due to the short distance between users, the point-to-point (P2P) measurement between the terminals to be located may be very accurate, so even if all the terminals to be located are relative to all available reference points Under the condition of non-line-of-sight, the accuracy of the positioning system can also be improved. Second, due to the characteristics of spatial diversity of mobile terminals, if at least one terminal to be positioned and one transmitting terminal are located under the condition of line-of-sight, they can be positioned through cooperative positioning. Avoid shadow effects, thereby improving positioning accuracy.

Relevant documents are: Khalaf-Allah M.Time of arrival(TOA)-based direct locationmethod[C]//Radar Symposium(IRS),2015 16th International.IEEE,2015:812-815;

Fang D,Chong S,Qian G.Research on Multipoint Positioning Based on TOACooperate with AOA Location Algorithm[J].DEStech Transactions on ComputerScience and Engineering,2016(itms);

Meng Y,Xu J,Huang Y,et al.Key factors of multi-station TDOA passivelocation study[C]//Intelligent Human-Machine Systems and Cybernetics(IHMSC),2015 7th International Conference on.IEEE,2015,2:220 -223;

Bohidar S, Behera S, Tripathy C R. A comparative view on received signalstrength(RSS) based location estimation in WSN[C]//Engineering and Technology(ICETECH), 2015 IEEE International Conference on. IEEE, 2015: 1-7;

Stephen Sander, Armin Daman, Kristin Mönching. Positioning Technology and Application in Wireless Communication System [M]. Beijing: Machinery Industry Press, 2016: 56-59;

Yun Z, Iskander M F. Ray tracing for radio propagation modeling: principles and applications[J]. IEEE Access, 2015, 3: 1089-1100.

SUMMARY OF THE INVENTION

The purpose of the present invention is to solve the problem that non-line-of-sight cooperative positioning partitions cannot be positioned in indoor positioning, and propose an indoor cooperative positioning method based on spatial analysis to improve the positioning accuracy of the direct positioning area of the line-of-sight environment.

The object of the present invention is achieved through the following technical solutions:

1. Design an improved ray tracing algorithm for positioning space analysis, preprocess the fixed reference point position in the space about the mirror point information in the space, and eliminate the calculation process of part of the plane by establishing the reflection surface filter set method. Specific steps include:

(1) Initialize the fixed reference point information, that is, the spatial three-dimensional position coordinates of the transmitter of the positioning signal can be obtained;

(2) Initialize the positioning reference signal information, that is, characteristic parameters such as signal strength can be obtained;

(3) Initialize the position information of the target node, that is, obtain the spatial three-dimensional position coordinates of the receiving end;

(4) Initialize the positioning space data, that is, obtain the relevant data information of the indoor environment affecting the positioning signal propagation, specifically including the plane composition of each object in the space, the plane equation parameters in each plane, the coordinates of the boundary bumps, the plane material, etc.;

(5) Establish a spatial plane filtering set, and put the planes that cannot participate in the propagation calculation into the set;

(6) preprocessing the mirror point set of the fixed reference point in space, and calculating the mirror point of the fixed reference point with respect to other planes in the space except the plane filter set;

(7) According to the different propagation modes of the positioning reference signal, the reflection scene goes to execute (8), the transmission scene goes to execute (9), and the diffraction scene goes to (10);

(8) Carry out reflection calculation, fix the reference point as a mirror reference point on the plane, and conduct an intersection test with the reflection plane through the connection between the mirror reference point and the target node to determine whether the intersection point is in the plane; if it is in the plane, according to the reflection material and the incident angle to calculate the loss of the reflected signal, and execute (9); if the calculated intersection point is not in the plane (may be in the plane extension area where the plane is located or on the extension line connecting the mirror reference point and the target node), the reflecting surface is invalid , will not be calculated;

(9) Carry out transmission calculation, calculate the attenuation value of the signal by passing through the material and thickness of the object, and execute (11);

(10) Carry out diffraction calculation, calculate the loss value of the signal through the material of the diffracted object and the diffraction angle, and execute (12);

(11) Statistical signals propagate paths in space in different ways, as well as the corresponding loss and delay of each path, and generate a path parameter table;

(12) Compare the signal strength in the diameter parameter table with the minimum signal strength constraint threshold. If the current signal strength is greater than the minimum signal strength threshold, it means that the signal strength can be calculated in the next step, and execute (5); if the signal strength is less than or equal to The minimum signal strength threshold means that the signal is attenuated to the receiving end and cannot be analyzed, so the calculation of the current signal is ended, and (11) is executed;

(13) Persistently store the current signal propagation path and the corresponding loss and delay for subsequent calculation and analysis.

2. Design an indoor space area determination algorithm based on improved ray tracing technology, determine whether a certain area in the space meets the required number of sight distances, and distinguish direct positioning zones and cooperative positioning zones. Specific steps include:

(1) Relevant parameters of the initialization judgment algorithm, the number of space partitions ND, the position sampling rate PLOS in the unit area (side length 1m) inside the partition, the minimum coverage rate of the partition line-of-sight environment Pmin, and the minimum signal threshold RSSmin;

(2) divide it into ND rectangular three-dimensional grids and number each grid according to the number of space partitions ND and the positioning space area, and denote it as Di;

(3) Calculate the corresponding number of line-of-sight path sampling points NLOS through the sampling rate PLOS in each partition, and randomly generate the coordinates of each line-of-sight path sampling point, denoted as P(x _i , y _i , z _i ) ;

(4) In the partition, according to the line-of-sight sampling points generated in step (3), the signal direct path propagation calculation is performed based on the ray tracing method in Section 2.2.2, where the sampling points can receive the line-of-sight signal of the fixed reference point positioning signal The standard is: the positioning signal propagates through direct rays and when it reaches the sampling point, the receiving intensity is greater than RSSmin;

(5) After all sampling points in the partition have completed the calculation of the LOS signal path based on ray tracing, count the number of reference points where the received LOS signal can satisfy 4 or more, and calculate the coverage rate PDi of the locatable sampling points , if PDi is greater than or equal to Pmin, then the area is a direct positioning partition of the line-of-sight environment; otherwise, the area is a cooperative positioning partition of a non-line-of-sight environment.

3. Design a two-stage cooperative positioning algorithm based on spatial analysis. In the first stage, the target node receives the positioning signals of all reference points, performs the initial positioning of the first stage, and judges the partition according to the positioning information. In the second stage, according to the partition and space division determined in the first stage, the target node determines whether to add the surrounding mobile reference node information to its own positioning reference information.

The purpose of initial positioning in the first stage is to determine the space partition where the target node is located. Since the target node positioning also has errors and is theoretically larger than the positioning error of the second stage, if the judgment partition is wrong, it will affect the result of the second stage cooperative positioning. The first-stage positioning is based on the coverage rate of the indoor partition in the common coverage area of the signal, and the partition where the target node is located is determined. As shown in Figure 1, the target node receives positioning signals from reference points R1, R2, and R3, and the estimated distances from each reference point to the target are L1, L2, and L3, respectively. The coverage of the three reference points R1, R2 and R3 is shown in Figure 1. The area covered by the three reference points is D1, D2, D3 and D4. According to the geometric area calculation, it can be concluded that the common signal coverage area ratio D3>D1>D2>D4 in the partition. From this, it can be inferred that the probability that the target node is located in the D3 area is relatively large, so it is determined that the area where the target node is located is the D3 area. The algorithm is described as follows:

(1) The target node obtains the reference node set _CR that can receive the positioning signal;

(2) Perform ranging estimation on the target node by each reference point in the reference point set _CR , and calculate the distance set _CL respectively;

(3) According to the sets _CR and _CL , determine the common coverage area _DA of the reference point;

(4) obtaining spatial partition information based on the spatial analysis algorithm, and obtaining the partition set _CD that can be covered by the area _DA with the most common coverage of the signals;

(5) calculate the coverage area of each partition in the set _{CD respectively, and calculate the partition coverage PD ;}

(6) If the partition with the largest coverage in the set _CD is unique, then adopt step (7); otherwise, enter step (8);

(7) Select the partition with the largest coverage, which is the partition where the target node is determined;

(8) Calculate the centroid coordinates of the common coverage area _DA and the centroid coordinates of the multiple maximum coverage zones. According to the Euclidean distance formula, the partition where the centroid with the shortest distance from the common coverage area _DA is located is determined as the partition where the target point is located.

The second stage positioning algorithm is described as follows:

(1) Data preprocessing definition, set the number of iterations k;

(2) Preprocessing of positioning space information, partitioning the indoor space, and judging whether each area is a direct positioning area or a cooperative positioning area;

(3) In the first stage, coarse positioning determines the partition where the target node is located. The target node receives the positioning information of the surrounding moving reference points and fixed reference points, and performs positioning according to the TDOA positioning algorithm to determine that the target node is in a certain area in step (2). a division;

(4) if the target node through step (2) is in the direct positioning sub-region where the LOS environment meets the positioning condition, then the second-stage positioning is performed based on the reference information of the fixed node again; otherwise, the second-stage positioning is not performed;

(5) After the target node completes the positioning, it sends its own position and positioning information to the surrounding target nodes and mobile reference;

(6) For the mobile reference point information that has received the surrounding position update, the target node has two options. If the target node is in the LOS environment, it does not need to be updated; step (5);

(7) If the number of iterations of positioning in the space is greater than the specified k or the global position information is no longer changed, the positioning is terminated.

The beneficial effects of the present invention are:

On the basis of traditional indoor cooperative positioning, the LOS environment analysis of indoor positioning space based on ray tracing method is supplemented, and different areas are divided and the number of LOS reference signals that can receive fixed reference points in each area is counted. Improve ray tracing efficiency with an improved ray tracing method for spatial analysis. By judging whether the area is a direct positioning area, it is decided whether to add the signal data of the surrounding mobile reference nodes into the positioning algorithm, so as to avoid the lowering of the positioning accuracy caused by the mobile reference nodes with large errors, and improve the overall positioning accuracy.

Description of drawings

Fig. 1 is a fixed reference point signal coverage map;

Figure 2 is a graph of ray tracing results;

Fig. 3 is a regional judgment result diagram;

4 is a schematic diagram of a cooperative positioning relationship;

Figure 5 is a comparison diagram of cooperative positioning results.

Detailed ways

The specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings:

An indoor collaborative positioning method based on spatial analysis, comprising the following steps:

(1) Using the improved ray tracing algorithm for positioning space analysis, preprocess the information of the fixed reference point position in the space about the mirror point information in the space, and establish the reflection surface filter set method;

(2) Using the indoor space area determination algorithm based on improved ray tracing technology, determine whether a certain area in the space meets the required number of sight distances, and distinguish the direct positioning zone and the cooperative positioning zone;

(3) Using a two-stage cooperative positioning algorithm based on spatial analysis, in the first stage, the target node receives the positioning signals of all reference points, performs initial positioning in the first stage, and judges the partition according to the positioning information, and in the second stage according to the first stage In the partition and space division determined in the first stage, the target node determines whether to add the surrounding mobile reference node information to its own positioning reference information.

The specific steps of the improved ray tracing algorithm for positioning space analysis:

(1.1) Initialize the fixed reference point information, and obtain the spatial three-dimensional position coordinates of the transmitter of the positioning signal;

(1.2) Initialize the positioning reference signal information, and obtain the signal characteristic parameters;

(1.3) Initialize the position information of the target node, and obtain the spatial three-dimensional position coordinates of the receiving end;

(1.4) Initialize the positioning space data, and obtain the data information that the indoor environment affects the propagation of the positioning signal, including the plane composition of each object in the space, the plane equation parameters in each plane, the coordinates of the boundary bumps, and the plane material;

(1.5) Establish a spatial plane filtering set, and put the planes that cannot participate in the propagation calculation into the set;

(1.6) Preprocess the mirror point set of the fixed reference point in space, and calculate the mirror point of the fixed reference point with respect to other planes in the space except the plane filter set;

(1.7) According to the different propagation modes of the positioning reference signal, the reflection scenario goes to execute (1.8), the transmission scenario goes to execute (1.9), and the diffraction scenario goes to (1.10);

(1.8) Perform reflection calculation, use the fixed reference point as a mirror reference point with respect to the plane, and conduct an intersection test with the reflection plane through the line connecting the mirror reference point and the target node to determine whether the intersection point is in the plane; if it is in the plane, according to the reflection material and the incident angle to calculate the loss of the reflected signal, and execute (1.9); if the calculated intersection point is not in the plane, the reflecting surface is invalid, and the calculation is not performed;

(1.9) Perform transmission calculation by calculating the attenuation value of the signal through the material and thickness of the object, and perform (1.11);

(1.10) Perform diffraction calculation Calculate the loss value of the signal through the material of the diffracting object and the diffraction angle, and execute (1.12);

(1.11) Statistical signals propagate paths in space in different ways, as well as the corresponding loss and delay of each path, and generate a path parameter table;

(1.12) Compare the signal strength in the diameter parameter table with the minimum signal strength constraint threshold, if the current signal strength is greater than the minimum signal strength threshold, execute (1.5); if the signal strength is less than or equal to the minimum signal strength threshold, end the current signal Calculation, execute (1.11);

(1.13) Persistently store the current signal propagation path and the corresponding loss and delay.

The specific steps of the indoor space area determination algorithm based on the improved ray tracing technology:

(2.1) The relevant parameters of the initialization judgment algorithm, the number of space partitions ND, the position sampling rate PLOS in the unit area (side length 1m) inside the partition, the minimum coverage rate of the partition line-of-sight environment Pmin, and the minimum signal threshold RSSmin;

(2.2) divide it into ND rectangular three-dimensional grids according to the number of space partitions ND and the positioning space area and number each grid, denoted as Di, and i is a positive integer;

(2.3) Calculate the corresponding number of line-of-sight path sampling points NLOS through the sampling rate PLOS in each partition, and randomly generate the coordinates of each line-of-sight path sampling point, denoted as P(x _i , y _i , z _i ) ;

(2.4) In the partition, carry out the signal direct path propagation calculation according to the line-of-sight sampling points generated in step (2.3);

(2.5) After all the sampling points in the partition have completed the calculation of the LOS signal path based on ray tracing, count the number of reference points where the received LOS signal can satisfy more than or equal to 4, and calculate the coverage rate PDi of the locatable sampling points , if PDi is greater than or equal to Pmin, then the area is a direct positioning partition of the line-of-sight environment; otherwise, the area is a cooperative positioning partition of a non-line-of-sight environment.

The criterion that the sampling point can receive the line-of-sight signal of the fixed reference point positioning signal in the step (2.4) is that the positioning signal propagates through direct rays and reaches the sampling point with a receiving intensity greater than RSSmin.

The specific steps of the first stage positioning algorithm:

(3.1.1) The target node obtains the reference node set _CR that can receive the positioning signal;

(3.1.2) each reference point in the reference point set _CR carries out ranging to the target node, and calculates the distance set _CL respectively;

(3.1.3) According to the sets _CR and _CL , determine the common coverage area _DA of the reference point;

(3.1.4) obtaining spatial partition information based on the spatial analysis algorithm, and obtaining the partition set _CD that can be covered by the area _DA with the most common coverage of the signals;

(3.1.5) Calculate the coverage area of each partition in the set _{CD respectively, and calculate the partition coverage rate PD ;}

( _3.1.6 ) If the partition with the largest coverage in the set CD is unique, then step (3.1.7) is adopted; otherwise, go to step (3.1.8);

(3.1.7) Select the partition with the largest coverage, which is the partition where the target node is determined;

( _3.1.8 ) Calculate the centroid coordinates of the common coverage area DA and the centroid coordinates of multiple zones with the largest coverage.

The specific steps of the second stage positioning algorithm:

(3.2.1) Data preprocessing definition, set the number of iterations k;

(3.2.2) Preprocessing of positioning space information, partitioning the indoor space, and judging whether each area is a direct positioning area or a cooperative positioning area;

(3.2.3) The first stage of coarse positioning determines the partition where the target node is located. The target node receives the positioning information of the surrounding moving reference points and fixed reference points, and performs positioning according to the TDOA positioning algorithm to determine that the target node is in step (3.2.2. ) in the direct positioning area or the cooperative positioning area;

(3.2.4) If the target node after step (3.2.2) is in the direct positioning partition where the LOS environment meets the positioning conditions, the second stage positioning is performed based on the reference information of the fixed node; otherwise, the second stage is not performed. position;

(3.2.5) After the target node completes the positioning, it sends its own position and positioning information to the surrounding target nodes and mobile reference;

(3.2.6) For the mobile reference point information that has received the surrounding position update, the target node has two options. If the target node is in the LOS environment, it does not need to be updated; And proceed to step (3.2.5);

(3.2.7) If the number of iterations of positioning in the space is greater than the specified k or the global position information does not change, the positioning is terminated.

Validity verification of spatial region determination algorithm based on improved ray tracing technology:

The number of fixed reference points NF=7, CF={F1, F2, F3, F4, F5, F6, F7}, the fixed reference point adopts the default honeycomb radius R=100m, and the position of each point is the default coordinate in Section 2.4.1. The transmit power signal strength of the fixed reference point is 23dBm, the frequency is 1.8GHz, and the minimum signal receiving strength is set to RSSmin=-105dBm. The number of partitions ND=90 will be surrounded by four points of N1, N9, N10 and N11 (excluding the area surrounded by four points of N4, N9, N10 and N5). The sampling rate PLOS=100%, that is, the number of sampling points selected in each partition is NLOS=1. In the present invention, the sampling point is set as the position of the center of each partition, and the minimum coverage rate Pmin=100 of the direct positioning partition of the line-of-sight environment can be determined. %.

Figures 2 and 3 show the verification results of the spatial region determination algorithm. Figure 2 shows that each path indicates the line-of-sight path that the sampling point can receive. Figure 2 shows the judgment results in each area, in which the red dot indicates that the area with a side length of 1m as the center is the direct positioning zone, and the blue and white dots indicate that it is in the cooperative positioning zone. The blue dot area indicates that only one line-of-sight path can be converted into a direct positioning partition. This part of the area can improve the positioning accuracy through cooperation, while the red dot area can directly complete the positioning without cooperation.

Comparison and verification of cooperative positioning algorithm errors:

Using the schematic diagram of cooperative positioning based on Fig. 4, the fixed reference points are F1, F2, F3, the fixed reference point adopts the default cellular radius R=100m, and the position of each point is the default coordinate in 2.4.1. The target nodes T1 and T2 can exchange information between them. Based on the spatial analysis, T1 is set to be in the direct positioning partition, T2 is set to the cooperative positioning partition, and the partitions where T1 and T2 are located are adjacent. Among them, T1 is located in a square partition with center coordinates (50, 50/3, 4.2) and a side length of 2m, and T2 is located in a square partition with center coordinates (50, 50/3-2, 4.2) and a side length of 2m, Considering the typical scene of indoor personnel, always keep the height coordinate z=4.2m. The connecting line in the fixed reference point and the target node graph indicates that there is a LOS propagation signal, and the ranging error obeys a Gaussian distribution with mean = 0 and standard deviation = by default. The ranging error of the target node (mobile reference node) follows a Gaussian distribution with mean = 0 and standard deviation = 1 by default. The simulation experiment is the result of 10,000 random target points T1 and T2 measurement errors.

Figure 5 shows the influence of the measurement error of the fixed reference point on the final positioning result, where the measurement error Noise varies from 1 to 10 m. It can be seen from Fig. 5 that the SACP algorithm proposed by the present invention is less affected by the measurement error than the traditional cooperative positioning CP algorithm. When Noise=1m, the accuracy proposed by the present invention is 0.4869 lower than the traditional cooperative positioning method RMSE meters, the positioning accuracy is improved by 34.6%. Based on the traditional CP algorithm, the SACP algorithm proposed by the present invention not only ensures that the target node in the cooperation area can complete the positioning, but also avoids the target node in the direct positioning area from being interfered by the data of the moving reference point in the cooperation area with a large position error. To a certain extent, the positioning effect is improved.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims (2)

1. an indoor collaborative positioning method based on space analysis, is characterized in that, comprises the following steps: (1) Using the improved ray tracing algorithm for positioning space analysis, preprocess the information of the fixed reference point position in the space about the mirror point information in the space, and establish the reflection surface filter set method; (2) Using the indoor space area determination algorithm based on improved ray tracing technology, determine whether a certain area in the space meets the required number of sight distances, and distinguish the direct positioning zone and the cooperative positioning zone; (3) Using a two-stage cooperative positioning algorithm based on spatial analysis, in the first stage, the target node receives the positioning signals of all reference points, performs initial positioning in the first stage, and judges the partition according to the positioning information, and in the second stage according to the first stage The partition and space division determined in the first stage, the target node judges whether to add the surrounding mobile reference node information to its own positioning reference information; The specific steps of the improved ray tracing algorithm for positioning space analysis: (1.1) Initialize the fixed reference point information, and obtain the spatial three-dimensional position coordinates of the transmitter of the positioning signal; (1.2) Initialize the positioning reference signal information, and obtain the signal characteristic parameters; (1.3) Initialize the position information of the target node, and obtain the spatial three-dimensional position coordinates of the receiving end; (1.4) Initialize the positioning space data, and obtain the data information that the indoor environment affects the propagation of the positioning signal, including the plane composition of each object in the space, the plane equation parameters in each plane, the coordinates of the boundary bumps, and the plane material; (1.5) Establish a spatial plane filtering set, and put the planes that cannot participate in the propagation calculation into the set; (1.6) Preprocess the mirror point set of the fixed reference point in space, and calculate the mirror point of the fixed reference point with respect to other planes in the space except the plane filter set; (1.7) According to the different propagation modes of the positioning reference signal, the reflection scenario goes to execute (1.8), the transmission scenario goes to execute (1.9), and the diffraction scenario goes to (1.10); (1.8) Perform reflection calculation, use the fixed reference point as a mirror reference point with respect to the plane, and conduct an intersection test with the reflection plane through the line connecting the mirror reference point and the target node to determine whether the intersection point is in the plane; if it is in the plane, according to the reflection material and the incident angle to calculate the loss of the reflected signal, and execute (1.9); if the calculated intersection point is not in the plane, the reflecting surface is invalid, and the calculation is not performed; (1.9) Perform transmission calculation by calculating the attenuation value of the signal through the material and thickness of the object, and perform (1.11); (1.10) Perform diffraction calculation Calculate the loss value of the signal through the material of the diffracting object and the diffraction angle, and execute (1.12); (1.11) Statistical signals propagate paths in space in different ways, as well as the corresponding loss and delay of each path, and generate a path parameter table; (1.12) Compare the signal strength in the diameter parameter table with the minimum signal strength constraint threshold, if the current signal strength is greater than the minimum signal strength threshold, execute (1.5); if the signal strength is less than or equal to the minimum signal strength threshold, end the current signal Calculation, execute (1.11); (1.13) Persistently store the current signal propagation path and the corresponding loss and delay; The specific steps of the indoor space area determination algorithm based on the improved ray tracing technology: (2.1) The relevant parameters of the initialization judgment algorithm, the number of space partitions ND, the position sampling rate PLOS in the unit area (side length 1m) inside the partition, the minimum coverage rate of the partition line-of-sight environment Pmin, and the minimum signal threshold RSSmin; (2.2) divide it into ND rectangular three-dimensional grids according to the number of space partitions ND and the positioning space area and number each grid, denoted as Di, and i is a positive integer; (2.3) Calculate the corresponding number of line-of-sight path sampling points NLOS through the sampling rate PLOS in each partition, and randomly generate the coordinates of each line-of-sight path sampling point, denoted as P(x _i , y _i , z _i ) ; (2.4) In the partition, carry out the signal direct path propagation calculation according to the line-of-sight sampling points generated in step (2.3); (2.5) After all the sampling points in the partition have completed the calculation of the LOS signal path based on ray tracing, count the number of reference points where the received LOS signal can satisfy more than or equal to 4, and calculate the coverage rate PDi of the locatable sampling points , if PDi is greater than or equal to Pmin, the area is the direct positioning partition of the line-of-sight environment; otherwise, the area is the cooperative positioning partition of the non-line-of-sight environment; The specific steps of the first-stage positioning algorithm: (3.1.1) The target node obtains the reference node set _CR that can receive the positioning signal; (3.1.2) each reference point in the reference point set _CR carries out ranging to the target node, and calculates the distance set _CL respectively; (3.1.3) According to the sets _CR and _CL , determine the common coverage area _DA of the reference point; (3.1.4) obtaining spatial partition information based on the spatial analysis algorithm, and obtaining the partition set _CD that can be covered by the area _DA with the most common coverage of the signals; (3.1.5) Calculate the coverage area of each partition in the set _{CD respectively, and calculate the partition coverage rate PD ;} ( _3.1.6 ) If the partition with the largest coverage in the set CD is unique, then step (3.1.7) is adopted; otherwise, go to step (3.1.8); (3.1.7) Select the partition with the largest coverage, which is the partition where the target node is determined; ( _3.1.8 ) Calculate the centroid coordinates of the common coverage area DA and the centroid coordinates of multiple zones with maximum coverage; The specific steps of the second-stage positioning algorithm: (3.2.1) Data preprocessing definition, set the number of iterations k; (3.2.2) Preprocessing of positioning space information, partitioning the indoor space, and judging whether each area is a direct positioning area or a cooperative positioning area; (3.2.3) The first stage of coarse positioning determines the partition where the target node is located. The target node receives the positioning information of the surrounding moving reference points and fixed reference points, and performs positioning according to the TDOA positioning algorithm to determine that the target node is in step (3.2.2. ) in the direct positioning area or the cooperative positioning area; (3.2.4) If the target node after step (3.2.2) is in the direct positioning partition where the LOS environment meets the positioning conditions, the second stage positioning is performed based on the reference information of the fixed node; otherwise, the second stage is not performed. position; (3.2.5) After the target node completes the positioning, it sends its own position and positioning information to the surrounding target nodes and mobile reference; (3.2.6) For the mobile reference point information that has received the surrounding position update, the target node has two options. If the target node is in the LOS environment, it does not need to be updated; And proceed to step (3.2.5); (3.2.7) If the number of iterations of positioning in the space is greater than the specified k or the global position information does not change, the positioning is terminated. 2. a kind of indoor cooperative positioning method based on space analysis according to claim 1, is characterized in that, the standard that sampling point can receive the line-of-sight signal of fixed reference point positioning signal in described step (2.4) is: The positioning signal propagates through direct rays and reaches the sampling point with a receiving intensity greater than RSSmin.

Images