Ultra-wideband base station deployment optimization method, system, computer equipment and medium

Abstract

The present invention provides an ultra-wideband base station deployment optimization method, system, computer equipment and medium, which belongs to the field of intelligent optimization technology for wireless base station deployment in shielded spaces. The method includes: determining the initial information of base stations and tags, including gridding according to the area and shape of the deployment scene, determining the number and position of tags, the number and initial position of base stations, and generating multiple groups of initial base station deployment positions; for each group of base station deployments, respectively calculating the Cramer-Rao lower bound and weighted position precision factor of each tag; constructing a base station deployment optimization fitness function; and outputting the optimal individual as the optimal base station deployment position based on a genetic algorithm. The present invention adopts the above-mentioned ultra-wideband base station deployment optimization method, system, computer equipment and medium, which can efficiently give the optimal base station deployment position according to the actual scene or three-dimensional real scene model, while maximizing the positioning accuracy, optimizing the overall deployment position of the base station and reducing the deployment cost.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent optimization of wireless base station deployment in shielded spaces (indoors, underground, tunnels, etc.), and in particular to an ultra-wideband base station deployment optimization method, system, computer equipment and medium. Background Art

[0002] While global satellite navigation systems can meet outdoor navigation and positioning needs, achieving high-precision positioning indoors is difficult due to low signal power and susceptibility to building obstruction. Ultra-wideband (UWB) technology, with its advantages of low power consumption, resistance to multipath interference, and high precision, has become a research hotspot in the field of indoor positioning. UWB technology is primarily used in scenarios such as indoor navigation and electronic fencing, ensuring the safety of personnel and important items in environments such as underground mines. However, UWB positioning technology is a local positioning technology that requires the deployment of wireless base stations within the target scene and the provision of tags for the target. Walls and obstacles in indoor scenes can affect UWB positioning accuracy. Therefore, optimizing the base station deployment structure is crucial to maximizing positioning accuracy while reducing the number of base stations.

[0003] In UWB base station deployment research, a hypothesis-based approach is often used to evaluate base station deployment locations by analyzing metrics such as the Cramer-Rao Lower Bound (CRLB) and Position Dilution of Precision (PDOP). CRLB, as an evaluation metric for node error variance, can be used to optimize base station deployment. PDOP quantifies the impact of range error on positioning error and is an important factor in optimizing base station deployment. Based on CRLB and PDOP, researchers at home and abroad have proposed various optimization schemes and introduced intelligent optimization algorithms such as genetic algorithms (GAs), simulated annealing, and differential evolution to improve deployment efficiency and positioning accuracy.

[0004] However, CRLB and PDOP have limitations in evaluating positioning accuracy. As an evaluation metric for node error variance, CRLB may not be equivalent when the environments of different nodes vary significantly, thus affecting evaluation accuracy. PDOP calculations are based on ideal spatial signals, but in reality, factors such as multipath and non-line-of-sight propagation can cause received signal quality to differ from the calculated assumptions. Therefore, while base station deployment methods based on CRLB or PDOP can improve positioning accuracy, there is still room for improvement. Summary of the Invention

[0005] The purpose of the present invention is to provide an ultra-wideband base station deployment optimization method, system, computer equipment and medium with a simple algorithm and high intelligence. It can efficiently give the optimal base station deployment location based on the actual scene or three-dimensional real scene model. While maximizing the positioning accuracy, it can optimize the overall deployment location of the base station and reduce the deployment cost.

[0006] To achieve the above object, the present invention provides an ultra-wideband base station deployment optimization method, comprising the following steps:

[0007] Step S1: Determine the initial information of base stations and tags, including gridding according to the area and shape of the deployment scenario, determining the number and positions of tags, the number and initial positions of base stations, and generating multiple sets of initial base station deployment positions;

[0008] Step S2: for each group of base station deployments, calculate the Cramer-Rao lower bound CRLB and weighted position precision dilution WPDOP of each tag respectively;

[0009] Step S3: constructing a base station deployment optimization fitness function;

[0010] Step S4: Output the optimal individual based on the genetic algorithm as the optimal base station deployment position.

[0011] Preferably, in step S1, the process of determining the number and position of tags includes:

[0012] Calculate the scene area and determine the scene shape through the 2D CAD drawing or 3D real scene model of the deployment scene;

[0013] Based on the area and shape of the deployment scenario, select the longest line segment in the plane area to form a rectangular area and establish a local coordinate system;

[0014] The rectangular area is gridded according to the preset standard, the coordinates of the grid points are used as the coordinates of the labels and the number of labels is recorded.

[0015] Preferably, in step S1, the process of determining the number and initial positions of base stations includes:

[0016] Take the four corners of the rectangular area as the base station deployment points and calculate the distance between any two points;

[0017] Determine whether the distance is greater than a preset threshold. If so, add a new base station to be deployed at the midpoint of the line connecting the two base station deployment points until all distances meet the conditions;

[0018] The number of base station deployment points that meet the conditions is taken as the number of base stations, and their coordinates are recorded.

[0019] Preferably, in step S1, the process of determining the initial base station deployment location includes:

[0020] A set of base station coordinate matrices is used as the first individual of the initial deployment position;

[0021] Perform mutation operation on the first individual to generate other individuals to form the initial population.

[0022] Preferably, in step S2, the process of calculating the Cramer-Rao lower bound CRLB and the weighted position dilution of precision WPDOP includes:

[0023] Construct a three-sided positioning model between the base station and the tag;

[0024] The Fisher information matrix is ​​used to solve the Cramer-Rao lower bound CRLB for each label;

[0025] The weighted position precision dilution WPDOP of each tag is solved using the weighted coefficient matrix.

[0026] Preferably, in step S3, the process of constructing the base station deployment optimization fitness function includes:

[0027] Multiply the Cramer-Rao lower bound CRLB and the weighted positional precision dilution WPDOP to get the comprehensive precision dilution ;

[0028] Take the comprehensive precision factor of all tags The mean of is used as the fitness function.

[0029] Preferably, in step S4, the process of outputting the optimal individual as the optimal base station deployment position based on the genetic algorithm includes:

[0030] Calculate the fitness value of each individual in the initial population;

[0031] Determine whether the convergence condition is met. If not, perform selection, crossover and mutation operations;

[0032] Update the population and iterate the optimization until the convergence condition is met or the optimal individual is output after the maximum number of iterations.

[0033] The present invention also provides an ultra-wideband base station deployment optimization system, comprising:

[0034] A processing module is used to determine the initial information of base stations and tags, including gridding according to the area and shape of the deployment scenario, determining the number and location of tags, the number and initial location of base stations, and generating multiple sets of initial base station deployment locations;

[0035] The precision dilution calculation module is used to calculate the Cramer-Rao lower bound CRLB and weighted position precision dilution WPDOP of each tag for each group of base station deployments;

[0036] Fitness function construction module, used to construct base station deployment optimization fitness function;

[0037] The genetic algorithm execution module is used to output the optimal individual as the optimal base station deployment position based on the genetic algorithm.

[0038] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory is used to store instructions, and the processor is used to execute the instructions to implement the ultra-wideband base station deployment optimization method as described above.

[0039] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the ultra-wideband base station deployment optimization method as described above is implemented.

[0040] Therefore, the present invention adopts the above-mentioned ultra-wideband base station deployment optimization method, system, computer device and medium, and the beneficial technical effects are as follows:

[0041] (1) Accurately match the actual scene requirements and improve the efficiency of base station deployment optimization: obtain the scene area and shape through the CAD drawing or 3D model of the deployment scene, grid the plane area of ​​the scene, determine the number and location of tags, and the number of base stations, set a group of base station deployment locations based on the number of base stations and experience, and add random variables to generate multiple groups of base station deployment locations, so that the initial deployment of base stations is more scientific, the subsequent optimization workload is reduced, and the deployment cost is reduced;

[0042] (2) Improve the reliability and positioning accuracy of the deployment location: Comprehensively consider the two factors that affect positioning accuracy, CRLB and WPDOP, and construct a comprehensive precision factor , and all tags The mean value is used as a fitness function to provide a more effective evaluation criterion for the genetic algorithm, thereby finding a more reliable optimal solution and improving the positioning accuracy of the base station deployment location;

[0043] (3) The method provided by the present invention has good applicability and high intelligence. In actual scenarios, it can efficiently provide the base station deployment location with low deployment cost and high positioning accuracy.

[0044] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is an overall flow chart of an ultra-wideband base station deployment optimization method of the present invention;

[0046] Figure 2 Gridding diagram for deployment scenarios;

[0047] Figure 3 This is the genetic algorithm flow chart;

[0048] Figure 4 is the experimental scene diagram, where Figure 4 (a) is a two-dimensional deployment scene diagram. Figure 4 (b) is a three-dimensional deployment real-life picture;

[0049] Figure 5 is the base station deployment location map, where: Figure 5 (a) is a two-dimensional deployment location diagram. Figure 5 (b) is a three-dimensional deployment location diagram;

[0050] Figure 6 This is the distribution diagram after base station deployment optimization, where: Figure 6 (a) is the optimized PDOP distribution diagram. Figure 6 (b) is the optimized CRLB distribution diagram. Figure 6 (c) in the figure is the optimized distribution map;

[0051] Figure 7 is the measurement result diagram;

[0052] Figure 8 is the error distribution diagram, where Figure 8 (a) in the figure is the standard deviation distribution diagram. Figure 8 (b) is the mean absolute error distribution diagram, Figure 8 (c) in the figure is the root mean square error distribution diagram;

[0053] Figure 9 It is the three-dimensional deployment position positioning result map, where: Figure 9 (a) in the figure is the result of positioning point 1. Figure 9 (b) is the result of positioning point 2. Figure 9 (c) in the figure is the positioning result of point 3. Figure 9 (d) is the positioning result of point 4;

[0054] Figure 10 is the three-axis error diagram, where Figure 10 (a) is the X-axis error diagram, Figure 10 (b) is the Y-axis error diagram. Figure 10 (c) in the figure is the Z-axis error diagram;

[0055] Figure 11 For the theoretical PDOP, CRLB, Distribution and error distribution diagram of corresponding deployment positions, where Figure 11 (a) is the theoretical PDOP distribution diagram, Figure 11 (b) is the error distribution diagram based on the PDOP scheme. Figure 11 (c) in the figure is the theoretical CRLB distribution diagram. Figure 11 (d) in the figure is the error distribution diagram based on the CRLB scheme. Figure 11 The (e) in the equation is theoretical. Distribution map, Figure 11 (f) is based on Error distribution diagram of the scheme. DETAILED DESCRIPTION

[0056] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0057] Unless otherwise defined, technical or scientific terms used in the present invention shall have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs.

[0058] Example 1

[0059] Figure 1 The overall flow chart of an ultra-wideband base station deployment optimization method of the present invention includes the following steps:

[0060] Step S1, determine the initial information of base stations and tags: grid the planar area of ​​the deployment scene according to the area and shape of the deployment scene, determine the number and location of tags, the number and initial location of base stations, set one group of base station deployment locations based on the number of base stations and experience, and add random variables to generate multiple groups of base station deployment locations. Each group of base station deployment locations is used as an individual in the initial population.

[0061] The process of determining the number and location of tags is as follows:

[0062] Calculate the scene area and determine the scene shape by deploying a 2D CAD drawing or a 3D real-life model of the scene;

[0063] According to the area and shape of the deployment scenario, select the longest line segment in its plane area, the length of which is m, draw a perpendicular line through the longest point of the line segment, the two end points of the perpendicular line do not exceed the boundary of the plane area, and the length is m, forming m rectangular area and establish a local coordinate system;

[0064] like Figure 2 As shown, according to The standard of m is used to grid the rectangular area, and the grid is grid area, then there is grid points, each grid point is regarded as a sample point, the coordinates of the sample point are used as the coordinates of the label, and the number of labels is recorded as .

[0065] The process of determining the number of base stations and their initial locations is as follows:

[0066] Take the four corners of the rectangular area as the base station deployment points and calculate the distance between any two points .

[0067] First, judge Is it greater than the preset threshold? If it is greater than, then add a new base station to be deployed at the midpoint of the line connecting the two base station deployment points (only one is added at the same location). In this case, the threshold Then, recalculate the distance between any two base station deployment points. If there is still a distance greater than the threshold, If the above operation is repeated, the above operation is repeated until the distances between all base stations to be deployed meet the requirements. Finally, the total number of base station deployment points that meet the conditions is taken as the number of base stations, recorded as and use its coordinates as the initial position of the base station.

[0068] The process of determining the initial base station deployment location is as follows:

[0069] The deployment location of a group of base stations is reflected as follows Base station coordinate matrix, that is, , as the first set of scenarios in the initial deployment location , that is, the first individual in the initial population; among them, For the first group of solutions The horizontal coordinate value of each base station, For the first group of solutions The vertical coordinate value of each base station, For the first group of solutions The vertical coordinate value of the base station.

[0070] The individual Perform mutation operation to generate O-1 individuals and form a population containing O individuals: Let the range of random numbers be , Based on the minimum base station spacing The positive number defined satisfies , which is the ratio of the minimum spacing to the number of base stations rounded up. The three-dimensional coordinates of each base station are added with a random value to obtain the second set of solutions in the initial deployment position , that is, the second individual in the initial population; repeat the previous step until the Oth group of solutions is generated After the end, the initial deployment position is formed , which is the initial population. The calculation formula for each individual is shown in formula (1):

[0071] (1);

[0072] Where, It is a random number uniformly distributed in the interval [-1, 1].

[0073] Step S2: For each group of base station deployments, the Cramer-Rao lower bound CRLB and weighted position precision dilution WPDOP of each tag are calculated respectively, including:

[0074] Construct a three-sided positioning model between the base station and the tag;

[0075] The Fisher information matrix is ​​used to solve the Cramer-Rao lower bound CRLB for each label;

[0076] The weighted position precision dilution WPDOP of each tag is solved using the weighted coefficient matrix.

[0077] Specifically:

[0078] Step S21, 1 group of solutions, that is, the vectors composed of the CRLB values ​​and WPDOP values ​​of S tags to M base stations under 1 individual and , forming a sequence and ,in and Respectively The CRLB value and WPDOP value of each label, Indicates transpose.

[0079] Step S22: Using the first set of solutions in the initial base station deployment For example, calculate the Tags CRLB and WPDOP, that is and .in The coordinate sequence of , No. Base stations The coordinates are ,Label The location is , calculate the The estimated distance between the base station and the tag As shown in formula (2):

[0080] (2);

[0081] Where, and Represent the estimated distance and theoretical distance between the base station and the tag, respectively. Represents the ranging error, assuming that the mean is 0 and the variance is The standard normal distribution of .

[0082] Step S23, CRLB is used to describe the theoretical lower limit of the variance of the tag estimation position. Cramer-Rao lower bounds for parameter estimates are given.

[0083] Step S24, Fisher Information Matrix is a The matrix of is expressed as follows:

[0084] (3);

[0085] Step S25: Get the label Cramero lower bound , the expression is shown in formula (4):

[0086] (4);

[0087] Where, Indicates taking the matrix traces.

[0088] Step S26, WPDOP describes the combined effect of the weighted estimated distance on the uncertainty of the position estimate, which is the covariance matrix Quantitative indicators, The expression of is shown in formula (5):

[0089] (5);

[0090] Where, 、 Represents the estimated distance Vector, theoretical position The vectors are all of shape , is the weight matrix.

[0091] Weight Matrix is a diagonal matrix whose elements As shown in formula (6):

[0092] (6);

[0093] Step S27: Get the label Weighted position dilution of precision , the expression is shown in formula (7):

[0094] (7);

[0095] Where, is the weighted coefficient matrix of the estimation error, is the offset coefficient matrix, that is, the Jacobian matrix, specifically: .

[0096] Step 3: Construct the base station deployment optimization fitness function: Multiply the CRLB and WPDOP obtained above to obtain the comprehensive precision factor of each tag. ,The mean of the comprehensive precision factors of all tags is used as the fitness function to reflect the pros and cons of the base station deployment.

[0097] Step S31: Construct a comprehensive precision factor by multiplying CRLB and WPDOP , which is the trade-off index of the system's comprehensive positioning capability, is expressed as shown in formula (8):

[0098] (8);

[0099] Where, For the The comprehensive precision coefficient of the tags.

[0100] Step S32: constructing the fitness function for base station deployment optimization: , the expression is shown in formula (9):

[0101] (9);

[0102] Where, Represents vector The mean of each element, For experienced After the iteration, the Individual, that is, experienced The first base station deployment position after the iteration Group plan.

[0103] Step 4: Figure 3 As described above, the optimal base station deployment location is solved based on the genetic algorithm: the fitness value of each individual in the population is calculated, individuals with smaller fitness values ​​are selected for crossover and mutation operations, the population is updated and iterative optimization is performed until the convergence condition is met or the maximum number of iterations is reached, and the optimal individual, that is, the optimal base station deployment location, is output.

[0104] Step S41: Based on the fitness function , calculate the initial population The fitness value of each individual in (the smaller the fitness value, the higher the fitness); the individual corresponding to the minimum fitness value is selected as the current optimal individual, that is, the optimal solution in the initial base station deployment position , let the global optimal individual , the expression is shown in formula (10):

[0105] (10);

[0106] Step S42: Determine Whether the convergence condition can be met, if so, exit the loop and output , otherwise enter the selection operation, convergence condition The calculation formula is shown in formula (11):

[0107] (11);

[0108] Step S43: Use roulette to select individuals with smaller fitness values ​​to enter the next generation, and calculate the selection probability of each individual , as shown in formula (12):

[0109] (12);

[0110] in, Indicates the number of the initial population The fitness function of each individual; An index representing the number of individuals in a population;

[0111] Step S44, then according to the selection probability Perform random sampling and select a certain number of individuals to enter the crossover operation.

[0112] Set intersection ,in is the product of the number of base stations and the coordinate dimension. and individuals The crossover of the two individuals constitutes the sequence and , forming a new individual , the sequence is , the calculation formula of the cross process is shown in formula (13):

[0113] (13);

[0114] Where, , . 、 、 and are new individuals The horizontal coordinate, vertical coordinate and vertical coordinate of the first base station in ; 、 、 and 、 、 Individual and individuals The horizontal coordinate, vertical coordinate and vertical coordinate of the first base station in .

[0115] Step S45: For new individuals , with a certain mutation probability Randomly change some of the genes of an individual. No. Elements The variation formula of is shown in formula (14):

[0116] (14);

[0117] Where, Represents the mutated element, Represents a random number, determined according to the mutation strategy.

[0118] Step S46: Add the generated new individuals to the population and eliminate individuals with larger fitness values ​​to complete the first iteration and form a new population, i.e., the new base station deployment location. ,in, For the new population individuals, i.e. the first individual at the new base station deployment location Group plan.

[0119] Step S47: Repeat step S41 to calculate the individual corresponding to the minimum fitness value of the new population ,like ,renew .

[0120] Step S48, repeat step S42, if the convergence condition is met, exit the loop output, if not, repeat steps S43-S46 to iteratively update the population until the population reaches the convergence condition or the number of iterations reaches the maximum number of iterations After the loop ends, output .

[0121] The present invention will be further described below through specific examples.

[0122] In order to verify the positioning performance of the optimal deployment location, unobstructed small indoor scenes and covered medium and large indoor scenes were selected as experimental scenes for 2D and 3D base station deployment, such as Figure 4 As shown, Figure 4 The scene in (a) has a solid wall on the top, two display boards on the left, a statue in the middle, and a solid wall on the right. Figure 4 The right side of (b) is Figure 4 In (a), the left side is empty, and the middle part is a display board and a statue, which act as a shield. The deployment location is as follows Figure 5 As shown, Figure 5 (a) is a two-dimensional deployment location diagram. Figure 5 (b) is a three-dimensional deployment location diagram, where the inverted triangle points, regular triangle points and diamond points represent the use of PDOP, CRLB and The specific coordinate data of the deployment location are shown in Table 1 and Table 2.

[0123] Table 1 Base station coordinates of two-dimensional deployment locations

[0124] ;

[0125] Table 2 Base station coordinates of three-dimensional deployment locations

[0126] ;

[0127] The simulation results show that the PDOP, CRLB, Distribution, such as Figure 6 As shown. Figure 6 (a) and Figure 6 As can be seen in (b), PDOP and CRLB are both minimum at the lower left corner, and gradually increase from the lower left corner to the upper right corner. Especially after exceeding the area surrounded by the base station, the increase is greater and reaches the maximum in the upper right corner. In addition, Figure 6 The contour lines in the left area of ​​(b) are sparse, which is caused by the uneven distribution of base stations. Figure 6 As can be seen in (c), It reaches its maximum value around the area and gradually decreases from the periphery to the center. The overall change trend is uniform, and the contour distribution is relatively symmetrical. The indicator distribution of the deployment location is ideal.

[0128] The experimental results were measured, and the static positioning results of the three deployment positions in the two-dimensional scene are as follows: Figure 7 As shown, the inverted triangle point, regular triangle point and diamond point represent the points based on PDOP, CRLB and The measurement results of the deployment position, the circular point is the theoretical position. It can be seen that the inverted triangle point is the most dispersed, with the maximum difference between the horizontal and vertical coordinates being 0.12 m and 0.36 m respectively, and it is the farthest away from the black circular point. The static positioning results of the deployment position based on PDOP are the most discrete and have the largest error. Both the regular triangle point and the diamond point are closer to the circular point, but compared with the diamond point, the regular triangle point is more dispersed, with the maximum difference between the horizontal and vertical coordinates being 0.15 m and 0.17 m respectively. The measurement results are most concentrated at the deployment location.

[0129] Calculate the standard deviation (STD), mean absolute error (MAE), and root mean square error (RMSE) of the three deployment locations. Figure 8 As shown, Figure 8 (a) in the figure is the standard deviation distribution diagram. Figure 8 (b) is the mean absolute error distribution diagram, Figure 8 (c) is the root mean square error distribution diagram, the horizontal axis is the sampling point, the vertical axis corresponds to three types of errors, the positive triangle point line, the inverted triangle point line and the diamond point line represent the three types of errors based on CRLB, PDOP and Error distribution of deployment position. It can be seen that the maximum STD of the inverted triangle point line and the regular triangle point line exceeds 0.1 m, while the STD of the diamond point line is no more than 0.1 m. The measurement results of the deployment position show the highest stability. In addition, the MAE and RMSE of the inverted triangle point line are not less than 0.2 m, except for one point where the MAE exceeds 0.2 m when x is greater than 0 and less than 50. The MAE and RMSE of the other positions of the regular triangle point line are less than 0.2 m, while the MAE and RMSE of the diamond point line are not greater than 0.1 m. Therefore, based on The positioning accuracy of the deployment location based on PDOP is the highest. The positioning results of the three are shown in Table 3. It can be seen that the positioning location based on PDOP has the largest deviation in the X axis, reaching 0.21 m. Although the positioning error in the Y axis does not exceed 0.1 m, the final error still exceeds 0.2 m. In addition, the positioning errors of the X and Y axes of the deployment location based on CRLB are both 0.04 m, slightly higher than that based on CRLB. Compared with the deployment locations based on CRLB and PDOP, the deployment locations based on The maximum error, STD, MAE, and RMSE of the deployment location were reduced by 52.0%, 30.0%, 30.0%, 25.8% and 71.8%, 30.8%, 80.7%, and 82.3%, respectively.

[0130] Table 3 Positioning results

[0131] ;

[0132] The static positioning results in the three-dimensional scene are as follows Figure 9 As shown, Figure 9 (a) in the figure is the result of positioning point 1. Figure 9 (b) is the result of positioning point 2. Figure 9 (c) in the figure is the positioning result of point 3. Figure 9(d) is the result of the fourth point positioning. The inverted triangle point, regular triangle point and diamond point represent the positioning results based on PDOP, CRLB and The measurement results of the deployment position are shown in Figure 1. The circular point is the theoretical position. It can be seen that on the Z axis, the difference between the diamond point and the circular point is no more than 0.2 m, while the difference between the inverted triangle point and the regular triangle point and the circular point is no less than 0.2 m; on the X and Y axes, the diamond point is closer to the circular point than the inverted triangle point and the regular triangle point; compared with the deployment position based on CRLB and PDOP, the deployment position based on The deployment location has higher accuracy.

[0133] Calculate the X, Y and Z axis errors for the three deployment positions, such as Figure 10 As shown, Figure 10 (a) is the X-axis error diagram, Figure 10 (b) is the Y-axis error diagram. Figure 10 (c) is the Z-axis error diagram, the horizontal axis is the point number, the vertical axis is the error of the X, Y and Z axes respectively, the positive triangle point, inverted triangle point and diamond point represent the error based on CRLB, PDOP and It can be seen that the X, Y and Z axis errors of the diamond point are no more than 0.1 m, 0.2 m and 0.2 m respectively, while the maximum errors of the positive triangle point and the inverted triangle point are no less than 0.2 m, 0.4 m, 0.6 m and 0.6 m, 1.0 m and 0.8 m respectively, and the maximum and minimum errors are no less than 0.1 m, 0.2 m, 0.4 m and 0.4 m, 0.8 m and 0.4 m respectively. Therefore, compared with the deployment positions based on CRLB and PDOP, the deployment positions based on The deployment position has higher positioning accuracy and stability in three axes. Statistics based on PDOP, CRLB and The maximum error, STD, MAE and RMSE of the scheme are shown in Table 4. It can be seen that based on The maximum error, STD, MAE and RMSE of the scheme are 0.42 m, 2.53 m, 0.28 m and 0.18 m respectively, which are 59.6%, 6.6%, 65.9%, 67.3% lower than those of the CRLB and PDOP schemes and 80.0%, 11.2%, 82.2% and 82.4% lower than those of the CRLB and PDOP schemes respectively. The deployment positions of the two are higher than those based on PDOP and CRLB.

[0134] Table 4 Based on PDOP, CRLB and Maximum error, STD, MAE and RMSE of the solution

[0135] ;

[0136] Dynamic experiments were conducted using three deployment locations, and the obtained PDOP, CRLB, The distribution and the error distribution of the corresponding deployment position are as follows: Figure 11 As shown in the figure, the black lines represent the equipotential lines of the values. As the values ​​increase, the color of the gradient area changes from dark to light. The normalized product of PDOP and CRLB is obtained. The values ​​are magnified proportionally for the sake of easy observation. The error distribution of the three schemes uses the mean absolute error as the error indicator. It can be seen that compared with the error distribution of PDOP, CRLB distribution and the corresponding deployment position, Distribution and based on The error distribution difference of the deployment position is the smallest, and based on The maximum error of the deployment position is the smallest.

[0137] In summary, whether it is the distribution relationship or the positioning accuracy, based on The deployment locations of represent the best performance and are more suitable for base station deployment optimization.

[0138] Example 2

[0139] An ultra-wideband base station deployment optimization system, comprising:

[0140] A processing module is used to determine the initial information of base stations and tags, including gridding according to the area and shape of the deployment scenario, determining the number and location of tags, the number and initial location of base stations, and generating multiple sets of initial base station deployment locations;

[0141] The precision dilution calculation module is used to calculate the Cramer-Rao lower bound CRLB and weighted position precision dilution WPDOP of each tag for each group of base station deployments;

[0142] Fitness function construction module, used to construct base station deployment optimization fitness function;

[0143] The genetic algorithm execution module is used to output the optimal individual as the optimal base station deployment position based on the genetic algorithm.

[0144] If the above functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0145] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0146] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, and then editing, interpreting, or processing in another suitable manner as necessary, and then storing it in a computer memory.

[0147] Therefore, the present invention adopts the above-mentioned ultra-wideband base station deployment optimization method, system, computer equipment and medium, which has a simple algorithm and a high degree of intelligence. It can efficiently give the optimal base station deployment position based on the actual scene or three-dimensional real scene model. While maximizing the positioning accuracy, it can optimize the overall deployment position of the base station and reduce the deployment cost.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for optimizing ultra-wideband base station deployment, characterized in that: The following steps are involved: Step S1: Determine the initial information of base stations and tags, including gridding according to the area and shape of the deployment scenario, determining the number and positions of tags, the number and initial positions of base stations, and generating multiple sets of initial base station deployment positions; Step S2: for each group of base station deployments, calculate the Cramer-Rao lower bound CRLB and weighted position precision dilution WPDOP of each tag respectively; Step S3: constructing a base station deployment optimization fitness function; Step S4: outputting the optimal individual as the optimal base station deployment position based on the genetic algorithm; In step S1, the process of determining the initial base station deployment location includes: A set of base station coordinate matrices is used as the first individual of the initial deployment position; Perform mutation operation on the first individual to generate other individuals to form the initial population; In step S2, the process of calculating the Cramer-Rao lower bound CRLB and the weighted position dilution of precision WPDOP includes: Construct a three-sided positioning model between the base station and the tag; The Fisher information matrix is ​​used to solve the Cramer-Rao lower bound CRLB for each label; The weighted position precision factor WPDOP of each tag is solved using the weighted coefficient matrix; In step S3, the process of constructing the base station deployment optimization fitness function includes: Multiply the Cramer-Rao lower bound CRLB and the weighted positional precision dilution WPDOP to get the comprehensive precision dilution ; Take the comprehensive precision factor of all tags The mean of is used as the fitness function.

2. The method for optimizing ultra-wideband base station deployment according to claim 1, wherein: In step S1, the process of determining the number and location of tags includes: Calculate the scene area and determine the scene shape through the 2D CAD drawing or 3D real scene model of the deployment scene; Based on the area and shape of the deployment scenario, select the longest line segment in the plane area to form a rectangular area and establish a local coordinate system; The rectangular area is gridded according to the preset standard, the coordinates of the grid points are used as the coordinates of the labels and the number of labels is recorded.

3. The method for optimizing ultra-wideband base station deployment according to claim 2, wherein: In step S1, the process of determining the number and initial positions of base stations includes: Take the four corners of the rectangular area as the base station deployment points and calculate the distance between any two points; Determine whether the distance is greater than a preset threshold. If so, add a new base station to be deployed at the midpoint of the line connecting the two base station deployment points until all distances meet the conditions; The number of base station deployment points that meet the conditions is taken as the number of base stations, and their coordinates are recorded.

4. The method for optimizing ultra-wideband base station deployment according to claim 1, wherein: In step S4, the process of outputting the optimal individual as the optimal base station deployment location based on the genetic algorithm includes: Calculate the fitness value of each individual in the initial population; Determine whether the convergence condition is met. If not, perform selection, crossover and mutation operations; Update the population and iterate the optimization until the convergence condition is met or the optimal individual is output after the maximum number of iterations.

5. An ultra-wideband base station deployment optimization system, characterized in that: The method for optimizing the deployment of an ultra-wideband base station according to any one of claims 1 to 4 comprises: A processing module is used to determine the initial information of base stations and tags, including gridding according to the area and shape of the deployment scenario, determining the number and location of tags, the number and initial location of base stations, and generating multiple sets of initial base station deployment locations; The precision dilution calculation module is used to calculate the Cramer-Rao lower bound CRLB and weighted position precision dilution WPDOP of each tag for each group of base station deployments; Fitness function construction module, used to construct base station deployment optimization fitness function; The genetic algorithm execution module is used to output the optimal individual as the optimal base station deployment position based on the genetic algorithm.

6. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory is used to store instructions, and the processor is used to execute the instructions to implement the ultra-wideband base station deployment optimization method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the ultra-wideband base station deployment optimization method according to any one of claims 1 to 4 is implemented.

Images