WO2023083041A1 - 一种定位方法、装置及存储介质 - Google Patents

一种定位方法、装置及存储介质 Download PDF

Info

Publication number
WO2023083041A1
WO2023083041A1 PCT/CN2022/128767 CN2022128767W WO2023083041A1 WO 2023083041 A1 WO2023083041 A1 WO 2023083041A1 CN 2022128767 W CN2022128767 W CN 2022128767W WO 2023083041 A1 WO2023083041 A1 WO 2023083041A1
Authority
WO
WIPO (PCT)
Prior art keywords
base station
equations
distance
signal
gradient descent
Prior art date
Application number
PCT/CN2022/128767
Other languages
English (en)
French (fr)
Inventor
胡兆兴
Original Assignee
中移(上海)信息通信科技有限公司
中移智行网络科技有限公司
中国移动通信集团有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中移(上海)信息通信科技有限公司, 中移智行网络科技有限公司, 中国移动通信集团有限公司 filed Critical 中移(上海)信息通信科技有限公司
Publication of WO2023083041A1 publication Critical patent/WO2023083041A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S5/00Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
    • G01S5/02Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves

Definitions

  • the present disclosure relates to the technical field of communication, and in particular to a positioning method, device and storage medium.
  • Neighbor method The simplest way is to directly select the location of the AP (Access Point, Access Point) with the highest signal strength, and the positioning result is the location of the currently connected AP stored in the hotspot location database;
  • the nearest neighbor method is simple and fast to implement, but the positioning accuracy cannot be guaranteed, and it depends on the deployment density of beacons, which has high cost and poor accuracy.
  • Trilateration (angle) measurement method The distance or angle between the target and the AP is obtained through various parameters of the signal, and the position is calculated geometrically. Including time-of-arrival method, relative time-of-arrival method, angle-of-arrival method, ranging method based on signal strength, and its hybrid algorithm.
  • Fingerprint method collect and model the signal features at each position in advance, and store them in the fingerprint database. During real-time positioning, match the current signal features with the data in the fingerprint library to determine the target location.
  • the traditional fingerprint positioning method uses the collected signal characteristic data to establish a signal strength database, and uses the Euclidean distance model or probability model to calculate the user matching degree in real time. This method is unscientific for the processing of signal fluctuations in the process of fingerprint collection.
  • the positioning accuracy of the fingerprint positioning method has certain limitations and cannot meet the needs of high-precision positioning. At the same time, the workload of fingerprint collection is huge, which seriously affects the efficiency and feasibility of indoor positioning system construction. .
  • the disadvantage of the related technology is that it is difficult to take into account the convenience of calculation and the high positioning accuracy.
  • Embodiments of the present disclosure provide a positioning method, device, and storage medium, so as to make calculation in positioning technology more convenient and positioning accuracy higher.
  • An embodiment of the present disclosure provides a positioning method, and the method includes:
  • the base station identifier and signal characteristic data After acquiring the base station identifier and signal characteristic data, associate the user equipment UE, the base station and the signal characteristic data, wherein the base station identifier and the signal characteristic data are base station identifiers received by the UE at the same time and signal feature data; data modeling is performed according to the base station location and the signal feature data associated with the UE, wherein the data modeling includes calculating the distance between the UE and the base station, and calculating the distance between the UE and the base station In the case of the distance between the UE and the base station, establish a set of equations for obtaining the position of the UE and the signal strength; in the case of using the triangulation method to determine the position of the UE, the gradient descent method is used, and the The minimum value of the function determined by the equation set is used as the position of the UE.
  • establishing a system of equations for calculating the position of the UE and the signal strength is based on a signal path loss propagation model.
  • a gradient descent method is used to determine the position of the UE through a set of equations, including: determining the position of the UE using the triangulation method
  • use the signal path loss propagation model to establish the equations of the four base stations that each UE is connected to at the same time; for the equations of the four base stations, three equations are obtained after subtracting each other in turn; according to The three equations calculate the position of the UE.
  • the minimum value of the function determined through the equation system is used as the position of the UE, the minimum value of the following formula is calculated:
  • the iterative formula of the following formula is used:
  • v n+1 ⁇ v n +(1- ⁇ )f'(x n - ⁇ v n )
  • x n is the current iteration position
  • x n+1 is the next iteration position
  • is the descending step size
  • is the learning rate, set to 0.9
  • 0 is 0
  • f'(x n ) is the derivative value of the function at x n .
  • using the gradient descent method using the minimum value of the function determined by the equation system as the position of the UE, includes: selecting initial points x, y; calculating partial derivatives of x, y Function; use the gradient descent method to calculate the position of the next point; perform iterative calculation and set the exit condition to judge whether the distance between the two points after calculation meets the exit condition; if the exit condition is met, the distance between the two points has converged, and the iteration is exited. Otherwise, repeat until the result is satisfied and jump out of the iteration; the position that satisfies the condition is the final position of the UE.
  • the step length to 1 meter
  • the method further includes: when acquiring base stations received by the UE at the same time, using a local outlier factor algorithm to eliminate discrete base stations.
  • An embodiment of the present disclosure provides a positioning device, including: a processor, configured to read the program in the memory, and perform the following process: after obtaining the base station identifier and signal characteristic data, link the user equipment UE, the base station, and the signal The characteristic data is associated, wherein, the base station identifier and the signal characteristic data are the base station identifier and signal characteristic data received by the UE at the same time; data is performed according to the base station location and the signal characteristic data associated with the UE Modeling, wherein the data modeling includes calculating the distance between the UE and the base station, and in the case of calculating the distance between the UE and the base station, establishing and obtaining the UE position and signal A set of equations of strength; in the case of using the triangulation method to determine the position of the UE, a gradient descent method is used, and the minimum value of the function determined through the set of equations is used as the position of the UE; the transceiver, Used to receive and send data under the control of the processor.
  • An embodiment of the present disclosure provides a positioning device, including: a data module, configured to associate a user equipment UE, a base station, and the signal characteristic data after obtaining the base station identifier and signal characteristic data, wherein the base station identifier And the signal characteristic data is the base station identification and signal characteristic data received by the UE at the same time; the modeling module is used to perform data modeling according to the base station location and the signal characteristic data associated with the UE, wherein, The data modeling includes calculating the distance between the UE and the base station, and in the case of calculating the distance between the UE and the base station, establishing an equation for obtaining the position of the UE and the signal strength
  • a positioning module configured to use a gradient descent method to use a minimum value of a function determined through a set of equations as the position of the UE when using a triangulation method to determine the position of the UE.
  • An embodiment of the present disclosure provides a computer program, including computer readable code.
  • a processor in the electronic device executes the program to implement the above one or more embodiments.
  • the targeting method in .
  • the applicability is good, using the gradient descent method to calculate the user position, compared with the traditional least square method, saves a lot of matrix operations, and at the same time, it can be applied to the solution of high-order non-linear equations, and can use multiple Various forms of equation solving and user position calculation.
  • High precision establish equations based on user signal strength, and realize user location calculation.
  • the fingerprint positioning method it is not limited to locating to the collected feature points, making user location calculation more accurate and positioning accuracy high.
  • FIG. 1 is a schematic diagram of an implementation flow of a positioning method in an embodiment of the present disclosure
  • FIG. 2 is a schematic diagram of the composition and structure of a positioning device in an embodiment of the present disclosure
  • FIG. 3 is a schematic diagram of the composition and structure of another positioning device in an embodiment of the present disclosure.
  • the number of base stations is screened through local factors and abnormal factors, the data with excellent data network type and high quality is selected for the next step of calculation, and the distance between the user and the base station is established through the differential calculation of the wireless signal propagation model
  • the solution of the equations is transformed into the problem of finding the minimum value of the function, and the gradient descent method is used to complete the calculation of the user's position, which has the advantages of convenient calculation and high precision.
  • Figure 1 is a schematic diagram of the implementation process of the positioning method, as shown in the figure, which may include:
  • Step 101 After obtaining the base station identifier and signal characteristic data, associate the user equipment UE, the base station and the signal characteristic data, wherein the base station identifier and the signal characteristic data are received by the UE at the same time The base station identification and signal characteristic data;
  • Step 102 perform data modeling according to the base station location and the signal characteristic data associated with the UE, wherein the data modeling includes calculating the distance between the UE and the base station, and calculating the UE In the case of the distance to the base station, establish a set of equations for obtaining UE position and signal strength;
  • the signal feature data may include the signal strength.
  • Step 103 In the case of using the triangulation method to determine the position of the UE, a gradient descent method is used, and the minimum value of the function determined through the equation set is used as the position of the UE.
  • LOF local outlier factor algorithm, Local Outlier Factor
  • the local anomaly factor algorithm is used to optimize and screen the network type, and eliminate The calculation error of the discrete base station for the user's position; and according to the wireless signal propagation model, the correlation between the user's position and the signal strength is established, that is, the equations of the user's position and the signal strength are established.
  • the improved gradient descent method is used to solve the equations, which can reduce the difficulty of calculating complex equations and improve the accuracy and efficiency of user position calculations.
  • Adopting the technical solution provided by this disclosure greatly saves the efficiency and feasibility of triangulation positioning system construction, can be applied to the solution of high-order non-linearizable equations, and can use various forms of equations to solve and calculate the user's position. It makes the calculation of the user's position more accurate and the positioning accuracy is high.
  • An embodiment of the present application provides a positioning method, and the method may include three steps of base station data screening, data modeling, and user position calculation.
  • the title of user will also be used. This is because some people in the industry are used to using users to describe solutions, so two descriptions are used in the embodiment at the same time.
  • the meaning of using user description and using UE are the same, in order to make those skilled in the art understand the implementation of this scheme better and more easily.
  • Step S22 base station data screening.
  • step S22 may include: when acquiring base stations received by the UE at the same time, using a local outlier factor algorithm to eliminate discrete base stations.
  • the base station identification and RSRP (Reference Signal Received Power, Reference Signal Receiving Power) data that the user can receive at the same time can be obtained, and according to the base station identification, combined with the base station address table, the user can be All base station data statistics associated at this moment form base station site distribution data, utilize LOF (local outlier factor algorithm) algorithm to carry out discrete base station screening for above-mentioned data, the processing mode of described step S22 can comprise the following steps:
  • Step S221 Treat all base station data as a set C of data points.
  • Step S222 Sequentially select two points p and o from the set, and calculate the distance d(p, o) between the two points.
  • Step S223 Calculate the K-th distance of point p, that is, the distance of the K-th farthest point from p, excluding p, denoted as d K (p), the calculation method can be shown in the following formula (1):
  • Step S224 Calculate the K-th neighborhood of point p, denoted as N K (p), that is, all points within the k-th distance of p, including the k-th distance. At the same time, calculate the number of points in the K neighborhood of point p, denoted as
  • Step S225 Calculate the reachable distance between each point in the set and point p, which is recorded as: reach_d K (p,o), and the calculation method can be shown in the following formula (2):
  • the k-th reachable distance from point o to point p at least the k-th distance of o, or the real distance between o and p, it should be noted that the reachable distance has directionality, reach_d K (p,o) and reach_d K (o,p) are not equal.
  • Step S226 Calculate the local reachable density of point p, denoted as lrd K (p), the calculation formula can be shown in the following formula (3):
  • the meaning of the above formula (3) is to represent the reciprocal of the average reachable distance from a point to p in the kth neighborhood of point p.
  • Step S227 Calculate the local outlier factor of point p, denoted as LOF K (p), the calculation formula can be shown in the following formula (4):
  • the above formula is the average of the ratio of the local reachable density of the neighborhood point N K (p) of point p to the local reachable density of point p.
  • the ratio When the ratio is closer to 1, it shows that the density of the neighborhood point of p is Almost, p may belong to the same cluster as the neighborhood; when the ratio is smaller than 1, it means that the density of p is higher than the density of its neighbor points, and p is a dense point; when the ratio is greater than 1, it means The density of p is smaller than the density of its neighbor points, the more likely p is an outlier.
  • Step S228 Calculate the local outlier factors of the K neighborhood of all points, and judge whether the current point is a discrete point according to preset conditions.
  • the K value of the K neighborhood needs to be clarified, and the threshold of the local discrete factor needs to be clarified. Therefore, the K value scheme is specially defined as shown in the following formula (5):
  • K is the number of points minus 1;
  • the threshold LOFT (C) of the local dispersion factor can be selected as 1.8 after experimental analysis, that is, when the local dispersion factor of a point is greater than 1.8, it is judged as an outlier point.
  • Step S24 data modeling.
  • establishing a set of equations for obtaining the position of the UE and the signal strength is based on a signal path loss propagation model of.
  • PL indicates the signal strength path loss (transmitting strength - receiving strength, generally the same location base station transmits strength is fixed, the receiving strength is different due to distance);
  • f is the frequency, the unit is MHZ (megahertz);
  • K is the path loss coefficient;
  • d represents The distance between the user and the base station;
  • L f is the floor penetration loss coefficient;
  • x ⁇ is the slow fading margin.
  • step 103 of "determining the position of the UE by using a gradient descent method and using a set of equations" includes:
  • Step S131 In the case of using the triangulation method to determine the position of the UE, use the signal path loss propagation model to establish equations for four base stations that each UE is connected to at the same time;
  • Step S132 For the equations of the four base stations, three equations are obtained after sequentially subtracting two by two;
  • Step S133 Calculate the location of the UE according to the three equations.
  • the minimum value of the function determined by the equation set in step 103 is used as the position of the UE, the minimum value of the following formula (7) is obtained:
  • the base station data After the base station data is screened, data modeling needs to be carried out according to the base station location and the signal characteristic data associated with the user.
  • the signal path loss propagation model the distance between the user and the base station is calculated, and at least three equations are established to obtain the user location.
  • This algorithm needs to calculate the distance based on the signal path loss, so as to infer the user's location, and is suitable for open scenes, such as exhibition halls, airports, etc.
  • triangulation can calculate the user's position according to formula (15).
  • the calculation method can be to shift the right side of the three equations in formula (15) to the left, so that the right side is equal to 0; and formula (15)
  • the three equations in find the absolute value and add up to obtain the equation in the following formula (16):
  • Step S26 The user solves the problem.
  • the iterative formula of the following formula (17) is used:
  • x n+1 is the position of the next iteration
  • is the descending step size
  • is the learning rate, which can generally be set to 0.9;
  • v n the amount of decrease to increase momentum, v 0 is 0;
  • f'(x n ) is the derivative value of the function at x n .
  • the minimum value point of the function can be calculated according to the f(X, Y) function formula.
  • the improved gradient descent method is used to obtain the user position.
  • the basic principle of the method is:
  • is the descending step size.
  • f'(x n ) is the derivative value of the function at x n .
  • the fixed step size has poor applicability to functions, slow convergence speed, and may also cause interval oscillations. Therefore, the improved gradient descent method will be used in the embodiment.
  • the gradient of the cost function at the next position is fully considered, and the step size is dynamically adjusted, which can effectively reduce the impact of oscillation convergence and increase the convergence speed. As shown in formula (17).
  • step 103 "using the gradient descent method, using the minimum value of the function determined by the equation system as the position of the UE" includes: selecting initial points x, y; calculating x , the partial derivative function of y; use the gradient descent method to calculate the position of the next point; perform iterative calculation, and set the jump-out condition to judge whether the distance between the two points after calculation satisfies the jump-out condition; if the jump-out condition is met, the distance between the two points has been Converge, jump out of the iteration, otherwise repeat until the result is satisfied and converge and jump out of the iteration; the position that satisfies the condition is the final position of the UE.
  • the user location calculation is completed.
  • the embodiment of the present disclosure also provides a positioning device and a computer-readable storage medium, which are similar to the positioning method, so the implementation of these devices can refer to the implementation of the method, and the repetition will not be repeated.
  • Figure 2 is a schematic structural diagram of the positioning device, as shown in the figure, the device includes:
  • the processor 200 is configured to read the program in the memory 220 and execute the following processes:
  • the base station identifier and signal characteristic data After acquiring the base station identifier and signal characteristic data, associate the user equipment UE, the base station and the signal characteristic data, wherein the base station identifier and the signal characteristic data are base station identifiers received by the UE at the same time and signal characteristic data;
  • the data modeling includes calculating the distance between the UE and the base station, and calculating the distance between the UE and the In the case of the distance between the base stations, a set of equations for obtaining the UE position and signal strength is established;
  • a gradient descent method is used, and the minimum value of the function determined through the equation system is used as the position of the UE;
  • the transceiver 210 is configured to receive and send data under the control of the processor 200 .
  • establishing a system of equations for calculating the position of the UE and the signal strength is based on a signal path loss propagation model.
  • a gradient descent method is used to determine the position of the UE through a set of equations, including: determining the position of the UE using the triangulation method
  • use the signal path loss propagation model to establish the equations of the four base stations that each UE is connected to at the same time; for the equations of the four base stations, three equations are obtained after subtracting each other in turn; according to The three equations calculate the position of the UE.
  • x n is the current iteration position
  • x n+1 is the next iteration position
  • is the descending step size
  • is the learning rate, generally set to 0.9
  • v 0 is 0
  • f'(x n ) is the derivative value of the function at x n .
  • the method also includes:
  • the bus architecture may include any number of interconnected buses and bridges, one or more processors represented by the processor 200 and various circuits of the memory represented by the memory 220 are linked together.
  • the bus architecture can also link together various other circuits such as peripherals, voltage regulators, and power management circuits, etc., which are well known in the art and therefore will not be described herein.
  • the bus interface provides the interface.
  • Transceiver 210 may be a plurality of elements, including a transmitter and a receiver, providing a means for communicating with various other devices over a transmission medium.
  • the processor 200 is responsible for managing the bus architecture and general processing, and the memory 220 can store data used by the processor 200 when performing operations.
  • An embodiment of the present disclosure also provides a positioning device, as shown in FIG. 3 , the device 30 includes:
  • the data module 31 is configured to associate the user equipment UE, the base station and the signal characteristic data after obtaining the base station identification and signal characteristic data, wherein the base station identification and signal characteristic data are the base station identification received by the UE at the same time and signal characteristic data;
  • the modeling module 32 is configured to perform data modeling according to the base station location and signal characteristic data associated with the UE, wherein the data modeling includes calculating the distance between the UE and the base station, and in the case of calculating the distance between the UE and the base station Next, establish a set of equations for obtaining UE position and signal strength;
  • the positioning module 33 is configured to use the gradient descent method to use the minimum value of the function determined through the equation set as the position of the UE in the case of using the triangulation method to determine the position of the UE.
  • the modeling module 32 is configured to, when calculating the distance between the UE and the base station, establish a system of equations for obtaining the position of the UE and the signal strength, which is based on the signal path loss The propagation model was established.
  • the positioning module 33 is configured to use the signal path loss propagation model to establish equations of four base stations that each UE is connected to at the same time when using triangulation to determine the position of the UE; The four equations are sequentially subtracted two by two to obtain three equations; the position of the UE is calculated according to the three equations.
  • the positioning module 33 is configured to calculate the minimum value of the formula (7) when the minimum value of the function determined by the equation system is used as the position of the UE:
  • the positioning module 33 is configured to adopt the gradient descent method, and when the minimum value of the function determined through the equation system is used as the position of the UE, the iterative formula of the following formula (17) is used:
  • x n is the current iteration position
  • x n+1 is the next iteration position
  • is the descending step size
  • is the learning rate, generally set to 0.9
  • v n is the amount of decreasing momentum, v 0 is 0
  • f'(x n ) is the derivative value of the function at x n .
  • the positioning module 33 is configured to select the initial point x, y; calculate the partial derivative function of x, y; use the gradient descent method to calculate the position of the next point; perform iterative calculation, and set the jump-out condition to judge the calculation Whether the distance between the last two points satisfies the jump-out condition; if the jump-out condition is met, the distance between the two points has converged, and the iteration is jumped out; otherwise, repeat until the result is converged and jump out of the iteration; the position that satisfies the condition is the final position of the UE.
  • the data module 31 is further configured to use a local outlier factor algorithm to eliminate discrete base stations when acquiring the base stations received by the UE at the same time.
  • each part of the device described above is divided into various modules or units by function and described separately.
  • the functions of each module or unit can be implemented in one or more pieces of software or hardware.
  • Embodiments of the present disclosure also provide a computer-readable storage medium, where the computer-readable storage medium stores a computer program for executing the above positioning method.
  • wireless signal strength is used to perform triangulation calculation
  • a signal propagation model is used to establish a system of equations between user position and signal strength, which lays a foundation for solving triangulation.
  • the gradient descent method is used to solve the equation system, which reduces the difficulty of calculating the user's position in the complex function equation system.
  • the user position estimation scheme provided uses the gradient descent method to complete the user position calculation, provides a complete calculation scheme, and improves the accuracy of the user position calculation.
  • the number of base stations is screened through local factors and abnormal factors, and the data with excellent data network type and high quality is selected for the next step of calculation, and the user position calculation is realized based on the improved gradient descent method.
  • Collect the wireless signal propagation model establish the correlation equations between the distance between the user and the base station and the signal strength, and convert the solution of the equations into the problem of finding the minimum value of the function according to the function characteristics, and use the gradient descent method to complete the calculation of the user position , therefore, it has at least one of the following advantages compared with traditional triangulation techniques:
  • the applicability is good, using the gradient descent method to calculate the user position, compared with the traditional least square method, saves a lot of matrix operations, and at the same time, it can be applied to the solution of high-order non-linear equations, and can use multiple Various forms of equation solving and user position calculation.
  • High precision establish equations based on user signal strength, and realize user location calculation.
  • the fingerprint positioning method it is not limited to locating to the collected feature points, and calculates the user's location to a more precise position, with high positioning accuracy.
  • an embodiment of the present disclosure also provides a computer program, including computer readable codes, and when the computer readable codes run in an electronic device, the processor in the electronic device executes the program to implement the embodiments of the present disclosure. Any targeting method provided.
  • the computer program product can be realized by hardware, software or a combination thereof.
  • the computer program product is embodied as a computer storage medium, and in another embodiment, the computer program product is embodied as a software product, such as a software development kit (Software Development Kit, SDK) and the like.
  • the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Accordingly, the present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) having computer-usable program code embodied therein.
  • These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions
  • the device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.
  • the user equipment UE, the base station and the signal characteristic data are associated, and data modeling is performed according to the base station location and the signal characteristic data associated with the UE , in the case of calculating the distance between the UE and the base station, establishing a set of equations for obtaining the position of the UE and the signal strength, and in the case of using the triangulation method to determine the position of the UE, using a gradient descent method , using the minimum value of the function determined by the equation group as the position of the UE; the positioning method has high efficiency, compared with the fingerprint positioning method, it saves the workload of fingerprint collection and greatly saves the efficiency of triangulation positioning system construction And feasibility, it has broad prospects in the field of triangulation positioning construction based on wireless signal strength.
  • the applicability is good, using the gradient descent method to calculate the user position, compared with the traditional least square method, saves a lot of matrix operations, and at the same time, it can be applied to the solution of high-order non-linear equations, and can use multiple Various forms of equation solving and user position calculation.
  • High precision establish equations based on user signal strength, realize user position calculation, in the fingerprint positioning method, it is not limited to positioning on the collected feature points, so that user position calculation is more accurate and positioning accuracy is high.

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

一种定位方法、装置及存储介质,包括:在获取到基站标识及信号特征数据之后,将用户设备UE、基站以及所述信号特征数据进行关联,其中,所述基站标识及所述信号特征数据是所述UE在同一时刻接收到的基站标识及信号特征数据(步骤101);根据基站位置及所述UE关联的所述信号特征数据进行数据建模,其中,所述数据建模中包括有计算所述UE到所述基站之间的距离,在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组(步骤102);在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置(步骤103)。

Description

一种定位方法、装置及存储介质
相关申请的交叉引用
本公开要求在2021年11月9日提交中国专利局、申请号为202111316692.7、申请名称为“一种定位方法、装置及存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本公开中。
技术领域
本公开涉及通信技术领域,特别涉及一种定位方法、装置及存储介质。
背景技术
随着时代飞速变迁,科学技术迅猛发展,信息服务质量效率提高,人们对于定位需求日益增加;在自然环境下存在因为没有卫星信号从而无法通过卫星定位实现位置定位的大量场景,同时在复杂环境下,如图书馆、商场、医院、体育馆、地下车库、货品仓库等场所对于人员以及物品的定位需求越来越多,因此基于泛在无线数据的定位技术应运而生,此类数据定位技术实现方案,现阶段定位技术多种多样,从实现方案分为有近邻法、三边(角)测量法和指纹法,方法如下所示:
近邻法:最简单的方式,直接选定那个信号强度最大的AP(接入点,Access Point)的位置,定位结果是热点位置数据库中存储的当前连接的AP的位置;
其不足在于:近邻法简单,实现快,但是定位精度得不到保证,依赖信标部署密度,成本高、精度差。
三边(角)测量法:通过信号的各种参数得到目标与AP的距离或者角度,用几何方法计算出位置。包括到达时间法、相对到达时间法、到达角度法、基于信号强度的测距方法,及其混合算法。
其不足在于:三角测量法理论上精度较高,相关算法集中在进行wifi、蓝牙等数据源的参数解算,使用方法多为最小二乘方法,此方法需要进行 复杂的矩阵运算,对于线性相关或者近似线性相关的数据存在盲区,对于高阶方程组解算适用性差。
指纹法:事先把各个位置上的信号特征采集并建模,存入指纹数据库。实时定位时,将当前的信号特征与指纹库中的数据进行匹配,从而确定目标位置。
其不足在于:传统指纹定位法利用采集的信号特征数据,建立信号强度数据库,并利用欧式距离模型或概率模型实时计算用户匹配度。此方法对于指纹采集过程中信号波动处理不科学,指纹定位方法定位精度有一定局限性,无法满足高精度定位需求;同时,指纹采集工作量巨大,严重影响了室内定位系统建设的效率和可行性。
综上,相关技术的不足在于,难以兼顾计算方便,以及定位精度高。
发明内容
本公开实施例提出了一种定位方法、装置及存储介质,用以使得定位技术中计算更方便,以及定位精度更高。
本公开提供以下技术方案:
本公开实施例提供了一种定位方法,所述方法包括:
在获取到基站标识及信号特征数据之后,将用户设备UE、基站以及所述信号特征数据进行关联,其中,所述基站标识及所述信号特征数据是所述UE在同一时刻接收到的基站标识及信号特征数据;根据基站位置及所述UE关联的所述信号特征数据进行数据建模,其中,所述数据建模中包括有计算所述UE到所述基站之间的距离,在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组;在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置。
在一些实施例中,所述在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组,是依据信号路损传播模型建立 的。
在一些实施例中,在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,通过等式方程组确定所述UE的位置,包括:在使用三角定位法确定所述UE位置的情况下,使用所述信号路损传播模型建立每个UE同时连接的四个基站的等式;对于所述四个基站的等式,采用依次两两相减后,得到三个等式;根据所述三个等式计算UE的位置。
在一些实施例中,所述通过等式方程组确定出的函数的极小值作为所述UE的位置时,是求下式的极小值:
Figure PCTCN2022128767-appb-000001
其中,UE位置为(X,Y,H),四个基站位置分别为(X 1,Y 1),(X 2,Y 2),(X 3,Y 3),(X 4,Y 4),Dm 21=DB 2-DB 1,Dm 31=DB 3-DB 1,Dm 41=DB 4-DB 1,K为路径损耗系数,发射强度为A,接收强度为DB,PL 1=A-DB 1,PL 2=A-DB 2,PL 3=A-DB 3,PL 4=A-DB 4,PL 1、PL 2、PL 3、PL 4分别为UE与四个基站根据路损传播模型建立的等式中的信号强度路径损耗。
在一些实施例中,所述采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置时,使用如下式的迭代公式:
x n+1=x n-αv n+1
v n+1=βv n+(1-β)f′(x n-αv n)
其中:x n:为当前迭代位置;x n+1:为当下一次迭代位置;α:为下降步长;β:为学习率,设置为0.9;v n:为增加动量的下降的量,v 0为0; f’(x n):为函数在x n处导数值。
在一些实施例中,所述采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置,包括:选择初始点x、y;求x、y的偏导函数;利用梯度下降法计算下一个点位置;进行迭代计算,并设置跳出条件,判断计算后两个点的距离是否满足跳出条件;满足跳出条件,则两个点的距离已收敛,跳出迭代,否则重复直到满足结果收敛跳出迭代;满足条件的位置为UE的最终位置。实施中,利用梯度下降法计算下一个点位置时,设置步长Step为1米;进行迭代计算时,设置跳出条件为T=0.2米。
在一些实施例中,所述方法还包括:获取所述UE同一时刻接收到的基站时,使用局部异常因子算法剔除离散基站。
本公开实施例提供了一种定位装置,包括:处理器,用于读取存储器中的程序,执行下列过程:在获取到基站标识及信号特征数据之后,将用户设备UE、基站以及所述信号特征数据进行关联,其中,所述基站标识及所述信号特征数据是所述UE在同一时刻接收到的基站标识及信号特征数据;根据基站位置及所述UE关联的所述信号特征数据进行数据建模,其中,所述数据建模中包括有计算所述UE到所述基站之间的距离,在计算所述UE到基站所述之间的距离的情况下,建立求取UE位置与信号强度的等式方程组;在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置;收发机,用于在所述处理器的控制下接收和发送数据。
本公开实施例提供了一种定位装置,包括:数据模块,用于在获取到基站标识及信号特征数据之后,将用户设备UE、基站以及所述信号特征数据进行关联,其中,所述基站标识及所述信号特征数据是所述UE在同一时刻接收到的基站标识及信号特征数据;建模模块,用于根据基站位置及所述UE关联的所述信号特征数据进行数据建模,其中,所述数据建模中包括有计算所述UE到所述基站之间的距离,在计算所述UE到所述基站之间的 距离的情况下,建立求取UE位置与信号强度的等式方程组;定位模块,用于在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置。
本公开实施例提供了一种计算机程序,包括计算机可读代码,当所述计算机可读代码在电子设备中运行时,所述电子设备中的处理器执行用于实现上述一个或多个实施例中的定位方法。
本公开实施例提供的技术方案中,通过采集无线信号,建立用户与基站之间距离和信号强度的关联方程组,根据函数特性,将方程组解算转换为求取函数的极小值问题,并利用梯度下降方法完成用户位置计算,因此,与传统三角定位技术相比至少具有以下优点之一:
效率高,相比指纹定位方法,节省了指纹采集的工作量,大大节省了三角定位系统建设的效率和可行性,在基于无线信号强度的三角定位建设领域具有广阔的前景。
适用性好,使用梯度下降方法进行用户位置计算,相比传统的最小二乘方法,省去了大量的矩阵运算,同时,能够适用于高阶不可线性化的方程组的解算,能够使用多种形式的方程组求解及用户位置计算。
精度高,基于用户信号强度建立方程组,实现用户位置计算,在指纹定位方法中不限于定位到采集的特征点上,使得用户位置计算更为精确,定位精度高。
应当理解的是,以上的一般描述和后文的细节描述仅是示例性和解释性的,而非限制本公开。
根据下面参考附图对示例性实施例的详细说明,本公开实施例的其它特征及方面将变得清楚。
附图说明
此处所说明的附图用来提供对本公开的理解,构成本公开的一部分,本公开的示意性实施例及其说明用于解释本公开,并不构成对本公开的不当限定。在附图中:
图1为本公开实施例中定位方法实施流程示意图;
图2为本公开实施例中一种定位装置的组成结构示意图;
图3为本公开实施例中另一种定位装置的组成结构示意图。
具体实施方式
以下将参考附图详细说明本公开的各种示例性实施例、特征和方面。附图中相同的附图标记表示功能相同或相似的元件。尽管在附图中示出了实施例的各种方面,但是除非特别指出,不必按比例绘制附图。
在这里专用的词“示例性”意为“用作例子、实施例或说明性”。这里作为“示例性”所说明的任何实施例不必解释为优于或好于其它实施例。
本文中术语“和/或”,仅仅是一种描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。另外,本文中术语“至少一种”表示多种中的任意一种或多种中的至少两种的任意组合,例如,包括A、B、C中的至少一种,可以表示包括从A、B和C构成的集合中选择的任意一个或多个元素。
另外,为了更好的说明本公开实施例,在下文的实施方式中给出了众多的细节。本领域技术人员应当理解,没有某些细节,本公开实施例同样可以实施。在一些实例中,对于本领域技术人员熟知的方法、手段、元件和电路未作详细描述,以便于凸显本公开实施例的主旨。以下将参考附图详细说明本公开的各种示例性实施例、特征和方面。附图中相同的附图标记表示功能相同或相似的元件。尽管在附图中示出了实施例的各种方面,但是除非特别指出,不必按比例绘制附图。
本公开实施例提供的技术方案中,通过局部因子异常因子进行基站数筛选,选择数据网型优、质量高的数据进行下一步计算,通过无线信号传播模型差分计算,建立用户与基站之间距离和信号强度的关联方程组,根据函数特性,将方程组解算转换为求取函数的极小值问题,并利用梯度下降方法完成用户位置计算,具有计算方便,精度高的优点。下面结合附图 对本公开的实施方式进行说明。
图1为定位方法实施流程示意图,如图所示,可以包括:
步骤101、在获取到基站标识及信号特征数据之后,将用户设备UE、基站以及所述信号特征数据进行关联,其中,所述基站标识及所述信号特征数据是所述UE在同一时刻接收到的基站标识及信号特征数据;
步骤102、根据基站位置及所述UE关联的所述信号特征数据进行数据建模,其中,所述数据建模中包括有计算所述UE到所述基站之间的距离,在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组;
其中,所述信号特征数据可以包括所述信号强度。
步骤103、在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置。
LOF(局部异常因子算法,Local Outlier Factor)通过计算“局部可达密度”来反映一个样本的异常程度,一个样本点的局部可达密度越大,这个样本点就越有可能是异常点。
在利用无线信号进行用户三角定位方法中,在获取无线基站的位置、用户接入基站的信号强度等信息的情况下,根据基站站址分布,使用局部异常因子算法进行网型优化和筛选,剔除离散基站对于用户位置的计算的误差;并且根据无线信号传播模型,建立用户位置与信号强度的关联关系,也就是建立用户位置与信号强度的方程组。方程组解算采用改进的梯度下降方法,能够很好的降低复杂方程组的计算难度,提升用户位置计算精度和效率。
采用本公开提供的技术方案,大大节省了三角定位系统建设的效率和可行性,能够适用于高阶不可线性化的方程组的解算,能够使用多种形式的方程组求解及用户位置计算,使得用户位置计算更为精确,定位精度高。
本申请实施例提供一种定位方法,所述方法可以包括基站数据筛选、 数据建模和用户位置求解三个步骤。在实例中,也会采用用户这一称谓,这是考虑到业界有人员习惯于使用用户来描述方案,因此实施例中同时采用两种描述,使用用户描述的含义与使用UE(用户设备,User Equipment)是相同的,是为了使本领域技术人员更好地、更容易理解本方案的实施。
步骤S22:基站数据筛选。
在一些实施例中,步骤S22可以包括:获取所述UE同一时刻接收到的基站时,使用局部异常因子算法剔除离散基站。
在一些实施例中,根据数据特征,可以获取用户在同一时刻能够接收到的基站标识及RSRP(参考信号接收功率,Reference Signal Receiving Power)数据,根据基站标识,结合基站站址表,可以将用户此时刻关联的所有基站数据统计,形成基站站址分布数据,针对上述数据利用LOF(局部异常因子算法)算法进行离散基站筛选,所述步骤S22的处理方式可以包括如下步骤:
步骤S221:将所有基站数据看作一个数据点的集合C。
步骤S222:从集合中顺序选两个点p和o,计算两个点的之间的距离d(p,o)。
步骤S223:计算点p的第K距离,也就是距离p第K远的点的距离,不包括p,记为d K(p),计算方法可以如下公式(1)所示:
d K(p)=d(p,o)公式(1);
且d(p,o)满足如下两个条件:
在集合中至少有不包括p在内的k个点o′∈C{x≠p},满足d(p,o′)≤d(p,o);
在集合中最多有不包括p在内的K-1个点o′∈C{x≠p},满足d(p,o′)<d(p,o)。
步骤S224:计算点p的第K邻域,记为N K(p),就是p的第k距离即以内的所有点,包括第k距离。同时计算点p的K邻域内的点的个数,记 为|N K(p)|。
步骤S225:计算集合内各个点与点p的可达距离,记为:reach_d K(p,o),计算方法可以如下公式(2)所示:
reach_d K(p,o)=max{d K(o),d(p,o)}公式(2);
点o到点p的第k可达距离,至少是o的第k距离,或者为o、p间的真实距离,需要注意的是可达距离具有方向性,reach_d K(p,o)和reach_d K(o,p)并不相等。
步骤S226:计算点p的局部可达密度,记为lrd K(p),计算公式可以如下公式(3)所示:
Figure PCTCN2022128767-appb-000002
上述公式(3)的含义为表示点p的第k邻域内点到p的平均可达距离的倒数。
步骤S227:计算点p的局部离群因子,记为LOF K(p),计算公式可以如下公式(4)所示:
Figure PCTCN2022128767-appb-000003
上述公式点p的邻域点N K(p)的局部可达密度与点p的局部可达密度之比的平均数,在这个比值越接近1的情况下,说明p的其邻域点密度差不多,p可能和邻域同属一簇;在这个比值越小于1的情况下,说明p的密度高于其邻域点密度,p为密集点;在这个比值越大于1的情况下,说明p的密度小于其邻域点密度,p越可能是异常点。
步骤S228:计算所有点的K邻域的局部离群因子,并根据预设条件判断当前点是否为离散点。根据算法要求,需要明确K邻域的K值,并且需要明确局部离散因子的阈值,因此特定义K值取值方案如下公式(5)所示:
Figure PCTCN2022128767-appb-000004
其中,在集合C中点的个数小于等于4个的情况下,不适用局部异常因子算法进行过滤,在C中点的个数大于等于5个的情况下,K取值为点的个数减1;
局部离散因子的阈值LOF T(C),经过实验分析,可以选择为1.8,即在点的局部离散因子大于1.8的情况下判定为离群点。
步骤S24:数据建模。
在一些实施例中,步骤S102中所述在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组,是依据信号路损传播模型建立的。
ITU(国际电信联盟,International Telecommunication Union)推荐使用的路损传播模型可以如下公式(6)所示:
PL=20log(f)+Klog(d)+L f-28db+x δ公式(6);
其中:PL标识信号强度路径损耗(发射强度-接收强度,一般同一场所基站发射强度固定,接收强度因距离原因不同);f为频率,单位MHZ(兆赫兹);K为路径损耗系数;d表示用户和基站之间的距离;L f为楼层穿透损耗系数;x δ为慢衰落余量。
在一些实施例中,步骤103“在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,通过等式方程组确定所述UE的位置”,包括:
步骤S131:在使用三角定位法确定所述UE位置的情况下,使用所述信号路损传播模型建立每个UE同时连接的四个基站的等式;
步骤S132:对于所述四个基站的等式,采用依次两两相减后,得到三个等式;
步骤S133:根据所述三个等式计算UE的位置。
在一些实施例中,步骤103中所述通过等式方程组确定出的函数的极小值作为所述UE的位置时,是求如下公式(7)的极小值:
Figure PCTCN2022128767-appb-000005
其中,UE位置为(X,Y,H),四个基站位置分别为(X 1,Y 1),(X 2,Y 2),(X 3,Y 3),(X 4,Y 4),Dm 21=DB 2-DB 1,Dm 31=DB 3-DB 1,Dm 41=DB 4-DB 1,K为路径损耗系数,发射强度为A,接收强度为DB,PL 1=A-DB 1,PL 2=A-DB 2,PL 3=A-DB 3,PL 4=A-DB 4,PL 1、PL 2、PL 3、PL 4分别为UE与四个基站根据路损传播模型建立的等式中的信号强度路径损耗。
在基站数据筛选过后,需要根据基站位置及用户关联的信号特征数据进行数据建模,根据信号路损传播模型,计算用户到基站之间的距离,建立至少三个等式,求取用户位置,该算法需要根据信号路损计算距离,从而推断用户位置,适用于空旷场景,比如展馆、机场等。
为了使用三角定位方法计算用户位置,要求满足每个用户同时连接四个基站,并使用所述信号路损传播模型建立等式,比如建立如下公式(8)所示的四个等式:
Figure PCTCN2022128767-appb-000006
对于上述四个等式,采用依次两两相减,得到三个等式,如下公式(9)所示:
Figure PCTCN2022128767-appb-000007
由PL的含义可知,PL=发射强度A-接收强度DB,因此设某场所发射强度为A,则有如下公式(10):
Figure PCTCN2022128767-appb-000008
则如下公式(11)所示,
Figure PCTCN2022128767-appb-000009
且如下公式(12)所示,
Figure PCTCN2022128767-appb-000010
因此公式(9)中的等式可以化简为如下公式(13)所示:
Figure PCTCN2022128767-appb-000011
令Dm 21=DB 2-DB 1,Dm 31=DB 3-DB 1,Dm 41=DB 4-DB 1,并对于步骤5中的公式进行去对数化简得到如下公式(14):
Figure PCTCN2022128767-appb-000012
设用户位置为(X,Y,H),四个基站位置分别为(X 1,Y 1),(X 2,Y 2),(X 3,Y 3),(X 4,Y 4),则公式(14)可以整理为如下公式(15):
Figure PCTCN2022128767-appb-000013
至此,三角定位可以根据公式(15)计算用户的位置,计算方法可以是将公式(15)中的三个等式右侧移项到左侧,使右侧等0;并对公式(15)中的三个等式求绝对值并相加得到如下公式(16)中的等式:
Figure PCTCN2022128767-appb-000014
则,如下公式(7)所示,令:
Figure PCTCN2022128767-appb-000015
上述用户位置求解即可转换为计算f(X,Y)的极小值点问题。
步骤S26:用户求解。
在一些实施例中,步骤103中所述采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置时,使用如下公式(17)的迭代公式:
Figure PCTCN2022128767-appb-000016
其中:
x n:为当前迭代位置;
x n+1:为当下一次迭代位置;
α:为下降步长;
β:为学习率,一般可以设置为0.9;
v n:为增加动量的下降的量,v 0为0;
f’(x n):为函数在x n处导数值。
可以根据f(X,Y)函数式,计算函数的极小值点,本实施例中使用改进的梯度下降法求取用户位置,该方法基本原理为:
设函数f(x),求的最小值点,根据迭代梯度下降方法原理,则:
x n+1=x n-αf’(x n)公式(18);
上述公式(18)中:
x n+1:为下一次迭代位置。
x n:为当前迭代位置。
α:为下降步长。
f’(x n):为函数在x n处导数值。
但是此方式在计算过程中,固定步长,对于函数适用性差,收敛速度慢,也可能会造成区间震荡。因此实施例中将使用改进的梯度下降方法,在计算下一个位置的情况下,充分考虑代价函数在下一个位置的梯度,进行动态调整步长,能够有效降低震荡收敛的影响,且提升收敛速度,如公式(17)所示。
在一些实施例中,步骤103中“所述采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置”,包括:选择初始点x、y;求x、y的偏导函数;利用梯度下降法计算下一个点位置;进行迭代计算,并设置跳出条件,判断计算后两个点的距离是否满足跳出条件;满足跳出条件,则两个点的距离已收敛,跳出迭代,否则重复直到满足结果收敛跳出迭代;满足条件的位置为所述UE的最终位置。
其中,利用梯度下降法计算下一个点位置时,可以设置步长Step为1米;进行迭代计算时,可以设置跳出条件为T=0.2米。
因此根据上述梯度下降原理,结合实施例中的数据,计算可以如下所 示:
选择初始点,令初始点为
Figure PCTCN2022128767-appb-000017
Figure PCTCN2022128767-appb-000018
计算f(X,Y)对于x、y的偏导函数,如下公式(19)和公式(20)所示:
Figure PCTCN2022128767-appb-000019
Figure PCTCN2022128767-appb-000020
利用改进的梯度下降方法计算下一个点位置,设置步长Step为1米,公式(21)如下所示:
Figure PCTCN2022128767-appb-000021
使用公式(20)进行迭代计算,并设置跳出条件为T=0.2米,判断计算后两个点的距离是否满足跳出条件,如下公式(22)令:
Figure PCTCN2022128767-appb-000022
则,判断d是否小于T,在是的情况下,满足跳出条件,结果已收敛,跳出迭代,在不是的情况下,以
Figure PCTCN2022128767-appb-000023
为起点重复公式(21)和公式(22)中的计算,直到满足结果收敛。
满足条件的
Figure PCTCN2022128767-appb-000024
为用户的最终位置,完成用户位置计算。
本公开实施例中还提供了一种定位装置、及计算机可读存储介质,与定位方法相似,因此这些设备的实施可以参见方法的实施,重复之处不再 赘述。
在实施本公开实施例提供的技术方案的情况下,可以按如下方式实施。
图2为定位装置结构示意图,如图所示,装置中包括:
处理器200,配置为读取存储器220中的程序,执行下列过程:
在获取到基站标识及信号特征数据之后,将用户设备UE、基站以及所述信号特征数据进行关联,其中,所述基站标识及所述信号特征数据是所述UE在同一时刻接收到的基站标识及信号特征数据;
根据基站位置及所述UE关联的所述信号特征数据进行数据建模,其中,所述数据建模中包括有计算所述UE到所述基站之间的距离,在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组;
在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置;
收发机210,配置为在处理器200的控制下接收和发送数据。
在一些实施例中,所述在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组,是依据信号路损传播模型建立的。
在一些实施例中,在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,通过等式方程组确定所述UE的位置,包括:在使用三角定位法确定所述UE位置的情况下,使用所述信号路损传播模型建立每个UE同时连接的四个基站的等式;对于所述四个基站的等式,采用依次两两相减后,得到三个等式;根据所述三个等式计算UE的位置。
在一些实施例中,所述通过等式方程组确定出的函数的极小值作为所述UE的位置时,是求如下公式(7)的极小值:
Figure PCTCN2022128767-appb-000025
Figure PCTCN2022128767-appb-000026
其中,UE位置为(X,Y,H),四个基站位置分别为(X 1,Y 1),(X 2,Y 2),(X 3,Y 3),(X 4,Y 4),Dm 21=DB 2-DB 1,Dm 31=DB 3-DB 1,Dm 41=DB 4-DB 1,K为路径损耗系数,发射强度为A,接收强度为DB,PL 1=A-DB 1,PL 2=A-DB 2,PL 3=A-DB 3,PL 4=A-DB 4,PL 1、PL 2、PL 3、PL 4分别为UE与四个基站根据路损传播模型建立的等式中的信号强度路径损耗。
在一些实施例中,所述采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置时,使用如下公式(17)的迭代公式:
Figure PCTCN2022128767-appb-000027
其中:x n:为当前迭代位置;x n+1:为当下一次迭代位置;α:为下降步长;β:为学习率,一般设置为0.9;v n:为增加动量的下降的量,v 0为0;f’(x n):为函数在x n处导数值。
在一些实施例,所述采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置,包括:选择初始点x、y;求x、y的偏导函数;利用梯度下降法计算下一个点位置,设置步长Step为1米;进行迭代计算,并设置跳出条件,判断计算后两个点的距离是否满足跳出条件,设置跳出条件为T=0.2米;满足跳出条件,则两个点的距离已收敛,跳出迭代,否则重复直到满足结果收敛跳出迭代;满足条件的位置为所述UE的最终位置。
在一些实施例中,利用梯度下降法计算下一个点位置时,设置步长Step为1米;进行迭代计算时,设置跳出条件为T=0.2米。
在一些实施例中,所述方法还包括:
获取所述UE同一时刻接收到的基站时,使用局部异常因子算法剔除离 散基站。
其中,在图2中,总线架构可以包括任意数量的互联的总线和桥,由处理器200代表的一个或多个处理器和存储器220代表的存储器的各种电路链接在一起。总线架构还可以将诸如外围设备、稳压器和功率管理电路等之类的各种其他电路链接在一起,这些都是本领域所公知的,因此,本文不再对其进行描述。总线接口提供接口。收发机210可以是多个元件,即包括发送机和接收机,提供用于在传输介质上与各种其他装置通信的单元。处理器200负责管理总线架构和通常的处理,存储器220可以存储处理器200在执行操作时所使用的数据。
本公开实施例中还提供了一种定位装置,如图3所示,所述装置30包括:
数据模块31,配置为在获取到基站标识及信号特征数据之后,将用户设备UE、基站以及信号特征数据进行关联,其中,所述基站标识及信号特征数据是UE在同一时刻接收到的基站标识及信号特征数据;
建模模块32,配置为根据基站位置及UE关联的信号特征数据进行数据建模,其中,数据建模中包括有计算UE到基站之间的距离,在计算UE到基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组;
定位模块33,配置为在使用三角定位法确定UE位置的情况下,采用梯度下降法,将通过等式方程组确定出的函数的极小值作为UE的位置。
在一些实施例中,所述建模模块32配置为在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组,是依据信号路损传播模型建立的。
在一些实施例中,所述定位模块33配置为在使用三角定位法确定所述UE位置的情况下,使用所述信号路损传播模型建立每个UE同时连接的四个基站的等式;对于所述四个等式,采用依次两两相减后,得到三个等式;根据所述三个等式计算UE的位置。
在一些实施例中,所述定位模块33配置为通过等式方程组确定出的函数的极小值作为所述UE的位置时,是求公式(7)的极小值:
Figure PCTCN2022128767-appb-000028
其中,UE位置为(X,Y,H),四个基站位置分别为(X 1,Y 1),(X 2,Y 2),(X 3,Y 3),(X 4,Y 4),Dm 21=DB 2-DB 1,Dm 31=DB 3-DB 1,Dm 41=DB 4-DB 1,K为路径损耗系数,发射强度为A,接收强度为DB,PL 1=A-DB 1,PL 2=A-DB 2,PL 3=A-DB 3,PL 4=A-DB 4,PL 1、PL 2、PL 3、PL 4分别为UE与四个基站根据路损传播模型建立的等式中的信号强度路径损耗。
在一些实施例中,所述定位模块33配置为采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置时,使用如下公式(17)的迭代公式:
Figure PCTCN2022128767-appb-000029
其中:x n:为当前迭代位置;x n+1:为当下一次迭代位置;α:为下降步长;β:为学习率,一般设置为0.9;v n:为增加动量的下降的量,v 0为0;f’(x n):为函数在x n处导数值。
在一些实施例中,所述定位模块33配置为选择初始点x、y;求x、y的偏导函数;利用梯度下降法计算下一个点位置;进行迭代计算,并设置跳出条件,判断计算后两个点的距离是否满足跳出条件;满足跳出条件,则两个点的距离已收敛,跳出迭代,否则重复直到满足结果收敛跳出迭代;满足条件的位置为所述UE的最终位置。
在一些实施例中,所述定位模块33配置为利用梯度下降法计算下一个点位置时,设置步长Step为1米;进行迭代计算时,设置跳出条件为T=0.2 米。
在一些实施例中,所述数据模块31还配置为获取所述UE同一时刻接收到的基站时,使用局部异常因子算法剔除离散基站。
为了描述的方便,以上所述装置的各部分以功能分为各种模块或单元分别描述。当然,在实施本公开时可以把各模块或单元的功能在同一个或多个软件或硬件中实现。
本公开实施例中还提供了一种计算机可读存储介质,所述计算机可读存储介质存储有执行上述定位方法的计算机程序。
综上所述,本公开实施例提供的技术方案中,利用无线信号强度进行三角定位计算,利用信号传播模型,建立用户位置与信号强度的方程组,为三角定位求解奠定基础。
在一些实施例中,利用了梯度下降方法进行方程组求解,降低复杂函数方程组用户位置计算难度。在一些实施例中,提供的用户位置估算方案,利用梯度下降方法完成用户位置计算,给出了完整的计算方案,提高用户位置计算的准确性。
由于本公开实施例提供的技术方案中,通过局部因子异常因子进行基站数筛选,选择数据网型优、质量高的数据进行下一步计算,并基于改进的梯度下降的方法实现用户位置计算,通过采集无线信号传播模型,建立用户与基站之间距离和信号强度的关联方程组,根据函数特性,将方程组解算转换为求取函数的极小值问题,并利用梯度下降方法完成用户位置计算,因此,与传统三角定位技术相比至少具有以下优点之一:
效率高,相比指纹定位方法,节省了指纹采集的工作量,大大节省了三角定位系统建设的效率和可行性,在基于无线信号强度的三角定位建设领域具有广阔的前景。
适用性好,使用梯度下降方法进行用户位置计算,相比传统的最小二乘方法,省去了大量的矩阵运算,同时,能够适用于高阶不可线性化的方 程组的解算,能够使用多种形式的方程组求解及用户位置计算。
精度高,基于用户信号强度建立方程组,实现用户位置计算,在指纹定位方法中不限于定位到采集的特征点上,将用户位置计算到更为精确的位置上,定位精度高。
相应的,本公开实施例还提供一种计算机程序,包括计算机可读代码,当所述计算机可读代码在电子设备中运行时,所述电子设备中的处理器执行用于实现本公开实施例提供的任一定位方法。
该计算机程序产品可以通过硬件、软件或其结合的方式实现。在一个实施例中,所述计算机程序产品体现为计算机存储介质,在另一个实施例中,计算机程序产品体现为软件产品,例如软件开发包(Software Development Kit,SDK)等等。
本领域内的技术人员应明白,本公开的实施例可提供为方法、系统、或计算机程序产品。因此,本公开可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本公开可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器和光学存储器等)上实施的计算机程序产品的形式。
本公开是参照根据本公开实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个 流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
显然,本领域的技术人员可以对本公开进行各种改动和变型而不脱离本公开的精神和范围。这样,倘若本公开的这些修改和变型属于本公开权利要求及其等同技术的范围之内,则本公开也意图包含这些改动和变型在内。
工业实用性
本公开实施例中,在获取到基站标识及信号特征数据之后,将用户设备UE、基站以及所述信号特征数据进行关联,根据基站位置及所述UE关联的所述信号特征数据进行数据建模,在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组,在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置;定位方法效率高,相比指纹定位方法,节省了指纹采集的工作量,大大节省了三角定位系统建设的效率和可行性,在基于无线信号强度的三角定位建设领域具有广阔的前景。适用性好,使用梯度下降方法进行用户位置计算,相比传统的最小二乘方法,省去了大量的矩阵运算,同时,能够适用于高阶不可线性化的方程组的解算,能够使用多种形式的方程组求解及用户位置计算。精度高,基于用户信号强度建立方程组,实现用户位置计算,在指纹定位方法中不限于定位到采集的特征点上,使得用户位置计算更为精确,定位精度高。

Claims (17)

  1. 一种定位方法,包括:
    在获取到基站标识及信号特征数据之后,将用户设备UE、基站以及所述信号特征数据进行关联,其中,所述基站标识及所述信号特征数据是所述UE在同一时刻接收到的基站标识及信号特征数据;
    根据基站位置及所述UE关联的所述信号特征数据进行数据建模,其中,所述数据建模中包括有计算所述UE到所述基站之间的距离,在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组;
    在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置。
  2. 如权利要求1所述的方法,其中,所述在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组,是依据信号路损传播模型建立的。
  3. 如权利要求2所述的方法,其中,在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,通过等式方程组确定所述UE的位置,包括:
    在使用三角定位法确定所述UE位置的情况下,使用所述信号路损传播模型建立每个UE同时连接的四个基站的等式;
    对于所述四个基站的等式,采用依次两两相减后,得到三个等式;
    根据所述三个等式计算UE的位置。
  4. 如权利要求2所述的方法,其中,所述通过等式方程组确定出的函数的极小值作为所述UE的位置时,是求下式的极小值:
    Figure PCTCN2022128767-appb-100001
    其中,UE位置为(X,Y,H),四个基站位置分别为(X 1,Y 1),(X 2,Y 2),(X 3,Y 3),(X 4,Y 4),Dm 21=DB 2-DB 1,Dm 31=DB 3-DB 1,Dm 41=DB 4-DB 1,K为路径损耗系数,发射强度为A,接收强度为DB,PL 1=A-DB 1,PL 2=A-DB 2,PL 3=A-DB 3,PL 4=A-DB 4,PL 1、PL 2、PL 3、PL 4分别为UE与四个基站根据路损传播模型建立的等式中的信号强度路径损耗。
  5. 如权利要求2所述的方法,其中,所述采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置时,使用如下式的迭代公式:
    x n+1=x n-αv n+1
    v n+1=βv n+(1-β)f′(x n-αv n)
    其中:
    x n:为当前迭代位置;
    x n+1:为当下一次迭代位置;
    α:为下降步长;
    β:为学习率;
    v n:为增加动量的下降的量,v 0为0;
    f’(x n):为函数在x n处导数值。
  6. 如权利要求5所述的方法,其中,所述采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置,包括:
    选择初始点x、y;
    求x、y的偏导函数;
    利用梯度下降法计算下一个点位置,设置步长Step为1米;
    进行迭代计算,并设置跳出条件,判断计算后两个点的距离是否满足跳出条件,设置跳出条件为T=0.2米;
    满足跳出条件,则两个点的距离已收敛,跳出迭代,否则重复直到满足结果收敛跳出迭代;
    满足条件的位置为所述UE的最终位置。
  7. 如权利要求1至6任一所述的方法,其中,所述方法还包括:
    获取所述UE同一时刻接收到的基站时,使用局部异常因子算法剔除离散基站。
  8. 一种定位装置,包括:
    处理器,用于读取存储器中的程序,执行下列过程:
    在获取到基站标识及信号特征数据之后,将用户设备UE、基站以及所述信号特征数据进行关联,其中,所述基站标识及所述信号特征数据是所述UE在同一时刻接收到的基站标识及信号特征数据;
    根据基站位置及所述UE关联的所述信号特征数据进行数据建模,其中,所述数据建模中包括有计算所述UE到所述基站之间的距离,在计算所述UE到基站所述之间的距离的情况下,建立求取UE位置与信号强度的等式方程组;
    在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置;
    收发机,用于在所述处理器的控制下接收和发送数据。
  9. 一种定位装置,包括:
    数据模块,用于在获取到基站标识及信号特征数据之后,将用户设备UE、基站以及所述信号特征数据进行关联,其中,所述基站标识及所述信 号特征数据是所述UE在同一时刻接收到的基站标识及信号特征数据;
    建模模块,用于根据基站位置及所述UE关联的所述信号特征数据进行数据建模,其中,所述数据建模中包括有计算所述UE到所述基站之间的距离,在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组;
    定位模块,用于在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置。
  10. 如权利要求9所述的装置,其中,所述建模模块配置为在计算所述UE到所述基站之间的距离的情况下,建立求取UE位置与信号强度的等式方程组,是依据信号路损传播模型建立的。
  11. 如权利要求10所述的装置,其中,所述定位模块配置为在使用三角定位法确定所述UE位置的情况下,采用梯度下降法,通过等式方程组确定所述UE的位置,包括:在使用三角定位法确定所述UE位置的情况下,使用所述信号路损传播模型建立每个UE同时连接的四个基站的等式;对于所述四个基站的等式,采用依次两两相减后,得到三个等式;根据所述三个等式计算UE的位置。
  12. 如权利要求10所述的装置,其中,所述定位模块配置为通过等式方程组确定出的函数的极小值作为所述UE的位置时,是求下式的极小值:
    Figure PCTCN2022128767-appb-100002
    其中,UE位置为(X,Y,H),四个基站位置分别为(X 1,Y 1),(X 2,Y 2),(X 3,Y 3),(X 4,Y 4),Dm 21=DB 2-DB 1,Dm 31=DB 3- DB 1,Dm 41=DB 4-DB 1,K为路径损耗系数,发射强度为A,接收强度为DB,PL 1=A-DB 1,PL 2=A-DB 2,PL 3=A-DB 3,PL 4=A-DB 4,PL 1、PL 2、PL 3、PL 4分别为UE与四个基站根据路损传播模型建立的等式中的信号强度路径损耗。
  13. 如权利要求10所述的装置,其中,所述定位模块配置为采用梯度下降法,将通过等式方程组确定出的函数的极小值作为所述UE的位置时,使用如下式的迭代公式:
    x n+1=x n-αv n+1
    v n+1=βv n+(1-β)f′(x n-αv n)
    其中:x n:为当前迭代位置;x n+1:为当下一次迭代位置;α:为下降步长;β:为学习率;v n:为增加动量的下降的量,v 0为0;f’(x n):为函数在x n处导数值。
  14. 如权利要求13所述的装置,其中,所述定位模块配置为选择初始点x、y;求x、y的偏导函数;利用梯度下降法计算下一个点位置,设置步长Step为1米;进行迭代计算,并设置跳出条件,判断计算后两个点的距离是否满足跳出条件,设置跳出条件为T=0.2米;满足跳出条件,则两个点的距离已收敛,跳出迭代,否则重复直到满足结果收敛跳出迭代;满足条件的位置为所述UE的最终位置。
  15. 如权利要求1至14任一所述的装置,其中,所述数据模块还配置为获取所述UE同一时刻接收到的基站时,使用局部异常因子算法剔除离散基站。
  16. 一种计算机可读存储介质,所述计算机可读存储介质存储有执行权利要求1至7任一所述方法的计算机程序。
  17. 一种计算机程序,包括计算机可读代码,当所述计算机可读代码在电子设备中运行时,所述电子设备中的处理器执行用于实现权利要求1-7中任一项所述的定位方法。
PCT/CN2022/128767 2021-11-09 2022-10-31 一种定位方法、装置及存储介质 WO2023083041A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111316692.7A CN113759311B (zh) 2021-11-09 2021-11-09 一种定位方法、装置及存储介质
CN202111316692.7 2021-11-09

Publications (1)

Publication Number Publication Date
WO2023083041A1 true WO2023083041A1 (zh) 2023-05-19

Family

ID=78784661

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/128767 WO2023083041A1 (zh) 2021-11-09 2022-10-31 一种定位方法、装置及存储介质

Country Status (2)

Country Link
CN (1) CN113759311B (zh)
WO (1) WO2023083041A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113759311B (zh) * 2021-11-09 2022-03-15 中移(上海)信息通信科技有限公司 一种定位方法、装置及存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100137005A1 (en) * 2008-11-29 2010-06-03 Inventec Appliances Corp. Method for positioning portable communication device
CN104301993A (zh) * 2013-07-15 2015-01-21 中国移动通信集团设计院有限公司 一种基站站址定位的方法及装置
CN105678990A (zh) * 2015-12-31 2016-06-15 赵旭 一种遥控器及其检测、控制方法
CN107820206A (zh) * 2017-11-15 2018-03-20 玉林师范学院 基于信号强度的非视距定位方法
CN108696931A (zh) * 2018-03-26 2018-10-23 北京数字联盟网络科技有限公司 基于移动通信网络的定位方法及计算机可读存储介质
CN109298396A (zh) * 2018-11-12 2019-02-01 广州杰赛科技股份有限公司 一种无人机定位方法、装置及计算机可读存储介质
CN113759311A (zh) * 2021-11-09 2021-12-07 中移(上海)信息通信科技有限公司 一种定位方法、装置及存储介质

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108627798B (zh) * 2018-04-04 2022-03-11 北京工业大学 基于线性判别分析和梯度提升树的wlan室内定位算法
CN111225439B (zh) * 2018-11-23 2022-11-22 中兴通讯股份有限公司 确定终端位置的方法、装置及存储介质
CN111343664B (zh) * 2018-12-18 2023-04-07 中国移动通信集团江西有限公司 用户定位方法、装置、设备及介质
CN112533144B (zh) * 2019-09-19 2023-07-18 中国移动通信集团辽宁有限公司 室内定位方法、装置、计算设备及计算机存储介质
CN113329492B (zh) * 2021-08-04 2021-11-19 深圳市海豚科技创新有限公司 基于tdoa的定位方法、装置、计算机设备及存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100137005A1 (en) * 2008-11-29 2010-06-03 Inventec Appliances Corp. Method for positioning portable communication device
CN104301993A (zh) * 2013-07-15 2015-01-21 中国移动通信集团设计院有限公司 一种基站站址定位的方法及装置
CN105678990A (zh) * 2015-12-31 2016-06-15 赵旭 一种遥控器及其检测、控制方法
CN107820206A (zh) * 2017-11-15 2018-03-20 玉林师范学院 基于信号强度的非视距定位方法
CN108696931A (zh) * 2018-03-26 2018-10-23 北京数字联盟网络科技有限公司 基于移动通信网络的定位方法及计算机可读存储介质
CN109298396A (zh) * 2018-11-12 2019-02-01 广州杰赛科技股份有限公司 一种无人机定位方法、装置及计算机可读存储介质
CN113759311A (zh) * 2021-11-09 2021-12-07 中移(上海)信息通信科技有限公司 一种定位方法、装置及存储介质

Also Published As

Publication number Publication date
CN113759311A (zh) 2021-12-07
CN113759311B (zh) 2022-03-15

Similar Documents

Publication Publication Date Title
US9869748B2 (en) Locating a mobile device
CN103068035B (zh) 一种无线网络定位方法、装置及系统
Zhao et al. Applying kriging interpolation for WiFi fingerprinting based indoor positioning systems
CN109275095B (zh) 一种基于蓝牙的室内定位系统、定位设备和定位方法
CN110557716B (zh) 一种基于对数正态模型的室内定位方法
WO2019062734A1 (zh) 基于Wi-Fi热点的室内定位方法及装置
US7983684B2 (en) Methods and apparatus for determining RF transmitter placement via local coverage optimization
CN102984745B (zh) Wi-Fi AP位置及路径损耗模型的联合估测方法
CN103220777A (zh) 一种移动设备定位系统
CN105072581A (zh) 一种基于路径衰减系数建库的室内定位方法
CN109379711B (zh) 一种定位方法
CN103581831A (zh) 基于WiFi以及移动终端的室内定位算法
CN103945531B (zh) 基于信息熵的WLAN室内定位Radio Map更新方法
Du et al. KF-KNN: Low-cost and high-accurate FM-based indoor localization model via fingerprint technology
WO2023083041A1 (zh) 一种定位方法、装置及存储介质
CN103987118A (zh) 基于接收信号强度信号ZCA白化的接入点k-means聚类方法
Machaj et al. Impact of optimization algorithms on hybrid indoor positioning based on GSM and Wi‐Fi signals
CN107635275A (zh) 基于sdn的室内目标定位中的ap选择方法
WO2023029899A1 (zh) 一种定位方法及装置
Lu et al. A study of indoor positioning systems using iBeacons with different transmission power levels
Bornholdt et al. Adaptive procedure for indoor localization using LoRa devices
CN105392194B (zh) 基于异构网络室内定位框架最优通信的能耗精度均衡方法
CN104678351A (zh) 一种基于ZigBee技术的室内定位系统的算法
CN108495262B (zh) 室内空间泛在定位信号指纹库稀疏表征和匹配定位方法
Oh et al. WiFi Positioning by Optimal k-NN in 3GPP Indoor Office Environment

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22891837

Country of ref document: EP

Kind code of ref document: A1