CN112911500A

CN112911500A - High-precision indoor positioning method based on multi-source data fusion

Info

Publication number: CN112911500A
Application number: CN202110057989.XA
Authority: CN
Inventors: 李卓群; 李东新
Original assignee: Hohai University HHU
Current assignee: Hohai University HHU
Priority date: 2021-01-15
Filing date: 2021-01-15
Publication date: 2021-06-04

Abstract

The invention relates to a high-precision indoor positioning method based on multi-source data fusion. The method comprises the following steps: antenna delay correction is carried out on a positioning base station, TOF message data is collected by using UWB positioning ranging data receiving equipment, correlation analysis is carried out, and information with reference value is separated; performing NLOS characteristic identification on the acquired ranging data through channel estimation, separating NLOS ranging data, and performing reconstruction processing on the separated NLOS data through an EKF algorithm; and fusing the processed TOF ranging data with the signal intensity to obtain primary fusion data, and solving to obtain optimal matching coordinate information through a corresponding data fusion rule and the secondary fusion data obtained through the estimator 2.

Description

High-precision indoor positioning method based on multi-source data fusion

Technical Field

The invention relates to the field of communication, in particular to a high-precision indoor positioning method based on multi-source data fusion.

Background

The indoor positioning is a research direction with wide prospect, and has the characteristics of small area, serious multipath propagation, easy change of positioning environment and the like. For outdoor environments, there are currently satellite positioning (GPS, galileo, beidou, etc.) or mobile base station positioning.

However, for indoor environment, on one hand, satellite signals cannot penetrate buildings to lose their effect, on the other hand, the positioning accuracy of the mobile base station is too low to meet the indoor accuracy requirement, and in addition, the multipath effect and the inevitable interference caused by people walking have become the focus of research in the industry on how to reduce the errors caused by these interferences.

Disclosure of Invention

In order to solve the above problems in the prior art, the present invention provides the following technical solutions:

a high-precision indoor positioning method based on multi-source data fusion comprises the following steps:

performing antenna delay correction on the positioning base station;

collecting TOF message data by using UWB positioning ranging data receiving equipment, carrying out correlation analysis, and separating information with reference value;

performing NLOS characteristic identification on the acquired ranging data through channel estimation, separating NLOS ranging data, and performing reconstruction processing on the separated NLOS data through an EKF algorithm;

and fusing the processed TOF ranging data with the signal intensity to obtain primary fusion data, and solving to obtain optimal matching coordinate information through a corresponding data fusion rule and the secondary fusion data obtained through the estimator 2.

The UWB positioning ranging data receiving equipment comprises UWB positioning base stations, tags and ranging data receiving modules, wherein the main base station receives ranging TOF messages sent by each base station through a 5G-WIFI local area network, sends ranging data of each base station and the tags to the ranging data receiving modules through the Ethernet, and finally sends the received data messages to the cloud server through 4G modules in the ranging data receiving modules according to a certain data format and stores the data in a background database.

The invention has the beneficial effects that: the high-precision indoor positioning method based on multi-source data fusion can flexibly utilize the fields in the corresponding data processing programs. The method comprehensively utilizes relevant theories in the aspects of signal processing, mathematical statistics, bionics and the like, and obtains a new estimated value by fusing optimal linear data with the minimum mean square error as a standard, wherein the error size of the new estimated value is smaller than the estimation error of each component of the new estimated value. The data fusion is adopted to process a plurality of data or information, the data which is more effective, more reliable and more in line with the requirements of users are combined, the local and incomplete sensing quantities provided by a plurality of sensors which are distributed at different positions and in different states are organically integrated, the information complementation is utilized to reduce the uncertainty so as to form relatively comprehensive description of the external environment, and therefore the scientificity of decision and planning of an intelligent system and the correctness and rapidity of response are improved.

Drawings

The invention and its advantageous technical effects are explained in detail below with reference to the accompanying drawings and various embodiments, in which:

FIG. 1 is a flow chart of a high-precision indoor positioning method based on multi-source data fusion according to the present invention;

FIG. 2 is a schematic block diagram of a positioning base station distribution;

FIG. 3 is a schematic diagram of a data fusion process;

FIG. 4 is a plot of mean square error before and after data fusion using the present method.

Detailed Description

FIG. 1 is a flow chart of the high-precision indoor positioning method based on multi-source data fusion of the invention. As shown in fig. 1, the high-precision indoor positioning method based on multi-source data fusion of the present invention includes:

firstly, antenna delay correction is carried out on a positioning base station; therefore, the crystal oscillator frequency difference can meet the requirement of the project.

And then, using UWB positioning ranging data receiving equipment to collect TOF message data, carrying out correlation analysis and separating information with reference value.

And then performing NLOS characteristic identification on the acquired ranging data through channel estimation, separating NLOS ranging data, and performing reconstruction processing on the separated NLOS data through an EKF algorithm.

And finally, fusing the processed TOF ranging data with the signal intensity to obtain primary fusion data, and solving to obtain optimal matching coordinate information through corresponding data fusion rules and the secondary fusion data obtained by the estimator 2.

The positioning precision mean square error is reduced to reach the order of magnitude after the fusion and before the fusion are compared, and the speed and the precision can be flexibly adjusted by changing the number of the positioning base stations.

Fig. 2 is a schematic block diagram of positioning base station distribution, and fig. 3 is a schematic diagram of a data fusion process.

The embodiments and specific operation of the present invention are described in detail with reference to fig. 1 to 3.

Before the positioning method is executed, the pre-debugging parameter configuration work is firstly carried out.

First, a positioning base station is erected and its parameters are configured, as shown in fig. 2, which shows the overall layout of the positioning base station. In the project, an 8-base station positioning scheme is adopted, and because the three-dimensional coordinates of the tags are measured, the height of at least one base station is higher than that of other base stations. In order to ensure larger coverage area, the positioning base station is placed at the corner of the construction site as much as possible.

And further configuring relevant parameters of the positioning base station. Firstly, a 48V direct-current power supply is used for supplying power to a positioning base station, the positioning base station is connected with a PC end through a network cable, a browser is opened through double click, and the Web configuration page developed by a project group is entered. The main parameters of configuration are the configuration of relevant parameters such as IP address, wireless mode, ranging mode and gateway. The IP addresses of all the base stations are set to be 192.168.1.1-192.168.1.254, the wireless mode is divided into an access point mode and a working station mode, the positioning base station as a main base station is configured to be in the access point mode, and other base stations are configured to be in the working station mode. The TOF ranging mode is adopted in the project, so that the TOF mode needs to be selected in the system type. In the configuration option of the gateway, the IP address and the port number of the server need to be consistent with the configuration of the IP address and the port number set by the ethernet module, otherwise, data cannot be uploaded.

And (4) relevant parameters of the three-dimensional visualization software of the cloud server are configured. And respectively configuring other parameters such as engineering parameters, grouting layer blocking parameters, material parameters, vibration quality parameters, grouting quality parameters, three-dimensional visualization parameters and the like, and then opening a cloud server to wait for real-time vibration data. And uploading the data to a cloud server through a 4G network and adding the data into a database.

Back-end execution data fusion processing

Firstly, extracting positioning data stored in a database into a data queue, firstly, collecting a ranging TOF message of a UWB base station, and uploading the data to a cloud terminal according to the following mode. The realization of the upper computer software generation is completed by a Visual Studio compiler and runs in a Windows operating system. The main function of the upper computer is to receive UWB ranging data and complete resolving the label position. The cloud server receives data uploaded by front-end hardware and stores the data in a database according to a certain format, wherein the data format is as follows: $ <1>, <2>, < Ti >, < A1>, < D1>, < N1>, < A2>, < D2>, < N2>, … …, < An >, < Dn >, < Nn >, < END >, < N > and < N > respectively

Wherein:

the $ represents the initial identifier of the message;

<1> indicates a device number; <2> represents time;

< Ti > represents the number of the i-th tag;

< An > represents the number of the nth positioning base station;

< Dn > represents the ranging distance between the nth base station and the tag Ti, and the unit is millimeter;

< Nn > indicates the signal strength obtained by the time of ranging;

the cloud Server imports data into a database, and the upper computer software and the SQL Server database realize data interaction through an Entity Framework (Entity Framework). Setting a connection character string in an EntityFramework in a configuration file app.config, storing the calculated position coordinates of the label in a database, and establishing a data channel between an upper computer and the database. The method comprises the steps that collected data are sequenced according to signal intensity fields in a descending mode, a positioning engine reads the first five pieces of data regularly, a preliminary estimation value is solved by using a least square algorithm, iterative estimation is carried out by using a Taylor series expansion algorithm, the estimation value is iterated to a set threshold value, primary data fusion is completed at the moment, an estimator 1 in the graph represents a least square and Tayor series hybrid algorithm container, the method is equivalent to an interface function, and TOF distance information is transmitted to a function interface, so that a primary fusion estimation value can be solved. The estimator 2 is an extended kalman filter, and performs filtering processing on the primary fusion estimation value set to obtain a secondary fusion estimation value, and at this time, a final coordinate estimation value is obtained.

FIG. 4 is a plot of mean square error before and after data fusion using the present method. The data processing results before and after fusion are shown in FIG. 4, assuming d_TOFAnd N_ERespectively ranging distance and for signal strength of the TOF and finds a new estimate d.

d＝f(d_TOF,N_E)＝ad_TOF+bN_E

a+b≤1

The corresponding objective function is:

according to Lagrange multiplier, finally obtaining

Substituting the above expression into the final expression for solving d:

it can further be derived that:

it can be seen from the above derivation formula that the above data fusion method is to weight 2 variables, where the magnitudes of the corresponding weighting parameters and the standard deviations of the corresponding components are inversely proportional, i.e., the smaller the estimated error value, the larger the weighting amount, and it can be seen from fig. 4 that the error magnitude of the new estimated value obtained by the optimal linear data fusion with the minimum mean square error as the standard is smaller than the estimated error of each component. The positioning accuracy obtained by adding the signal intensity value is higher than that obtained by using a single TOF ranging distance, and the stability is also greatly improved. Therefore, the multi-base station data fusion has better application value.

It will be appreciated by those skilled in the art that the foregoing is only an embodiment of the present invention, and the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily made by those skilled in the art within the technical scope of the present invention will be covered by the present invention.

Claims

1. A high-precision indoor positioning method based on multi-source data fusion comprises the following steps:

performing antenna delay correction on the positioning base station;

2. The method of claim 1, wherein the UWB positioning and ranging data receiving device comprises UWB positioning base stations, tags and ranging data receiving modules, the main base station receives ranging TOF messages sent by each base station through a 5G-WIFI local area network, sends the ranging data of each base station and each tag to the ranging data receiving modules through an ethernet, and finally sends the received ranging data messages to the cloud server according to a certain data format through a 4G module in the ranging data receiving modules and stores the data in the background database.