CN113645565B - Indoor positioning method based on hexagonal closest packing structure - Google Patents
Indoor positioning method based on hexagonal closest packing structure Download PDFInfo
- Publication number
- CN113645565B CN113645565B CN202110771836.1A CN202110771836A CN113645565B CN 113645565 B CN113645565 B CN 113645565B CN 202110771836 A CN202110771836 A CN 202110771836A CN 113645565 B CN113645565 B CN 113645565B
- Authority
- CN
- China
- Prior art keywords
- equipment
- rssi
- beacon
- filtering
- positioning
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/023—Services making use of location information using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/021—Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/33—Services specially adapted for particular environments, situations or purposes for indoor environments, e.g. buildings
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
Abstract
The invention discloses an indoor positioning method based on a hexagonal closest packing structure, and belongs to the technical field of indoor positioning services. According to the method, the beacon equipment is arranged in the space to be positioned on the basis of the hexagonal closest-packed topological structure, and compared with a common simple cubic-packed topological structure, the space utilization rate of the beacon equipment and the space coverage rate of an effective signal range are greatly improved; the filtering method improves the traditional Kalman filtering method suitable for a linear Gaussian system, improves the receiving precision of the RSSI and reduces the interference of environmental factors on the precision of an indoor positioning system; by adopting the fingerprint database positioning combined with the arrangement of the beacon equipment and the filtering method, the problems of high environmental sensitivity, poor reusability and the like of the traditional fingerprint database are solved, and a higher indoor positioning target is realized.
Description
Technical Field
The invention belongs to the technical field of indoor positioning service, and particularly relates to an indoor positioning method based on a hexagonal closest packing structure.
Background
With the rapid development of modern communication technology and internet technology, people no longer meet the lagging positioning means which are used in the past and have low precision, more limited areas and poor ductility; in addition, with the continuous realization of the concept of "everything interconnection", various intelligent devices need to normally operate on the basis of acquiring user position information, and the demand of people for position information with higher precision is increasing day by day. Therefore, how to accurately and continuously provide real-time location information of users becomes one of the hot issues generally focused by people in this year.
High-precision GPS (Global Positioning System) and Positioning that can theoretically achieve outdoor centimeter-level precision, and Positioning precision has been practically controlled to within one meter. However, due to the problems of signal shielding, attenuation, reflection and the like, it is difficult for the GPS technology to achieve accurate positioning in indoor scenes, underground scenes and scenes with dense buildings.
For indoor positioning, the current common solutions are: infrared-based indoor positioning, wi-Fi-based indoor positioning, BLE (Bluetooth Low Energy) -based indoor positioning, visible light-based indoor positioning, and the like. In general, the essence of these indoor positioning methods is that positioning is performed based on RSSI (Received Signal Strength Indication) acquired by a device to be positioned.
Through search, the Chinese patent application numbers are: CN201710084372.0, application date is: the patent application of invention of 2017, 2 month and 16 days discloses an indoor positioning optimization algorithm based on WiFi, and the implementation process of the algorithm is divided into two stages: a training stage: collecting RSSI data offline, and converting the RSSI data into an RSSI fingerprint database; separating the RSSI fingerprint database from the x and y coordinate values; constructing a non-recurrent neural network continuously-trained RSSI value part to obtain a new data set and form a new RSSI fingerprint database; and (3) a testing stage: and (3) the RSSI fingerprint of a certain position is subjected to fingerprint library matching through a non-recurrent neural network which is the same as that in the training stage by a KNN classification algorithm to classify the x coordinate value and the y coordinate value respectively, so that a final positioning result is obtained. The method is suitable for the noise environment by increasing the scale of the data set by constructing a non-recursive neural network and continuously generating a new data set through the data set acquired at one time, but the method has high requirements on the size of the data set and can realize good effect only by needing the larger data set.
Also, for example, the Chinese patent application number is: cn201710164175.X, application date is: an invention patent of 3, month and 20 days in 2017 discloses a rapid KNN indoor WiFi positioning method based on random forests, and the method specifically comprises the following steps: dividing the positioning area into a plurality of sub-areas, and setting a plurality of positioning coordinate points in each sub-area; the method comprises the steps that a terminal collects RSSI fingerprint information and coordinate information of each coordinate point, the RSSI fingerprint information and the coordinate information are transmitted to a server through a wireless network, and a fingerprint database is built; the server judges the type of the area where the target is located through an integrated random forest algorithm; and matching by adopting a KNN algorithm according to the category of the target. The method adopts a KNN classification algorithm to construct a fingerprint database, the KNN algorithm needs to know the integral information of the data, and the effect is sensitive to the size of the data set.
Also, for example, the Chinese patent application number is: CN201711308570.7 filed as: the invention patent of 12 and 11 months in 2017 discloses an indoor positioning method based on a Bluetooth RSSI value position fingerprint database, which comprises the steps of firstly carrying out reasonable layout of Bluetooth base stations in a specific space, carrying out data acquisition on the Bluetooth base stations, preprocessing the acquired data and the data of points to be measured by means of Gaussian fitting screening, kalman filtering and the like, improving sampling precision, and simultaneously matching the information of the points to be measured with the position fingerprint database by a K-NNSS method to realize positioning. The Bluetooth base station layout mode is adopted, the space utilization rate is low, filtering is performed by adopting a Gaussian distribution filtering method, the filtering effect is general, the influence caused by the actual environment is not considered, and the fluctuation of the filtering result is large under different environments; the fingerprint library is constructed by using a K-NNSS method, and the K-NNSS algorithm needs to know the overall information of data, and the effect of the K-NNSS algorithm is sensitive to the size of a data set.
Also, for example, the Chinese patent application number is: CN201911210986.4 filed as: the invention patent of 12 months and 2 days in 2019 discloses a positioning method and system based on WiFi fusion prediction, and the method comprises the steps of collecting RSSI from each access point at a reference point; establishing an improved Kalman model, training the improved Kalman model in different environments, inputting the acquired data into the trained improved Kalman model, and filtering to obtain a fingerprint database; the mobile terminal acquires RSSI and performs filtering by using an improved Kalman model; and obtaining a positioning result of the WiFi positioning system according to the fingerprint database and the WKNN positioning algorithm based on the position constraint at the last moment. The method utilizes the neural network to train the covariance in the prediction process and the parameter of the measured noise covariance in the Kalman gain process so as to lead the covariance in the prediction process and the parameter of the measured noise covariance to be optimal, but the method solves the parameters of the covariance in the prediction process and the parameter of the measured noise covariance through theoretical calculation without considering the influence of actual environmental factors; and a KNN classification algorithm is used for constructing a fingerprint database, the overall information of the data needs to be known, and the effect of the KNN classification algorithm is sensitive to the size of the data set.
In summary, the prior art has the following disadvantages:
on one hand, most of the existing positioning systems adopt a space structure similar to a simple cube in the setting of the positioning device, and the space utilization rate of the space structure is low (about 39%), so that a user is difficult to acquire accurate RSSI in more areas in the space.
On the other hand, theoretically, according to the electromagnetic wave attenuation model in the free space, the RSSI acquired by the user using the signal receiving device should have a positive correlation with the distance between the user and the signal transmitting device. However, due to the problems of medium penetration attenuation, interference sources in the space, synchronous reception after reflection and the like, the RSSI acquired by the user by using the signal receiving equipment is inaccurate, so that the problems of low positioning accuracy, large floating of positioning results and the like are caused.
In addition, the traditional RSSI fingerprint database positioning system needs to manually traverse all areas of the environment to be positioned and acquire an RSSI vector group, the offline construction process consumes high labor cost, and the fingerprint database needs to be reconstructed after the environment to be positioned changes, the original fingerprint database can hardly be used in a new environment, and the reusability of the fingerprint database is poor.
Disclosure of Invention
Aiming at the problems that the space utilization rate of the space structure of the existing positioning system is low, an accurate RSSI value is difficult to obtain and the reusability of a fingerprint database is poor, the invention provides an indoor positioning method based on a hexagonal closest packing structure, and the effective space coverage rate in the positioning system is improved on the basis of the hexagonal closest packing structure; meanwhile, the method can carry out filtering processing on the RSSI value acquired by the equipment to be filtered under a specific environment condition, solves the problems of inaccurate and large floating of the acquired RSSI value, and realizes reusability of a fingerprint database positioning method on the basis of a hexagonal closest packing structure, thereby improving positioning precision.
In order to solve the above problems, the present invention adopts the following technical solutions.
An indoor positioning method based on a hexagonal closest packing structure comprises the following steps:
the first step,
Determining a maximum reception distance d for a beacon device r : in an environment to be positioned, repeatedly using a signal receiving device for multiple times to receive a signal sent by a beacon device, acquiring RSSI corresponding to the signal, and defining the maximum distance at which the signal receiving device can receive the signal sent by the beacon device and acquire the RSSI corresponding to the signal as a maximum receiving distance d r ；
Determining a maximum filtering distance d between signal receiving devices f : within the error range, the maximum distance between two signal receiving devices which meet the condition that the signals sent by the surrounding beacon devices are received and have the same attenuation degree is defined as the maximum filtering distance d f ；
Step two, determining the maximum construction distance: maximum construction distance d s ＝min{d r ,d f }；
Step three, arranging beacon equipment:
a1: according to the maximum construction distance d s The maximum effective coverage area of each beacon device is regarded as d, taking the beacon device as the sphere center and the radius s The sphere of (2); placing the spheres according to a hexagonal closest packing structure, and adjusting the positions of the beacon devices to enable the beacon devices to meet the topological structure relationship of the hexagonal closest packing, so as to obtain the primitive cells of the hexagonal closest packing structure formed by the beacon devices;
a2: according to the primitive cells obtained in the step A1, taking the geometric center of the hexagonal closest packed structure of the primitive cells as an origin, establishing a space coordinate system in an environment to be positioned, and periodically translating the primitive cells along three directions of x, y and z of the space coordinate system to ensure that the hexagonal closest packed structure of the primitive cells is densely distributed in the whole environment to be positioned, thereby finishing the arrangement of beacon equipment;
a3: combining the maximum construction distance d according to the space coordinate system established in the step A2 s And the geometric property of the hexagonal closest packing structure, calculating the absolute position coordinates of each beacon device in the space coordinate system, and storing the absolute position coordinates in the server;
step four, fingerprint library positioning: comprising an off-line construction phase and an on-line positioning phase, wherein,
the off-line construction phase comprises: using maximum constructional distance d s The beacon equipment arrangement method in the step three is adopted to complete the arrangement of the beacon equipment in the environment to be positioned; receiving signals sent by beacon equipment in a hexagonal closest packing structure by using signal receiving equipment, and acquiring an RSSI vector group r off For RSSI vector set r off Each RSSI vector component in the RSSI vector is filtered; repeatedly acquiring RSSI vector group r corresponding to different positions for multiple times off Constructing a training set; processing the training set by using a naive Bayes algorithm to construct a classifier;
the on-line positioning stage comprises: after a user enters an environment to be positioned, receiving signals sent by surrounding beacon equipment by using signal receiving equipment, acquiring RSSI (received signal strength indicator) corresponding to the signals, and selecting the beacon equipment corresponding to the RSSI with the minimum absolute value as positioning reference equipment; for RSSI vector set r on Each RSSI vector component in the RSSI vector group is filtered, and the processed RSSI vector group r is processed on Inputting the data into a classifier constructed in an offline construction stage to obtain the relative position coordinates of the user and the positioning reference equipment; and calculating the absolute position coordinate of the user by combining the absolute position coordinate of the positioning reference equipment to finish positioning.
Further, in step four, the filtering method includes:
defining a filtering reference device and a device to be filtered: at a distance equal to or less than the maximum filtering distance d f If one of the two signal receiving devices acquires the theoretical value and variance of RSSI of the surrounding beacon deviceIf so, the signal receiving equipment can be used as filtering reference equipment, and the other signal receiving equipment can be used as equipment to be filtered; establishing communication connection between filtering reference equipment and equipment to be filtered; and the filtering reference equipment transmits the theoretical value and the variance of the RSSI of the surrounding beacon equipment to the equipment to be filtered, and the equipment to be filtered uses the theoretical value and the variance of the RSSI transmitted by the filtering reference equipment to finish filtering based on a Kalman filtering method.
Further, the specific filtering step includes:
b1: the prediction equation of the filtering reference device is as follows:
wherein the content of the first and second substances,
predicted value X of RSSI obtained by the k filtering reference equipment i For filtering results of RSSI obtained by the ith filtering reference device,
for the variance of the filtering result of the RSSI acquired by the filtering reference device at the i-th time, when k =1,
a measurement value equal to the first time;
prediction value of filtering reference device
The variance of (c) is:
the filtering reference equipment sends the predicted value and the predicted value variance to the equipment to be filtered;
b2: the predicted noise variance of the device to be filtered is:
b3: the measurement equation of the device to be filtered is as follows:
Z k ＝H k X k +v k ，
wherein, Z k Measurement value H of RSSI obtained for equipment to be filtered k For measuring the noise drive matrix, X k For this filtering result, v k To measure noise;
b4: the measured noise variance of the device to be filtered is R k (ii) a The filtering gain of the equipment to be filtered is as follows:
G k ＝P k H k T (H k P k H k T +R k ) -1 ；
b5: the filtering result of the equipment to be filtered for the obtained RSSI at this time is as follows:
b6: the filtering of the RSSI acquired by the equipment to be filtered can be completed by iterating the process for multiple times in one filtering period.
Further, in step four, the specific steps of the offline construction stage include:
c1: according to the maximum construction distance d s Finishing the arrangement of the beacon equipment in the environment to be positioned;
c2: repeatedly measuring for multiple times in an environment to be positioned, selecting origin equipment, and taking beacon equipment positioned on 12 coordination numbers with the origin equipment as a central atom in a hexagonal closest packing structure as coordination equipment; establishing fingerprint library primitive cells by taking the original point equipment as an original point and combining 12 coordination equipment;
c3: creating an empty training data set S, wherein the data format in the S is bidirectional mapping from a three-dimensional vector group to a twelve-dimensional vector group; wherein the three-dimensional vector represents a position coordinate vector, and the twelve-dimensional vector group represents an RSSI vector group;
c4: taking the origin device as the center of sphere and the maximum structure distance d s Randomly selecting a point in a spherical range of a radius, recording a position coordinate vector of the point, and recording the position coordinate vector as P: (x, y, z);
c5: at the position coordinate P: at the position (x, y, z), using a signal receiving device to receive signals sent by all coordination devices in the primitive cell of the fingerprint library, acquiring the corresponding RSSI of the signals, and recording all the acquired RSSI as a vector group r off ＝(r 1 ,r 2 ,r 3 ……r 12 ) (ii) a In the process, the origin equipment is used as filtering reference equipment, the signal receiving equipment is used as equipment to be filtered, and the RSSI vector group r acquired by the signal receiving equipment is subjected to filtering off Each RSSI vector component in the RSSI vector is subjected to filtering processing;
c6: according to the RSSI vector group r obtained in the step C5 off Constructing a set r of slave RSSI vectors off Bidirectional mapping relation to position coordinate vector
Adding the bidirectional mapping relation into a training data set S;
c7: repeating the off-line construction steps C4-C6 for multiple times to obtain a training data set S ′ And for the training data set S ′ Carrying out deduplication processing on the data in the step (1);
c8: from the training data set S obtained in step C7 ′ Constructing a classifier by using a naive Bayes algorithm;
the specific steps of the on-line positioning stage comprise:
d1: after a user enters an environment to be positioned, using signal receiving equipment to receive signals sent by surrounding beacon equipment, acquiring RSSI (received signal strength indicator) corresponding to the signals, and selecting the beacon equipment corresponding to the RSSI with the minimum absolute value as positioning reference equipment;
d2: according to the fingerprint library primitive cells in the off-line construction stage step C2, the positioning reference equipment is used as origin equipment of the fingerprint library primitive cells, 12 beacon equipment which is positioned in the hexagonal closest packing structure and has 12 coordination numbers with the positioning reference equipment as a central atom are used as coordination equipment, the origin equipment is used as an origin, and all 12 coordination equipment are combined to establish the fingerprint library primitive cells;
d3: according to the fingerprint library primitive cell obtained in the step C2, a user receives signals sent by all coordination devices in the fingerprint library primitive cell by using signal receiving equipment, acquires RSSI corresponding to the signals, and records all acquired RSSI as a vector group r on ＝(r 1 ,r 2 ,r 3 ……r 12 ) (ii) a In the process, the positioning reference equipment is used as filtering reference equipment, the signal receiving equipment is used as equipment to be filtered, and the RSSI vector group r acquired by the signal receiving equipment is subjected to filtering on Each RSSI vector component in the RSSI vector is filtered;
d4: the RSSI vector group r after filtering processing on Inputting the relative position coordinate vector of the user and the positioning reference equipment into a Bayes classifier constructed in the step C8 of the off-line construction stage, and recording the relative position coordinate vector as a vector p = (x, y, z);
d5: according to the position coordinate vector p = (x, y, z) obtained in the step D4 of the online positioning stage, the position coordinate vector p and the absolute position coordinate (x) of the origin device are compared b ,y b ,z b ) Adding up to obtain the absolute position coordinate (x) of the user u ,y u ,z u ) And finishing positioning.
Further, in the step C2, the selection criteria of the origin device are as follows:
in an environment to be located, at a distance d from a beacon device s Respectively receiving the RSSI of each beacon device by using a signal receiving device, calculating the mean value of the RSSI of each beacon device and calculating the variance of the mean value; repeating the process for multiple times, calculating the mean value of the RSSI variance of each beacon device, and selecting the beacon device corresponding to the minimum mean value as the original point device; if the minimum mean value corresponds to a plurality of beacon devices, performing secondary sequencing on the plurality of beacon devices corresponding to the minimum mean value;
the sub-ordering specifically operates as: and measuring the distance between each beacon device corresponding to the minimum mean value and the geometric center of the hexagonal closest packing structure, and selecting the beacon device corresponding to the minimum distance as the origin device.
Compared with the prior art, the invention has the beneficial effects that:
(1) According to the indoor positioning method based on the hexagonal closest packing structure, the beacon equipment is arranged by using the topological structure with higher space utilization rate, namely the hexagonal closest packing structure, so that the coverage rate of the beacon equipment is improved, and meanwhile, a structural beacon equipment arrangement method basis is provided for fingerprint library positioning.
(2) According to the indoor positioning method based on the hexagonal closest packing structure, the improved Kalman filtering algorithm can be used for completing filtering processing on a nonlinear signal system under a specific environment condition, the RSSI acquisition precision of signal receiving equipment is improved, the positioning precision of the positioning algorithm based on the RSSI is further improved, and meanwhile, a filtering method foundation is provided for fingerprint database positioning.
(3) Compared with the traditional fingerprint database positioning method, the fingerprint database positioning method has the following two advantages: firstly, the data characteristic dimension of the fingerprint database is higher, so that the positioning precision is improved; secondly, due to the topological characteristic of the hexagonal closest packing structure in the space, each beacon device can serve as an origin device in an original cell of the fingerprint library to provide positioning reference information, and can also serve as a coordination device to send signals to a signal receiving device to obtain RSSI (received signal strength indicator), a user can use any beacon device in the space as the positioning reference device in an environment to be positioned, and only the positioning reference device and the coordination device actually participate in positioning, so that the reusability of the fingerprint library is realized.
Drawings
FIG. 1 is a flow chart of the beaconing device of the present invention;
FIG. 2 is a flow chart of RSSI filtering according to the present invention;
FIG. 3 is a flow chart of the fingerprint database positioning offline stage according to the present invention;
FIG. 4 is a flow chart of the fingerprint repository location online phase of the present invention;
FIG. 5 is a general flowchart of an indoor positioning method of the present invention;
FIG. 6 is a schematic diagram of a cell of the hexagonal closest packing structure of the present invention;
FIG. 7 is a schematic representation of the coordination number of the hexagonal closest packing structure of the present invention;
FIG. 8 is a schematic diagram of the translation in space of hexagonal closest-packed cells according to the present invention;
FIG. 9 is a schematic view of a single layer of hexagonal closest-packed structure of the present invention densely packed in space;
FIG. 10 is a schematic diagram of a spatial rectangular coordinate system constructed by the fingerprint database positioning method according to the present invention;
FIG. 11 is a diagram illustrating an exemplary fingerprint database locating method according to the present invention;
FIG. 12 is a diagram illustrating an example of a fingerprint database locating method according to the present invention.
Detailed Description
The invention is further described with reference to specific embodiments and the accompanying drawings.
Examples
The embodiment provides an indoor positioning method based on a hexagonal closest packing structure, as shown in fig. 5, which mainly includes three methods: the beacon device positioning method based on the hexagonal closest packing structure, the RSSI filtering method based on Kalman filtering and the reusable fingerprint database positioning method based on the two methods are adopted. The beacon equipment setting method is based on a hexagonal closest-packed topological structure and is set in a space to be positioned, and compared with a common simple legislative-packed topological structure, the space utilization rate of beacon equipment and the space coverage rate of an effective signal range are greatly improved; the filtering method is based on a Kalman filtering method suitable for a linear Gaussian system, and realizes a filtering method aiming at a nonlinear system, so that the receiving precision of RSSI is improved, and the interference of environmental factors on the precision of an indoor positioning system is reduced; by combining the beacon equipment setting method, the invention realizes a reusable fingerprint database positioning method, solves the problems of high environmental sensitivity, poor reusability and the like of the traditional fingerprint database, and realizes a higher indoor positioning target.
The beacon equipment setting method based on the hexagonal closest packing structure specifically comprises the following steps:
hardware devices used in an indoor positioning system are mainly classified into two types: a signal transmitting apparatus and a signal receiving apparatus. We usually pre-arrange the signalling device as a beacon device in the environment to be located; when positioning is needed, the signal receiving equipment is used for acquiring signals and related information, such as RSSI (received signal strength indicator) values, sent by surrounding beacon equipment, and positioning is further completed by combining a related positioning algorithm.
The electromagnetic wave radiation model of the beacon device (signal transmission device) is a sphere with the beacon device as the center of the sphere, and under ideal conditions, signal receiving devices located within the sphere can accurately and effectively receive signals sent by the beacon device and acquire the RSSI corresponding to the signals. However, due to the influence of real-world factors, generally, the farther a signal receiving device is from a beacon device, the poorer the accuracy of the RSSI acquired by the signal receiving device is. And in the received error range, defining the maximum distance, at which the signal receiving equipment can accurately and effectively receive the signal sent by the beacon equipment and acquire the RSSI corresponding to the signal, as the maximum receiving distance. Therefore, the maximum range (i.e., the maximum effective coverage range of the beacon device) in which the signal receiving device can accurately and effectively receive the signal sent by the beacon device and acquire the RSSI corresponding to the signal can be regarded as a sphere with the beacon device as the center of the sphere and the maximum receiving distance as the radius. How to reasonably arrange beacon devices in an environment to be positioned so that the space coverage and the space utilization rate of the beacon devices are as large as possible becomes a problem to be researched. To solve this problem, this embodiment provides an indoor positioning system based on a hexagonal closest-packed structure, and as shown in fig. 1, the specific process of the beacon device setting method is as follows:
a1: defining a maximum reception distance d r : according to the physical characteristics of the beacon equipment and the environmental characteristics of the environment to be positioned, in the environment to be positioned, the signal receiving equipment is repeatedly used for multiple times to receive the signal sent by the beacon equipment, the RSSI corresponding to the signal is obtained, and the signal receiving equipment can accurately and effectively receive the signal sent by the beacon equipment and obtain the RSS corresponding to the signalThe maximum distance of I is defined as the maximum receiving distance d r 。
A2: according to the maximum receiving distance d obtained in A1 r The maximum effective coverage area of each beacon device can be regarded as the sphere center of the beacon device and the radius d r The sphere of (2). Placing the spheres in a hexagonal closest packing configuration, as shown in fig. 6; and the positions of the beacon devices are appropriately adjusted so that the beacon devices satisfy the topological relation of hexagonal closest packing, and as shown in fig. 7, the cells of hexagonal closest packing formed by the beacon devices are obtained.
A3: and (3) establishing a space coordinate system in the environment to be positioned according to the primitive cells obtained in the step A2, and periodically translating the primitive cells along three directions of x, y and z shown in the figure, so that the hexagonal closest-packed structure of the primitive cells is densely distributed in the whole environment to be positioned (forms a crystal structure similar to that in solid physics), and the characteristic that the topological structure of the beacon device of the whole environment to be positioned can be summarized by using the characteristic of the primitive cells of the beacon device is met, namely, the setting of the beacon device is completed. Wherein the hexagonal closest-packed monolayer structure is shown in figure 9.
A4: according to the space coordinate system established in A3, combining the maximum receiving distance d r And the geometric properties of the hexagonal closest packing structure, the coordinates of each beacon device in the space coordinate system are calculated and stored in the server side (for later use in combination with a fingerprint library positioning algorithm).
The beacon device setting method simulates a simulation process from a primitive cell to a crystal in solid matters, and realizes the setting of the beacon device based on a hexagonal closest packing structure. The indoor positioning system is different from a common simple cubic packing structure in the indoor positioning system, and the overall space utilization rate of the beacon equipment in the hexagonal closest packing structure is higher. Meanwhile, from the coordination number, the coordination number of the beacon device in the simple cubic structure is 6, and the coordination number of the beacon device in the hexagonal closest packing structure is 12, so that more accurate and effective RSSI data (the coordination number is the number of the beacon devices around the beacon device and closest to the beacon device) are provided for the use of an indoor positioning algorithm.
For most indoor positioning systems based on RSSI, the positioning algorithm is relatively perfect, and higher positioning accuracy can be ensured under ideal conditions. However, due to the fact that signals penetrate through medium attenuation, interference sources exist in space, synchronous receiving is achieved after reflection, and the like, the RSSI acquired by the signal receiving equipment is not accurate enough, and therefore the positioning accuracy is low. In order to solve the problem, a filtering method may be used, that is, the RSSI obtained by the signal receiving device is filtered to obtain a more accurate RSSI, and then the RSSI is input into a positioning algorithm for positioning.
For a linear gaussian system, the kalman filter algorithm is a common and effective filtering method, and is essentially a high-efficiency iterative filter, which can estimate the state of a dynamic system from a series of measurements that do not completely contain noise. But for non-linear signal systems or non-gaussian noise systems, such classical kalman filtering is very inefficient. In practical application scenarios, the signal to be filtered is often a more complex nonlinear system; due to the diversity of signals in the space and the complexity of the actual scene space, the noise generated in the signal measurement process is often not simple gaussian noise. The two problems result in that the classic Kalman filtering algorithm is difficult to be directly applied to the actual environment, and in order to solve the problems, the invention provides the RSSI filtering method based on Kalman filtering.
First, the following background and definitions are given:
in a positioning system, the degree of attenuation of signals received by two closely spaced signal receiving devices into the surrounding environment should be the same within an acceptable error range. At this time, if the absolute position of one of the signal receiving devices in the positioning system is known, the RSSI obtained by the positioning system can be obtained by the RSSI ranging formula: r = a log d + b;
wherein r is RSSI obtained by the signal receiving equipment, d is the distance between the signal receiving equipment and the beacon equipment, a is a correction coefficient related to the receiving sensitivity and the antenna gain of the signal receiving end, and b is the signal sending in the environmentThe transmitting end and the receiving end are separated by one meter of signal strength. a. b is affected by actual environmental conditions and can be obtained by repeated measurements. Through repeated measurement, the variance sigma of the signal received by the signal receiving equipment under the current environment condition can be obtained 2 . Therefore, the signal model received by the signal receiving equipment is the mathematical expectation r and the variance σ 2 The gaussian model of (1).
If the theoretical value and variance of the RSSI of the surrounding beacon devices acquired by one of the signal receiving devices are known, the signal receiving device can be used as a filtering reference device, and the other signal receiving device can be used as a device to be filtered. Under the condition that the distance between the filtering reference device and the device to be filtered is close, the attenuation degrees of the ambient signals received by the filtering reference device and the device to be filtered are the same within the acceptable error range, so that the signal model received by the known filtering reference device can be used as the signal prediction model of the device to be filtered, and the filtering method based on Kalman filtering is performed by combining the signal measurement model of the device to be filtered.
As shown in fig. 2, the specific process of the filtering method is as follows:
firstly, in an actual environment, through multiple measurements, a maximum effective distance satisfying a condition that the attenuation degrees of ambient signals received by a filtering reference device and a device to be filtered are the same within an accepted error range is determined, and the distance is defined as a maximum filtering distance and is recorded as d f 。
B1: for the filtering reference device, the prediction equation of the filtering reference device is as follows:
wherein the content of the first and second substances,
predicted value X of RSSI obtained by the k filtering reference equipment i For filtering results of RSSI obtained by the ith filtering reference device,
for the variance of the filtering result of the RSSI obtained by the i-th filtering reference device, when k =1,
a measurement value equal to the first time; and obtaining the RSSI predicted value of the filtering reference equipment by adopting an inverse variance weighted average method. The method of weighted average through inverse variance can ensure the predicted value
The variance of (a) is minimized, thereby making the predicted value closer to the theoretical value.
Prediction value of filtering reference device
The variance of (c) is:
as stated by way of background and by definition, the signal prediction model of the device to be filtered can be replaced by the signal prediction model of the filtering reference device.
The filtering reference equipment and the equipment to be filtered establish communication connection through a low-power Bluetooth (BLE) technology, and the predicted value and the variance of the current time are sent to the equipment to be filtered, so that the prediction equation of the equipment to be filtered is the prediction equation of the filtering reference equipment and the predicted value of the equipment to be filtered
The variance of (1) is a predicted value of a filtering reference device
The variance of (c).
B2: the predicted noise variance of the device to be filtered is:
b3: for the device to be filtered, the measurement equation of the device to be filtered is as follows:
Z k ＝H k X k +v k ，
wherein Z is k Measurement value H of RSSI obtained for equipment to be filtered k For measuring the noise drive matrix, X k For the result of this filtering, v k To measure noise; the measurement equation of the device to be filtered reflects the signal scalability. Wherein the measurement noise v k The larger the measurement, the more inaccurate the measurement is; otherwise, the noise v is measured k The smaller the measurement, the more accurate the measurement.
B4: the measured noise variance of the device to be filtered is R k (ii) a Due to measurement noise v in the measurement equation of the device to be filtered k Is obtained by a plurality of measurements, and thus the measurement noise variance R can be obtained k 。
B5: according to the predicted noise variance and the measured noise variance of the equipment to be filtered (the predicted noise variance and the measured noise variance of the equipment to be filtered are subjected to inverse variance weighted average), the filtering income G of the equipment to be filtered at this time can be obtained k Comprises the following steps:
G k ＝P k H k T (H k P k H k T +R k ) -1 ；
b6: the filtering result of the equipment to be filtered for the obtained RSSI at this time is as follows:
b7: the filtering of the RSSI acquired by the equipment to be filtered can be completed by iterating the process for multiple times in one filtering period.
The filtering method is similar to the classic Kalman filtering method in flow, and is different in that the variance of a prediction equation of filtering reference equipment is used for replacing a prediction variance obtained by iteration in the classic Kalman filtering method, and a nonlinear signal prediction model of equipment to be filtered is essentially replaced by a linear Gaussian signal prediction model of the filtering reference equipment, so that the filtering process is simplified, the application range of the classic Kalman filtering and the derivation method thereof is enlarged, and the iteration property of a Kalman filter is not changed.
The fingerprint library positioning method is a positioning algorithm commonly used in an indoor positioning system, and the principle is that signal characteristic information (signal cluster) acquired by a position to be positioned is compared with signal characteristic information prestored in a fingerprint library, the most similar signal characteristic information is found, and position information mapped by the signal characteristic information is found, so that positioning is completed.
The fingerprint database positioning method has the advantages that the sensitivity degree to the position information is high, so that the positioning with higher precision can be realized; on the contrary, because of its high degree of dependence on the environment, when the surrounding environment changes, the original fingerprint database cannot be applied to the new environment, and the fingerprint database needs to be reconstructed. This results in that although the traditional fingerprint database algorithm has a high positioning accuracy, it is still tedious, and requires a lot of manpower to construct and maintain. In order to solve the above problems, the present invention provides a reusable fingerprint database positioning method by combining the beacon device setting method based on the hexagonal closest packing structure and the RSSI filtering method based on the kalman filter mentioned above.
The specific process of the fingerprint database positioning method is as follows: the fingerprint database positioning is divided into two stages, an off-line construction stage and an on-line positioning stage, wherein,
as shown in fig. 3, the specific steps of the offline construction phase include:
c1: maximum reception distance d obtained according to beacon device setting method based on hexagonal closest-packed structure r And the maximum filtering distance d obtained by the RSSI filtering method based on Kalman filtering f (in general, d r >d f ) Defining a maximum construction distance d s ，d s ＝min{d r ,d f Will construct the maximum distance d s As the radius of the maximum effective coverage range of the beacon equipment, completing the environment to be positionedArrangement of the signal transmission device. Therefore, the spatial layout of the beacon equipment can be ensured to meet the topological structure of hexagonal closest packing, and the filtering requirement in the RSSI filtering method based on Kalman filtering is also met.
C2: and selecting a beacon device with stronger environment characteristic representativeness from the beacons in the environment to be positioned by repeatedly measuring in the environment to be positioned for a plurality of times, and defining the beacon device as an origin device. And establishing the primitive cell of the fingerprint library by taking the original point device as an original point and combining all beacon devices belonging to the coordination number of the original point device in the hexagonal closest packing structure. In the hexagonal closest packing structure, the coordinating devices are 12 beacon devices in 12 coordination numbers with the origin device as the central atom.
The origin equipment is selected according to the following criteria:
in an environment to be located, at a distance from the beacon device d s Respectively receiving the RSSI of each beacon device by using a signal receiving device, calculating the mean value of the RSSI of each beacon device and calculating the variance of the mean value; repeating the process for multiple times, calculating the mean value of the RSSI variance of each beacon device, and selecting the beacon device corresponding to the minimum mean value as the original point device; if the minimum mean value corresponds to a plurality of beacon devices, performing secondary sequencing on the plurality of beacon devices corresponding to the minimum mean value;
the sub-sequence is specifically operated as: and measuring the distance between each beacon device corresponding to the minimum mean value and the geometric center of the hexagonal closest packing structure, and selecting the beacon device corresponding to the minimum distance as the origin device.
To explain specifically by using the figure, the spatial structure and the spatial rectangular coordinate system of the primitive cells of the fingerprint library are shown in fig. 10, a sphere 1 is an origin device, spheres 2 to 13 are beacon devices belonging to the coordination number of the origin device in the hexagonal closest packing structure, the beacon devices are defined as coordination devices, and the radii of the spheres 1 to 13 are d s 。
The geometry of spheres 1 to 13 in FIG. 10 is the same as the geometry of spheres 1 to 13 in FIG. 6, i.e. the geometry of the coordination number of the hexagonal closest packing structure. In order to show the geometrical relationship between the origin device and the coordination device more clearly, the positions of the spheres 1 and 2-13 are shown as being separated from each other. However, due to the nature of the hexagonal closest packing structure, sphere 1 is actually tangent to each of spheres No. 2 to 13, i.e., a structure satisfying the coordination number in hexagonal closest packing that is tangent between the origin device and the coordination device.
C3: creating an empty training data set S, wherein the data format in the S is bidirectional mapping from a three-dimensional vector group to a twelve-dimensional vector group; wherein the three-dimensional vector represents a position coordinate vector and the twelve-dimensional vector group represents an RSSI vector group.
C4: taking the origin device as the center of sphere and the maximum structure distance d s Randomly selecting a point in a spherical range of a radius, recording a position coordinate vector of the point, and recording the position coordinate vector as P: (x, y, z).
C5: at the position coordinate P: at the position of (x, y, z), using a signal receiving device to receive signals sent by all coordination devices in the primitive cell of the fingerprint library, acquiring the corresponding RSSI of the signals, and recording the RSSI as a vector group r off ＝(r 1 ,r 2 ,r 3 ……r 12 ) (ii) a In the process, the origin equipment is used as filtering reference equipment, the signal receiving equipment is used as equipment to be filtered, and the RSSI vector group r acquired by the signal receiving equipment is filtered by using the RSSI filtering method based on Kalman filtering off Each RSSI vector component in (1) is filtered.
C6: according to the RSSI vector group r obtained in the step C5 off Constructing a set r of slave RSSI vectors off Bidirectional mapping relation to position coordinate vector
And adds the bi-directional mapping to the training data set S.
C7: repeating the off-line construction steps C4-C6 for multiple times to obtain a training data set S ′ And for the training data set S ′ The data in (2) is processed for deduplication.
C8: from the training data set S obtained in step C7 ′ The classifier is constructed using a naive bayes algorithm.
It is worth mentioning that, unlike a common fingerprint database positioning algorithm, the primitive cell fingerprint database constructed in the offline stage of the fingerprint database positioning method is a relative fingerprint database, that is, a three-dimensional coordinate vector corresponding to each RSSI vector group in the primitive cell fingerprint database is not an absolute position coordinate in an environment to be positioned, but a position coordinate vector relative to an origin device in a primitive cell of the fingerprint database. And adding the position coordinate vector and the absolute position coordinate of the origin equipment to obtain the position coordinate which needs to be finally positioned.
As shown in fig. 4, the specific steps of the online positioning stage include:
d1: after a user enters an environment to be positioned, a signal receiving device is used for receiving signals sent by surrounding signal sending devices, RSSI (received signal strength indicator) corresponding to the signals is obtained, and the signal sending device corresponding to the RSSI with the minimum absolute value is selected as positioning reference device.
D2: according to the definition in the step C2 of the off-line construction phase, the positioning reference device is used as an origin device of the primitive cell of the fingerprint library, 12 beacon devices with the coordination numbers of 12 located in the hexagonal closest packing structure and with the positioning reference device as a central atom are used as coordination devices, the origin device is used as an origin, and all 12 coordination devices are combined to establish the primitive cell of the fingerprint library.
D3: according to the fingerprint library primitive cell obtained in the step, a user receives signals sent by all coordination devices in the fingerprint library primitive cell by using signal receiving equipment, acquires RSSI corresponding to the signals and records the RSSI as a vector group r on ＝(r 1 ,r 2 ,r 3 ……r 12 ). In the process, the positioning reference equipment is used as filtering reference equipment, the signal receiving equipment is used as equipment to be filtered, and an RSSI vector group r acquired by the signal receiving equipment is filtered by using an RSSI filtering method based on Kalman filtering on Each RSSI vector component in (1) is filtered.
D4: the RSSI vector group r after filtering processing on Inputting the position coordinate vector into a Bayes classifier constructed in the step C8 of the off-line construction stage, outputting a group of position coordinate vectors which are recorded as vectors p = (x, y, z), wherein the vector group is the position of the user relative to the positioning reference equipmentAnd (4) coordinates.
D5: according to the position coordinate vector p = (x, y, z) obtained in the step D4 of the online positioning stage, the position coordinate vector p = (x, y, z) and the absolute position coordinate (x) of the positioning reference device are obtained b ,y b ,z b ) Adding up to obtain the absolute position coordinate (x) of the user u ,y u ,z u ) And finishing positioning. Wherein the absolute position coordinates (x) of the reference device are located b ,y b ,z b ) In the off-line construction stage, the beacon equipment is stored to a server side in the process of finishing the setting of the beacon equipment in an environment to be positioned according to a beacon equipment setting method based on a hexagonal closest packing structure, and the beacon equipment is directly read at the server side according to a major and a minor (or other identity codes) acquired from a broadcast packet sent by positioning reference equipment.
The specific implementation scheme is as follows:
the method comprises the following steps: after entering an environment to be positioned, a user uses signal receiving equipment to receive signals sent by surrounding beacon equipment and acquires RSSI corresponding to the signals. As shown in fig. 11, spheres 1 to 13 represent beacon devices that have been arranged in an environment to be located, sphere 14 represents a user and a signal receiving device used by the user, and after the user acquires RSSI corresponding to surrounding beacon devices using the signal receiving device, the RSSI corresponding to the beacon device located at sphere 1 in the figure is found to be minimum. The user therefore defines the beacon device located at the sphere 1 as the positioning reference device.
Step two: as shown in fig. 12, after acquiring the RSSI corresponding to the surrounding beacon devices, the user finds that the RSSI corresponding to the beacon device located at the sphere 1 in the figure is minimum. Therefore, the user defines the beacon device located at the sphere 1 in the figure as a positioning reference device, uses the positioning reference device as an origin, uses 12 beacon devices located at 12 coordination numbers of the hexagonal closest packing structure with the positioning reference device as a central atom as coordination devices (i.e., beacon devices located at the spheres 2 to 13), and uses the positioning reference device as an origin to construct a fingerprint library primitive by combining the 12 coordination devices.
Step three: according to the fingerprint library primitive cell obtained in the above-mentioned step, user uses signal receiving equipment to receive all coordination in fingerprint library primitive cellThe signals from the devices (i.e., the beacon devices located at spheres 2-13 in fig. 12) are acquired and the corresponding RSSI of these signals is recorded as vector set r on ＝(r 1 ,r 2 ,r 3 ……r 12 ). In the process, the positioning reference equipment is used as filtering reference equipment, the signal receiving equipment is used as equipment to be filtered, and the RSSI vector group r acquired by the signal receiving equipment is subjected to RSSI filtering method based on Kalman filtering on Each RSSI vector component in (1) is filtered.
Step four: according to RSSI vector group r obtained in the online positioning stage on ＝(r 1 ,r 2 ,r 3 ……r 12 ) Inputting the position coordinate vector into a Bayes classifier constructed in an offline construction stage to obtain a group of position coordinate vectors output by a fingerprint library, and recording the position coordinate vectors as a vector p = (x, y, z), wherein the vector group is the position coordinate of the user relative to the positioning reference equipment.
Step five: according to the position coordinate vector p = (x, y, z) obtained in the online positioning stage, the position coordinate vector p = (x, y, z) and the absolute position coordinate (x) of the positioning reference equipment are compared b ,y b ,z b ) Adding up to obtain the absolute position coordinate (x) of the user u ,y u ,z u ) And finishing positioning. Assuming that the absolute position coordinate of the positioning reference device (i.e., the beacon device located at the sphere 1 in fig. 12) in the rectangular coordinate system of the environment space to be positioned is (1,1,1), according to the position coordinate vector p = (0,1,0) obtained in the online positioning stage step D4, the absolute position coordinate of the user in the rectangular coordinate system of the environment space to be positioned is (1,2,1). Wherein the absolute position coordinates (x) of the reference device are located b ,y b ,z b ) And step one, the identification codes are stored in the server side, and the identification codes are directly read at the server side according to the major and minor (or other identification codes) acquired from the broadcast packet sent by the positioning reference equipment.
The flow chart of the fingerprint database positioning method is shown in fig. 3 and 4, and reusability of the fingerprint database positioning method is realized on the basis of a beacon device setting method based on a hexagonal closest packing structure and an RSSI filtering method based on Kalman filtering: because the hexagonal closest packing structure in the environment to be positioned is obtained by translation and close distribution of cells in the x, y and z directions, any beacon device has a coordination structure (except an environment boundary) shown in fig. 7, any beacon device can be set as a positioning reference device, positioning is completed according to the flow of the online positioning stage, and reusability of a fingerprint library positioning method is realized; meanwhile, in a hexagonal closest packing structure, the coordination number of each beacon device is 12, which means that each three-dimensional coordinate can be mapped to a vector set which is at least twelve-dimensional when a fingerprint library is constructed, so that the prediction effect when a naive Bayes classifier is used is improved, and the positioning accuracy of a fingerprint library positioning method is improved.
The examples described herein are merely illustrative of the preferred embodiments of the present invention and do not limit the spirit and scope of the present invention, and various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the design concept of the present invention shall fall within the protection scope of the present invention.
Claims (5)
1. An indoor positioning method based on a hexagonal closest packing structure is characterized in that: the method comprises the following steps:
the first step,
Determining a maximum reception distance d for a beacon device r : in an environment to be positioned, repeatedly using a signal receiving device for multiple times to receive a signal sent by a beacon device, acquiring RSSI corresponding to the signal, and defining the maximum distance at which the signal receiving device can receive the signal sent by the beacon device and acquire the RSSI corresponding to the signal as a maximum receiving distance d r ；
Determining a maximum filtering distance d between signal receiving devices f : within the error range, the maximum distance between two signal receiving devices which satisfy the same attenuation degree of signals sent by surrounding beacon devices is defined as the maximum filtering distance d f ；
Step two, determining the maximum construction distance: maximum construction distance d s ＝min{d r ,d f }；
Step three, arranging beacon equipment:
a1: according to the maximum construction distance d s The maximum effective coverage area of each beacon device is regarded as d, taking the beacon device as the sphere center and the radius s The sphere of (2); placing the spheres according to a hexagonal closest packing structure, and adjusting the positions of the beacon devices to enable the beacon devices to meet the topological structure relationship of the hexagonal closest packing, so as to obtain the primitive cells of the hexagonal closest packing structure formed by the beacon devices;
a2: according to the primitive cells obtained in the step A1, taking the geometric center of the hexagonal closest packed structure of the primitive cells as an origin, establishing a space coordinate system in an environment to be positioned, and periodically translating the primitive cells along three directions of x, y and z of the space coordinate system to ensure that the hexagonal closest packed structure of the primitive cells is densely distributed in the whole environment to be positioned, thereby finishing the arrangement of beacon equipment;
a3: combining the maximum construction distance d according to the space coordinate system established in the step A2 s And the geometric property of the hexagonal closest packing structure, calculating the absolute position coordinates of each beacon device in the space coordinate system, and storing the absolute position coordinates in the server;
step four, fingerprint library positioning: comprising an off-line construction phase and an on-line positioning phase, wherein,
the off-line construction phase comprises: using maximum constructional distance d s Finishing the arrangement of the beacon equipment in the environment to be positioned by adopting the beacon equipment arrangement method in the third step; receiving signals sent by beacon devices in a hexagonal closest packing structure by using signal receiving devices, and acquiring an RSSI vector group r off For RSSI vector set r off Each RSSI vector component in the RSSI vector is filtered; repeatedly acquiring RSSI vector group r corresponding to signal receiving equipment positioned at different positions for multiple times off Constructing a training set; processing the training set by using a naive Bayes algorithm to construct a classifier;
the on-line positioning stage comprises: after a user enters an environment to be positioned, the signal receiving equipment is used for receiving signals sent by surrounding beacon equipment, RSSI (received signal strength indicator) corresponding to the signals is obtained, and absolute RSSI (received signal strength indicator) is selectedThe beacon device corresponding to the RSSI with the minimum value is a positioning reference device; for RSSI vector set r on Each RSSI vector component in the RSSI vector group is filtered, and the processed RSSI vector group r is processed on Inputting the relative position coordinate vector of the user and the positioning reference equipment into a classifier constructed in an offline construction stage; and calculating the absolute position coordinate of the user by combining the absolute position coordinate of the positioning reference equipment to finish positioning.
2. The indoor positioning method based on the hexagonal closest-packed structure as claimed in claim 1, wherein: in the fourth step, the filtering method comprises:
defining a filtering reference device and a device to be filtered: at a distance equal to or less than the maximum filtering distance d f If one of the two signal receiving devices obtains a theoretical value and a variance of the RSSI of the surrounding beacon device, the signal receiving device can be used as a filtering reference device, and the other signal receiving device can be used as a device to be filtered; establishing communication connection between filtering reference equipment and equipment to be filtered; and the filtering reference equipment transmits the theoretical value and the variance of the RSSI of the surrounding beacon equipment to the equipment to be filtered, and the equipment to be filtered uses the theoretical value and the variance of the RSSI transmitted by the filtering reference equipment to finish filtering based on a Kalman filtering method.
3. The indoor positioning method based on the hexagonal closest-packed structure as claimed in claim 2, wherein: the specific filtering step comprises:
b1: the prediction equation of the filtering reference device is as follows:
wherein the content of the first and second substances,
predicted value X of RSSI obtained by the k filtering reference equipment i Is the ith timeFiltering results of the filtering of the RSSI obtained by the reference device,
for the variance of the filtering result of the RSSI obtained by the i-th filtering reference device, when k =1,
a measurement value equal to the first time;
prediction value of filtering reference device
The variance of (c) is:
the filtering reference equipment sends the predicted value and the predicted value variance to the equipment to be filtered;
b2: the predicted noise variance of the device to be filtered is:
b3: the measurement equation of the device to be filtered is as follows:
Z k ＝H k X k +v k ，
wherein Z is k Measurement value H of RSSI obtained for equipment to be filtered k For measuring the noise drive matrix, X k For the result of this filtering, v k To measure noise;
b4: the measured noise variance of the device to be filtered is R k (ii) a The filtering gain of the equipment to be filtered is as follows:
G k ＝P k H k T (H k P k H k T +R k ) -1 ；
b5: the filtering result of the equipment to be filtered for the obtained RSSI at this time is as follows:
b6: the filtering of the RSSI acquired by the equipment to be filtered can be completed by iterating the process for multiple times in one filtering period.
4. The indoor positioning method based on the hexagonal closest packing structure as claimed in claim 3, wherein: in the fourth step, the specific steps of the offline construction stage include:
c1: according to the maximum construction distance d s Finishing the arrangement of the beacon equipment in the environment to be positioned;
c2: repeatedly measuring for multiple times in an environment to be positioned, selecting origin equipment, and taking beacon equipment positioned on 12 coordination numbers with the origin equipment as a central atom in a hexagonal closest packing structure as coordination equipment; establishing a fingerprint library primitive cell by taking an original point device as an original point and combining 12 coordination devices;
c3: creating an empty training data set S, wherein the data format in the S is bidirectional mapping from a three-dimensional vector group to a twelve-dimensional vector group; wherein, the three-dimensional vector represents a position coordinate vector, and the twelve-dimensional vector group represents an RSSI vector group;
c4: taking the origin device as the center of sphere and the maximum structure distance d s Randomly selecting a point in a spherical range of a radius, recording a position coordinate vector of the point, and recording the position coordinate vector as P: (x, y, z);
c5: in the position coordinate vector P: at the position (x, y, z), using a signal receiving device to receive signals sent by all coordination devices in the primitive cell of the fingerprint library, acquiring the corresponding RSSI of the signals, and recording all the acquired RSSI as a vector group r off ＝(r 1 ,r 2 ,r 3 ……r 12 ) (ii) a In the process, the origin equipment is used as filtering reference equipment, the signal receiving equipment is used as equipment to be filtered, and the RSSI vector group r acquired by the signal receiving equipment is subjected to filtering off Each RSSI vector component in (a) is filteredProcessing;
c6: according to the RSSI vector group r obtained in the step C5 off Constructing a set r of slave RSSI vectors off Bidirectional mapping relation to position coordinate vector
Adding the bidirectional mapping relation into a training data set S;
c7: repeating the offline construction steps C4-C6 for multiple times to obtain a training data set S ', and performing duplication elimination processing on data in the training data set S';
c8: constructing a classifier by using a naive Bayes algorithm according to the training data set S' obtained in the step C7;
the specific steps of the online positioning stage comprise:
d1: after a user enters an environment to be positioned, receiving signals sent by surrounding beacon equipment by using signal receiving equipment, acquiring RSSI (received signal strength indicator) corresponding to the signals, and selecting the beacon equipment corresponding to the RSSI with the minimum absolute value as positioning reference equipment;
d2: according to the fingerprint library primitive cells in the off-line construction stage step C2, the positioning reference equipment is used as origin equipment of the fingerprint library primitive cells, 12 beacon equipment which is positioned in the hexagonal closest packing structure and has 12 coordination numbers with the positioning reference equipment as a central atom are used as coordination equipment, the origin equipment is used as an origin, and all 12 coordination equipment are combined to establish the fingerprint library primitive cells;
d3: according to the fingerprint library primitive cell obtained in the step C2, a user receives signals sent by all coordination devices in the fingerprint library primitive cell by using signal receiving equipment, acquires RSSI corresponding to the signals, and records all acquired RSSI as a vector group r on ＝(r 1 ,r 2 ,r 3 ……r 12 ) (ii) a In the process, the positioning reference equipment is used as filtering reference equipment, the signal receiving equipment is used as equipment to be filtered, and the RSSI vector group r acquired by the signal receiving equipment is subjected to filtering on Each RSSI vector component in the RSSI vector is filtered;
d4: the RSSI vector group r after filtering processing on Inputting the relative position coordinate vector of the user and the positioning reference equipment into a Bayes classifier constructed in the step C8 of the off-line construction stage, and recording the relative position coordinate vector as a vector p = (x, y, z);
d5: according to the position coordinate vector p = (x, y, z) obtained in the step D4 of the online positioning stage, the position coordinate vector p and the absolute position coordinate (x) of the origin device are compared b ,y b ,z b ) Adding up to obtain the absolute position coordinate (x) of the user u ,y u ,z u ) And finishing positioning.
5. The indoor positioning method based on the hexagonal closest-packed structure as claimed in claim 4, wherein: in the step C2, the selection criteria of the origin device are:
in an environment to be located, at a distance from the beacon device d s Respectively receiving the RSSI of each beacon device by using a signal receiving device, calculating the mean value of the RSSI of each beacon device and calculating the variance of the mean value; repeating the process for multiple times, calculating the mean value of the RSSI variance of each beacon device, and selecting the beacon device corresponding to the minimum mean value as the original point device; if the minimum mean value corresponds to a plurality of beacon devices, performing secondary sequencing on the plurality of beacon devices corresponding to the minimum mean value;
the sub-sequence is specifically operated as: and measuring the distance between each beacon device corresponding to the minimum mean value and the geometric center of the hexagonal closest packing structure, and selecting the beacon device corresponding to the minimum distance as the origin device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110771836.1A CN113645565B (en) | 2021-07-08 | 2021-07-08 | Indoor positioning method based on hexagonal closest packing structure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110771836.1A CN113645565B (en) | 2021-07-08 | 2021-07-08 | Indoor positioning method based on hexagonal closest packing structure |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113645565A CN113645565A (en) | 2021-11-12 |
| CN113645565B true CN113645565B (en) | 2022-10-04 |
Family
ID=78416884
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110771836.1A Active CN113645565B (en) | 2021-07-08 | 2021-07-08 | Indoor positioning method based on hexagonal closest packing structure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113645565B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116502912B (en) * | 2023-04-23 | 2024-01-30 | 甘肃省人民医院 | Method and device for detecting potential distribution of medicinal plants, storage medium and electronic equipment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107801147A (en) * | 2017-07-21 | 2018-03-13 | 西安工程大学 | One kind is based on the adaptive indoor orientation method of the improved multizone of RSSI rangings |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3499989B1 (en) * | 2015-03-27 | 2021-10-20 | PCMS Holdings, Inc. | System and method for updating location data for localization of beacons |
| US10925029B2 (en) * | 2016-12-22 | 2021-02-16 | Huawei Technologies Co., Ltd. | Wi-Fi access point-based positioning method and device |
-
2021
- 2021-07-08 CN CN202110771836.1A patent/CN113645565B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107801147A (en) * | 2017-07-21 | 2018-03-13 | 西安工程大学 | One kind is based on the adaptive indoor orientation method of the improved multizone of RSSI rangings |
Non-Patent Citations (3)
| Title |
|---|
| 基于RSSI的加权蜂窝形状质心定位算法;邹东尧等;《轻工学报》;20170115(第01期);全文 * |
| 基于最优信标组的扩展卡尔曼定位算法;孟文超等;《传感技术学报》;20110415(第04期);全文 * |
| 基于蓝牙信标的室内定位;李辉等;《电脑知识与技术》;20200115(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113645565A (en) | 2021-11-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110012428B (en) | Indoor positioning method based on WiFi | |
| CN108512621B (en) | Wireless channel modeling method based on neural network | |
| CN107333238B (en) | Indoor fingerprint rapid positioning method based on support vector regression | |
| CN109068267B (en) | Indoor positioning method based on LoRa SX1280 | |
| CN110536245B (en) | Deep learning-based indoor wireless positioning method and system | |
| CN106793082A (en) | A kind of positioning of mobile equipment method in WLAN/ bluetooth heterogeneous network environments | |
| CN112218330B (en) | Positioning method and communication device | |
| CN102802260A (en) | WLAN indoor positioning method based on matrix correlation | |
| Ning et al. | Outdoor location estimation using received signal strength-based fingerprinting | |
| Adege et al. | Applying Deep Neural Network (DNN) for large-scale indoor localization using feed-forward neural network (FFNN) algorithm | |
| CN108828643B (en) | Indoor and outdoor seamless positioning system and method based on grey prediction model | |
| CN106028290A (en) | WSN multidimensional vector fingerprint positioning method based on Kriging | |
| CN110213003A (en) | A kind of wireless channel large-scale fading modeling method and device | |
| Del Corte-Valiente et al. | Localization approach based on ray-tracing simulations and fingerprinting techniques for indoor–outdoor scenarios | |
| CN113645565B (en) | Indoor positioning method based on hexagonal closest packing structure | |
| CN110247719A (en) | The playback of 5G time varying channel and emulation mode based on machine learning | |
| Cui et al. | Indoor Wi-Fi positioning algorithm based on location fingerprint | |
| Huan et al. | Indoor location fingerprinting algorithm based on path loss parameter estimation and bayesian inference | |
| Dong et al. | A wifi fingerprint augmentation method for 3-d crowdsourced indoor positioning systems | |
| CN111263295B (en) | WLAN indoor positioning method and device | |
| Zhang et al. | Towards floor identification and pinpointing position: A multistory localization model with wifi fingerprint | |
| CN108111973A (en) | A kind of indoor orientation method and device obtained based on real time fingerprint | |
| CN113194401B (en) | Millimeter wave indoor positioning method and system based on generative countermeasure network | |
| CN114679683A (en) | Indoor intelligent positioning method based on derivative fingerprint migration | |
| CN114710742A (en) | Indoor positioning method for constructing fingerprint map based on multi-chain interpolation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |