CN108495262B - Sparse representation and matching positioning method for indoor space ubiquitous positioning signal fingerprint database - Google Patents
Sparse representation and matching positioning method for indoor space ubiquitous positioning signal fingerprint database Download PDFInfo
- Publication number
- CN108495262B CN108495262B CN201810271486.0A CN201810271486A CN108495262B CN 108495262 B CN108495262 B CN 108495262B CN 201810271486 A CN201810271486 A CN 201810271486A CN 108495262 B CN108495262 B CN 108495262B
- Authority
- CN
- China
- Prior art keywords
- positioning
- fingerprint
- signal
- matching
- positioning signal
- 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
- H04W4/30—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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
- H04W64/006—Locating users or terminals or network equipment for network management purposes, e.g. mobility management with additional information processing, e.g. for direction or speed determination
Abstract
The invention belongs to the field of navigation positioning and position service, and discloses a sparse representation and matching positioning method for an indoor space ubiquitous positioning signal fingerprint library, wherein a group of observed values are collected at certain time intervals for each positioning source to form an observed value matrix; obtaining a ubiquitous signal characteristic representation about the location source position from the observation value matrix; in a positioning area, acquiring observation value matrixes at a plurality of positions, and establishing a ubiquitous signal characteristic representation matrix associated with the positions; processing the collected positioning signal observation value to obtain a positioning characteristic quantity, and matching the characteristic quantity with the characteristic fingerprint in the positioning characteristic fingerprint library; and calculating the current position of the mobile terminal equipment and displaying. The method adopts a ubiquitous signal fingerprint characteristic sparse representation fingerprint database building method and a sparse representation fingerprint matching positioning method, and is easy to collect a fingerprint database in a crowd funding mode and popularize and apply.
Description
Technical Field
The invention belongs to the field of navigation positioning and position service, and particularly relates to a sparse representation and matching positioning method for an indoor space ubiquitous positioning signal fingerprint database.
Background
Currently, the current state of the art commonly used in the industry is such that:
with the growth and development of the mobile communication industry, users are on a greatly increasing trend for location information service requirements based on diversification and individuation while enjoying the great convenience brought to social production and life by wireless networks. Meanwhile, with the wide popularization of mobile terminals, powerful support is provided for efficient and portable position estimation. Thanks to the market growth in the early stages, users have become accustomed to using mobile terminals as a living service entrance and making full use of the convenience of location services. With the deep integration of life services and mobile terminals and the exploration of the scenic operation, the portal attributes of the mobile terminals will be further expanded. In an open outdoor environment without shielding, the global positioning system and the cellular network positioning system become the most promising positioning method at present with mature technology and good system performance. However, in a complex indoor environment, wireless signals are affected by multipath effects, and the global positioning system and the cellular network positioning system cannot well meet the requirements of people on positioning performance.
Therefore, a plurality of domestic and foreign famous academic research institutions carry out intensive research on indoor positioning systems, and various indoor positioning systems are provided according to different wireless signal characteristics, such as an indoor positioning system based on bluetooth, an indoor positioning system based on radio frequency tags, an indoor positioning system based on UWB, a wireless local area network indoor positioning technology and the like.
The indoor positioning system based on the Bluetooth not only requires the mobile equipment to have the Bluetooth function, but also needs to relocate the Bluetooth access point, and the method is not put into market promotion at present because the whole manufacturing cost is too high; indoor positioning systems based on radio frequency tags are generally suitable for short-range positioning and do not have communication capability;
the indoor positioning system based on the UWB has good market competitive advantages in the aspects of positioning accuracy, positioning cost, system energy consumption, service quality and the like, is a potential indoor positioning technology at present, but the UWB technology is not mature at present, a plurality of core problems need to be solved urgently, and no practical available hardware equipment exists.
Compared with the indoor positioning systems, the wireless local area network indoor positioning system does not need extra hardware equipment, and can realize high-precision indoor positioning of a user only by installing corresponding data acquisition software on the mobile equipment, so that the wireless local area network indoor positioning system can effectively make up for the defects of the outdoor positioning system.
In summary, the problems of the prior art are as follows:
(1) in the prior art, a high-precision positioning system needs special hardware, and the cost of the whole system is high.
(2) At present, the ubiquitous signal fingerprint matching indoor positioning system becomes a research hotspot of domestic and foreign colleges and universities and enterprise research and development institutions based on mature theoretical basis, low implementation cost and potential market application value.
The fingerprint matching method is based on Bayes principle or signal domain distance shortest principle. No matter which principle is adopted, the fingerprint matching method needs to collect a sufficient number of ubiquitous signal observed values in an offline stage, and establish a signal characteristic fingerprint library for online stage matching positioning. Advantages of the fingerprint matching method include 1) no need of additional infrastructure, low cost; 2) the fingerprint database is easy to collect and popularize and apply in a crowd funding mode. However, in application, the process of collecting the observation value to generate the fingerprint library is complicated, manual field operation is required, and the fingerprint library is easily interfered by an operation environment. These problems reduce the usability and market acceptance of fingerprint matching methods.
The difficulty and significance for solving the technical problems are as follows:
the difficulty of solving the problems of the prior art is as follows: in the fingerprint matching and positioning process, the accuracy of the fingerprint feature library determines the accuracy of the positioning result. In the traditional method, the accuracy degree of the fingerprint feature library is closely related to the number of sampling samples trained by the fingerprint feature library; to improve the accuracy of the fingerprint feature library, it is required to increase the number of samples collected by the fingerprint feature library, which results in low efficiency of training the fingerprint library, and causes time and labor consuming deployment, maintenance and update of the whole positioning system and increased cost. And the positioning precision is reduced due to the reduction of the number of the training samples of the fingerprint database. Therefore, in the traditional method, the accuracy of the fingerprint database is improved, and the cost of the system is reduced, which are in a pair of contradiction;
the significance brought by the prior art problem is solved: by the sparse feature characterization method, the feature characterization model parameters are calculated by using fewer fingerprint library training samples, so that better positioning accuracy can be maintained, the sample collection amount of the fingerprint library is reduced, the collection cost of the positioning feature fingerprint library and the deployment, maintenance and updating cost of the whole positioning system are reduced, and the usability and market acceptance degree of the fingerprint matching positioning method and system are improved.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a sparse representation and matching positioning method for an indoor space ubiquitous positioning signal fingerprint database.
The invention is realized in this way, a sparse representation and matching positioning method of indoor space ubiquitous positioning signal fingerprint database, comprising:
training a positioning signal feature fingerprint library: collecting a group of observed values at certain time intervals for each positioning source to form an observed value matrix; obtaining a ubiquitous signal characteristic representation about the location of the positioning source from an observation matrix; in a positioning area, acquiring observation value matrixes at a plurality of positions, and establishing a ubiquitous signal characteristic representation matrix associated with the positions;
calculating on-line matching and positioning; processing the collected positioning signal observation value to obtain a positioning characteristic quantity, and matching the characteristic quantity with the characteristic fingerprint in the positioning characteristic fingerprint library; and calculating the current position of the mobile terminal equipment and displaying.
Further, the sparse representation and matching positioning method of the indoor space ubiquitous positioning signal fingerprint database specifically comprises the following steps:
step 1, selecting one or more positioning signal sources for positioning: designing position distribution and signal power intensity of a positioning signal source, and arranging the positioning signal source in a positioning area; when the opportunistic signals existing in the regional space are utilized, the signal source arrangement and setting operation is omitted;
step 2, collecting the observation value of the positioning signal source of the fingerprint database: collecting positioning signal source observed values in a space by one or more mobile devices, and calibrating a space reference position of the collected observed values; the space position calibration comprises the steps of determining a plurality of reference points with known coordinate values in space, receiving an observed value of a positioning signal source and a signal source identity identification code at each reference point position by using a mobile terminal, and finally recording position information of each reference point, observed values of a plurality of signal sources observed by the point and the signal source identity identification codes;
step 3, generating a virtual positioning signal observation value: generating a virtual positioning signal observation value by using the collected positioning signal source observation value and a reference position corresponding to the positioning signal source observation value; the method comprises the following steps that observed values of different positioning sources observed by the mobile equipment at the same moment are subjected to difference calculation, or the observed values of the same positioning source are subjected to difference calculation at different spatial positions by the mobile equipment;
step 4, generating a ubiquitous signal observation value fingerprint database: calculating fingerprint feature library parameters including Bayesian density model parameters by using original observation or virtual positioning observation values of the same reference position;
step 5, distributing a matching positioning fingerprint database to the users: the generated matching positioning fingerprint database is published to other user mobile equipment for use, or is configured in a cloud server for providing positioning service;
step 6, positioning a user to collect an observation value: the user mobile equipment acquires a positioning signal source observed value in an environment, then generates a corresponding positioning signal virtual observed value, inquires a fingerprint feature library parameter according to an original observed value or a virtual observed value, and reconstructs fingerprint feature distribution;
step 7, fingerprint matching and positioning: acquiring and generating a positioning signal virtual observation value by the user mobile equipment, matching the original observation value and the virtual observation value with the positioning fingerprint characteristics, and determining the spatial position of the user mobile equipment at the current moment by adopting a selected matching algorithm;
and 8, repeating the steps 6 to 7 at the next moment until the positioning task is finished.
Further, the mobile terminal device comprises a mobile phone, a tablet computer, a bracelet or other mobile terminals;
furthermore, the matching algorithm comprises a Bayesian estimation method, an observed value domain shortest distance method and an algorithm based on other positioning principles; the method is characterized in that the positioning fingerprint features are described by a sparse representation model, and representation model parameters are calculated from fewer sampling samples; and in the real-time positioning stage, a positioning signal is reconstructed from the sparse representation model, and matching positioning calculation is carried out.
The positioning signal source comprises a WiFi signal, a Bluetooth signal and an earth magnetic field signal.
Another object of the present invention is to provide a computer program for implementing the sparse representation of the fingerprint library of the indoor space ubiquitous positioning signal and the matching positioning method.
The invention also aims to provide an information data processing terminal for realizing the sparse representation and the matching positioning method of the indoor space ubiquitous positioning signal fingerprint database.
It is another object of the present invention to provide a computer-readable storage medium comprising instructions which, when executed on a computer, cause the computer to perform the indoor-space-ubiquitous-location-signal-fingerprint-library sparse-characterization and matching-location method.
Another object of the present invention is to provide a sparse representation and matching location system for fingerprint library of ubiquitous indoor space location signals, comprising:
positioning a signal source device; a positioning source device that generates and transmits a positioning signal;
a positioning signal observation sensor; a sensor for receiving and observing the positioning signal;
a location signal feature fingerprint library computing device; calculating the positioning signals acquired by the positioning signal observed quantity acquisition equipment to generate a positioning signal characteristic fingerprint library;
locating a signal feature fingerprint library; calculating a generated positioning signal characteristic fingerprint database according to the collected positioning signal observed quantity, and using the positioning signal characteristic fingerprint database for matching positioning calculation;
moving the positioning terminal device; the positioning device comprises a positioning signal observation sensor and a device for observing and recording a positioning signal and calculating the position of the positioning device by utilizing computing resources;
matching the positioning computing device; calculating and processing a positioning signal acquired by positioning mobile terminal equipment;
a data storage medium; storing positioning signal characteristic fingerprint database data generated by calculation of positioning signal characteristic fingerprint database calculation equipment;
a location map visualization user interface; and graphically displaying the data processed by the matching positioning computing equipment.
The invention also aims to provide an information data processing terminal carrying the sparse characteristic of the indoor space ubiquitous positioning signal fingerprint database and the matching positioning system.
In summary, the advantages and positive effects of the invention are：
The invention provides a high-efficiency fingerprint library characteristic sparse representation and calculation method and a corresponding fingerprint matching positioning algorithm, aiming at the problems of complicated process of acquiring observation values on site to generate a fingerprint library and large field workload in a ubiquitous signal fingerprint matching indoor positioning method. As shown in table 1, the efficiency of on-site fingerprint library acquisition is improved, the workload of on-site operation is reduced, 30 or more characteristic fingerprint library samples can be required by using 20 samples instead of the conventional method, the work efficiency of on-site operation is improved by more than 50%, and the availability of the matching positioning method is improved; the fingerprint matching and positioning algorithm based on sparse representation improves the matching and positioning accuracy, and as shown in the experiment of table 1, 95% of positioning errors are reduced from 6.18 meters to 3.84 meters, and the accuracy is improved by 37.9%.
TABLE 1 comparison of conventional positioning method based on WiFi ubiquitous signals with sparse characterization and matching positioning method of fingerprint database of the present invention
| Conventional methods | The method of the invention | |
| Number of samples taken | 30 | 20 |
| Positioning root mean square error (meter) | 2.87 | 1.94 |
| Location 95% error (rice) | 6.18 | 3.84 |
| Maximum positioning error | 10.15 | 4.39 |
Compared with the prior art, the invention also has the following beneficial effects:
(1) by adopting a ubiquitous signal fingerprint feature sparse representation fingerprint database building method, the number of observed values required in fingerprint feature representation calculation is reduced, the workload of field acquisition is reduced, and the usability of a fingerprint matching indoor positioning method is improved;
(2) and the ubiquitous signal fingerprint characteristic sparse characteristic and the corresponding fingerprint matching positioning method are adopted, so that the indoor positioning precision and reliability are improved.
(3) The ubiquitous signal fingerprint characteristic sparse representation fingerprint database building method and the sparse representation fingerprint matching positioning method are adopted, and the fingerprint database is easy to collect in a crowd funding mode and popularize and apply.
Drawings
Fig. 1 is a schematic diagram of a topological relationship structure formed by a plurality of ubiquitous positioning signal sources and a mobile positioning terminal according to the present invention.
Fig. 2 is a schematic diagram of a data processing method for indoor space ubiquitous signal characteristic sparse representation, a fingerprint database efficient database building method and a corresponding ubiquitous signal fingerprint matching indoor positioning method according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The system works in personal mobile equipment, such as a mobile phone, a tablet computer, a bracelet or other mobile terminals, and is used for determining the positions of the mobile equipment and users thereof and developing intelligent position service application; the method of the invention utilizes a sensor module embedded in the mobile equipment to collect the observation quantity of the ubiquitous positioning signal source in the environment and adopts the fingerprint matching principle to determine the position of the mobile equipment. The method reduces the number of observation quantity samples required when the fingerprint database is established in the fingerprint matching method, thereby reducing the workload of data acquisition of the fingerprint database; the invention also comprises a new fingerprint matching method based on the established fingerprint database, and the positioning precision and reliability of fingerprint matching are improved. Therefore, the invention improves the usability, reliability and positioning precision of WiFi or Bluetooth and other fingerprint matching indoor positioning technologies.
The invention is further described below with reference to specific assays.
According to the sparse representation and matching positioning method for the indoor space ubiquitous positioning signal fingerprint library, the ubiquitous signal space characteristic library can be represented by collecting a small amount of ubiquitous positioning source signals, and the spatial position of personal mobile equipment is determined by adopting a fingerprint matching positioning principle; the ubiquitous signal characteristic sparse representation algorithm is adopted, so that the accuracy and reliability of fingerprint matching positioning are improved; the mobile device comprises a mobile phone, a tablet computer, a bracelet or other mobile terminals. The invention simultaneously supports one or more fingerprint matching positioning signal sources, wherein the positioning signal sources comprise WiFi signals, Bluetooth signals, geomagnetic signals and the like.
The sparse characteristic and matching positioning method of the indoor space ubiquitous positioning signal fingerprint library, provided by the embodiment of the invention, is characterized in that ubiquitous positioning source signals in an environment and corresponding reference positions of the ubiquitous positioning source signals are collected to generate a ubiquitous signal fingerprint characteristic library, and the spatial position of personal mobile equipment is determined by adopting a positioning principle of fingerprint matching; the mobile device comprises a mobile phone, a tablet computer, a bracelet or other mobile terminals.
According to the sparse representation and matching positioning method for the indoor space ubiquitous positioning signal fingerprint library, the Bayes probability density of the signal is reconstructed by adopting less ubiquitous positioning signal observed quantity and corresponding reference positions, and the sparse representation and matching positioning method is used for determining the space position of the mobile equipment.
According to the sparse representation and matching positioning method for the indoor space ubiquitous positioning signal fingerprint library, the Bayes density model is established by adopting less ubiquitous positioning signal observed quantities and corresponding reference positions, and the Bayes probability density of the signal is reconstructed by utilizing a group of model parameters to determine the space position of the mobile equipment.
According to the sparse representation and matching positioning method for the indoor space ubiquitous positioning signal fingerprint library, the distributed features of a group of limited number of parameter description signals are determined by adopting less ubiquitous positioning signal observation quantities and corresponding reference positions, and a lightweight fingerprint feature library is established and used for determining the space position of the mobile equipment.
According to the sparse representation and matching positioning method of the indoor space ubiquitous positioning signal fingerprint library, ubiquitous positioning source signals in an environment are observed, and the space position of personal mobile equipment is determined by utilizing a lightweight fingerprint feature library and adopting a positioning principle of fingerprint matching; the personal mobile device comprises a mobile phone, a tablet computer, a bracelet or other mobile terminals.
The sparse representation and matching positioning method for the indoor space ubiquitous positioning signal fingerprint library provided by the embodiment of the invention observes ubiquitous positioning source signals in an environment, reconstructs ubiquitous positioning signal fingerprint features by using a limited number of parameters by using a lightweight fingerprint feature library, and determines the space position of personal mobile equipment by using a Bayesian probability density of the ubiquitous signals and a positioning principle of fingerprint matching. The matching algorithm comprises a Bayesian estimation method, an observed value domain shortest distance method, an algorithm based on other positioning principles, and an improved algorithm based on the methods.
According to the sparse representation and matching positioning method for the indoor space ubiquitous positioning signal fingerprint library, ubiquitous positioning source signals in an environment are observed, the lightweight fingerprint feature library is utilized, the Bayes probability density of the ubiquitous signals is reconstructed by adopting a limited number of parameters, and the fingerprint matching positioning is carried out by utilizing the Bayes density instead of the probability distribution.
The sparse representation and matching positioning method for the indoor space ubiquitous positioning signal fingerprint database provided by the embodiment of the invention simultaneously supports one or more fingerprint matching positioning signal sources, wherein the positioning signal sources comprise WiFi signals, Bluetooth signals, earth magnetic field signals and the like.
According to the sparse representation and matching positioning method for the indoor space ubiquitous positioning signal fingerprint database, provided by the embodiment of the invention, the positioning signal source comprises virtual observed quantities formed by WiFi signals, Bluetooth signals, earth magnetic field signals and the like, and the virtual observed quantities comprise but are not limited to the difference value of two original observed quantities.
The sparse representation and matching positioning method for the indoor space ubiquitous positioning signal fingerprint database provided by the embodiment of the invention comprises the following steps of:
step 1, selecting one or more positioning signal sources for positioning: designing position distribution, signal power intensity and the like of a positioning signal source, and arranging the positioning signal source in a positioning area; when the opportunistic signals existing in the regional space are utilized, the signal source arrangement and setting operation is omitted;
step 2, collecting the observation value of the positioning signal source of the fingerprint database: collecting positioning signal source observed values in a space by one or more mobile devices, and calibrating a space reference position of the collected observed values; the space position calibration comprises the steps of determining a plurality of reference points with known coordinate values in space, receiving an observed value of a positioning signal source and a signal source identity identification code at each reference point position by using a mobile terminal, and finally recording position information of each reference point, observed values of a plurality of signal sources observed by the point and the signal source identity identification codes;
step 3, generating a virtual positioning signal observation value: the method comprises the steps that the observed values of the positioning signal sources and the corresponding reference positions of the observed values are acquired, and virtual positioning signal observed values are generated;
step 4, generating a ubiquitous signal observation value fingerprint database: fingerprint feature library parameters, including but not limited to bayesian density model parameters, are calculated using raw observations or virtual positioning observations of the same reference location.
Step 5, distributing a matching positioning fingerprint database to the users: the generated matching positioning fingerprint database is published to other user mobile equipment for use, or is configured in a cloud server for providing positioning service;
step 6, positioning a user to collect an observation value: the user mobile equipment acquires a positioning signal source observed value in an environment, then generates a corresponding positioning signal virtual observed value, inquires a fingerprint feature library parameter according to an original observed value or a virtual observed value, and reconstructs fingerprint feature distribution;
step 7, fingerprint matching and positioning: acquiring and generating a positioning signal virtual observation value by the user mobile equipment, matching the original observation value and the virtual observation value with the positioning fingerprint characteristics, and determining the spatial position of the user mobile equipment at the current moment by adopting a selected matching algorithm;
and 8, repeating the steps 6 to 7 at the next moment until the positioning task is finished.
The invention is further described below with reference to the accompanying drawings.
Fig. 1 is a schematic diagram of a topological relationship structure formed by a ubiquitous positioning signal source and a mobile positioning terminal according to the present invention.
The sparse representation and matching positioning method for the indoor space ubiquitous positioning signal fingerprint database provided by the embodiment of the invention comprises two stages:
(1) training a positioning signal characteristic fingerprint library; (2) and (5) performing online matching positioning calculation. In the training stage of the positioning fingerprint feature library, the positioning terminal collects a group of observed values of each positioning source at a certain time interval to form an observed value vector. The observations collected by the positioning terminals at different locations are different for a given positioning signal source. Therefore, at a certain position, the positioning terminal collects the observed values of a plurality of positioning sources to form an observed value matrix. This observation matrix is associated with the location from which a ubiquitous signature characterization for the location can be derived. In a positioning area, observation value matrixes are collected at a plurality of positions, and a ubiquitous signal characteristic characterization matrix associated with the positions is established. In the positioning stage, a positioning signal observation value acquired by the mobile terminal device is processed to obtain a positioning characteristic quantity, the characteristic quantity is matched with the characteristic fingerprint in the positioning characteristic fingerprint library, and the current position of the mobile terminal device is calculated by the matching positioning calculation device and is displayed on a map visual user interface. On the basis, related position service application is developed by combining with a specific application scene. The matching algorithm comprises a Bayesian estimation method, an observed value domain shortest distance method, an algorithm based on other positioning principles and the like.
The sparse feature and matching positioning system for the indoor space ubiquitous positioning signal fingerprint database provided by the embodiment of the invention comprises the following components:
positioning a signal source device; a positioning source device that generates and transmits a positioning signal;
a positioning signal observation sensor; a sensor for receiving and observing the positioning signal;
a location signal feature fingerprint library computing device; calculating the positioning signals acquired by the positioning signal observed quantity acquisition equipment to generate a positioning signal characteristic fingerprint library;
locating a signal feature fingerprint library; calculating a generated positioning signal characteristic fingerprint database according to the collected positioning signal observed quantity, and using the positioning signal characteristic fingerprint database for matching positioning calculation;
moving the positioning terminal device; the positioning device comprises a positioning signal observation sensor and a device for observing and recording a positioning signal and calculating the position of the positioning device by utilizing computing resources;
matching the positioning computing device; calculating and processing a positioning signal acquired by positioning mobile terminal equipment;
a data storage medium; storing positioning signal characteristic fingerprint database data generated by calculation of positioning signal characteristic fingerprint database calculation equipment;
a location map visualization user interface; and graphically displaying the data processed by the matching positioning computing equipment.
The positioning signal observed quantity acquisition equipment is connected with a positioning signal observed quantity acquisition terminal and is used for acquiring positioning signals; calculating the collected positioning signals through the positioning signal characteristic fingerprint database computing equipment to generate a positioning signal characteristic fingerprint database; the characteristic fingerprint database is stored in a data storage medium; and the matching positioning computing equipment is used for computing and processing the positioning signals acquired by the positioning mobile terminal equipment and displaying the positioning signals on a visual user interface of a position map.
FIG. 2 is a schematic diagram of a data processing method for sparse representation and matching location method of an indoor space ubiquitous location signal fingerprint database according to an embodiment of the present invention,
firstly, a ubiquitous positioning signal source observed value in an environment is collected through a mobile terminal, and a positioning characteristic fingerprint database is generated through a positioning signal characteristic fingerprint database computing device. Taking a bayesian fingerprint matching method as an example, when a fingerprint database of a traditional bayesian algorithm is established, RSSI values of APs received at each fingerprint point are required to be calculated and probability distribution of each AP is recorded. At RiAP at fingerprint point is named as AmHas a signal strength RSSI value of XnThe probability of (d) can be expressed as:
in the formula (I), the compound is shown in the specification,
is an observed value X representing signal strength RSSInNumber of occurrences in training data at ith fingerprint Point, NiRepresenting the total number of training samples collected at the ith fingerprint point. The entire fingerprint library can be represented as:
D＝[R1，R2，...，Rw](2)
where W is the total number of fingerprint points collected for that area.
In order to improve the calculation efficiency, the influence of the singularity of the RSSI value of the signal strength is weakened. The traditional algorithm divides the RSSI into 9 ranges when establishing the fingerprint database, each range can be regarded as a bin, and thus each bin is recorded by taking the bin as a unit when establishing the fingerprint databaseProbability distribution of individual APs; and when the AP is matched and positioned, finding out the corresponding bin acquisition probability according to the received signal strength RSSI value of the AP. At this time, AP of the ith fingerprint point is B of AmnIndividual bins can be represented as:
in the formula, En-1And EnAre respectively BnThe left and right limits of (a) and (b),
represents the RSSI value at (E)n-1,En]Number of occurrences within the range.
The invention adopts a Weibull Bayesian density model to describe the probability density of the RSSI of each AP received on each fingerprint point, and the probability density function expression is as follows:
the functional expression for calculating the probability distribution is:
x is a variable of this function, K is a shape parameter, λ is a scale parameter, and θ is a shift parameter.
Parameter estimation of the Weibull bayesian model can be done with a limited sample size of RSSI values. The calculation formula of the parameters (λ, K, θ) of the model is as follows:
in the formula (I), the compound is shown in the specification,
is a set of RSSI values oiThe mean value of (a); STD is standard deviation; Γ is the gamma function; (K +0.15) is
An approximate expression where K is 1.5. ltoreq. K.ltoreq.2.5.
When the fingerprint database is established in the fingerprint matching method based on the Weibull signal model, only three parameters of each AP on a fingerprint point need to be recorded, and the data volume is greatly reduced. And during matching positioning, probability distribution is dynamically calculated based on a Weibull Bayesian density model according to the RSSI values of all APs received in real time. In the positioning stage, the received RSSI value of the AP is added with 5dB and subtracted with 5dB, and the corresponding probability is calculated according to a Bayesian density model as follows:
in the formula, x is represented at a fingerprint point RiThe smart phone receives an RSSI value with AP name Am. Three parameters required by a Weibull signal model are recorded in the fingerprint library, namely, the probability distribution of the RSSI value of the smart phone with the AP name Am received at the fingerprint point Ri is simulated.
The fingerprint matching method of the invention applies Bayes theory and maximum likelihood estimation method. The principle is that a conditional probability is used for establishing a model for the position fingerprint, then a Bayesian inference mechanism is adopted to estimate the position of the positioning terminal, also called Bayesian probability algorithm, and the basic principle is as follows:
in the formula, x represents a certain fingerprint point in the fingerprint database; y represents the RSSI value of the AP received by the smart phone at the positioning point; p (x | y) represents the probability that a location point will appear at fingerprint point x when the RSSI value is y; p (y | x) represents the probability of the RSSI value being y at fingerprint point x; p (x) represents the probability of a fingerprint point. As shown in equation (13), when the value of p (x | y) is the maximum, that is, when the RSSI value received by the anchor point is y, the probability that the fingerprint point x appears is the maximum, that is, the fingerprint point x is the best matched with the anchor point, which can be output as the positioning result. Therefore, the bayesian probability method is to find the maximum value of p (x | y), where x is the positioning result, and the formula is:
to obtain the maximum value of p (y | x), it can be known that p (x) and p (y) are the same at each fingerprint point through the bayesian formula, i.e. it can be converted to solve the maximum value of p (y | x), which represents the probability of RSSI value of each AP received at fingerprint point x, because each AP is independent, so as to find the maximum value of probability product of RSSI value of each AP, the formula is as follows:
in the formula, x represents a certain fingerprint point in the fingerprint database; y isjAnd the RSSI value of the jth AP received by the smart phone at the positioning point is represented. Therefore, the conditional probability product of all APs at each fingerprint point is obtained, the probability maximum value is found out according to the maximum likelihood estimation method, and the corresponding fingerprint point is the positioning result.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When used in whole or in part, can be implemented in a computer program product that includes one or more computer instructions. When loaded or executed on a computer, cause the flow or functions according to embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center via wire (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL), or wireless (e.g., infrared, wireless, microwave, etc.)). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that includes one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (7)
1. The sparse representation and matching positioning method for the indoor space ubiquitous positioning signal fingerprint library is characterized by comprising the following steps of:
training a positioning signal feature fingerprint library: collecting a group of observed values at certain time intervals for each positioning source to form an observed value matrix; obtaining a ubiquitous signal characteristic representation about the location of the positioning source from an observation matrix; in a positioning area, acquiring observation value matrixes at a plurality of positions, and establishing a ubiquitous signal characteristic representation matrix associated with the positions;
calculating on-line matching and positioning; processing the collected positioning signal observation value to obtain a positioning characteristic quantity, and matching the characteristic quantity with the characteristic fingerprint in the positioning signal characteristic fingerprint library; calculating the current position of the mobile terminal equipment and displaying;
the sparse representation and matching positioning method of the indoor space ubiquitous positioning signal fingerprint database specifically comprises the following steps:
step 1, selecting one or more positioning signal sources for positioning: designing position distribution and signal power intensity of a positioning signal source, and arranging the positioning signal source in a positioning area; when the existing opportunistic signals in the area space are utilized, the signal source arrangement and setting operation is omitted;
step 2, collecting the observation value of the positioning signal source of the fingerprint database: collecting positioning signal source observed values in a space by one or more mobile devices, and calibrating a space reference position of the collected observed values; the space reference position calibration comprises the steps of determining a plurality of reference points with known coordinate values in space, receiving an observed value of a positioning signal source and a signal source identification code at each reference point position by using a mobile terminal, and finally recording position information of each reference point, observed values of a plurality of signal sources observed by the point and the signal source identification codes;
step 3, generating a virtual positioning signal observation value: generating a virtual positioning signal observation value by using the collected positioning signal source observation value and a reference position corresponding to the positioning signal source observation value; the method comprises the following steps that observed values of different positioning sources observed by the mobile equipment at the same moment are subjected to difference calculation, or the observed values of the same positioning source are subjected to difference calculation at different spatial positions by the mobile equipment;
step 4, generating a ubiquitous signal observation value fingerprint database: calculating fingerprint feature library parameters including Weibull Bayes density model parameters by using the original observation and the virtual positioning observation values of the same reference position;
step 5, distributing a matching positioning fingerprint database to the users: the generated matching positioning fingerprint database is published to other user mobile equipment for use, or is configured in a cloud server for providing positioning service;
step 6, positioning a user to collect an observation value: the user mobile equipment acquires a positioning signal source observed value in an environment, then generates a corresponding positioning signal virtual observed value, inquires a fingerprint feature library parameter according to an original observed value and the virtual observed value, and reconstructs fingerprint feature distribution;
step 7, fingerprint matching and positioning: acquiring and generating a positioning signal virtual observation value by the user mobile equipment, matching the original observation value and the virtual observation value with the positioning fingerprint characteristics, and determining the spatial position of the user mobile equipment at the current moment by adopting a selected matching algorithm;
step 8, repeating the steps 6 to 7 at the next moment until the positioning task is finished;
the selected matching algorithm including the positioning principle is adopted, and the method specifically comprises the following steps: the positioning fingerprint features are described by Weibull Bayes density model parameters, and the characterization model parameters are calculated from fewer sampling samples; and in the real-time positioning stage, a positioning signal is reconstructed from Weibull Bayes density model parameters, and matching positioning calculation is carried out.
2. The sparse representation of indoor space ubiquitous positioning signal fingerprint repository and the matching positioning method of claim 1, wherein the mobile terminal device comprises a mobile phone, a tablet, a bracelet or other mobile terminal.
3. A computer program for implementing the sparse representation of the fingerprint library of the indoor space ubiquitous location signal and the matching location method according to any one of claims 1 to 2.
4. An information data processing terminal for implementing the sparse representation and matching positioning method of the indoor space ubiquitous positioning signal fingerprint database according to any one of claims 1 to 2.
5. A computer readable storage medium comprising instructions which, when executed on a computer, cause the computer to perform the indoor space-ubiquitous location signal fingerprint bank sparse representation and matching location method of any one of claims 1-2.
6. A system for implementing the sparse representation of the fingerprint library of the indoor space ubiquitous positioning signal and the matching positioning method according to any one of claims 1 to 2, the system comprising:
positioning a signal source device: a positioning source device that generates and transmits a positioning signal;
positioning signal observation sensor: a sensor for receiving and observing the positioning signal;
positioning signal feature fingerprint library computing device: calculating the positioning signals acquired by the positioning signal observed quantity acquisition equipment to generate a positioning signal characteristic fingerprint library;
positioning signal feature fingerprint database: calculating a generated positioning signal characteristic fingerprint database according to the collected positioning signal observed quantity, and using the positioning signal characteristic fingerprint database for matching positioning calculation;
mobile positioning terminal equipment: the positioning device comprises a positioning signal observation sensor and a device for observing and recording a positioning signal and calculating the position of the positioning device by utilizing computing resources;
the matching location computing device: calculating and processing a positioning signal acquired by positioning mobile terminal equipment;
data storage medium: storing positioning signal characteristic fingerprint database data generated by calculation of positioning signal characteristic fingerprint database calculation equipment;
location map visualization user interface: and graphically displaying the data processed by the matching positioning computing equipment.
7. An information data processing terminal carrying the system according to claim 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810271486.0A CN108495262B (en) | 2018-03-29 | 2018-03-29 | Sparse representation and matching positioning method for indoor space ubiquitous positioning signal fingerprint database |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810271486.0A CN108495262B (en) | 2018-03-29 | 2018-03-29 | Sparse representation and matching positioning method for indoor space ubiquitous positioning signal fingerprint database |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108495262A CN108495262A (en) | 2018-09-04 |
| CN108495262B true CN108495262B (en) | 2021-03-02 |
Family
ID=63316926
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810271486.0A Active CN108495262B (en) | 2018-03-29 | 2018-03-29 | Sparse representation and matching positioning method for indoor space ubiquitous positioning signal fingerprint database |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108495262B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115638795B (en) * | 2022-12-26 | 2023-03-10 | 武汉大学 | Indoor multi-source ubiquitous positioning fingerprint database generation and positioning method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102333372A (en) * | 2011-09-15 | 2012-01-25 | 中国科学院计算技术研究所 | Real-time positioning method and system based on radio frequency fingerprints |
| CN103402256A (en) * | 2013-07-11 | 2013-11-20 | 武汉大学 | Indoor positioning method based on WiFi (Wireless Fidelity) fingerprints |
| CN103916954A (en) * | 2013-01-07 | 2014-07-09 | 华为技术有限公司 | Probability locating method and locating device based on WLAN |
| CN105208651A (en) * | 2015-08-17 | 2015-12-30 | 上海交通大学 | Wi-Fi position fingerprint non-monitoring training method based on map structure |
| CN107333243A (en) * | 2017-08-14 | 2017-11-07 | 柳景斌 | A kind of mobile device fingerprint matching localization method for exempting from hardware demarcation |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8942725B2 (en) * | 2012-12-14 | 2015-01-27 | Apple Inc. | Location determination using a state space estimator |
-
2018
- 2018-03-29 CN CN201810271486.0A patent/CN108495262B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102333372A (en) * | 2011-09-15 | 2012-01-25 | 中国科学院计算技术研究所 | Real-time positioning method and system based on radio frequency fingerprints |
| CN103916954A (en) * | 2013-01-07 | 2014-07-09 | 华为技术有限公司 | Probability locating method and locating device based on WLAN |
| CN103402256A (en) * | 2013-07-11 | 2013-11-20 | 武汉大学 | Indoor positioning method based on WiFi (Wireless Fidelity) fingerprints |
| CN105208651A (en) * | 2015-08-17 | 2015-12-30 | 上海交通大学 | Wi-Fi position fingerprint non-monitoring training method based on map structure |
| CN107333243A (en) * | 2017-08-14 | 2017-11-07 | 柳景斌 | A kind of mobile device fingerprint matching localization method for exempting from hardware demarcation |
Non-Patent Citations (1)
| Title |
|---|
| 终端异质下位置指纹的鲁棒性研究;谢代军，孔范增，胡捍英;《计算机工程》;20140531;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108495262A (en) | 2018-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Luo et al. | Enhancing Wi-Fi fingerprinting for indoor positioning using human-centric collaborative feedback | |
| CN104270816B (en) | The adaptive dynamic fingerprint base construction method of LED visible light indoor locating system | |
| CN107302793B (en) | Positioning method, server, terminal and system based on wireless signals | |
| CN109379711B (en) | positioning method | |
| Anagnostopoulos et al. | A reproducible comparison of RSSI fingerprinting localization methods using LoRaWAN | |
| CN111935820B (en) | Positioning implementation method based on wireless network and related equipment | |
| CN109068274B (en) | Method for positioning complex indoor environment target under assistance of fine-grained fingerprint quality | |
| CN112218330A (en) | Positioning method and communication device | |
| CN103987118A (en) | Access point k-means clustering method based on received signal strength signal ZCA whitening | |
| Lemic et al. | Web-based platform for evaluation of RF-based indoor localization algorithms | |
| CN106162559A (en) | For estimating equipment and the method for position in a wireless communication system | |
| CN111523777A (en) | Novel smart city system and application method thereof | |
| Cui et al. | Improved genetic algorithm to optimize the Wi-Fi indoor positioning based on artificial neural network | |
| CN109547929A (en) | Distributed sensor node positioning method based on conjugate gradient method | |
| CN115100376A (en) | Electromagnetic spectrum map implementation method based on improved inverse distance interpolation method | |
| Chen et al. | A wifi indoor localization method based on dilated cnn and support vector regression | |
| CN108495262B (en) | Sparse representation and matching positioning method for indoor space ubiquitous positioning signal fingerprint database | |
| CN106792527A (en) | A kind of position data processing method, device and computing device | |
| CN116528282B (en) | Coverage scene recognition method, device, electronic equipment and readable storage medium | |
| CN112526572B (en) | Network switching method and positioning system for indoor and outdoor seamless navigation | |
| CN109143161A (en) | High-precision indoor orientation method based on mixed-fingerprint Environmental Evaluation Model | |
| CN114513805A (en) | Wireless modeling method and system | |
| CN110320493B (en) | Indoor positioning method, device, electronic equipment and computer storage medium | |
| CN104635206B (en) | A kind of method and device of wireless location | |
| CN113890833B (en) | Network coverage prediction method, device, equipment and storage medium |
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 |