KR20210090058A - Method and system for building indoor positioning database and positioning indoor with indoor positioning database - Google Patents

Abstract

The present invention relates to a method for building an indoor positioning database and performing indoor positioning to reduce the amount of communications with a server and an apparatus thereof. According to an embodiment of the present invention, the indoor positioning method comprises the following steps: measuring a received signal strength (RSS) from access points (APs) measured at a first location in an indoor space; selecting a predetermined number of APs as first APs associated with a first location in an order of the intensity of the RSS; determining a weighted centroid (WC) of the first APs at the first location on the basis of the RSS from the first APs; and comparing information of related APs for each reference point in the indoor space, the RSS from the related APs, and data of a database storing the WC of the related APs at the reference point with the information of the first APs, the RSS from the first APs, and the WC of the first APs at the first location to estimate location information of the first location.

Description

Building a database for indoor positioning, indoor positioning method and system {METHOD AND SYSTEM FOR BUILDING INDOOR POSITIONING DATABASE AND POSITIONING INDOOR WITH INDOOR POSITIONING DATABASE}

The present disclosure is for indoor positioning that allows to establish a database for indoor positioning based on radio frequency signal strength and a weighted centroid (WC) when positioning based on radio frequency (RF), and to estimate an indoor location It relates to database construction, indoor positioning method and system.

Today, as social and commercial interest in Location Based Service (LBS) increases, research and development of various outdoor and indoor location measurement systems is increasing. Location information, both indoors and outdoors, is at the heart of an ergonomic future-sustainable computing framework, which can provide multiple possibilities for a wide range of computing environments.

Meanwhile, GNSS (Global Navigation Satellite System) has been a representative technology in a location measurement system. However, in the case of a technology for estimating a location by receiving a GNSS signal, since it is difficult to receive a satellite signal when it is located inside a building or between high-rise buildings, it is difficult to estimate the exact location of the device. Accordingly, research and development of various indoor positioning systems (Indoor Positioning Service, IPS) is increasing.

Location-based services, particularly indoor location measurement, can estimate indoor location using Wi-Fi, Bluetooth Low Energy (BLE), ZigBee, Ultra-Wide Band (UWB), visible light, and the Earth's magnetic field. In particular, among indoor positioning technologies, there is a method of using a Bluetooth low energy beacon. It consumes less energy and is efficient, and is now built into many smartphones and tablets.

In particular, the fingerprinting method has become a major choice for location-based service design because location estimation is good and line-of-sight from an access point is not required. That is, this fingerprinting method is widely used in indoor location measurement due to its high stability.

However, the fingerprinting method builds a database for indoor positioning including an offline (training) step, performs positioning in an online (test) step, accumulates data for all access points deployed indoors in the database, and for positioning perform calculations Accordingly, the fingerprinting method has a problem in that the calculation process is complicated and the size of the database may be increased.

The above-mentioned background art is technical information possessed by the inventor for derivation of the present invention or acquired in the process of derivation of the present invention, and cannot necessarily be said to be a known technique disclosed to the general public prior to filing of the present invention.

An object of the present disclosure is to establish a database for indoor positioning by performing clustering based on radio frequency signal strength (RSS) and weighted centroid (WC) when positioning based on radio frequency, It is intended to estimate the indoor location.

A method for building a database for indoor positioning according to an embodiment of the present disclosure includes measuring Received Signal Strength (RSS) from access points measured at a reference point in an indoor space, and accessing a predetermined number of accesses in the order of increasing the size of the RSS selecting the access points associated with the reference point as access points associated with the reference point, determining a weighted centroid (WC) of the associated access points at the reference point based on RSS from the associated access points; information of the associated access points and the Storing RSS from the related access points and the WC of the related access points in association with the location information of the reference point in a database, and repeating the above steps for each reference point in the indoor space.

Determining the WC of the associated access points includes calculating a distance from the reference point to each of the associated access points based on RSS from the associated access points, and at the reference point based on the calculated distance. determining the WC of the associated access points.

After the repeating step, generating training data comprising information of each reference point's related access points stored in the database and RSS from the respective reference point's related access points and the WC of the related access points at each reference point It may further include the step of

Here, the training data may be data labeled with location information of each reference point.

After the generating, the method may further include training the constructed initial neural network model as a neural network model for indoor positioning using the training data.

Here, the neural network model for indoor positioning may be configured to output location information of a specific location when information of related access points at a specific location, RSS from related access points, and WC of related access points at the specific location are input. have.

The step of measuring the RSS includes the steps of measuring the RSS measured from each of the access points a plurality of times, calculating an average value of the RSS measured a plurality of times, and selecting the average value as the RSS value from each of the access points. may include

An indoor positioning method according to an embodiment of the present disclosure includes the steps of measuring RSS from access points measured at a first location in an indoor space; selecting as first associated access points, determining the WC of the first access points at the first location based on RSS from the first access points, and the number of associated access points per reference point in the indoor space Information of the first access points and data in a database in which information, RSS from the associated access points, and the WC of the associated access points at the reference point are stored, RSS from the first access points and the first access point at the first location and estimating location information of the first location by comparing their WCs.

Here, the database includes cluster information in which reference points are grouped, and the estimating includes: searching for a reference point having a WC closest to the WC of the first access points in the database, and belonging to the cluster of the searched reference points calculating the location information of the first location based on the difference between the WC at reference points and the WC of the first access points and the difference between the RSS of the related access points at the reference points and the RSS of the first access points may include steps.

In addition, in the step of calculating the location information of the first location, the WC of the corresponding reference point among the reference points of the cluster is the closest to the WC of the first access points and related access points of the first access points and the first access points are RSS size and calculating the location information of the first location based on the number of access points matching each other when they are listed in an order according to .

In addition, the step of estimating the location information of the first location includes the step of estimating the location information of the first location based on the neural network model for indoor positioning learned using the training data based on the built initial neural network model. may include

The neural network model for indoor positioning may be configured to output location information of a specific location upon input of information of related access points at a specific location, RSS from related access points, and WC of related access points at the specific location.

In addition, the step of measuring the RSS includes measuring the RSS measured from each access point a plurality of times, calculating an average value of the RSS measured a plurality of times, and selecting the average value as the RSS value from each access point may include the step of

An indoor positioning device according to an embodiment of the present disclosure includes a communication interface for receiving RSS from access points, one or more processors, and a memory connected to the one or more processors, the memory, when executed by the processor, the processor to measure RSS from access points measured at a first location in an indoor space, and select a certain number of access points in the order of increasing RSS size as first access points related to the first location, Determine the WC of the first access points in the first location based on the RSS from the first access points, information of the access points related to each reference point in the indoor space, the RSS from the related access points, and the reference point By comparing the data of the database in which the WC of the related access points is stored with the information of the first access points, the RSS from the first access points, and the WC of the first access points at the first location, the location information of the first location It is possible to store codes that cause (cause) to estimate .

The proposed system can reduce the physical size of the database and the amount of data communication with the server at the execution stage by increasing the localization accuracy by using various fingerprinting functions. In addition, affinity propagation clustering can minimize the search space of RP, reduce computational cost, and smooth out noisy RSS through exponential averaging.

1 is a schematic illustration of an indoor positioning system according to an embodiment of the present disclosure.
2 is a block diagram schematically illustrating an indoor positioning system according to an embodiment of the present disclosure.
3 is a view for explaining the overall flow of an offline phase of building a database for indoor positioning and an online phase of performing indoor positioning according to an embodiment of the present disclosure.
4 is a diagram for describing in detail an offline step in the indoor positioning method according to an embodiment of the present disclosure.
5 is a diagram for describing in detail an online step according to an embodiment of the present disclosure.
6 is a diagram for explaining a weighted center in indoor positioning according to an embodiment of the present disclosure.
7 is a view for explaining an exemplary indoor space in which indoor positioning is performed according to an embodiment of the present disclosure.
8 shows a comparison of a fingerprint database according to an embodiment of the present disclosure with a conventional fingerprint database.
9 is a flowchart illustrating a process of building a database for indoor positioning in an offline stage according to an embodiment of the present disclosure.
10 is a flowchart illustrating a process of performing indoor positioning in an online step according to an embodiment of the present disclosure.
11 to 13 are graphs for explaining the superiority of the method according to an embodiment of the present disclosure compared to the conventional method.

Advantages and features of the present invention, and methods for achieving them will become apparent with reference to the detailed description in conjunction with the accompanying drawings. However, it should be understood that the present invention is not limited to the embodiments presented below, but may be implemented in a variety of different forms, and includes all transformations, equivalents, and substitutes included in the spirit and scope of the present invention. . The embodiments presented below are provided to complete the disclosure of the present invention, and to fully inform those of ordinary skill in the art to the scope of the present invention. In describing the present invention, if it is determined that a detailed description of a related known technology may obscure the gist of the present invention, the detailed description thereof will be omitted.

The terms used in the present application are only used to describe specific embodiments, and are not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise. In the present application, the terms include or have is intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features, number, step , it should be understood that it does not preclude in advance the possibility of the existence or addition of an operation, component, part, or combination thereof. Terms such as first, second, etc. may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

Hereinafter, embodiments according to the present invention will be described in detail with reference to the accompanying drawings, and in the description with reference to the accompanying drawings, the same or corresponding components are given the same reference numerals, and overlapping descriptions thereof are omitted. decide to do

First, the present embodiment relates to a two-step fingerprinting-based indoor positioning method for increasing the accuracy of indoor positioning. In this embodiment, although indoor positioning may be performed using a beacon as an access point, the present invention is not limited thereto, and various wireless communication modules may be applied.

Indoor positioning can be divided into an offline process (location learning process) and an online process (measurement process). In general, in RF-based positioning such as WiFi and Bluetooth beacons, fingerprint data in the form of an RF signal is measured in an offline process, it is stored in a database, and an RF signal measured at the current location by a user carrying a positioning terminal in the online process. can estimate the current position by comparing it with the RF signal stored in the database. In this case, the mainly used fingerprint data is an RF signal strength value (RSS), but in this embodiment, a weighted centroid (WC) in addition to the RSS may be additionally used as the fingerprint data to estimate the current location.

That is, in the offline process, RSS and WC measured at a reference point (RP) may be stored in the DB as data for comparison. On the other hand, in the online process, the current location can be estimated by comparing the stored RSS and WC information with data measured by the user terminal at a specific location. In this case, the reference point may be selected through a clustering process, and thus a smaller number of reference points may be used compared to the prior art.

In other words, in the present embodiment, a received signal strength (RSS) of a beacon may be converted into a distance using a propagation model, and a weighted center (WC) of a neighboring beacon may be estimated. And in this embodiment, instead of storing the RSS of all beacons placed indoors, by storing the estimated WC along with the ranking of beacons based on the distance of RSS and surrounding beacons in the server database, a database for indoor positioning It can be built and used to perform indoor positioning.

In addition, the indoor positioning database construction and indoor positioning method according to an embodiment of the present disclosure are performed by a computer, and the computer stores a computer program that functions the computer to perform the indoor positioning database construction and indoor positioning method.

In addition, the computer refers to a computing device in a broad sense including a smart device (user terminal) such as a smart phone or a tablet PC as well as a general personal computer.

In addition, the computer program may be provided by being stored in a separate recording medium, and the recording medium may be specially designed and configured for the present invention or may be known and used by a person skilled in the field of computer software. .

For example, the recording medium includes a hard disk, a magnetic medium such as a floppy disk and a magnetic tape, an optical recording medium such as a CD and DVD, a magneto-optical recording medium capable of both magnetic and optical recording, a ROM, a RAM, and a flash memory. and the like, alone or in combination, may be a hardware device specially configured to store and execute program instructions.

In addition, the computer program may be a computer program composed of program instructions, a local data file, a local data structure, etc. alone or in combination, and may be executed by a computer using an interpreter as well as machine code such as generated by a compiler. It may be a computer program written in a high-level language code that can be

In addition, the computer program may be stored in a server capable of transmitting the computer program through a communication network.

1 is a schematic illustration of an indoor positioning system according to an embodiment of the present disclosure.

Referring to FIG. 1 , the indoor positioning system may include a wireless receiving device 10 , a server 20 , a wireless transmitting device 30 , and a network 40 .

In this embodiment, the wireless receiving device 10 is a device capable of receiving a signal having a radio signal strength (RSS) from the wireless transmitting device 30, and can perform positioning while moving indoors and receive a positioning result. It may be a user terminal with Also, in this embodiment, the server 20 may be connected to the wireless receiving device 10 and/or the wireless transmitting device 30 through the network 40 .

Here, the computer program may be stored in the wireless receiving device 10 or stored in the server 20 . If the computer program is stored in the wireless receiving device 10, offline data is stored in the server 20, and the wireless receiving device 10 receives the offline data from the server 20 and the The location of the wireless receiving device 10 may be determined.

However, the offline data may be stored in the wireless receiving device 10 , and in this case, the location can be determined only by the wireless receiving device 10 without the need for the server 120 . In addition, when the computer program is stored in the server 20, the server 20 can receive the RSS from the wireless reception device 10 to position the location of the wireless reception device 10, and the positioning result may be transmitted to the wireless receiving device 10 .

The wireless receiving device 10 receives a wireless signal transmitted from a plurality of wireless transmitting devices 30 and based on this, a weighted centroid localization (WCL) technology and fingerprinting using a radio map (Radio Map) (Fingerprinting) technology can be combined to perform a function of estimating one's current location.

In addition, the wireless receiving device 10, for example, the user terminal may be provided with a service for operating or controlling the indoor positioning system through an authentication process after accessing the indoor positioning application or the indoor positioning site. In this embodiment, the user terminal that has completed the authentication process may operate the indoor positioning system and control the operation of the indoor positioning system.

In this embodiment, the user terminal is a desktop computer, a smartphone, a notebook computer, a tablet PC, a smart TV, a mobile phone, a personal digital assistant (PDA), a laptop, a media player, a micro server, a global positioning system (GPS) device operated by a user , e-book terminals, digital broadcast terminals, navigation devices, kiosks, MP3 players, digital cameras, home appliances, and other mobile or non-mobile computing devices, but is not limited thereto. In addition, the user terminal may be a wearable terminal such as a watch, glasses, a hair band and a ring having a communication function and a data processing function. The user terminal is not limited to the above, and a terminal capable of web browsing may be borrowed without limitation.

The server 20 may be a server for operating an indoor positioning system. In addition, the server 20 may be a database server that provides big data necessary for applying various artificial intelligence algorithms and data for operating the indoor positioning system. In addition, the server 20 may include a web server or an application server for remotely controlling the operation of the indoor positioning system using an indoor positioning application installed in the wireless receiving device 10 or an indoor positioning web browser.

Here, artificial intelligence (AI) is a field of computer science and information technology that studies how computers can do the thinking, learning, and self-development that can be done with human intelligence. It could mean making it possible to imitate intelligent behavior.

Also, artificial intelligence does not exist by itself, but has many direct and indirect connections with other fields of computer science. In particular, in modern times, attempts are being made to introduce artificial intelligence elements in various fields of information technology and use them to solve problems in that field.

Machine learning is a branch of artificial intelligence, which can include fields of study that give computers the ability to learn without explicit programming. Specifically, machine learning can be said to be a technology that studies and builds a system and an algorithm for learning based on empirical data, making predictions, and improving its own performance. Algorithms in machine learning can take the approach of building specific models to make predictions or decisions based on input data, rather than executing strictly set static program instructions.

The server 20 may receive and analyze service request information from the indoor positioning system, generate service response information corresponding to the service request information, and transmit it to the indoor positioning system.

A plurality of wireless transmission devices 30 may be provided, may be installed at a predetermined specific location in the room, and may perform a function of generating and transmitting a wireless signal. The wireless transmitting device 30 is preferably composed of, for example, a beacon signal, that is, a beacon generator that transmits a beacon message as a specific RF (Radio Frequency) signal, but is not limited thereto. For example, ZigBee, Bluetooth (Bluetooth), UWB (Ultra-Wideband), RFID (Radio Frequency Identification), Wi-Fi or infrared communication (Infrared Data Association, IrDA) using any one of the short-range wireless communication method It may be implemented as a device, an access point (AP), a base station, or the like.

The network 40 may serve to connect the wireless receiving device 10 , the server 20 , and the wireless transmitting device 30 in the indoor positioning system. Such a network 40 is, for example, a wired network such as local area networks (LANs), wide area networks (WANs), metropolitan area networks (MANs), integrated service digital networks (ISDNs), or wireless LANs, CDMA, Bluetooth, Wi- It may encompass a wireless network such as Fi or satellite communication, but the scope of the present invention is not limited thereto. In addition, the network 40 may transmit and receive information using short-distance communication and/or long-distance communication. Here, the short-distance communication may include Bluetooth, radio frequency identification (RFID), infrared data association (IrDA), ultra-wideband (UWB), ZigBee, and Wireless fidelity (Wi-Fi) technologies. Communication may include code division multiple access (CDMA), frequency division multiple access (FDMA), time division multiple access (TDMA), orthogonal frequency division multiple access (OFDMA), single carrier frequency division multiple access (SC-FDMA) technology. can

Network 40 may include connections of network elements such as hubs, bridges, routers, switches, and gateways. Network 40 may include one or more connected networks, eg, multiple network environments, including public networks such as the Internet and private networks such as secure enterprise private networks. Access to network 40 may be provided via one or more wired or wireless access networks. Furthermore, the network 40 may support an Internet of Things (IoT) network and/or 5G communication that exchanges and processes information between distributed components such as things.

2 is a block diagram schematically illustrating an indoor positioning system according to an embodiment of the present disclosure.

As shown in FIG. 2 , the indoor positioning system may include a communication interface 100 , a sensing unit 200 , a processor 300 , a memory 400 , and a processing unit 500 .

The communication interface 100 interworks with the network 40 to provide a transmission/reception signal between the wireless receiving device 10, the wireless transmitting device 30, and/or the server 20 in the indoor positioning system in the form of packet data. can be In addition, the communication interface 100 may support various things intelligent communication (internet of things (IoT), internet of everything (IoE), internet of small things (IoST), etc.), and M2M (machine to machine) communication, V2X (vehicle to everything communication) communication, D2D (device to device) communication, etc. may be supported.

The sensing unit 200 may be a sensing module for receiving a wireless signal. In addition, the sensing unit 200 may include a camera module, a Light Imaging Detection and Ranging (LIDAR), an ultrasonic sensor, a Radio Detection and Ranging (RADAR), and an infrared sensor. The sensing unit 200 may sense environmental information around the wireless receiving device 10 through a plurality of sensor modules. According to an embodiment, the sensing module may further include other components in addition to the described components, or may not include some of the described components.

The processor 300 is a kind of central processing unit and may control the operation of the entire indoor positioning system by driving control software mounted on the memory 400 .

In this embodiment, the processor 300 measures the RSS from the access points measured at the reference point of the indoor space in the offline step of building the database for indoor positioning, and a predetermined number of accesses in the order of the size of the RSS. The points may be selected as access points associated with the reference point. In addition, the processor 300 may determine the WC of the related access points at the reference point based on the RSS from the related access points. In addition, the processor 300 may store information of the related access points, RSS from the related access points, and the WC of the related access points in association with the location information of the reference point, and store the above process in the reference points of the indoor space. can be repeated every time.

Then, the processor 300 measures RSS from access points measured at a first location, which is an arbitrary location in an indoor space, in an online step for indoor positioning, and selects a certain number of access points in order of increasing RSS size. The first access points associated with the first location may be selected. And the processor 300 may determine the WC of the first access points in the first location based on the RSS from the first access points. In addition, the processor 300 includes information of the access points related to each reference point of the indoor space, RSS from the related access points, and data of the first access points and data of the database in which the WCs of the access points related to the reference point are stored. By comparing RSS from the first access points and the WC of the first access points at the first location, location information of the first location may be estimated.

Meanwhile, the processor 300 may refer to, for example, a data processing device embedded in hardware having a physically structured circuit to perform a function expressed as a code or an instruction included in a program. As an example of the data processing apparatus embedded in the hardware as described above, a microprocessor, a central processing unit (CPU), a processor core, a multiprocessor, an application-specific integrated (ASIC) circuit) and a processing device such as a field programmable gate array (FPGA), but the scope of the present invention is not limited thereto.

In addition, in the present embodiment, the processor 300 includes information of the related access points of each reference point stored in the database and RSS from the related access points of each reference point and the WC of the related access points at each reference point. training data can be generated. In this case, the training data may be labeled with location information of each reference point. Also, the processor 300 may train the constructed initial neural network model as a neural network model for indoor positioning using training data. In this case, the neural network model for indoor positioning may be configured to output location information of a specific location when information of related access points at a specific location, RSS from related access points, and WC of related access points at a specific location are input.

That is, the processor 300 may perform machine learning such as deep learning, and the memory 400 may store data used for machine learning, result data, and the like.

Here, deep learning technology, which is a type of machine learning, can learn by going down to a deep level in multiple stages based on data. Deep learning can represent a set of machine learning algorithms that extract core data from a plurality of data as the level increases.

The deep learning structure may include an artificial neural network (ANN). For example, the deep learning structure is composed of a deep neural network (DNN) such as a convolutional neural network (CNN), a recurrent neural network (RNN), and a deep belief network (DBN). can be The deep learning structure according to the present embodiment may use various well-known structures. For example, the deep learning structure according to the present invention may include CNN, RNN, DBN, and the like. RNN is widely used in natural language processing and the like, and it is an effective structure for processing time-series data that changes with time, and it is possible to construct an artificial neural network structure by stacking layers at every moment. DBN may include a deep learning structure composed of multiple layers of restricted boltzman machine (RBM), a deep learning technique. By repeating RBM learning, when a certain number of layers is reached, a DBN having the corresponding number of layers can be configured. CNNs can include models that simulate human brain functions based on the assumption that when a person recognizes an object, it extracts the basic features of an object, then performs complex calculations in the brain and recognizes an object based on the result. .

On the other hand, learning of the artificial neural network can be accomplished by adjusting the weight of the connection line between nodes (and adjusting the bias value if necessary) so that a desired output is produced with respect to a given input. In addition, the artificial neural network may continuously update a weight value by learning. In addition, a method such as back propagation may be used for learning the artificial neural network.

That is, the indoor positioning system may be equipped with an artificial neural network, and the processor 300 may include an artificial neural network, for example, a deep neural network (DNN) such as CNN, RNN, DBN, etc. . As a machine learning method of such an artificial neural network, both unsupervised learning and supervised learning may be used. The processor 150 may control the artificial neural network structure to be updated after learning according to a setting.

The memory 400 may be connected to one or more processors 300 to store codes that, when executed by the processor 300 , cause the processor 300 to support various functions of the indoor positioning system.

That is, the memory 400 may store a plurality of application programs (or applications) driven in the indoor positioning system, information for the operation of the indoor positioning system, and commands. At least some of these applications may be downloaded from an external server through wireless communication.

In this embodiment, the memory 400 may perform a function of temporarily or permanently storing data processed by the processor 300 . Here, the memory 400 may include magnetic storage media or flash storage media, but the scope of the present invention is not limited thereto. Such memory 400 may include internal memory and/or external memory, and may include volatile memory such as DRAM, SRAM, or SDRAM, one time programmable ROM (OTPROM), PROM, EPROM, EEPROM, mask ROM, flash ROM, NAND flash memory, or non-volatile memory such as NOR flash memory, SSD, compact flash (CF) card, SD card, Micro-SD card, Mini-SD card, Xd card, or flash drive such as a memory stick , or a storage device such as HDD.

Meanwhile, in the present embodiment, the memory 400 may include a radio map database 570, which will be described later, but may be provided separately.

The processing unit 500 may be provided outside the processor 300 as shown in FIG. 2 , may be provided inside the processor 300 to operate like the processor 300 , and may be provided inside the server 20 of FIG. 1 . may be provided. Hereinafter, a detailed operation of the processing unit 500 for indoor positioning will be described with reference to FIGS. 3 to 5 .

3 is a diagram for explaining the overall flow of an offline phase of building a database for indoor positioning and an online phase of performing indoor positioning.

The proposed method of this embodiment may consist of two phases: a training/offline phase and an execution/online phase, and in both phases, the scanned RSS A propagation model can be used to transform values into distances.

The propagation model used here may be in the form of Equation 1 below.

는 거리 d 에서 dBm 단위의 RSS이고, A는 표준 거리(1m로 정의됨)에서 수신되는 RSS이며,

는 평균 0과 분산

를 갖는 가우시안 분포 랜덤 변수이고, n 은 경로 손실 지수(exponent)이다. here,

is the RSS in dBm at distance d , A is the RSS received at a standard distance (defined as 1m),

is the mean 0 and variance

is a Gaussian distributed random variable with , and n is the path loss exponent.

And in a preferred embodiment, the following Equation 2 can be derived from Equation 1.

를 획득하는데 사용될 수 있다.Here, the measured distance d is a weight used to calculate a weighting center (WC), which will be described later.

can be used to obtain

In this regard, FIG. 6 for explaining a method of obtaining a weighted center in indoor positioning according to an embodiment of the present disclosure exemplifies a case of obtaining a weighted center when there are three beacons.

Through previously known positions of the access points and their respective weights, a weighting center may be calculated by Equation 3 below.

Here, (x _w , y _w ) is the estimated weighting center, (x _j , y _j ) is the previously known coordinates of the access point j, and d _j is the reference point or access point j in the user terminal for indoor positioning. is the distance to , g is the degree of weight, and u is the number of access points (eg, beacons) considered for weight centroid estimation.

에 N 개의 기준 지점(RP)들이 포함될 수 있으며, B 개의 비콘(무선 송신 기기, 30)들의 RSS 데이터는

RP에서 q 번 획득될 수 있다. 이는,

와 같이 나타낼 수 있으며, 여기서,

이다.Meanwhile, referring to FIG. 7 for explaining an exemplary indoor space in which indoor positioning is performed according to an embodiment of the present disclosure, in the offline stage, the indoor space (testbed) is divided into N virtual grids of the same size, and the center of each grid may indicate the reference point RP. Therefore, the indoor space

may include N reference points (RP) , and RSS data of B beacons (wireless transmitting device, 30)

It can be obtained q times in RP. this is,

It can be expressed as, where

to be.

That is, in the present embodiment, RSS measured from each access point (APs) may be measured a plurality of times, and an average value of RSS measured a plurality of times (eg, q times) may be calculated. You can select from RSS values from access points. That is, errors can be reduced by using methods such as averaging or moving average.

Here, RSS data of the B beacons (wireless transmitting device, 30) may be used in the clustering process. In this embodiment, the clustering process may be referred to as affinity propagation clustering (APC).

At this time, only u beacons may be used for the fingerprinting information in the order of the RSS data largest (here, u < B). It should be noted that RSS data is obtained from four directions at a reference point and the user terminal can be held in the “texting/messaging” position.

The radio map formed by RSS acquisition in the offline stage can be expressed as Equation 4 below.

는

RP(

)에서 비콘(무선 송신 기기, 30) b 로부터 얻어진 평균 RSS일 수 있다. here,

is

RP(

) may be the average RSS obtained from the beacon (wireless transmitting device, 30) b.

In addition, in the offline stage, the rankings of the access points may be recorded in the order of increasing RSS measured at the reference point.

Referring to FIG. 8 illustrating a comparison between a conventional fingerprint database and a fingerprint database according to an embodiment of the present disclosure, characteristics of the fingerprint database according to an embodiment of the present disclosure can be seen more clearly.

As shown in Fig. 8(a), the conventional fingerprint database includes the coordinates of the reference point and RSS from all beacons observed at the reference point, whereas in the embodiment of the present disclosure as in Fig. 8(b). The corresponding fingerprint database may include the coordinates of the reference point, the RSS and ranking of u beacons adjacent to the reference point, and data on the weighting centers of the u beacons.

Accordingly, since the embodiment of the present disclosure uses only the necessary RSS of beacons, it is possible to enable more accurate positioning while saving computational resources.

That is, in the offline stage, information on u access points, ranking of u access points, and weight centers related to u access points are recorded in the database in order of increasing RSS values based on the RSS measured at each reference point. can be Such a database may be stored in the server 20 and then used by communication with the wireless receiving device 10 or the user terminal during indoor positioning, or may be stored in the wireless receiving device 10 or the user terminal itself.

과 같이 나타날 수 있다. Next, in the online stage, online RSS data observed at an unknown location r' in the room is

may appear as

이며, t는 시간(time instant)이고, EA는 지수 평균(exponential averaging)일 수 있으며, 이를 통해 2개의 관측이 시작되자마자 바로 평탄화된(smoothed) RSS를 얻을 수 있다.here,

, where t is time instant, and EA can be exponential averaging, so that as soon as two observations start, a smoothed RSS can be obtained.

_{The EA} for the RSS data sample at the instant t(RSS EA ) may be given by Equation 5 below.

는 평탄화 팩터(smoothing factor)로 0과 1 사이에서 결정될 수 있다.

가 클수록 평탄화 레벨은 감소하고,

가 0에 가까워질수록 평탄화 효과는 커지며, 최근 RSS 관측에 대해 보다 덜 민감하게 반응하게 된다.here,

may be determined between 0 and 1 as a smoothing factor.

The higher the value, the lower the flattening level

The closer to 0, the greater the flattening effect, and the less sensitive it is to recent RSS observations.

After measuring the RSS observed at position r', one can define a set H of cluster heads and define the number of reference points that are members of the cluster as N'(N'<N). First, the cluster head RP _H may be determined by Equation 6 below.

은 알려지지 않은 r' 위치에서 근접한 u' 개의 비컨들로부터의 관측된 가중 중심이고,

은 m번째 클러스터-헤드 RP의 저장된 가중치 중심을 나타낸다.here,

is the observed weighting center from u' adjacent beacons at an unknown r' location,

denotes the stored weight center of the m-th cluster-head RP.

After the cluster head RP is selected, Euclidean distances can be calculated according to Equations 7 and 8 below based on the online-observed fingerprint data and the stored fingerprint data for N' RPs.

where p is a member of the selected cluster head.

N' RPs are _{arranged in ascending order of D wcl} , and the first k RPs (positions known as J _z [x _z , y _z ]) are selected to estimate _{T WC through Equation 9 below.}

A similar operation may be _{repeated using D RSSl to estimate T RSS} using the relationship in Equation 10 (where u in Equation 8 is the stored beacon in RP for the _{minimum D WCl ).}

The position of the user terminal in an arbitrary position may be finally estimated using Equation 11 below.

이고,

는 매칭된 비콘의 총 수, 즉, 위치 r'에서 최소 D_wcl을 가지는 RP에 대한 저장된 비콘들의 랭킹과 매칭하는 온라인-관측된 비콘들의 개수를 나타낸다.here,

ego,

denotes the total number of matched beacons, i.e., the number of online-observed beacons that match the ranking of stored beacons for the RP with _{minimum D wcl at position r'.}

이고 최소 D_wcl을 가지는 RP에서의 비콘들의 저장된 랭킹은

일 수 있다. 이러한 경우, 온라인의 정렬된 랭킹과 오프라인의 정렬된 랭킹 사이에 단 하나의 비콘(세 번째 비콘)만 매칭되므로

는 1이다.For example, if beacons observed online at unknown location r' (eg, where u=4) are sorted in descending order according to their ranking.

ego The stored ranking of beacons in the RP with a minimum D _{wcl is}

can be In this case, only one beacon (the third beacon) matches between the online sorted ranking and the offline sorted ranking, so

is 1

에 추가로 기초하여 r'의 위치 정보를 연산할 수 있다.That is, among the members of the cluster head, the WC of the corresponding reference point is the beacons of the beacons observed online at the location r', the beacons of the reference point closest to the WC and the beacons observed online at the location r' are listed in order according to the RSS The number of beacons that match each other when

It is possible to calculate the position information of r' based in addition to .

4 is a diagram for describing in detail an offline step in the indoor positioning method according to an embodiment of the present disclosure. Also, FIG. 5 is a diagram for describing in detail an online step according to an embodiment of the present disclosure.

The processing unit 500 includes a BLE scan unit 510 , an extraction unit 520 , a clustering unit 530 , a BLE selection unit 540 , a ranking determination unit 550 , a WC estimation unit 560 , and a radio map database 570 . ), a cluster head estimator 580 and a position estimator 590 may be included.

At this time, the BLE scan units 510-1 and 510-2, the extractors 520-1 and 520-2, the BLE selectors 540-1 and 540-2, and the WC estimators 560-1 and 560-2 ) and radio map databases 570-1 and 570-2 are divided into an offline stage and an online stage for convenience of explanation, but may be implemented in the same component.

In the online step, the BLE scan unit 510-1 may measure Received Signal Strength (RSS) from access points measured at a reference point (RP) in an indoor space. In this embodiment, the RSS value may be measured using a propagation model, and the RSS value may be converted into a distance.

The extraction unit 520-1 calculates the average value of the RSS values to extract a standardized feature, and the clustering unit 530 calculates clustering data using APC based on the feature information, and the BLE selector 540-1 may select a predetermined number of BLE beacons in the order of increasing average RSS.

The clustering unit 530 may divide a series of elements into clusters and select a representative cluster head for each cluster. In this embodiment, the clustering unit 530 may allocate the same opportunity to become a cluster head to all elements.

The ranking determiner 550 determines a ranking for the selected BLE beacons according to the averaged RSS size, and the WC estimator 560 calculates weighted center data for the selected BLE beacons.

These data are stored in the radio map database 570-1, so that they can be utilized for indoor positioning.

In the online phase after the offline phase, referring to FIG. 5 , a user terminal located at an arbitrary position r' may measure RSS from beacons through the BLE scan unit 510 - 2 .

The extractor 520-2 obtains an averaged RSS using the RSS data measured multiple times, and the BLE selector 540-2 selects u' adjacent r' BLE beacons in the order of the averaged RSS size. have.

The WC estimator 560 - 2 calculates a first weight center of the selected beacons, and the cluster head estimator 580 uses the radio map database 570 - 2 to use the first weight center among cluster heads stored in the offline stage. Estimate the cluster head having the weighted center closest to .

The position estimator 590 may estimate the position of r' by using the weight center information and the RSS information in the estimated cluster head.

9 is a flowchart illustrating a process of building a database for indoor positioning in an offline stage according to an embodiment of the present disclosure.

First, the RSS from the access points is measured at the set collimation points of the indoor space (S110). Here, RSS may be a value obtained by averaging a plurality of measurement values instead of one measurement.

According to the size of the measured RSS, a predetermined number of access points may be selected as access points (beacons) related to the reference point in the order of increasing size ( S120 ).

A distance from a reference point to each of the access points may be obtained based on the RSS from the selected related access points, and a weighting center of the related access points may be determined using the distance ( S130 ).

During this process, clustering may be performed using APC, and a cluster head may be determined among reference points in the cluster.

Information of the related access points, RSS (ranking information according to RSS) from the related access points, and the WC of the related access points are stored in the database in association with the location information of the reference point (S140).

Here, the above processes may be performed by a processor of a device (mobile type) for building a database for indoor positioning, and the database may be built on an external server such as a local server.

10 is a flowchart illustrating a process of performing indoor positioning in an online step according to an embodiment of the present disclosure.

After the database for indoor positioning is built in the offline step, the user terminal may measure RSS from the access points at any first location in the indoor space (S210). Here, RSS may be a value obtained by averaging a plurality of measurement values instead of one measurement.

The user terminal may select a predetermined number of access points in the order of increasing the size of the measured RSS as the first access points related to the first location (S220).

A distance from the first location to the first access points may be determined based on the RSS from the selected first access points, and a WC of the first access points at the first location may be determined based on the distance ( S230 ).

The location information of the first location may be estimated by comparing the data of the database built in the offline step with information of the first access points, RSS from the first access points, and WC of the first access points ( S240 ).

Here, the information of the first access points may be IDs of the access points, RSS rankings, and the like.

The above process may be performed by a processor of the user terminal.

11 to 13 are graphs for explaining the superiority of the method according to an embodiment of the present disclosure compared to the conventional method.

11 to 13 are data obtained by performing positioning in various indoor spaces and measuring positioning errors. The method proposed in the present disclosure is superior to the case of NN, Wk-NN, and Wk-NN+APC, which are conventional methods. performance can be seen.

In addition, the proposed system can reduce the physical size of the database and the amount of data communication with the server at the execution stage by increasing the localization accuracy by using various fingerprinting functions. In addition, affinity propagation clustering can minimize the search space of RP, reduce computational cost, and smooth out noisy RSS through exponential averaging.

The embodiments according to the present invention described above may be implemented in the form of a computer program that can be executed through various components on a computer, and such a computer program may be recorded in a computer-readable medium. In this case, the medium includes a hard disk, a magnetic medium such as a floppy disk and a magnetic tape, an optical recording medium such as CD-ROM and DVD, a magneto-optical medium such as a floppy disk, and a ROM. , RAM, flash memory, and the like, hardware devices specially configured to store and execute program instructions.

Meanwhile, the computer program may be specially designed and configured for the present invention, or may be known and used by those skilled in the computer software field. Examples of the computer program may include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.

In the specification of the present invention (especially in the claims), the use of the term "above" and similar referential terms may be used in both the singular and the plural. In addition, when a range is described in the present invention, each individual value constituting the range is described in the detailed description of the invention as including the invention to which individual values belonging to the range are applied (unless there is a description to the contrary). same as

The steps constituting the method according to the present invention may be performed in an appropriate order, unless there is an explicit order or description to the contrary. The present invention is not necessarily limited to the order in which the steps are described. The use of all examples or exemplary terms (eg, etc.) in the present invention is merely for the purpose of describing the present invention in detail, and unless defined by the claims, the scope of the present invention is limited by the examples or exemplary terminology. it's not going to be In addition, those skilled in the art will recognize that various modifications, combinations, and changes may be made in accordance with design conditions and factors within the scope of the appended claims or their equivalents.

Therefore, the spirit of the present invention should not be limited to the above-described embodiments, and the scope of the spirit of the present invention is not limited to the scope of the scope of the present invention. will be said to belong to

10: wireless receiving device (user terminal)
20 : Server
30: wireless transmitting device (access point)
40: network
100: communication interface
200: sensing unit
300 : processor
400 : memory
500: processing unit
510 (510-1, 510-2): BLE scan unit
520 (520-1, 520-2): extraction unit
530: clustering unit
540 (540-1, 540-2): BLE selector
550: ranking decision unit
560: WC (weighted centroid) estimator
570(570-1, 570-2): Radio map database
580: cluster head estimator
590: location estimation unit

Claims (15)

A) measuring Received Signal Strength (RSS) from access points measured at a reference point in an indoor space;
B) selecting a predetermined number of access points as the access points related to the reference point in the order of the size of the RSS;
C) determining a weighted centroid (WC) of the associated access points at the reference point based on RSS from the associated access points;
E) storing the information of the related access points and RSS from the related access points and the WC of the related access points in association with the location information of the reference point in a database; and
F) comprising repeating steps A) to E) for each reference point of the indoor space,
How to build a database for indoor positioning.
The method of claim 1,
Determining the WC of the related access points comprises:
calculating a distance from the reference point to each of the associated access points based on the RSS from the associated access points; and
determining the WC of the related access points at the reference point based on the calculated distance;
How to build a database for indoor positioning.
The method of claim 1,
After step F),
generating training data comprising information of each reference point's associated access points stored in the database and RSS from each reference point's associated access points and the WC of the associated access points at each reference point and,
The training data is labeled with the location information of each reference point,
How to build a database for indoor positioning.
4. The method of claim 3,
After the generating step,
Further comprising the step of training the built initial neural network model as a neural network model for indoor positioning using the training data,
The neural network model for indoor positioning is configured to output location information of a specific location upon input of information of related access points at a specific location, RSS from related access points, and WC of related access points at the specific location,
How to build a database for indoor positioning.
The method of claim 1,
Measuring the RSS comprises:
measuring RSS measured from each of the access points a plurality of times, and calculating an average value of the RSS measured a plurality of times; and
selecting the average value as the RSS value from each of the access points;
How to build a database for indoor positioning.
measuring RSS from access points measured at a first location in an indoor space;
selecting a predetermined number of access points as first access points associated with the first location in an order of increasing RSS size;
determining the WC of the first access points in the first location based on the RSS from the first access points; and
Information of the access points related to each reference point of the indoor space, RSS from the related access points, and data of the database in which the WC of the related access points at the reference point are stored and information of the first access points, the first estimating location information of the first location by comparing RSS from access points and WC of the first access points at the first location;
Indoor positioning method.
7. The method of claim 6,
The database includes cluster information in which reference points are grouped,
The estimating step is
searching for a reference point having a WC closest to the WC of the first access points in the database; and
Based on the difference between the WC of the first access points and the WC at the reference points belonging to the cluster of the searched reference points and the difference between the RSS of the related access points at the reference points and the RSS of the first access points Comprising the step of calculating the location information of the first location,
Indoor positioning method.
7. The method of claim 6,
The step of calculating the location information of the first location,
Access that matches each other when the WC of the reference point among the reference points of the cluster is the closest to the WC of the first access points and the related access points of the reference point and the first access points are listed in order according to the RSS size calculating location information of the first location further based on the number of points;
Indoor positioning method.
9. The method of claim 8,
The step of estimating the location information of the first location,
Comprising the step of estimating the location information of the first location based on the neural network model for indoor positioning learned using the training data based on the initial neural network model built,
The neural network model for indoor positioning is configured to output location information of a specific location upon input of information of related access points at a specific location, RSS from related access points, and WC of related access points at the specific location,
Indoor positioning method.
7. The method of claim 6,
Measuring the RSS comprises:
measuring RSS measured from each of the access points a plurality of times, and calculating an average value of the RSS measured a plurality of times; and
selecting the average value as the RSS value from each of the access points;
Indoor positioning method.
a communication interface for receiving RSS from access points;
one or more processors; and
a memory coupled to the one or more processors;
The memory, when executed by the processor, causes the processor to:
measuring RSS from access points measured at a first location in an indoor space;
selecting a predetermined number of access points as the first access points related to the first location in the order of the size of the RSS,
determine the WC of the first access points in the first location based on the RSS from the first access points;
Information of the access points related to each reference point of the indoor space, RSS from the related access points, and data of the database in which the WC of the related access points at the reference point are stored and information of the first access points, the first comparing RSS from access points and WC of the first access points at the first location to store codes that cause to estimate location information of the first location,
indoor positioning system.
12. The method of claim 11,
The memory, when executed by the processor, causes the processor to:
Calculate a distance from the first location to each of the first access points based on the RSS from the first access points, and WC of the first access points at the first location based on the calculated distance storing codes that cause to determine
indoor positioning system.
12. The method of claim 11,
The memory, when executed by the processor, causes the processor to:
the first location based on training data comprising information of each reference point's associated access points stored in the database and RSS from the respective reference point's associated access points and the WC of the associated access points at each reference point stores the codes that cause to estimate the location information of
The training data is labeled with the location information of each reference point,
indoor positioning system.
14. The method of claim 13,
The memory, when executed by the processor, causes the processor to:
stores codes that cause to estimate the location information of the first location based on a neural network model for indoor positioning learned using the training data based on the initial neural network model built,
The neural network model for indoor positioning is configured to output location information of a specific location upon input of information of related access points at a specific location, RSS from related access points, and WC of related access points at the specific location,
indoor positioning system.
12. The method of claim 11,
The memory, when executed by the processor, causes the processor to:
Codes that measure the RSS measured from each access point a plurality of times, calculate an average value of the RSS measured a plurality of times, and select the average value as the RSS value from each access point to measure the RSS. to save,
indoor positioning system.

Images