KR100857248B1 - Apparatus and method for generating mobile position information and Apparatus and method for recognizing mobile position - Google Patents

Abstract

An apparatus and method for generating moving object position information and an apparatus and method for recognizing moving body position are provided. An apparatus for generating moving object position information includes: a signal obtaining unit configured to receive a plurality of signal strength information respectively received from a mobile node by at least one fixed node existing at a predetermined position; A preprocessor for smoothing a plurality of signal strength information received for each fixed node; And a feature vector processing unit for obtaining the average value of the smoothed plurality of signal strength information for each fixed node, and configuring a set of the average values for each fixed node as a feature vector, wherein the apparatus for recognizing a moving position includes: a predetermined position; A signal acquisition unit configured to receive a plurality of signal strength information respectively received by the at least one fixed node from the mobile node; A preprocessor for smoothing a plurality of signal strength information received for each fixed node; A feature vector processing unit obtaining a mean value of the smoothed plurality of signal strength information for each fixed node and configuring a set of the average values for each fixed node as a feature vector; And a location recognizing unit recognizing a relative position of the mobile node with respect to the fixed node based on the feature vector.

Radio Frequency (RF), Position Aware, Fixed Node, Mobile Node, Smoothing

Description

Apparatus and method for creating location and Apparatus and method recognizing location of mobile object}

1 is a block diagram showing the configuration of an apparatus for generating moving object position information according to an embodiment of the present invention.

2 is a block diagram showing the configuration of an apparatus for recognizing a moving position according to an embodiment of the present invention.

3 is a flowchart illustrating a process of generating moving object position information according to an exemplary embodiment of the present invention.

4 is a diagram illustrating a 4D feature vector value of a mobile node stored in a DB according to an embodiment of the present invention.

5 is a flowchart illustrating a moving position recognition process according to an embodiment of the present invention.

<Explanation of symbols on main parts of the drawings>

200: device for recognizing the position of the moving object

201: signal acquisition unit

202: preprocessing unit

203: feature vector processing unit

204: DB

205: control unit

206: location recognition unit

207: interface unit

The present invention relates to a device and a method for generating mobile location information and a device and method for recognizing mobile location, and more particularly, by analyzing a pattern of radio frequency (RF) signal strength information according to distance And an apparatus and method for generating location information that can reduce system construction cost by recognizing a user's location and using an inexpensive RF module, and an apparatus and method for recognizing a mobile body location.

Location-recognition technologies and systems currently being researched and implemented can be broadly classified into wide area-based location recognition methods and area-based location recognition methods according to location recognition areas.

The wide area-based location recognition method can recognize the location in a relatively wide area such as GPS and location recognition according to the access area of the mobile communication network base station. Location recognition method using signal strength and signal arrival time.

The method using the RF signal is a method of locating the position using the strength of the RF signal or the propagation delay of the RF signal. A typical method is a radar (RADAR) system developed by Microsoft.

The radar is an RF-based method for recognizing and tracking a user's location in a building. The radar measures user coordinates using signal strength information collected from multiple receivers.

The radar measures the signal strength and the signal-to-noise ratio of the signals transmitted by the WLAN devices in the access point based on the IEEE 802.11 WLAN environment and calculates the two-dimensional position of the WLAN devices in the indoor environment.

The use of the radar method has the advantage that the number of devices required for location measurement is small and the WLAN environment in the building is used without a separate device. However, since all the location measurement targets must support the WLAN, small devices or It is difficult to apply to devices with a limited power source such as a battery, and has a disadvantage of being relatively expensive.

In addition, a few companies sell systems that measure the position of RF signals by reaching the time of arrival, and Pinpoint's 3D-iD systems are known to be 1 to 3 meters in precision.

An infrared positioning system includes an active badge system that installs an infrared receiver on the ceiling of an office and attaches an active badge having a badge-type infrared transmitter to users.

Each of the active badges has a unique identification number, and when the identification number is periodically transmitted in the infrared, the infrared receivers installed on the ceiling detect the transmitted infrared signal and identify a specific user's location.

One or more infrared receivers in each office are installed in a fixed location to form a network, which receives the user's identification number and forwards it to location management software.

The active badge is equipped with a microprocessor and allows bidirectional communication. The relatively simple system configuration makes it possible to construct a positioning system at a low cost.

In addition, since only a unique identification number is transmitted, the signal generation time is very short, and since the initialization time is slightly different for each medium, a signal is rarely generated in several mediums in the same space at the same time.

However, infrared communication is only possible within a limited distance, is expensive to install and maintain, and performance deteriorates when sunlight shines in offices with windows.

In addition, as the number of users increases, not only the collision rate increases but also the system needs to be extended. It is mainly used to install only one receiver in a certain area and figure out which area the person wearing the badge is in.

In the method using ultrasonic waves, an ultrasonic transmitter called bat is attached to a person or an object, and an ultrasonic receiver is attached to the ceiling of an office.

The ultrasonic transmitter and the ultrasonic receiver each have a unique identification number, the ultrasonic receiver is connected to a network server, and the transmitter has a separate wireless transceiver.

Active Bats use triangulation to locate transmitters based on measured distances between transmitters and receivers.

That is, a transmitter attached to a person or an object generates an ultrasonic wave of a short pulse to measure the arrival time to the receiver on the ceiling, and multiply the arrival time by the sound velocity to easily calculate the distance between the transmitter and the receiver.

Measuring the distance between the transmitter and three or more receivers provides enough information to determine the three-dimensional position of the transmitter, so attach two or more transmitters to a person or object to determine the relative positions of the transmitters. You can also get directions.

Ultra-wide band (UWB) localization method is a method that requires much less infrastructure than the ultrasonic positioning system.

UWB transmits very short radio pulses continuously, occupying a broadband spectrum of several GHz while having very low power density.

RFID (Radio Frequency Identification) system is largely composed of RFID tag, reader, data processing system.

The RFID tag attached to a body, a thing, a building, etc. transmits the identification number of the object to the RFID reader by calling the RFID reader, and sends it to the data processing system to display necessary information on a terminal or peripheral device available to the user.

The RFID tag consists of a microchip and a coil antenna and is divided into active and passive types according to the presence or absence of a battery.

Generally, the passive method is chosen in consideration of the cost.

The tag communicates only when the reader calls, the reader consists of the RF module, control unit and coupling elements, and all readers are connected to the data processing system.

The ID is stored in an RFID tag attached to the object, and information about the object is obtained from a data processing system connected to a network by using the ID as a pointer.

On the other hand, there is also a method using a combination of RF and ultrasound.

Cricket, developed by MIT, is a handset-based positioning system that attaches a device called a beacon to the ceiling to send signals for distance measurement, and attaches a receiver to people or objects targeted for distance measurement.

The beacon simultaneously transmits an ultrasonic signal and an RF signal for distance measurement.

Since the ultrasonic wave and the RF signal have different propagation speeds, the RF signal arrives at the receiver first and the ultrasonic signal is received thereafter.

The time difference between the two signals can be used to measure the distance between the beacon and the receiver, and the distance between three or more beacons and the receiver can be measured using triangulation to calculate the position of a person or object.

Using an ultrasonic signal and an RF signal simultaneously provides the advantage that distance can be measured using the time difference of arrival of the two signals even if the time synchronization between the beacon and the receiver is not correct.

The various location recognition methods and techniques described above may be used in various ways depending on the application.

GPS and mobile network methods are advantageous in a wide area, but it takes time for location recognition, high communication costs, and cannot be used in areas with poor communication with GPS satellites and GPS receivers.

In particular, the basic error range of GPS is 5 ~ 10m, and the position recognition by mobile communication network base station is less accurate because the error range is over 500m.

In the case of the RFID-based method, the tag purchase cost is relatively inexpensive, but all the readers must be installed at the desired location, but the cost is also considerable. That is, it is not suitable as a location recognition method because of its low efficiency in terms of economics.

The method such as location recognition using infrared light has high precision but can be applied to R & D use or business model that requires precision because of high purchase price.

As described above, the existing research based on the local area recognizes the position by focusing on the correlation with the signal strength and distance.

However, since the relationship between the distance and the signal strength information is not necessarily inversely, this method has a limitation in obtaining a high recognition rate, and many existing studies have raised such a problem.

An object of the present invention is to provide a location information by analyzing a pattern of signal strength information according to a distance through a device and method for generating mobile location information and a device and method for recognizing mobile location.

Another object of the present invention is to reduce the cost of constructing and maintaining a location recognition system by using an inexpensive RF module through a device and method for generating mobile location information and a device and method for recognizing mobile location.

Still another object of the present invention is to be able to process a large area such as a parking lot as well as a narrow area such as an office or a home through an apparatus and method for generating moving object position information and an apparatus and method for recognizing a moving object position.

The objects of the present invention are not limited to the above-mentioned objects, and other objects that are not mentioned will be clearly understood by those skilled in the art from the following description.

In order to achieve the above object, the apparatus for generating mobile object position information according to an embodiment of the present invention obtains a signal in which at least one fixed node existing at a predetermined position receives a plurality of signal strength information received from the mobile node, respectively; part; A preprocessor for smoothing a plurality of signal strength information received for each fixed node; And a feature vector processing unit for obtaining the average value of the smoothed plurality of signal strength information for each fixed node and configuring the set of average values for each fixed node as a feature vector.

In order to achieve the above object, the apparatus for recognizing a moving object position according to an embodiment of the present invention is a signal acquisition unit for receiving a plurality of signal strength information received from the mobile node at least one fixed node existing at a predetermined position ; A preprocessor for smoothing a plurality of signal strength information received for each fixed node; A feature vector processing unit obtaining a mean value of the smoothed plurality of signal strength information for each fixed node and configuring a set of the average values for each fixed node as a feature vector; And a location recognizing unit recognizing a relative position of the mobile node with respect to the fixed node based on the feature vector.

In order to achieve the above object, the method for generating mobile object position information according to an embodiment of the present invention obtains a signal that receives a plurality of signal strength information received from the mobile node at least one fixed node existing at a predetermined position, respectively step; A preprocessing step of smoothing the plurality of signal strength information received for each fixed node; And a feature vector processing step of obtaining the average value of the smoothed plurality of signal strength information for each fixed node and configuring the set of average values for each fixed node as a feature vector.

In order to achieve the above object, a method of recognizing a moving object position according to an embodiment of the present invention is a signal acquisition step of receiving a plurality of signal strength information received from the mobile node at least one fixed node existing at a predetermined position, respectively; ; A preprocessing step of smoothing the plurality of signal strength information received for each fixed node; A feature vector processing step of obtaining the average value of the smoothed plurality of signal strength information for each fixed node and configuring the set of average values for each fixed node as a feature vector; And a location recognizing step of recognizing a relative position of the mobile node with respect to the fixed node based on the feature vector.

Specific details of other embodiments are included in the detailed description and the drawings.

Advantages and features of the present invention and methods for achieving them will be apparent with reference to the embodiments described below in detail with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but can be implemented in various different forms, and only the embodiments make the disclosure of the present invention complete, and the general knowledge in the art to which the present invention belongs. It is provided to fully inform the person having the scope of the invention, which is defined only by the scope of the claims. Like reference numerals refer to like elements throughout.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram showing the configuration of an apparatus for generating moving object position information according to an embodiment of the present invention.

The apparatus 100 for generating mobile object position information according to an exemplary embodiment of the present invention includes a signal obtaining unit 101 for receiving a signal strength information received from a mobile node by a fixed node, and a preprocessor for smoothing the signal strength information. (102), a feature vector processor (103) constituting a feature vector based on the smoothed signal strength information, a DB (104) in which the feature vector is stored, and a controller (105) for controlling each device.

Hereinafter, the term '~ part' used in the embodiment of the present invention refers to software or a hardware component such as an FPGA or an ASIC, and '~ part' performs certain roles.

However, '~' is not meant to be limited to software or hardware.

 '~ Portion' may be configured to be in an addressable storage medium or may be configured to play one or more processors.

Thus, as an example, '~' means components such as software components, object-oriented software components, class components, and task components, and processes, functions, properties, procedures, and the like. Subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables.

The functionality provided within the components and the 'parts' may be combined into a smaller number of components and the 'parts' or further separated into additional components and the 'parts'.

In addition, the components and '~' may be implemented to play one or more CPUs in the device or secure multimedia card.

In an embodiment of the present invention, the fixed node and the mobile node are RF modules for detecting a change in the moving object, and the fixed node includes an RF receiver and receives an RF signal for the movement of the moving object transmitted from the mobile node through the RF receiver. Receives and transmits to the signal acquisition unit 101.

The fixed node is fixedly installed in a predetermined space and position, and the number is not limited to a specific number can be installed a plurality.

 The mobile node includes an RF transmitter and transmits an RF signal to the fixed node at predetermined time intervals according to a user-specified sensitivity and response time.

The mobile nodes may be installed or attached to a moving object to detect the movement, and the number of mobile nodes is not limited if it is possible to distinguish each mobile node.

The RF signal scheme of the fixed node and the mobile node may be an analog scheme or a digital scheme, and the present invention is not limited to any one scheme, but the digital scheme is used as a preferred embodiment.

When the fixed node receives the signal strength information transmitted from the mobile node, the corresponding information is stored in the signal acquisition unit 101.

The preprocessor 102 smoothes the signal strength information received from the signal acquirer 101 through the controller 105.

Smoothing is an operation that weakens or eliminates these fluctuations or discontinuities and smooths them when there is a slight variation or discontinuity in the data due to coarse sampling or noise. As a result, it may change little by little depending on the surrounding environment conditions, so use the signal value as many times as possible to generate stable signal value without noise.

In such a case, when the RF signal value is input in real time, the signal strength value may be irregularly input depending on the condition.

For example, if there is motion around a node, such as when a person passes by, the signal strength value may change irregularly.

In this case, a smoothing method is used as a preprocessing step to remove noise.

The feature vector processor 103 obtains an average value of the smoothed signal strength information transmitted from the preprocessor 102 through the controller 105 and configures the feature vector to use the average value as a vector element in a specific direction. The feature vector is normalized and stored in the DB 104 through the control unit 105.

In the present invention, the feature value is the minimum information necessary for distinguishing a specific mobile node from other mobile objects that are close to the mobile node. In the preferred embodiment of the present invention, RF signal strength information transmitted from the mobile node to fixed nodes at predetermined intervals is preferred. Is used as the feature value, and the set of feature values of the mobile node relative to the fixed node is defined as the feature vector.

In addition, in the present invention, the feature value is not limited to RF signal strength information transmitted by the mobile node, and may be used as the feature value as long as it is the minimum information necessary to distinguish a specific mobile node from other mobile objects in proximity to the mobile node.

2 is a block diagram showing the configuration of an apparatus for recognizing a moving position according to an embodiment of the present invention.

The apparatus 200 for recognizing a moving object position according to an exemplary embodiment of the present invention may include a signal obtaining unit 201 for receiving signal strength information received from a mobile node by a fixed node, and a preprocessor for smoothing signal strength information ( 202, the feature vector processor 203 constituting the feature vector based on the smoothed signal strength information, the DB 204 storing the feature vector, the controller 205 controlling each device, and the distance value based on the feature vector. And a location recognizing unit 206 for determining the location of the mobile node and an interface unit 207 showing the location of the mobile node on the screen.

The role of the signal acquisition unit 201, the preprocessor 202, the feature vector processor 203, and the DB 204 in the apparatus 200 for recognizing a moving object position according to an exemplary embodiment of the present invention is the moving object described with reference to FIG. 1. It is the same as the apparatuses 201 to 204 for generating location information.

The feature vector newly configured by the feature vector processing unit 203 is transmitted to the location recognition unit 206 through the control unit 205, and the newly constructed feature vector and DB (received from the vector processing unit 203) are transferred from the location recognition unit 206. The distance value and the position of the mobile node are recognized by comparing the existing feature vectors normalized and stored in step 204).

The distance value calculated by the position recognizing unit 206 and the position of the mobile node are displayed on the terminal screen which the user can see through the interface unit 207 via the control unit 205.

3 is a flowchart illustrating a process of generating moving object position information according to an exemplary embodiment of the present invention.

As mentioned in the part described with reference to FIG. 1, the number of fixed nodes and mobile nodes is not limited to a specific number, but for convenience of description, it is assumed that the number of fixed nodes is four and the number of mobile nodes is one.

Four fixed nodes receive one or more RF signal strength values from the mobile node once per second, and these values are input to the signal acquisition unit 101 (S301).

In this case, the RF signal strength transmitted from the mobile node is transmitted at predetermined time intervals, and the transmission time interval can be adjusted by the user. However, in the following embodiments of the present invention, the mobile node transmits one or more RF signals at least once per second. It is assumed that it transmits.

In step S301, when the RF signal strength value is input in real time, the RF signal strength value may irregularly enter according to a condition, and the preprocessor 202 removes the noise by a smoothing method through a preprocessing process for removing the noise ( S302).

In the smoothing method according to an embodiment of the present invention, a method of calculating an average value with a proximity signal value may be used and may be expressed as Equation (1).

Formula (1)

은 평활화를 한 결과 값이고

과

는 특징 벡터를 구성하는 요소 값으로서 특정 시간에 고정 노드로부터 받은 신호 값을 의미한다.

Is the result of smoothing

and

Denotes a signal value received from a fixed node at a specific time as an element value constituting the feature vector.

과

는

을 만족하는 파라미터이다.

and

Is

This parameter satisfies.

For example, assume that a fixed node receives the following RF signal strength values for a certain time.

In this case, the RF signal strength value may be a digital method, an analog method, and is not limited to a specific method.

Hereinafter, the present invention will be described using a digital method as a preferred embodiment of the value of the RF signal strength.

[206 202 200 208 204 208 204 206 202 204 202 206]

과

를 각각 1/2이라고 할 때, 위 값들을 근접 신호 값과의 평균값을 연산하는 평활화 방법을 통해 아래와 같은 결과값을 얻을 수 있다.

and

When each is 1/2, the following result can be obtained through the smoothing method of calculating the average of the above values with the proximity signal values.

[206 204 201 204 206 206 206 205 204 203 203 204]

In addition, through the process of calculating the sum of the distance between the signal strength value and the signal strength value received for a predetermined time, the smoothing can be performed by judging the distinguished signal value as noise and expressed by Equation (2). Can be.

Formula (2)

은 일정 시간동안 받은 신호 세기 값의 수이고

는 그 중

번째 신호 세기 값을 의미하며

는

번째 신호 세기 값과 나머지 신호 세기 값들의 거리의 합을 나타낸다.

Is the number of signal strength values received over time

Among them

Signal strength value

Is

The sum of the distance between the first signal strength value and the remaining signal strength values.

In this way, the position recognition rate can be increased even when noise occurs by smoothing the signal values obtained for a predetermined time in one fixed node.

 Since the number of fixed nodes is assumed to be 4, the same process is performed for the values obtained from the remaining 3 fixed nodes.

After S302, the vector information processing unit 103 configures the feature vector based on the average value of the values obtained after the smoothing in the preprocessing unit 102 (S303).

For example, suppose you have the following feature vectors:

는 특징 벡터를 구성하는 요소로서

번째 고정 노드에서 얻은 값에서 구한 요소 값이고

은 고정 노드의 수를 의미하며 본 발명의 실시예에서는 4가 된다.

Is a component of the feature vector.

The element value from the value obtained from the first fixed node

Denotes the number of fixed nodes and is 4 in the embodiment of the present invention.

는 식(3)과 같이 정의한다.Feature vector element

Is defined as Eq. (3).

Formula (3)

은 입력 받은 신호 값의 수(본 발명의 실시예의 경우 12)이며,　

는

번째 고정 노드에서 얻은

번째 신호 값을 의미한다. In equation (3)

Is the number of received signal values (12 for an embodiment of the present invention),

Is

Obtained from the first fixed node

The first signal value.

For example, the above smoothed values

[206 204 201 204 206 206 206 205 204 203 203 204]

요소 값은 204.3이 된다. Obtained from the values obtained from one fixed node as above.

The element value is 204.3.

값이 201.7, 184.3, 89.1이라고 한다면 결과적으로 특징 벡터 f는 다음과 같다. Obtained from the remaining three fixed nodes when there are four fixed nodes

If the values are 201.7, 184.3, and 89.1, the feature vector f is

f = (204.3 201.7 184.3 89.1)

For reference, the range of the signal value of the RF sensor module used in the present invention is not limited to a specific range, and in the present invention, the signal value of the RF sensor module is basically a signal value larger than 0 and smaller than 255 in a preferred embodiment. Assume that the RF sensor module is designed to produce.

After S303, the vector information processing unit 103 normalizes the signal value to a real value between 0 and 1 using the elements of the feature vector described above (S304), and stores the normalized values in the DB (S305).

In this case, the range of normalization values for normalization is not limited to only a range of specific values, and the present invention will be described using a real value between 0 and 1 as a preferred range of normalization values.

The equation for normalizing the signal value is shown in equation (4).

Formula (4)

는 정규화 이전의 값,

은 정규화한 값이며

은 최소 RF 신호 값이고

는 최대 RF 신호 값이다.

Is the value before normalization,

Is a normalized value

Is the minimum RF signal value

Is the maximum RF signal value.

Using the maximum and minimum signal values, all RF signal values are normalized to real values between 0 and 1.

For example, the above-described feature vector is as follows.

f = (204.3 201.7 184.3 89.1)

At this time, the feature vector fN normalized by Equation (4) is as follows.

fN  = (1.0 0.98 0.83 0)

If it is assumed that the apparatus for generating vehicle position information according to an embodiment of the present invention is used in a parking lot that can park 100 cars, in this case, the number of positions to be recognized is 100 because the positions of the cars will be recognized.

When four fixed nodes receiving RF signals from the car are installed in the parking lot, 100 four-dimensional feature vectors of the mobile node are obtained through the above-described processes (S301 to S305) as shown in FIG. Can be stored.

This is the modeling of knowledge for location recognition, which is the result of learning from an AI perspective.

In this way, the feature vector table for position recognition, that is, the learning model, is obtained in the learning process.

5 is a flowchart illustrating a process of recognizing a moving position according to an embodiment of the present invention.

The process of constructing and normalizing the feature vector to recognize the mobile body position and storing the feature vector in the DB 204 is the same as the process of generating the mobile body position information described with reference to FIG. 3 (S301 to S305).

The feature vector values normalized and stored in the DB 204 are used as a model for recognizing the position of the mobile node through the position recognition unit 206. The recognition algorithms include Euclidean distance, neural network, and SVM (Support Vector Machine). ) May be used.

The feature vector newly configured by the feature vector processor 203 to recognize the position of the moving object is transmitted to the position recognizer 206 through the controller 205 (S506),

The position recognizing unit 206 compares the newly constructed feature vector received from the vector processing unit 203 with the existing feature vector normalized and stored in the DB 204 (S507).

The distance value and the position of the mobile node are recognized (S508).

S501 to S503 correspond to a process of recognizing from an artificial intelligence perspective.

In other words, the feature vector extracted in the recognition process obtains a distance by comparing with the feature vector registered as a model, and in the present invention, the Euclidean distance is used as shown in Equation (5) below.

Formula (5)

는

번째 특징 벡터의

번째 벡터 요소 값을 의미한다. In equation (5)

Is

Of the first feature vector

Second vector element value.

는 이동 노드가 임의의 위치에 있을 때 고정 노드로부터 획득한 특징 벡터 중

번째 요소 값이며,

는 모형과 현재 얻은 특징 벡터 간의 유클리드 거리 값을 의미한다.　

Of the feature vectors obtained from the fixed node when the mobile node is in any position

Element value,

Denotes the Euclidean distance value between the model and the currently obtained feature vector.

When the mobile node is located at an arbitrary position, 100 distance values can be obtained by comparing 100 models in FIG. 4.

At this time, the mobile node can be recognized as being at the position having the smallest value and can be expressed by Equation (6) below.

Formula (6)

값은 100이 되므로 도 4를 이용하여 구한 유클리드 거리 100 개 중 가장 작은 값을 이동 노드의 위치로 인식한다.In the embodiment of the present invention in the case of equation (6)

Since the value becomes 100, the smallest value among the 100 Euclidean distances obtained using FIG. 4 is recognized as the position of the mobile node.

Through the above-described process, the location recognizing unit 206 recognizes the relative position of the mobile node with respect to the fixed node, and applies the location of the mobile node to the map on the terminal on which the location of the fixed node is displayed. The absolute position of the mobile node relative to the node is calculated (S509).

 After S504, the location recognizing unit 206 displays the location of the mobile node in a map on the terminal visible to the user through the interface unit 207 (S510).

Although embodiments of the present invention have been described above with reference to the accompanying drawings, those skilled in the art to which the present invention pertains may implement the present invention in other specific forms without changing the technical spirit or essential features thereof. I can understand that. Therefore, it should be understood that the embodiments described above are exemplary in all respects and not restrictive.

According to the apparatus and method for generating mobile body position information and the apparatus and method for recognizing mobile body position as described above, there are one or more of the following effects.

Unlike the prior art, which considers only the correlation between the strength and the distance of the RF signal, the positional information is analyzed by analyzing the pattern of the signal strength information according to the distance, thereby increasing the accuracy of the position.

The use of inexpensive RF modules also reduces the cost of building and maintaining location-aware systems.

Even at low cost, it can handle not only a narrow area such as an office or a house but also a large area such as a parking lot.

Claims (20)

A signal obtaining unit which receives a plurality of signal strength information respectively received from a mobile node by at least one fixed node existing at a predetermined position; A preprocessor for smoothing a plurality of signal strength information received for each fixed node; And And a feature vector processing unit for obtaining the average value of the smoothed plurality of signal strength information for each fixed node, and configuring a set of the average values for each fixed node as a feature vector. delete The method of claim 1, The feature vector is expressed by

(S is the value before normalization,

Is a normalized value,

Is the minimum of the average values,

Is normalized by the maximum of the average values) and stored in a DB, and used as a model for recognizing the position of the mobile node. A signal obtaining unit which receives a plurality of signal strength information respectively received from a mobile node by at least one fixed node existing at a predetermined position; A preprocessor for smoothing a plurality of signal strength information received for each fixed node; A feature vector processing unit obtaining a mean value of the smoothed plurality of signal strength information for each fixed node and configuring a set of the average values for each fixed node as a feature vector; And And a position recognizer that recognizes a position of the mobile node relative to the fixed node based on the feature vector. The method of claim 4, wherein The smoothing is an expression

(

Is the result of smoothing,

and

Signal strength information received from the fixed node at a specific time;

and

Is

Is a parameter that satisfies The preprocessing unit obtains a sum of a difference from the remaining signal strength information values for each of the smoothed plurality of signal strength information values, and compares the remaining values with the remaining values to determine the noise as noise. The apparatus for recognizing the position of the moving object. The method of claim 4, wherein And the fixed node receives the signal strength information from the mobile node at predetermined intervals. delete delete The method of claim 4, wherein The feature vector is expressed by

(S is the value before normalization,

Is a normalized value,

Is the minimum of the average values,

Is normalized by the maximum of the average values) and stored in a DB, and used as a model for recognizing the position of the mobile node. The method of claim 4, wherein The position recognition unit

(

Is the j th vector element value of the i th feature vector registered in the DB as a model,

Is the j th vector element value of the feature vector obtained from the fixed node when the mobile node is at an arbitrary position

Is the Euclidean distance value between the model and the currently obtained feature vector, where M is the smallest value among the result values calculated by the number of vector elements of the feature vector. Recognizing device. A signal acquiring step of receiving a plurality of signal strength information respectively received from a mobile node by at least one fixed node existing at a predetermined position; A preprocessing step of smoothing the plurality of signal strength information received for each fixed node; And And a feature vector processing step of obtaining an average value of the smoothed plurality of signal strength information for each fixed node, and configuring a set of the average values for each fixed node into a feature vector. The method of claim 11, wherein The smoothing is an expression

(

Is the result of smoothing,

and

Signal strength information received from the fixed node at a specific time;

and

Is

Is a parameter that satisfies The preprocessing step obtains the sum of the differences with the remaining signal strength information values for each of the smoothed plurality of signal strength information values, compares the remaining values with the remaining values, and determines a value that is remarkably different as noise. And processing the moving object position information. The method of claim 11, The feature vector processing step is

(S is the value before normalization,

Is a normalized value,

Is the minimum of the average values,

Is normalized to the feature vector according to the maximum of the average values) and stored in a DB, and used as a model for recognizing the position of the mobile node. A signal acquiring step of receiving a plurality of signal strength information respectively received from a mobile node by at least one fixed node existing at a predetermined position; A preprocessing step of smoothing the plurality of signal strength information received for each fixed node; A feature vector processing step of obtaining the average value of the smoothed plurality of signal strength information for each fixed node and configuring the set of average values for each fixed node as a feature vector; And And recognizing a position of the mobile node relative to the fixed node based on the feature vector. The method of claim 14, The smoothing is formula

(

Is the result of smoothing,

and

Signal strength information received from the fixed node at a specific time;

and

Is

Is a parameter that satisfies The preprocessing step obtains the sum of the differences with the remaining signal strength information values for each of the smoothed plurality of signal strength information values, compares the remaining values with the remaining values, and determines a value that is remarkably different as noise. And processing the moving object position. The method of claim 14, The signal acquiring step is a method for recognizing a moving position where the fixed node receives the signal strength information from the mobile node at predetermined intervals. delete The method of claim 14, The feature vector processing step is

(S is the value before normalization,

Is a normalized value,

Is the minimum of the average values,

And normalizing the feature vector according to the maximum of the average values) and storing the feature vector in a DB, and using the feature vector as a model for recognizing the position of the mobile node. The method of claim 14, The location recognition step is

(

Is the j th vector element value of the i th feature vector registered in the DB as a model,

Is the j th vector element value of the feature vector obtained from the fixed node when the mobile node is at an arbitrary position

Recognizing that the mobile node is located at the position having the smallest value among the results calculated by the Euclidean distance value between the model and the currently obtained feature vector, M is the number of vector elements of the feature vector) , How to recognize the moving object position. According to claim 1, The smoothing is formula

(

Is the result of smoothing,

and

Signal strength information received from the fixed node at a specific time;

and

Is

Is a parameter that satisfies The preprocessing unit obtains a sum of a difference from the remaining signal strength information values for each of the smoothed plurality of signal strength information values, and compares the remaining values with the remaining values to determine the noise as noise. And an apparatus for generating moving object position information.

Images