KR100946774B1 - Signal Strength Prediction Method for WLAN-based Positioning - Google Patents

Abstract

According to the present invention, in a method for estimating received signal strength using path loss prediction in WLAN-based positioning, a location value of an access point (AP), a location value of a reference point (RP), and a path in an indoor space The distance between the access point AP and the reference point RP and the distance between the access point AP and the prediction point PP are calculated using the position values of the prediction point PP, which is the loss prediction target, respectively. Using the path loss and the calculated distance between the (AP) and the reference point (RP) for predicting the path loss between the access point (AP) and the prediction point (PP), as shown in Equation 1 A signal strength prediction method is provided.

According to the disclosed signal strength prediction method for WLAN-based positioning, since the received signal strength is predicted using the signal strength of the actual propagation environment, the received signal strength prediction value can be improved, and measured values received from a plurality of reference points. Is applied to the received signal strength prediction model, provides accurate and fast signal strength prediction value regardless of the propagation environment, reduces the time required for database construction, low computational complexity, and improves WLAN-based location performance. You can.

WLAN, positioning, access point, reference point, database

Description

Prediction method of signal intensity for WLAN-based positioning

The present invention relates to a signal strength prediction method for WLAN-based positioning, and more particularly, to a WLAN-based positioning for predicting wireless signal strength using a mathematically modeled method in a WLAN-based positioning in an indoor environment. It relates to a signal strength prediction method.

With the development of mobile communication technology, technologies for positioning mobile terminals in various communication networks have been actively studied. In June 1996, the Federal Communications Commission (FCC) in the United States required all wireless network service providers to provide location information to public agencies such as fire and police stations in case of an emergency. Or, by 2001, about 67% of all 911 calls were provided with the terminal's location information within an error range of 100m.

Recently, as the demand for location-based services increases in Korea, the necessity of technology for finding a more accurate location is emerging.

Currently, location-based services are mainly based on GPS satellites and mobile communication network. However, the GPS-based positioning system has a low reception rate in indoor and building close areas, and the mobile communication network-based positioning system has an error of several hundred meters, so the error range is very large for use in emergency rescue and disaster services. .

In order to solve this problem, wireless LAN-based location positioning technology is drawing attention in the indoor environment.

Due to the imperfection of the indoor propagation environment, the WLAN-based positioning technology generally uses a location positioning method based on probabilistic modeling (Finger printing method). However, probabilistic modeling must establish the propagation characteristics of the indoor environment as a database before performing the positioning, and the database changes seriously whenever an environmental change such as a wall / door / furniture occurs. There is a problem of building a database using manpower.

The existing signal strength prediction model is as follows.

Electromagnetic waves are subjected to very complicated paths by the reflection, diffraction and scattering of various kinds of building materials and structures such as walls / doors / furnitures etc. in the indoor radio environment from the transmitting antenna to the receiving antenna. Therefore, received radio waves have complex nonlinear characteristics.

In order to predict the received signal under such a complicated environment, the existing proposed signal strength prediction algorithm is largely based on the deterministic model based on the physical principle of electromagnetic waves and the empirical model based on statistical data.

Ray tracing is a representative algorithm of the deterministic model. In the propagation path between the transmitter and the receiver, each ray exhibits various propagation phenomena such as refraction, diffraction, and scattering. This technique assumes the radio wave radiated from the transmission point as a ray, and the ray scatters during propagation to the radio wave environment. In the model, the reflection, transmission and diffraction coefficients are modeled by repeating the reflection, transmission, and diffraction geometrically, and the ray propagates based on the coefficients calculated through the modeling process. In this way, the traced rays predict the strength of the signal received at each receiving point.

Since the ray tracing technique uses various variables based on propagation characteristics to predict signal strength, the predicted signal strength has accurate characteristics. However, the time required to estimate signal strength is long because of the geometric change in propagation in the propagation environment.

On the other hand, the one-slope model (OSM) of the empirical model is the easiest model to calculate the average signal level using only the distance between the transceivers without detailed information on the layout of the building. In this case, the path loss may be expressed as a distance function between the transmitter and the receiver, which is expressed by Equation a below.

Equation a

는 One-Slope 모델을 사용하였을 경우 송신기와 수신기 사이의 경로손실,

는 송신기로부터 떨어진 기준거리(일반적으로 기준거리는 1m),

는 기준거리에서의 기준 경로손실(Reference path loss),

은 경로손실계수(Path loss exponent),

는 송신기와 수신기 사이의 거리이다. 물론

와

은 통신 환경에 따라 각각 다르게 주어진다. here,

In case of using One-Slope model, path loss between transmitter and receiver is

Is the reference distance away from the transmitter (usually 1m).

Is the reference path loss at the reference distance,

Is the path loss exponent,

Is the distance between the transmitter and the receiver. sure

Wow

Is given differently depending on the communication environment.

)을 가산한다면, 수신기에 수신된 수신신호 세기를 예측할 수 있으며, 이러한 One-Slope 모델은 기준거리, 송수신기 사이의 거리와 같은 거리 요소가 주요 변수를 차지한다.The path loss obtained from the transmitting power of the transmitter (

), The received signal strength received by the receiver can be estimated. In this one-slope model, distance factors such as the reference distance and the distance between the transceivers occupy the main variables.

That is, the One-Slope model has a relatively small number of variables to consider in order to predict the signal strength, so that the signal strength can be predicted relatively simply. However, since the characteristics of the propagation environment are not accurately considered, it is difficult to accurately estimate the signal strength.

Another model is the Cost 231 Multi-Wall Model (MWM).

Multi-Wall model is a model that predicts the signal strength in consideration of the losses caused by passing through the walls and layers in addition to the free space loss can be defined as Equation b.

[Equation b]

이고,

이다.)(here,

ego,

to be.)

는 Multi-Wall 모델을 사용하였을 경우 송신기와 수신기 사이 의 경로손실,

는 송신기로부터 거리가 1m일 때의 자유 공간 경로손실,

는 벽을 투과하며 발생한 신호의 경로손실,

는 건물의 층을 투과하며 발생한 신호의 경로손실이다.At this time,

In case of using Multi-Wall model, path loss between transmitter and receiver,

Is the free space path loss when the distance from the transmitter is 1m,

Is the path loss of the signal

Is the path loss of the signal that passes through the building floor.

는 각기 다른 재질의 벽 투과 손실 지수를 나타내며,

는 i라는 종류의 벽의 개수,

는 층간 전송 손실 지수이며,

는 신호 전송시 투과하는 층의 수이다.Also,

Represents the wall penetration loss index of different materials,

Is the number of walls of type i,

Is the interlayer transmission loss index,

Is the number of layers that transmit during signal transmission.

In the case of the Multi-Wall model (MWM), a relatively accurate signal strength prediction is possible because various variables are considered in the path loss prediction compared to the One-Slope model (OSM). However, since it is necessary to accurately grasp the indoor structure in order to predict the signal strength, it is difficult to predict the signal strength.

The present invention was created in order to solve the above-mentioned problems, by using the signal strength extracted from the positioning target region to predict the path loss to reduce the time and effort for measuring the actual propagation environment when predicting the signal strength. An object of the present invention is to provide a signal strength prediction method for WLAN-based positioning that can improve complexity.

Signal strength prediction method for WLAN-based positioning of the present invention for achieving the above object, in a method for predicting the received signal strength using the path loss prediction in the WLAN-based positioning, access in the indoor space The distance between the access point AP and the reference point RP using the position value of the point AP, the position value of the reference point RP, and the position value of the prediction point PP as a path loss prediction target, respectively Computing the distance between the AP and the prediction point PP, and using the path loss and the calculated distance between the access point (AP) and the reference point (RP), as shown in Equation 1 Predicts the path loss between the and the prediction point (PP).

[Equation 1]

(P _{1_est} : path loss between AP and PP, P _{1_1} : path loss between AP and RP, d _{1_P} : distance between AP and PP, d _{1 _1} : distance between AP and RP, n: between AP and PP Path loss coefficient of

In addition, when the number of reference points RP is large, the path loss between the AP and the prediction point PP is estimated using the average of the path loss calculated through each reference point RP. Equation 1 may be redefined by Equation 2.

[Equation 2]

(N: number of RP, P _{1 _j} : path loss between AP and j th RP, d _{1 _j} : distance between AP and j th RP)

When the number of the reference point RP and the access point AP is large, the path loss between the i th access point AP and the prediction point PP may be defined as Equation 3 below.

[Equation 3]

(P _{i _ est} : path loss between i th AP and PP, P _{i _j} : path loss between i th AP and j th RP, d _{i _} j: distance between i th AP and j th RP, d _{i _P} is the distance between the i th AP and the PP)

On the other hand, the present invention, before predicting the path loss between the access point (AP) and the prediction point (PP), the position value of the access point (AP), the position value of the reference point (RP), the access point An access computed using a received signal strength for a signal transmitted from the AP and received at the reference point RP, a transmit power transmitted from the access point AP, and a received signal strength of the reference point RP The method may further include measuring a path loss between the point AP and the reference point RP in the indoor space and constructing it into a database.

In addition, the present invention is a signal transmitted from the access point (AP) on a plurality of reference points by dividing the indoor space by a predetermined range before predicting the path loss between the access point (AP) and the prediction point (PP) Measuring the received signal strength with respect to the data, and constructing the measured received signal strength with a position value of each reference point in a database; And after predicting a path loss between the access point AP and the prediction point PP, receiving the prediction point PP calculated through a path loss value between the access point AP and the prediction point PP. The method may further include comparing a signal strength with a received signal strength in the database and providing a position value of a reference point indicating the most similar received signal strength as a positioning result of the prediction point PP.

According to the signal strength prediction method for WLAN-based positioning according to the present invention, since the received signal strength is predicted using the signal strength of the actual propagation environment, the received signal strength prediction value can be improved.

In addition, according to the present invention, by applying the measured values received from a plurality of reference points to the received signal strength prediction model provides accurate and fast signal strength prediction value, regardless of the propagation environment, and the time required to build a database It can reduce the complexity of the calculation and improve the WLAN-based positioning performance.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Prior to this, terms or words used in the present specification and claims should not be construed as being limited to the common or dictionary meanings, and the inventors should properly explain the concept of terms in order to best explain their own invention. Based on the principle that it can be defined, it should be interpreted as meaning and concept corresponding to the technical idea of the present invention.

Therefore, the embodiments described in the specification and the drawings shown in the drawings are only the most preferred embodiment of the present invention and do not represent all of the technical idea of the present invention, various modifications that can be replaced at the time of the present application It should be understood that there may be equivalents and variations.

1 is a diagram illustrating an arrangement of measurement points for constructing a database when estimating a terminal using probabilistic modeling. 3 is an exemplary diagram of an AP and an RP according to another embodiment of a signal prediction method according to an embodiment of the present invention, and FIGS. 4 and 5 are diagrams of a performance test environment using the signal prediction method according to the present invention. It is a block diagram of one embodiment and another embodiment.

The present invention relates to a method for predicting received signal strength using path loss prediction in WLAN based positioning.

In the terminal location estimation method using probabilistic modeling which is currently used most, a database must be constructed to perform location estimation as shown in FIG. 1.

In general database construction, as shown in FIG. 1, the indoor space to be positioned is divided into a predetermined range, and then a point (P) for measuring signal strength is set. By arranging the terminal at a fixed point P according to user setting, the terminal measures the signal strength received from the access point (AP) from each access point (AP), and together with the position coordinates of the point P, the received signal strength Save as a database.

However, this method takes a long time to build a database, and requires several times of calibration at the same point P to improve the accuracy of the measured values due to the instability of the radio environment. Whenever your environment changes, you need to build a new database. In addition, when the terminal requests a location after establishing a database, the positioning server compares the measured data with the database and provides the most suitable location value.

Here, in order to improve the performance of the WLAN-based positioning system, it is necessary to reduce the time required for the database construction time and to build a database close to the actual measurement value. Accordingly, the present invention proposes a signal strength prediction algorithm for a wireless LAN-based indoor positioning system.

As introduced in the background section above, the proposed signal strength prediction algorithm has a complex modeling process and low accuracy. In this paper, we propose a new signal strength prediction algorithm that improves the complexity of the deterministic model in signal strength prediction and improves the shortcomings of empirical models that require time and effort to measure the actual propagation environment.

)와 기준 노드에서의 손실값(

) 대신 실제 환경에서 측정된 기준노드의 전파 특성값을 사용하는 방법이다. 또한, 기존의 Deterministic 모델은 측위 대상 위치와는 상관없는 다른 환경에서 측정된 값들을 이용하여 만들어진 수학적 모델링인데 반해, 본 발명은 실제 측위 대상지역에서 추출된 신호 세기를 활용함으로써 대상 지역의 전파 특성을 포함하게 된다.The proposed prediction method of the present invention utilizes a One-Slope Model (OSM) that can be used without detailed information inside a building, and calculates a reference distance in the process of calculating path loss.

) And the loss value at the reference node (

Instead, the propagation characteristic value of the reference node measured in the actual environment is used. In addition, the conventional deterministic model is a mathematical modeling made using values measured in other environments irrespective of the positioning target position, whereas the present invention utilizes the signal strength extracted from the actual positioning target region to improve the propagation characteristics of the target region. It will be included.

In the present invention, as shown in FIG. 2, a position value of an access point (AP) and a number of measured reference points (RP) in an indoor space to be positioned, and at the access point AP, The signal strength of the positioning target point (PP) is predicted using the received signal strength reaching the reference point RP side.

, j번째 참조포인트(RP)의 위치를

, 신호를 예측할 대상인 예측포인트(PP)의 위치를

라고 가정한다. 여기서, 상기 AP, RP 및 P는 고정된 위치로 정의되어 있다.Locate the i th access point (AP) in the area

, the position of the j reference point (RP)

, The location of the prediction point (PP)

Assume that Here, the AP, RP and P is defined as a fixed position.

First, referring to the case where the number of reference points RP is one as shown in FIG. 2 (a), in order to predict the path loss of the prediction point PP, the position value of the access point AP1 and the reference point AP are predetermined. Using (RP; RP1), the distance between the access point (AP: AP1) and the reference point (RP; RP1) is calculated as in Equation 1a below.

Equation 1a

In the same principle, the distance between the access point AP1 and the prediction point PP is expressed by Equation 1b below.

[Equation 1b]

) 값을 이용하여, 액세스포인트(AP;AP1)와 예측포인트(PP) 사이의 경로손실(

)을 예측할 수 있으며 이는 다음의 수학식 1c와 같다. Meanwhile, Equations 1a, 1b, and the path loss between the access point AP1 and the reference point RP1RP1

), The path loss between the access point AP1 and the prediction point PP

) Can be predicted as shown in Equation 1c below.

Equation 1c

That is, P _{1 _ est} is the path loss (dB) between AP and PP, P _{1 _1} is the path loss (dB) between AP and RP, d _{1 _P} is the distance (m) between AP and PP, d _{1 _1} Is the distance (m) between AP and RP, and n is the path loss coefficient between AP and PP.

Like the one-slope model (OSM), the proposed invention has a path loss coefficient determined by the measurement environment, but the reference path loss value is determined differently according to the measurement environment. When estimating path loss using only one reference point (RP) RP1 as shown in FIG. 2A, a signal received from the access point (AP) to the reference point (RP) RP1 is poor Experience a severe damage to the signal received at the reference point (RP).

Therefore, simply predicting the signals of all the prediction points PP using only one reference point RP1 may include unstable signal prediction values (path loss value and received signal strength value) due to the propagation environment. Concept).

In order to compensate for this unstable signal prediction value, as shown in FIG. 2B, a plurality of reference points RP (for example, RP1 and RP2) may be used to estimate signal strength.

Therefore, the path loss of the prediction point PP can be predicted as shown in Equation 2a through the average of the path loss values obtained through the two RPs in FIG. 2 (b).

Equation 2a

That is, according to the present invention, the number of reference points (RP) is a plurality, and access using the average of the path loss calculated through each of the plurality of reference points (RP; RP1, RP2, ..., RPj) The path loss between the point AP and the prediction point PP may be predicted, and accordingly, Equation 2a may be redefined by Equation 2b.

[Equation 2b]

Here, N: the number of RP, P _{1 _} j is the path loss (dB) between the AP and j-th RP, d _{1 _} j is the distance (m) between the AP and j-th RP.

On the other hand, as shown in Figure 3, when the number of the reference point (RP) and the access point (AP) is a large number, the path loss between the i th access point (AP) and the prediction point (PP) is defined as in Equation 3a Can be.

[Equation 3]

Here, P _{i _ est} is the distance between the i-th AP and the path loss (dB) between PP, P _{i _j} is a path loss between the i-th AP and the j-th RP, d _{i _j} is the i-th AP and the j-th RP (m), d _{i _} P is the distance m between the i th AP and the PP.

As described above, when using multiple access points (APs), it is possible to more accurately calculate the path loss by using the path loss value calculated for each access point (AP). For example, the final path loss may be calculated using an average of path losses calculated for each access point to provide more accurate positioning results.

The present invention is an algorithm for positioning a location of an indoor mobile terminal, and an algorithm for presenting a signal strength prediction model on a WLAN-based positioning system. The prediction point (PP) described above is in communication with the location of a mobile terminal to be positioned in an indoor space. do.

)은, 예측포인트(PP)의 경로손실(

) 연산 이전에 미리 측정된 값으로서, 후술하겠지만 데이터베이스로 기 구축된 데이터에 해당된다.Meanwhile, the path loss between the access point AP1 and the reference point RP1RP1 in Equation 1c

) Is the path loss of the predicted point (PP)

This value is measured in advance before the operation, and corresponds to data that has been previously constructed in a database, as will be described later.

)을 예측하기 이전에, 별도로 데이터베이스 구축 단계를 거칠 수 있다. More specifically, the path loss between the access point AP and the prediction point PP

), You can go through the database construction step separately.

), '참조포인트(RP)의 위치값'(

), '상기 액세스포인트(AP)에서 전송되어 참조포인트(RP)에 수신된 신호에 대한 수신신호 세기', 액세스포인트(AP)에서 전송된 송신전력과 참조포인트(RP)의 수신신호 세기를 이용하여 연산되는 '액세스포인트(AP)와 참조포인트(RP) 사이의 경로손실(

)'을 실내 공간상에서 미리 측정하여 데이터베이스화하는 단계이다.This database construction step means the location value of the access point (AP).

), 'Position value of reference point (RP)' (

), 'Received signal strength of the signal transmitted from the access point (AP) received at the reference point (RP)', using the transmit power transmitted from the access point (AP) and the received signal strength of the reference point (RP) 'Path loss between access point (AP) and reference point (RP)

) Is the step of pre-measure and database the indoor space.

)은, AP의 신호 송신전력으로부터 RP의 수신신호 세기를 차감함으로써 계산 가능하다.For example, the path loss between an access point (AP) and a reference point (RP) (

) Can be calculated by subtracting the received signal strength of the RP from the signal transmission power of the AP.

)을 예측할 수 있다.Summary of the Invention As described above, the access point (AP) is used by using the position value of the access point (AP) in the indoor space, the position value of the reference point (RP), and the position value of the prediction point (PP) that is a path loss prediction target. Computing the distance between the AP and the reference point (RP) and the distance between the access point (AP) and the prediction point (PP), respectively, the path loss between the access point (AP) and the reference point (RP) and the calculated distance By using the path loss between the access point (AP) and the prediction point (PP) as shown in equation (1c)

) Can be predicted.

If the predicted path loss is used, the calculation with the transmission power transmitted from the AP (for example, the transmission power of the AP-the received signal strength at the path loss = PP between the AP and the PP, or the aforementioned A complex equation in which other elements are added to the equation can be used to predict the received signal strength at the prediction point PP.

The received signal strength predicted in this way is compared with the received signal strength stored in a database for each location (reference point) in the indoor space, and the position value corresponding to the most similar received signal strength is determined as a result of the positioning of the prediction point PP. Of course, it can be provided.

)을 예측하기 이전, 실내 공간을 일정 범위로 나눈 다수 개의 기준포인트(예를 들면 도 2의 포인트, P) 상에서 액세스포인트(AP)로부터 각각의 기준포인트 측에 전송된 신호에 대한 수신신호 세기를 미리 측정하여, 측정된 수신신호 세기를 각 기준포인트의 위치값과 함께 데이터베이스로 구축할 수 있다.To this end, the present invention, the path loss between the access point (AP) and the prediction point (PP) (

), The received signal strength of the signal transmitted from the access point AP to each reference point on a plurality of reference points (for example, point P in FIG. 2) divided by the indoor space into a predetermined range. By measuring in advance, the measured received signal strength may be constructed as a database together with the position value of each reference point.

)을 예측한 이후에는, 액세스 포인트(AP)와 예측포인트(PP) 사이의 경로손실(

) 값을 통해 연산되는 예측포인트(PP)의 수신신호 세기를, 상기 데이터베이스 내의 수신신호 세기와 비교하여 가장 유사한 수신신호 세기를 나타내는 기준포인트의 위치값을 예측포인트(PP)의 측위 결과로 제공할 수 있다. 물론, 본 발명의 기술범주 내에서 이러한 측위 결과를 제공하기 위한 보다 다양한 실시예가 존재할 수 있다.And, as shown in Equation 1c, the path loss (

), The path loss between the access point AP and the prediction point PP

The received signal strength of the predicted point PP, which is computed through the value of), is compared with the received signal strength in the database to provide the position value of the reference point indicating the most similar received signal strength as the positioning result of the predicted point PP. Can be. Of course, there may be more various embodiments to provide such positioning results within the technical scope of the present invention.

All of the series of operations disclosed in the present invention as described above can be made in a separate positioning server (not shown) that performs communication (various information exchange) with each terminal, AP or RP on the PP, which is obvious to those skilled in the art. It is true.

For example, the positioning server (not shown) may be configured for all the above-described databases, path loss calculation and prediction, prediction of received signal strength through predicted path loss, and the current position value (x, y) of the prediction point PP from the terminal. ) Reception of information, and the like.

In order to analyze the performance of the method of the present invention as described above, the second floor of the ○○ University Material Center was selected as the test site.

4 and 5 are two test environments (test environments 1,2) for performance testing of the signal strength prediction method of the present invention.

The distance between the reference points (P) is 2m, test environment 1 is open area (n = 2 ~ 3.2), test environment 2 is a laboratory where there are many obstacles (Large number of obstacles, n = 3) ~ 3.4).

AP1 of FIG. 4 (Test Environment 1) is received in the LOS environment, and AP3 and 4 are received through the wall, so the NLOS (Non Light Of Sight) environment and AP2 of FIG. 5 (Test Environment 2) are OLOS (Obstructed Line Of). It is received in a Sight environment, and AP3, 5 is received through a wall, so it is an NLOS (Non Light Of Sight) environment.

Each experimental space installed a total of three WLAN access points (AP), one at the light of sight (LOS), the other two at one point and two points through the wall. In addition, the received signal strength was checked using an HP dv4000 notebook equipped with IEEE 802.11a / b / g Cisco Aironet.

The path loss coefficient was set differently according to the experimental environment. Since the signals are received through the wall, the loss index due to the wall is considered in the existing Multi-Wall model (MWM).

And, the received signal strength was measured 10 times at each experimental point (P).

The proposed method was verified by substituting 10 average values of the signal strength measured at the reference point (RP) and compared with the results of the One-Slope model (OSM) and the Multi-Wall model (MWM). Each result expressed by comparing the performance is expressed as an average value of the received signal strength expected values obtained at all points.

[Table 1]: Comparison of the difference between the PP received signal strength prediction value and the actual measured value according to the number of RP (test environment 1, open area)

[Table 2]: Comparison of the difference between the PP received signal strength prediction value and the actual measured value according to the number of RP (Test environment 2, Large number of obstacles)

Tables 1 and 2 compare the received signal strength prediction values according to RP. The signal strength prediction model is a propagation model in which the signal strength varies greatly by the distance between the moving object and the reference point. Radio waves are attenuated by various factors that interfere with radio waves in a mobile communication channel environment.

Therefore, there are more obstacles between the moving object and the reference point than the LOS environment (test environment 1) where radio waves experience less interference (test environment 2). A lot of radio wave changes occur in Therefore, the received signal strength estimate shows better results in the NLOS (Non Line of Sight) environment, which experiences more obstacles or propagation changes than the LOS environment.

In addition, in Tables 1 and 2, it can be seen that as the number of RPs increases, the difference (prediction error) between the received signal strength prediction value and the actual measurement value decreases. This means that as the number of RPs increases, the incomplete propagation environment is compensated for, so that the difference between the received signal strength prediction value and the actual measured value can be significantly reduced.

However, simply increasing the number of RP does not improve the received signal strength prediction value. The indoor propagation environment is incompletely changed, and the remaining signal strengths are affected by the location of the RP and the number of RPs to predict received signal strength. The performance of the received signal strength prediction value may be improved.

[Table 3]: Comparison of received signal strength difference between the proposed method and the conventional method

(Test Environment 1, Open area)

[Table 4]: Comparison of received signal strength difference between the proposed method and the conventional method

(Test Environment 2, Large number of obstacles)

Table 3 and Table 4 compare the received signal strength prediction performance between the conventional One-Slope model (OSM), Multi-Wall model (MWM) and the method of the present invention based on the data measured in each test environment.

The existing Multi-Wall model (MWM) predicts the received signal strength by taking into account the losses caused by walls and layers in addition to the free space loss, so the one-slope model when the loss due to the wall is not considered (nw = 0) Has the same signal strength prediction as (OSM). In this case, since the signal strength prediction value is determined by the fixed reference strength and the reference distance value, the prediction value has an error with the value measured in the real environment.

On the other hand, since the proposed algorithm of the present invention predicts the received signal strength by using the signal strength of the actual propagation environment, the proposed algorithm has an improved received signal strength prediction value compared to the conventional algorithm.

Next, if one wall (nw = 1) or two (nw = 2) is considered, the One-Slope model (OSM) does not consider another loss factor, so the received signal strength estimates The error is more than 10dB. The Multi-Wall model (MWM) shows better performance than the one-slope model (OSM) because it additionally takes into account the losses due to the wall, but since the factors used to predict the received signal strength are already modeled, the received signal The intensity estimates do not reflect the real environment.

However, the method of the present invention provides a much more accurate received signal strength prediction value than the conventionally presented algorithm by applying the measured values received from the plurality of reference points to the received signal strength prediction model without being greatly affected by the propagation environment. .

As described above, although the present invention has been described by way of limited embodiments and drawings, the present invention is not limited thereto and is intended by those skilled in the art to which the present invention pertains. Of course, various modifications and variations are possible within the scope of equivalents of the claims to be described.

1 is an exemplary diagram of measurement point arrangement for building a database when estimating a terminal using probabilistic modeling;

2 is an exemplary diagram when the number of RPs is one or two in the signal strength prediction method for WLAN-based positioning according to an embodiment of the present invention;

3 is an exemplary diagram of an AP and an RP according to another embodiment of a signal prediction method according to an embodiment of the present invention;

4 and 5 are configuration diagrams of one embodiment and another embodiment of the performance test environment using the signal prediction method of the present invention.

Claims (5)

A method for predicting received signal strength using path loss prediction in WLAN-based positioning, The position value of the access point AP in the indoor space, the position value of the reference point RP, and the position value of the prediction point PP as a path loss prediction target are respectively used. The distance and the distance between the access point AP and the prediction point PP are respectively calculated, and the path loss between the access point AP and the reference point RP and the calculated distance are accessed as shown in Equation 1 below. Predict the path loss between the point (AP) and the predicted point (PP), When the number of reference points RP is large, the path loss between the AP and the prediction point PP is predicted using the average of the path loss calculated through each reference point RP. Equation 1 is redefined by Equation 2, characterized in that the signal strength prediction method for WLAN-based positioning.

(P _{1_est} : path loss between AP and PP, P _{1_1} : path loss between AP and RP, d _{1_P} : distance between AP and PP, d _{1_1} : distance between AP and RP, n: between AP and PP Path loss factor)

(N: number of RP, P _{1_j} : path loss between AP and j th RP, d _{1_j} : distance between AP and j th RP) delete The method of claim 1, The number of the reference point (RP) and the access point (AP) is a plurality, the path loss between the i-th access point (AP) and the prediction point (PP) is characterized in that defined by Equation 3, WLAN Signal strength prediction method for based positioning.

(P _{i_est:} i the path loss between the first AP and the PP, P _{i_j:} distance between the i-th AP and the j-th RP, d _{i_P::} i the path loss between the first AP and the j-th RP, d _{i_j} i-th AP and Distance between PP) The method according to claim 1 or 3, Before estimating the path loss between the access point AP and the prediction point PP, the position value of the access point AP, the position value of the reference point RP, and transmission from the access point AP And the access point AP computed using the received signal strength of the signal received at the reference point RP, the transmission power transmitted from the access point AP, and the received signal strength of the reference point RP. And measuring the path loss between reference points (RP) in the indoor space in advance and constructing it into a database. The method of claim 4, wherein Before estimating the path loss between the access point AP and the prediction point PP, the received signal strength of the signal transmitted from the access point AP on a plurality of reference points obtained by dividing the indoor space into a predetermined range. Measuring in advance to build the measured received signal strength together with a position value of each reference point in a database; And After predicting the path loss between the access point AP and the prediction point PP, the received signal of the prediction point PP calculated through the path loss value between the access point AP and the prediction point PP. And comparing the strength with the received signal strength in the database to provide a position value of a reference point indicating the most similar received signal strength as a positioning result of the prediction point (PP). Signal strength prediction method.

Images