KR100831556B1 - Method of correction NOLS error for wireless positioning system - Google Patents

Abstract

1. TECHNICAL FIELD OF THE INVENTION

The present invention relates to an invisible path error correction method for a wireless positioning system and a computer readable recording medium storing a program for realizing the method.

2. The technical problem to be solved by the invention

The present invention provides an invisible path error correction method and method for a wireless positioning system for improving the positioning accuracy by improving the positioning error due to the invisible path of the radio wave generated when determining the position of the terminal in a wireless communication environment. To provide a computer-readable recording medium that records a program for realization.

3. Summary of Solution to Invention

According to the present invention, in the invisible path error correction method applied to a wireless positioning system, the existing moments are obtained by obtaining second moments for received signals that may be assumed to include errors due to invisible path propagation. A first step of comparing the theoretical values of the model; A second step of obtaining a visible path distance value and a parameter value of a scattering model according to the comparison result of the first step; And a third step of determining a position of the terminal by improving an error due to an invisible path through a known positioning method based on the obtained values.

4. Important uses of the invention

The present invention is used in a wireless positioning system and the like.

Wireless positioning system, invisible path error, TOA estimate, delay spread, MVMA algorithm, weighting algorithm

Description

Invisible path error correction method for wireless positioning system {Method of correction NOLS error for wireless positioning system}             

1 is an exemplary configuration of a positioning service system to which the present invention is applied.

2 is a cell arrangement diagram illustrating a distance between a true distance and a base station in an invisible path error correction method according to an embodiment of the present invention.

3 is an exemplary view showing a measured distance circle including an error due to invisible path propagation according to an embodiment of the present invention.

4 is a performance comparison graph when the cell radius of the invisible path error correction method according to an embodiment of the present invention 5Km.

5 and 6 are graphs comparing the performance when the cell radius is 2Km in the invisible path error correction method according to an embodiment of the present invention.

7 and 8 are graphs comparing performance when the cell radius is 1Km in the invisible path error correction method according to an embodiment of the present invention.

9 is an exemplary view showing the position of the base station, the terminal and the scatterer in the circular scatterer model of the invisible path error correction method according to another embodiment of the present invention.

10 is an exemplary view showing the positions of the base station, the terminal and the scatterer in the disk-type scatterer model of the invisible path error correction method according to another embodiment of the present invention.

FIG. 11 is an exemplary diagram illustrating the positions of a base station, a terminal, and a scatterer in a Gaussian scatterer model of an invisible path error correction method according to another embodiment of the present invention. FIG.

12A and 12B are exemplary views illustrating a circular scatterer model and a disk-shaped scatterer model received in an invisible path error correction method according to another embodiment of the present invention.

13 is a performance comparison graph for the ROS algorithm in the invisible path error correction method according to another embodiment of the present invention.

14 is a performance comparison graph of the DOS algorithm of the invisible path error correction method according to another embodiment of the present invention.

15 is a performance comparison graph for a Gaussian algorithm among invisible path error correction methods according to another embodiment of the present invention.

* Explanation of symbols for the main parts of the drawings

11 mobile station (MS) 12 base station (BTS)

13: control station (BSC) 14: location-related service server

15: database

The present invention is the wireless subscriber network (WLL), broadband wireless subscriber network (B-WLL), mobile telephone network (Cellular), personal mobile communication network (PCS), other mobile telephone networks currently in use, European and North American methods To determine the location of a terminal in a wireless communication system including a next generation mobile communication network (IMT-2000) such as International Mobile Telecommunication-2000 (IMT-2000), Universal Mobile Telecommunication Service (UMTS), etc. The present invention relates to an invisible path error correction method for improving the positioning accuracy by improving the radio positioning error due to the invisible path propagation, and a computer-readable recording medium recording a program for realizing the method.

That is, the present invention belongs to a technique for improving positioning accuracy by eliminating various positioning error factors present in a wireless positioning system for determining a position of a terminal in a wireless communication network including a code division multiple access (CDMA) mobile communication network. The positioning error factors and techniques for solving them are as follows.

1) multipath error

In the city center, the error caused by multipath is the biggest error factor in the system of measuring the angle of arrival (AOA) or the signal strength. In addition, multipaths in time-based positioning systems using Time of Arrival (TOA) or Time Difference Of Arrival (TDOA) methods cause errors in estimating accurate TOA or TDOA measurements. Traditional delay estimators, which use cross-correlation to estimate measurements, are affected by multipaths, especially when the difference between the multipath signal and the signal that arrives directly is within 1 chip. To reduce this multipath error, traditional delay estimators use very high resolution techniques such as "root-MUSIC (MUltiple SIgnal Classification) or TLS-ESPRIT (Total Least Square-Estimation of Signal Parameters via Rotational Invariance Techniques)". Detect multipath components not detected.

2) hearability problems

In order to estimate the position of the mobile station using TDOA on the N dimension, a distance or time difference between the base station and the mobile station of at least N + 1 different positions is required. That is, wireless positioning using a CDMA mobile communication network on a two-dimensional plane is possible only by estimating the time difference between a mobile station and a base station at three or more different positions. Therefore, the ability of estimating the difference in time delay from other base stations acts as an important factor, which is called the hearing capability. CDMA mobile communication networks are interference limited systems. Therefore, the CDMA mobile communication network has to isolate the cells in order to maximize the subscriber capacity and provide good quality service to the user. Therefore, the link between the mobile station belonging to the home cell and other cells becomes more difficult. In addition, a strong base station transmission signal of a home cell acts as a strong interference signal to a weak transmission signal of another cell, thereby deteriorating a reception ability, thereby increasing an estimation error of a difference in time delay from another base station. In order to solve this problem, the transmission signal of the home cell base station is temporarily stopped to remove the idle period downlink (IPDL) or the home cell base station signal received from the terminal to secure the reception capability of another base station. Therefore, there is an interference cancellation technology that secures hearing capability.

3) time synchronization error between base stations

The time synchronization error between base stations is directly connected to the position error of the terminal. Since the GPS (Global Positioning System) maintains the accuracy of the time between satellites about 10 ^-11 , it is possible to lower the positioning error factor by less than 3cm by maintaining the time synchronization between the base stations using the GPS.

4) Invisible path error

A signal received without the visible path signal component is reflected and diffracted to follow a path that is longer than the direct path length through which the visible path signal component passes. This greatly reduces the accuracy of wireless positioning systems using TOA and TDOA, even with ultra-high resolution technology in a multipath-free environment. According to the measurement results in the Global System for Mobile Telecommunication (GSM) system, statistical errors due to non line of site (NLOS) propagation in the city range from 400 to 700 m. CDMA systems affected by the propagation environment are not an exception to this result.

However, such NLOS propagation has a bias component in TOA or TDOA measurements, which cannot be solved by the positioning algorithm alone. That is, the existing positioning algorithms can estimate the exact position only when there is a visible path component between the base station and the terminal, so that the position of the terminal by the measured value of the invisible path is far from the actual position. One way to reduce the invisible path error is to calculate the location algorithm by reducing the weight of the invisible path signal components. Each pseudo distance measurement is weighted to calculate the least squares method.

Therefore, a method for solving the problem of deterioration of positioning performance due to invisible path error that acts as the largest error factor among the above-described positioning error factors is required.

In general, the location determining technology may be classified into the following network-based, terminal-based, dedicated network-based, and other technologies.

Techniques for determining location will be described in more detail to aid in the understanding of the present invention.

First of all, the network-based technology is a method of determining a location using a base station constituting a mobile communication network. The necessary parameters are the angle of arrival of radio waves (AOA), time of arrival (TOA) and time difference of arrival ( TDOA), amplitude, phase, signal strength, and the like. The calculation of the location of these parameters in the network side is called a network-based technology, and the parameters can be classified into a case of receiving at the base station and a method of receiving at the terminal and transferring them to the network side.

When using a network-based technology, the location can be determined without changing the existing terminal without additional equipment.

The Federal Communications Commission (FCC) mandate of the US states that by October 1, 2001, network-based technologies will be able to locate within a 100-meter radius for wireless E911 (Enhanced 911) services.                         

Next, referring to the terminal-based technology, it may be referred to as a technology based on the Global Positioning System (GPS). That is, a method of determining one's own location by using signal information of GPS satellites received by a GPS module built in a terminal, a method of determining one's own location by a GPS module, a terminal having a built-in GPS module, There is a method of determining the location by transmitting and receiving data with the network equipment that has information about the GPS satellites.

In the case of using an external GPS, since GPS must be purchased separately, cost is increased and the size of the terminal becomes large, making it difficult to carry.

On the other hand, when looking at the technology based on the dedicated network, which is a technology using a dedicated network for positioning, and uses a different frequency than the general wireless communication system. For example, a dedicated network in the US LMS (location and monitoring) band is used, and a separate signal is used for location determination. This is a kind of road sign, and roadside base stations of the Intelligent Transport System (ITS) are also included in this category.

Dedicated network-based technology generally lacks economic feasibility because a separate network must be constructed for location verification. On the other hand, high accuracy can be achieved if the signal is also designed for a specific application and installed in some areas.

In addition to the above-described methods, there is a radio camera technology that determines a location by receiving a signal transmitted from a terminal in a network similar to the network-based technology.

RadioCamera technology finds a location by comparing a fingerprint of a wireless signal form of the current user with a wireless signal form known in advance for a specific region. The technology uses a database of user signals received via multipath and radio signal propagation that is already established to enable continuous location measurements regardless of the user's own location requirements.

In addition, although no special positioning technique is used, the positioning is performed by an identifier (ID) of a base station or a sector in the base station area. That is, it is possible to identify its own location by the sector ID in the base station or the base station area which is served by the digital cellular mobile communication system or the PCS system. However, since the size of a sector in a base station or a base station area in a cellular mobile communication system or a PCS varies, the accuracy of the position varies greatly.

In addition, hybrid schemes using the advantages of both network-based and terminal-based schemes have been actively studied in recent years.

Among these technologies, the present invention can be applied to a scheme in which both network-based and terminal-based schemes are combined in a general mobile communication network. In particular, the terminal can directly calculate the position using the received signal or transfer the received signal to the network side. Ideal for calculations. In other words, assuming that the terminal simultaneously receives three or more forward pilot signals necessary for its location determination at any position in the CDMA network, these signals include errors due to invisible path propagation. To improve the technology.

At present, several methods have been proposed as error improvement algorithms by invisible path of wireless positioning system using mobile communication network.

The first method is to use the variance of the received signal to distinguish the base station from the terminal and the visible path / invisible path. That is, since the signal including the invisible path propagation has a larger dispersion than the signal due to the visible path propagation, the visible path / invisible path signal is divided by the dispersion. Kalman filter technology is used to reconstruct a signal from a base station in an invisible path from a visible path to a signal. This method needs to know the variance in advance in order to distinguish the visible path / invisible path base station, and there is a problem that continuous signal reception is required for signal reconstruction.

The second method is to reconstruct the signals received from the mobile terminal close to the visible path signal. This method requires the terminal to be mobile and also knows the statistical characteristics of the noise and invisible path signals in advance.

The third method is to give weight to the rest. Estimates are made by making several subsets of signals received from the base station, and using these values, different weights are calculated according to the rest to calculate an average value. The rest serves as a parameter for distinguishing the invisible path base station, and the weight serves to improve the error due to the invisible path. This method can only operate when there are 4 to 5 or more signals received from the base station.

A similar weighting method uses AOA, which requires several signals to be estimated. As another method, a method of estimating the position assuming that only one signal among a plurality of base station signals is received from an invisible path is also proposed.

In addition, various methods have been proposed to improve the positioning error due to the signal received from the base station in the invisible path, but it can be said that the assumption for the algorithm to operate is not practical. That is, it is difficult for the actual terminal to exceed three or more signals due to the problem of Hearability, or it is assumed that four to five or more signals are received, or that the terminal is moving. The algorithm is difficult to operate when moving at a stationary or pedestrian speed.

On the other hand, in the CDMA mobile communication networks such as cellular, PCS and IMT-2000, the domestic territory is narrow and the number of subscribers is large, and the cell size is mostly composed of pico and micro cells. In this case, when a wireless positioning system is implemented, a visible path is not secured between a subscriber and three adjacent base stations necessary for positioning. This is true even in a macrocell environment. In general, in the wireless positioning, positioning is performed by assuming a signal received from the visible path, and it is very difficult to secure the visible path in all three or more signals to be received in the general environment of the wireless positioning. Therefore, the received signal includes an invisible path error, which causes a positioning error.

Therefore, in the current technical field, a method for improving the positioning error caused by the error due to the invisible path, which can be assumed to be always included in the received signal in the radio positioning, is essential.

The present invention has been proposed to meet the requirements as described above, and to improve the positioning accuracy by improving the positioning error due to the invisible path of the radio wave generated when determining the position of the terminal in a wireless communication environment An object of the present invention is to provide an invisible path error correction method for a system and a computer readable recording medium recording a program for realizing the method.

The present invention for achieving the above object, in the invisible path error correction method applied to the wireless positioning system, for each received signal that can be assumed to include the error due to the invisible path propagation, the constraint condition Creating and obtaining a weight value through an optimization process; And a second step of determining a position of the terminal by improving an error due to an invisible path through a known positioning method based on the obtained weight value.

In addition, the present invention, in the invisible path error correction method applied to the wireless positioning system, by obtaining the second moment (moments) for the received signals that can be assumed to include the error due to the invisible path propagation A first step of comparing the theoretical values of the existing scattering model; A second step of obtaining a visible path distance value and a parameter value of a scattering model according to the comparison result of the first step; And a third step of determining a position of the terminal by improving an error due to an invisible path through a known positioning method based on the obtained values.

In addition, the present invention provides a system for creating a constraint and obtaining a weight value through an optimization process for each received signal that can be assumed to include an error due to invisible path propagation in a wireless positioning system having a processor. 1 function; And a computer readable recording medium having recorded thereon a program for realizing a second function of determining a position of a terminal by correcting an error due to an invisible path through a known positioning method based on the obtained weight value. to provide.

In addition, the present invention, in the wireless positioning system having a processor, for the received signals that can be assumed to include an error due to the invisible path propagation, the second moment (moments) to obtain the theoretical of the existing scattering model A first function of comparing with values; A second function of obtaining a visible path distance value and a parameter value of a scattering model according to a comparison result in the first function; And a computer-readable recording medium having recorded thereon a program for realizing a third function of determining a position of a terminal by improving an error due to an invisible path through a known positioning method based on the obtained values. do.

The present invention is a technique for improving the positioning accuracy required for the implementation of a wireless positioning system for determining the position of a wireless terminal in a CDMA mobile communication network, and solves the positioning performance degradation caused by the TOA estimation error of the invisible path radio wave which occurs frequently in the urban radio environment. It is a technique for doing.

Accordingly, in the present invention, an invisible path error correction method for a wireless positioning system for improving a positioning error due to invisible path propagation of a radio wave generated when determining a location of a terminal in a wireless communication environment. In this paper, we propose a method to improve the positioning accuracy by improving the positioning error due to the invisible path propagation in the wireless terminal.In another embodiment, the inference value and the measured value using the moment of the probability statistics theory based on the scattering model We propose a method to improve the positioning accuracy using MVMA.

The above objects, features and advantages will become more apparent from the following detailed description taken in conjunction with the accompanying drawings. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is an exemplary configuration diagram of a positioning service system to which the present invention is applied. In the drawing, "11" is a mobile station (MS), "12" base station (BTS), and "13" is a control station. (BSC: Base Station Controller), "14" represents a Location Service (LCS) server, and "15" represents a Geographic Information System (GIS) database (DB), respectively.

Since the CDMA mobile communication system environment is only known in the art, detailed description thereof will be omitted here.

The mobile station (MS) 11 is within the range of any base station 12, and the information of the mobile station 11 is transmitted to the switching center (MSC) of the CDMA network via the base station 12 and the control station 13 controlling it. The information transmitted from the switching center is transmitted to the mobile station 11 via the control station 13 and the base station 12.

The mobile station 11 is a terminal that the user can carry and communicate with while moving.

The base station (BTS) 12 connects the mobile station 11 to the control station 13, and includes a digital channel unit (DCU), a timing / frequency control unit (TCU), and a radio frequency. It consists of a Radio Frequency Unit (RFU) and may additionally include a Global Positioning System (GPS).

The base station 12 performs a wired / wireless conversion function of communicating with the mobile station 11 over the air and communicating with the control station 13 by wire.

The control station (BSC) 13 connects the base station 12 to the switching center to coordinate the connection between the base stations 12 and serves as a signal processing function for communication between the base station 12 and the switching station 14.

The exchange station connects with the control station 13 to perform call setup and release functions of the mobile station 11, and performs various functions related to call processing and additional services.

Now, look at the function of each component for the positioning technology as follows.

The base station 12 performs a wired / wireless conversion function of communicating with the mobile station 11 over the air and communicating with the control station 13 by wire. At this time, the mobile station 11 receives the positioning signal from the radio and transmits the positioning signal to the control station 13 through the wire, and transmits the charging class of the outgoing signal according to the confirmed position to the mobile station 11.

The control station (BSC) 13 connects the base station 12 to the CDMA mobile communication network to coordinate the connection between the base stations 12, and serves as a signal processing function for communication between the base station 12 and the CDMA mobile communication network. At this time, the relevant information for positioning is transmitted to the location-related service server 14 through the CDMA mobile communication network.

The location service server 14 receives the location request from the control station 13, searches the GIS DB 15, and transfers the searched result to the control station 13 through the CDMA mobile communication network.

The mobile station 11 displays the charging class data according to the position received from the base station 12 on the screen.

Now, an operation process for an invisible path error correction method according to an embodiment of the present invention will be described.

The present invention has the following features to improve the above-mentioned problems, and proposes an algorithm for improving the error due to the invisible path propagation generated in the mobile communication network.

First, the signal received from the base station does not require prior information as to whether it is received from the visible path / invisible path.

Second, no statistics are needed for the signal received from the invisible path.

Third, there is no need for continuous signal reception.

Fourth, no more than four or five signals need to be received (three signals can improve error).

In general, the range equation between the i-th base station and the terminal is as shown in Equation 1 below.

In general, the distance equation of Equation 1 includes an error due to the invisible path propagation due to the characteristics of the mobile communication channel, and displays the distance including the error in relation to the true range l _i . If it can be expressed as shown in the following [Equation 2].

의 값을 갖는다. 상기 [수학식 2]를 상기 [수학식 1]에 대입하고 양 변을 제곱하면 하기의 [수학식 3]과 같다.Here, about invisible path propagation

Has the value of. Substituting [Equation 2] into [Equation 1] and squared both sides is shown in [Equation 3].

Variables are defined as shown in [Equation 4] to facilitate the expression development.                     

값들을 구하여 기지국으로부터 수신된 신호를 가시경로 거리를 전파한 수식으로 나타낼 수 있도록 하여 위치를 추정하는 것이다. 이렇게 하여 비가시경로 전파에 의한 오차를 개선한 단말기 위치를 구할 수 있게 된다.The present invention

By estimating the values, the position of the signal received from the base station can be represented by the equation of propagation of the visible path distance. In this way, it is possible to obtain the terminal position which has improved the error due to the invisible path propagation.

값들을 구하기 위해 필요한 파라미터 값들 간의 관계를 구하기 위해 도 2와 같은 일반적인 셀 배치에 상황을 고려하면, 참 거리(true range)와 비가시경로에 의한 거리의 관계를 알 수 있다. 도 2에서

로 표시할 수 있으므로, 이 식에서 양변을 제곱하고 삼각함수의 관계식(trigonometric relationship)들을 이용하면 하기의 [수학식 5]와 같다.

Considering the situation in the general cell arrangement as shown in FIG. 2 to obtain the relationship between the parameter values necessary to obtain the values, it is possible to know the relationship between the true range and the distance due to the invisible path. In Figure 2

Since both sides are squared in this equation and trigonometric relationships of trigonometric functions are used, Equation 5 below.

로 나누어 정규화하면 다음의 [수학식 7]을 구할 수 있다.From Equation 2, the following Equation 6 can be obtained, and these equations are substituted into Equation 5 above.

Normalizing by dividing by gives Equation 7 below.

를 정의하였으므로 상기의 [수학식 7]로부터

에 대한 제한 식을 다음의 [수학식 8]과 같이 구할 수 있다.As shown in [Equation 4] above

Since it is defined above from [Equation 7]

The limiting equation for can be obtained as in Equation 8 below.

Here, when Equation 8 is written in the form of a matrix, Equation 9 is obtained.

이고,here,

ego,

,

와 같이 나타낼 수 있다.

,

Can be expressed as:

를 구하기 위해 또 다른 제약조건(constraints)을 구하면 다음과 같다.

To find another constraint (constraints) to obtain

보다 작으면 BS₁과 BS₂로부터의 거리에 따른 원은 교차하지 않는다. 참 거리에 의한 원은 한 점에서 교차하여야 하므로 이러한 상황은 불가능하다. 따라서 BS₁에서의 비가시경로 오차

는

보다 클 수 없다. 같은 방법으로 하면

는

보다 클 수 없다. 따라서

에 대한 상한(upper bound)은 다음의 [수학식 10]과 같다.For example, since the invisible path error always has a positive value, the measured distance is larger than the true distance, and the position of the terminal appears in an area where each circle overlaps (U, V, and W). area). If BS ₂ is a visible path and the distance by the received signal is true, then the true distance from BS ₁

If it is smaller, the circles along the distance from BS ₁ and BS ₂ do not intersect. This situation is not possible because the circle by the true distance must intersect at one point. Therefore, invisible path error in BS ₁

Is

Cannot be greater than In the same way

Is

Cannot be greater than therefore

The upper bound on is given by Equation 10 below.

와

에 대한 상한을 구하면 다음의 [수학식 11]과 같다.In the same way

Wow

The upper limit for is given by Equation 11 below.

이므로

이 된다. 따라서

이 가질 수 있는 최소값은 다음의 [수학식 12]와 같이 주어진다.By the way,

Because of

Becomes therefore

The minimum value it can have is given by Equation 12 below.

이므로, 하기의 [수학식 13]과 같이 구할 수 있고

에

에 대해 동일하게 하한(lower bound)을 구하면 하기의 [수학식 14]와 같다.From FIG. 3

Therefore, it can be obtained as shown in Equation 13 below.

on

The lower bound is equally obtained for Equation 14 below.

에 대한 하한은 다음의 [수학식 15]와 같이 나타낼 수 있다.So you want to save

The lower limit for may be expressed as in Equation 15 below.

라고 하면 최적화하여야 할 비용 함수는 하기의 [수학식 16]과 같이 표시할 수 있다.On the other hand, the actual terminal position determination can be seen as an optimization problem under such constraints. That is, the terminal position range based on the signal received in FIG. 3 exists within points U, V, and W, and a cost function to be optimized is a portion consisting of points U, V, and W formed at intersections of distances. . Coordinates of points U, V, and W, respectively.

In this case, the cost function to be optimized may be expressed as in Equation 16 below.

Substituting Equation 4 above into Equation 3 above to obtain a true range (terminal location) is shown in Equation 17 below.

와 같이 쓸 수 있고, 이를 행렬 형태로 다시 쓰면 다음의 [수학식 18]과 같다.[Equation 17] can be summarized by the LLOP (Linear Line of Position) algorithm,

It can be written as

와 같이 쓸 수 있다. 따라서 단말기 위치는 [수학식 19]를 통해 다음과 같이 구할 수 있다.here,

Can be written as: Therefore, the location of the terminal can be obtained as follows through Equation 19.

의 함수로 쓰면 하기의 [수학식 20]과 같다.Substituting Equation 19 above into Equation 16 as a cost function

When written as a function of [Equation 20].

here,

이고,

이다. 이 식을 행렬 형태로 다시 쓰면 하기의 [수학식 21]과 같다.

ego,

to be. If this equation is rewritten in matrix form, Equation 21 is used.

과 같이 나타낼 수 있다. here,

It can be expressed as

The equation for the optimized vector representing the terminal position to be finally obtained can be expressed as Equation 22 below.

이다. here,

to be.

는 상기의 [수학식 22]로 주어지는 제한조건 하에서

와 같이 구해진다. 구해진 벡터에서

를 상기의 [수학식 19]에 대입하여 단말기 위치를 구할 수 있다. 이와 같이, 본 발명의 일시예에 따른 비가시경로 오차 보정 방법에 대한 시뮬레이션 결과를 살펴보면, 도 4는 본 발명의 다른 실시예에 따른 비가시경로 오차 보정 방법 중 셀 반경이 5Km인 경우의 성능 비교 그래프로서, 가중치 알고리즘(Weighted Algorithm)이 본 발명에 따른 알고리즘에 의한 성능이고, 비가중치 알고리즘(Unweighted Algorithm)은 일반적인 경우의 성능으로, 상기 도 4에서는 LLOP 알고리즘을 사용한 것이다.Therefore, the optimized vector you want to find

Under the constraints given by Equation 22 above

Obtained as From the obtained vector

The position of the terminal can be obtained by substituting [Equation 19] above. As described above, referring to simulation results of the method for compensating the invisible path error according to an exemplary embodiment of the present invention, FIG. 4 compares the performance when the cell radius is 5 km in the invisible path error compensating method according to another embodiment of the present invention. As a graph, a weighted algorithm is performance by the algorithm according to the present invention, and an unweighted algorithm is performance in a general case, and the LLOP algorithm is used in FIG. 4.

5 and 6 are graphs comparing performance when the cell radius is 2Km in the invisible path error correction method according to an exemplary embodiment of the present invention. In FIG. 6, the invisible path error is larger than that in FIG. 5. The case is shown. Therefore, the larger the invisible path error, the better the error improvement performance of the proposed weighted algorithm.

7 and 8 are graphs comparing performance when the cell radius is 1 Km in the invisible path error correction method according to an embodiment of the present invention. Large case is shown. Therefore, the larger the invisible path error, the better the error improvement performance of the proposed weighting algorithm. However, it can be seen that as the cell radius decreases, the degree of performance improvement also decreases.

Meanwhile, an operation process for an invisible path error correction method using MVMA among the invisible path error correction methods according to another embodiment of the present invention will be described below.

The present invention uses Mean and Variance Matching Algorithm (MVMA) to improve errors caused by invisible path propagation in a mobile communication network. This technique applies a commonly used scattering model to radiolocation so that the difference converges to zero by comparing the mean and variance, the second moment of the received signal, with the mean and variance from the measured values. At this time, the distance of the visible path signal obtained is applied to a general positioning algorithm to determine the position of the terminal having improved invisible path error.

This algorithm has the following features with respect to the existing algorithm.

First, it is also applicable to the case where three received signals are the minimum signals for positioning.

Second, it can be applied to all terminal states (stopping, walking speed, movement, etc.).

Third, it is applicable to both positioning from the base station in the forward link and the base station in the reverse link.

Fourth, it can be applied even without prior statistical values for the signal received from the invisible path.

Fifth, it can be applied without receiving continuous signal.

Sixth, it is possible to apply without the visible path / invisible path prior information of the received signal.

Therefore, the operation of the invisible path error correction method using the MVMA according to the present invention will be described.

1) Scattering Model

A typical channel model used by a standardization institution for performance verification and comparison of radiolocation algorithms is a scattering model, for example, a CDMA testbed (CODIT) model of a universal mobile telecommunication system (UMTS).

Although the position of the scatterer depends on the terminal location, there are general models representing the position of the scatterer. These models are the Ring Of Scatterers (ROS), Disk Of Scatterers (DOS) and Gaussian Scattering (DOS) models.                     

These models obtain the mean and variance from the signals received from the terminal (from the base station), and then use them to calculate the true range R _i from which the invisible path error is eliminated and finally to calculate the terminal position. have. In general, the range equation between the i- th base station and the terminal is expressed as Equation 27 below.

In general, the distance equation of Equation 23 includes an error due to propagation of the invisible path due to the characteristics of the mobile communication channel, and when the distance l _i including the error is expressed as a relationship with the true distance, It can be expressed as in Equation 24 below.

의 값을 갖는다.Here, about invisible path propagation

Has the value of.

A) Ring of Scatterers Model

In the circular scatterer model, the scatterers are located on a circle having a radius r _DOS around the terminal as shown in FIG. 9. Distribution of the angle of arrival θ of the signal received at the terminal has a uniform distribution (uniform distribution) in the range [0, 2] as shown in Equation 25 below.

로 나타낼 수 있다. 도 9로부터 코사인(cosine) 법칙을 이용하여 θ와 수신된 거리간의 관계를 수식으로 나타내면 다음의 [수학식 26]과 같다. If the true range between the i th base station and the terminal is R _i , and the actual propagated distance (the distance from the received signal) is l _i , the error due to the invisible path propagation is

It can be represented as. Using the cosine law from FIG. 9, the relationship between θ and the received distance is expressed by the following Equation (26).

를 대입하여 θ를 비가시경로 오차

의 함수로 정리하면 하기의 [수학식 27]과 같다.In [Equation 26] above

By substituting the error ratio in the θ pm

When summarized as a function of [Equation 27].

Probability Density Function (PDF: Probability Density Function) of the measured distance can be represented by Equation 28 below, and by substituting θ , the equation can be expressed by Equation 29 below.

In the same manner, the invisible path component function is expressed by Equation 30 below.

B) disk of Scatterers Model

In this model, the scatterers are located on a circular disk with a fixed radius R _d around the terminal position as shown in FIG. 10. The distance r _DOS from the terminal to the scatterer is uniformly distributed in the range [0, R _d ], and the signal arrival angle θ to the terminal has a uniform distribution in the range [0, 2π]. 10 can be expressed as in Equation 31 below.

In order to represent this as a function of the measured signal l _i , converting ( x, y ) to a function of ( l _i , φ ), the joint probability density function (joint PDF) can be expressed as Equation 32 below. .

In order to calculate the probability density function p _DOS ( l _i ) to be obtained from Equation 32, the absolute value of Jacobian is calculated in Equation 31, and Equation 33 After substituting and arranging as shown in [Equation 35], and integrating with [ phi ] as shown in [Equation 36] below.

Here, φ ⁽¹⁾ and φ ⁽²⁾ are the same as in the following Equations 37 and 38.

Finally, p _DOS ( l _i ) to be obtained may be expressed as in Equation 39 below.

이고, here,

ego,

이다.

to be.

C) Gaussian Scattering Model

에 의해 결정된다.In the Gaussian scatterer model, the standard deviation of the Gaussian distribution is different from the circular or disc-shaped scatterer model, whereas the scattering environment is determined by the radius of the circle or disk.

Determined by

11 illustrates a Gaussian scatterer model. Finally, in order to obtain a PDF for l _i , when the PDF of the scatterer position is displayed in FIG. 11, Equation 40 is as follows.

이다. 가우시안 산란체 모델의 PDF를 x, y의 함수가 아닌 r_{gaus s}, θ의 함수로 나타내기 위해 자코비안 J를 이용하여 나타내면 하기의 [수학식 41]과 같다. here,

to be. In order to express the PDF of the Gaussian scatterer model as a function of r _{gaus s} and θ , not as a function of x and y, it is represented by Equation 41 below.

Substituting Jacobian J into the above Equation 41, the combined PDF of r _{gaus s} and θ is represented by Equation 42 below.

To obtain P _gauss (l _i ) from Equation 42, p _gauss is expressed as a function of l _i and other variables. In FIG. 11, according to the cosine law, (l _i -r _gauss ) ² may be represented by Equation 43 below, and r _gauss may be summarized as shown in Equation 44 below.

Since the combined PDF of l _i and θ can be represented by Equation 45 below, this is given by Equation 46 below.

lowest (marginal) PDF of l _i is for θ, such as do not harm the guhana closed form (closed-form) by integrating the equation 46] of the above for θ is present, [Equation 47] below The numerical summation is calculated as shown in Equation 47 below.

2) Mean and Variance Matching Algorithm (MVMA)

The present invention can be applied to both the direction from the base station to the terminal (forward link) and the positioning from the terminal to the base station (reverse link: reverse link), but in the present description, for a signal transmitted from the base station to the terminal for convenience. Only the signal received from the positioning, i.e., the terminal, will be described.

Actual application examples of the present invention are shown in Figure 12a and 12b.

In FIG. 12a, the i th time of arrival (TOA) received by the terminal through the multipath propagation signal transmitted from the base station BS _j is represented by l _ji, and the radius of the circle is represented by r _{RO S.} In FIG. 12B, l _ji denotes the i th TOA in which the signal transmitted from the base station BS _j is received by the terminal through multipath propagation, and R _d denotes the radius of the scatterer disk. 12A and 12B, the distribution form of TOA is shown in the TOA profile as a dotted line in each model. The probability distribution of the scatterer model in FIGS. 12A and 12B is for the TOA and is not related to the strength or amplitude of the received signal. A Gaussian scatterer model can also be obtained similarly to FIGS. 12A and 12B.

Based on the PDF of the scatterer model obtained in the previous section, the true (or visible path) distance R _i between the base station and the terminal to be calculated and the radius of the scatterer are determined. In order to simply indicate the radius of the scatterer, it is represented by r, and is represented by Equation 48 below.

Since the representation of the second moments in MVMA should be the same as the moments calculated from the measured TOA values, R _i and r are calculated based on this.

The mean of a signal including an error due to the invisible path propagation received from the base station from FIGS. 12A and 12B is represented by Equation 49 below, and the measured variance is represented by Equation 50 below. ], The mean is equal to [Equation 51] below, and the variance is shown in [Equation 52] below.                     

However, the circular scatterer model can obtain a "closed-form" solution for both values, but the disk-shaped scatterer model and the Gaussian model cannot obtain the "closed-form" solution. Thus, a model that cannot obtain a "closed-form" solution may be solved using other mathematical techniques.

As described above, after R _i and r are obtained through MVMA, the terminal location is obtained through a linear line of position (LLOP), least square, Taylor series, and the like.

FIG. 13 is a performance comparison graph of the ROS algorithm among the invisible path error correction methods according to another embodiment of the present invention, and shows that the performance of the MVMA algorithm is excellent for all scatterer numbers, and FIG. 14 is the present invention. As a graph comparing the performance of the DOS algorithm among the invisible path error correction methods according to another embodiment of the present invention, when the scatterer radius is larger than about 200 m, the performance of the MVMA algorithm is excellent for all the scatterer numbers. 15 is a performance comparison graph for a Gaussian algorithm among the invisible path error correction methods according to another embodiment of the present invention, and shows that the performance of the MVMA algorithm applied to the Gaussian environment is excellent.

A) MVMA for the ROS Model

When the above Equation 51 is applied to the circular scatterer model, it is the same as Equation 53 below, which is integrated by parts and developed using Equation 54 below. The estimated average value can be obtained as shown in Equation 55.

When the variance is obtained in the same manner, Equation 56 is used.

과

에 대해 스티페스트 디센트(steepest descent) 방법과 같은 기울기(gradient) 알고리즘을 이용하여 R_i 와 r_DOS 를 구한다. 이를 수식으로 나타내면 하기의 [수학식 57]과 같다.Here, in order to obtain R _i and r _DOS , the mean and the variance of Equations 55 and 56 must be equal to the mean and the variance from the measured TOA profile. However, since these equations are nonlinear equations, direct solutions are not easy. Therefore, we use the difference between the calculated and measured values in the mean and variance. In other words,

and

R _i and r _DOS are calculated using a gradient algorithm such as the steepest descent method. This is represented by the following formula (57).

Where β is the step-size to adjust to converge the algorithm.

The steepest descent algorithm continues to run until it converges to the solutions for R _i and r _DOS . When R ₁ , R ₂ , and R ₃ are obtained, the position of the terminal can be obtained by a method such as LLOP and least squares.

B) MVMA for the DOS Model

PDF of the disk-shaped scatterer model from N (l _i ) and D (l _i ) in Equation 39 is a function of l _j , R _j, and R _d . Therefore, the average and the variance of the measured TOA are a function of R _j and R _d . In using MVMA to find R _j and R _d , the "closed-form" equation for mean and variance cannot be obtained, so l _j is divided into smaller intervals as shown in Equations 58 and 59 below. Integrate to find.

Here, Δ n means a small interval obtained by dividing l _j and is a interval used for the Rimann sum.

R _j and R _d are obtained by equalizing the average and the variance from the measured values with Equations 58 and 59 above. In other words, R _j and R _d is the average of the measured values and the variance μ _{DOS, j} and σ _{DOS, j} by the difference between the equation (58) and [Equation 59] You can get it together.

However, since the "closed-form" expression for Equation 60 does not exist, an error surface having minimum values in R _j and R _d is formed to solve the solution. The error plane is configured as in Equation 61 below.

일 때 구할 수 있다.In Equation 61, the minimum value of the error plane is that when each term is 0, that is,

Can be obtained when

The reason for using the standard deviation instead of the variance in the error plane is that the variance is larger than the mean, which can dominate the overall minimization process.

However, since the solution must be obtained numerically for the mean and the variance, the error plan must be obtained numerically as well. In this case, the Nedler-Mead algorithm, which is a simplex method, can be used to find the minimum value. After obtaining R _i through an algorithm that can obtain the minimum value, the location of the terminal can be determined by applying it to general positioning algorithms such as LLOP and least square.

C) MVMA for the Gaussian Model

When MVMA is applied to a Gaussian model, the PDF for P _gauss (l _i ) is given by Equation 47 above. When the integration is performed to find the average and the variance of the measured signal, Equations 62 and 63 are used.

Here, Δ n represents the step size of the “Riemann” summation, and the maximum limit of the integral may theoretically have a value larger than the cell radius, but is set to 2 R _j for convenience in consideration of logical and practical aspects. Solving the above [Equation 47] by substituting the above [Equation 62] and [Equation 63], the result is the same as [Equation 64] and [Equation 65].

In the MVMA for the Gaussian model, the equations (64) and (65) are matched with the mean and the variance of the measured signals, as in the MVMA for the disc-shaped scatterer model, to obtain R _j and σ _g . Even in this case, since the "closed-form" solution does not exist, the solution is solved numerically, and the error plane is formed in the disk scatterer model as in the MVMA. The Nedler-Mead algorithm may be applied, and the position of the terminal may be determined by a method such as LLOP and least square after obtaining R _j and σ _g .

The method of the present invention as described above may be implemented as a program and stored in a computer-readable recording medium (CD-ROM, RAM, ROM, floppy disk, hard disk, magneto-optical disk, etc.).

The present invention described above is not limited to the above-described embodiments and the accompanying drawings, and various substitutions, modifications, and changes are possible in the art without departing from the technical spirit of the present invention. It will be clear to those of ordinary knowledge.

As described above, the present invention implements a wireless positioning system in a mobile communication network including a CDMA scheme such as cellular, PCS, and IMT-2000. By improving the accuracy by resolving errors, it enables positioning services requiring precise resolution, and expands the coverage service area so that users can receive the same level of service as they are currently receiving through landline while on the go. It works.

Therefore, the present invention can provide a variety of services by identifying the location of the wireless terminal in such an environment, it is possible to provide a high-accuracy service by improving the radio positioning error due to the invisible path radio waves included in the received signal. That is, from the user's point of view, various location-related services, such as travel information guidance, vehicle tracking, location information service, location-based advertising, etc. in a low-level area, can be accurately known to the terminal location, so that information within a desired distance from the actual terminal location can be obtained. Since only the user can correctly send and receive, there is an effect of improving the quality of service.

In addition, the present invention improves the accuracy of the location service (LCS) or location based service (LCS) by improving the error due to the invisible path in the newly introduced IMT-2000 system, and provides various services. As the accuracy of the terminal location increases, only necessary information can be transmitted and received in terms of systems and networks, thereby preventing excessive system installation and unnecessary waste of radio resources for radio location services.

Claims (15)

In the non-visible path error correction method applied to the wireless positioning system, For each received signal that can be assumed to include an error due to invisible path propagation, a first step of creating a constraint and obtaining a weight value through an optimization process; And A second step of determining the position of the terminal by improving an error due to an invisible path through a known positioning method based on the obtained weight value Invisible path error correction method for a wireless positioning system comprising a. The method of claim 1, When the error due to the invisible path is improved, An invisible path error correction method for a wireless positioning system, characterized in that it improves an error due to an invisible path without distinguishing whether it is received from a visible path or an invisible path. The method of claim 1, When the error due to the invisible path is improved, An invisible path error correction method for a wireless positioning system, characterized in that for improving the error due to invisible path propagation without statistical characteristics of the signal received from the invisible path. The method of claim 1, When the error due to the invisible path is improved, Invisible path error correction method for a wireless positioning system, characterized in that for improving the error due to invisible path propagation without receiving continuous signal. The method according to any one of claims 1 to 4, When the error due to the invisible path is improved, An invisible path error correction method for a wireless positioning system, characterized by a solution based on a linear line of position (LLOP) algorithm for a visible path distance. In the non-visible path error correction method applied to the wireless positioning system, For received signals that may be assumed to contain errors due to invisible path propagation, a first step of obtaining a second moment and comparing it with theoretical values of an existing scattering model; A second step of obtaining a visible path distance value and a parameter value of a scattering model according to the comparison result of the first step; And Based on the obtained values, a third step of determining the position of the terminal by improving the error due to the invisible path through a known positioning method Invisible path error correction method for a wireless positioning system comprising a. The method of claim 6, When the error due to the invisible path is improved, Invisible path error correction method for a wireless positioning system, characterized in that any one of the circular, disk-shaped, Gaussian scatterer model is introduced. The method of claim 6, When the error due to the invisible path is improved, Invisible path error for wireless positioning system, which uses the second moment to match the mean and variance obtained from any one of the circular, disk and Gaussian scatterer models with the mean and variance obtained from the received signal. Calibration method. The method of claim 8, When the error due to the invisible path is improved, In case of the circular scatterer model, a gradient algorithm such as a steepest descent method is used to solve the invisible path error correction method for the wireless positioning system. The method of claim 8, When the error due to the invisible path is improved, A method for correcting invisible path errors for a wireless positioning system, characterized by constructing an error plane and numerically matching the mean and the variance when a closed-form solution cannot be obtained. The method of claim 10, In the error plane, An invisible path error correction method for a wireless positioning system, characterized in that the magnitude is balanced using an absolute value of standard deviation without using variance. The method of claim 11, In the error plane, Invisible path error correction method for a wireless positioning system characterized in that the minimum value is obtained by applying the Nedler-Mead algorithm. The method according to any one of claims 6 to 12, When the error due to the invisible path is improved, An invisible path error correction method for a wireless positioning system, characterized by a solution of a visible path distance by an LLOP algorithm. In a wireless positioning system with a processor, A first function, for each received signal that can be assumed to include an error due to invisible path propagation, to create a constraint and obtain a weight value through an optimization process; And A second function of determining a position of a terminal by improving an error due to an invisible path through a known positioning method based on the obtained weight value A computer-readable recording medium having recorded thereon a program for realizing this. In a wireless positioning system with a processor, For received signals that may be assumed to contain errors due to invisible path propagation, a first function of obtaining second moments and comparing them with theoretical values of an existing scattering model; A second function of obtaining a visible path distance value and a parameter value of a scattering model according to a comparison result in the first function; And Based on the obtained values, a third function for determining the position of the terminal by improving the error due to the invisible path through a known positioning method A computer-readable recording medium having recorded thereon a program for realizing this.

Images