KR20130021514A - Method and system for wireless positioning - Google Patents

Abstract

PURPOSE: A wireless positioning method and a system thereof are provided to improve the accuracy of tag location estimation by independently estimating measurement noise while estimating the tag location by applying an adaptive extended Kalman filter. CONSTITUTION: A location recognition signal is transmitted(S10). The location recognition signal, including multiple pulses with predetermined identical time intervals, is transmitted from a tag to each of multiple signal receiving parts. Reception time information and measurement noise information for the location recognition signal are transmitted(S20). The reception time information and measurement noise information for the location recognition signal are generated at each of the multiple signal receiving parts according to the transmitted location recognition signal. The location of the tag is estimated in real time by using the transmitted reception time information and measurement noise information(S30). [Reference numerals] (AA) Start; (BB) End; (S10) Transmitting a location recognition signal from a tag; (S20) Transmitting reception time information and measurement noise information from a plurality of signal receiving parts; (S30) Estimating the location of the tag in real time

Description

Method and system for wireless positioning

The present invention relates to a wireless positioning method and system. More particularly, the present invention relates to a wireless positioning method and system capable of more accurately estimating the position of a tag in real time based on a time difference of arrival (TDOA).

When using radio waves with excellent time resolution, such as Ultra WideBand (UWB) for wireless positioning, highly accurate position estimation of 30 cm or less with respect to a wireless positioning object is possible, and a representative wireless positioning method is a time of arrival (TOA) method. And TDOA (Time Difference Of Arrival).

First, in the case of the TOA method, the position of the tag is estimated by measuring the propagation movement time between a positioning object such as a tag and three or more receivers (Beacon). Accordingly, in the case of the TOA method, the synchronization between the tag and the receivers is performed. The tag has a feature that both transmission and reception of radio waves are required.

On the other hand, the TDOA method is a method of estimating the position of a tag by using a time difference with respect to the time when a signal transmitted from a positioning object such as a tag reaches each receiver. Since the synchronization is not required separately, the tag can be configured only by the transmitter, so that the tag can be miniaturized easily.

As described above, in the case of the TOA method and the TDOA method where position recognition is performed by directly measuring the arrival time of the signal arriving at the receiver, highly accurate position estimation is possible depending on the characteristics of the radio wave used. Since elements are included, there is a problem that the position estimation performance is very sensitive to changes in the wireless channel environment and the hardware performance of the transceiver.

Therefore, to solve this problem, various filtering methods such as a Kalman filter or a particle filter (PF) have been developed to estimate the position, but in the case of applying the Kalman filter, the TOA method and the TDOA method are applied. As such, when a nonlinear element is included in a system, noise prediction cannot be easily performed.

In addition, when the particle filter is applied, a large amount of computation is required in the process of generating and processing a plurality of particles due to the characteristics of a filter that removes noise that is not modeled through a plurality of randomly generated particles and nonlinearity of the system. If the position of dozens or hundreds of tags needs to be estimated in real time several times per second, or if the hardware resources required for filtering operations using a low power embedded processor are very limited, there is a problem.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and it is possible to independently estimate measurement noise generated from a plurality of receivers in the process of estimating the position of a tag in real time by applying an Adaptive Extended Kalman Filter. It is an object of the present invention to provide a wireless positioning method and system that can enable a more shaped position estimation performance.

Wireless positioning method according to a preferred embodiment of the present invention for achieving the above object comprises the steps of: (a) transmitting a position recognition signal including a plurality of pulses having a predetermined same time interval from each of a plurality of signal receivers; (b) transmitting reception time information and measurement noise information for the position recognition signal generated by each of the plurality of signal receivers according to the transmitted position recognition signal; And (c) estimating the position of the tag in real time using the transmitted reception time information and the measured noise information.

In addition, the wireless positioning system according to a preferred embodiment of the present invention includes a tag for transmitting a position recognition signal including a plurality of pulses having a predetermined same time interval; A plurality of signal receivers each receiving a position recognition signal transmitted from the tag to generate reception time information and measurement noise information for the position recognition signal; And a position estimator for estimating the position of the tag in real time after receiving the generated reception time information and the measured noise information from the plurality of signal receivers.

According to the present invention, the position recognition signal transmitted from the tag to the plurality of signal receivers is configured to include a plurality of pulses having a predetermined same time interval, thereby applying an adaptive extended Kalman filter to independently estimate the position of the tag. In this way, the measurement noise can be predicted, thereby improving the position estimation accuracy of the tag.

In addition, by transmitting the location recognition signal in the UWB band and packet form, even when using a large number of tags at the same time, it is possible to perform the position estimation for each tag in real time.

1 is a block diagram of a wireless positioning system according to a preferred embodiment of the present invention;
2 is a reference diagram of a location recognition signal transmitted from a tag and a location recognition signal received from a plurality of signal receivers;
3 is a reference diagram for measurement noise distribution,
4 is a detailed block diagram of the position estimator 30 of FIG. 1;
5 is a flow chart for a wireless positioning method according to a preferred embodiment of the present invention;
6 is a detailed flowchart of S30 of FIG. 5, and
7 is a reference graph for a tag position estimation result.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. First, in adding reference numerals to the components of each drawing, it should be noted that the same reference numerals are used as much as possible even if displayed on different drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. In addition, preferred embodiments of the present invention will be described below, but the technical idea of the present invention may be implemented by those skilled in the art without being limited or limited thereto.

1 is a block diagram of a wireless positioning system according to a preferred embodiment of the present invention, FIG. 2 is a reference diagram for a position recognition signal transmitted from a tag and a position recognition signal received at a plurality of signal receivers, and FIG. 3 is measurement noise. See also distribution.

As shown in FIG. 1, the wireless positioning system 1 according to a preferred embodiment of the present invention includes a tag 10, a plurality of signal receivers 20a, 20b, 20c, and 20d, and a position estimator 30. do.

The tag 10 transmits a position recognition signal including a plurality of pulses having a predetermined same time interval to the plurality of signal receivers 20a, 20b, 20c, and 20d.

At this time, as shown in (a) of FIG. 2, the position recognition signal transmitted from the tag 10 includes a plurality of pulses having the same predetermined time interval T _d , and (b) and (b) of FIG. As shown in c), the position recognition signal transmitted to the signal receiver 20a and the signal receiver 20d includes noise on a propagation channel, sampling noise, noise due to a mismatch between synchronizations between the receiver, and the signal receiver 20a. Measurement noise caused by hardware jitter has a different shape when compared with the position recognition signal shown in FIG.

In addition, the position recognition signal may be a signal of the UWB (Ultra Wide Band) band and may be transmitted to the plurality of signal receivers 20a, 20b, 20c, and 20d in the form of a packet, by applying a plurality of tags instead of a single tag. Even when configuring a wireless positioning system, a plurality of signal receivers 20a, 20b, 20c, and 20d may further distinguish a unique identification signal from each other.

In addition, since the transmission period of the location recognition signal transmitted from the tag 10 may be 1Hz to 5Hz, and the length of the packet constituting the location recognition signal is very short, a plurality of tags even when using the basic ALOHA Mac protocol For example, positioning signals can be transmitted simultaneously from hundreds of tags.

The reason for configuring the position recognition signal transmitted from the tag 10 to the plurality of signal receivers 20a, 20b, 20c, and 20d as described above includes a plurality of pulses having a predetermined same time interval. .

First, if a signal receiving unit, which is a reference among the plurality of signal receiving units, is m and any one of the other signal receiving units is n, the tag 10 transmits a k-th position recognition signal to the signal receiving units m and n, respectively. The TDOA value may be expressed as Equation 1 below.

Here, T _{m , k} is the time the position recognition signal is moved from the tag 10 to the signal receiver m, T _{n , k} is the time the position recognition signal is moved from the tag 10 to the signal receiver n, N _{m, k} denotes measurement noise generated in the position recognition signal reception process of the signal receiver m, and N _{n , k} denotes measurement noise generated in the position recognition signal reception process of the signal receiver n, N _{m , Let k} , N _{n , k be} Gaussian forms with mean zero and variance σ _{m, k} and σ _{n, k} , respectively.

In this case, the measurement noise may include noise on a propagation channel, sampling noise, noise due to mismatch of synchronization between signal receivers, and noise due to signal jitter hardware jitter, and particularly manufactured in a CMOS form. Deviation in the manufacturing process of the signal receiver may cause considerable noise during the high-speed sampling operation of the GHz class and may exhibit different characteristics for each signal receiver.

In addition, since sampling noise, synchronization noise, and hardware noise are dominant compared to noise on a propagation channel, N _{m , k} and N _{n , k} can be approximated as independent of each other. Thus _, TDOA _{mn , k} has an average of 0 and variance. It can have a Gaussian noise σ _{mn , k , where} σ _{m, k} + σ _{n, k} . As shown in FIG. 3, a Gaussian-shaped distribution is obtained for a measured noise distribution graph of about 10,000 actual measured TDOA values. I can confirm that I have.

In addition, in order to estimate σ _{mn , k} , which is the measurement noise of TDOA _{mn , k} , the k + 1th position, the next position recognition signal, is located at a predetermined time interval T _D after the k-th position recognition signal is transmitted from the tag 10. When the recognition signal is transmitted to the signal receivers m and n, and after receiving consecutive signals for each signal receiver (m and n), the time differences TDOA _{m , k, k + 1} and TDOA _{n , k, k + 1} are calculated. It can be expressed as Equation 2 below.

In addition, the difference between TDOA _{m , k, k + 1} and TDOA _{n , k, k + 1} can be expressed by Equation 3 below.

In this case, N _{m , k + 1} -N _{n , k + 1} can be modeled as Gaussian noise with an average of 0 and a variance of σ _{mn , k + 1} ≒ (σ _{m, k + 1} + σ _{n, k + 1} ). N _{m , k} -N _{n , k} can be modeled as Gaussian noise with mean 0 and variance σ _{mn , k} ≒ (σ _{m, k} + σ _{n, k} ), N _{m , k + 1 Assume that} -N _{n , k + 1} and N _{m , k} -N _{n , k} are independent of each other and stationary for a predetermined time interval (T _D ), then TDOA _{m , k, k + 1} -TDOA _{n , k, k + 1} can be approximated with Gaussian noise with a mean of 0 and a variance of 2σ _{mn , k} (≒ σ _{mn , k + 1} + σ _{mn , k} )

As such, the magnitudes of the measured noises σ _{mn , k} for the position recognition signals transmitted to the kth signals in the signal receivers m and n can be estimated by receiving a continuous position recognition signal having a fixed time interval for each signal receiver. do.

Therefore, in the case of the wireless positioning system 1 according to the preferred embodiment of the present invention, the position recognition signal including a plurality of pulses having the same predetermined time interval is transmitted in the form of a packet to measure the position recognition signal for each signal receiver. The noise can be estimated, and as a result, the measurement noise can be estimated independently in the process of estimating the position of the tag 10 in real time by applying an adaptive extended Kalman filter. It will be described later with reference to 4.

The plurality of signal receivers 20a, 20b, 20c, and 20d respectively receive the position recognition signals transmitted from the tag 10 to receive the reception time information of the position signals and the measurement noise generated during the reception of the position signals. After generating the measured noise information for the transmission to the position estimator (30).

In this case, although not shown in FIG. 1, the plurality of signal receivers 20a, 20b, 20c, and 20d may be connected to each other through a clock sync line to receive the position recognition signal according to the same synchronization signal. Can be.

The position estimator 30 estimates the position of the tag 10 in real time using the reception time information and the measured noise information transmitted from each of the plurality of signal receivers 20a, 20b, 20c, and 20d.

In this case, the position estimator 30 estimates the position of the tag 10 by using a time difference at which a signal transmitted from the tag 10 reaches each of the plurality of signal receivers 20a, 20b, 20c, and 20d. The location of the tag 10 may be estimated in real time using a Time Difference Of Arrival method.

In addition, a detailed configuration of the position estimating unit 30 will be described below with reference to FIG. 4.

4 is a detailed block diagram of the position estimator 30 of FIG. 1.

As shown in FIG. 3, the position estimator 30 includes a TDOA calculator 32, a TDOA noise calculator 34, and a TDOA data processor 36.

The TDOA calculator 32 is a predetermined reference signal receiver among a plurality of signal receivers 20a, 20b, 20c, and 20d using the reception time information transmitted from the signal receivers 20a, 20b, 20c, and 20d, respectively. A TDOA (Time Difference Of Arrival) value of other signal receivers 20b, 20c, and 20d except for the reference signal receiver for 20a is calculated.

In this case, the TDOA values of the other signal receivers 20b, 20c, and 20d with respect to the reference signal receiver 20a may be calculated as in Equation 4 below.

Here, t (k + 1) means the TDOA values of the other signal receivers 20b, 20c, and 20d with respect to the reference signal receiver 20a at k + 1 time.

The TDOA noise calculator 34 uses the measured noise information transmitted from the plurality of signal receivers 20a, 20b, 20c, and 20d, respectively, to other signal receivers 20b, 20c, and 20b for the reference signal receiver 20a. Calculate the TDOA noise value of 20d).

In this case, the TDOA noise values of the reference signal receiver 20a and the other signal receivers 20b, 20c, and 20d may be calculated as in Equation 5 below.

Here, v (k + 1) means the TDOA noise values of the other signal receivers 20b, 20c, and 20d with respect to the reference signal receiver 20a at k + 1 time.

The TDOA data processor 36 estimates the position of the tag 10 in real time using the TDOA value calculated by the TDOA calculator 32 and the TDOA noise value calculated by the TDOA noise calculator 34.

In this case, the TDOA data processor 36 may estimate the position of the tag 10 in real time by applying the TDOA value and the TDOA noise value to an Adaptive Extended Kalman Filter.

In addition, the detailed process of estimating the position of the tag 10 in real time by applying the adaptive extended Kalman filter by the TDOA data processor 36 is as follows.

Prior to real-time estimation of the position of the tag 10 performed by the TDOA data processor 36, a predetermined position estimation model of the tag 10 is applied, which is represented by Equation 6 below.

Here, s (k + 1) means the state of the tag 10 at k + 1 time, that is, the position and speed of the tag 10 at k + 1 time, and w (k) Denotes process noise that is not modeled, that is, an acceleration component of the tag 10, and Δ denotes a time between k and k + 1, that is, a location-aware signal in the form of a packet transmitted from the tag 10. Means the cycle of.

The TDOA value calculated by the TDOA calculator 32 according to Equation 6 may be expressed as Equation 7 below.

Here, t (k + 1) is the TDOA value of the other signal receivers 20b, 20c, and 20d to the reference signal receiver 20a at k + 1 time, and v (k + 1) is at k + 1 time. The TDOA noise value, h (s (k + 1)) of the other signal receivers 20b, 20c, and 20d with respect to the reference signal receiver 20a indicates the relationship between s (k + 1) and t (k + 1). In other words, the difference between the distances between the other signal receivers 20b, 20c, and 20d from the position p (k + 1) of the current tag 10 and the reference signal receiver 20a is shown. The value divided by).

The TDOA data processor 36 applies the Adaptive Extended Kalman Filter to calculate the position of the tag 10 in real time, and the position of the tag 10 in the previous state estimated by the Adaptive Extended Kalman Filter is It can be expressed as Equation 8 below.

은 이전 데이터를 근거로 추정되는 태그(10)의 스테이트(state), 다시 말해서 추정되는 태그(10)의 위치를 의미하며, P는

의 공분산을 의미하고, Q는 TDOA 데이터 처리부(36)에서 수행되는 태그(10)의 위치 추정 시 발생되는 처리 노이즈(process noise)의 공분산을 의미한다.From here,

Denotes a state of the tag 10 estimated based on previous data, that is, a position of the estimated tag 10, P is

Q denotes a covariance of, and Q denotes a covariance of process noise generated when the position of the tag 10 performed by the TDOA data processor 36 is estimated.

In addition, the position of the tag 10 updated by applying the adaptive extended Kalman filter in the TDOA data processor 36 may be represented by Equation 9 below.

Where ▽ h denotes Jacobian in Essence for applying the nonlinear function h to the adaptive extended Kalman filter, K means Kalman Gain, and R (K + 1) is measured The covariance of the noise (measurement noise), and y (k + 1) refers to the position of the tag 10 is updated in the k + 1 time.

At this time, the Kalman gain is determined by the covariance Q of the process noise and the covariance R of the measurement noise, and when the measurement noise is large, K has a small value, resulting in a new observation value. This makes it possible to reduce the proportion of innovation (y).

In other words, the Kalman gain depends on the relative magnitudes of the measurement noise and the processing noise. In the present invention, each signal receiving unit transmits a position recognition signal including a plurality of pulses having a predetermined same time interval in the form of a packet. By independently estimating the measurement noise for the position-aware signal, it is possible to apply the optimal Kalman gain every loop of the adaptive extended Kalman filter.

을 사용하는 것이 가능하다.In addition, the covariance R of the measured noise may be expressed as Equation 10 below, and the covariance of a smooth form as shown in Equation 11 below.

It is possible to use

Here, σ _{ab , k + 1} is the TDOA noise variance value of the reference signal receiver 20a and the signal receiver 20b at k + 1 time _, and σ _{ac , k + 1} is the reference signal receiver 20a at k + 1 time. ) And the TDOA noise variance value of the signal receiver 20c, and σ _{ad , k + 1} mean the TDOA noise variance values of the reference signal receiver 20a and the signal receiver 20d at k + 1 time.

Here, lambda means a smoothing intensity determination coefficient having a value between 0 and 1.

In addition, the estimated position and covariance of the tag 10 at k + 1 time updated by the adaptive extended Kalman filter can be expressed by Equation 12 below.

5 is a flowchart of a radio positioning method according to a preferred embodiment of the present invention.

As shown in FIG. 5, a location recognition signal including a plurality of pulses having the same predetermined time interval from the tag 10 is transmitted to the plurality of signal receivers 20a, 20b, 20c, and 20d in S10.

In S20, the plurality of signal receivers 20a, 20b, 20c, and 20d generate the reception time information and the measurement noise information for the position recognition signal, and transmit the generated signal to the position estimator 30.

If the position estimation unit 30 estimates the position of the tag 10 by using the reception time information and the measured noise information in S30, the termination is performed.

At this time, the detailed process of S30 will be described with reference to FIG.

6 is a detailed flowchart for S30.

As shown in FIG. 6, in S32, the TDOA calculator 32 uses the reception time information transmitted from the plurality of signal receivers 20a, 20b, 20c, and 20d to transmit the plurality of signal receivers 20a, 20b, 20c, and the like. The TDOA (Time Difference Of Arrival) values of the other signal receivers 20b, 20c, and 20d with respect to the predetermined reference signal receiver 20a are calculated.

In this case, the method of calculating the TDOA (Time Difference Of Arrival) values of the other signal receivers 20b, 20c, and 20d with respect to the reference signal receiver 20a by the TDOA calculator 32 in S32 has been described above. Is omitted.

In operation S34, the TDOA noise calculator 34 uses the measured noise information transmitted from the plurality of signal receivers 20a, 20b, 20c, and 20d to determine other signal receivers 20b, 20c, and 20b for the reference signal receiver 20a. Calculate the TDOA noise value of 20d).

In this case, the method of calculating the TDOA noise values of the other signal receivers 20b, 20c, and 20d with respect to the reference signal receiver 20a by the TDOA noise calculator 34 in S34 has been described above. .

If the TDOA data processor 36 estimates the position of the tag 10 in real time using the TDOA value and the TDOA noise value in S36, the termination is completed.

In this case, in S36, the TDOA data processor 36 may apply an adaptive extended Kalman filter to the TDOA value and the TDOA noise value to estimate the position of the tag 10 in real time. Since the method of applying the Adaptive Extended Kalman Filter has been described above, a detailed description thereof will be omitted.

7 is a reference graph for a tag position estimation result.

As shown in Fig. 7, the moving path of the actual tag 10 (dashed line portion on the graph), the moving path of the tag 10 estimated by the conventional method (solid line portion indicated by ○ on the graph), and the wireless position of the present invention. Comparing the moving path of the tag 10 estimated by the system 1 (solid line portion indicated by X on the graph), the moving path of the tag 10 estimated by the wireless location system 1 of the present invention is compared to the conventional method. Compared to the moving path of the actual tag 10 can be confirmed.

The wireless positioning method and system of the present invention is configured in the form of a plurality of pulses having a predetermined same time interval for the position recognition signal transmitted from the tag 10 to the plurality of signal receivers 20a, 20b, 20c, 20d. By applying the adaptive extended Kalman filter in the TDOA data processing unit 36 to estimate the measurement noise independently in the process of estimating the position of the tag 10 in real time, the accuracy of position estimation for the tag can be improved. Have

In addition, even when a large number of tags are used simultaneously by transmitting the location recognition signals transmitted from the tag 10 to the plurality of signal receivers 20a, 20b, 20c, and 20d in the form of UWB signals and packets in the GHz band. It is possible to perform the position estimation for in real time.

Therefore, in the case of the wireless positioning system 1 of the present invention, it can be used for workplace safety management, in-house logistics management, hospital expensive equipment management, and patient management, and by using the UWB signal of the GHz band as a location recognition signal, a single radio positioning is performed. Each system can have a precise location-aware resolution of 30cm or less within the 20m x 20m range.

The above description is merely illustrative of the technical idea of the present invention, and various modifications, changes, and substitutions may be made by those skilled in the art without departing from the essential characteristics of the present invention. This will be possible. Therefore, the embodiments disclosed in the present invention and the accompanying drawings are intended to illustrate and not to limit the technical spirit of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments and the accompanying drawings . The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of the present invention.

(1): wireless positioning system (10): tags
(20a, 20b, 20c, 20d): signal receiving unit 30: position estimating unit
(32): TDOA calculation section 34: TDOA noise calculation section
36: TDOA data processing unit

Claims (8)

(a) transmitting, from a tag, a position recognition signal comprising a plurality of pulses having a predetermined same time interval to each of the plurality of signal receivers;
(b) transmitting reception time information and measurement noise information for the position recognition signal generated by each of the plurality of signal receivers according to the transmitted position recognition signal; And
(c) estimating the position of the tag in real time using the transmitted reception time information and the measured noise information. The method of claim 1,
The step (c)
(c1) calculating time difference of arrival (TDOA) values of signal receivers other than the reference signal receiver for a predetermined reference signal receiver among the plurality of signal receivers using the reception time information;
(c2) calculating TDOA noise values of the other signal receivers with respect to the reference signal receiver using the measured noise information; And
(c3) estimating the position of the tag in real time using the TDOA value and the TDOA noise value. The method of claim 2,
The step (c3)
And estimating the position of the tag in real time by applying an adaptive extended Kalman filter to the TODA value and the TDOA noise value. The method of claim 1,
In the step (a)
The location recognition signal is a UWB (Ultra WideBand) band signal, the wireless positioning method characterized in that it is transmitted to the plurality of signal receivers in the form of a packet. A tag for transmitting a position recognition signal comprising a plurality of pulses having a predetermined same time interval;
A plurality of signal receivers each receiving a position recognition signal transmitted from the tag to generate reception time information and measurement noise information for the position recognition signal; And
And a position estimator for estimating the position of the tag in real time after receiving the generated reception time information and the measured noise information from the plurality of signal receivers. 6. The method of claim 5,
The position estimating unit,
A TDOA calculator configured to calculate a time difference of arrival (TDOA) value of signal receivers other than the reference signal receiver for a reference signal receiver that is predetermined among the plurality of signal receivers using the reception time information, and the measured noise information A TDOA noise calculator for calculating TDOA noise values of the other signal receivers with respect to the reference signal receiver, and a TDOA data processor for estimating the position of the tag in real time using the TDOA value and the TDOA noise value. Wireless positioning system comprising a. The method according to claim 6,
And the TDOA data processor estimates the position of the tag in real time by applying an Adaptive Extended Kalman Filter to the TDOA value and the TDOA value. 6. The method of claim 5,
The location recognition signal is a UWB (Ultra WideBand) band signal, the wireless positioning system, characterized in that transmitted to the plurality of signal receivers in the form of a packet.

Images