JP4997637B2 - Position estimation system and program - Google Patents

Description

  The present invention relates to a position estimation system and program for estimating the position of a target using a target wave generation source and a plurality of nodes having a wave reception function, and in particular, an intersection of distance circles is unnecessary, and a sensor node The present invention relates to a position estimation system and program with few restrictions on positions and numbers and search areas.

  BFA (Believable Factor Algorithm) has been proposed as a target position estimation algorithm in an environment outside the line-of-sight (see, for example, Non-Patent Document 1). This non-line-of-sight (NLOS) environment is different from the line-of-sight (LOS) environment where the signal emitted by the target arrives as a direct wave. It refers to an environment where there are shielding objects between them and waves do not reach directly, but reach as reflected waves or diffracted waves.

FIG. 9 is a diagram for explaining a conventional position estimation algorithm. In the conventional proposed position estimation algorithm BFA, first, at the three sensor nodes N1, N2, and N3, the time at which a signal emitted from a target is received at a known time is measured, and the time difference (TOA) between the known time and the received time is measured. : Each time circle is drawn based on (Time Of Arrival), and intersections A, B, and C are derived. Assuming that the target exists in the region surrounded by the intersections A, B, C, the cost function f (X) is
f (X) = (1- αA) ‖X-XA‖ 2 + (1-αB) ‖X-XB‖ 2 + (1-αC) ‖X-XC‖ 2
Where X: Target position coordinates
Xi: Position coordinates of intersections A, B, C
.alpha.i: Distance ‖X-Xi‖ ² related to reliability BF (Believable Factor)
Where αi is
αi = 1− | rNLOSi−di | / di,
(| RNLOSi−di | <<<< rLOSi-di | & di ≧ rNLOSi)
1- | rNLOSi-di | / rNLOSi,
(| RNLOSi-di | <<<< rLOSi-di | & di <rNLOSi)
1- | rLOSi-di | / di,
(| RLOSi-di | <<<< rNLOSi-di | & di≥rLOSi)
1- | rLOSi-di | / rLOSi,
(| RLOSi-di | <<<< rNLOSi-di | & di << rLOSi)
However, rLOSi, rNLOSi: Estimated distance from target to sensor node Ni in LOS, NLOS environment based on received signal strength RSS (Received Signal Strength) at sensor node Ni
di: As an estimated distance from the target based on TOA to the sensor node Ni, a coordinate X at which the cost function f (X) is minimum is obtained.
H. Yan, H. Han-ying, and Z. Shan, "A TOA based believable factor mobile location algorithm," IEEEWCNC pp. 260-263, 2004.

  The above-described position estimation system has restrictions on the position and number of sensor nodes and the search area, such as the position cannot be estimated when there is no intersection of distance circles, or the number of sensor nodes needs to be three. There is a disadvantage that it is not suitable for position estimation of a target in a sensor network.

  In view of the above problems, an object of the present invention is to provide a position estimation system and program that do not require intersections of distance circles and have few restrictions on the positions and number of sensor nodes and search areas.

The position estimation system of the present invention receives a wave emitted at a known time by a target whose position is to be estimated, and measures three or more nodes that measure the reception time and received signal strength. Target coordinates are estimated based on an estimated distance from the node to the target (hereinafter referred to as “TOA estimated distance”) calculated from a time difference between the measured reception time and the known time (hereinafter, this estimated point). (Referred to as “TOA temporary estimation point”.) Target coordinates are estimated based on an estimated distance from the node to the target calculated from the first position estimation means and the received signal strength measured by each of the nodes ( Hereinafter, this estimated point is referred to as “RSS temporary estimated point.”) For each node, the distance di to the TOA temporary estimated point for each node A reliability calculation means for calculating the reliability BF of the node from the distance ri to the RSS temporary estimation point by the following formula , and each TOA estimated distance is weighted and corrected by the reliability BF calculated by the reliability calculation means. And a third position estimating means for estimating the target coordinates based on the TOA estimated distance corrected by the weight correcting means.
BF = 1- | ri-di | / di, (di≥ri)
1- | ri-di | / ri, (di <ri)

  The first, second and third position estimating means calculate a root mean square of the difference between the estimated distance from the node to the target and the actual distance from the node to each coordinate at each coordinate. The target position can be estimated with high accuracy by estimating the coordinate having the minimum root mean square or the average value of the plurality of coordinates as the target coordinates when there are a plurality of the minimum coordinates. Can do.

  Further, the second position estimation means calculates a plurality of estimated distances for each node based on the received signal strength and a plurality of distance attenuation constant candidates in a predetermined range, so that the distance attenuation constant is unknown. Even if it exists, the position of the target can be estimated with high accuracy.

Further, the position estimation system of the present invention receives a wave emitted at a known time from three or more wave generation sources at a target whose position is to be estimated, and measures the reception time and received signal strength. And target coordinates based on an estimated distance from the wave source to the target (hereinafter referred to as “TOA estimated distance”) calculated from a time difference between the reception time measured by the measurement means and the known time. (Hereinafter, this estimated point is referred to as “TOA temporary estimated point”) calculated from the first position estimating means and the received signal intensity measured by the measuring means, from the wave source to the target. Target coordinates are estimated based on the estimated distance (hereinafter, this estimated point is referred to as “RSS temporary estimated point”), and for each wave generation source, TO By the following equation from the distance ri to the distance di and RSS temporary estimated point to the temporary estimated point, the reliability calculation means calculates the reliability BF of the node, the reliability BF calculated by the reliability calculating means Weighting correcting means for weighting and correcting each TOA estimated distance, and third position estimating means for estimating a target coordinate based on the TOA estimated distance corrected by the weighting correcting means.
BF = 1- | ri-di | / di, (di≥ri)
1- | ri-di | / ri, (di <ri)

  The present invention is also a program for causing a computer to function as the system.

  According to the present invention, the position of the target can be estimated with high accuracy regardless of the line-of-sight or the non-line-of-sight environment. Furthermore, there is no need to distinguish whether the sensor node (or wave generation source) is in the line-of-sight environment or the non-line-of-sight environment, and the position of the target is estimated using all information of more than three sensor nodes. be able to.

  The best mode for carrying out the present invention will be described below in detail with reference to the accompanying drawings.

  FIG. 1 is a configuration diagram of a position estimation system according to an embodiment of the present invention. In the figure, a field F has a target 5 that emits radio waves, and four sensor nodes 1 to 4 that receive the radio waves are arranged. The server 6 includes a communication function 61, a calculation function 62, and a position estimation function 63.

  The positions of the sensor nodes 1 to 4 in the field F are stored in advance in the calculation function 62 of the server 6. The sensor nodes 1 to 4 have a function of measuring the reception time and received signal strength of the radio wave emitted by the target 5 at a known time, and the measured reception time and received signal strength are respectively transmitted to the communication function 61 of the server 6. The calculation function 62 is notified through this.

  FIG. 2 is a flowchart showing functional operations of the calculation function 62 and the position estimation function 63 of the server 6. The functional operations of the calculation function 62 and the position estimation function 63 of the embodiment will be described using the flowchart of FIG.

  In the “calculate temporary estimated point Xt by TOA” step S1, the calculation function 62 receives the TOA from the sensor nodes 1 to 4 via the communication function 61 and estimates the distance to the target (hereinafter referred to as “TOA estimated distance”). Calculate The position estimation function 63 estimates target coordinates as temporary estimated points Xt based on the TOA estimated distance.

  FIG. 3 is a diagram for explaining a method for estimating the target coordinates based on the estimated distance. Here, a method for estimating the position based on the TOA estimated distance will be described. The positions of the four sensor nodes N1 to N4 are known, and the time when radio waves are emitted from the target T is also known. The estimated TOA distance is represented by a black line, the actual distance from each grid point G to the sensor nodes N1 to N4 is represented by a hatching line, and the difference between the two distances is represented by a white line. At this time, using the difference in distance as an error, a lattice point G having a minimum root mean square error (RMSE) is obtained as a temporary estimated point where the target exists. Further, when there are a plurality of grid points G at which the RMSE is minimum, an average value of these coordinates is obtained as a temporary estimated point. The target position can be similarly estimated based on the estimated distance by RSS and the estimated distance by modified TOA, which will be described later.

  Next, in the “calculate temporary estimated point Xr by RSS” step S2 in FIG. 2, the calculation function 62 receives the RSS from the sensor nodes 1 to 4 via the communication function 61 and estimates the distance to the target (hereinafter “RSS estimation”). Calculate the distance. At that time, assuming that the distance attenuation constant of the radio wave intensity is in a predetermined range (for example, 2.0 to 4.0 power), the distance attenuation constant is changed in a predetermined step (for example, every 0.2), and a plurality of RSSs are obtained. Find the estimated distance. The position estimation function 63 estimates target coordinates as temporary estimated points Xr based on the plurality of RSS estimated distances in a predetermined range. As this estimation method, the method described with reference to FIG. 3 can be used.

In "Calculate BF from Xt, Xr" step S3, the position estimation function 63 calculates the reliability (BF) αi.
αi = 1− | ri−di | / di, (di ≧ ri)
1- | ri-di | / ri, (di <ri)
Ri: RSS estimated distance based on RSS and set distance attenuation constant
d i: Estimated distance of TOA Accordingly, sensor nodes having greatly different ri and di have small reliability α i, and sensor nodes having approximate ri and di have large reliability α i and approximate one.

  In the “weighting correction of TOA with BF” step S4, the position estimation function 63 corrects the TOA estimated distance by weighting with αi. As a result, the error due to the NLOS environment can be reduced by correcting and shortening the estimated TOA distance by the sensor node having a low reliability αi.

  In “calculate final estimated point X by modified TOA” step S5, the position estimation function 63 estimates target coordinates as final estimated point X based on the weighted corrected TOA estimated distance. As this estimation method, the method described with reference to FIG. 3 can be used.

  FIG. 4 is a diagram illustrating a method for estimating target coordinates based on the corrected TOA estimated distance. By estimating the coordinates of the target T as the estimated position X by the distance circle (solid line) based on the corrected TOA estimated distance obtained by weighting and correcting the distance circle (broken line) based on the TOA estimated distance, accurate position estimation can be performed.

5-8 is a figure which shows the result of having simulated the effect of this invention. The horizontal axis indicates an estimation error (m) between the actual target position and the estimated position, and the vertical axis indicates a cumulative distribution function (CDF) of the estimation error. The characteristic on the left is a good characteristic with few errors.
[Simulation specifications]
Field: 13.0 × 14.0m
Number of targets: 1 (random placement)
Number of sensor nodes: 5, 7 (random arrangement)
Distance attenuation constant: LOS: 2.6, NLOS: 3.8
Measurement RSS in unit distance: 0dBm
TOA NLOS uniform distribution noise: U (0, 1.60 × 10 ^-8 )
TOA normal distribution noise: N (0, (1.70) ² )
RSS normal distribution noise: N (0, (6.1 × 10 ^-9 ) ² )
Field grid step size: 1.0m
Distance attenuation constant step amount: 0.2
[Graph]
TOA: A position estimation method using only the arrival time RSS: A position estimation method using only the received signal strength, and a method in which the distance attenuation constant is estimated in each sensor Invention: The method of the present invention Ideal: The distance attenuation constant in the present invention When given as known BFA: A method in which three sensor nodes are used and the distance attenuation constant is known in the line-of-sight environment and in the line-of-sight environment. FIG. 5 shows two sensor nodes in the line-of-sight environment and sensors in the line-of-sight environment. If there are 3 nodes,
FIG. 6 shows that when there are five sensor nodes in the outside environment,
FIG. 7 shows that when there are three sensor nodes in the line-of-sight environment and four sensor nodes in the non-line-of-sight environment,
FIG. 8 shows the characteristics when there are five sensor nodes in the line-of-sight environment and two sensor nodes in the non-line-of-sight environment.

  From the figure, it can be seen that the algorithm of the present invention improves the position estimation accuracy compared to the conventional algorithm. This is presumably because the error due to the non-line-of-sight environment was reduced by weighting and correcting the TOA estimated distance in the present invention. The algorithm of the present invention can use three or more sensor nodes for target position estimation. Therefore, the algorithm of the present invention is considered to have improved position estimation accuracy compared to BFA. Also, from the figure, BFA has the worst position estimation accuracy among all methods. This is because the BFA uses only three sensor nodes for position estimation, but considering the average of errors, it is considered that the number of sensor nodes is desired to be as large as possible.

  In addition, this invention is not limited to the said Example.

  In the configuration diagram of the distance estimation system in FIG. 1, the example in which the number of sensor nodes is four, that is, sensor nodes 1 to 4, has been described. It goes without saying that the position can be estimated with

  In the above-described example, the target 5 emits radio waves and the sensor nodes 1 to 4 convert the distance using the intensity of the received radio waves. However, the target 5 need not be limited to radio waves. The sensor nodes 1 to 4 may receive a sound wave and convert it by distance conversion.

  Moreover, it can also be set as the structure which reverses a wave generation source and a sensor node, and estimates a position within a target.

  Note that the position estimation system of the present invention is also realized by a program for causing a computer to function as the position estimation system. This program may be stored in a computer-readable recording medium.

  The recording medium on which the program is recorded may be the ROM of the server 6 shown in FIG. 1, or a program reading device such as a CD-ROM drive is provided as an external storage device. It may be a CD-ROM or the like that can be read by insertion.

  The recording medium may be a magnetic tape, a cassette tape, a flexible disk, a hard disk, an MO / MD / DVD, or a semiconductor memory.

It is a block diagram of the position estimation system by one Example of this invention. It is a flowchart which shows the functional operation | movement of the calculation function of a server, and a position estimation function. It is a figure explaining the method of estimating a target coordinate based on an estimated distance. It is a figure explaining the method of estimating a target coordinate based on correction TOA estimation distance. It is a figure (the 1) which shows the result of having simulated the effect of this invention. It is FIG. (2) which shows the result of having simulated the effect of this invention. It is FIG. (The 3) which shows the result of having simulated the effect of this invention. It is FIG. (4) which shows the result of having simulated the effect of this invention. It is a figure explaining the conventional position estimation algorithm.

Explanation of symbols

1-4 Sensor node 5 Target 6 Server 61 Communication function 62 Calculation function 63 Position estimation function

Claims (7)

Three or more nodes that receive a wave emitted at a known time by a target whose position is to be estimated and measure its reception time and received signal strength;
Target coordinates are estimated based on an estimated distance (hereinafter referred to as “TOA estimated distance”) from the node to the target, which is calculated from a time difference between the reception time measured by each node and the known time (hereinafter referred to as “TOA estimated distance”). The estimated point is referred to as “TOA temporary estimated point”.) First position estimating means;
Target coordinates are estimated based on the estimated distance from the node to the target calculated from the received signal strength measured by each node (hereinafter, this estimated point is referred to as “RSS temporary estimated point”). Position estimation means;
Reliability calculation means for calculating the reliability BF of the node from the distance di to the TOA temporary estimation point and the distance ri to the RSS temporary estimation point for each node according to the following equation:
BF = 1- | ri-di | / di, (di≥ri)
1- | ri-di | / ri, (di <ri)
Weighting correction means for weighting and correcting each TOA estimated distance by the reliability BF calculated by the reliability calculation means;
A position estimation system comprising: third position estimation means for estimating target coordinates based on the TOA estimated distance corrected by the weight correction means.   The first, second and third position estimating means calculate the root mean square of the difference between the estimated distance from the node to the target and the actual distance from the node to each coordinate at each coordinate, The position estimation system according to claim 1, wherein when there are a plurality of coordinates having a minimum root mean square or a plurality of the minimum coordinates, an average value of the plurality of coordinates is estimated as a target coordinate.   3. The second position estimation unit calculates a plurality of estimated distances for each node based on the received signal strength and a plurality of distance attenuation constant candidates in a predetermined range. Position estimation system. Measuring means for receiving a wave emitted at a known time from three or more wave generation sources in a target whose position is to be estimated, and measuring the reception time and the received signal strength;
Target coordinates are estimated based on an estimated distance from the wave generation source to the target (hereinafter referred to as “TOA estimated distance”) calculated from the time difference between the reception time measured by the measuring means and the known time. (Hereinafter, this estimated point is referred to as a “TOA temporary estimated point”.) First position estimating means;
Target coordinates are estimated based on the estimated distance from the wave source to the target calculated from the received signal strength measured by the measuring means (hereinafter, this estimated point is referred to as “RSS temporary estimated point”). Second position estimating means;
Reliability calculation means for calculating the reliability BF of the node from the distance di to the TOA temporary estimation point and the distance ri to the RSS temporary estimation point by the following equation for each wave generation source;
BF = 1- | ri-di | / di, (di≥ri)
1- | ri-di | / ri, (di <ri)
Weighting correction means for weighting and correcting each TOA estimated distance by the reliability BF calculated by the reliability calculation means;
A position estimation system comprising: third position estimation means for estimating target coordinates based on the TOA estimated distance corrected by the weight correction means.   The first, second and third position estimating means calculate the root mean square of the difference between the estimated distance from the wave generating source to the target and the actual distance from the wave generating source to each coordinate at each coordinate. 5. The position according to claim 4, wherein when there are a plurality of coordinates having the minimum root mean square or a plurality of the minimum coordinates, an average value of the plurality of coordinates is estimated as a target coordinate. Estimation system.   5. The second position estimation unit calculates a plurality of estimated distances for each wave generation source based on the received signal strength and a plurality of distance attenuation constant candidates in a predetermined range. 5. The position estimation system according to 5.   The program for functioning a computer as a system in any one of Claims 1 thru | or 6.

Images