JP4308779B2 - Signal identification method and lightning location method - Google Patents

Description

  The present invention relates to a method of collating which recorded signal at each observation point is the same signal observation when recording signals at a plurality of observation points, and in particular using the collation. The present invention relates to a signal identification method for locating signal sources.

Iwanami Lecture Information Science 21 “Pattern Recognition and Graphic Processing” (Makoto Nagao) Chapter 2 Section 4 “Speech Recognition” pp.71-89

  In general, signal localization at a plurality of points is required for position determination of an unspecified source. Specific application examples include lightning location determination by aerodynamic observation and epicenter identification by seismic observation.

  When searching for the same signal, after setting an appropriate search space, see if the feature quantity that does not depend on the observation point of the signal, for example, the waveform, duration, signal strength that does not decay, etc., match similar signals. I was exploring. If the signal is transmitted at a high speed, the reception time may also be seen. In that case, time synchronization was performed between observation points using a large scale device such as GPS.

  However, if multiple signals are received in a short time, multiple signals with the same characteristics may be received. If the search space is too large, the possibility of misjudgment increases, and it is too small as necessary. The possibility that the same signal cannot be found increases. Therefore, the optimal size of the search space had to be determined well.

  On the other hand, when judging by the reception time, unless the signal reception interval is sufficiently sparse, it is required that the transmission speed is extremely fast like light and the time synchronization accuracy between observation points is extremely high. There is a problem that the range of possible fields is narrow and the cost becomes very high in order to realize such time synchronization. Moreover, even if it is such a very fast signal, it is not an essential solution because there will always be some error in the reception time depending on the distance between the source and the observation point. In any case, it was difficult to limit the reception time, and it was necessary to set a search space with a certain width.

  The present invention has been made in view of the above problems of the prior art, and an object of the present invention is to provide a signal identification method that can easily set an optimal search space that is not easily affected by time errors.

In the signal identification method of the present invention, signals intermittently transmitted from a plurality of unspecified signal transmission sources are observed at two points X and Y, and the target signal A in the observation record of the point X is the point. In the method of identifying a candidate signal that may be the same as the target signal A in the observation record of Y, the candidate of the point Y around the same time when the target signal was detected at the point X The search range is a signal search range, and the search space is set to twice the quotient of the distance between the point X and the point Y and the signal transmission speed, and a plurality of candidate signals are detected according to the search result. One of the candidate signals detected in this case is set as a new target signal B, and at the point where the target signal B is observed, the point of the other point is centered on the same time when the target signal B is detected. The candidate signal for the target signal B The search range is set as a new search space, and the search is performed again. If there is only one candidate signal having the same possibility as the target signal A, or the number of candidate signals in the observation record at each observation point The re-searching is repeated until the wrinkles of are met . (Claim 1)

  Further, the signal identification method of the present invention has means for determining whether the source of the target signal is closer to the point X or the point Y, and in the case where it is possible to determine which is closer, the target When the signal is close to the point X, the search space for the target signal is limited after the reception time at the point X, and within the range of the distance between the point X and the point Y and the signal transmission speed quotient. When a search space is set and the target signal is close to the point Y, the search space for the target signal is limited to a time before the reception time at the point X, and the distance between the point X and the point Y In the case where the search space is set within the range of the quotient of the signal transmission speed and it is not possible to determine which one is close to the target signal, one of the candidate signals found on the search space is assumed to be the same. The source of the target signal is Re-determining whether it is closer to X or the point Y, and if there is a contradiction that the target signal and the candidate signal are the same, removing one of the candidate signals from the same signal candidate of the target signal Features. (Claim 2)

  Furthermore, the signal identification method of the present invention is characterized in that a maximum deviation width of time synchronization accuracy at the point X and the point Y is added to the search space. (Claim 3)

  The lightning location evaluation method of the present invention is a method for determining the location of lightning by performing aerial observations at a plurality of points, and identifying observation signals between observation points using the above signal identification method. It is characterized by. (Claim 5)

According to the first aspect of the present invention, the maximum deviation time of the reception time that can occur at each observation point is calculated from the signal transmission speed and the distance between the observation points, and the search space is set to the time before and after the deviation time from the reception time. The risk of misjudgment is eliminated by limiting. It should be noted that the quotient of “distance between X and Y” and “signal transmission speed”, that is, that the signal transmission time from X to Y has the maximum deviation width can be shown by simple geometric proof. Is possible.
Furthermore, when multiple candidate signals are found in the search space and the same signal cannot be identified, the same search is performed recursively in each candidate signal in order to identify the same signal, and each observation point In many cases, signal candidates that have the same possibility as the first signal are combined into one by repeating until the number of candidate signals in the observation record in 合 う matches or until the candidate signal is determined by the formula You can narrow down.

  The invention according to claim 2 is provided with means for determining a relative positional relationship with the observation point of the signal transmission source, so that it can be understood which observation point the signal reaches first. The risk of misjudgment is further reduced by setting the search space set in claim 1 as the center at one-half, or by providing means for removing a signal that contradicts the reception time from the candidates.

  According to the third aspect of the present invention, an erroneous determination or a search omission due to a synchronization shift is achieved by adding a time synchronization accuracy shift width to the range of the search space set in the first and second inventions described above. The time synchronization accuracy requested can be lowered to some extent.

  As an example of how to determine which observation point is closer to the source, look at the direction of the signal at the time of observation, and if you are coming from the opposite direction to the other observation point, determine that you are closer to yourself. If the signal strength is attenuated according to the distance, it can be judged by looking at the signal strength at each observation point. Even in the case of the former method, if a signal is coming from the other observation point, check the direction of the signal at the other observation point and receive it from the opposite direction to the first observation point. Then you can see that it is far from the first observation point.

According to the fourth aspect of the present invention, accurate lightning position locating can be realized without an expensive time synchronizer by performing electromagnetic wave identification by the above-described method as preprocessing for lightning locating.

FIG. 1 is a flowchart showing an example of the signal identification method of the present invention.
First, the basic time width of the search space is determined in the basic search space setting. Here, the search space is a time range for searching for the same candidate signal at a point with a different target signal to be identified. In the present invention, the basic time width is a quotient of the distance between observation points and the signal transmission speed. 2 times the maximum time synchronization error between observation points.

  Next, the target signal (here, the target signal A first) is determined, and then the search space is determined based on the characteristics of the target signal. For example, the search space is limited before or after the reception time of the target signal depending on the positional relationship between the signal transmission source and each observation point. If there is no such feature, the basic time width becomes the time width of the search space as it is. The center of the search space where no limitation is made is the reception time of the target signal.

  Next, candidate signals having the same invariant features as the target signal are searched, and candidate signals having contradictions in reception time are excluded as a result of comparison on the assumption that the target signal and each candidate signal are the same.

  If a candidate signal B having the same invariant characteristics as the target signal A is found as a result of the search, the search is performed in a search space where the candidate signal B is the target signal B. As a result of the search, when the target signal B and the candidate signal C form a one-to-one confirmed pair, if the candidate signal C is the target signal A, the pair of the candidate signal B and the target signal A is confirmed and the processing ends. .

  If the candidate signal C is not a pair with the target signal A, the confirmed pair is deleted, and it is confirmed whether there is a “determined pair”.

  Here, the “determined pair” refers to a set of a target signal and a candidate signal for which only one candidate signal having the same invariant characteristics as the target signal can be confirmed as a result of the search.

  When there are no confirmed pairs, it is checked whether there is an unsearched signal, that is, a signal that has been found by the search but has not been searched by itself, and if it remains, search again as the target signal. If it does not remain, the identification is impossible and the processing ends.

  Hereinafter, the identification of electromagnetic waves in static observation for lightning location will be described as an example.

  2 to 6 are diagrams showing an example of a process for identifying static electricity caused by the same lightning as the static electricity signal A on the X and Y observation records along the time series. Represents the search space from the signal at the apex. Further, the range of triangles having the X-side signal as a vertex is a search from X to Y, and the range of triangles having the Y-side signal as a vertex is a search from Y to X. The length of the base of the triangle represents the time width of the search space. This basic time width is twice the quotient of “distance between X and Y” and “signal transmission speed”. Here, if the distance between X and Y is 300 km and the signal speed is the speed of light (300,000 km / sec), the basic time width is 0.001 × 2 = 0.002 seconds. The time synchronization error was ± 0.0005 seconds, and the search space time width was 0.002+ (0.005 × 2) = 0.003 seconds.

FIG. 2 shows a first embodiment of the present invention.
The signal A is the target signal, and the search space remains the basic search space. As a result of the search, only signal B is a candidate, so that it becomes a confirmed pair, and it is determined that signal B is the same as signal A.

FIG. 3 shows a second embodiment of the present invention.
In this example, there is provided means capable of recognizing in 180 degrees whether the static reception direction at the observation point is the X direction or the Y direction. When the receiving direction of the signal A was seen at the point X, it was a direction opposite to the point Y, so it was clearly determined that the lightning point was closer to the point X than the point Y.

  Therefore, the search space can be limited after the reception time at the point X where the signal A is observed. However, it is allowed up to 0.0005 seconds before the time synchronization error from the reception time.

  As a result, two candidate signals B and C were found, but since the signal C was received from the opposite direction to the point X, the lightning point was clearly a static that was closer to the point Y than the point X. Can be removed from candidates due to time conflicts. Thereby, it is determined that the signal B is the same as the signal A.

FIG. 4 shows a third embodiment of the present invention.
In this example, the search space remains the basic search space, and the intensity that attenuates according to the transmission distance is observed at the observation point.

As a result of the search, three candidate signals B, C, and D were found.
Here, assuming that the signal A and the signal B are the same, since the signal intensity is stronger in the signal A, the lightning point is determined to be closer to the point X. If the signal A and the signal C are the same, the lightning point is Assuming that the signal A and the signal D are the same, it is determined that the lightning point is close to the point X.

  However, even if the signal B and the signal D allow 0.0005 seconds, which is a time synchronization error, from the reception time of the signal A at the point X, the signal B is received before the signal A and the signal D is received after the reception time of the signal A, respectively. Has been. Therefore, the signal C that does not cause a contradiction remains, so that the signal C is determined to be the same as the signal A.

FIG. 5 shows a fourth embodiment of the present invention.
In this example, the search space remains the basic search space.

  As a result of the search, two candidate signals B and C were found, but there is no determinant which candidate signal is the same as the signal A. Therefore, when searching from the candidate signals B and C toward the point X in the reverse direction, no new candidate signal was found from the signal B, and a new signal D was found from the signal C. Therefore, the signal B and the signal A are a definite pair, and the signal B is determined to be the same as the signal A.

FIG. 6 shows a fifth embodiment of the present invention.
In this example, it is assumed that the search space remains the basic search space, and the reception direction of the static electricity at the observation point can be recognized in units of 180 degrees in the X direction or the Y direction.

  Here, the signal A was observed from the Y direction, and candidate signals B, C, and D were found as a result of the search. Signal B was observed from the Y direction, and signals C and D were observed from the X direction. Since there is no contradiction at this point, there is no determinant which is the same as A. Therefore, when searching in the reverse direction from each candidate signal B, C, D to the point X, a new signal E is obtained from the signal B, and new signals F, G, H are obtained from the signals C, D. found. Here, the signal E is received from the Y direction, and assuming that the signal B and the signal E are the same, the lightning point is closer to the point Y than the point X. The signal E is earlier than the point Y at the point X. A contradiction arises with what is being observed. Therefore, the signal E is excluded from the candidates, the signal B and the signal A are in a definite pair, and the signal B is determined to be the same as the signal A.

  Note that the combination of the signals C and D and the signals F, G and H is unknown at this point, but does not need to be pursued since the original purpose has been achieved.

FIG. 7 shows a sixth embodiment of the present invention.
In this example, the search space remains the basic search space.

  As a result of the search, candidate signals B, C, and D were found, but there is no determinant which is the same as signal A. Then, when searching in the reverse direction from the signals B, C, D to the point X, the signals E, F, G are newly found, but here also a confirmed pair cannot be made.

  Therefore, when searching from each signal E, F, G toward the point Y, a new signal H was found from the signal E, and nothing was found from other signals.

  At this time, since there is no candidate other than the signal D from the signal G, it becomes a confirmed pair, and subsequently, the signal C and the signal F also become a confirmed pair. Then, since signals B, C, and D, which were candidates for signal A, are determined as another pair and the pair has been deleted, it is determined that signal B is the same as signal A.

  Next, a seventh embodiment of the lightning location method of the present invention will be described.

  Until the identification of each signal by static electricity was the same as in Example 6, it was confirmed that static electricity A and static electricity B were electromagnetic waves from the same lightning.

  The representative location method includes the meeting method by two-point observation and the time difference method by three-point observation. Here, the meeting method is used on the assumption that the direction of the signal can be confirmed at each observation point.

  Since the receiving directions of the static electricity A and the static electricity B are known, if the intersection point when the straight line is extended from the point X and the point Y to the direction is obtained, it can be known as the lightning point. In the case of using the time difference method, a point Z is provided, and after signal identification according to the present invention is performed at the point X and the point Z or at the point Y and the point Z, each signal is determined from the difference in arrival time between the observation points. Calculate the distance between the observation point and the lightning point. Thereby, a thunderstorm point can be specified.

  As described above, according to the present invention, it is possible to accurately perform signal identification even if there is a slight difference in reception time or time synchronization error. It can be realized accurately at low cost without an apparatus. In addition to lightning location localization, the present invention can also be applied to seismic source localization by seismic wave observation, detection of abnormalities in a living body by nerve pulse observation, and grasping of space phenomena by cosmic ray observation.

It is a flowchart which shows an example of the signal identification method of this invention. It is a figure which shows Example 1 of this invention. It is a figure which shows Example 2 of this invention. It is a figure which shows Example 3 of this invention. It is a figure which shows Example 4 of this invention. It is a figure which shows Example 5 of this invention. It is a figure which shows Example 6 of this invention.

Claims (4)

Signals transmitted intermittently from a plurality of unspecified signal transmission sources are observed at two points X and Y, and the target signal A in the observation record of the point X is the target in the observation record of the point Y In a method for identifying a candidate signal that may be identical to signal A ,
The search space is a range in which the candidate signal at the point Y is searched around the same time at which the target signal is detected at the point X.
The search space is set to twice the quotient of the distance between the point X and the point Y and the signal transmission speed ,
One of the candidate signals detected when a plurality of the candidate signals are detected as a result of the search is set as a new target signal B;
At a point where the target signal B is observed, a range for searching the candidate signal for the target signal B at the other point is set as a new search space around the same time when the target signal B is detected. Do a re-search,
The signal identification method characterized by repeating the re-search until there is only one candidate signal having the same possibility as the target signal A or until the number of candidate signals in observation records at each observation point matches. . Means for determining whether the source of the target signal is closer to the point X or the point Y;
When the target signal is close to the point X in the case where the target signal is close to the point X, the search space for the target signal is limited after the reception time at the point X, and the point X and the point Y And set the search space in the range of the distance and the quotient of the signal transmission speed,
When the target signal is close to the point Y, the search space for the target signal is limited to a point before the reception time at the point X, and the quotient of the distance between the point X and the point Y and the signal transmission speed is obtained. Set the search space by range,
When it is not possible to determine which one is close, it is assumed that the target signal and one of the candidate signals found in the search space are the same, and the source of the target signal is the point X or the point Y. Re-determine whether it is close to
2. The signal identification method according to claim 1, wherein when there is a contradiction that the target signal and the candidate signal are identical, one of the candidate signals is excluded from the same signal candidate of the target signal. The signal identification method according to claim 1, wherein a maximum deviation width of time synchronization accuracy at the point X and the point Y is added to the search space. 4. A method for determining the position of a lightning by performing aerial observation at a plurality of points, wherein the observation signal between observation points is identified using the signal identification method according to any one of claims 1 to 3. A lightning location method characterized by that.

Images