JP2616730B2 - Sound source motion analyzer - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention is used for underwater acoustic sounding. The present invention is used for measuring the speed, course, and position of an object moving in water. In particular, the present invention relates to a technique of performing measurement by one actual measurement.

[0002]

2. Description of the Related Art As disclosed in Japanese Patent Application Laid-Open No. 2-66481, a conventionally known sound source motion analysis apparatus is disclosed in Japanese Patent Application Laid-Open No. 2-66481 in order to analyze the position, course and speed of a sound source moving in water. The sound source signal is measured three or more times in a time series with the receiver, and the measured values are used to solve simultaneous equations including the course, speed, and relative orientation of the sound source and each receiver as unknowns. Technology was used.

[0003]

In the conventional motion analysis device, it is necessary to establish a simultaneous equation of an order higher than the unknown in order to obtain a plurality of unknowns, and three or more measurements are required in time series.

[0004] Also, in this equation, during the measurement,
Since it is assumed that the sound source is performing a constant velocity linear motion, it is difficult to analyze the motion of the sound source whose speed or course is changing.

[0005] The present invention has been made in such a background, and the speed of the sound source, the course,
It is an object of the present invention to provide a sound source motion analysis device capable of measuring a position. SUMMARY OF THE INVENTION It is an object of the present invention to provide a sound source motion analysis device capable of analyzing a sound source even when the sound source is moving other than at a constant velocity linear motion.

[0006]

SUMMARY OF THE INVENTION The present invention is characterized in that the position, course, and speed of a sound source moving in water can be analyzed in a time series in a single measurement of a sound source signal. The motion analysis apparatus of the present invention does not use a technique of establishing and analyzing a simultaneous equation of an order equal to or greater than the unknown number to obtain the unknown number, and calculates the unknown number by calculating the error between the measured value and the measured value of the replica obtained assuming the unknown number. Estimation technology is used. That is,
The present invention provides a receiver arranged at a plurality of positions for receiving sound waves arriving from a single sound source in water, and means for analyzing the position of the sound source according to the Doppler frequency observed via the receiver. This is a sound source motion analysis device provided with:

Here, it is a feature of the present invention that the analyzing means includes a parameter setting section for setting a parameter value corresponding to the unknown using a random number, and a Doppler for inputting the set random value. A replica calculator for calculating a replica Doppler frequency of the frequency, an error calculator for inputting the Doppler frequency actually measured by the receiver and the replica Doppler frequency and calculating an error thereof, and comparing the error with a threshold. A calculation end determination unit that determines the end of the calculation by using the error and a previously output error from the error calculation unit, and when the difference is equal to or less than an allowable value, re-inputs the parameter value to the parameter setting unit. An error change evaluator and, when the difference is equal to or more than an allowable value, an output of the error change evaluator is input, and an increase in the difference is compared with an allowable value, and the result is allowed. An error re-evaluation unit that re-inputs the value of the parameter to the parameter setting unit when the value is equal to or less than a value, and a parameter retreat unit that re-inputs the value of the previous parameter to the parameter setting unit when the increment is equal to or more than an allowable value. The analyzing means updates the parameter corresponding to the unknown according to the value of the parameter re-input to the parameter setting unit and a new random number value until the calculation completion determination unit determines the calculation completion. To analyze repeatedly.

[0008] A three-way receiver, the course A of the sound source
h, speed S, oscillation frequency F, course Ah of the receiver
The relative azimuths with respect to are respectively A1, A2, A3, the underwater sound speed is sw, and the replica Doppler frequencies are F1, F
Assuming that 2, F3, F1 = [sw / (sw + S · cos (A1-Ah))]
× F F2 = [sw / (sw + S · cos (A2-Ah))]
× F F3 = [sw / (sw + S · cos (A3-Ah))]
× F 2 is desirable.

It is preferable that the parameter setting unit updates the parameter according to a parameter set update value = 1 set value (or an initial value) before a loop ± a random number calculation value.

[0010]

Function All unknown values are assumed and set by random number calculation. Calculate the measured value of the replica based on them. The error between the measured value of each receiver and the measured value of the replica is calculated. Evaluate the change in error and accept the set value if the error decreases. Further, even if the error increases, the set value is accepted if it is below the reference. Based on the accepted measured values, the set values are updated again by random numbers, and the process is repeated until the error becomes extremely small. As a result, the optimal values of all unknowns that minimize the error from the actually measured values are estimated.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The construction of an embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a block diagram of an apparatus according to an embodiment of the present invention. FIG. 2 is a diagram showing a positional relationship between a sound source and a receiver according to the embodiment of the present invention.

According to the present invention, there are provided receivers 11 arranged at three positions for receiving sound waves coming from one sound source 20 in water.
This is a sound source motion analysis device including: a first analysis unit 13 for analyzing a position of a sound source 20 according to Doppler frequencies f1, f2, and f3 observed via the receivers 11 to 13.

Here, the feature of the present invention is that the analyzing unit 1 includes parameters F, Ah, S,
A parameter setting unit 2 for setting the values of A1, A2, and A3 using random numbers, and a replica Doppler frequency F of the Doppler frequencies f1, f2, and f3 by inputting the set random numbers.
1, F2, F3, and Doppler frequencies f1, f actually measured by the receivers 11 to 13.
2, f3 and the replica Doppler frequencies F1, F2, F
3, an error calculation unit 4 for calculating the error, a calculation end determination unit 5 for comparing the error with a threshold value to determine the end of the calculation, and an error previously output from the error calculation unit 4 When the difference is equal to or less than the allowable value, the error change evaluator 6 re-inputs the values of the parameters F, Ah, S, A1, A2, and A3 to the parameter setting unit 2, and when the difference is equal to or larger than the allowable value. The output of the error change evaluator 6 is input and the increase in the difference is compared with an allowable value. When the result is equal to or smaller than the allowable value, the values of the parameters F, Ah, S, A1, A2, and A3 are sent to the parameter setting unit 2. An error re-evaluation unit 7 for re-input,
When the increment is equal to or larger than the allowable value, the previous parameter F ₋₁ ,
A parameter retreating unit 8 for re-inputting the values of Ah _-1 , S _-1 , A1 _-1 , A2 _-1 , and A3 _-1 to the parameter setting unit 2;
The analysis unit 1 determines whether the parameters F, Ah, S, A1, A2, A3 or F ₋₁ , Ah ₋₁ , which are re-input to the parameter setting unit 2 until the calculation end determination unit 5 determines the calculation end.
The parameter corresponding to the unknown is updated according to the values of S ₋₁ , A 1 ₋₁ , A 2 ₋₁ , and A 3 ₋₁ and the new random number, and the analysis is repeated.

Next, the operation of the embodiment of the present invention will be described. A sound source moving underwater while radiating an acoustic signal of frequency f at course ah and speed s is observed by three receivers 11 to 13. The relative azimuth between the receiver 11 and the sound source is a1, the measured Doppler frequency measured by the receiver 11 is f1, the relative azimuth between the receiver 12 and the sound source is a2, and the measured Doppler frequency measured by the receiver 12 is f2, the relative azimuth between the receiver 13 and the sound source is a
3. The measured Doppler frequency measured by the receiver 13 is f3
And The underwater sound speed is sw.

In FIG. 1, an analysis unit 1 of a sound source motion analysis apparatus includes replica Doppler frequencies F1, F2, F3 calculated from the values of parameters F, Ah, S, A1, A2, A3.
And Doppler frequencies f1, f2, f3 as measured data
The parameters F, Ah, S, A
This is a parameter estimator that sequentially changes the values of 1, A2, and A3 and repeatedly executes the processing.

The parameter setting unit 2 sets a set of parameters Ah, S, F, A1, A2, and A3 corresponding to the unknowns ah, s, f, a1, a2, and a3 in the motion analysis by random number calculation. In this setting, in the first setting, based on the initial value previously input by the operator as the expected value of the parameter, in the second and subsequent settings, the value of the parameter set returned from the loop by the error evaluation is set. The setting value is updated based on the setting. That is,
The updated value of the parameter set = the set value (or the initial value) one loop before ± the calculated random number, and the replica calculator 3 receives the receivers 11 to 13 based on the set value of the parameter set.
Are calculated. The calculation formula is: F1 = [sw / (sw + S · cos (A1-Ah))]
× F F2 = [sw / (sw + S · cos (A2-Ah))]
× F F3 = [sw / (sw + S · cos (A3-Ah))]
× F. The error calculator 4 calculates the replica Doppler frequency F
1 to F3 are input and the vectors (F1, F2, F
3) and the vectors (f1,...) Of the measured Doppler frequencies f1 to f3 obtained from the sound source signals measured by the receivers 11 to 13
The error σ of f2, f3) is calculated. In the iterative process, the error calculator 4 calculates the replica Doppler frequency F1,
Although F2 and F3 are input and updated each time, the first input values obtained by one measurement are used as the actually measured Doppler frequencies f1, f2, and f3. The calculation formula is σ = Σ ^k (Fk−fk) (where k = 1, 2, 3). The calculation end determination unit 5 evaluates the error σ, and
If the error σ is larger than the end reference value, the repetition processing is continued, and σ is input to the error change evaluator 6.

The error change evaluator 6 calculates a change Δσ between σ input in the previous loop and newly input σ, and when the error decreases, sets the parameter set value set by the parameter setting unit 2 to the value. Accept and input the value to the parameter setting unit 2 again. When the error increases, Δσ is input to the error re-evaluation unit 7. That is, error decrease Δσ = σ _n−1 −σ _n <0 → acceptance of set value of parameter set Error increase Δσ = σ _n−1 −σ _n ≧ 0 → Δσ input to error re-evaluation unit 7. The error re-evaluation unit 7 evaluates the increase error Δσ,
If Δσ is equal to or less than the increase allowable value, the parameter measurement unit 2
The set value of the parameter set set in the step is accepted, and the value is input to the parameter setting unit 2 again.

If the increase error Δσ is larger than the allowable increase value, the parameter retreat unit 8 returns the set value of the parameter set to the set value of the previous loop and inputs it to the parameter set unit 2. That is, Δσ ≦ increase allowable value → acceptance of parameter set value Δσ> increase allowable value → parameter reversing section 8 changes parameter set value (retreat). The parameter setting unit 2 inputs the set value of the accepted parameter set or the set value of the parameter set returned to the set value of the previous loop, and updates the parameter set by random number calculation based on this set value. Repeat the process.

When the calculation end determination unit 5 satisfies σ ≦ end reference value, the repetition processing ends, and the output unit 9 outputs the parameters F0, Ah0, S0, A10, A20, A
The set of 30 is the unknown f, ah, s, a1, a2, a
3 and are output as the three optimal parameter sets.
In the output unit 9, each receiver position, A10, A
The sound source position (X0, Y0, Z0) is calculated from 20, A30.

FIG. 3 is a diagram showing an example of a change in the error. As can be seen from FIG. 3, even if the error is increased by the error re-evaluation unit 7, if the error is smaller than the allowable increase value, the set value is changed. By accepting, it is possible to approach the overall error minimum point without staying at the local error reduction point, and to perform accurate estimation.

As described above, the measured Doppler frequencies f1, f2,... Obtained from the acoustic signals measured by the receivers 11 to 13
f3 and the unknowns in the motion analysis are represented by parameters F and A
Parameters F0, Ah0 for reducing errors by repeatedly performing error calculation of replica Doppler frequencies F1, F2, F3 calculated as h, S, A1, A2, A3 and changing parameters F, Ah, S, A1, A2, A3. , S0,
Since the motion analysis is performed by estimating A10, A20, and A30, the motion analysis can be performed only by performing the measurement of the acoustic signal once in time series. Further, since the motion and the position of the sound source are obtained from the measurement data only once in a time series, it is possible to analyze even when the sound source is moving other than the constant velocity linear motion.

[0022]

As described above, according to the present invention,
The speed, course and position of the sound source can be measured according to one actual measurement. Accordingly, analysis can be performed even when the sound source is moving by means other than the constant velocity linear motion.

[Brief description of the drawings]

FIG. 1 is a block diagram of an apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing a positional relationship between a sound source and a receiver according to the embodiment of the present invention.

FIG. 3 is a diagram showing an example of a change in an error.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 analysis unit 2 parameter setting unit 3 replica calculation unit 4 error calculation unit 5 calculation end determination unit 6 error change evaluation unit 7 error re-evaluation unit 8 parameter retreat unit 9 output unit 11 to 13 receiver 20 sound source

Claims (3)

(57) [Claims] 1. A receiver arranged at a plurality of positions for receiving sound waves arriving from one sound source in water, and analyzing the position of the sound source according to a Doppler frequency observed via the receiver. And a parameter setting unit that sets a value of a parameter corresponding to the unknown using a random number, and a replica Doppler of the Doppler frequency by inputting the set random value. A replica calculator for calculating the frequency, an error calculator for inputting the Doppler frequency actually measured by the receiver and the replica Doppler frequency and calculating the error, and comparing the error with a threshold to terminate the calculation. A calculation end determining unit for determining, and comparing this error with an error previously output from the error calculating unit, and when the difference is equal to or less than an allowable value, the parameter An error change evaluator for re-inputting a value to the parameter setting unit, and when the difference is equal to or more than an allowable value, an output of the error change evaluator is input, and an increase in the difference is compared with an allowable value, and the result is set to an allowable value. An error re-evaluation unit that re-inputs the value of the parameter to the parameter setting unit when the value is equal to or less than a value,
A parameter retreating unit for re-inputting the value of the previous parameter to the parameter setting unit when the increment is equal to or more than the allowable value, wherein the analyzing unit analyzes the parameter until the calculation end determination unit determines the calculation end. A sound source motion analysis apparatus characterized by updating a parameter corresponding to the unknown according to a value of the parameter re-input to the setting unit and a new random number, and repeatedly analyzing the parameter. 2. A receiver comprising three receivers, wherein a direction Ah, a speed S, an oscillation frequency F, and a relative orientation of the receiver with respect to the direction Ah of the sound source are A1 and A, respectively.
2, A3, the underwater sound velocity is sw, and the replica Doppler frequencies are F1, F2, and F3. F1 = [sw / (sw + S.cos (A1-Ah))]
× F F2 = [sw / (sw + S · cos (A2-Ah))]
× F F3 = [sw / (sw + S · cos (A3-Ah))]
The sound source motion analysis device according to claim 1, wherein X is F. 3. The parameter setting unit according to claim 1, wherein an update value of the parameter set = 1 a set value (or an initial value) one loop before.
2. The method according to claim 1, wherein the parameter is updated according to the calculated value of the random number.
A sound source motion analysis device as described in the above.