JPH0212081A - System for detecting underwater acoustic signal - Google Patents

Abstract

PURPOSE:To detect a weak signal in the vicinity of a strong signal by re- detecting the mean value and threshold of noises by calculation after replacing a once detected signal with the mean value of noises in the vicinity of the detected signal. CONSTITUTION:A noise averaging circuit 2 finds a 1st mean noise value of the frequency component data of underwater sounds inputted from a storage circuit 1 and a threshold circuit 3 finds a 1st threshold from the means value. A signal detection circuit 4 detects a signal by comparing the frequency component data with the 1st threshold and, when the frequency component data are lower than the 1st threshold, a data replacement circuit 5 replaces the frequency component data with the 1st means noise value and outputs the 1st mean noise value as replacing data. A noise averaging circuit 6 finds a 2nd mean noise value of the replacing data and a threshold circuit 7 finds a 2nd threshold from the mean value. Then a signal detection circuit 8 detects a signal by comparing the replacing data with the 2nd threshold.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a passive sonar device, and particularly to a method for detecting underwater acoustic signals.

[Conventional technology 1 Conventionally, this type of underwater sound! The signal detection method was as follows.

First, the noise average value MP of the frequency component data of underwater sound is determined using equation 1. Note that there are N cells of frequency components,
It is assumed that the frequency cells are arranged in ascending order.

r-chi 1r+2 Tatami, X i : level of i-th cell 1cI (i
=l-N) Mp: ratio for the p-th cell111〕(p=1 to N)W: number of cells used for noise average value calculation, p<=W/2 or p> =N-, W/2, P=W/2+2, respectively.

p=N-W/2-2 Ht¥hei Use the average value.

In addition, a total of three cells, including the cell for calculating the miscellaneous B-mo average, are excluded from the calculations.

Next, threshold 1: Ti is determined from Equation 2 based on 31 u ''p By (fi.

Ti=KIIMi @Ken・Shiki Mashashi, Ni: Constant Then, by comparing the threshold and the frequency component, anything exceeding the threshold of 7 was detected as a signal. (
See Equation 3) Signal: X i > = T i Jl signal'i: X l < T l s +
+ E-formula 3 [Problems to be solved] In the conventional detection signal detection method for two consecutive underwater acoustic signals, there was a problem when there was a weak signal in the vicinity of a strong signal.

That is, when calculating the noise average value for a weak signal B in the vicinity of a strong signal A as shown in FIG.
Since the component of the strong signal A is included in the signal, the average noise value of the apparent L increases, which in turn raises the threshold E, which sometimes makes it impossible to detect the weak signal B.

The present invention has been made in view of the above problems, and -
, i3. After replacing the detected signal with the nearby noise average value, the noise average and threshold are calculated and detected again. 2. Underwater α-sound signal that can also detect strong signals and weak signals that exist in the vicinity of the signal. The D objective is to provide a detection method.

[Means for solving the problem] -1; In order to achieve a multi-day goal, the underwater acoustic signal detection method of the present invention includes a memory for storing frequency component data of underwater sounds, and a memory for storing frequency component data input from this storage means. 1st
a first noise averaging f stage for calculating a noise average value; a first threshold means for calculating a first threshold from the first noise average value; and a first noise averaging stage for detecting a signal by comparing the frequency component data with the first threshold. a signal detection means, a data replacement means for comparing the frequency component data with a first threshold and replacing the frequency component data with a first noise average value and outputting it as replacement data when the frequency component data is greater than or equal to the first threshold; 239 tone average value means for calculating a second noise average value of data; and a second threshold 1 for calculating a second threshold from the second noise average value.
The present invention has a configuration including a second threshold and second signal detection means for detecting a signal by comparing the replacement data with a second threshold.

[Example] Next, the present invention will be explained with reference to the drawings.

FIG. 1 shows one embodiment of the present invention. 1 is a memory circuit, 2 and 6 are first and second noise averaging circuits, 3 and 7 are first and second storage average circuits, 4 and 8 are first
and a second signal detection circuit.

5 is a data replacement circuit.

Frequency component data (Xi
: 1=1-N) is temporarily stored in the storage circuit 1, and then applied to the first H tone equalization circuit 2, the first signal detection circuit 4, and the data replacement circuit 5 through the output 11 and I7. In the first noise equalization circuit 2, the first noise average value Mi (i=
l~N).

The first noise average value 21 (M
i) is sent to the threshold generation circuit 3 and the data replacement circuit 5. In the first slave 2-step circuit 3, the first threshold Ti (i=1 to N) is calculated using Equation 2 above.

The first threshold 31 (Ti) obtained by the first threshold circuit 3 is sent to the first signal detection circuit 4 and the data replacement circuit 5.

The first signal detection circuit 4 detects a signal using Equation 3 above. The detection result obtained here is the same as the detection result of the conventional method, and in order to distinguish it from the detection result obtained later, the first
A first detected value 41, called a detected value, is output from the first output terminal 200.

The data replacement circuit 5 replaces the cell detected as a signal by the first detection among the frequency component data with the nearby hill mochi 1. That is, the replacement is performed using Equation 4 with Xi as the replacement data.

Xi'= Xi:ff Xi<TtMi:if
Xi>zTi ΦψEquation 4 The replacement data 51 (Xi') is sent to the second signal detection circuit 8, and also to the 231st noise equalization circuit 6, where it is used as the second noise equalization &amp;
i is calculated. Calculations in the second noise equalization circuit 6 are performed using Equation 5.

and Xi': level of the 4th cell (i blind 1-N
) Mp: the second noise average value for the pth cell [1 (p=1-N) W: the number of cells used to calculate the fi sound average value, where p<=W
/2 or if p>=N-W/2.

respectively, p=W/2+2゜ A noise average value of p=N-W/2-2 is used.

The second miscellaneous perspiration average value 61 is sent to the threshold/rud circuit 7. In the second threshold circuit 7, a second threshold is calculated using equation 6.

Second threshold T i' = K * M i.

The second threshold 71 of the +1#- equation is sent to the second signal detection circuit 8. The second signal detection circuit 8 detects a signal using Equation 4 (7).

Signal: Xi'>=Ti' Non-signal: Xi'<Tt' *a* Equation 7 As a result,
Signal detection can be performed at the second threshold threshold based on the second noise average value that does not include the influence of strong signals, and the result is the second
A secondary detector 81 is obtained from the output terminal 300.

[Effects of the Invention] As explained above, the present invention has the following advantages: - After replacing the detected signal with the average noise value of one neighborhood, the noise modulus, threshold, and threshold are calculated again and the detection is performed again. This has the effect of detecting weak signals that exist near strong signals.

[Brief explanation of the drawing]

FIG. 1 shows an embodiment of the present invention, and FIG. 2 shows an explanation of problems caused by the conventional method. l: Memory circuit 2: First noise equalization circuit 3: First threshold circuit 4: First signal detection circuit 5: Data, 22 conversion circuit 6: Second noise? Equalization circuit 7: Second threshold circuit 8: Second signal detection circuit Representative Patent attorney Yoshihiro Watanabe

Claims (1)

[Claims] a storage means for storing frequency component data of underwater sounds; a first noise averaging means for calculating a first noise average value of the frequency component data inputted from the storage means; and a first noise averaging means for calculating a first threshold from the first noise average value. a first threshold means, a first signal detection means for comparing the frequency component data with a first threshold and detecting a signal; and comparing the frequency component data and the first threshold and detecting a frequency when the frequency component data is equal to or higher than the first threshold. a data replacement means for replacing component data with a first noise average value and outputting it as replacement data; a second noise average value means for calculating a second noise average value of this replacement data; and a second threshold value from the second noise average value. a second threshold means for determining
An underwater acoustic signal detection system comprising: second signal detection means for detecting a signal by comparing the replacement data with a second threshold.