Robust acoustic target depth attribute discrimination method in complex marine environment and storage medium

Abstract

The invention discloses a robust acoustic target depth attribute discrimination method in a complex marine environment and a storage medium, and relates to the technical field of underwater target depth classification, in particular to an acoustic target depth attribute discrimination method and a storage medium. The objective of the invention is to solve the problem that the existing target depth classification needs a vertical array to cover most waveguide depths, otherwise, the target depth classification accuracy is low. The method comprises the following steps of: 1, acquiring a vertical array receiving signal, and performing beam forming on the receiving signal to obtain a sound field elevation angle structure; 2, calculating the modal elevation cosine of the critical modal; 3, according to the modal elevation cosine of the critical modal in the step 2 and the sound field elevation structure in the step 1, calculating a notch modal-free modal sound field elevation structure strength ratio; and 4, taking the notch mode-free mode sound field elevation structural strength ratio in the step 3 as a decision metric, comparing the decision metric with a classifier threshold, and giving a criterion of a water surface target or an underwater target.

Description

Technical Field

[0001] The present invention relates to the technical field of underwater target depth classification, and in particular to a robust acoustic target depth attribute discrimination method and storage medium in a complex ocean environment. Background Art

[0002] The purpose of target depth classification is to distinguish between surface targets and underwater targets by analyzing sonar received signals, providing information support for underwater acoustic detection and confrontation. From the perspective of underwater acoustic physics, using the sound propagation characteristics of radiated noise to distinguish between surface and underwater targets is a hot and difficult issue in underwater acoustic research. The negative jump layer waveguide is a typical shallow sea summer waveguide. There are both free and notched simple normal waves in the waveguide, and the sound propagation characteristics are affected by the sending and receiving depth. As a sound propagation feature, the sound field elevation angle structure has a dependence on the target depth. Therefore, the relevant characteristics of the sound field elevation angle structure can be used to solve the problem of target depth classification.

[0003] Since the vertical array can sample the sound field in the vertical direction, it is convenient to extract the difference characteristics of the sound field of targets at different depths. The target depth classification method often relies on the vertical array. However, in order to obtain better feature extraction and classification effects, the vertical array usually needs to cover most of the waveguide depth to fully sample the simple normal wave. Otherwise, the performance of target depth classification will be seriously reduced. Therefore, it is difficult to realize the extraction of relevant characteristics of the sound field elevation structure and target depth classification under the small aperture vertical array. Summary of the invention

[0004] The purpose of the present invention is to solve the problem that the existing target depth classification requires a vertical array to cover most of the waveguide depth, otherwise the target depth classification accuracy is low, and to propose a robust acoustic target depth attribute discrimination method and storage medium in a complex ocean environment.

[0005] The specific process of the robust acoustic target depth attribute discrimination method in complex ocean environments is as follows:

[0006] Step 1: Obtain the vertical array receiving signal, perform beamforming on the received signal, and obtain the elevation angle structure of the sound field;

[0007] Step 2: Calculate the modal elevation angle cosine of the critical mode;

[0008] Step 3: Calculate the notched mode-free mode sound field elevation angle structure intensity ratio according to the modal elevation angle cosine of the critical mode in step 2 and the sound field elevation angle structure in step 1;

[0009] Step 4: The notch mode-free mode sound field elevation angle structure intensity ratio in step 3 is used as a decision metric and compared with the classifier threshold to give the criterion for surface target or underwater target.

[0010] A computer storage medium stores at least one instruction, which is loaded and executed by a processor to implement the robust acoustic target depth attribute discrimination method in a complex ocean environment.

[0011] The beneficial effects of the present invention are:

[0012] The present invention proposes the concept and calculation method of the elevation angle structural intensity ratio of the trapped mode to the free mode sound field for the first time, and uses it as a decision metric to construct a shallow sea negative jump layer waveguide target depth classifier based on a small aperture vertical array.

[0013] It has the following beneficial effects:

[0014] 1) The present invention can realize the effective classification of surface targets and underwater targets in shallow sea negative jump layer waveguides, does not require a vertical array to cover most of the waveguide depth, has high accuracy in target depth classification, and has a simple implementation method and is easy to operate.

[0015] 2) The present invention has good adaptability to small-aperture vertical arrays, has good classification effects on vertical arrays covering a depth less than the lower half of the water layer, and reduces the requirements on the aperture of the vertical array.

[0016] 3) The calculation of the elevation angle structural intensity ratio of the notched mode-free mode sound field has a low dependence on the prior knowledge of the environment, and the classifier still has a robust classification performance when the environmental parameters are not accurately measured. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of a robust acoustic target depth attribute discrimination method in a complex ocean environment according to the present invention;

[0018] Figure 2 is a schematic diagram of vertical array reception, φ m is the modal elevation angle of the critical mode order m, z s is the target depth;

[0019] Figure 3 is the measured sound velocity profile, z is the array element depth, and c is the sound velocity at the vertical array coverage depth;

[0020] Figure 4 is the elevation angle structure diagram of the target’s sound field, (a) is the elevation angle structure diagram of the surface target’s sound field, (b) is the elevation angle structure diagram of the underwater target’s sound field, φ is the array scanning angle;

[0021] Figure 5 is the variation law of the elevation angle structure strength ratio of the trap mode to the free mode sound field with the target depth, TFMR is the elevation angle structure strength ratio of the trap mode to the free mode sound field, Z is the array element depth, Z lim is the critical depth;

[0022] Figure 6 is the probability distribution of the elevation angle structure intensity ratio of the notched mode to the free mode sound field at a typical target depth;

[0023] Figure 7 is the effect of vertical array aperture on the performance of the classification method, N is the number of test array elements, and Pd is the detection probability;

[0024] Figure 8 is the influence of signal-to-noise ratio on the performance of the classification method, SNR is signal-to-noise ratio, and dB is decibel. DETAILED DESCRIPTION

[0025] Specific implementation method 1: The specific process of the robust acoustic target depth attribute discrimination method in a complex ocean environment in this implementation method is as follows:

[0026] Step 1: Obtain the vertical array receiving signal, perform beamforming on the received signal, and obtain the elevation angle structure of the sound field;

[0027] Step 2: Based on the normal wave theory, calculate the modal elevation cosine of the critical mode;

[0028] Step 3: Calculate the notched mode-free mode sound field elevation angle structure intensity ratio according to the modal elevation angle cosine of the critical mode in step 2 and the sound field elevation angle structure in step 1;

[0029] Step 4: The notch mode-free mode sound field elevation angle structure intensity ratio in step 3 is used as a decision metric and compared with the classifier threshold to give the criterion for surface target or underwater target.

[0030] Specific implementation method 2: This implementation method is different from the specific implementation method 1 in that in step 1, a vertical array receiving signal is obtained, and beamforming is performed on the received signal to obtain the elevation angle structure of the sound field; the specific process is:

[0031] Step 1.1: Express the received signal y(z,t) of each element of the uniform vertical array as:

[0032] y(z,t)=s(t)*p(z,t)+n(t) (1)

[0033] Where s(t) is the target radiated noise, p(z,t) is the channel impulse response at each array element, z is the array element depth, n(t) is the received noise, t is time, and * is the convolution symbol;

[0034] Step 1.2: Perform beamforming on the received signal y(z,t) to obtain the sound field elevation structure B y (cosφ,f).

[0035] The other steps and parameters are the same as those in the first embodiment.

[0036] Specific implementation method three: This implementation method is different from specific implementation methods one or two in that in step 1.2, the received signal y(z, t) is beamformed to obtain the sound field elevation structure B y (cosφ,f); the expression is:

[0037]

[0038] Where N is the number of elements in the uniform vertical array, y n (r,f) is the Fourier transform result of the signal y(z,t) received by the nth array element, f is the frequency, r is the target distance, w is the target distance, n (cosφ,f) is the weighting vector, and φ is the array scanning angle.

[0039] The other steps and parameters are the same as those in the first or second embodiment.

[0040] Specific implementation method 4: This implementation method is different from any one of the specific implementation methods 1 to 3 in that the weighted vector w n The expression of (cosφ,f) is:

[0041]

[0042] Where j is the imaginary unit, j 2 =-1; k is the wave number, k = 2πf / c, c is the sound velocity at the vertical array coverage depth; Δz is the array element spacing, z 0 It is the depth of array element No. 0.

[0043] The other steps and parameters are the same as those in Specific Embodiments 1 to 3.

[0044] Specific implementation mode 5: This implementation mode is different from any one of the specific implementation modes 1 to 4 in that, in step 2, the modal elevation angle cosine of the critical mode is calculated based on the normal wave theory; the specific process is:

[0045] Step 2.1: Based on the normal wave theory, calculate the critical mode order in the waveguide:

[0046]

[0047] Where: m 0 is the critical mode order, the order m=1,2,...,M, M is the total order of the simple normal wave; v pm is the mth-order mode phase velocity; c max is the maximum value of the speed of sound in the waveguide;

[0048] Step 2.2: Calculate the critical mode order m 0 The cosine of the modal elevation angle.

[0049] The other steps and parameters are the same as those in Specific Embodiments 1 to 4.

[0050] Specific implementation method 6: This implementation method is different from the specific implementation methods 1 to 5 in that the critical modal order m is calculated in step 2.2. 0 The modal elevation cosine of is expressed as:

[0051]

[0052] Where μ is the cosine of the modal elevation angle of the critical mode, For m 0 Order horizontal wave number; is the critical mode order m 0 The modal elevation angle.

[0053] The other steps and parameters are the same as those in Specific Implementation Methods 1 to 5.

[0054] Specific implementation method seven: This implementation method is different from any one of specific implementation methods one to six in that, in step 3, the notch mode-free mode sound field elevation angle structure intensity ratio is calculated according to the modal elevation angle cosine of the critical mode in step 2 and the sound field elevation angle structure in step 1; the specific process is:

[0055] According to the modal elevation angle cosine μ of the critical mode in step 2 and the sound field elevation angle structure B in step 1 y (cosφ,f), calculate the elevation angle structural strength ratio R of the trapped mode-free mode sound field a (f), the expression is:

[0056]

[0057] The numerator is the intensity of the sound field elevation structure in the notch mode angle region, and the denominator is the intensity of the sound field elevation structure in the free mode angle region.

[0058] The other steps and parameters are the same as those in Specific Embodiments 1 to 6.

[0059] Specific implementation eight: This implementation is different from any one of specific implementations one to seven in that in step 4, the notch mode-free mode sound field elevation angle structure intensity ratio in step 3 is used as a decision metric and compared with the classifier threshold to give a criterion for surface targets or underwater targets; the specific process is as follows:

[0060] The elevation angle structural strength ratio R of the notched mode-free mode sound field in step 3 a (f) as a decision metric;

[0061] The decision metric R a (f) Compared with the classifier threshold η:

[0062]

[0063] If the decision metric R a (f) is less than or equal to the classifier threshold η, then it is judged as H 0 , H 0 Indicates that the target is on the water surface;

[0064] If the decision metric R a (f) is greater than the classifier threshold η, then it is judged as H 1 , H 1 Indicates that the target is underwater.

[0065] The other steps and parameters are the same as those in Specific Embodiments 1 to 7.

[0066] Specific implementation method 9: This implementation method is different from any one of specific implementation methods 1 to 8 in that the classifier threshold is 0.5≤η≤1.

[0067] The other steps and parameters are the same as those in Specific Embodiments 1 to 8.

[0068] Specific implementation ten: This implementation is a computer storage medium, which stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement a robust acoustic target depth attribute discrimination method in a complex ocean environment as described in any one of items 1 to 9.

[0069] It should be understood that the instructions include computer program products, software or computerized methods corresponding to any method described in the present invention; the instructions can be used to program a computer system, or other electronic device. Computer storage media may include readable media on which instructions are stored, which may include but are not limited to magnetic storage media, optical storage media; magneto-optical storage media include read-only memory ROM, random access memory RAM, erasable programmable memory (e.g., EPROM and EEPROM) and flash memory layers, or other types of media suitable for storing electronic instructions.

[0070] Embodiment 1:

[0071] according to Figures 1 to 6 As shown, the present invention provides a robust acoustic target depth attribute discrimination method in a complex ocean environment. To verify the feasibility of the present invention, the following simulation experiment is designed.

[0072] Simulation conditions: The measured sound velocity profile is as follows Figure 3 As shown. The bottom of the vertical array is anchored to the bottom of the water through an anchoring device. The number of array elements is 22, and the array element interval is 5m. The coverage depth is 107m~212m, covering about 48% of the sea depth. The signal frequency is 130Hz. Assume that the target depth is 1~216m, and the distance between the target and the vertical array is 5km.

[0073] The simulation experiment includes the following steps:

[0074] Step 1: Obtain the vertical array receiving signal and perform beamforming on the receiving signal corresponding to each target depth. The elevation angle structure of the sound field of the surface target at a depth of 9m is as follows: Figure 4 As shown in (a), the elevation angle structure of the sound field of an underwater target at a depth of 54m is as follows: Figure 4 As shown in (b). The modal elevation cosine of the critical mode is calculated based on the simple normal wave theory, which is marked with a black dotted line in the figure, and the free mode angle area and the notch mode angle area are marked. Figure 4 Middle (a), Figure 4 As can be seen from (b), the high-intensity sound field elevation structure of the surface target in the present invention appears in the free mode angle region, and the high-intensity sound field elevation structure of the underwater target appears in the notch mode angle region.

[0075] Step 2: According to the modal elevation cosine of the critical mode in the previous step and the acoustic field elevation structure, calculate the intensity ratio of the notched mode to the free mode acoustic field elevation structure at each target depth. Figure 5 The variation law of the elevation angle structure intensity ratio of the notched mode to the free mode sound field with the target depth is given. Figure 5 It can be seen that in the present invention, the value of the elevation angle structural intensity ratio of the trapped wave mode to the free mode sound field of the surface target is smaller than that of the underwater target.

[0076] Step 3: Take the elevation angle structural intensity ratio of the notched mode to free mode sound field in step 2 as the decision metric, set the signal-to-noise ratio to 10 dB, perform 200 Monte Carlo experiments at each target depth, and statistically analyze the probability distribution of the decision metric at each target depth. Figure 6 The probability distribution of decision metrics at several typical target depths is given. Figure 6 It can be seen that when the target depth is 15m or less, the values ​​of the decision metric are concentrated below 0.3, and the probability distribution curve is sharp; for other target depths, the values ​​of the decision metric are greater than 0.5, and the probability distribution curve is relatively flat. Compared with the threshold η=0.4, it can correctly give the judgment of the surface target or the underwater target. The feasibility of the invention is verified.

[0077] Embodiment 2:

[0078] Assume that the test array aperture increases from 25m (N=6, covering about 12% of the sea depth) to 105m (N=22, covering about 48% of the sea depth). The depth ranges of surface targets and underwater targets are 1m~15m and 16m~216m, and 1000 target depths are randomly selected according to uniform distribution. The signal-to-noise ratio is 10dB. Monte Carlo experiments are performed 200 times at each target depth. Other simulation conditions are the same as those in specific embodiment 1. Drawing on the relevant concepts of binary detection, the detection probability Pd represents the probability of correctly classifying an underwater target, and the false alarm probability Pf represents the probability of classifying a surface target as an underwater target. Assume that the false alarm probability is 10 -3 , Figure 7 The influence of vertical array aperture on the detection probability of the classification method is given.

[0079] Depend on Figure 7 It can be seen that even if about 14% of the ocean depth is covered (N=7), the classification method can still maintain a robust depth classification performance with a detection probability of about 0.81, indicating that the classification performance of the method is less constrained by the aperture, verifying the effectiveness of the invention.

[0080] Embodiment three:

[0081] Assume the signal-to-noise ratio is -5dB to 35dB. The 6-element vertical array is anchored to the bottom of the water, covering about 12% of the sea depth. Other simulation conditions are the same as those in the second embodiment. Assume the false alarm probability is 10 -3 , Figure 8 The effect of signal-to-noise ratio on the detection probability of the classification method is given.

[0082] Depend on Figure 8 It can be seen that when the SNR is high (>10dB), the classification method provides a detection probability of more than 0.8, which verifies the effectiveness of the invention.

[0083] The present invention may also have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.

Claims

1. A robust method for distinguishing acoustic target depth attributes in complex ocean environments, characterized by: The specific process of the method is: Step 1: Obtain the vertical array receiving signal, perform beamforming on the received signal, and obtain the elevation angle structure of the sound field; Step 2: Calculate the modal elevation angle cosine of the critical mode; Step 3: Calculate the notched mode-free mode sound field elevation angle structure intensity ratio according to the modal elevation angle cosine of the critical mode in step 2 and the sound field elevation angle structure in step 1; Step 4: The notch mode-free mode sound field elevation angle structure intensity ratio in step 3 is used as a decision metric and compared with the classifier threshold to give the criterion for surface target or underwater target.

2. The robust acoustic target depth attribute discrimination method in a complex ocean environment according to claim 1 is characterized by: In step 1, a vertical array receiving signal is obtained, and beamforming is performed on the receiving signal to obtain the elevation angle structure of the sound field; the specific process is: Step 1.1: Express the received signal y(z,t) of each element of the uniform vertical array as: y(z,t)=s(t)*p(z,t)+n(t) (1) Where s(t) is the target radiated noise, p(z,t) is the channel impulse response at each array element, z is the array element depth, n(t) is the received noise, t is time, and * is the convolution symbol; Step 1.2: Perform beamforming on the received signal y(z,t) to obtain the sound field elevation structure B y (cosφ,f).

3. The robust acoustic target depth attribute discrimination method in a complex ocean environment according to claim 2 is characterized by: In step 1.2, beamforming is performed on the received signal y(z, t) to obtain the sound field elevation structure B y (cosφ,f); the expression is: Where N is the number of elements in the uniform vertical array, y n (r,f) is the Fourier transform result of the signal y(z,t) received by the nth array element, f is the frequency, r is the target distance, w is the target distance, n (cosφ,f) is the weighting vector, and φ is the array scanning angle.

4. The robust acoustic target depth attribute discrimination method in a complex ocean environment according to claim 3 is characterized by: The weight vector w n The expression of (cosφ,f) is: Where j is the imaginary unit, j 2 =-1; k is the wave number, k=2πf / c, c is the sound velocity of the vertical array coverage depth; Δz is the array element spacing, z0 is the depth of array element No.

0.

5. The robust acoustic target depth attribute discrimination method in a complex ocean environment according to claim 4 is characterized by: In step 2, the modal elevation cosine of the critical mode is calculated; the specific process is: Step 2.1: Calculate the critical mode order in the waveguide: Where: m0 is the critical mode order, order m = 1, 2, ..., M, M is the total order of the simple normal wave; v pm is the mth-order mode phase velocity; c max is the maximum value of the speed of sound in the waveguide; Step 2.2: Calculate the modal elevation angle cosine of the critical modal order m0.

6. The robust acoustic target depth attribute discrimination method in a complex ocean environment according to claim 5 is characterized by: In step 2.2, the modal elevation angle cosine of the critical modal order m0 is calculated, and the expression is: Where μ is the cosine of the modal elevation angle of the critical mode, k rm0 is the m0th order horizontal wave number; φ m0 is the modal elevation angle of the critical mode order m0.

7. The robust acoustic target depth attribute discrimination method in a complex ocean environment according to claim 6 is characterized by: In step 3, the intensity ratio of the notched mode to the free mode sound field elevation angle structure is calculated according to the modal elevation angle cosine of the critical mode in step 2 and the sound field elevation angle structure in step 1; the specific process is: According to the modal elevation angle cosine μ of the critical mode in step 2 and the sound field elevation angle structure B in step 1 y (cosφ,f), calculate the elevation angle structural strength ratio R of the trapped mode-free mode sound field a (f), the expression is: The numerator is the intensity of the sound field elevation structure in the notch mode angle region, and the denominator is the intensity of the sound field elevation structure in the free mode angle region.

8. The robust acoustic target depth attribute discrimination method in a complex ocean environment according to claim 7 is characterized by: In the step 4, the notch mode-free mode sound field elevation angle structure intensity ratio in step 3 is used as a decision metric, and compared with the classifier threshold to provide a criterion for a surface target or an underwater target; The specific process is: The elevation angle structural strength ratio R of the notched mode-free mode sound field in step 3 a (f) as a decision metric; The decision metric R a (f) Compared with the classifier threshold η: If the decision metric R a (f) If it is less than or equal to the classifier threshold η, it is judged as H0, which means that the target is on the water surface; If the decision metric R a (f) If it is greater than the classifier threshold η, it is judged as H1, which means that the target is underwater.

9. The robust acoustic target depth attribute discrimination method in a complex ocean environment according to claim 8 is characterized by: The classifier threshold is 0.5≤η≤1.

10. A computer storage medium, characterized in that: The storage medium stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the robust acoustic target depth attribute discrimination method in a complex ocean environment described in any one of items 1 to 9.

Images