Target depth and speed estimation method based on cepstrum multipath matching

Abstract

The invention provides a target depth and speed estimation method based on cepstrum multipath matching. The method comprises the following steps: firstly, estimating a multipath time delay process generated by a multipath propagation effect based on a sound signal received by a hydrophone; traversing combinations of different target depths and target speeds in an expected target position area to form hypothetical parameter conditions, and solving a multipath time delay process for each hypothetical parameter condition; and matching the multipath time delay history obtained according to the plurality of hypothesis parameter conditions with the multipath time delay history estimated according to the sound signal, and taking the target depth and the target speed corresponding to the matcher as estimation results. By using the method, depth and speed estimation can be performed on the underwater target under the condition of unknown depth.

Description

Technical Field

[0001] The present invention relates to the fields of underwater acoustic signal processing and sonar technology, and specifically relates to a method for estimating the depth and velocity of a target based on cepstrum multipath matching. Background Art

[0002] When an underwater unmanned platform locates a surface target, since its own depth is known and the depth of the surface target is known, the combination of azimuth and pitch angle estimation and depth can be used to locate it. For an underwater target, due to the unknown depth, existing underwater unmanned systems generally can only measure its direction, and it is difficult to estimate the depth and velocity of an underwater target with unknown depth, so positioning cannot be performed. Summary of the Invention

[0003] In view of this, the present invention provides a method for estimating the depth and velocity of a target based on cepstrum multipath matching, which can estimate the depth and velocity of an underwater target under the condition of unknown depth.

[0004] In order to solve the above technical problems, the present invention is implemented as follows.

[0005] A method for estimating the depth and velocity of a target based on cepstrum multipath matching includes:

[0006] Step 1: Estimate the multipath delay history generated by the multipath propagation effect by using cepstrum analysis method based on the acoustic signal received by the hydrophone;

[0007] Step 2: Traverse the combinations of different target depths and target velocities within the expected target position area to form hypothetical parameter conditions, and calculate the multipath delay history for each hypothetical parameter condition;

[0008] Step 3: Match the multipath delay histories calculated according to multiple hypothetical parameter conditions with the multipath delay history estimated from the acoustic signal, and use the corresponding target depth and target velocity of the matcher as the estimation result.

[0009] Preferably, the Step 1 includes:

[0010] Step S11: Filter and amplify the acoustic signal received by the hydrophone and then sample it;

[0011] Step S12: Obtain the cepstrum sequence of each frame of the acoustic signal;

[0012] Step S13: Obtain the negative peaks of the cepstrum sequences of K frames to obtain the multipath delay history.

[0013] Preferably, in the Step 2, the multipath delay history is calculated for each hypothetical parameter condition by using the following formula:

[0014]

[0015] r t = v|t - T CPA |

[0016] where τ s is the time delay calculated using the target depth z and target speed v in the assumed parameter conditions; r t is the horizontal distance between the sound source and the hydrophone receiving point at time t; T CPA is the time when the sound source and the hydrophone receiving point are closest. Take the time corresponding to the maximum value in the multipath time delay history obtained in step 1. z0 is the known depth of the hydrophone receiving point, and c is the sound speed in water; t takes multiple times to obtain multiple time delays, forming a multipath time delay history.

[0017] Preferably, step 3 is: perform matching through a cost function; take the assumed parameter conditions that make the cost function value maximum, and the corresponding target depth and speed as the estimation results.

[0018] Preferably, the cost function J is:

[0019]

[0020] where K is the number of time delay points extracted; τ i is the time delay of the i-th frame in the multipath time delay history estimated from the acoustic signal, and τ si is the time delay of the i-th frame in the multipath time delay history obtained according to the assumed parameter conditions.

[0021] Beneficial effects:

[0022] (1) Through the multipath time delay matching search of depth and speed parameters within a certain range, the present invention realizes the estimation of the depth and speed of an underwater target in the case of unknown depth.

[0023] (2) The accuracy of estimating the multipath time delay using the cepstrum analysis method does not depend on the limitation of the signal bandwidth. Compared with the autocorrelation multipath time delay estimation method, the cepstrum analysis method is more conducive to application in a strong line spectrum environment.

[0024] (3) The present invention can realize the estimation of the target depth and speed through a single hydrophone, and the system is simple.

[0025] (4) The present invention combines cepstrum analysis, multipath time delay history calculation, and cost function J calculation to realize search matching, with small computational complexity and high computational accuracy, which is conducive to application on an underwater unmanned platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a flowchart of the method for estimating the target depth and speed of the present invention.

[0027] Figure 2 It is the cepstrum waterfall plot in the simulation of the present invention;

[0028] Figure 3 It is the multipath delay plot extracted in the simulation of the present invention;

[0029] Figure 4 It is the cost function value plot of only sea surface reflection in the simulation of the present invention. Specific embodiments

[0030] The following combines the drawings and gives embodiments to describe the present invention in detail.

[0031] The present invention provides a method for estimating target depth and speed based on cepstrum multipath delay matching. The core idea is: under the condition that only the sea surface multipath delay can be measured, combined with the sea depth information, through the multipath delay matching search of depth and speed parameters within a certain range, the estimation of target depth and speed is realized. The estimation of target depth and speed can be achieved through a single hydrophone. The system is simple, with a small amount of calculation and high calculation accuracy, which is beneficial to be applied on underwater unmanned platforms.

[0032] Figure 1 The flowchart of the method for estimating target depth and speed based on cepstrum multipath delay matching of the present invention is shown. As shown in the figure, it includes:

[0033] Step 1: Based on the acoustic signal received by the hydrophone, use cepstrum analysis method to estimate the delay history generated by the multipath propagation effect of the acoustic signal.

[0034] Step 2: Traverse the combinations of different target depths and target speeds within the expected target position area, form hypothetical parameter conditions, and perform a search based on the hypothetical parameter conditions to obtain the multipath delay history under the corresponding hypothetical parameter conditions.

[0035] Step 3: Match the multipath delay history obtained according to multiple hypothetical parameter conditions with the multipath delay history estimated from the acoustic signal, and use the corresponding target depth and target speed of the matcher as the estimation result.

[0036] The specific implementation scheme of the present invention is as follows:

[0037] Step 1 is specifically:

[0038] Step S11: Filter and amplify the signal received by the hydrophone and then sample it, and the sampling frequency is f s .

[0039] Step S12: Use cepstrum analysis to estimate the multipath delay:

[0040] Frame the received signals of the hydrophone. The time length of one frame of signals is T0 (unit: s), and a discrete signal sequence x(n) is obtained, where n = 1, 2, …, N, and N = T0·f s , and N is an even number.

[0041] First, perform FFT operation on the sequence x(n), n = 1, 2, …, N to obtain a complex sequence y(k), where k = 1, 2, …, N. Then, take the square of the absolute value of y(k) to obtain the power spectrum P(k); then take the log logarithm of P(k) to obtain the sequence Y(k). Perform inverse Fourier transform on Y(k) and take its real part to obtain the cepstrum sequence C(n), where n = 0 to N - 1. The cepstrum sequence C(n) is symmetric about the left and right, and only the first N / 2 points are used for subsequent processing.

[0042] Step S13: Extract the multipath time delay.

[0043] The cepstrum sequence of the first frame obtained by the method of step S12 is denoted as C1(n), where n = 1, 2, …, N / 2. Similarly, the cepstrum sequence of the second frame is denoted as C2(n), and the cepstrum sequence of the i-th frame is denoted as C i (n), where i = 1, 2, …, K, and K is the total number of frames for data processing.

[0044] When negative peaks are searched in the set interval range in three consecutive frames of cepstrum sequences, analyze the corresponding time delay values of these three negative peaks according to certain criteria. For targets from far to near, if the three time delay values increase in sequence and their differences are less than a threshold value determined by the maximum possible speed of the target and the frame length, it is determined that the multipath time delay is caused by a moving sound source, and enter the locking mode.

[0045] In the locking mode, search for negative peaks in the range [-τ l , τ u before and after the time delay of the previous frame. The value of this range is: before the target passes abeam, τ l is close to 0, and τ u is 3 times the time delay difference between the previous two frames. After the target passes abeam, τ u is close to 0, and τ i takes 3 times the time delay difference between the previous two frames.

[0046] Complete the negative peaks of the K-frame cepstrum sequence according to this method to obtain K time delays τ i (i = 1, 2, ……, K), which form the multipath time delay history.

[0047] Step 2 is specifically: Assume the calculation of multipath time delay under different depths and speeds.

[0048] The relationship between the sea surface reflection multipath time delay τ s and the target depth and speed:

[0049]

[0050] Among them, τ s is the time delay calculated using the target depth z and target speed v in the assumed parameter conditions; r t = v|t - T CPA | is the horizontal distance between the sound source and the receiving point at time t, and T CPA is the moment when the sound source is closest to the hydrophone receiving point, and it takes the moment corresponding to the maximum value in the multipath time delay history obtained in step 1. z0 is the known depth of the hydrophone receiving point, and c is the sound speed in water.

[0051] In this way, for each set of assumed depth z and speed v, t in formula (1) takes multiple moments, and a set of time delay sequences τ si (i = 1, 2, ……, K) are obtained, which form the multipath time delay history. Different combinations of depth and speed parameters are set in the expected target position area for search to obtain multiple sets of multipath time delay histories under different assumed parameter conditions.

[0052] Step 3 is specifically as follows: In this step, the matching of the multipath time delay history is carried out by setting a cost function. The multipath time delay histories obtained under different assumed conditions are matched with the multipath time delay history estimated from the sound signal, the cost function value is calculated, and the assumed parameter conditions that make the cost function value the largest are taken, and the corresponding target depth and speed are used as the estimation results.

[0053] The cost function established in this step is:

[0054]

[0055] Among them, K is the number of time delay points extracted, which is also the number of data frames mentioned above. τ i is the time delay of the i-th frame in the multipath time delay history estimated from the sound signal, and τ si is the time delay of the i-th frame in the multipath time delay history obtained according to the assumed parameter conditions.

[0056] The method of the present invention is verified by simulation: Assume that the sea depth is 100 m, there is an underwater sound source with a depth of 30 m and a speed of 4 kn, and the receiving depth is 60 m. The cepstrum waterfall diagram of the simulated underwater moving target is as Figure 2 shown. The multipath time delay is extracted using the peak from the cepstrum waterfall diagram, as Figure 3 shown. Figure 4 is the cost function value for the matching between the time delay difference between the sea surface reflection and the direct wave and the estimated time delay value. The maximum value is 4.3434. At this time, the depth of the target is 13.95 m and the speed is 1.85 m / s. The depth estimation error is 1.05 m, and the speed is 1.85 m / s. The speed estimation error is 0.05 m / s, which can meet the application requirements, indicating the feasibility of the method.

[0057] The above specific embodiments only describe the design principle of the present invention. The shapes and names of the components in this description can be different and are not restricted. Therefore, those skilled in the art of the present invention can modify or make equivalent replacements to the technical solutions recorded in the foregoing embodiments; and these modifications and replacements do not depart from the gist and technical solutions of the present invention, and shall all fall within the protection scope of the present invention.

Claims

1. A method for estimating target depth and velocity based on cepstrum multipath matching, characterized in that Including: Step 1: Estimate the multipath delay profile generated by the multipath propagation effect using cepstrum analysis based on the acoustic signal received by the hydrophone; Step 2: Traverse combinations of different target depths and target speeds within the expected target position area to form assumed parameter conditions, and calculate the multipath delay profile for each assumed parameter condition; Step 3: Match the multipath delay profiles calculated according to multiple assumed parameter conditions with the multipath delay profile estimated from the acoustic signal, and use the target depth and target speed corresponding to the matcher as the estimation result.

2. The method according to claim 1, characterized in that, The said Step 1 includes: Step S11: Sample the acoustic signal received by the hydrophone after filtering and amplifying; Step S12: Obtain the cepstrum sequence of each frame of the acoustic signal; Step S13: Obtain the negative peaks of the cepstrum sequences of K frames to obtain the multipath delay profile.

3. The method according to claim 1, wherein In the said Step 2, the multipath delay profile is calculated for each assumed parameter condition using the following formula: r t = v|t - T CPA | where τ s is the time delay calculated using the target depth z and the target speed v in the assumed parameter conditions; r t is the horizontal distance between the sound source and the hydrophone receiving point at time t; T CPA is the moment when the sound source and the hydrophone receiving point are the closest, taking the moment corresponding to the maximum value in the multipath time delay history obtained in step 1, z0 is the known depth of the hydrophone receiving point, c is the sound speed in water; t takes multiple moments to obtain multiple time delays, forming a multipath time delay history.

4. The method according to any one of claims 1 to 3, characterized in that The said Step 3 is: Perform matching through a cost function; select the assumed parameter condition that makes the cost function value the largest, and use the corresponding target depth and speed as the estimation result.

5. The method according to claim 4, wherein The said cost function J is: where K is the number of extracted time delay points; τ i is the time delay of the i-th frame in the multipath time delay profile estimated from the acoustic signal, τ si is the time delay of the i-th frame in the multipath time delay profile obtained according to the assumed parameter conditions.

Images