A Doppler estimation and compensation method for high-speed mobile underwater acoustic communication

Abstract

The present invention discloses a Doppler estimation and compensation method for high-speed mobile underwater acoustic communication, comprising: receiving an underwater acoustic communication signal in a specific frame format with a Doppler effect, locally refining and amplifying the frequency of the CW signal through a Zoom-FFT algorithm, and determining the Doppler within a preset range; estimating the frequency deviation of the LFM signal through an AF algorithm, and compensating for the Doppler frequency shift through resampling; eliminating the remaining Doppler effect after compensation through symbol residual phase cancellation, converting the transmitted bipolar symbol into a complex domain symbol, transforming the complex domain symbol information, and when used for demodulation, the complex domain symbol cancels the phase between the symbols through inverse transformation. The present invention completes the estimation and compensation of large Doppler and residual Doppler, solves the problems of low Doppler estimation accuracy under low signal-to-noise ratio, long matching time for multiple correlators, difficulty in large Doppler compensation, and residual Doppler signal phase compensation under low complexity, and provides key technical support for the stable operation of high-speed mobile underwater acoustic communication systems.

Description

Technical Field

[0001] The present invention relates to the field of underwater acoustics, and in particular to a Doppler estimation and compensation method for high-speed mobile underwater acoustic communications. Background Art

[0002] As the importance of the ocean field in the national strategic position becomes more and more prominent, the application of marine underwater unmanned system platforms is increasing. Among them, unmanned boats, AUVs, unmanned ships, and UUVs can perform tasks such as ocean observation and shipwreck search in relatively complex marine environments. The information exchange of these marine unmanned platforms is inseparable from underwater acoustic communication. The underwater acoustic channel has the characteristics of severe Doppler spread, limited spectrum resources, and low signal-to-noise ratio. The acoustic wave information transmission environment is relatively harsh, and the stability of underwater acoustic communication is difficult to guarantee. Therefore, higher requirements are placed on stable marine communication technology.

[0003] Because the propagation speed of underwater sound is much lower than that of electromagnetic waves, underwater acoustic communications between mobile platforms are severely affected by Doppler interference. This Doppler effect can cause phase rotation of the demodulated signal within the communication system, severely impacting the stability of underwater acoustic communications. Therefore, suppressing the Doppler effect in underwater acoustic communication systems between rapidly moving platforms is a core research direction for ensuring stable operation of underwater communication systems. Extensive research has been conducted both domestically and internationally on Doppler estimation and compensation techniques for underwater acoustic communications between mobile platforms.

[0004] Patent publication number CN108833312A discloses a time-varying sparse underwater acoustic channel estimation method based on the delay-Doppler domain. This method includes modeling the time-varying underwater acoustic channel in the delay-Doppler domain to obtain a sparse two-dimensional representation. Based on the delay-Doppler domain representation, a block-by-block training mode is adopted, combined with the Schmidt orthogonal matching pursuit algorithm to iteratively optimize the sparse underwater acoustic channel impulse response function in the delay-Doppler domain. The estimated delay-Doppler domain underwater acoustic channel information is used to construct a minimum mean square error equalizer based on the channel information at the receiving end to recover the transmitted signal.

[0005] Patent publication number CN110247867A discloses an underwater acoustic Doppler estimation method and device, and an underwater acoustic communication method and system, which includes sending two hyperbolic frequency modulation signals at the transmitting end, and using the matched filter output of the first hyperbolic frequency modulation signal and the matched filter output of the second hyperbolic frequency modulation signal to perform re-correlation at the receiving end, thereby obtaining the matched filter output of the first hyperbolic frequency modulation signal and the overall relative time shift of the second hyperbolic frequency modulation signal, and then estimating the Doppler factor.

[0006] The patent with publication number CN112822135A discloses a single-carrier high-speed underwater acoustic communication method, which includes the following steps: (1) single-carrier signal coding modulation, in which the transmitter converts the original information sequence into serial-to-parallel and maps it to the QPSK phase set, and then adds a training sequence before the mapped phase sequence and modulates it with the carrier before sending it out; (2) channel estimation, in which the receiver completes signal detection and synchronization and estimates the channel using the locally known training sequence; (3) frequency domain zero-forcing equalization processing, in which the estimated channel is used to construct a weight matrix in the frequency domain and multiply it with the received signal to achieve channel equalization processing; (4) decision feedback equalization processing, in which the signal after frequency domain zero-forcing equalization processing is inverse Fourier transformed back to the time domain and finally decoded using a decision feedback equalizer with an embedded phase-locked loop.

[0007] The paper [Hu Yaohui et al., A Combined Differential Spread Spectrum Underwater Acoustic Communication Method for Underwater Mobile Communications [J], Journal of Electronics and Information Technology, 2022] discloses a method for solving the Doppler problem by implementing polarity differential decoding between two-dimensional combinations using combined differentials. Polarity modulation is then added after energy mapping decoding to improve spectrum efficiency. The paper [Wei Zhuoqun, Research on Mobile Robust Spread Spectrum Underwater Acoustic Communication Technology [D], 2018] discloses a delay-Doppler estimation method based on a narrowband mutual ambiguity function that can accurately estimate the Doppler of mobile nodes in underwater communications. It uses a direct-spread spectrum signal between every two symbols for real-time Doppler estimation and compensation, and employs resampling for Doppler compensation. However, this method is not suitable for eliminating residual Doppler effects in special situations. The paper [Qin Hao, Research on Communication Reliability of Underwater Mobile Nodes [D], 2016] discloses an improved Pisarenko harmonic decomposition method based on polynomials for Doppler estimation, but its high complexity makes it unsuitable for communications with varying Doppler frequencies.

[0008] The Doppler effect in underwater acoustic communications on high-speed mobile platforms cannot be fully compensated, or the Doppler tolerance of existing Doppler estimation and compensation algorithms cannot be effectively addressed. This leads to inaccurate estimation at low signal-to-noise ratios, long estimation times, incomplete residual Doppler compensation, and a complex algorithm. Existing Doppler estimation techniques include block Doppler estimation, which uses matched filtering to calculate the Doppler estimate using LFM signals before and after the signal. This method has low estimation accuracy, resulting in an average Doppler that cannot fully compensate for the signal's Doppler. Ambiguity function Doppler estimation algorithms have a long estimation time and poor real-time performance. Ambiguity functions are used to perform Doppler estimation, and residual Doppler is eliminated using a decision feedback equalizer embedded in a phase-locked loop. However, this algorithm is complex and slow, making it unsuitable for downshifting. Summary of the Invention

[0009] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a Doppler estimation and compensation method for high-speed mobile underwater acoustic communication.

[0010] The object of the present invention is achieved through the following technical solutions:

[0011] The present invention provides a Doppler estimation and compensation method for high-speed mobile underwater acoustic communication, comprising the following steps:

[0012] S1. Receive signals transmitted by a transmitting end, wherein the signals include a continuous single-frequency CW signal, a linear frequency modulated pulse LFM signal, and an underwater acoustic communication signal in a specific frame format with a Doppler effect, and perform local refinement and amplification of the frequency of the CW signal using a Zoom-FFT algorithm for fast and accurate Doppler estimation, thereby determining the Doppler of the underwater acoustic communication signal within a preset range;

[0013] S2. The frequency deviation of the LFM signal is estimated by using the ambiguity function AF to reduce the Doppler search step size, further refine the Doppler estimation, and compensate for the Doppler frequency shift by resampling.

[0014] S3. Completely eliminate the remaining Doppler effect after compensation by canceling the residual phase of the code element, convert the transmitted bipolar code element into a complex domain code element, transform the complex domain code element information, and when used for demodulation, cancel the phase between the complex domain code elements through inverse transformation;

[0015] Step S2 specifically includes:

[0016] The local LFM signal is subjected to a small step Doppler factor Δ based on the Doppler range determined by the Zoom-FFT algorithm. n (Δ0Δ1...Δ M ) changes, the local LFM signal after the Doppler change is correlated with the received LFM signal, the maximum value of the ambiguity function AF of the signal is selected, and the Doppler estimated based on the Zoom-FFT-AF combined Doppler algorithm is used to resample the underwater acoustic communication signal to compensate for the Doppler frequency offset;

[0017] The fuzzy function AF is defined as τ is the signal delay; f d is the Doppler shift;

[0018] The time function s(t) of the LFM signal is expressed as Where A is the signal amplitude, T is the transmission period of each code element, and f0 represents the carrier frequency;

[0019] If A=1, the ambiguity function AF of the local LFM signal is:

[0020]

[0021] The Doppler factor Δ of the received LFM signal and the local LFM signal is n (Δ0Δ1...ΔM ) is correlated with the signal after the change, and its fuzzy function AF expression can be expressed as: Where r is the received LFM signal.

[0022] Furthermore, the step S1 specifically includes:

[0023] Perform spectrum shifting on the CW signal cw0(n): Move to the center frequency f k At this point, we get the signal cw(n), where is the unit rotation factor;

[0024] Calculate the sampling ratio D = f s / f k , where f s is the sampling frequency;

[0025] Calculate the total number of sampling points N = DN of the CW signal cw0(n) r , where N r For analysis points;

[0026] Through the digital low-pass filter, the signal cw(n) is retained in the frequency band f s / (2D)The frequency band that needs to be refined is obtained by c (n), Where h(n) is the impulse response and the cutoff frequency of the digital low-pass filter is f c ≤f s / (2D);

[0027] If the sampling frequency is f s / D, for signal y c (n) Take a data point every (D-1) points to obtain the coarse-grained signal rc(m);

[0028] Perform N-point complex FFT on the coarse-grained signal rc(m), intercept the refined spectrum fft result, move it in the opposite direction, connect it front and back, and convert it into the actual frequency to obtain the refined FFT transformation result F CW (k), the obtained frequency is subtracted from the original frequency to obtain the approximate Doppler frequency shift.

[0029] Preferably, the elimination of the residual Doppler effect after compensation by canceling the residual phase of the code symbols in step S3 specifically includes: at the transmitting end, performing complex domain code symbol transformation on the PSK code symbols:

[0030]

[0031] in is the bipolar complex domain code element, is the information before transformation, is the intermediate transformation codeword, is the transformed code element, θ is the phase information corresponding to the code element a, is the phase information corresponding to the y code element, α is the phase information corresponding to the x code element, and then Perform spectrum spreading, raised cosine filtering, up-sampling, and up-conversion modulation to obtain a transmit signal;

[0032] The received underwater acoustic communication signal is demodulated, raised cosine filtered and despread based on the receiving end, and the complex domain code elements before judgment are inversely transformed based on the transmitting end:

[0033]

[0034] in To offset the I / Q path pre-decision data once, is the complex domain amplitude information of the transformed code element, β (n) is the corresponding phase information, is the data before I / Q path decision after symbol phase cancellation transformation, β'n is Its own phase information, and then Take out the real part and the imaginary part to make a decision and get the code element information.

[0035] The beneficial effects of the present invention are:

[0036] 1) The present invention uses the Zoom-FFT algorithm to quickly and accurately estimate the CW signal frequency, locks the signal Doppler within a small range, and uses the LFM signal AF algorithm to further search and estimate the Doppler with a small step size, solving the problem of estimating large Dopplers with high computational complexity and low precision under low signal-to-noise ratio.

[0037] 2) The present invention utilizes the characteristics of the phase information between code elements to perform a specific transformation on the complex domain code element information of the transmitted data. During demodulation, the phases between the complex domain code elements are offset by a specific inverse transformation. The algorithm has low complexity and is easy to implement. It solves the problem of residual Doppler compensation after Doppler compensation and provides key technical support for stable underwater mobile communications. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A schematic flow chart of a Doppler estimation and compensation method for high-speed mobile underwater acoustic communication according to an embodiment of the present invention;

[0039] Figure 2 A schematic diagram of a transmission signal structure of a Doppler estimation and compensation method for high-speed mobile underwater acoustic communication according to an embodiment of the present invention;

[0040] Figure 3 Schematic diagram of the Doppler estimation process of the Zoom-FFT algorithm according to an embodiment of the present invention;

[0041] Figure 4 1. A schematic diagram of the Doppler estimation structure of the AF algorithm according to an embodiment of the present invention;

[0042] Figure 5 1 is a flowchart of a Zoom-FFT-AF Doppler joint estimation algorithm according to an embodiment of the present invention;

[0043] Figure 6 Schematic diagram of Doppler factor estimation NMSE of Zoom-FFT-AF and AF algorithm according to an embodiment of the present invention;

[0044] Figure 7 Figure 2 shows a simulation diagram of signal demodulation constellations with and without Doppler compensation using the residual symbol phase cancellation algorithm according to an embodiment of the present invention, where (a) is without Doppler compensation and (b) is with Doppler compensation.

[0045] Figure 8 Schematic diagram of the bit error rate compensation estimated by different Doppler factors under the corresponding algorithm of an embodiment of the present invention, where (a) is a Doppler factor of 0.002 and (b) is a Doppler factor of 0.02. DETAILED DESCRIPTION

[0046] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0047] The present invention provides a Doppler estimation and compensation method for high-speed mobile underwater acoustic communication, the overall flow diagram of which is shown as follows: Figure 1 As shown in the figure, a signal with Doppler effect is first subjected to a zoom-FFT algorithm to locally refine and amplify the frequency of a continuous single-frequency CW signal, enabling rapid and relatively accurate Doppler estimation. This results in the Doppler of the communication information being localized within a narrow range. Using a linear frequency modulated (LFM) signal and an ambiguity function (AF), the algorithm's search step size is reduced, further refining the Doppler estimate. At this point, the Doppler estimate is relatively accurate. Resampling is then used to compensate for the Doppler shift, but the compensated signal still has residual Doppler effects, which seriously affect the demodulation quality. Finally, symbol residual phase cancellation is used to completely eliminate the Doppler effect. By leveraging the phase information between symbols, the transmitted bipolar symbols are converted into complex-domain symbols. A specific transformation is then performed on the complex-domain symbol information to achieve phase cancellation between the complex-domain symbols during demodulation, ensuring stable transmission of underwater acoustic communication information. The continuous single-frequency CW signal and the linear frequency modulated (LFM) signal are the signals preceding the underwater acoustic communication signal.

[0048] For example, CW signals are easily drowned by complex ocean noise at low signal-to-noise ratios. The frequency of the CW signal is roughly estimated by the Zoom-FFT algorithm. The LFM signal still has good delay and frequency shift resolution at low signal-to-noise ratios. The schematic diagram of the transmitted signal structure is shown in the figure. Figure 2 As shown, the CW signal performs a coarse Doppler estimation, and the LFM signal performs a fine estimation, which basically eliminates the Doppler effect of the signal.

[0049] For example, the Zoom-FFT algorithm can locally refine and amplify the frequency of the signal to obtain higher resolution. Compared with direct FFT, it has less data and shorter computing time. The Zoom-FFT algorithm realizes frequency estimation through the frequency shift method. The flowchart of the Doppler estimation of the Zoom-FFT algorithm is as follows: Figure 3 shown.

[0050] Specifically, the spectrum of the CW signal cw0(n) is shifted: Then calculate the sampling ratio D = f s / f k , also known as the refinement factor, f s is the sampling frequency, f k is the center frequency. cw0 total number of sampling points N = DN r , N r The low-pass filtering before sampling can prevent the frequency after sampling from aliasing. The low-pass digital filter is designed and the impulse response h(n) is calculated to obtain the signal The frequency band signal that needs to be analyzed is filtered out through a low-pass digital filter. The cutoff frequency of the low-pass digital filter is f c ≤f s / (2D). Assume the original sampling frequency is f s , the number of sampling points is N, then the frequency resolution is f s / N, the resampling frequency is f s / D, the number of sampling points N remains unchanged, then the frequency resolution is f s / (D·N), at the original sampling frequency f s Under this condition, the frequency resolution after processing is increased by D times, and the resampled output rc(m)=y c (Dn). Finally, perform N-point FFT on rc(m). The spectrum line at this time is not the spectrum line of the actual signal. It needs to be moved in the opposite direction, intercepted and connected front and back to become the actual spectrum, and recombined to obtain the refined frequency F CW (k).

[0051] Specifically, the frequency shift and time delay of the signal can be obtained by analyzing the ambiguity function (AF) of the signal. AF describes the joint characteristics of the signal in the time and frequency domains. When the delay and Doppler characteristic parameters of the constructed original signal are completely consistent with those of the signal passing through the underwater acoustic channel, the AF value at this time delay and Doppler is the largest, at its origin. Therefore, the maximum value of AF can be used to estimate the Doppler frequency shift of the signal passing through the channel. AF is defined as Among them, τ is the signal delay; f d is the Doppler frequency shift, and the time function of the LFM signal can be expressed as Where A is the signal amplitude, T is the transmission period of each symbol, and f0 is the carrier frequency. If A = 1, the AF of the local LFM signal is:

[0052]

[0053] The schematic diagram of the Doppler estimation structure of the AF algorithm is as follows: Figure 4 As shown, it is composed of several groups of correlators. The local LFM signal of correlator 0 is LFM((1+Δ0)t), the local LFM signal of correlator 1 is LFM((1+Δ1)t), and the local LFM signal of correlator M is LFM((1+Δ M )t), the received LFM signal is compared with the local LFM signal by the Doppler factor Δ n (Δ0Δ1...Δ M ) is correlated with the signal after the change, and its fuzzy function expression can be expressed as: Where r is the received LFM signal. The maximum AF value is then determined and selected, representing the Doppler shift caused by the signal passing through the channel. Doppler shift estimation requires a larger Doppler search over the original LFM signal, as the Doppler estimation range is unknown. This algorithm requires a smaller Doppler search step size to accurately estimate the Doppler factor, thus requiring a larger number of correlators. This results in a longer estimation time, making it unsuitable for real-time communication systems.

[0054] For example, after the Zoom-FFT algorithm estimates the approximate Doppler range, the AF algorithm searches and estimates in a smaller Doppler range with a small step size, which can reduce the estimation time and increase the estimation accuracy. The schematic diagram of the Zoom-FFT-AF Doppler joint estimation process is shown in the figure. Figure 5As shown in the figure, the solid line indicates that the AF algorithm has a range of ±5 and a step size of 1, and the dotted line indicates that the Zoom-FFT-AF algorithm has a range of ±5 and a step size of 0.1. The CW signal and the LFM signal take T = 0.05s, under the conditions of a signal-to-noise ratio of -10db, a moving speed of 2-30m / s, and the same correlator, the Zoom-FFT-AF (joint estimation) algorithm searches for a speed step of 0.1m / s and a search range of ±0.5m / s; the AF algorithm searches for a speed step of 1m / s and a search range of ±5m / s. The simulation results of the normalized mean square error of Doppler estimation by the AF algorithm and the Zoom-FFT-AF algorithm are shown in the figure. Figure 6 As shown in the figure, it can be obtained that if the Doppler range is determined in advance, the Doppler estimation accuracy of the Zoom-FFT-AF algorithm will be better under the premise of the same computational complexity.

[0055] Specifically, the transmitted symbol is recorded as a(n), which is obtained by converting the binary transmission signal into bipolar information, and then converting it into BPSK (a+ib, a is ±1, b is 0) or QPSK (a+ib, a, b is ) format to get the code element This code element is a complex number, which is the information before the specific transformation. In the form of a+ib, it is transformed into

[0056]

[0057] is the code element information to be modulated next, which is a complex signal, where n is 0, 1, ..., The code element a, y, x is in the form of a+b, θ, α is the phase information corresponding to the code element a, y, and x. Then the signal is spread spectrum, upsampled by raised cosine filtering, multiplied by the carrier for upconversion, and the signal frame format is constructed and transmitted. The received signal is r(t), the carrier frequency is f0, and the demodulated signal is expressed as:

[0058]

[0059] The baseband signal after low-pass filtering and despreading is:

[0060]

[0061] Specifically, the despread x'(t) is the baseband symbol information before the channel decision, f d It is the Doppler frequency shift. At this time, the Doppler information can be regarded as a cosine signal, which modulates the baseband information. The information is periodically phase-reversed in the envelope formed by the modulation. The corresponding discrete signal is X'(n) is the complex domain code element information, α (n) for Symbol phase information, T is each symbol transmission period, its phase information α (n) It is the original phase information of the baseband information and does not affect the stability of the system. X'(n) is in the form of a'+ib', where a' and b' are the bipolar code element information before the signal is demodulated and judged. According to the code element transformation process at the transmitter, the primary phase cancellation transformation can be expressed as:

[0062]

[0063] To offset the I / Q path pre-decision data once, is the complex domain amplitude information of the transformed code element, β (n) =j2πf d T+α (n) -α (n-1) for Contains Doppler phase information, baseband signal There are still some f left in d T’s phase information will cause the information phase to shift, making it difficult to demodulate. From the symbol transformation process at the transmitter, it is known that the phase cancellation transformation between symbols needs to be performed again. The cancellation result is expressed as:

[0064]

[0065] is the data before I / Q decision after symbol phase cancellation transformation, β' n for The baseband signal completely eliminates the Doppler effect, and then The real and imaginary parts are taken out for judgment to obtain the code element information. The code element phase discussed above does not affect the information demodulation, so no special processing is required.

[0066] For example, under the conditions of sampling frequency 48000, bandwidth 4000hz, spreading period 31, Doppler size 0.002, and signal-to-noise ratio 0dB, the signal demodulation constellation simulation diagram with and without code element residual phase cancellation algorithm for Doppler compensation is shown as follows: Figure 7 As shown in the figure, (a) is without Doppler compensation, and (b) is with Doppler compensation. From the simulated constellation, it can be seen that under the influence of small Doppler, the phase of the signal constellation without Doppler estimation and compensation technology rotates. The signal demodulation constellation with code element residual phase cancellation technology is better and the signal phase information can be clearly distinguished.

[0067] For example, since the code source residual phase cancellation algorithm requires that the code element time length must be much smaller than the envelope time length of the Doppler modulated baseband signal in order to achieve Doppler compensation between code elements, it is impossible to compensate for larger Doppler signals. Under the Doppler preliminary compensation of the Zoom-FFT-AF Doppler joint estimation algorithm, the remaining Doppler of the signal is already very small, which can fully meet the Doppler tolerance of the code element residual phase cancellation algorithm. Under the conditions of sampling frequency 48000, bandwidth 4000hz, spread spectrum period 31, and Doppler factor size of 0.002 and 0.02, the compensation bit error rate of different Doppler factors under the corresponding algorithm is as follows: Figure 8 As shown in Figure 1, (a) is a Doppler factor of 0.002. The asterisk line in Figure (a) represents no Doppler estimation compensation, and the circled line represents symbol residual phase cancellation. (b) is a Doppler factor of 0.02. The circled line in Figure (b) represents symbol residual phase cancellation, and the asterisk line represents Zoom-FFT-AF symbol residual phase cancellation. The simulation results show that when the Doppler factor is small, the symbol residual phase cancellation algorithm can effectively compensate for the Doppler frequency offset. When the Doppler factor is large, the symbol residual phase cancellation algorithm is no longer able to compensate for the Doppler frequency offset, exceeding the algorithm's Doppler tolerance. The Zoom-FFT-AF symbol residual phase cancellation algorithm can effectively compensate for the frequency offset caused by large Doppler.

[0068] The foregoing description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein and should not be construed as excluding other embodiments. Rather, the present invention can be used in various other combinations, modifications, and environments and can be modified within the scope of the concept described herein through the above teachings or techniques or knowledge in the relevant field. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention are intended to be protected by the appended claims.

Claims

1. A Doppler estimation and compensation method for high-speed mobile underwater acoustic communication, characterized in that: The following steps are involved: S1. Receive signals transmitted by a transmitting end, wherein the signals include a continuous single-frequency CW signal, a linear frequency modulated pulse LFM signal, and an underwater acoustic communication signal in a specific frame format with a Doppler effect, and perform local refinement and amplification of the frequency of the CW signal using a Zoom-FFT algorithm for fast and accurate Doppler estimation, thereby determining the Doppler of the underwater acoustic communication signal within a preset range; S2. The frequency deviation of the LFM signal is estimated by using the ambiguity function AF to reduce the Doppler search step size, further refine the Doppler estimation, and compensate for the Doppler frequency shift by resampling. S3. Completely eliminate the remaining Doppler effect after compensation by canceling the residual phase of the code element, convert the transmitted bipolar code element into a complex domain code element, transform the complex domain code element information, and when used for demodulation, cancel the phase between the complex domain code elements through inverse transformation; Step S2 specifically includes: The local LFM signal is subjected to a small step Doppler factor Δ based on the Doppler range determined by the Zoom-FFT algorithm. n (Δ0Δ1...Δ M ) changes, the local LFM signal after the Doppler change is correlated with the received LFM signal, the maximum value of the ambiguity function AF of the signal is selected, and the Doppler estimated based on the Zoom-FFT-AF combined Doppler algorithm is used to resample the underwater acoustic communication signal to compensate for the Doppler frequency offset; The fuzzy function AF is defined as τ is the signal delay; f d is the Doppler shift; The time function s(t) of the LFM signal is expressed as Where A is the signal amplitude, T is the transmission period of each code element, and f0 represents the carrier frequency; If A=1, the ambiguity function AF of the local LFM signal is: The Doppler factor Δ of the received LFM signal and the local LFM signal is n (Δ0Δ1...Δ M ) is correlated with the signal after the change, and its fuzzy function AF expression can be expressed as: Where r is the received LFM signal.

2. The Doppler estimation and compensation method for high-speed mobile underwater acoustic communication according to claim 1, characterized in that: The step S1 specifically includes: Perform spectrum shifting on the CW signal cw0(n): Move to the center frequency f k At this point, we get the signal cw(n), where is the unit rotation factor; Calculate the sampling ratio D = f s / f k , where f s is the sampling frequency; Calculate the total number of sampling points of the CW signal cw0(n) where N r For analysis points; Through the digital low-pass filter, the signal cw(n) is retained in the frequency band f s / (2D)The frequency band that needs to be refined is obtained by c (n), Where h(n) is the impulse response and the cutoff frequency of the digital low-pass filter is f c ≤f s / (2D); If the sampling frequency is f s / D, for signal y c (n) Take a data point every (D-1) points to obtain the coarse-grained signal rc(m); Perform N-point complex FFT on the coarse-grained signal rc(m), intercept the refined spectrum fft result, move it in the opposite direction, connect it front and back, and convert it into the actual frequency to obtain the refined FFT transformation result F CW (k), the obtained frequency is subtracted from the original frequency to obtain the approximate Doppler frequency shift.

3. The Doppler estimation and compensation method for high-speed mobile underwater acoustic communication according to claim 2, characterized in that: The elimination of the residual Doppler effect after compensation by canceling the residual phase of the code in step S3 specifically includes: performing complex domain code symbol transformation on the PSK code symbol at the transmitting end: in is the bipolar complex domain code element, is the information before transformation, is the intermediate transformation codeword, is the transformed code element, θ is the phase information corresponding to the code element a, is the phase information corresponding to the y code element, α is the phase information corresponding to the x code element, and then Perform spectrum spreading, raised cosine filtering, up-sampling, and up-conversion modulation to obtain a transmit signal; The received underwater acoustic communication signal is demodulated, raised cosine filtered and despread based on the receiving end, and the complex domain code elements before judgment are inversely transformed based on the transmitting end: in To offset the I / Q path pre-decision data once, is the complex domain amplitude information of the transformed code element, β (n) is the corresponding phase information, is the data before I / Q decision after symbol phase cancellation transformation, β' n for Its own phase information, and then Take out the real part and the imaginary part to make a decision and get the code element information.

Images