An Underwater Acoustic Communication Dual-Mode Demodulation Method and System Based on Minimum Frequency Shift Keying

Abstract

The present application provides an underwater acoustic communication dual-mode demodulation method and system based on minimum frequency shift keying. The method includes: receiving an underwater acoustic signal, performing preprocessing, and performing channel estimation on the training sequence in the underwater acoustic signal; calculating a phase change index according to the channel estimation result; when the phase change index is higher than a set bit error rate threshold, processing the received signal using a non-coherent demodulation mode; when the phase change index is lower than the set bit error rate threshold, processing the received signal using a coherent demodulation mode. The advantages of the present application are as follows: it solves the problem of insufficient applicability of a single demodulation mode in different underwater acoustic communication environments; it ensures that the system maintains good communication performance under different channel conditions; it can avoid the problem of information lag, breaks through the limitations of the traditional single demodulation mode, and significantly improves the reliability and robustness of the system in a variable underwater acoustic channel.

Description

Technical Field

[0001] This application belongs to the field of underwater acoustic communication technology, and particularly relates to an underwater acoustic communication dual-mode demodulation method and system based on minimum frequency shift keying. Background Art

[0002] The underwater acoustic channel is a channel with complex temporal, spatial, and frequency variations. Its main characteristics include absorption attenuation, multipath effect, Doppler frequency shift, time-variation, and ambient noise, etc. The drastic fluctuations in time, space, and frequency of underwater acoustic communication pose great difficulties and challenges to underwater acoustic communication. In underwater acoustic communication, modulation and demodulation are the key links to realize underwater information transmission, and optimizing the modulation and demodulation process to reduce the bit error rate is an effective way to improve communication reliability.

[0003] Traditional underwater acoustic communication systems usually adopt a single mode for demodulation, mainly including two methods: coherent demodulation and non-coherent demodulation. Coherent demodulation performs well in the case of obvious multipath effect. Especially after introducing an equalizer with a phase-locked loop (PLL), it can effectively resist strong phase changes. Because coherent demodulation depends on a reference signal synchronized with the phase at the transmitting end, the receiving end needs to accurately know the phase and frequency of the transmitted signal. In an environment with drastic phase changes, the phase-locked loop may be difficult to keep up with the rapid phase changes, resulting in a decline in the performance of coherent demodulation. Non-coherent demodulation does not depend on signal phase information but demodulates by detecting the amplitude or envelope of the signal. Therefore, in an environment with drastic phase changes, non-coherent demodulation has more advantages. However, in the case of low signal-to-noise ratio or large channel delay spread, the performance of non-coherent demodulation will be significantly reduced. Therefore, the demodulation method at the receiving end has an important impact on communication performance. The existing single-mode demodulation methods have their own limitations in a complex underwater acoustic communication environment, and there has not yet appeared an adaptive selection system model that combines coherent and non-coherent demodulation.

[0004] In the field of wireless communication, a variety of feedback-based adaptive modulation methods have been developed. These methods can dynamically adjust the modulation method according to the real-time changes of channel conditions to ensure the effectiveness and stability of communication. However, the application of these technologies in the underwater acoustic communication environment has significant limitations. Since the propagation speed of the underwater acoustic channel is much lower than that of the wireless channel, the channel feedback information usually requires a long transmission time, which leads to the lag of the feedback information, making it difficult for the feedback-based adaptive modulation method to adapt to the rapidly changing underwater acoustic channel environment in real time. Therefore, the traditional feedback mechanism-dependent adaptive modulation method does not work well in underwater acoustic communication and often cannot cope with drastic channel fluctuations and variable ambient noise. Therefore, in the field of underwater acoustic communication, there is an urgent need to develop an adaptive communication system that does not depend on the feedback mechanism and can dynamically adapt to channel changes to better cope with the challenges in the underwater acoustic environment and achieve more efficient and reliable communication. Summary of the Invention

[0005] The purpose of this application is to overcome the defect that it is difficult to ensure both stability and reliability simultaneously under complex and variable underwater acoustic channel conditions with the existing single demodulation mode.

[0006] To achieve the above purpose, this application proposes a dual-mode demodulation method for underwater acoustic communication based on minimum shift keying, including:

[0007] Step 1: Receive the underwater acoustic signal, perform preprocessing, and conduct channel estimation on the training sequence in the underwater acoustic signal.

[0008] Step 2: Calculate the phase change index according to the channel estimation result.

[0009] Step 3: When the phase change index is higher than the set bit error rate threshold, process the received signal using the non-coherent demodulation mode; when the phase change index is lower than the set bit error rate threshold, process the received signal using the coherent demodulation mode.

[0010] As an improvement of the above method, at the transmitting end of the underwater acoustic signal, the MSK transmitted signal consists of a training sequence and an information sequence.

[0011] As an improvement of the above method, the preprocessing includes:

[0012] Perform band-pass filtering on the received underwater acoustic signal to suppress environmental noise and out-of-band interference.

[0013] Dynamically correct the frequency offset caused by underwater acoustic propagation through the Doppler frequency shift compensation algorithm.

[0014] As an improvement of the above method, the calculation of the phase change index includes:

[0015] The calculation formula for the phase change index DI is:

[0016]

[0017] where α represents an adjustable factor; represents the equivalent average relative velocity: N represents the length of the equivalent average relative velocity sequence;

[0018]

[0019] where ν r (n) represents the equivalent relative velocity at the nth moment:

[0020]

[0021] where f c represents the carrier center frequency; v represents the sound speed; represents the equivalent Doppler frequency shift, Δθ represents the change in phase, fs represents the sampling frequency;

[0022] Δν r (n) represents the change rate of the equivalent relative velocity:

[0023] Δν r (n) = ν r (n) - ν r (n - 1)

[0024] represents the average fluctuation of the equivalent relative velocity:

[0025]

[0026] As an improvement of the above method, the coherent demodulation mode processes the received signal, including:

[0027] Step 3a-1: Perform roll-off filtering on the received signal;

[0028] Step 3a-2: Correlate the filtered signal with the equivalent carrier at nT b ≤t≤(n + 2)T b within the time, where n is the serial number of the transmitted symbol, and T b is the reciprocal of the symbol rate. After correlation, the equivalent baseband signal of MSK is obtained;

[0029] Step 3a-3: Perform time-domain decision feedback equalization based on a phase-locked loop on the equivalent baseband signal to obtain the equalized signal;

[0030] Step 3a-4: Multiply the equalized signal by to obtain If and have the same sign, it is judged as 1, otherwise it is judged as 0.

[0031] As an improvement of the above method, the non-coherent demodulation mode processes the received signal, including:

[0032] Step 3b-1: Complete the envelope detection of the preprocessed signal through four base function correlators f1(t), f2(t), f3(t), f4(t); the four base function correlators are:

[0033] f1(t) = cos(2πf1t)

[0034] f2(t) = sin(2πf1t)

[0035] f3(t) = cos(2πf2t)

[0036] f4(t) = sin(2πf2t)

[0037] Among them, f1 and f2 are two equivalent frequency points of MSK, and f1 = f c -1 / 4T b , f2 = f c +1 / 4T b , T b is the reciprocal of the symbol rate; f c represents the carrier center frequency; t represents time;

[0038] At the end of each signal interval, sample the outputs of the four correlators to obtain sample values z1, z2, z3, and z4, and send the sample values to the detector;

[0039] Step 3b-2: The detector makes a detection decision based on the size of the envelope; the envelope is

[0040] This application also provides an underwater acoustic communication dual-mode demodulation system based on minimum shift keying, which is implemented based on the above method. The system includes:

[0041] A channel estimation module, which is used to receive the underwater acoustic signal in the underwater acoustic communication sea area, perform preprocessing, and estimate the channel for the training sequence;

[0042] A calculation phase change index module, which is used to calculate the phase change index according to the channel estimation result;

[0043] A demodulated signal module, which is used to process the received signal in a non-coherent demodulation mode when the phase change index is higher than the set bit error rate threshold; and process the received signal in a coherent demodulation mode when the phase change index is lower than the set bit error rate threshold.

[0044] Compared with the prior art, the advantages of this application are as follows:

[0045] 1) Innovatively propose a dual-mode demodulation scheme: Combining the characteristics of continuous phase modulation, for the first time, introduce a coherent and non-coherent dual-mode demodulation method at the receiving end, effectively combining the advantages of both, and solving the problem of insufficient applicability of a single demodulation mode in different underwater acoustic communication environments;

[0046] 2) Adaptive demodulation based on the phase change index: By proposing the phase change index, realize intelligent demodulation mode selection, as Figure 4 shown, to ensure that the system maintains good communication performance under different channel conditions;

[0047] 3) Dual-mode adaptive demodulation improves the reliability of the system: By implementing dual-mode adaptive demodulation at the receiving end, avoid the problem of information lag, break through the limitations of the traditional single demodulation mode, and significantly improve the reliability and robustness of the system in a changing underwater acoustic channel. Description of the Drawings

[0048] Figure 1 The figure shows a flow chart for selecting a demodulation method;

[0049] Figure 2(a) shows a schematic diagram of the coherent demodulation method;

[0050] Figure 2(b) shows a schematic diagram of the non - coherent demodulation method;

[0051] Figure 3(a) shows a reference diagram of the bit error rate and phase change index for coherent demodulation;

[0052] Figure 3(b) shows a reference diagram of the bit error rate and phase change index for non - coherent demodulation;

[0053] Figure 4 The figure shows a simulation performance comparison diagram of bit error rate - signal - to - noise ratio;

[0054] Figure 5 The figure shows a static channel impulse response diagram;

[0055] Figure 6(a) shows a variation diagram with a relative speed range of - 0.05 to 0.15 knots;

[0056] Figure 6(b) shows a variation diagram with a relative speed range of - 0.25 to 0.1 knots. Detailed implementation manners

[0057] The technical solutions of the present application will be described in detail below with reference to the accompanying drawings.

[0058] The present invention proposes an underwater acoustic communication dual - mode demodulation method and system based on minimum shift keying. This method dynamically evaluates the channel state and adopts an adaptive switching mechanism to cope with various channel environments and improve the reliability of communication.

[0059] In the present invention, to achieve dual - mode demodulation, we introduce minimum shift keying (MSK) as the core modulation technology to meet the requirements of reliable communication in complex underwater acoustic environments. MSK signals have the characteristics of both phase - modulated signals and frequency - modulated signals, enabling both coherent and non - coherent demodulation modes at the receiving end.

[0060] The object of the present invention is achieved through the following technical solutions, and its overall architecture is as Figure 1As shown in the figure. Through the dual-mode adaptive demodulation method, the present invention innovatively combines MSK modulation with two demodulation modes, coherent and non-coherent, and realizes the dynamic evaluation of the channel phase change and the adaptive demodulation selection for complex underwater acoustic channels. By introducing a phase change metric index and using high-order statistical analysis methods, the phase jitter is modeled and quantified to comprehensively characterize the phase perturbation degree of the channel. Based on the phase change index, a threshold decision system based on the expected bit error rate (BER) is designed at the receiving end. This system determines the optimal threshold value of the phase change metric value through the non-linear mapping model between the phase change index and the BER, realizing the adaptive adjustment in the demodulation mode selection, so that the system can automatically switch to the optimal demodulation mode in different channel environments.

[0061] The dual-mode demodulation method for underwater acoustic communication based on minimum shift keying includes:

[0062] (1) Signal reception and preprocessing

[0063] At the transmitting end, the MSK transmitted signal x(n) consists of the training sequence x p (n) and the information sequence x m (n). After passing through the underwater acoustic channel, the underwater acoustic MSK signal at the receiving end can be expressed as:

[0064] y(n) = x(n) * h(n) + w(n) (1)

[0065] where * represents convolution, and x(n), y(n), and w(n) represent the transmitted signal, the received signal, and the additive white Gaussian noise respectively, and h(n) represents the underwater acoustic channel impulse response.

[0066] For the received MSK signal, multi-stage preprocessing operations are performed. First, the signal is band-pass filtered to suppress environmental noise and out-of-band interference and ensure the spectral purity of the signal; in addition, the Doppler frequency shift compensation algorithm is used to dynamically correct the frequency shift caused by underwater acoustic propagation, thereby enhancing the time consistency of the signal. Finally, the preprocessed training sequence y p (n) is used for channel estimation, and common channel estimation methods such as the least mean square algorithm (LMS), the recursive least squares algorithm (RLS), etc. can be used.

[0067] (2) Calculation of phase change index

[0068] For the obtained channel estimation result, take the phase of its main arrival path and represent the phase as θ. An effective method for calculating the phase change index is proposed for the phase change to effectively characterize the phase perturbation of the channel. First, the equivalent Doppler frequency shift is defined as the derivative of the phase with respect to time:

[0069]

[0070] where Δθ is the change in phase, and f dopp is the equivalent Doppler frequency shift. Then, considering the influence of the equivalent Doppler frequency shift, the equivalent relative velocity between the transceiver can be expressed as:

[0071]

[0072] where f c is the carrier center frequency, and v is the speed of sound, generally taken as 1500 m / s. Thus, the calculation formula for the equivalent average relative velocity is:

[0073]

[0074] Furthermore, the change rate of the equivalent relative velocity is defined, that is, the difference between the equivalent relative velocity at a certain moment and the equivalent relative velocity at the previous moment. The formula is:

[0075] Δν r (n) = ν r (n) - ν r (n - 1) (5)

[0076] According to the calculation formula of volatility, the average fluctuation of the equivalent relative velocity is expressed as:

[0077]

[0078] The index for measuring the phase change is defined as:

[0079]

[0080] where N is the length of the equivalent average relative velocity sequence, and α is an adjustable factor. The phase change index consists of two parts: the mean value of the equivalent relative velocity after Doppler compensation and the standard deviation of the equivalent relative velocity fluctuation. One is the mean value of the equivalent relative velocity after Doppler compensation, which represents the average offset of the equivalent Doppler shift and reflects the numerical value of the equivalent Doppler; the other is the standard deviation of the equivalent relative velocity fluctuation, which characterizes the speed of the phase change. In practical applications, the value of α can be taken as 0.1.

[0081] (3) Demodulation method selection

[0082] Step 1) Select several historical channels in this experimental sea area, obtain the correlation data between the coherent demodulation bit error rate and the phase change index under the conditions of the experimental sea area, and draw the corresponding scatter plots as shown in Figures 3(a) and 3(b). Under the condition of a fixed signal-to-noise ratio, the coherent demodulation bit error rate and the phase change index show a highly positive correlation, with the correlation coefficient reaching above 0.9, while the non-coherent demodulation bit error rate and the phase change index are almost uncorrelated, with the correlation coefficient less than 0.1.

[0083] Step 2) By setting a specific bit error rate threshold, the corresponding phase change index threshold values under different signal-to-noise ratio conditions are extracted from FIGS. 3(a) and 3(b). In the actual received signal processing process, the system first obtains the phase change of the main arrival path by using the channel estimation result obtained from the training sequence. Through the method in step "(2) Phase change index calculation", the phase change index value in the current environment is calculated.

[0084] Step 3) Compare the phase change index value with the threshold value corresponding to the preset expected bit error rate: when the phase change index exceeds the threshold value, the system will automatically switch to the non-coherent demodulation mode to reduce the demodulation error caused by phase fluctuations; conversely, when the phase change index is lower than the threshold value, the coherent demodulation mode is selected to make full use of the performance advantages of coherent demodulation under low phase perturbation conditions.

[0085] This demodulation mode selection strategy based on the phase change index ensures that the system can adaptively select the optimal demodulation scheme in a complex ocean channel environment, thereby maximizing the overall performance and stability of the communication system.

[0086] (4) Demodulation

[0087] When the coherent demodulation mode is selected, the received signal is processed as follows (as shown in FIG. 2(a)):

[0088] Step 1) Matched filtering: Since the signal waveform at the transmitter is roll-off filtered, the received signal at the receiver also needs to be subjected to corresponding matched filtering. The received signal is processed by a matched filter to maximize the signal-to-noise ratio, and the filtered signal r(t) is obtained;

[0089] Step 2) Correlation demodulation: The filtered signal r(t) is correlated with the equivalent carrier in the time interval from nT b ≤t≤(n + 2)T b where n is the serial number of the transmitted symbol and T b is the reciprocal of the symbol rate. After correlation, the equivalent baseband signal r n of MSK is obtained;

[0090] Step 3) Equalization: To compensate for channel fading and multipath effects, the equivalent baseband signal is subjected to time-domain decision feedback equalization based on a phase-locked loop to obtain the equalized signal

[0091] Step 4) Detection: Multiply the output of the equalizer by to obtain If and have the same sign, the decision is 1; otherwise, it is 0.

[0092] When the non - coherent demodulation mode is selected, the received signal is processed through the following steps (as shown in Fig. 2(b)):

[0093] Step 1) Envelope detection: The envelope detection of the filtered signal r(t) can be completed by four basis - function correlators f1(t), f2(t), f3(t), and f4(t). The correlators are written as:

[0094]

[0095] where f1 and f2 are two equivalent frequency points of MSK, f1 = f c -1 / 4T b and f2 = f c +1 / 4T b ; t represents time.

[0096] At the end of each signal interval, the outputs of the four correlators are sampled to obtain sample values z1, z2, z3, and z4, and the sample values are sent to the detector.

[0097] Step 2) Detection: Define the envelope as The detector makes a detection decision based on the size of the envelope.

[0098] The present invention provides an effective criterion through the phase - change index for adaptively selecting coherent or non - coherent demodulation modes under different channel conditions. Through real - time channel evaluation, the demodulation mode can be dynamically optimized without relying on check information, thereby significantly improving the transmission reliability of the underwater acoustic communication system. Compared with the traditional single - demodulation method, the technical solution of the present invention significantly enhances the robustness and communication performance of the system.

[0099] As Figure 1 shown, Embodiment 1 of the present invention proposes a dual - mode demodulation method for the receiving end based on MSK. This embodiment uses an MSK underwater acoustic communication system as the application background, and verifies the effectiveness of the present invention through simulation. It includes the following steps:

[0100] Step 1, Select a typical shallow - sea static channel as the simulation channel as Figure 5 shown, and use the obtained channel to simulate the underwater acoustic MSK communication process. Specifically, the simulation parameters are: the transmitted signal is a minimum - shift - keying (MSK) signal, the symbol rate is set to 10 bps, the center frequency of the carrier is 450 Hz, the total number of transmitted symbols is 500, the duration of the MSK signal is 50 seconds, the length of the training sequence is 10 seconds, and the total transmission duration of the entire signal is 60 seconds.

[0101] Step 2, add phase changes in the simulation. Since the derivative of the phase is proportional to the equivalent relative velocity, the phase changes are added in the form of equivalent Doppler.

[0102] Specifically, decompose the equivalent relative velocity into an equivalent average velocity and an increment relative to this average velocity. Based on the continuity of velocity, the instantaneous velocity at any given moment can be expressed as:

[0103]

[0104] where ν(n) represents the velocity at the current moment, is the average velocity, Δv max is the maximum fluctuation amplitude of the relative average velocity, and μ is a random number ranging from -0.5 to 0.5. The average relative velocity ranges from 0 to 0.15 m / s, and the absolute value of the relative amplitude ranges from 0 to 3 m / s. In the simulation, the bit signal-to-noise ratio is set to 10 dB, and the relationship between the bit signal-to-noise ratio and the signal-to-noise ratio is:

[0105]

[0106] where, is the bit signal-to-noise ratio, R b is the symbol rate, R s is the baud rate, and f s is the sampling frequency.

[0107] Step 3, calculate the value of the phase change index using the equivalent relative velocity obtained from the simulation in Step 2. Two typical relative velocity changes are shown in Figures 6(a) and 6(b). In Figure 6(a), the relative velocity ranges from -0.05 to 0.15 knots, with weak volatility, and the value of the phase change index is 0.0132. In Figure 6(b), the relative velocity ranges from -0.25 to 0.1 knots, with strong volatility, and the value of the phase change index is 0.1788. According to the implementation process of the dual-mode adaptive demodulation system at the receiving end, assuming that the desired demodulation bit error rate is less than 0.01, when the bit signal-to-noise ratio is 10 dB, the phase change index threshold is 0.05. The phase change index value in Figure 6(a) is less than the threshold, so coherent demodulation is selected; the phase change index value in Figure 6(b) is greater than the threshold, so non-coherent demodulation is selected.

[0108] Step 4: Coherently and non-coherently demodulate the MSK signals with the two equivalent relative velocities in Fig. 6(a) and Fig. 6(b). When the bit signal-to-noise ratio is 10 dB, after adding the equivalent relative velocity in Fig. 6(a), the bit error rate of coherent demodulation is 0.0040, and the bit error rate of non-coherent demodulation is 0.0200; after adding the equivalent relative velocity in Fig. 6(b), the bit error rate of coherent demodulation is 0.4500, and the bit error rate of non-coherent demodulation is 0.0200. The demodulation results also verify from the side that coherent demodulation is greatly affected by phase changes, while non-coherent demodulation is hardly affected by phase changes. From the demodulation results, it can be seen that the proposed dual-mode demodulation method for the receiving end of underwater acoustic communication based on MSK can achieve better communication performance in different underwater acoustic communication environments.

[0109] This application also provides a dual-mode demodulation system for underwater acoustic communication based on minimum shift keying, which is implemented based on the above method. The system includes:

[0110] A channel estimation module, which is used to receive the underwater acoustic signals in the underwater acoustic communication sea area, perform preprocessing, and estimate the channel for the training sequence;

[0111] A module for calculating the phase change index, which is used to calculate the phase change index according to the channel estimation result;

[0112] A demodulated signal module, which is used to process the received signal in the non-coherent demodulation mode when the phase change index is higher than the set bit error rate threshold; and process the received signal in the coherent demodulation mode when the phase change index is lower than the set bit error rate threshold.

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application and not to limit them. Although this application has been described in detail with reference to the embodiments, those of ordinary skill in the art should understand that any modification or equivalent replacement of the technical solutions of this application does not depart from the spirit and scope of the technical solutions of this application, and they should all be covered by the scope of the claims of this application.

Claims

1. A dual-mode demodulation method for underwater acoustic communication based on minimum frequency shift keying, comprising: Step 1: Receive the underwater acoustic signal, perform preprocessing, and perform channel estimation on the training sequence in the underwater acoustic signal; Step 2: Calculate the phase change index based on the channel estimation result; Step 3: When the phase change index is higher than the set bit error rate threshold, the received signal is processed in a non-coherent demodulation mode; When the phase change index is lower than the set bit error rate threshold, the coherent demodulation mode is used to process the received signal; The calculating of the phase change index comprises: The calculation formula of phase change index DI is: Among them, α represents the adjustable factor; represents the equivalent average relative velocity: N represents the length of the equivalent average relative velocity sequence; Among them, ν r (n) represents the equivalent relative speed at the nth moment: Among them, f c represents the carrier center frequency; v represents the speed of sound; represents the equivalent Doppler frequency shift, Δθ represents the change in phase, and f s Indicates the sampling frequency; Δν r (n) represents the rate of change of the equivalent relative velocity: Dn r (n)=n r (n)-n r (n-1) Express the average fluctuation of the equivalent relative velocity:

2. The dual-mode demodulation method for underwater acoustic communication based on minimum frequency shift keying according to claim 1 is characterized in that: At the underwater acoustic signal transmitting end, the MSK transmitting signal consists of a training sequence and an information sequence.

3. The dual-mode demodulation method for underwater acoustic communication based on minimum frequency shift keying according to claim 1 is characterized in that: The pre-processing comprises: Perform bandpass filtering on the received underwater acoustic signal to suppress environmental noise and out-of-band interference; The frequency offset caused by underwater acoustic propagation is dynamically corrected through the Doppler shift compensation algorithm.

4. The dual-mode demodulation method for underwater acoustic communication based on minimum frequency shift keying according to claim 1 is characterized in that: The coherent demodulation mode processes the received signal, including: Step 3a-1: performing roll-off filtering on the received signal; Step 3a-2: Compare the filtered signal with the equivalent carrier In nT b ≤t≤(n+2)T b time, where n is the serial number of the transmitted symbol, T b is the inverse of the symbol rate, and the equivalent baseband signal of MSK is obtained after correlation; Step 3a-3: performing time-domain decision feedback equalization based on a phase-locked loop on the equivalent baseband signal to obtain an equalized signal; Step 3a-4: Compare the equalized signal with Multiply to get like and If they have the same sign, the judgment is 1, otherwise it is 0.

5. The dual-mode demodulation method for underwater acoustic communication based on minimum frequency shift keying according to claim 1 is characterized in that: The non-coherent demodulation mode processes the received signal, including: Step 3b-1: Complete envelope detection of the preprocessed signal through four basis function correlators f1(t), f2(t), f3(t), and f4(t); the four basis function correlators are: f1(t)=cos(2πf1t) f2(t)=sin(2πf1t) f3(t)=cos(2πf2t) f4(t)=sin(2πf2t) Among them, f1 and f2 are two equivalent frequency points of MSK, f1 = f c -1 / 4T b , f2=f c +1 / 4T b , T b is the inverse of the symbol rate; f c represents the carrier center frequency; t represents time; At the end of each signal interval, the outputs of the four correlators are sampled to obtain sample values ​​z1, z2, z3, and z4, and the sample values ​​are sent to the detector; Step 3b-2: The detector makes a detection decision based on the size of the envelope; the envelope is 6. A dual-mode demodulation system for underwater acoustic communication based on minimum frequency shift keying, implemented based on any method described in claims 1-5, characterized in that: The system comprises: A channel estimation module is used to receive the underwater acoustic signal of the underwater acoustic communication sea area, perform preprocessing, and perform channel estimation on the training sequence; A phase change index calculation module, used to calculate the phase change index according to the channel estimation result; and The demodulation signal module is used to process the received signal in a non-coherent demodulation mode when the phase change index is higher than the set bit error rate threshold; when the phase change index is lower than the set bit error rate threshold, the coherent demodulation mode is used to process the received signal.

Images