A multi-carrier underwater communication method based on pulse stretching technology

Abstract

The application discloses a multi-carrier underwater communication method based on pulse stretching technology. Information to be transmitted is converted into a binary array with 8 bits as a byte by an ASCII encoding unit, and the binary array is changed into a new array with 10 bits as a byte after passing through an 8 / 10B encoding unit, wherein the number of 0 and 1 is similar. After OOK modulation of the multi-carrier of a modulation unit, a stretching coefficient k is solved by a Gaussian envelope approximation and parameter solving unit, the modulated signal is stretched according to the stretching coefficient k in a stretching unit, the signal is transmitted in water and received by a receiving end, and after demodulation of a demodulation unit and 10B / 8B encoding and ASCII decoding, the transmitted information is restored. The obtained signal has a low peak-to-average ratio, which is beneficial to long-distance transmission under water.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of communication, and relates to a multi-carrier underwater communication method based on pulse stretching technology. BACKGROUND

[0002] It is very difficult to carry out wireless communication under water, and the key reason is that the commonly used communication frequencies are in the absorption range of water molecules. Water is a polar molecule, and the positive and negative charge centers thereof do not coincide, so that the water molecules can change the direction under the action of the electric field in the electromagnetic wave. Once the frequency of the electromagnetic wave is close to the inherent vibration frequency of the water molecules, the water molecules can absorb most of the energy in the electromagnetic wave and convert it into their own vibration. The commonly used communication electromagnetic wave frequencies are almost in the absorption frequency range of the water molecules, resulting in very fast attenuation of the electromagnetic wave in water. In addition, the underwater noise, multipath effect and complex underwater environment and other factors also have adverse effects on the reliability of underwater wireless communication. Therefore, for the fields of ocean exploration, ocean resource development and military reconnaissance, how to solve the problem of underwater communication has always been a difficult problem.

[0003] The acoustic signal is the most suitable signal for transmission under water, and the signal is converted into an acoustic wave by a receiving transducer on the surface of the water, and the acoustic wave is converted into an electric signal by a transducer on the other side to complete the communication task. However, even the acoustic signal most suitable for transmission under water, the high-frequency acoustic signal still has serious attenuation in water.

[0004] Due to the serious attenuation of the high-frequency acoustic signal, the usable bandwidth of the underwater acoustic communication is extremely narrow, and in order to take into account the transmission rate, a multi-carrier modulation mode is often used to fully utilize the bandwidth.

[0005] When the power of the underwater acoustic transducer is constant, the peak-to-average ratio is an important index limiting the communication distance of the multi-carrier. When the peak-to-average ratio is too high, the underwater acoustic transducer works in a low-power state for a long time, and only reaches a high power in the moment before and after the peak value. In order to ensure that the signal is not distorted, the rated power of the underwater acoustic transducer must be higher than the peak power appearing for a short time, which greatly limits the working state of the underwater acoustic transducer. SUMMARY

[0006] The purpose of the present application is to provide a multi-carrier underwater communication method based on pulse stretching technology, which greatly reduces the peak-to-average ratio of the signal by using the pulse stretching mode, and makes the main power of the signal in a single transmission period concentrated in the center position of the period, thereby reducing the influence of underwater multipath transmission.

[0007] In order to achieve the above purpose, the technical solution adopted by the present application is:

[0008] A multi-carrier underwater communication method based on pulse stretching technology, comprising a multi-carrier underwater communication system, the system comprising an ASCII encoding unit, an 8B / 10B encoding unit, a modulation unit, a Gaussian envelope approximation and parameter solving unit, a stretching unit, a demodulation unit, a 10B / 8B encoding unit, and an ASCII decoding unit; information to be transmitted is converted into a binary array of 8 bits per byte by the ASCII encoding unit, and after passing through the 8 / 10B encoding unit, the binary array becomes a new array of 10 bits per byte with similar occurrences of 0 and 1, and after OOK modulation of the new array by the multi-carrier modulation unit, the stretching coefficient k is solved by the Gaussian envelope approximation and parameter solving unit, and in the stretching unit, the modulated signal is stretched according to the stretching coefficient k, the signal is transmitted in water and received by the receiving end, and after demodulation by the demodulation unit and 10B / 8B encoding and ASCII decoding, the transmitted information is restored.

[0009] Preferably, the data peak equalization module is composed of the ASCII encoding unit, the 8B / 10B encoding unit, and the modulation unit.

[0010] The 8-bit data of the binary array will become a 10-bit array after 8b / 10b encoding, and the number of 0 and 1 in the 10-bit data appears in three cases: 5 0s and 5 1s, 6 0s and 1, and 4 0s and 6 1s, and when the non-uniformity of the previous byte is -2, the non-uniformity of the next byte will be adjusted to +2, thereby ensuring that the number of 0 and 1 is basically equal in a long time.

[0011] Preferably, the signal s(t) of a single transmission period after multi-carrier OOK modulation is as shown in formula (1):

[0012]

[0013] Wherein M is a row matrix composed of 0 and 1 after 8 / 10B encoding of the data, n is the number of carriers, f is the carrier frequency, t is the time, and T is the length of a single transmission period.

[0014] When the amplitude and initial phase of each carrier are the same, s(0) in s(t) is the pulse peak value of a single transmission period, and if 8b / 10b encoding is not performed, the value range of s(0) is 0-n.

[0015] After 8b / 10b encoding, the value range of s(0) is 2 / n-1-2 / n+1.

[0016] Preferably, the peak-to-average ratio optimization module is composed of the Gaussian envelope approximation and parameter solving unit and the stretching unit.

[0017] After peak equalization after passing through the ASCII encoding unit, 8B / 10B encoding unit, modulation unit, the peak-to-average ratio of a single cycle transmission is reduced by the peak-to-average ratio optimization module, and the reduction of the peak-to-average ratio is adjusted by adjusting the phase;

[0018] When adjusting the parameters of the Gaussian envelope approximation and the parameter solving unit, as shown in formula (2), the signal is made to disperse by adding a phase shift term in the frequency domain after Fourier transform: Wherein, p(f) is a second-order polynomial with f as the independent variable.

[0019] Preferably, since the signal form of the superposition of multiple single-frequency signals is a plurality of impact functions after Fourier transform, it is difficult to solve the optimal phase shift parameter after multiplying the phase shift term. In order to solve the above problem, the original signal also needs to be approximated by a Gaussian envelope in the peak-to-average ratio optimization module; Since the fitting degree of the Gaussian envelope approximation is high near the peak point of the signal, and the peak-to-average ratio is concerned about the peak point, the expression of the Gaussian approximation is shown in formula (3), wherein σ is the Gaussian envelope width, f0 is the center frequency of the signal band, and the optimal phase shift parameter and the phase shift term expression can be obtained by the expression after inverse Fourier transform of the expression after Fourier transform and multiplication of the phase shift term of the Gaussian approximation:

[0020]

[0021] Then, the expression of the Gaussian envelope approximation signal after Fourier transform is shown in formula (4):

[0022]

[0023] Then, the expression of the Gaussian envelope multiplied by the phase shift term is shown in formula (5);

[0024]

[0025] Since p(f) is a quadratic polynomial, the expression after inverse Fourier transform of formula (5) is shown in formula (6), wherein

[0026]

[0027] As can be seen from formula (6), the expression is still in the form of Gaussian envelope multiplied by oscillation term, and the peak value becomes The symmetry axis becomes t=p'(f0), and the frequency of the oscillation term becomes f0-p(f0).

[0028] Preferably, after obtaining the expression after inverse Fourier transform according to (6), in order to make the peak value as low as possible, the symmetry axis is unchanged, the peak value is still at s(0), and the oscillation term frequency is unchanged, and σ is reduced in a single cycle.p <T, according to formula (2) and formula (3), the result shown in formula (7) can be obtained:

[0029] p(f0)=0

[0030] p'(f0)=0 (5)

[0031] Therefore, the expression of p(f) is as shown in formula (8): p(f) = kf 2 -2kf0f(8), where k is the coefficient to be determined; by combining equations (6) and (8), the peak expression after inverse Fourier transform is shown in equation (9):

[0032]

[0033] Combining equations (4) and (9), we can get the condition for the value of k as shown in equation (10):

[0034]

[0035] When the k value increases, the peak value decreases rapidly, and when the k value increases to the peak value, it slowly decreases.

[0036] Preferably, select σ g = 0.8T, so that the data is concentrated in the middle 80% of a single communication cycle, avoiding inter-symbol interference when no idle cycle is used;

[0037] The signal after peak-to-average ratio optimization is shown in formula (7):

[0038]

[0039] The beneficial effects of the present invention are:

[0040] The multi-carrier underwater communication method based on pulse stretching technology in the present invention adopts 8 / 10B coding to reduce the signal peak difference caused by different transmission information in a single cycle equalization. The pulse stretching method greatly reduces the signal's average peak ratio and makes the main power of the signal concentrated at the center of the cycle within a single transmission cycle, thereby reducing the impact of underwater multipath transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a flowchart of a multi-carrier underwater communication method based on pulse stretching technology.

[0042] Figure 2 It is the relationship between the broadening coefficient k value and the peak value.

[0043] Figure 3 This is a flowchart of Example 4 of a multi-carrier underwater communication method based on pulse stretching technology.

[0044] Figure 4 is the signal after 8B / 10B coding and modulation and the comparison of the spread signal. DETAILED DESCRIPTION

[0045] The application will be described in detail below with reference to the accompanying drawings:

[0046] Embodiment 1

[0047] In combination Figures 1 to 4 A multi-carrier underwater communication method based on pulse spreading technology, comprising a multi-carrier underwater communication system, the system comprising an ASCII encoding unit, an 8B / 10B encoding unit, a modulation unit, a Gaussian envelope approximation and parameter solving unit, a spreading unit, a demodulation unit, a 10B / 8B encoding unit, and an ASCII decoding unit; the peak value equalization module of the data composed of the ASCII encoding unit, the 8B / 10B encoding unit, and the modulation unit; and the peak-to-average ratio optimization module composed of the Gaussian envelope approximation and parameter solving unit and the spreading unit.

[0048] Embodiment 2

[0049] On the basis of the above embodiment disclosure, the present embodiment further discloses the following:

[0050] When the multi-carrier underwater communication system communicates, the communication method is as follows:

[0051] The information to be transmitted is converted into a binary array of 8 bits per byte by the ASCII encoding unit, and after passing through the 8 / 10B encoding unit, it becomes a new array of 10 bits per byte with similar occurrences of 0 and 1. After OOK modulation of the multi-carrier by the modulation unit, the spreading coefficient k is solved by the Gaussian envelope approximation and parameter solving unit, and the modulated signal is spread according to the spreading coefficient k in the spreading unit. The signal is transmitted in water and received by the receiving end, and after demodulation by the demodulation unit and 10B / 8B encoding unit and ASCII decoding unit, the transmitted information is restored.

[0052] The 8-bit data of the binary array will become a 10-bit array after 8b / 10b coding. There are three cases of the number of 0 and 1 bits in the 10-bit data: 5 0s and 5 1s, 6 0s and 1, and 4 0s and 6 1s. And when the non-uniformity of the previous byte is -2, the non-uniformity of the next byte will be adjusted to +2, so as to ensure that the number of 0 and 1 is basically equal in a long time.

[0053] The signal s(t) of a single transmission period after multi-carrier OOK modulation is shown in formula (1):

[0054]

[0055] Wherein M is a row matrix composed of 0 and 1 after the data is encoded by 8 / 10B, n is the number of carriers, f is the carrier frequency, t is time, T is the length of a single transmission period; the s(0) in s(t) is the pulse peak value of a single transmission period when the amplitude and initial phase of each carrier are the same, and the value range of s(0) is 0~n if no 8b / 10b encoding is performed; after 8b / 10b encoding, the value range of s(0) is 2 / n-1~2 / n+1.

[0056] After the peak value is equalized through the ASCII encoding unit, the 8B / 10B encoding unit and the modulation unit, the peak-to-average ratio of a single period transmission is reduced through the peak-to-average ratio optimization module, and the reduction of the peak-to-average ratio is adjusted by adjusting the phase.

[0057] Embodiment 3

[0058] Based on the disclosure of the above embodiments, the present embodiment further discloses the following:

[0059] When adjusting the parameters of the Gaussian envelope approximation and the parameter solving unit, as shown in formula (2), the signal is made to disperse by adding a phase shift term in the frequency domain after Fourier transform: Wherein p(f) is a second-order polynomial with f as the independent variable.

[0060] The signal form superimposed by a plurality of single-frequency signals is a plurality of impulse functions after Fourier transform, and it is difficult to solve the optimal phase shift parameter after multiplying the phase shift term. In order to solve the above problems, the original signal also needs to be approximated by a Gaussian envelope in the peak-to-average ratio optimization module. Since the fitting degree of the Gaussian envelope approximation is high near the peak value point of the signal, and the peak-to-average ratio is concerned about the peak value point, the expression of the Gaussian approximation is shown in formula (3), wherein σ is the Gaussian envelope width, f0 is the center frequency of the signal band, and the optimal phase shift parameter and the phase shift term expression can be solved through the expression after inverse Fourier transform of the expression of the Gaussian approximation after Fourier transform and multiplying the phase shift term:

[0061]

[0062] Then, the expression of the Gaussian envelope approximation signal after Fourier transform is shown in formula (4):

[0063]

[0064] Then, the expression of the Gaussian envelope multiplied by the phase shift term is shown in formula (5);

[0065]

[0066] Since p(f) is a quadratic polynomial, the expression of the inverse Fourier transform of equation (5) is shown in equation (6), where

[0067]

[0068] As can be seen from equation (6), the expression is still in the form of a Gaussian envelope multiplied by an oscillation term, and the peak value becomes The symmetry axis becomes t = p'(f0), and the frequency of the oscillation term becomes f0-p(f0).

[0069] After obtaining the expression of the inverse Fourier transform according to equation (6), in order to further reduce the peak value as much as possible, the symmetry axis is unchanged, the peak value is still at s(0), and the frequency of the oscillation term is unchanged, and in a single period, σ p According to equations (2) and (3), the result shown in equation (7) can be obtained:

[0070] p(f0) = 0

[0071] p'(f0) = 0 (9)

[0072] Therefore, the expression of p(f) is shown in equation (8): p(f) = kf 2 -2kf0f (8), where k is an undetermined coefficient; by combining equation (6) and equation (8), the expression of the peak value after the inverse Fourier transform is shown in equation (9):

[0073]

[0074] By combining equation (4) and equation (9), the condition for the value of k is shown in equation (10):

[0075]

[0076] When the value of k increases, the peak value rapidly decreases, and when the value of k increases to the peak value, the peak value slowly decreases.

[0077] Select the value of k to make σ g = 0.8T, so that the data is concentrated in the middle 80% position of a single communication period, and the inter-symbol interference is avoided without using a null cycle;

[0078] The signal optimized by the peak-to-average ratio is shown in equation (11):

[0079]

[0080] Since the optimal working frequency band of the underwater acoustic transducer used in the present study is 500-700 Hz, the center frequency is taken as 600 Hz, the frequency interval is 1 Hz, and the period is T as an example. After the data "Tianjin university!!" is encoded by 8B / 10B, the modulated signal is as shown in the black curve in FIG. 1. Figure 4 The signal after the spreading by the spreading factor k is as shown in the gray curve in FIG. 2. As can be seen from the figure, the peak-to-peak value of the signal after the spreading is obviously reduced, and the part with strong energy is concentrated in the middle part of the transmission period. The peak-to-average ratio is reduced from 23.01 dB to 11.75 dB, which is beneficial to the long-distance transmission of the signal under the condition that the rated power is unchanged. Figure 4

[0081] Example 4

[0082] In the multi-carrier underwater communication, a multi-carrier underwater communication method based on the pulse spreading technology is provided. The unit procedures required in the method include ASCII encoding, 8B / 10B encoding, modulation, Gaussian envelope approximation and parameter solving, spreading, demodulation, 10B / 8B encoding, and ASCII decoding. In the communication method, the information to be transmitted is converted into an 8-bit binary array by ASCII encoding; the array is converted into a new array of 10 bits per byte by 8 / 10B encoding, in which the number of 0 and 1 is similar; the modulated multi-carrier is modulated by OOK, and the spreading factor k is solved by Gaussian envelope approximation and parameter solving; in the spreading, the modulated signal is spread according to the spreading factor k; the signal is transmitted in water and received by the receiving end, and the transmitted information is restored after demodulation, 610B / 8B encoding, and ASCII decoding.

[0083] As shown in formula 001, the information to be transmitted is converted into ASCII code, and becomes 8-bit data per byte.

[0084] x[1], x[2], x[3]...x[8]; 001;

[0085] In the 8B / 10B encoding process, the first three bits x[1], x[2], x[3] are converted into four-bit data shown in formula 002 by 3B / 4B encoding through the mapping relationship of Table 1. The last five bits of data x[4], x[5], x[6], x[7], x[8] are converted into six-bit data shown in formula 003 by 5B / 6B encoding through the mapping relationship of Table 2. In the mapping process, the mapping mode of RD=-1 with fewer bits with a value of 1 is always selected.

[0086] X[7], X[8], X[9], X

[10] ; 002;

[0087] ​X[1], X[2], X[3], X[4], X[5], X[6]; 003;

[0088] Table 1 3B / 4B encoding mapping table:

[0089]

[0090] Table 2 5B / 6B encoding mapping table:

[0091]

[0092] The 4B and 6B data after encoding 6B in front 4B in back re-spliced into 10-bit data, complete 8B / 10B encoding. The data after re-splicing is shown in formula 004.

[0093] X[1], X[2], X[3], X[4], X[5], X[6], X[7], X[8], X[9], X

[10] 004

[0094] The signal s(t) of a single transmission cycle after multi-carrier OOK modulation is shown in formula 005; wherein M is a row matrix composed of 0 and 1 after 8 / 10B encoding of data, n is the number of carriers, f is the carrier frequency, t is time, and T is the length of a single transmission cycle.

[0095]

[0096] When the amplitude and initial phase of each carrier are the same, s(0) is the pulse peak value of a single transmission cycle. If 8b / 10b encoding is not performed, the value range of s(0) is 0-n, with a large floating. After 8b / 10b encoding, the value range of s(0) is 2 / n-1-2 / n+1, with a very small floating. Although this slightly reduces the effective information transmitted by each transmission cycle, the maximum value of different information peaks is reduced from n to n / 2+1, which is more conducive to long-distance transmission.

[0097] Although the peak value in multi-carrier OOK modulation can be effectively reduced by 8b / 10b encoding, the peak-to-average ratio is still large, and the transducer still works in a low power state for a long time. Therefore, other means are still needed to reduce the peak-to-average ratio of a single cycle transmission.

[0098] In multi-carrier OOK modulation, the frequency of the signal actually plays a role in transmitting information. The phase of each carrier does not affect the orthogonal relationship between signals or the intensity of each carrier. Therefore, it is feasible to reduce the peak-to-average ratio by adjusting the phase.

[0099] Reference the pulse stretching technique in pulsed laser. As shown in equation 006, the signal can be made "dispersive" by adding a phase shift term in the frequency domain after Fourier transform. Where p(f) is a second order polynomial with f as the independent variable, since the signal form of multiple single frequency signals superimposed is a plurality of impulse functions after Fourier transform. After multiplied by the phase shift term, it is difficult to solve the optimal phase shift parameter. Where the following results are obtained:

[0100] In order to solve the above problems, the original signal can be approximated by a Gaussian envelope. As can be seen, the Gaussian envelope approximation has a high fitting degree near the peak point of the signal, and the peak-to-average ratio is concerned just at the peak point. Where σ is the Gaussian envelope width, and f0 is the center frequency of the signal band; Fourier transform the expression after Gaussian approximation and multiply by the phase shift term. The optimal phase shift parameter and the phase shift term expression can be obtained by the expression after inverse Fourier transform:

[0101]

[0102] The expression after Fourier transform of the Gaussian envelope approximation signal is shown in the following equation:

[0103]

[0104] The expression of the Gaussian envelope multiplied by the phase shift term is shown in the following equation:

[0105]

[0106] Since p(f) is a second order polynomial, the expression after inverse Fourier transform of equation 009 is shown in equation 100;

[0107]

[0108] Where As can be seen in equation 101, the expression is still in the form of Gaussian envelope multiplied by oscillation term; its peak value becomes The symmetry axis becomes t=p'(f0), and the frequency of the oscillation term becomes f0-p(f0).

[0109] Example 5

[0110] After transformation, we hope: 1) the peak value is as low as possible; 2) the symmetry axis is unchanged, and the peak value is still at s(0); 3) the oscillation term frequency is unchanged; 4) since the approximation discussed is within a single period, σ p <T.

[0111] According to the symmetry axis unchanged, the peak value is still at s(0), and the oscillation term frequency is unchanged, the result shown in equation 101 is obtained.

[0112] p(f0)=0

[0113] p'(f0)=0;101;

[0114] Thus, the expression of p(f) is shown in formula 102: p(f) = kf 2 -2kf0f; 102, where k is the coefficient to be determined. The peak expression after inverse Fourier transform is shown in Equation 103:

[0115]

[0116] By approximation, they are all within a single period, and σ p <T, we can also know that the condition for the value of k is as shown in formula 104: 104; When the k value increases, the peak value decreases rapidly; when the k value increases to a certain extent, the peak value decreases slowly.

[0117] Since the propagation time of underwater acoustic signals is affected by the multipath effect, in order to prevent the multipath effect from causing inter-symbol interference, a blank period is often inserted between two transmission cycles to prevent the interference. In this study, due to the use of pulse stretching technology, it is possible to adjust the parameters so that most of the energy of a single transmission cycle is concentrated in the middle of the transmission cycle, thereby avoiding the use of empty cycles, reducing communication complexity, and increasing the transmission rate to a certain extent. Therefore, this study selected σ g = k value of 0.8T. This allows most of the energy to be concentrated in the middle 80% of a single communication cycle, thus avoiding inter-symbol interference when idle cycles are not used.

[0118] The peak-to-average ratio optimized signal is shown below:

[0119] The multi-carrier underwater communication method based on pulse stretching technology in this invention uses 8 / 10B coding to reduce the difference in signal peaks caused by different transmitted information during single-cycle equalization. Pulse stretching significantly reduces the signal's average-to-peak ratio and concentrates the main power of the signal at the center of a single transmission cycle, reducing the impact of underwater multipath transmission.

[0120] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by technicians in this technical field within the essential scope of the present invention should also fall within the scope of protection of the present invention.

Claims

1. A multi-carrier underwater communication method based on pulse stretching technology, characterized in that: The multi-carrier underwater communication system includes an ASCII encoding unit, an 8B / 10B encoding unit, a modulation unit, a Gaussian envelope approximation and parameter solving unit, a stretching unit, a demodulation unit, a 10B / 8B encoding unit, and an ASCII decoding unit group; The information to be transmitted is converted into a binary array of 8 bits per byte through the ASCII encoding unit. After the binary array passes through the 8 / 10B encoding unit, it becomes a new array of 10 bits per byte. After the new array passes through the multi-carrier OOK modulation of the modulation unit, the Gaussian envelope approximation and parameter solution unit solves the stretching coefficient k. In the stretching unit, the modulated signal is stretched according to the stretching coefficient k. The signal is transmitted in water and received by the receiving end. After demodulation by the demodulation unit, 10B / 8B encoding, and ASCII decoding, the transmitted information is restored. The Gaussian envelope approximation and parameter solving unit and the broadening unit constitute the data peak-to-average ratio optimization module; After peak equalization is performed through the ASCII encoding unit, 8B / 10B encoding unit, and modulation unit, the peak-to-average ratio optimization module is used to reduce the peak-to-average ratio of single-cycle transmission. Reducing the peak-to-average ratio requires adjusting the phase. Among them, when the Gaussian envelope approximation and the parameter adjustment of the parameter solving unit are performed, as shown in formula (2), the signal is dispersed by adding a phase shift term in the frequency domain after Fourier transformation: Here, p(f) is a second-order polynomial with f as the independent variable.

2. The multi-carrier underwater communication method based on pulse stretching technology according to claim 1, characterized in that: The ASCII encoding unit, 8B / 10B encoding unit, and modulation unit constitute the data peak equalization module; After 8b / 10b encoding, the 8-bit data in a binary array becomes a 10-bit array. The number of 0s and 1s in the 10-bit data can be divided into three categories: 5 0s and 5 1s, 6 0s and 1s, and 4 0s and 6 1s. When the disparity of the previous byte is -2, the disparity of the next byte is adjusted to +2 to ensure that the number of 0s and 1s is equal over a long period of time.

3. The multi-carrier underwater communication method based on pulse stretching technology according to claim 1, characterized in that: The signal s(t) of a single transmission cycle after multi-carrier OOK modulation is shown in formula (1): Where M is the row matrix composed of 0s and 1s after data is 8 / 10B encoded, n is the number of carriers, f is the carrier frequency, t is time, and T is the duration of a single transmission cycle; When the amplitude and initial phase of each carrier are the same, the s(0) data in s(t) is the pulse peak value of a single transmission cycle; If 8b / 10b encoding is not performed, the value range of s(0) is 0 to n; If 8b / 10b encoding is performed, the value range of s(0) is 2 / n-1 to 2 / n+1.

4. The multi-carrier underwater communication method based on pulse stretching technology according to claim 1, characterized in that: In the peak-to-average ratio optimization module, the original signal also needs to be approximated by a Gaussian envelope. Since the Gaussian envelope approximation has a higher fitting degree near the peak point of the signal, and the peak-to-average ratio focuses on the peak point, the expression of the Gaussian approximation is shown in formula (3), where σ is the Gaussian envelope width and f0 is the center frequency of the signal band. After Fourier transforming the expression after Gaussian approximation and multiplying it by the phase shift term, the optimal phase shift parameter and phase shift term expression can be obtained through the expression after inverse Fourier transform: Then, the expression after Fourier transform of the Gaussian envelope approximation signal is shown in formula (4): Then, the expression of the multiplication of the Gaussian envelope and the phase shift term is shown in Equation (5); Since p(f) is a quadratic polynomial, the inverse Fourier transform of equation (5) is expressed as equation (6), where It can be seen from formula (6) that the expression is still in the form of Gaussian envelope multiplied by oscillation term, and its peak value becomes The axis of symmetry becomes t=p′(f0), and the frequency of the oscillation term becomes f0-p(f0).

5. The multi-carrier underwater communication method based on pulse stretching technology according to claim 4, characterized in that: After obtaining the inverse Fourier transform expression according to (6), in order to reduce the peak value as much as possible, under the condition that the symmetry axis remains unchanged, the peak value remains at s(0), and the oscillation term frequency remains unchanged, within a single cycle σ p <T, according to equations (2) and (3), we can get the result shown in equation (7): p(f0) = 0, p′(f0) = 0 (7); Therefore, the expression of p(f) is as shown in formula (8): p(f) = kf 2 -2kf0f(8), where k is the coefficient to be determined; by combining equations (6) and (8), the peak expression after inverse Fourier transform is shown in equation (9): Combining equations (4) and (9), we can get the condition for the value of k as shown in equation (10): When the k value increases, the peak value decreases rapidly, and when the k value increases to the peak value, it slowly decreases.

6. The multi-carrier underwater communication method based on pulse stretching technology according to claim 5, characterized in that: Select σ g = 0.8T, so that the data is concentrated in the middle 80% of a single communication cycle, avoiding inter-symbol interference when no idle cycle is used; The signal after peak-to-average ratio optimization is shown in formula (11):

Images