Spread spectrum underwater acoustic communication method based on OTFS-IM

Abstract

The invention provides a spread spectrum underwater acoustic communication method based on OTFS-IM. The spread spectrum underwater acoustic communication method comprises the following steps: grouping communication information at a signal transmitting end to obtain grouped communication information; generating an information symbol matrix of a delay-Doppler domain according to the packet communication information; converting the information symbol matrix to obtain a transmitting signal; processing a receiving signal matched with the transmitting signal at a signal receiving end to obtain a signal receiving matrix; determining a synchronous receiving matrix according to the signal receiving matrix; segmenting the synchronous receiving matrix according to groups to obtain a receiving matrix; and decoding each group of receiving matrixes to obtain decoding information, and recovering communication information based on each group of information corresponding to the decoding information. Information to be transmitted is transmitted to a receiving end through spread spectrum code selection, spread spectrum code position selection and spread spectrum code phase; the related signal receiving matrixes are weighted and combined according to the time delay-Doppler domain channel gain, the spread spectrum symbols are judged according to the synchronous receiving matrixes, and the performance of the received communication information is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of underwater acoustic communication and signal processing, and particularly relates to a spread-spectrum underwater acoustic communication method based on OTFS-IM. Background Technique

[0002] Although the underwater acoustic channel as a transmission medium can play an important role in ocean resource development and the military field, however, the channel not only has the multipath effect, but is also limited by characteristics such as time-varying effect and bandwidth limitation, which seriously restricts the transmission rate, stability and reliability of the underwater acoustic communication system. To address the above challenges, the transmission schemes of high-speed underwater acoustic communication technology are mainly based on orthogonal frequency division multiplexing technology (OFDM). Compared with the traditional single-carrier communication method, OFDM effectively copes with the frequency-selective fading caused by the multipath effect through the frequency-domain parallel transmission mechanism, and at the same time maintains the characteristic of relatively low transceiver complexity. Especially in a dynamic ocean environment, the orthogonality between OFDM subcarriers is destroyed in a fast time-varying fading channel under complex moving conditions, resulting in serious inter-carrier interference (ISI).

[0003] In view of the technical drawbacks of the above OFDM, on the one hand, scholars have proposed OFDM technology based on index modulation (OFDM-IM) from the perspective of controlling subcarrier activity, which improves the system in terms of energy efficiency and anti-frequency shift performance. On the other hand, scholars have proposed filter bank-based multicarrier modulation (FBMC) technology from the perspective of adjusting the subcarrier orthogonality condition and the time-frequency point distance. The FBMC technology introduces a pulse filter with good time-frequency localization to solve the out-of-band radiation and fading problems of OFDM. However, the non-ideal pulse filter will simultaneously bring additional interference. Therefore, the index modulation technology is applied to FBMC, and FBMC technology based on index modulation (FBMC-IM) is proposed. FBMC-IM divides the subcarriers into several subcarrier groups, where a part of the subcarriers are activated and transmit constellation data, and the remaining subcarriers are in a silent state, so that the activated subcarriers are sparsely distributed.

[0004] Recently, a new two-dimensional modulation scheme - orthogonal time-frequency space modulation (OTFS) has been proposed and has become a reliable communication scheme for high-mobility communication. Compared with the OFDM modulation that modulates information in the time-frequency domain, OTFS modulates information in the delay-Doppler domain. It not only has excellent delay and Doppler flexibility, but also ensures that the symbols in the data frame will experience almost constant attenuation. It is worth noting that many studies have shown that OTFS has a significant performance improvement compared with the OFDM modulation scheme in the time-varying channel of the underwater acoustic channel.

[0005] In addition, although the traditional Orthogonal Time Frequency Space with Index Modulation (OTFS-IM) method has advantages such as low power consumption, simple implementation at the transmitter, and high spectral efficiency, and is suitable for high-speed mobile scenarios and large-scale MIMO systems, the traditional OTFS-IM method is vulnerable to interference from partially activated symbols in index modulation, especially in a low signal-to-noise ratio environment, which may also limit the bit error rate performance. At the same time, the OTFS-IM system needs to perform demodulation in the time-delay Doppler domain at the receiver, and complex detection algorithms (such as sparse signal detection, iterative detection, MP detection) are required to decode the index modulation information, resulting in a large amount of computation. Summary of the Invention

[0006] The purpose of the present invention is to solve the deficiency of poor performance of the existing OTFS-IM technology under low signal-to-noise ratio, and provide a spread-spectrum underwater acoustic communication method based on OTFS-IM. The packet communication information at the transmitter transmits the information to be transmitted to the receiver through spread-spectrum code selection, spread-spectrum code position selection, and the phase information of the spread-spectrum code respectively; the receiver uses a two-dimensional Rake receiver based on the time-delay Doppler domain, weights and combines the relevant signal reception matrix according to the time-delay Doppler domain channel gain, and judges the spread-spectrum symbol based on the synchronous reception matrix, thereby improving the performance of the received communication information and effectively reducing the complexity of processing communication signals.

[0007] To achieve the above purpose, the technical solution provided by the present invention is:

[0008] A spread-spectrum underwater acoustic communication method based on OTFS-IM includes: a modulation method at the transmitter of the underwater acoustic communication signal and a demodulation method at the receiver. Specifically:

[0009] The transmitter modulation method includes: (1) grouping the transmitted communication information at the signal transmitter to obtain grouped communication information; (2) generating an information symbol matrix in the time-delay Doppler domain according to the grouped communication information; (3) converting the information symbol matrix into a time-frequency domain signal, converting the time-frequency domain signal into a time-domain complex signal, and obtaining a transmitted signal based on the time-domain complex signal.

[0010] The receiving - end demodulation method includes: (1) Processing the received signal matching the transmitted signal at the signal receiving - end to obtain a delay - Doppler signal receiving matrix; (2) Determining a synchronous receiving matrix according to the signal receiving matrix; (3) Dividing the synchronous receiving matrix into groups to obtain a receiving matrix for each group; (4) Decoding each receiving matrix of each group respectively, and based on the decoded information obtained by decoding the receiving matrix, obtaining the information of each group, and restoring the communication information based on the information of each group.

[0011] As a further limitation of the present invention, the transmitting - end modulation method specifically includes: (1) Grouping the communication information to be transmitted through a bit splitter at the signal transmitting - end, and the obtained grouped communication information includes index bits, mapping bits, and phase - information bits; (2) Performing a zero - setting operation on the grids corresponding to the unactivated data according to the grouped communication information to generate an information symbol matrix in the delay - Doppler domain; (3) Transforming the information symbol matrix into the time - frequency domain signal through an ISFFT, and transforming the time - frequency domain signal into the time - domain complex signal using a Heisenberg transform; Based on the time - domain complex signal, adding a CP prefix to it to obtain the transmitted signal.

[0012] As a further limitation of the present invention, (2) in the transmitting - end modulation method specifically includes:

[0013] According to the index bits in the grouped communication information, controlling the activation of the grids in the delay - Doppler domain, placing the spreading sequences selected by the mapping bits of the grouped communication information into the activated grids, and performing a zero - setting operation on the grids corresponding to the unactivated data in the spreading sequences to generate an information symbol matrix in the delay - Doppler domain.

[0014] As a further limitation of the present invention, in the transmitting - end modulation method, (1) further includes: Grouping the information bits to be transmitted; specifically including:

[0015] Based on the OTFS - IM framework, for effective implementation of index modulation, dividing the bit information A in the communication information into g groups through a bit splitter, each group including bit information p, expressed as A = pg; Mapping the bit information p of each group to a sub - grid in the time - delay - Doppler domain respectively, and the number of columns n of each sub - grid is N / g, where N represents the number of Doppler dimensions; Dividing the bit information p of each group into index bit p 1 、mapping bit p 2 and phase - information bit p 3 , so the bit information p transmitted by each sub - grid is expressed as: p = p 1 +p 2 +p 3 .

[0016] As a further limitation of the present invention, the modulation of the packet communication information in (1) of the transmitter modulation method specifically includes:

[0017] (1) p 1 index bits are used to select the positions of the active symbols in each of the sub-grids. By default, k symbol positions are selected for activation, corresponding to the bit data transmitted by the index selector for p 1 index bits Denoted as

[0018] (2) p 2 index bits are used for Gold sequence symbol mapping, and the bit data transmitted by its mapper is denoted as p 2 = klog 2 W, where k represents k symbol positions and W represents the number of Gold sequences that can be selected in the sub-grid;

[0019] (3) p 3 phase information bits control the positive and negative of the Gold spreading sequence.

[0020] As a further limitation of the present invention, in order to effectively perform index modulation in the transmitter modulation method (1), it further includes:

[0021] For the β-th sub-network in the sub-grid of the delay-Doppler domain, the index bit p 1 is input into the index selector to select and activate k column index bits from n columns. The corresponding numbers in the β-th sub-network are represented as:

[0022] I β = {i β,1 , i β,2 , …, i β,k} Formula (1)

[0023] In Formula (1), I β represents the activation data in the β sub-networks, i β,1 represents the first activated column selected, i β,2 represents the second activated column selected, i β,k represents the activation data among the k columns selected; where: i β,l ∈ [1, 2, …, n], β = [1, 2, …, g], l = [1, 2, …, k];

[0024] The mapping bit p 2 entering the mapper and the mapped spreading sequence output after passing through the mapper are represented as:

[0025] S β = {s β,1 , sβ,2 ,…,s β,k} Formula (2)

[0026] In formula (2), S β represents the mapped spreading sequence of the β-th subnetwork, s β,1 represents the spreading sequence placed in the first active column within this sub-grid, s β,2 represents the spreading sequence placed in the second active column within this sub-grid, s β,k represents the spreading sequence placed in the k-th active column within this sub-grid; where s β,α ∈±c (i) , β = [1, 2, …, g], α = [1, 2, …, k], i = [1, 2, …, W], c (i) represents the i-th Gold sequence, s β,α represents the spreading sequence placed in the α-th active column within this sub-grid, β represents the OTFS sub-grid serial number, g represents the number of OTFS sub-grids, α represents the serial number of the active column within the sub-grid, and k represents the number of active columns within the sub-grid;

[0027] Specifically, s in formula (2) β,α is to control the phase of the spreading sequence through a single phase information bit p 3 and is expressed as:

[0028]

[0029] In formula (3), s β,α represents the α-th column in the β-th sub-grid, c (i) represents the i-th Gold sequence;

[0030] After the above-mentioned index modulation is performed, the data output by the β-th sub-grid based on the OTFS-IM framework is expressed as:

[0031] X β = [x β (1), x β (2), …, x β (n)] T Formula (4)

[0032] In formula (4), X β represents the data output by the β-th sub-grid, x β (1) represents the data of the first column of this sub-grid, x β (2) represents the data of the second column of this sub-grid, x β (n) represents the data of the n-th column of this sub-grid,

[0033] As a further limitation of the present invention, the receiving - end demodulation method specifically includes:

[0034] (1) At the signal receiving end, remove the CP prefix from the received signal that matches the transmitted signal, and sequentially perform the Wigner transform and the SFFT transform on it to obtain a time - domain signal, and transform the time - domain signal to obtain a delay - Doppler signal reception matrix;

[0035] (2) According to the known underwater acoustic communication channel, find the Doppler frequency shift corresponding to the maximum peak in the signal reception matrix based on the delay - Doppler to obtain the synchronization position, and perform circular shift on it based on the synchronization position to obtain a synchronized reception matrix;

[0036] (3) Divide the synchronized reception matrix according to the Doppler index upper limit and each group of position index information by group to obtain the reception matrix of each group;

[0037] (4) Decode each group of the reception matrices using a two - dimensional Rake receiver in the time - delay - Doppler domain to obtain the decoded information obtained from each group of the reception matrices; according to the Gold sequence position, spreading sequence number, and decision positive - negative value respectively included in the decoded information of each group, obtain the information of each group; restore the communication information based on the index bit, mapping bit, and single - phase information bit respectively included in the information of each group.

[0038] As a further limitation of the present invention, (3) in the receiving - end demodulation method is specifically:

[0039] According to the Doppler index upper limit and each group of position index information, calculate the correlation using frequency - domain correlation, and divide the synchronized reception matrix after circular shift by group to obtain the reception matrix of each group, expressed as:

[0040]

[0041] In formula (5), Y β represents the reception matrix of each group, K es represents the Doppler index upper limit, represents a set of complex - number matrices of M rows and N columns;

[0042] Based on the reception matrix Y of each group in formula (5) β , use the two - dimensional Rake receiver of the receiving - end demodulation method (4) to calculate the FFT for it and calculate and store the FFT of the i - th Gold sequence of each sub - grid in the corresponding reception matrix Y β ; the frequency - domain correlation vector of the corresponding reception matrix Y β is expressed as:

[0043]

[0044] In Equation (6), r β,ε represents a vector element, and r β,ε = [r 0 , r 1 , …, r M-1 T , ε = [0, 1, …, 2K es - 1], r 0 represents the first element of vector r β,ε , r 1 represents the second element of vector r β,ε , r M-1 represents the Mth element of vector r β,ε , T represents the transpose symbol, ε represents the number of columns, and K es represents the upper limit of the Doppler index, where i = [1, 2, …, W].

[0045] As a further limitation of the present invention, in (4) of the receiving - end demodulation method, each group of received matrices is decoded using a two - dimensional Rake receiver in the time - delay - Doppler domain, including:

[0046] Based on each group of the received matrices, by weighting, the contributions of different channel components in the time - domain - Doppler domain are combined, using the channel tap gain as the weight, to perform a weighted calculation on the frequency - domain - related vector element r β,ε , and the expression is:

[0047]

[0048] In Equation (7), q u,v represents the element in the uth row and vth column of the decision matrix Q, h ε represents the channel matrix gain, h ε (t) represents the tth element in h ε , r β,ε (m) represents the mth element of the vector element r β,ε ;

[0049] By comparing the maximum absolute values of the outputs of n×W branches, the maximum - value index is determined, which is expressed as:

[0050]

[0051] In Equation (8), represents the determined maximum - value index, Q represents the decision matrix, which is a set of q u,v , u represents the row number of the decision matrix, u = [1, 2, …, n], n represents the number of rows of the decision matrix, and its value is v represents the column number of the decision matrix, v = [1, 2, …, W], and W represents the number of columns of the decision matrix, and its value is​

[0052] Based on the determined maximum index Obtain the corresponding index bit and mapping bit, and determine the corresponding phase information bit; specifically, the decision rule of the phase information bit is expressed as:

[0053]

[0054] In formula (9), represents the decision information bit, u represents the row number of the decision matrix, v represents the column number of the decision matrix, and Q represents the decision matrix.

[0055] The advantages of the present invention are:

[0056] 1. The packet communication information at the transmitting end of the present invention transmits the information to be transmitted to the receiving end through spreading code selection, spreading code position selection, and the phase information of the spreading code respectively; the receiving end uses a two-dimensional Rake receiver based on the time-delay Doppler domain, weights and combines the relevant signal reception matrix according to the time-delay Doppler domain channel gain, and judges the spreading symbol based on the synchronous reception matrix, thereby improving the performance of the received communication information and effectively reducing the complexity of processing communication signals.

[0057] 2. In the case of underwater acoustic channel communication with low signal-to-noise ratio and significant Doppler effect, the present invention makes full use of the strong anti-interference ability and low bit error rate of the spreading Gold sequence. When modulating each group of communication information, the Gold sequence is used as the carrier of the phase information bit. First, the selected spreading sequence is controlled according to the mapping bit, and the positive and negative of the spreading sequence are controlled by the phase information bit; then, the activation of the grid in the delay-Doppler domain is controlled according to the index bit, the spreading sequence is placed in the activated grid, and the unactivated grid is set to zero; the receiving end judges the spreading symbol from two dimensions of the time-delay Doppler domain based on the Rake receiver in the order of time domain first and then frequency domain, reducing the complexity of signal processing.

[0058] 3. The present invention combines the index modulation technology with the OTFS system, and proposes an index-based OTFS technology modulation (OTFS-IM). In the delay-Doppler domain, the information is divided into index information and constellation information for transmission, and the position of the constellation information in the delay-Doppler grid is flexibly controlled through the index information, achieving a balance between spectral efficiency and energy efficiency and forming the required trade-off solution.

[0059] 4. The present invention introduces the OTFS-IM technology into underwater acoustic communication. Using the ML detection algorithm at the receiving end, compared with the traditional OFDM-IM modulation scheme and OTFS scheme, it can obtain better bit error rate performance.

[0060] Additional aspects and advantages of the present invention will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present invention. Brief Description of the Drawings

[0061] The above and / or additional aspects and advantages of the present invention will become apparent and be readily understood from the description of the embodiments in conjunction with the following drawings, where:

[0062] Figure 1 : Flowchart of a spread-spectrum underwater acoustic communication method based on OTFS-IM provided by the present invention;

[0063] Figure 2 : Flowchart of OTFS-IM spread-spectrum communication for underwater communication provided by the present invention;

[0064] Figure 3 : Principle flowchart of a low-complexity two-dimensional Rake receiver provided by the present invention;

[0065] Figure 4 : Diagrams of the autocorrelation function and ambiguity function of the DSSS signal generated using the Gold sequence provided by the present invention;

[0066] Figure 5 : Data analysis diagram of the bit error rate using the MP detection algorithm and the two-dimensional Rake receiver scheme when the spread-spectrum sequence length M = 64 provided by the present invention;

[0067] Figure 6 : Data analysis diagram of the bit error rate of the OTFS-IM-2D Rake receiver under different lengths of spread-spectrum sequences and different moving speeds provided by the present invention;

[0068] Figure 7 : When p 1 = 2, p 2 = 2, p 3 = 1, partial index modulation mapping table. Detailed Description of the Embodiments

[0069] The embodiments of the present invention will be described in detail below. The embodiments are exemplary and are intended to explain the present invention, but should not be construed as limiting the present invention.

[0070] Please refer to Figure 1 , the embodiments of the present invention provide a spread-spectrum underwater acoustic communication method based on OTFS-IM, including: the modulation method at the transmitter end and the demodulation method at the receiver end of the underwater acoustic communication signal. Specifically:

[0071] (1) The modulation method at the transmitter end of the embodiments of the present invention includes:

[0072] (11) Group the transmitted communication information at the signal transmitting end to obtain grouped communication information; specifically: at the signal transmitting end, use a bit splitter to group the transmitted communication information, and the grouped communication information of each group includes index bits, mapping bits, and phase information bits.

[0073] More specifically, use a bit splitter to group the transmitted communication information, and each group includes index bits, mapping bits, and phase information bits. Let the phase information bit vector input per frame be A ∈ [b 1 , b 2 , …, b Num , b γ ∈ [0, 1], γ = 1, 2, …, Num, where b Num represents the Num-th phase information bit, and Num represents the total number of phase information bits transmitted per frame. The bit splitter first divides the Num phase information bits into g groups, each group having p bits of information (Num = pg), and then divides each group of p bits into p 1 index bits, p 2 mapping bits, and p 3 phase information bits.

[0074] (12) Generate an information symbol matrix in the delay-Doppler domain according to the grouped communication information; specifically: generate an information symbol matrix in the delay-Doppler domain according to the index bits, mapping bits, and phase information bits in the grouped communication information, and perform a zeroing operation on the grids corresponding to the unactivated data.

[0075] More specifically, generate an information symbol matrix in the delay-Doppler domain according to the index bits, mapping bits, and phase information bits, and zero the unactivated columns in the sub-grids. Let the length of the selected Gold spreading sequence be M, and the spreading codes with good autocorrelation are divided into several groups, called spreading code groups, and each group has spreading codes.

[0076] Considering that the grouped communication information needs to be synchronized, an additional group is added at the beginning of each frame, that is, (g + 1) groups are transmitted, and each group is controlled by p 1 index bits to control 2 p1 activation positions, then the information symbol matrix in the delay-Doppler domain is where M = 0, 1, …, M - 1, n = 0, 1, …, N - 1, M represents the number of subcarriers, which is equal to the length of the spreading sequence, and N represents the number of symbols. In the embodiment of the present invention, the spreading sequence is placed in the information symbol matrix by columns, and it is divided into (g + 1) sub-grids by groups, and each sub-grid represents β = 1, 2, …, g + 1, where β represents the index of the sub - grid. Due to the limited number of spreading codes, the limited set of spreading codes is sequentially assigned to each sub - grid. If the number of spreading codes is insufficient, the assignment is cyclic. In the embodiments of the present invention, except for the first synchronization sub - grid, each sub - grid corresponds to p 1 index bits, p 2 mapping bits, and p 3 phase - information bits. The default activation position of the synchronization sub - grid is the first column, and its spreading code does not require phase - information - bit control. Except for the synchronization sub - grid, each sub - grid first selects the position of the activation column according to its corresponding index bits, and all unactivated columns are set to zero.

[0077] (13) Transform the information - symbol matrix into a time - frequency domain signal, transform the time - frequency domain signal into a time - domain complex signal, and obtain the transmitted signal based on the time - domain complex signal; specifically: use ISFFT (Inverse Symplectic Finite Fourier Transform) to transform the information - symbol matrix into a time - frequency domain signal, and use the Heisenberg transform to transform the time - frequency domain signal into a time - domain complex signal; based on the time - domain complex signal and add a CP prefix to it to obtain the transmitted signal.

[0078] (2) The demodulation method at the receiving end in the embodiments of the present invention includes:

[0079] (21) Process the received signal that matches the transmitted signal at the signal receiving end to obtain a delay - Doppler signal - receiving matrix; specifically: at the signal receiving end, remove the CP prefix from the received signal that matches the transmitted signal and sequentially perform the Wigner transform and the SFFT (Sparse Fast Fourier Transform) to obtain a time - domain signal, and transform the time - domain signal to obtain a delay - Doppler signal - receiving matrix.

[0080] The above - mentioned (21) in the embodiments of the present invention specifically includes: according to the index bits in the packet - communication information, control the activation of the grids in the delay - Doppler domain, place the spreading sequence selected by the mapping bits of the packet - communication information in the activated grids, and perform a zero - setting operation on the grids corresponding to the unactivated data in the spreading sequence to generate an information - symbol matrix in the delay - Doppler domain.

[0081] In addition to specifically including the above solutions, the (21) of the embodiments of the present invention further includes: grouping the transmitted information bits; specifically: based on the OTFS-IM framework, in order to effectively perform index modulation, the bit information A in the communication information is divided into g groups by a bit splitter, and each group includes bit information p, expressed as A = pg; the bit information p of each group is respectively mapped to a sub-grid in the time-delay Doppler domain, and the number of columns n of each sub-grid is N / g, where N represents the number of Doppler dimensions; the bit information p of each group is divided into index bits p 1 , mapped bits p 2 and phase information bits p 3 , so the bit information p transmitted by each sub-grid is expressed as: p = p 1 + p 2 + p 3 . Preferably: p 1 index bits are used to select the positions of the active symbols in each sub-grid, and by default, k symbol positions are selected for activation, corresponding to the bit data transmitted by the index selector of p 1 expressed as p 2 index bits are used for Gold sequence symbol mapping, and the bit data transmitted by its mapper is expressed as p 2 = klog 2 W, where k represents k symbol positions and W represents the number of Gold sequences that can be selected by the sub-grid; p 3 phase information bits control the positive and negative of the Gold spreading sequence. The embodiments of the present invention are based on OTFS-IM and spread spectrum communication, enabling the spread spectrum underwater acoustic communication method to simultaneously have the good anti-Doppler characteristics of OTFS and the good anti-interference ability of spread spectrum communication.

[0082] The above of the embodiments of the present invention for effectively performing index modulation specifically includes:

[0083] First, for the β-th sub-network in the sub-grid of the time-delay Doppler domain, input the index bit p 1 into the index selector to select and activate k column index bits from n columns, and the corresponding numbers in the β-th sub-network are expressed as:

[0084] I β = {i β,1 , i β,2 , …, i β,k} Formula (1)

[0085] In Formula (1), I β represents the activation data in the β sub-networks, and i β,1 represents the first selected activation column, i β,2 ​Indicates the second selected active column, i β,k Indicates the k-th selected active column; where: i β,l ∈[1, 2, …, n], β = [1, 2, …, g], l = [1, 2, …, k];

[0086] Then, according to the mapping bit p entering the mapper 2 Select a spreading code from the allocated set of spreading codes, and control the sign of the spreading code by the phase information bit, and place it in the active position. The mapping bit p entering the mapper 2 And the mapped spreading sequence output after passing through the mapper is expressed as:

[0087] S β = {s β,1 , s β,2 , …, s β,k} Formula (2)

[0088] In formula (2), S β Represents the mapped spreading sequence of the β-th sub-network, s β,1 Represents the spreading sequence placed in the first active column within this sub-grid, s β,2 Represents the spreading sequence placed in the second active column within this sub-grid, s β,k Represents the spreading sequence placed in the k-th active column within this sub-grid; where, s β,α ∈ ±c (i) , β = [1, 2, …, g], α = [1, 2, …, k], i = [1, 2, …, W], c (i) Represents the i-th Gold sequence, s β,α Represents the spreading sequence placed in the α-th active column within this sub-grid, β represents the OTFS sub-grid serial number, g represents the number of OTFS sub-grids, α represents the serial number of the active column within the sub-grid, k represents the number of active columns within the sub-grid; specifically, the phase of the spreading sequence is controlled by a single phase information bit p 3 , and s β,α in formula (2) is expressed as:

[0089]

[0090] In formula (3), s β,α Represents the α-th column in the β-th sub-grid, c (i) Represents the i-th Gold sequence;

[0091] After the above-mentioned index modulation is performed, the data output from the β-th sub-grid based on the OTFS-IM framework is expressed as:

[0092] X β = [x β(1), x β (2), …, x β (n)] T Equation (4)

[0093] In Equation (4), X β represents the data output by the β-th sub-grid, and x β (1) represents the data of the first column of this sub-grid, and x β (2) represents the data of the second column of this sub-grid, and x β (n) represents the data of the n-th column of this sub-grid,

[0094] Next, use ISFFT (Inverse Symplectic Finite Fourier Transform) to transform the information symbol matrix into a time-frequency domain signal, use the Heisenberg transform to transform it into a time-domain complex signal, and add a CP prefix to obtain the transmitted signal. Through the inverse symplectic finite Fourier transform (ISFFT), map the information symbol matrix X DD [m, n] in the delay-Doppler domain to the digital modulation symbol X TF [l, k] in the time-frequency domain, which is expressed as:

[0095]

[0096] In the formula, X TF represents the digital modulation signal in the time-frequency domain, l = 0, 1, …, M - 1, k = 0, 1, …, N - 1.

[0097] Then, perform the Heisenberg transform and pulse shaping on the digital modulation symbol X TF [l, k] in the time-frequency domain to transform the time-frequency domain signal into a time-domain signal, which is expressed as:

[0098]

[0099] In the formula, g tx represents the selected transmit pulse shaping function, t represents time, T represents the symbol period, and Δf represents the subcarrier spacing. The transmitted signal x(t) is added with a cyclic prefix (CP) to obtain the final time-domain transmitted signal s(t). The length of the CP sequence depends on the maximum multipath delay.

[0100] (22) Determine the synchronous reception matrix according to the signal reception matrix; specifically: according to the known underwater acoustic communication channel, find the Doppler shift corresponding to the maximum peak based on the delay-Doppler for the signal reception matrix, that is, the synchronous position. Perform a cyclic shift on it based on the synchronous position to obtain the synchronous reception matrix, ensuring that the first column is the point with the maximum peak.

[0101] (23) Split the synchronous reception matrix by group to obtain the reception matrix of each group; specifically: according to the Doppler index upper limit and the position index information of each group, split the synchronous reception matrix by group to obtain the reception matrix of each group.

[0102] More specifically, in the embodiment of the present invention, according to the Doppler index upper limit and the position index information of each group, the correlation is calculated using frequency-domain correlation, and the synchronous reception matrix after synchronous cyclic shift is split by group to obtain the reception matrix of each group, expressed as:

[0103]

[0104] In formula (5), Y β represents the reception matrix of each group, K es represents the Doppler index upper limit, represents the set of complex matrices of M rows and N columns;

[0105] Use frequency-domain correlation to calculate the correlation. Based on the reception matrix Y of each group in formula (5) β , use the two-dimensional Rake receiver of the receiver demodulation method (4). According to the Doppler index upper limit K es and the position index information of each group, split the received matrix after synchronous shift by group to obtain the received matrix of each group and calculate the FFT of the received matrix Y β , calculate and store the FFT of the i-th Gold sequence of each sub-grid in the corresponding received matrix Y β ; the frequency-domain correlation vector of the corresponding received matrix Y β is expressed as:

[0106]

[0107] In formula (6), r β,ε represents the vector element, r β,ε = [r 0 , r 1 , …, r M-1 T , ε = [0, 1, …, 2K es - 1], r 0 represents the first element of the vector r β,ε , r 1 represents the second element of the vector r β,ε , r M-1 represents the M-th element of the vector r β,ε , T represents the transpose symbol, ε represents the number of columns, K es represents the Doppler index upper limit, where i = [1, 2, …, W].

[0108] ​Please refer to Figure 2 In specific applications, in the signal receiving end of the embodiments of the present invention, the CP prefix is removed from the received signal, and then the Wigner transform and the SFFT (Sparse Fast Fourier Transform) transform are sequentially performed to transform the time-domain signal into a received matrix in the delay-Doppler domain. Through the Wigner transform, the time-domain received matrix r(t) after removing the CP is transformed into the time-frequency domain, and the expression is:

[0109] Y TF [l,k]=∫r(t)g rx (t-kT)e -j2πlΔft dt

[0110] In the formula, g rx represents the received pulse shaping filter.

[0111] Through the symplectic finite Fourier transform SFFT, the time-frequency domain signal Y TF [l,k] is mapped to the delay-Doppler domain, which is expressed as:

[0112]

[0113] In the embodiments of the present invention, according to the known underwater acoustic channel, the Doppler frequency shift corresponding to the maximum peak is found, that is, the synchronization position. According to the synchronization position, is circularly shifted to obtain the synchronized received matrix to ensure that the first column is the point with the maximum peak. In the embodiments of the present invention, according to the Doppler index upper limit and each group of position index information, the OTFS group splitter is used to split the synchronized received matrix into groups to obtain the received matrix of each group β=1,2,…,g + 1. In the embodiments of the present invention, the two-dimensional Rake receiver in the delay-Doppler domain is used to decode the received matrix Y β of each group respectively. The spreading code groups for decoding each group correspond to the transmitting end, and the synchronization sub-grid does not need to be decoded.

[0114] (24) Decode the received matrix of each group using a two-dimensional Rake receiver respectively. According to the decoded information obtained by decoding the received matrix, obtain the information of each group, and use a bit synthesizer to restore the communication information based on the information of each group. Specifically: use a two-dimensional Rake receiver in the time-delay Doppler domain to decode the received matrix of each group respectively, and obtain the decoded information obtained from the received matrix of each group; according to the Gold sequence position, spreading sequence number, and decision positive and negative values respectively included in the decoded information of each group, obtain the information of each group respectively; restore the communication information based on the index bits, mapping bits, and single-phase information bits respectively included in the information of each group. In the embodiment of the present invention, by receiving the communication signal sent by the transmitting end through the above-mentioned two-dimensional Rake receiver in the time-delay Doppler domain, not only can the spread-spectrum underwater acoustic communication method ensure a low bit error rate even in complex underwater acoustic channels with low signal-to-noise ratio and significant Doppler effect during signal processing, thereby ensuring the effect of underwater covert communication; moreover, its receiver complexity is relatively low, and the processing time is better than that of traditional underwater acoustic communication methods.

[0115] More specifically, the above-mentioned decoding of the received matrix of each group using a two-dimensional Rake receiver in the time-delay Doppler domain in the embodiment of the present invention specifically includes:

[0116] Based on the received matrix of each group, combine the contributions of different channel components in the time-domain Doppler domain in a weighted manner, use the channel tap gain as the weight, and perform weighted calculation on the vector element r β,ε The expression is:

[0117]

[0118] In formula (7), q u,v represents the element in the u-th row and v-th column of the decision matrix Q, h ε (t) represents the t-th element in h ε , r β,ε (m) represents the m-th element in the vector element r β,ε , t = mod(M - m + 1, M) + 1, and the channel matrix H dd The expression is u = [1, 2,..., n], v = [1, 2,..., W].

[0119] By comparing the maximum absolute values of the outputs of n×W branches, determine the maximum value index, which is expressed as:

[0120]

[0121] In formula (8), represents the determined maximum value index, Q represents the decision matrix, and this decision matrix Q is a set of q u,v , which is expressed as u represents the row number of the decision matrix, u = [1, 2, …, n], where n represents the number of rows of the decision matrix, and its value is β = 2, 3, …, v = [1, 2, …, W], where W represents the number of columns of the decision matrix, and its value is

[0122] Based on the determined maximum value index Obtain the corresponding index bit and mapping bit, and determine the corresponding phase information bit; specifically, the decision rule of the phase information bit is expressed as:

[0123]

[0124] In formula (9), represents the decision information bit, u represents the row number of the decision matrix, v represents the column number of the decision matrix, and Q represents the decision matrix.

[0125] Preferably, for the OTFS-IM system with n = 4, k = 1, W = 4 (i.e., p 1 = 2, p 2 = 2, p 3 = 1), according to the Figure 7 shown index modulation mapping table to decide the index bit and mapping bit. Finally, use a bit synthesizer to merge the decision bits of each group to restore the estimated phase information bit matrix

[0126] More specifically, the embodiment of the present invention generates a random channel matrix suitable for underwater according to the characteristics of large propagation delay and significant Doppler effect in the underwater acoustic channel, and transmits the transmitted signal through the underwater acoustic channel, adding multipath, Doppler and noise. The transmitted signal x(t) passes through a time-varying channel with an underwater acoustic channel impulse response of h(τ, υ), and this channel response corresponds to the delay-time response g tx (τ, t). The signal data transmitted to the receiving end is the received time-domain signal:

[0127] r(t) = ∫∫h(τ, υ)s(t - τ)e j2πυ(t-τ) dτdτ + ω(t)

[0128] = ∫g(τ, t)s(t - τ)dτ + ω(t)

[0129] In the formula, r(t) represents the signal data, g tx (τ, t) represents the delay-time response, g(τ, t) = ∫ υ h(τ, υ)e j2 πυ(t-τ) dτ, τ represents the actual time delay, υ represents the actual Doppler frequency shift, t represents time, ω(t) represents the time-domain form of Gaussian white noise, and j represents the imaginary symbol.

[0130] The DD-domain channel impulse response of the embodiment of the present invention reveals the principle of DD-domain channel modeling, and the expression is as follows:

[0131]

[0132] In the formula, h(τ, υ) represents the DD-domain channel impulse response, P represents the number of propagation paths, and h i represents the gain of the i-th propagation path, τ i represents the delay of the i-th propagation path, υ represents the actual Doppler frequency shift, and υ i represents the Doppler frequency shift of the i-th propagation path, and δ represents the Dirichlet function.

[0133] The relationship between the delay, Doppler sampling offset of the i-th propagation path in the DD-domain grid and the actual delay and Doppler frequency shift is expressed as:

[0134]

[0135] In the formula, M represents the length of the selected Gold spreading sequence, Δf represents the Doppler frequency shift of each block, υ i represents the Doppler frequency shift, represents the i-th Doppler tap, represents the normalized frequency shift, N represents the number of grid columns, and T represents the duration of each block;

[0136] According to the time-domain signal r(t) received by the receiving end and the channel impulse response h(τ, υ) in the embodiment of the present invention, the relationship expression between the time-domain transmitted signal s(t) and the time-domain received signal r(t) is further obtained as:

[0137]

[0138] Figure 4 Shows the autocorrelation function and ambiguity function of generating DSSS signals using the Gold sequence. It can be seen that the Gold sequence has good autocorrelation performance, and the ambiguity function also has a relatively sharp peak, which is very suitable for spread-spectrum communication and underwater systems, especially having significant advantages in multipath interference and noise environments. At the same time, the Gold sequence is also vulnerable to Doppler interference, and the OTFS technology can make up for this deficiency.

[0139] The embodiment of the present invention uses MATLAB to simulate the bit error rate of the OTFS-IM spread-spectrum communication scheme in an underwater acoustic channel, and analyzes the effectiveness of the above-mentioned spread-spectrum underwater acoustic communication method of the embodiment of the present invention. The simulation parameter settings are as follows: the index bit p 1 = 2, the mapped bit p 2= 2, that is, 5 bits are transmitted in each group. The number of groups per frame g = 7, and a total of 35 bits are transmitted. The embodiment of the present invention is divided into 8 sub-grids, and each sub-grid has activation positions, and there are spreading codes in the spreading code group. The length of the spreading code can be selected as M = 32, 64, 128, 256, and the number of simulation times is 1000 times. Figure 5 is the bit error rate when using the MP detection algorithm and the two-dimensional Rake receiver scheme when the spreading sequence length M = 64. Under the same parameters, the operation time of using the MP algorithm is 13.6 seconds, while the operation time of using the two-dimensional Rake receiver is only 0.074 seconds. Compared with the MP algorithm, the two-dimensional Rake receiver exchanges a gain loss of about 5 dB for a nearly 183-fold increase in operating speed. Figure 6 is the bit error rate of the OTFS-IM-2D Rake receiver under different lengths of spreading sequences and different moving speeds. The red line, pink line and purple line respectively represent that the upper limit of the channel speed is 1 m / s, 4 m / s and 7 m / s, characterizing the influence of different degrees of Doppler on the communication scheme. It can be seen that the larger the upper limit of the moving speed, the lower the bit error rate. However, overall, the influence of Doppler interference on the OTFS-IM spreading scheme is not significant. Moreover, the longer the spreading sequence used, the lower the bit error rate. Especially when M = 256, the bit error rate is reduced to zero at -5 dB.

[0140] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present invention.

Claims

1. A spread spectrum underwater acoustic communication method based on OTFS-IM, characterized in that: include: The transmitter modulation method and the receiver demodulation method of underwater acoustic communication signals are as follows: The transmitting end modulation method comprises: (1) grouping the communication information to be transmitted at the signal transmitting end to obtain grouped communication information; (2) generating an information symbol matrix in the delay-Doppler domain according to the grouped communication information; (3) converting the information symbol matrix into a time-frequency domain signal, converting the time-frequency domain signal into a time-domain complex signal, and obtaining a transmitting signal based on the time-domain complex signal; The receiving end demodulation method includes: (1) processing the received signal matching the transmitted signal at the signal receiving end to obtain a delay-Doppler signal receiving matrix; (2) determining a synchronous receiving matrix based on the signal receiving matrix; (3) dividing the synchronous receiving matrix into groups to obtain a receiving matrix for each group; (4) decoding the receiving matrix of each group using a two-dimensional Rake receiver, obtaining information of each group based on the decoded information obtained by decoding the receiving matrix, and restoring the communication information based on the information of each group using a bit synthesizer.

2. The method for spread spectrum underwater acoustic communication based on OTFS-IM according to claim 1, characterized in that: The transmitting end modulation method specifically includes: (1) grouping the communication information to be sent through a bit splitter at the signal transmitting end, and the obtained grouped communication information includes index bits, mapping bits and phase information bits; (2) performing a zeroing operation on the grids corresponding to the inactivated data in the grouped communication information according to the grouped communication information, and generating an information symbol matrix in the delay-Doppler domain; (3) converting the information symbol matrix into the time-frequency domain signal through ISFFT, and converting the time-frequency domain signal into the time-domain complex signal using Heisenberg transform; based on the time-domain complex signal, adding a CP prefix to it to obtain the transmitting signal.

3. The method for spread spectrum underwater acoustic communication based on OTFS-IM according to claim 2, characterized in that: The transmitting end modulation method (2) specifically includes: According to the index bit in the packet communication information, the activation of the grid in the delay-Doppler domain is controlled, the spread spectrum sequence selected by the mapping bit of the packet communication information is placed in the activated grid, the grid corresponding to the inactivated data in the spread spectrum sequence is zeroed, and the information symbol matrix in the delay-Doppler domain is generated.

4. The method for spread spectrum underwater acoustic communication based on OTFS-IM according to claim 2, characterized in that: In the transmitting end modulation method, (1) further includes: grouping the transmitted information bits; which specifically includes: Based on the OTFS-IM framework, in order to effectively perform index modulation, the bit information A in the communication information is divided into g groups through a bit divider, each group includes bit information p, expressed as A=pg; the bit information p of each group is respectively mapped to a sub-grid in the delay-Doppler domain, and the number of columns of each sub-grid is n=N / g, where N represents the number of Doppler dimensions; the bit information p of each group is divided into an index bit p1, a mapping bit p2 and a phase information bit p3, so the bit information p transmitted by each sub-grid is expressed as: p=p1+p2+p3.

5. The method for spread spectrum underwater acoustic communication based on OTFS-IM according to claim 4, characterized in that: The modulation of the packet communication information in (1) in the transmitter modulation method specifically includes: (1) p1 index bits are used to select the position of the activated symbol in each of the sub-grids. By default, k symbol positions are selected for activation. The bit data transmitted by the index selector corresponding to the p1 index bits is Expressed as (2) p2 index bits are used for Gold sequence symbol mapping, and the bit data transmitted by the mapper is expressed as p2 = klog2W, where k represents the k symbol position and W represents the number of Gold sequences that can be selected by the subgrid; (3) The p3 phase information bits control the positive or negative of the Gold spreading sequence.

6. The method for spread spectrum underwater acoustic communication based on OTFS-IM according to claim 4, characterized in that: The transmitting end modulation method (1) is to effectively perform index modulation, and further includes: For the β-th sub-network in the delay-Doppler domain sub-grid, the index bit p1 is input into the index selector to select k columns of index bits from n columns and activate them. The corresponding number in the β-th sub-network is expressed as: I β ={i β,1 ,i β,2 ,…,i β,k } Formula(1) In formula (1), I β represents the activation data in β sub-networks, i β,1 Indicates the first activated column selected, i β,2 Indicates the second selected activation column, i β,k represents the selected kth activation column; where: i β,l ∈[1,2,…,n], β=[1,2,…,g], l=[1,2,…,k]; The mapping bit p2 entering the mapper and the mapping spread spectrum sequence output after passing through the mapper are expressed as: S β ={s β,1 ,s β,2 ,…,s β,k } Formula (2) In formula (2), S β represents the mapping spread spectrum sequence of the βth sub-network, s β,1 represents the spreading sequence placed in the first active column of the subgrid, s β,2 represents the spreading sequence placed in the second active column of this subgrid, s β,k represents the spreading sequence placed in the kth activation column in the subgrid; where s β,α ∈±c (i) , β=[1,2,…,g], α=[1,2,…,k], i=[1,2,…,W], c (i) represents the i-th Gold sequence, s β,α represents the spreading sequence placed in the αth activated column in the subgrid, β represents the OTFS subgrid number, g represents the number of OTFS subgrids, α represents the number of activated columns in the subgrid, and k represents the number of activated columns in the subgrid; Specifically, s in formula (2) β,α The phase of the spread spectrum sequence is controlled by a single phase information bit p3, which is expressed as: In formula (3), s β,α represents the αth column in the βth subgrid, c (i) represents the i-th Gold sequence; After the index modulation performed above, the data output by the βth subgrid based on the OTFS-IM framework is expressed as: X β =[x β (1),x β (2),…,x β (n)] T Formula (4) In formula (4), X β Represents the data output by the βth subgrid, x β (1) represents the data in the first column of the subgrid, x β (2) represents the data in the second column of the subgrid, x β (n) represents the data of the nth column of the subgrid, 7. The method for spread spectrum underwater acoustic communication based on OTFS-IM according to claim 1, characterized in that: The receiving end demodulation method specifically includes: (1) at a signal receiving end, removing the CP prefix from a received signal matching the transmitted signal and sequentially performing Wigner transform and SFFT transform to obtain a time domain signal, and transforming the time domain signal to obtain a delay-Doppler signal receiving matrix; (2) according to the known underwater acoustic communication channel, the Doppler frequency shift corresponding to the maximum peak of the signal receiving matrix is ​​found according to the delay-Doppler pair to obtain a synchronization position, and based on the synchronization position, a cyclic shift is performed to obtain a synchronization receiving matrix; (3) dividing the synchronous receiving matrix into groups according to the Doppler index upper limit and each group of position index information to obtain a receiving matrix for each group; (4) Decoding the receiving matrices of each group using a two-dimensional Rake receiver in the delay-Doppler domain to obtain decoding information of the receiving matrices of each group; obtaining information of each group based on the Gold sequence position, spread spectrum sequence number and positive and negative judgment values ​​respectively contained in the decoding information of each group; and restoring the communication information based on the index bit, mapping bit and single phase information bit respectively contained in the information of each group.

8. The OTFS-IM-based spread spectrum underwater acoustic communication method according to claim 7, characterized in that: (3) in the receiving end demodulation method is specifically: According to the Doppler index upper limit and each group of position index information, the correlation is calculated using frequency domain correlation, and the synchronous receiving matrix after synchronous cyclic shift is divided into groups to obtain the receiving matrix of each group, which is expressed as: In formula (5), Y β represents the receiving matrix of each group, K es represents the upper limit of the Doppler index, Represents a set of complex matrices with M rows and N columns; Based on the receiving matrix Y of each group in formula (5) β , using the two-dimensional Rake receiver of the receiving end demodulation method (4), calculating the FFT and calculating and storing the corresponding receiving matrix Y β The FFT of the ith Gold sequence of each subgrid in ; the corresponding receiving matrix Y β The frequency domain correlation vector of is expressed as: In formula (6), r β,ε Represents a vector element, r β,ε =[r0,r1,…,r M-1 ] T ,ε=[0,1,…,2K es -1], r0 represents the vector r β,ε The first element, r1 represents the vector r β,ε The second element, r M-1 Represents vector r β,ε The Mth element of es represents the upper limit of Doppler index, where i = [1, 2, …, W].

9. The method for spread spectrum underwater acoustic communication based on OTFS-IM according to claim 7, characterized in that: In (4) of the receiving end demodulation method, each group of receiving matrices is decoded using a two-dimensional Rake receiver in the delay-Doppler domain, including: Based on the receiving matrix of each group, the contributions of different channel components in the delay-Doppler domain are combined in a weighted manner, and the channel tap gain is used as the weight to calculate the vector element r related to the frequency domain. β,ε For weighted calculation, the expression is: In formula (7), q u,v represents the uth row and vth column element of the decision matrix Q, h ε represents the channel matrix gain, h ε (t) represents h ε The tth element in r β,ε (m) represents the vector element r β,ε The mth element in ; By comparing the maximum absolute values ​​of the n×W branch outputs, the maximum value index is determined, which is expressed as: In formula (8), represents the determined maximum value index, Q represents the decision matrix, which is q u,v The set of u represents the row number of the decision matrix, u=[1,2,…,n], n represents the number of rows of the decision matrix, and its value is v represents the column number of the decision matrix, v = [1, 2, ..., W], W represents the number of decision matrix columns, and its value is Based on the determined maximum index Obtain the corresponding index bit and mapping bit, and determine the corresponding phase information bit; specifically, the decision rule of the phase information bit is expressed as: In formula (9), represents the information bit of the decision, u represents the row number of the decision matrix, v represents the column number of the decision matrix, and Q represents the decision matrix.

Images