Underwater sound OTFS communication system based on double-domain joint channel estimation
By adopting dual-domain joint channel estimation technology in the water acoustic OTFS communication system, using channel estimation and compensation in the Doppler delay domain and the time delay domain, the problem of interference between carriers in time-varying channels is solved, and the problem of OFDM technology is improved. The spectrum efficiency and stability of the communication system are achieved.
Patent Information
- Application Number
- CN202411010292.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-26
- Publication Date
- 2024-12-03
AI Technical Summary
OFDM technology in high-speed hydroacoustic communication systems causes inter-carrier interference due to time-varying channels, reducing spectrum efficiency. The existing methods shorten the symbol duration to reduce channel changes, but lead to a decrease in spectrum efficiency.
Using a water acoustic OTFS communication system based on dual-domain joint channel estimation, channel estimation is performed in the Doppler time delay domain and the time delay domain, and channel compensation is performed using pilot carriers and pilot sequences, and the time-changing channel is converted to non-fading channel, realizes the constant channel gain.
It improves the communication performance and stability of the hydroacoustic communication system in high-speed mobile and fast time-varying channel environments, and makes up for the performance degradation of OFDM technology under complex time-varying channel conditions.
Smart Images

Figure CN119071110A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of underwater acoustic channels, and in particular to an underwater acoustic OTFS communication system based on dual-domain joint channel estimation. Background Art
[0002] Acoustic waves are currently the only effective carrier for long-distance information transmission underwater. Underwater acoustic channels face many challenges, such as limited bandwidth, large delay spread, frequency selective fading, and susceptibility to the Doppler effect. Since OFDM (Orthogonal Frequency Division Multiplex, OFDM) modulation and demodulation have low implementation complexity, they have been widely used in high-speed underwater acoustic communication systems. Orthogonal frequency division multiplexing is a multi-carrier modulation technology in which the multiplexed subcarrier frequencies are orthogonal to each other. Due to orthogonality, the peak value of each subcarrier will coincide with the zero value of other subcarriers. Therefore, even if the subcarriers partially overlap with each other in the spectrum, they will not be interfered with by other subcarriers at the peak of the required subcarrier. At the same time, the OFDM system avoids the bandwidth efficiency loss that is common in systems using non-orthogonal carrier sets, greatly improving the spectrum efficiency of underwater acoustic communication systems compared to early systems.
[0003] However, time-varying channels destroy the orthogonality between subcarriers, resulting in inter-carrier interference, which greatly reduces the performance of OFDM. The existing method is to shorten the OFDM symbol duration to alleviate the channel variation on each symbol, however, this method will lead to a serious reduction in spectrum efficiency. Summary of the invention
[0004] In view of the above problems and technical requirements, this application proposes an underwater acoustic OTFS communication system based on dual-domain joint channel estimation. The technical solution of this application is as follows:
[0005] An underwater acoustic OTFS communication system based on dual-domain joint channel estimation, the underwater acoustic OTFS communication system includes a transmitter and a receiver connected to an underwater acoustic time-varying channel:
[0006] The transmitter is used for performing symbol modulation on a binary bit stream to be transmitted to obtain a symbol vector, placing the symbol vector in a Doppler grid in a Doppler delay domain to obtain each OTFS frame of a Doppler two-dimensional transmission sequence, wherein the Doppler grid includes a grid of M rows and N columns, M is the number of subcarriers, N is the number of OTFS symbols, and the subcarriers in the Doppler two-dimensional transmission sequence include a data carrier and an inserted null carrier and a pilot carrier; performing OTFS modulation on the Doppler two-dimensional transmission sequence to obtain a time domain transmission signal s(t) and transmitting the signal through an underwater acoustic time-varying channel;
[0007] A receiver is used to perform OTFS demodulation on a time domain received signal r(t) received from an underwater acoustic time-varying channel to obtain a time delay domain received signal and a Doppler delay domain received signal; in the Doppler delay domain, an estimated value of a delay tap coefficient of the underwater acoustic time-varying channel is obtained according to a pilot carrier of the Doppler delay domain received signal. In the time delay domain, the pilot carrier and delay tap coefficient estimation value of the time delay domain received signal are obtained. The estimated value of the full channel impulse response is obtained, and signal detection is performed according to the time delay domain received signal and the estimated value of the full channel impulse response, and a binary bit stream estimated value is output.
[0008] A further technical solution is that the first several symbol vector rows of the Doppler grid are data carriers, the last several symbol vector rows of the Doppler grid are inserted empty carriers, and the pilot carrier is inserted at the empty carriers of the last several symbol vector rows.
[0009] A further technical solution is that the OTFS symbol x(q, l) of any l-th row and q-th column in the Doppler grid for each OTFS frame of the Doppler two-dimensional transmission sequence is:
[0010]
[0011] Among them, x d (q, l) represents the data carrier OTFS symbol at the lth row and qth column in the Doppler grid, x p Indicates the mth p Line n p The column represents the OTFS symbol of the inserted pilot carrier, 0 represents the OTFS symbol of the inserted null carrier, l g is the maximum channel delay spread index and reaches the predetermined threshold, n p It is any integer value in the range [0, N-1].
[0012] A further technical solution is to determine the maximum channel delay spread index l g include:
[0013] The maximum additional delay τ is determined by simulation to ensure that the communication reliability and communication rate of the underwater acoustic time-varying channel meet the design requirements. β , the maximum additional delay τ β is the multipath power decaying from the initial value to The delay of is greater than the maximum multipath power P max Small βdB multipath power;
[0014] With the maximum additional delay τ β The corresponding delay tap index is used as the maximum channel delay spread index l g.
[0015] A further technical solution is that the receiver obtains the delay tap coefficient estimation value l of the underwater acoustic time-varying channel in the Doppler delay domain according to the pilot carrier of the Doppler delay domain received signal, including:
[0016] The relationship between the OTFS symbol y(q, l) of any l-th row and q-th column in the Doppler delay domain received signal and the OTFS symbol x(q, l) of the l-th row and q-th column in the Doppler two-dimensional transmission sequence is determined as follows:
[0017]
[0018] Among them, q max is the range of fractional Doppler coefficients after Doppler compensation; a line with an amplitude of ξ q′,l′ =h′(q′,l′)η(q,l) path, whose Doppler tap coefficient is q′ and delay tap coefficient is l′, then d(q′,l′)=1, otherwise d(q′,l′)=0; where, h(q′, l′) represents the channel amplitude with Doppler tap coefficient q′ and delay tap coefficient l′, η(q, l) represents the channel phase with Doppler tap coefficient q and delay tap coefficient l, w(q, l) represents the channel noise with Doppler tap coefficient q and delay tap coefficient l, j is an imaginary parameter, and mod represents the remainder;
[0019] Based on the relationship between y(q, l) and x(q, l), the relationship between the pilot carrier in the Doppler delay domain received signal and the Doppler two-dimensional transmission sequence is determined as follows:
[0020] Y p =X p h+w p
[0021] Among them, w p is the noise:
[0022] Y p =[y(0, m p ), y(0, m p +1),…y(0,m p +l g ),
[0023] y(1,m p ), g(1, m p +1),…y(1,m p +lg),
[0024] ...y(N-1, m p +l g )] Y
[0025]
[0026] For any x q′,l′ have:
[0027]
[0028] The compressed sensing method is used to estimate the sparse equivalent channel h and reconstruct the delay tap coefficient estimate.
[0029] A further technical solution is that the transmitter performs OTFS modulation on the Doppler two-dimensional transmission sequence to obtain a time domain transmission signal s(t) including:
[0030] The Doppler two-dimensional transmission sequence is subjected to an inverse symplectic finite Fourier transform to obtain each data frame of the delayed two-dimensional transmission sequence placed in the delay grid, the delay grid includes a grid of M rows and N columns, the subcarriers in each data frame of the delayed two-dimensional transmission sequence include a data carrier, a null carrier and a pilot carrier obtained by the inverse symplectic finite Fourier transform, and the subcarriers in the delay grid in each data frame of the delayed two-dimensional transmission sequence from the 0th row to the M-(2l g +1)-1 row is the data carrier, l=m p =M-(l g +1) row is the pilot carrier, the M-(2l g +1) Go to the mth row p -1 row and mth row p The +1th to M-1th rows are empty carriers;
[0031] The time-delayed two-dimensional transmission sequence is subjected to Heisenberg transformation and parallel-to-serial conversion to obtain the time-domain transmission signal s(t).
[0032] A further technical solution is that each data frame of the received signal in the time delay domain is in the delay grid of M rows and N columns from row 0 to row m. p -1 line is the data carrier, mth line is the p Rows from row 1 to row M-1 are pilot carriers; the receiver obtains the full channel impulse response estimate in the time delay domain including:
[0033] The correlation between the pilot carrier in the time delay domain received signal and the pilot carrier in the time delay two-dimensional transmission sequence is determined according to the delay tap coefficient estimate l, and the full channel impulse response estimate is calculated according to the correlation using the least squares channel estimation method.
[0034] A further technical solution is that the receiver uses the least squares channel estimation method to calculate the full channel impulse response estimation value according to the correlation relationship, including:
[0035] The correlation between the pilot carrier in the time delay domain received signal and the pilot carrier in the time delay two-dimensional transmission sequence is determined as:
[0036]
[0037] in, It is the first in the time-delayed two-dimensional transmission sequence. The pilot carrier of row qth column, is the first The pilot carrier of the row and column q, 0≤q≤N-1; L is the set of delay tap coefficients in the delay Doppler domain; is the channel impulse response function after discretization sampling;
[0038] According to the correlation relationship, the least squares channel estimation method is used. Calculate the estimated value of the channel impulse response function corresponding to the estimated value of the delay tap coefficient l after discretization sampling
[0039] Through spline domain interpolation The full channel impulse response estimate is obtained by interpolation.
[0040] A further technical solution is that the value of the number N of OTFS symbols at least meets the Doppler reflection requirement, and the values of the number M of subcarriers and the number N of OTFS symbols meet the requirement that the OTFS frame duration is less than the channel coherence time of the underwater acoustic time-varying channel.
[0041] Its further technical solution is that the receiver performs OTFS demodulation on the time domain received signal r(t) to obtain a time delay domain received signal and a Doppler delay domain received signal, including: performing serial-to-parallel conversion on the time domain received signal r(t) to obtain a time delay domain received signal, and performing Wigner transform and symplectic finite Fourier transform on the time delay domain received signal to obtain a Doppler delay domain received signal.
[0042] The beneficial technical effects of this application are:
[0043] The present application discloses an underwater acoustic OTFS communication system based on dual-domain joint channel estimation. The OTFS technology is applied to the design of the underwater acoustic communication system. The information symbols sent by the OTFS communication system are generated in the Doppler delay domain. The received signal can be considered to be obtained by two-dimensional convolution of the transmitted signal and the Doppler delay domain channel. The Doppler delay domain has strong delay tolerance and Doppler tolerance, and has the potential for full diversity, which is one of the keys to support reliable communication. In addition, OTFS modulation can convert a time-varying fading channel into a non-fading and time-independent channel in the Doppler delay domain, so that the channel gain of all modulation symbols therein is substantially constant. The present application designs a frame structure for an underwater acoustic OTFS communication system, and then performs Doppler estimation and compensation processing on the received signal in the receiver, so that the underwater acoustic time-varying channel experienced by the signal satisfies the under-expansion condition, and finally performs dual-domain joint channel estimation in the Doppler delay domain and the time delay domain on this basis. This method can effectively improve the communication performance and stability of the underwater acoustic communication system in an environment where the communication machine moves at high speed and the underwater acoustic channel changes rapidly, and make up for the defect that the system performance of OFDM technology is significantly reduced under the conditions of complex and fast time-varying underwater acoustic channels. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 It is a schematic diagram of a system block diagram of an underwater acoustic OTFS communication system according to an embodiment of the present application.
[0045] Figure 2 It is a schematic diagram of the data frame structure involved in an underwater acoustic OTFS communication system of an embodiment of the present application. DETAILED DESCRIPTION
[0046] The specific implementation of the present application is further described below in conjunction with the accompanying drawings.
[0047] The present application discloses an underwater acoustic OTFS communication system based on dual-domain joint channel estimation. The underwater acoustic OTFS communication system includes a transmitter and a receiver connected to an underwater acoustic time-varying channel. Please refer to Figure 1 The logical block diagram of the transmitter and receiver is shown:
[0048] 1. Transmitter
[0049] The transmitter is used to obtain the binary bit stream Sin to be transmitted, and then symbol modulates the binary bit stream Sin to be transmitted to obtain a symbol vector, and then places the symbol vector in the Doppler grid in the Doppler delay domain to obtain each OTFS frame of the Doppler two-dimensional transmission sequence. A common practice is to use an LDPC encoder to encode the binary bit stream Sin to be transmitted, and then modulate the codeword into a symbol vector composed of OTFS symbols through a symbol modulator.
[0050] Please refer to Figure 2The Doppler grid consists of M rows and N columns, where M is the number of subcarriers and the length of the delay domain, and the resolution is Δf is the subcarrier spacing. N is the number of OTFS symbols and the length of the Doppler domain, with a resolution of T is the OTFS symbol period, and Δf and T satisfy TΔdf=1.
[0051] The underwater acoustic OTFS communication system of the present application is implemented based on the premise that the channel delay-Doppler response remains constant during the duration of the OTFS frame, which means that the full channel impulse response and related parameters remain unchanged in an OTFS frame, which requires that the length of a single OTFS frame does not exceed the channel coherence time of the underwater acoustic time-varying channel, and the values of M and N directly affect the length of the OTFS frame, so the values of M and N are also very important. The channel coherence time of the underwater acoustic time-varying channel is often very short, which requires that N and M cannot be too large at the same time to ensure reliable communication in time-varying and overspread channels. In the communication scenario of the underwater acoustic time-varying channel, the bandwidth of the underwater acoustic OTFS communication system is limited, and the Doppler spread maximum of the under-spread underwater acoustic time-varying channel is significant. Therefore, under under-spread conditions, a smaller N value cannot meet the Doppler resolution requirements. Therefore, for the underwater acoustic OTFS communication system of the present application, considering that M is sufficient to avoid obvious fractional delays, it is possible to reduce M and increase N to enhance the Doppler tolerance of the system. Therefore, when designing the values of M and N, the value of the number of OTFS symbols N at least meets the Doppler reflection requirement, and ensures that the values of the number of subcarriers M and the number of OTFS symbols N satisfy that the OTFS frame duration is less than the channel coherence time of the underwater acoustic time-varying channel.
[0052] The subcarriers in the Doppler two-dimensional transmission sequence include data carriers and inserted null carriers and pilot carriers. In one embodiment, in order to achieve better communication effect, the OTFS frame structure is designed for the Doppler two-dimensional transmission sequence. Please refer to Figure 2 In the Doppler two-dimensional transmission sequence, the first several symbol vector rows located in the Doppler grid are data carriers. In order to avoid inter-block interference in the time domain, empty carriers are inserted in the last several symbol vector rows of the Doppler grid. In addition, it is necessary to insert a pilot carrier for channel estimation. This embodiment directly inserts the pilot carrier into the empty carrier of the last several symbol vector rows to avoid additional data overhead caused by the pilot carrier.
[0053] Furthermore, the pilot carrier is inserted at the middle row position in the empty carrier, so that the pilot carrier and the data carrier of the current frame are separated by several rows of empty carriers to avoid interference between the pilot carrier and the data carrier of the current frame. And the pilot carrier and the data carrier of the next frame are separated by several rows of empty carriers to avoid interference between the pilot carrier and the data carrier of the next frame. The present application uses x(q, l) to represent the OTFS symbol of any lth row and qth column in the Doppler grid of a single OTFS frame of a Doppler two-dimensional transmission sequence, 0≤q≤N-1, 0≤q≤N-1. Then Figure 2 As shown, x(q, l) at different positions is:
[0054]
[0055] Among them, x d (q, l) represents the data carrier OTFS symbol at the lth row and qth column in the Doppler grid. p Indicates the mth p Line n p The OTFS symbol of the inserted pilot carrier is shown in the column. 0 represents the OTFS symbol of the inserted null carrier. p It is any integer value in the range of [0, N-1], indicating that the mth p A pilot carrier is inserted into any column of a row.
[0056] l g is the maximum channel delay spread index and reaches the predetermined threshold, the last 2l of the Doppler grid g +1 symbol vector row is inserted into the empty carrier, and is composed of Figure 2 It can be seen that in each OTFS frame of the Doppler two-dimensional transmission sequence, the pilot carrier is located at the mth p The interval between the line and the data carrier of the current frame is l g The interval between the line carrier and the data carrier of the next frame is l g Line empty carrier. g The predetermined threshold that needs to be reached is a minimum value for avoiding interference between the pilot carrier and the data carrier, thereby avoiding the influence of the inserted pilot carrier on the data carrier.
[0057] With maximum delay τ max The corresponding delay tap index is used as the maximum channel delay spread index l g The inter-block interference between the pilot carrier and the data carrier can be completely avoided, but it will lead to a very low communication rate and is not conducive to the communication effect of the underwater acoustic OTFS communication system. The maximum channel delay spread index l g The larger the value, the lower the communication rate, and the maximum channel delay spread index l gThe smaller the value, the more likely it is that inter-block interference will occur and the lower the communication reliability will be. Therefore, it is necessary to comprehensively consider the maximum channel delay spread index l g In one embodiment, the maximum additional delay τ is determined by simulation to ensure that the communication reliability and communication rate of the underwater acoustic time-varying channel meet the design requirements. β , the maximum additional delay τ β is the multipath power decaying from the initial value to The delay of is greater than the maximum multipath power P max The multipath power of small βdB, that is Then with the maximum additional delay τ β The corresponding delay tap index is used as the maximum channel delay spread index l g , thus we can determine the minimum value of l when ensuring the communication reliability and communication rate of the underwater acoustic time-varying channel meet the design requirements g .
[0058] Then, the Doppler two-dimensional transmission sequence is modulated by OTFS to obtain the time domain transmission signal s(t) and sent through the underwater acoustic time-varying channel. Including:
[0059] Perform an inverse symplectic finite Fourier transform (ISFFT) on the Doppler two-dimensional transmission sequence to obtain each data frame of the delayed two-dimensional transmission sequence placed in the delay grid. The delay grid includes M rows and N columns of grids, and the length of the delay domain in the delay grid is M and the length of the time domain is N. The subcarriers of each data frame in the delayed two-dimensional transmission sequence include data carriers, null carriers and pilot carriers obtained by inverse symplectic finite Fourier transform. Please refer to Figure 2 Schematic diagram of the delay grid of each data frame in the two-dimensional transmission sequence of the delay, from the 0th row to the M-(2l g +1)-1 row is the data carrier, M=m p =M-(l g +1) row is the pilot carrier, the M-(2l g +1) Go to the mth row p -1 row and mth row p +1 to M-1 rows are empty carriers. Represents the symbol located in the lth row and qth column of the delay grid in a single data frame in the delayed two-dimensional transmission sequence.
[0060] Then the delayed two-dimensional transmission sequence is subjected to Heisenberg transform and parallel-to-serial conversion. In practice, it may also need to undergo up-conversion processing. In addition, since the underwater acoustic time-varying channel is under-expanded, it has a long delay spread and a significant Doppler effect. Therefore, it is often necessary to insert a linear frequency modulation signal in the frame header for signal synchronization, and then obtain the time domain transmission signal s(t) and send it through the underwater acoustic time-varying channel.
[0061] Receiver
[0062] The receiver receives the time domain received signal r(t) from the underwater acoustic time-varying channel, please refer to Figure 2 Due to the time-varying effect of the underwater acoustic time-varying channel, the signal content of the received time-domain received signal r(t) and the time-domain transmitted signal s(t) sent by the transmitter through the underwater acoustic time-varying channel has changed.
[0063] Perform OTFS demodulation on the time domain received signal r(t) to obtain the time delay domain received signal and the Doppler delay domain received signal. Including:
[0064] First, the time domain received signal r(t) is preprocessed by down-conversion and then converted into a time delay domain received signal by serial-to-parallel conversion. The signal form of the time delay domain received signal is similar to the time delay two-dimensional transmission sequence. Each data frame of the time delay domain received signal includes symbols placed in a delay grid of M rows and N columns. Represents the symbol located in the lth row and qth column of the delay grid in each data frame of the received signal in the time delay domain. Figure 2 Due to the delay expansion effect of the underwater acoustic time-varying channel, the data carrier in the delayed two-dimensional transmission sequence will expand and leak to the empty carrier, and the pilot carrier will also expand and leak to the empty carrier. Therefore, each data frame of the received signal in the time delay domain is in the delay grid with M rows and N columns from row 0 to row m. p -1 line is the data carrier, mth line is the p Rows up to row M-1 are pilot carriers.
[0065] Then, the time delay domain received signal is subjected to Wigner transform and Symplectic Finite Fourier Transform (SFFT) to obtain the Doppler delay domain received signal. The signal form of the Doppler delay domain received signal is also similar to the Doppler two-dimensional transmission sequence. Each OTFS frame of the Doppler delay domain received signal includes symbols placed in the Doppler grid of M rows and N columns. y(q, l) represents any OTFS symbol of the lth row and qth column in the Doppler delay domain received signal.
[0066] The present application combines the Doppler delay domain and the time delay domain to perform dual-domain joint channel estimation. First, in the Doppler delay domain, the delay tap coefficient estimation value of the underwater acoustic time-varying channel is obtained according to the pilot carrier of the Doppler delay domain received signal. include:
[0067] The relationship between the OTFS symbol y(q, l) of any l-th row and q-th column in the Doppler delay domain received signal and the OTFS symbol x(q, l) of the l-th row and q-th column in the Doppler two-dimensional transmission sequence is determined as follows:
[0068]
[0069] Among them, q max is the range of fractional Doppler coefficients after Doppler compensation. q′,l′ =h′(q′, l′)η(q, l), the Doppler tap coefficient is q′ and the delay tap coefficient is l′, then d(q′, l′)=1, otherwise d(q′, l′)=0. h(q′, l′) represents the channel amplitude with Doppler tap coefficient q′ and delay tap coefficient l′, j is an imaginary parameter. η(q, l) represents the channel phase with Doppler tap coefficient q and delay tap coefficient l, w(q, l) represents the channel noise with Doppler tap coefficient q and delay tap coefficient l. mod represents the remainder.
[0070] Based on the above relationship between y(q, l) and x(q, l), the relationship between the pilot carrier in the Doppler delay domain received signal and the Doppler two-dimensional transmission sequence can be determined as follows:
[0071] Y p =X p h+w p
[0072] Among them, w p is the noise:
[0073] Y p =[y(0, mp), y(0, m p +1),…y(0,m p +l g ),
[0074] y(1,m p ), y(1, m p +1),…y(1,m p +l g ),
[0075] …g(N-1,m p +l g )] T
[0076]
[0077] X p For any x q′,l′ have:
[0078]
[0079] Since underwater acoustic time-varying signals are usually sparse in the Doppler delay domain, that is, their equivalent channel h contains only a limited number of non-zero elements, Y p As a known measurement, X p As the observation matrix, the compressed sensing method can be used to estimate the sparse equivalent channel h and reconstruct the delay tap coefficient estimate l.
[0080] Then, in the time delay domain, the full channel impulse response estimation value is obtained according to the pilot carrier of the time delay domain received signal and the delay tap coefficient estimation value l. This includes: determining the correlation between the pilot carrier in the time delay domain received signal and the pilot carrier in the delayed two-dimensional transmission sequence according to the delay tap coefficient estimation value l, and then calculating the full channel impulse response estimation value according to the correlation using the least squares channel estimation method according to the correlation.
[0081] The correlation between the pilot carrier in the time delay domain received signal and the pilot carrier in the time delay two-dimensional transmission sequence is:
[0082]
[0083] in, It is the first in the time-delayed two-dimensional transmission sequence. The pilot carrier of row qth column, is the first The pilot carrier of the row and column q, 0≤q≤N-1. L is the set of delay tap coefficients in the delay-Doppler domain. is the channel impulse response function after discretization sampling;
[0084] Then, based on this correlation, the least squares channel estimation method can be used to calculate the corresponding delay tap coefficient estimate after discretization sampling according to the following formula: The estimated value of the channel impulse response function
[0085]
[0086] Finally, the spline domain interpolation The full channel impulse response estimate can be obtained by interpolation.
[0087] After obtaining the full channel impulse response estimate, signal detection can be performed based on the time delay domain received signal and the full channel impulse response estimate, and a binary bit stream estimate can be output. This part is the same as the method implemented by the existing OTFS system reusing the equalizer and decoder, and this application will not go into details.
[0088] The above is only a preferred embodiment of the present application, and the present application is not limited to the above embodiments. It is understood that other improvements and changes directly derived or associated by those skilled in the art without departing from the spirit and concept of the present application should be considered to be included in the protection scope of the present application.
Claims
1. An underwater acoustic OTFS communication system based on dual-domain joint channel estimation, characterized in that: The underwater acoustic OTFS communication system includes a transmitter and a receiver connected to an underwater acoustic time-varying channel: The transmitter is used for performing symbol modulation on a binary bit stream to be transmitted to obtain a symbol vector, placing the symbol vector in a Doppler grid in a Doppler delay domain to obtain each OTFS frame of a Doppler two-dimensional transmission sequence, wherein the Doppler grid includes a grid with M rows and N columns, M is the number of subcarriers, N is the number of OTFS symbols, and the subcarriers in the Doppler two-dimensional transmission sequence include data carriers and inserted null carriers and pilot carriers; performing OTFS modulation on the Doppler two-dimensional transmission sequence to obtain a time domain transmission signal s(t) and transmitting the signal through the underwater acoustic time-varying channel; The receiver is used to perform OTFS demodulation on a time domain received signal r(t) received from an underwater acoustic time-varying channel to obtain a time delay domain received signal and a Doppler delay domain received signal; in the Doppler delay domain, a delay tap coefficient estimation value l of the underwater acoustic time-varying channel is obtained according to a pilot carrier of the Doppler delay domain received signal; in the time delay domain, a full channel impulse response estimation value is obtained according to a pilot carrier and a delay tap coefficient estimation value l of the time delay domain received signal; and signal detection is performed according to the time delay domain received signal and the full channel impulse response estimation value, and a binary bit stream estimation value is output.
2. The underwater acoustic OTFS communication system according to claim 1, characterized in that: The first several symbol vector rows of the Doppler grid are data carriers, the last several symbol vector rows of the Doppler grid are inserted empty carriers, and the pilot carriers are inserted at the empty carriers of the last several symbol vector rows.
3. The underwater acoustic OTFS communication system according to claim 2, characterized in that: The OTFS symbol x(q, l) of any l-th row and q-th column in the Doppler grid of each OTFS frame of the Doppler two-dimensional transmission sequence is: Among them, x d (q, l) represents the data carrier OTFS symbol at the lth row and qth column in the Doppler grid, x p Indicates the mth p Line n p The column represents the OTFS symbol of the inserted pilot carrier, 0 represents the OTFS symbol of the inserted null carrier, l g is the maximum channel delay spread index and reaches the predetermined threshold, n p It is any integer value in the range of [0, N-1].
4. The underwater acoustic OTFS communication system according to claim 3, characterized in that: Determine the maximum channel delay spread index l g include: The maximum additional delay τ is determined by simulation to ensure that the communication reliability and communication rate of the underwater acoustic time-varying channel meet the design requirements. β , the maximum additional delay τ β is the multipath power decaying from the initial value to The delay of is greater than the maximum multipath power P max Small βdB multipath power; With the maximum additional delay τ β The corresponding delay tap index is used as the maximum channel delay spread index l g .
5. The underwater acoustic OTFS communication system according to claim 3, characterized in that: The receiver obtains a delay tap coefficient estimation value of an underwater acoustic time-varying channel in a Doppler delay domain according to a pilot carrier of a Doppler delay domain received signal. include: The relationship between the OTFS symbol y(q, l) of any l-th row and q-th column in the Doppler delay domain received signal and the OTFS symbol x(q, l) of the l-th row and q-th column in the Doppler two-dimensional transmission sequence is determined as follows: Among them, q max is the range of fractional Doppler coefficients after Doppler compensation; a line with an amplitude of ξ q′,l′ =h′(q′,l′)η(q,l) path, whose Doppler tap coefficient is q′ and delay tap coefficient is l′, then d(q′,l′)=1, otherwise d(q′,l′)=0; where, h(q′, l′) represents the channel amplitude with Doppler tap coefficient q′ and delay tap coefficient l′, η(q, l) represents the channel phase with Doppler tap coefficient q and delay tap coefficient l, w(q, l) represents the channel noise with Doppler tap coefficient q and delay tap coefficient l, j is an imaginary parameter, and mod represents the remainder; Based on the relationship between y(q, l) and x(q, l), the relationship between the pilot carrier in the Doppler delay domain received signal and the Doppler two-dimensional transmission sequence is determined as follows: Y p =X p h+w p Among them, w p is the noise: Y p =[y(0,m p ),y(0,m p +1),…y(0,m p +l g ), y(1,m p ),y(1,m p +1),…y(1,m p +l g ),…y(N-1,m p +l g )] T For any x q′,l′ have: The compressed sensing method is used to estimate the sparse equivalent channel h and reconstruct the delay tap coefficient estimate.
6. The underwater acoustic OTFS communication system according to claim 3, characterized in that: The transmitter performs OTFS modulation on the Doppler two-dimensional transmission sequence to obtain a time domain transmission signal s(t), including: The Doppler two-dimensional transmission sequence is subjected to an inverse symplectic finite Fourier transform to obtain each data frame of the delayed two-dimensional transmission sequence placed in a delay grid, wherein the delay grid includes a grid of M rows and N columns, and the subcarriers in each data frame of the delayed two-dimensional transmission sequence include a data carrier, a null carrier, and a pilot carrier obtained by the inverse symplectic finite Fourier transform, and the subcarriers in the delay grid in each data frame of the delayed two-dimensional transmission sequence from the 0th row to the M-(2l g +1)-1 row is the data carrier, l=m p =M-(l g +1) row is the pilot carrier, the M-(2l g +1) Go to the mth row p -1 row and mth row p The +1th to M-1th rows are empty carriers; The time-delayed two-dimensional transmission sequence is subjected to Heisenberg transformation and parallel-to-serial conversion to obtain a time-domain transmission signal s(t).
7. The underwater acoustic OTFS communication system according to claim 6, characterized in that: Each data frame of the received signal in the time delay domain is from row 0 to row m in the delay grid of M rows and N columns. p -1 line is the data carrier, mth line is the p Rows from row 1 to row M-1 are pilot carriers; The receiver obtains a full channel impulse response estimation value in the time delay domain, including: According to the delay tap coefficient estimate The correlation between the pilot carrier in the time delay domain received signal and the pilot carrier in the delayed two-dimensional transmission sequence is determined, and a full channel impulse response estimation value is calculated based on the correlation using a least squares channel estimation method.
8. The underwater acoustic communication system according to claim 7, characterized in that: The receiver uses the least squares channel estimation method to calculate the full channel impulse response estimation value according to the correlation relationship, including: The correlation between the pilot carrier in the time delay domain received signal and the pilot carrier in the time delay two-dimensional transmission sequence is determined as: in, It is the first in the time-delayed two-dimensional transmission sequence. The pilot carrier of row qth column, is the first signal in the time delay domain received The pilot carrier of the row and column q, 0≤q≤N-1; L is the set of delay tap coefficients in the delay Doppler domain; is the channel impulse response function after discretization sampling; According to the correlation, the least squares channel estimation method is used according to Calculate the corresponding delay tap coefficient estimate after discretization sampling The estimated value of the channel impulse response function Through spline domain interpolation The full channel impulse response estimate is obtained by interpolation.
9. The underwater acoustic OTFS communication system according to claim 1, characterized in that: The value of the number N of OTFS symbols at least meets the Doppler reflection requirement, and the values of the number M of subcarriers and the number N of OTFS symbols meet the requirement that the duration of the OTFS frame is less than the channel coherence time of the underwater acoustic time-varying channel.
10. The underwater acoustic OTFS communication system according to claim 1, characterized in that: The receiver performs OTFS demodulation on the time domain received signal r(t) to obtain a time delay domain received signal and a Doppler delay domain received signal, including: performing serial-to-parallel conversion on the time domain received signal r(t) to obtain a time delay domain received signal, and performing Wigner transform and symplectic finite Fourier transform on the time delay domain received signal to obtain the Doppler delay domain received signal.
Citation Information
Cited By
Inter-Doppler interference suppression method based on Gaussian learning factor particle swarm optimization
CN119892157A
