MIMO-OTFS combined time synchronization and channel estimation method

Abstract

The invention relates to the technical field of orthogonal time-frequency-space communication systems, in particular to an MIMO-OTFS combined time synchronization and channel estimation method, which comprises the following steps: for a plurality of transmitting antennas, generating a pilot frequency sequence based on a bipolar maximum length sequence; obtaining time delay extension of the channel according to the prior channel information; constructing a time delay-Doppler domain sending signal based on the time delay spread and the pilot frequency sequence, performing OTFS modulation on the time delay-Doppler domain sending signal, and then sending the time delay-Doppler domain sending signal; performing time synchronization detection on the received signal transmitted through the channel based on the bipolar maximum length sequence, and acquiring timing offset based on synchronization time positions of multiple paths; based on the synchronization time position and the timing offset of the plurality of paths, obtaining a time delay tap, a Doppler frequency shift and a channel gain of each path of each transmitting antenna; according to the invention, the pilot frequency overhead can be saved in channel estimation and the accuracy of timing offset estimation can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of orthogonal time-frequency-space communication systems, and in particular to a MIMO-OTFS joint time synchronization and channel estimation method. Background Art

[0002] With the rapid development of next-generation wireless networks, the application scenarios of wireless communications present the characteristics of high carrier frequency and high mobility. In such high Doppler frequency shift scenarios, fast time-varying multipath channels, i.e., dual-selective channels, will degrade the performance of widely adopted modulation schemes such as Orthogonal Frequency Division Multiplexing (OFDM). This is because such modulation schemes are based on the assumption of time-invariant channels. To address this problem, the literature [Hadani R, Rakib S, Tsatsanis M, et al. Orthogonal time frequency space modulation [C] / / 2017 IEEE Wireless Communications and Networking Conference (WCNC). IEEE, 2017: 1-6.] provides a promising Orthogonal Time Frequency Space (OTFS) modulation scheme based on the dual-selective channel assumption. OTFS characterizes the channel and modulates information symbols in the two-dimensional Delay-Doppler (DD) domain. Therefore, this new modulation scheme can accurately estimate multipath delay and Doppler frequency shift through channel estimation, and exhibits strong robustness in high-mobility communication scenarios.

[0003] In any practical communication system, including OTFS systems, synchronization and channel estimation are crucial technologies. Regarding OTFS channel estimation, the literature [Raviteja P, Phan KT, Hong Y. Embedded pilot-aided channel estimation for OTFS in delay–Doppler channels[J]. IEEE transactions on vehicular technology, 2019, 68(5): 4906-4917.] provides an embedded pilot-aided algorithm (EPA), which performs channel estimation in the DD domain and solves the fractional Doppler problem by additionally arranging guard symbols. The scheme provided in the literature [Yuan W, Li S, Wei Z, et al. Data-aided channel estimation for OTFS systems with a superimposed pilot and data transmission scheme[J]. IEEE wireless communications letters, 2021, 10(9): 1954-1958.] superimposes pilot and data symbols, and estimates the channel and detects data symbols through iteration. However, these methods do not study the synchronization problem. Moreover, for the MIMO-OTFS system, the pilot overhead of EPA increases linearly with the number of transmitting antennas. Therefore, when the MIMO scale increases, the problem of excessive pilot overhead will arise, and a more efficient solution is needed to ensure the performance of the system.

[0004] Regarding the synchronization problem, the solution provided in the literature [Raviteja P, Phan KT, Hong Y. Embedded pilot-aided channel estimation for OTFS in delay–Doppler channels[J]. IEEE transactions on vehicular technology, 2019, 68(5): 4906-4917.] shows that the embedded pilot signal exhibits the characteristics of constant amplitude and equal phase difference in the delay-time (DT) domain. Utilizing this characteristic, the solution provided in the literature [Bayat M, Farhang A. Time and frequency synchronization for OTFS[J]. IEEE Wireless Communications Letters, 2022, 11(12): 2670-2674.] uses the absolute correlation function to perform timing offset (TO) estimation, thereby determining the strongest path. However, when the first path is not the strongest, this method cannot locate the first multipath component, resulting in TO estimation errors. In addition, due to the poor autocorrelation characteristics of the constant amplitude sequence in the DT domain, this method requires a two-step process in the delay and time dimensions to perform TO estimation.

[0005] For MIMO-OTFS systems, due to the presence of multiple transmit and receive antennas, not only synchronization is required between each pair of transmit-receive antennas, but also interference and mutual influence between antennas must be considered. Errors in timing offset (TO) estimation may accumulate between multiple antennas, resulting in a degradation of the overall system performance. Therefore, the synchronization problem in MIMO-OTFS systems requires a more precise solution to ensure the stability of the system. Summary of the invention

[0006] In view of the above problems, the present invention provides a MIMO-OTFS joint time synchronization and channel estimation method, which solves the technical problems in the prior art that when the MIMO scale increases in the MIMO-OTFS system, the pilot overhead is too large and the accuracy of the timing offset estimation is limited.

[0007] The present invention provides a MIMO-OTFS joint time synchronization and channel estimation method, comprising the following steps:

[0008] Step S1: for multiple transmitting antennas, generate a different pilot sequence for each transmitting antenna based on a bipolar maximum length sequence; obtain the delay spread of the channel according to the priori channel information; construct a delay-Doppler domain transmission signal of each transmitting antenna based on the delay spread and the pilot sequence of each transmitting antenna, and perform OTFS modulation on the delay-Doppler domain transmission signal and then transmit it to each transmitting antenna;

[0009] Step S2, performing time synchronization detection on the received signal after channel transmission based on the bipolar maximum length sequence, obtaining synchronization time positions of multiple paths, and obtaining timing offsets based on the synchronization time positions of the multiple paths;

[0010] Step S3: based on the synchronization time positions of the multiple paths and the timing offset, obtain the delay tap, Doppler frequency shift and channel gain of each path of each transmitting antenna.

[0011] Preferably, step S1 specifically includes:

[0012] Step S1-1: for each transmitting antenna, generate a bipolar maximum length sequence with a different initial state, the bipolar maximum length sequence of each transmitting antenna is in a cyclic shift relationship, and perform discrete Fourier transform on the bipolar maximum length sequence to obtain a pilot sequence of each transmitting antenna;

[0013] Step S1-2, acquiring a delay spread of a channel according to a priori channel information, obtaining a guard symbol width parameter from the delay spread, and determining a guard symbol based on the guard symbol width parameter; determining a data symbol from the data to be transmitted; and constructing a delay-Doppler domain transmission signal of each transmitting antenna from the pilot sequence of each transmitting antenna, the guard symbol, and the data symbol;

[0014] Step S1-3: Perform OTFS modulation on the delay-Doppler domain transmission signal to obtain a time domain transmission signal, which is sent to the channel by each transmitting antenna.

[0015] Preferably, in step S1-1, for the nth t The expression of the pilot sequence is:

[0016]

[0017]

[0018] Among them, n t is the transmit antenna index, n t =1,2,…,N t , N t is the number of transmitting antennas, For nth tThe pilot sequence of the transmitting antennas, k is the Doppler index, DFT{·} represents the discrete Fourier transform, is zero-filled MLS, n is the time index, and the value range is n=0,1,…N-1, is bipolar MLS, N is the number of grid points in the second transform domain;

[0019] The bipolar MLS is constructed by replacing the amplitude 0 in the maximum length sequence with Replace the amplitude 1 in MLS with It is found that the dual-polarity MLS of different transmitting antennas have different initial states, P MLS The preset power.

[0020] Preferably, in step S1-2, the expression of the delay-Doppler domain transmission signal of each transmitting antenna is:

[0021]

[0022] in, For nth t The delay-Doppler domain of the transmitting antennas sends signals, l, k are the delay index and Doppler index respectively, l MLS is the index of the delay position of the pilot sequence, L is the protection symbol width parameter, For nth t The value of L is not less than the delay corresponding to the delay extension.

[0023] Preferably, step S1-3 specifically includes: Perform OTFS modulation to obtain the time domain signal And add a length of L RCP The reduced cyclic prefix is ​​used to obtain the time domain transmission signal, which is sent to the channel by each transmitting antenna.

[0024] Preferably, step S2 specifically includes:

[0025] Step S2-1, extracting a time-delay-time domain row sequence from a received signal after channel transmission to obtain a row sequence;

[0026] Step S2-2: Based on the bipolar maximum length sequence of each transmitting antenna, an autocorrelation positioning method is used to perform time synchronization detection on all the row sequences to obtain the synchronization time position of each path of each transmitting antenna in the channel;

[0027] Step S2-3: determine the minimum value of the synchronization time positions of the paths as the synchronization time position of the first path, and obtain the timing offset based on the synchronization time position of the first path.

[0028] Preferably, step S2-1 specifically includes extracting the row sequence using the following expression:

[0029]

[0030] in, For is the row sequence of the starting time position, is to extract the starting time position of the row sequence, r[·] is the received signal, and M is the number of grid points in the first transform domain.

[0031] Preferably, step S2-2 specifically includes:

[0032] Use the dual-polarity maximum length sequence of each transmitting antenna in turn with Calculate, for the nth t transmit antennas, and obtain the complex exponential sequence

[0033]

[0034] right Perform N-point DFT to obtain the sequence to be judged calculate The normalized maximum value of :

[0035]

[0036] in, For is the normalized maximum value of the row sequence at the starting time position, |·| represents the modulus of the complex number, and max{·} represents the maximum value of the sequence;

[0037] Set the peak judgment threshold The starting time position of all row sequences Perform N on all transmit antennas t Second judgment, when satisfied When the transmitting antenna number n t Record as n t,λ , λ is the antenna index that meets the threshold condition, and all n antennas that meet the condition are t,λ Add to collection Record the corresponding Arrange in order from smallest to largest: As the synchronization time position of all paths in the channel, i is the path index, and the value range is i=1, 2, …, P, and P is the total number of paths that can be distinguished in the channel.

[0038] Preferably, in step S2-3, the calculation expression of the timing offset is:

[0039]

[0040] in, is the timing offset, is the synchronization time position of the first path.

[0041] Preferably, in step S3, the i-th path and the n-th path t,λ The expressions of the delay tap, Doppler frequency shift and channel gain of each path of the transmitting antenna are:

[0042]

[0043]

[0044]

[0045]

[0046] in, For nth t,λ The delay tap of the i-th path of the transmit antenna, For nth t,λ The Doppler shift of the ith path of the transmitting antenna; n t,λ The value range is Y is a row sequence vector, [·] T It means taking the transpose, is the channel gain vector of the i-th path, Expressed as N n For synchronized antenna assembly The number of elements of Is a collection The nth element in the matrix G is the design matrix, and the nth row and nth t,λ The column elements are For nth t,λ Dual-polarity MLS with 1 transmit antenna;

[0047] Indicates that the linear minimum mean square error method is used to estimate the channel gain vector, where σ 2 is the white noise power, σ 2 It is obtained from the prior channel information, and I is the unit matrix.

[0048] Summarize the channel parameter set n t =1,2,…,N t , as the channel estimation result.

[0049] Compared with the prior art, the present invention has at least the following beneficial effects:

[0050] (1) The present invention sets an MLS with different initial states as a pilot signal for each transmitting antenna of the MIMO system, and sets the MLS of each transmitting antenna to a cyclic shift relationship. When the number of transmitting antennas is less than the number of Doppler grid points, the overhead is consistent with that of a single transmitting antenna. Therefore, the pilot overhead of the present invention avoids a linear increase with the number of transmitting antennas, thereby saving the pilot overhead.

[0051] (2) The present invention uses MLS as a pilot signal and performs correlation operation between the received signal and the local MLS in the DT domain, thereby achieving time synchronization without TO estimation in the time dimension.

[0052] (3) The present invention uses the same pilot signal to provide a closed-form solution for channel gain and off-grid Doppler estimation, and completes channel estimation using the results obtained by time-synchronous correlation operations.

[0053] (4) The present invention adopts a grid-free scheme to estimate Doppler frequency shift, thereby enhancing the estimation accuracy in the case of fractional Doppler. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The drawings are only for the purpose of illustrating particular embodiments and are not to be construed as limiting the invention.

[0055] Figure 1 This is a flow chart of the MIMO-OTFS joint time synchronization and channel estimation method provided by the present invention.

[0056] Figure 2 A schematic diagram of the pilot patterns of the transmitter and receiver provided by the present invention.

[0057] Figure 3 A schematic diagram of extracting a received signal by the time synchronization method provided by the present invention.

[0058] Figure 4 The present invention provides a channel estimation method and a schematic diagram of the variation trend of the bit error rate with the data signal-to-noise ratio under perfect time synchronization and channel state information. DETAILED DESCRIPTION

[0059] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. In addition, the present invention can also be implemented in other ways different from those described herein, and therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0060] In order to solve the time synchronization and channel estimation problems of MIMO-OTFS system, the present invention allocates maximum length sequences (MLS) of different initial states as pilot signals to multiple transmitting antennas, uses the autocorrelation characteristics of MLS to estimate the TO and channel delay taps of each transmitting antenna, and provides closed-form estimation expressions of Doppler frequency shift and channel gain of each propagation path of each transmitting antenna. Compared with the existing time synchronization and channel estimation methods, the present invention provides a new joint time synchronization channel estimation method, realizes accurate time synchronization and channel estimation under MIMO-OTFS system, and saves the pilot overhead of multiple transmitting antennas.

[0061] For the transmitter of the MIMO-OTFS communication system, for each transmitting antenna, the OTFS time domain signal can be obtained by performing an inverse finite symplectic Fourier transform (ISFFT) on the two-dimensional delay-Doppler (DD) domain signal and then performing a Heisenberg transform. When both the transmission pulse and the reception pulse are rectangular windows, the generation process of the OTFS time domain signal can be simplified. The simplified generation process is: perform an inverse discrete Fourier transform (IDFT) on the DD domain signal in the Doppler dimension to obtain a two-dimensional delay-time (DT) domain signal, and then output the DT domain signal in series to obtain the OTFS time domain signal.

[0062] The data symbol is obtained by mapping the data to be transmitted to a specific waveform symbol through modulation. The data symbol represents the data to be transmitted. DD [l,k] is a DD domain transmitted signal consisting of quadrature amplitude modulation (QAM) data symbols, such as Figure 2 As shown, the DD domain signal is represented as a two-dimensional cell of size MxN, where l and k are delay index and Doppler index respectively, and the value ranges are l=0,1,…,M-1, k=0,1,…N-1 respectively, where M and N are the number of grid points in the first transform domain and the number of grid points in the second transform domain respectively. The process of transforming the DD domain signal to obtain the DT domain signal can be expressed as: Among them, X DT [l,n] is the DT domain transmission signal obtained by transformation, n is the time index, the value range is n=0,1,…N-1, k′ is the Doppler index used for accumulation calculation. In order to obtain the OTFS time domain signal, the DT domain signal needs to be output serially in columns, and the process is expressed as: s[n′]=X DT [l,n], n'=l+nM, where s[n'] is the OTFS time domain signal, n' is the index of the OTFS time domain signal, and the value range is n'=0,1,…,MN-1. By traversing the DT domain to send the signal X DTThe values ​​of the l,n indexes in [l,n] are arranged serially in columns, and the final output is an OTFS time domain signal.

[0063] For a dual-selective channel with fast time variation and multipath, its response to a pulse with a delay of τ and a Doppler of ν is recorded as h(τ,ν). Then the received signal r(t) obtained by sending the signal s(t) through the channel can be expressed as follows:

[0064] r(t)=∫∫h(τ,ν)s(t-τ)e j2πν(t-τ) dνdτ

[0065] The delay resolution and Doppler resolution are and The above formula is sampled at intervals of and The discrete form of the time domain input-output relationship of the dual-selection channel can be obtained:

[0066]

[0067] Where r[n′] is the received signal in the time domain, P is the number of paths in the channel, i is the path index, and l i , k i and h i are the delay tap, Doppler tap and channel gain of the ith path respectively. By performing serial-to-parallel conversion on the input and output signals in the above formula, we can obtain the DT domain input-output relationship, that is, let n′=nM+l (assuming l i <l), r[n′]=Y DT [l,n], by traversing the index n′ in the received time domain signal and performing serial-to-parallel conversion, the DT domain received signal Y is obtained DT [l,n].

[0068] The input-output relationship of the DT domain can be expressed as:

[0069]

[0070] The above derivation is performed under the assumption that there is no TO. Therefore, in order to recover the transmitted signal s[n′] from r[n′], it is necessary to implement joint time synchronization channel estimation for each transmitting antenna.

[0071] In order to illustrate the effectiveness of the method proposed by the present invention, the above technical solution of the present invention is described in detail below through a specific embodiment. Figure 1 As shown, the present invention discloses a MIMO-OTFS joint time synchronization and channel estimation method, and the specific implementation steps are as follows:

[0072] Step S1, generating a pilot sequence based on a bipolar maximum length sequence; acquiring a delay spread of a channel according to prior channel information; constructing a delay-Doppler domain transmission signal based on the delay spread and the pilot sequence, and performing OTFS modulation on the delay-Doppler domain transmission signal before transmitting it.

[0073] At the transmitting end, each transmitting antenna generates a bipolar MLS with a different initial state. The bipolar MLS generated by each transmitting antenna is a cyclic shift relationship with a cycle period of N-1. Discrete Fourier Transform (DFT) is performed on the bipolar MLS to obtain the frequency domain sequence of the bipolar MLS as a pilot sequence. According to prior information, the delay spread of the channel is obtained, and the positions of the pilot sequence, protection symbols and data symbols in the DD domain are determined according to the delay spread. The pilot sequence, protection symbols and data symbols constitute a DD domain transmission signal, and OTFS modulation is performed on the DD domain transmission signal.

[0074] In some embodiments, a priori information is used to estimate the delay spread characteristics of the wireless channel. Delay spread is the time delay difference between each path after the signal propagates through multiple paths. The priori information can be historical channel characteristics, device movement information, environmental characteristics, or spectrum characteristics.

[0075] In this step, for a MIMO-OTFS system with N Doppler frequency shift components, for the nth t There are transmitting antennas, generating a signal with a length of N-1 and a power of P MLS Bipolar where n t =1,2,…,N t , N t is the total number of transmitting antennas, m = 0, 1, ..., N-2, the bipolar MLS is generated by replacing 0 in MLS with -1, keeping 1 as +1, and then changing the amplitude according to the power. The value of the MLS amplitude in this step is The bipolar MLS of each transmitting antenna is different from each other, and each bipolar MLS is in a cyclic shift relationship, and the cycle period is N-1.

[0076] By setting the MLS of each transmitting antenna as a cyclic shift relationship, the channel of each antenna can be estimated separately by decoding different cyclic shifts during reception without the need for additional pilot signals. Therefore, when the number of transmitting antennas is less than the number of Doppler grid points, the present invention can achieve the overhead of the pilot signal (i.e., the occupied time and spectrum resources) being consistent with the case of a single transmitting antenna.

[0077] right The last digit is padded with zero to obtain zero-padded MLS:

[0078]

[0079] Zero padding Perform discrete Fourier transform and get the nth t Doppler domain MLS for each transmitting antenna:

[0080]

[0081] where DFT{·} denotes discrete Fourier transform.

[0082] like Figure 2 As shown, the nth t The Doppler domain of the transmitting antenna Set up the DD domain grid on the first MLS In order to prevent the data symbol and MLS from interfering with each other, a guard symbol needs to be placed between the two. The guard symbol value is 0. According to the prior channel information, the delay extension of the channel is obtained, and the guard symbol width is determined to be 2L+1, where the value of L is not less than the delay corresponding to the delay extension, and the DD domain signal is obtained:

[0083]

[0084] Among them, l MLS is the index of the delay position of the pilot sequence, L is the protection symbol width parameter, is the data symbol.

[0085] Then Perform OTFS modulation to obtain the time domain signal s[n′] and add a length of L RCP The reduced cyclic prefix is ​​used to obtain the time domain signal and send it.

[0086] Step S2: performing time synchronization detection on the received signal after channel transmission based on the bipolar maximum length sequence, obtaining the synchronization time positions of multiple paths, and obtaining the timing offset based on the synchronization time positions of the multiple paths.

[0087] like Figure 2 As shown, with transmitting antenna n t For example, according to the relationship between the input signal and the output signal of the OTFS system, since there are multiple paths in the channel, in the received signal of the DT domain, the MLS will be dispersed in the delay The different rows of DT domain grid correspond to the i-th path. The mathematical expression of the row sequence is as follows:

[0088]

[0089] in, is the transmitting antenna nt The received signal in the DT domain row sequence, For zero-filled MLS.

[0090] It can be observed that the non-zero rows on the DT grid at the receiving end are in the form of MLS multiplied by complex exponential sequences. Therefore, the good autocorrelation characteristics of MLS are used to locate get The estimation of TO can be completed.

[0091] In order to locate in the DT domain The delay position, such as Figure 3 As shown, the present invention extracts the sequence part that can be used as a row in the DT domain grid from the time domain received signal r[n']:

[0092]

[0093] in, For is the row sequence of the starting time position, The starting time position of the currently extracted row sequence.

[0094] Then use zero-filled MLS with The corresponding bit multiplication is performed to realize the delay positioning. The process can be expressed as: When located When:

[0095]

[0096]

[0097]

[0098] in, For is a complex exponential sequence of the starting time position.

[0099] therefore,

[0100] Locate each transmitting antenna The complex exponential sequence obtained when Perform N-point DFT to obtain the sequence to be judged Its modulus There will be an obvious peak. When k is an integer, the peak is at k = k i Get, its value is

[0101] calculate The normalized maximum value of

[0102] in, For nth t The transmitting antenna is the normalized maximum value of the row sequence at the starting time position, |·| represents the modulus of the complex number, and max{·} represents the maximum value of the sequence.

[0103] Set the peak judgment threshold by Extract each row sequence for the self-incrementing index and perform N t Second judgment, when satisfied When the transmitting antenna number n t Record as n t,λ , and n t,λ Add to collection To synchronize the antenna set, record the corresponding Get all satisfaction of

[0104] According to the above process, Extract each partial sequence for the index, and then detect the synchronization time position of all paths in the channel, which are arranged in ascending order as follows: Assume the synchronization time position of the first path is available Time synchronization can be completed according to TO.

[0105] Traditional time synchronization methods usually require two steps because there may be residual time dimension TO after the first step of synchronization. In this step, the present invention performs correlation operation on the received signal and the local MLS. Based on the good autocorrelation of MLS, the system can directly extract information more accurately from the correlation result, which simplifies the synchronization steps and improves the accuracy.

[0106] S3. Based on the synchronization time positions of the multiple paths and the timing offset, obtain the delay tap, Doppler frequency shift and channel gain of each path.

[0107] The delay taps are estimated according to the synchronized time positions of all the acquired paths; for each path, the delay taps, Doppler frequency shift and channel gain of the channel are obtained according to the closed-form solution of Doppler frequency shift and channel gain, and then the channel estimation is completed.

[0108] In this step, for The synchronization time position of each path obtained by satisfying the peak judgment threshold of the transmitting antenna The delay tap and Doppler shift of each path are obtained by the following expressions:

[0109]

[0110]

[0111] in, For nth t,λ The delay tap of the i-th path of the transmit antenna, For nth t,λ The Doppler shift of the ith path of the transmitting antenna is n t,λ The value range is λ is the synchronization antenna index, It means searching for the k value that corresponds to the maximum value.

[0112] For the i-th path, record the column vector of size N×1 is a row sequence vector of size Column vector of is the gain vector, where For collection The number of elements of Is a collection elements in; G is the design matrix, and the size of matrix G is The nth row and nth t,λ The column elements are For nth t,λ Dual-polarity MLS with two transmit antennas.

[0113] The channel gain vector is estimated using the linear minimum mean square error (LMMSE) method. where σ 2 is the white noise power, which can be obtained from the prior channel information, and I is the unit matrix.

[0114] i=1,2,…,P, P is the total number of distinguishable paths in the channel, and finally the channel parameter set is obtained. n t =1,2,…,N t , as the channel estimation result.

[0115] In this step, based on the closed-form solution of channel gain and fractional Doppler estimation provided by the present invention in the above content, the estimation of Doppler shift can be free from the discretization restriction of the frequency grid, thereby enhancing the estimation accuracy in the case of fractional Doppler.

[0116] In the MIMO-orthogonal time-frequency-space system of the present invention, the implementation steps of the joint time synchronization and channel estimation technology of a single transmitting antenna can be summarized as follows:

[0117] Input: received signal r[n′], local Detection Threshold l MLS , delay spread L and RCP length L RCP .

[0118] Output: Timing offset and channel parameter set

[0119] initialization: stop=∞and trigger=True.

[0120] Step 1: Extract from the received signal r[n′] And do related operations right Do the DFT operation to get

[0121] Step 2, calculate The normalized maximum value of :

[0122] Step 3: If Then determine whether Trigger=True is satisfied. If so, trigger=False,let Then the judgment is completed. and get And the elements Add to collection

[0123] Step 4: Order Repeat steps 1 to 3 until

[0124] The present invention also provides experimental results on the effectiveness of the OTFS joint time synchronization and channel estimation method under a single transmitting antenna.

[0125] The channel estimation method of the present invention is compared with the following benchmark estimation methods: (1) Embedded Pilot Assisted Method (EPA), which refers to the decision made by executing the embedded pilot-based channel estimation algorithm under perfect time synchronization; (2) Best Performance, which refers to the decision made under perfect time synchronization and perfect channel information. In addition, the present invention uses the bit error rate (BER) metric to evaluate the performance of the channel estimation method.

[0126] Figure 4 The variation trend of BER of all channel estimation methods with data signal-to-noise ratio is given. It can be seen from the figure:

[0127] For all the channel estimation comparison methods, their BER decreases with the increase of data signal-to-noise ratio. As the pilot power increases, the BER of each method decreases, because the increase of pilot power will improve the accuracy of channel estimation, and when the pilot power of this method reaches more than 35dB, the performance is close to the ideal estimation method, which is significantly better than other estimation methods.

[0128] Although the specific embodiments of the present invention depict various actions or steps in a specific order, this should be understood as requiring such actions or steps to be performed in the specific order shown or in a sequential order, or requiring all illustrated actions or steps to be performed to obtain the desired results. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Certain features described in the context of a separate embodiment can also be implemented in a single implementation in combination. On the contrary, the various features described in the context of a single implementation can also be implemented in multiple implementations individually or in any suitable sub-combination. The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or replacements that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

[0129] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A MIMO-OTFS joint time synchronization and channel estimation method, characterized in that: The following steps are involved: Step S1: for multiple transmitting antennas, generate a different pilot sequence for each transmitting antenna based on a bipolar maximum length sequence; and obtain a delay spread of the channel according to a priori channel information; Based on the delay spread and the pilot sequence of each transmitting antenna, construct a delay-Doppler domain transmission signal of each transmitting antenna, perform OTFS modulation on the delay-Doppler domain transmission signal and then transmit it from each transmitting antenna; Step S2, performing time synchronization detection on the received signal after channel transmission based on the bipolar maximum length sequence, obtaining synchronization time positions of multiple paths, and obtaining timing offsets based on the synchronization time positions of the multiple paths; Step S3: based on the synchronization time positions of the multiple paths and the timing offset, obtain the delay tap, Doppler frequency shift and channel gain of each path of each transmitting antenna.

2. The MIMO-OTFS joint time synchronization and channel estimation method according to claim 1, characterized in that: Step S1 specifically includes: Step S1-1: for each transmitting antenna, generate a bipolar maximum length sequence with a different initial state, the bipolar maximum length sequence of each transmitting antenna is in a cyclic shift relationship, and perform discrete Fourier transform on the bipolar maximum length sequence to obtain a pilot sequence of each transmitting antenna; Step S1-2, acquiring a delay spread of a channel according to a priori channel information, obtaining a guard symbol width parameter from the delay spread, and determining a guard symbol based on the guard symbol width parameter; determining a data symbol from the data to be transmitted; and constructing a delay-Doppler domain transmission signal of each transmitting antenna from the pilot sequence of each transmitting antenna, the guard symbol, and the data symbol; Step S1-3: Perform OTFS modulation on the delay-Doppler domain transmission signal to obtain a time domain transmission signal, which is sent to the channel by each transmitting antenna.

3. The MIMO-OTFS joint time synchronization and channel estimation method according to claim 2, characterized in that: In step S1-1, for the nth t The expression of the pilot sequence is: Among them, n t is the transmit antenna index, n t =1,2,…,N t , N t is the number of transmitting antennas, For nth t The pilot sequence of the transmitting antennas, k is the Doppler index, DFT{·} represents the discrete Fourier transform, is zero-filled MLS, n is the time index, and the value range is n=0,1,…N-1, is bipolar MLS, N is the number of grid points in the second transform domain; The bipolar MLS is constructed by replacing the amplitude 0 in the maximum length sequence with Replace the amplitude 1 in MLS with It is found that the dual-polarity MLS of different transmitting antennas have different initial states, P MLS The preset power.

4. The MIMO-OTFS joint time synchronization and channel estimation method according to claim 3, characterized in that: In step S1-2, the expression of the delay-Doppler domain transmission signal of each transmitting antenna is: in, For nth t The delay-Doppler domain of the transmitting antennas sends signals, l, k are the delay index and Doppler index respectively, l MLS is the index of the delay position of the pilot sequence, L is the protection symbol width parameter, ] is the nth t The value of L is not less than the delay corresponding to the delay extension.

5. The MIMO-OTFS joint time synchronization and channel estimation method according to claim 4, characterized in that: Step S1-3 specifically includes: right Perform OTFS modulation to obtain the time domain signal And add a length of L RCP The reduced cyclic prefix is ​​used to obtain the time domain transmission signal, which is sent to the channel by each transmitting antenna.

6. The MIMO-OTFS joint time synchronization and channel estimation method according to any one of claims 1 to 5, characterized in that: Step S2 specifically includes: Step S2-1, extracting a time-delay-time domain row sequence from a received signal after channel transmission to obtain a row sequence; Step S2-2: Based on the bipolar maximum length sequence of each transmitting antenna, an autocorrelation positioning method is used to perform time synchronization detection on all the row sequences to obtain the synchronization time position of each path of each transmitting antenna in the channel; Step S2-3: determine the minimum value of the synchronization time positions of the paths as the synchronization time position of the first path, and obtain the timing offset based on the synchronization time position of the first path.

7. The MIMO-OTFS joint time synchronization and channel estimation method according to claim 6, characterized in that: Step S2-1 specifically includes extracting the row sequence using the following expression: in, For is the row sequence of the starting time position, is to extract the starting time position of the row sequence, r[·] is the received signal, and M is the number of grid points in the first transform domain.

8. The MIMO-OTFS joint time synchronization and channel estimation method according to claim 7, characterized in that: Step S2-2 specifically includes: Use the dual-polarity maximum length sequence of each transmitting antenna in turn with Calculate, for the nth t transmit antennas, and obtain the complex exponential sequence right Perform N-point DFT to obtain the sequence to be judged calculate The normalized maximum value of : in, For is the normalized maximum value of the row sequence at the starting time position, |·| represents the modulus of the complex number, and max{·} represents the maximum value of the sequence; Set the peak judgment threshold The starting time position of all row sequences Perform N on all transmit antennas t Second judgment, when satisfied When the transmitting antenna number n t Record as n t,λ , λ is the antenna index that meets the threshold condition, and all n antennas that meet the condition are t,λ Add to collection Record the corresponding Arrange in order from smallest to largest: As the synchronization time position of all paths in the channel, i is the path index, and the value range is i=1, 2, …, P, and P is the total number of paths that can be distinguished in the channel.

9. The MIMO-OTFS joint time synchronization and channel estimation method according to claim 8, characterized in that: In step S2-3, the calculation expression of the timing offset is: in, is the timing offset, is the synchronization time position of the first path.

10. The MIMO-OTFS joint time synchronization and channel estimation method according to claim 9, characterized in that: In step S3, the i-th path, the n-th t,λ The expressions of the delay tap, Doppler frequency shift and channel gain of each path of the transmitting antenna are: in, For nth t,λ The delay tap of the i-th path of the transmit antenna, For nth t,λ The Doppler shift of the ith path of the transmitting antenna; n t,λ The value range is Y is a row sequence vector, [·] T It means taking the transpose, is the channel gain vector of the i-th path, Expressed as For synchronized antenna assembly The number of elements of Is a collection The nth element in the matrix G is the design matrix, and the nth row and nth t,λ The column elements are For nth t,λ Dual-polarity MLS with 1 transmit antenna; Indicates that the linear minimum mean square error method is used to estimate the channel gain vector, where σ 2 is the white noise power, σ 2 It is obtained from the prior channel information, and I is the unit matrix. Summarize the channel parameter set As the channel estimation result.

Images