Low peak-to-average ratio superposition pilot frequency transmission and channel estimation method in scattering OTFS system

Abstract

The invention discloses a low peak-to-average ratio superposition pilot frequency transmission and channel estimation method in a scattering OTFS system, and relates to the technical field of communication. At a sending end, information data are coded, modulated and pre-coded, data symbols are converted to a frequency Doppler domain, pilot frequency is superposed on the frequency Doppler domain, obtained data are subjected to parallel-serial conversion to obtain frequency domain data, and the frequency domain data are subjected to inverse Fourier transform and converted to a time domain to be sent. At a receiving end, Fourier transform is carried out on a received time domain signal to obtain a frequency domain received signal, serial-parallel conversion is carried out on the frequency domain signal to obtain frequency Doppler domain information, a signal related to pilot frequency is taken out, channel estimation is carried out by utilizing a matching and dichotomy method to obtain channel information, and data is converted to a time delay Doppler domain for detection. The channel estimation performance is excellent by superimposing the pilot frequency in the frequency Doppler domain, the signal formed in the time domain is smaller than the signal formed by placing the pilot frequency in the time delay Doppler domain, and more excellent performance can be obtained by detecting in the time delay Doppler domain.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and particularly to a method for transmitting and channel estimating superimposed pilots with low peak-to-average power ratio in a scattered OTFS system. Background Art

[0002] Orthogonal Time Frequency Space (OTFS) modulation technology is a new modulation technology developed based on Orthogonal Frequency Division Multiplexing (OFDM) modulation technology. Its main change is to use two-dimensional Fourier transform in the time-frequency domain to construct a new delay-Doppler domain, and re-modulate the signals in the time-frequency domain of the OFDM system in the delay-Doppler domain, so as to combat the destruction of orthogonality between OFDM subcarriers caused by high-speed movement and high frequency bands. The OTFS modulation technology has significant performance advantages in high-speed mobile scenarios, meeting the application scenarios of the terahertz frequency band and a terminal mobile speed of 1000 km / h required by the sixth-generation mobile communication system, and is a key technology in the field of communication.

[0003] The main problems of OTFS lie in its channel estimation and detection. The detection result directly determines the bit error rate performance of the system, and the quality of channel estimation has a decisive impact on the detection result. Therefore, OTFS channel estimation is an important issue in OTFS research. The most common existing OTFS channel estimation scheme is the embedded pilot channel estimation, which places a guard interval between the pilot and the data to make the pilot only affected by noise rather than the data, so as to achieve excellent performance. However, due to the influence of the guard interval, it will reduce the spectrum utilization efficiency of the system. In order to increase the spectrum utilization efficiency of the system, reference [1] proposed an embedded pilot structure without a guard interval, and improved the performance through iteration and interference cancellation between the data and the pilot. However, due to the mutual influence between the data and the pilot, its performance is still greatly reduced compared with the embedded pilot. Reference [2] proposed a sparse embedded pilot channel estimation scheme without a guard interval, and its performance is improved compared with the case of a single pilot. However, due to the reduction of the pilot energy for detection, its performance in detecting the position in the delay-Doppler domain will decline when the pilot power is reduced.

[0004] Therefore, the problem to be solved by the present invention is how to ensure high channel estimation performance while ensuring high spectrum utilization efficiency in a high-speed mobile troposcatter channel.

[0005] [1]W.Yuan,S.Li,Z.Wei,J.Yuan and D.W.K.Ng,"Data-Aided ChannelEstimation for OTFS Systems With a Superimposed Pilot and Data TransmissionScheme,"in IEEE Wireless Communications Letters,vol.10,no.9,pp.1954-1958,Sept.2021,doi:10.1109 / LWC.2021.3088836.

[0006] [2]W.Liu,L.Zou,B.Bai and T.Sun,"Low PAPR channel estimation for OTFSwith scattered superimposed pilots,"in China Communications,vol.20,no.1,pp.79-87,Jan.2023,doi:10.23919 / JCC.2023.01.007. Summary of the Invention

[0007] In view of this, the present invention proposes a method for low peak-to-average power ratio (PAPR) superimposed pilot transmission and channel estimation in a scattered OTFS system. This method solves the problems of high PAPR of overlapping pilots and performance degradation caused by mutual interference between data and pilots.

[0008] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0009] A method for low PAPR superimposed pilot transmission and channel estimation in a scattered OTFS system, wherein the transmitting process includes the following steps:

[0010] Step a1, the transmitting end obtains the information at the transmitting end;

[0011] Step a2, the transmitting end performs channel coding and mapping on the information at the transmitting end to form complex data of multi-level quadrature amplitude modulation;

[0012] Step a3, the transmitting end precodes the complex data to convert it to the frequency Doppler domain, and converts the pilot sequence to the frequency Doppler domain. Then, the two are superimposed in the frequency Doppler domain; the superimposed data is converted from parallel to serial to obtain frequency-domain data, and the frequency-domain data is finally converted to the time domain through inverse Fourier transform. Then, a cyclic prefix is added and pulse shaping is performed to form a transmitted signal data block, and the transmitted signal data block is transmitted to the receiving end through the channel;

[0013] The receiving process includes the following steps:

[0014] Step b1, the receiving end first performs Fourier transform on the received time-domain signal to obtain the received frequency-domain signal;

[0015] Step b2, the receiving end performs serial-to-parallel conversion on the received frequency-domain signal to obtain frequency Doppler domain information, and extracts the signals related to the pilots;

[0016] Step b3, the receiving end uses matching and the bisection method for channel estimation to obtain the channel estimation result;

[0017] Step b4, the receiving end converts the received frequency-domain signal to the delay-Doppler domain for signal detection and interference cancellation;

[0018] Step b5, iteratively execute Step b2 to Step b4 until the difference between the channel estimation results in two adjacent iterations is less than a preset threshold,

[0019] Step b6, demodulate and channel decode the data after signal detection to obtain the original transmitted data, and complete the transmission of low peak-to-average ratio superimposed pilots and channel estimation in the scattered OTFS system.

[0020] Furthermore, in Step a3, the specific method for the transmitting end to perform precoding on the complex data to convert it to the frequency Doppler domain, convert the pilot sequence to the frequency Doppler domain, and then superimpose the two in the frequency Doppler domain is as follows:

[0021] The transmitting end performs precoding on the complex data to convert it to the frequency Doppler domain:

[0022] S d = F M X;

[0023] where X is the complex data, and F M is the M-point Fourier transform matrix, and M is the number of subcarriers in the OTFS system;

[0024] Convert the pilot sequence to the frequency Doppler domain:

[0025]

[0026] where is the pilot sequence; K max is the maximum Doppler shift; N is the number of complex data symbols after precoding;

[0027] Superimpose the two in the frequency Doppler domain:

[0028] S t = S d + S p .

[0029] Further, the specific manner of step b2 is as follows:

[0030] In step b201, the receiving end performs serial-to-parallel conversion on the received frequency-domain signal to obtain frequency Doppler domain information:

[0031] R t = reshae(R, M, N);

[0032] where R is the received frequency-domain signal; reshae() represents serial-to-parallel conversion;

[0033] In step b202, set a variable u1 representing the number of iteration times, and let u1 = 1; extract the signal related to the pilot:

[0034]

[0035] where L max is the maximum time delay, and K max is the maximum Doppler frequency shift.

[0036] Further, the specific manner of step b3 is as follows:

[0037] In step b301, set a variable u2 representing the number of iteration times, and let u2 = 1; for each column of R p , divide each element by the original pilot sequence, and then apply the inverse Fourier transform to the result:

[0038]

[0039] where g = 1:2K max + 1; the superscript H represents conjugate transpose;

[0040] Let:

[0041]

[0042] In step b302, for the path with time delay id r , according to , the position index of the maximum value minus K max is denoted as the estimated Doppler frequency shift where d r is the time delay resolution, and i is the variable indicating the path; if the iteration variable u1 = 1, then i = 1:L max + 1; otherwise, i is the variable indicating the valid path currently retained in 1:L max + 1;

[0043] Set the left boundary Set the right boundary Set the midpoint

[0044] Step b303: Construct two matching sequences and

[0045]

[0046] where \(k = -K\) max :K max ; \(j\) represents the imaginary unit;

[0047] Step b304: Let

[0048] where \(E\) p is the power of the pilot sequence \(P\) v ;

[0049] Step b305: Judge whether the result of is less than the preset threshold \(\delta1\). If so, let \(\tau\) i = id r ; And execute Step b306 for the path with time delay id r ;

[0050] Otherwise, make a decision on the path with time delay id r :

[0051] If then let Otherwise

[0052] After that, let and return to execute Step b303;

[0053] Step b306: Calculate the channel coefficient:

[0054]

[0055] Step b307: Set the threshold where \(\sigma\) 2 is the noise power, and \(E\) s is the signal power of the complex data \(X\) at the transmitter;

[0056] Judge whether the iteration variable \(u2 = 1\) holds. If so, then:

[0057] Record the \(i\)-th path that satisfies as the valid path, and perform interference cancellation on :

[0058]

[0059] Reject the \(i\)-th path that does not satisfy ;

[0060] If the iteration variable u2 = 1 does not hold, for the currently retained valid paths, interference cancellation is performed:

[0061]

[0062] Step b308, for the currently retained valid paths, set u2 = u2 + 1, and return to execute step b302 until all the currently retained valid paths satisfy Thus, the time delay τ i , Doppler frequency shift v i , and channel coefficient h i constitute the time delay τ, Doppler frequency shift v, and channel coefficient h respectively.

[0063] Furthermore, the specific manner of step b4 is as follows:

[0064] Step b401, by performing an M-point inverse Fourier transform on the received frequency-domain signal R and compensating the phase, the signal in the time delay-Doppler domain can be obtained:

[0065]

[0066] where q = 0:MN - 1; I N is the N-order identity matrix; represents the Kronecker product; (q) M represents taking the remainder of q with respect to M points;

[0067] Step b402, by performing a Fourier transform on the pilot signal in the frequency-Doppler domain and compensating the phase offset, the interference cancellation signal of the received signal can be obtained:

[0068]

[0069] Step b403, after interference cancellation, the signal containing the data part in the received signal can be expressed as:

[0070] y d = y - H est P DD ;

[0071] where,

[0072]

[0073] F N is the N-point Fourier transform matrix, I M is the M-order identity matrix, P represents the set of indicator variables of the currently retained valid paths, is the MN-order identity matrix I MN arranged according to τi Step b404 obtained by performing a cyclic left shift, performing signal detection with the minimum mean square error:

[0074]

[0075] Step b405, the extracted data is used to cancel the interference on the pilot signal, obtaining:

[0076]

[0077] Furthermore, the specific manner of step b5 is:

[0078] Let:

[0079]

[0080] Let u1 = u1 + 1, and return to execute step b202 until the difference in H between two adjacent iteration processes est is less than a preset threshold.

[0081] Due to the adoption of the above technical solution, the beneficial effects of the present invention compared with the prior art are as follows:

[0082] 1. By placing pilots in the frequency - Doppler domain, the present invention enables a lower peak - to - average power ratio of the transmitted data, and better channel estimation performance and bit error rate at the receiving end.

[0083] 2. The iterative channel estimation scheme of the present invention achieves excellent channel estimation performance with low complexity, and can overcome the problem of frequency leakage caused by fractional Doppler. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 It is a schematic diagram of the placement manner of OTFS data and pilots in the frequency - Doppler domain in an embodiment of the present invention.

[0085] Figure 2 It is a schematic diagram of the comparison of CCDF curves under different pilot formats in an OTFS system in an embodiment of the present invention.

[0086] Figure 3 It is a schematic diagram of the comparison of NMSE curves under different pilot formats in an OTFS system in an embodiment of the present invention.

[0087] Figure 4 It is a schematic diagram of the comparison of BER curves under different pilot formats in an OTFS system in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0088] The following further describes the content of the present invention in conjunction with the accompanying drawings and specific embodiments.

[0089] Method for Transmitting and Channel Estimating Low Peak-to-Average Ratio Superimposed Pilots in a Scattering OTFS System. The transmitting process includes the following steps:

[0090] Step a1, the transmitter obtains the transmitter information;

[0091] Step a2, the transmitter performs channel coding and mapping on the transmitter information to form complex data of multi-level quadrature amplitude modulation;

[0092] Step a3, the transmitter precodes the complex data to transform it into the frequency Doppler domain, and transforms the pilot sequence into the frequency Doppler domain. Then, the two are superimposed in the frequency Doppler domain; the superimposed data is subjected to serial-to-parallel conversion to obtain frequency-domain data, and the frequency-domain data is subjected to inverse Fourier transform to finally be transformed into the time domain. Then, a cyclic prefix is added and pulse shaping is performed to form a transmitted signal data block, and the transmitted signal data block is transmitted to the receiver through the channel;

[0093] The receiving process includes the following steps:

[0094] Step b1, the receiver first performs Fourier transform on the received time-domain signal to obtain the received frequency-domain signal;

[0095] Step b2, the receiver performs serial-to-parallel conversion on the received frequency-domain signal to obtain frequency Doppler domain information, and extracts the signal related to the pilot;

[0096] Step b3, the receiver uses matching and the bisection method for channel estimation to obtain the channel estimation result;

[0097] Step b4, the receiver transforms the received frequency-domain signal into the delay Doppler domain for signal detection and interference cancellation;

[0098] Step b5, iteratively execute steps b2 to b4 until the difference between the channel estimation results in two adjacent iterations is less than a preset threshold,

[0099] Step b6, demodulate and perform channel decoding on the data after signal detection to obtain the original transmitted data, completing the transmission and channel estimation of low peak-to-average ratio superimposed pilots in the scattering OTFS system.

[0100] Further, as Figure 1 shown, the specific method for the transmitter to precode the complex data to transform it into the frequency Doppler domain, and transform the pilot sequence into the frequency Doppler domain, and then superimpose the two in the frequency Doppler domain in step a3 is:

[0101] The transmitter precodes the complex data to transform it into the frequency Doppler domain:

[0102] S d =F M X;

[0103] Among them, X is complex data, and F M is an M-point Fourier transform matrix, and M is the number of subcarriers of the OTFS system;

[0104] Convert the pilot sequence to the frequency-Doppler domain:

[0105]

[0106] Among them, is the pilot sequence; K max is the maximum Doppler shift; N is the number of complex data symbols after precoding;

[0107] Superimpose the two in the frequency-Doppler domain:

[0108] S t = S d + S p .

[0109] After vectorizing the signal, we obtain the frequency-domain signal S = vec(S t ), where vec() represents serial-to-parallel conversion. Finally, perform an MN-point inverse Fourier transform on the frequency-domain signal to convert it to the time domain for transmission.

[0110] For traditional threshold-based channel estimation algorithms, when the pilot energy is large, the delay and Doppler position (located on the delay-Doppler domain grid) can be accurately estimated. However, when there is a fractional Doppler shift, since the threshold only considers the signal power and noise power, some weak-energy paths may not be detected, which will reduce the performance of channel estimation. To solve this problem, we propose a channel estimation algorithm combining threshold-based matching and the dichotomy method.

[0111] First, we obtain the received pilot data from the frequency-Doppler domain and divide it by the local pilot. Then, perform a Fourier transform in the frequency domain. Next, for each delay, we apply the matching and threshold-based decision rules to obtain the corresponding delay and Doppler shift. The following is the specific process of channel estimation and iteration.

[0112] We can convert the data from the delay-Doppler domain to the frequency-Doppler domain, or directly use the pre-coded OFDM transmission method to obtain the frequency-Doppler domain data.

[0113] Furthermore, the specific manner of step b2 is:

[0114] Step b201, the receiving end performs serial-to-parallel conversion on the received frequency-domain signal to obtain the frequency-Doppler domain information:

[0115] Rt = reshae(R, M, N);

[0116] Where R is the received frequency-domain signal; reshae() represents serial-to-parallel conversion;

[0117] Step b202, set the variable u1 representing the iteration count, and let u1 = 1; extract the signal related to the pilot:

[0118]

[0119] Where L max is the maximum time delay, and K max is the maximum Doppler frequency shift.

[0120] Furthermore, the specific manner of step b3 is as follows:

[0121] Step b301, set the variable u2 representing the iteration count, and let u2 = 1; for each column of R p , divide each element by the original pilot sequence element-wise, and then apply the inverse Fourier transform to the result:

[0122]

[0123] Where g = 1:2K max + 1; the superscript H represents conjugate transpose;

[0124] a:b:c means taking values from the curve [a, c] at an interval of b;

[0125] a:c means taking values from the curve [a, c] at an interval of 1;

[0126] (:, j) means taking the j-th column of the matrix;

[0127] . / means element-wise division of the matrix or vector

[0128] Let:

[0129]

[0130] Where is constructed by rows,

[0131] is constructed by columns;

[0132] Step b302, for the path with time delay id r , according to the position index of the maximum value in minus K max the value of is denoted as the estimated Doppler frequency shift Where d rLet be the time delay resolution, and i be the variable indicating the path; if the iteration variable u1 = 1, then i = 1:L max +1; otherwise, i is the variable indicating the valid path currently reserved in 1:L max +1;

[0133] Set the left boundary Set the right boundary Set the midpoint

[0134] Step b303, construct two matching sequences and

[0135]

[0136] where k = -K max :K max ; j represents the imaginary unit;

[0137] Step b304, let

[0138] where E p is the power of the pilot sequence P v ;

[0139] Step b305, judge whether the result of is less than the preset threshold δ1. If so, let τ i = id r ; and execute Step b306 for the path with time delay id r ;

[0140] Otherwise, make a decision on the path with time delay id r :

[0141] If then let Otherwise

[0142] After that, let and return to execute Step b303;

[0143] Step b306, calculate the channel coefficient:

[0144]

[0145] Step b307, set the threshold where σ 2 is the noise power, and E s is the signal power of the complex data X at the transmitter;

[0146] Determine whether the iteration variable u2 = 1 holds. If it holds, then:

[0147] Record the i-th path that satisfies as a valid path, and perform interference cancellation on :

[0148]

[0149] Reject the i-th path that does not satisfy ;

[0150] If the iteration variable u2 = 1 does not hold, then for the currently retained valid paths, perform interference cancellation on :

[0151]

[0152] Step b308, for the currently retained valid paths, set u2 = u2 + 1, and return to execute step b302 until all the currently retained valid paths satisfy Thus, the time delay τ, Doppler frequency shift v, and channel coefficient h respectively composed of τ i , v i , h i are obtained.

[0153] Furthermore, the specific manner of step b4 is as follows:

[0154] Step b401, by performing an M-point inverse Fourier transform on the received frequency-domain signal R and performing phase compensation, the signal in the time delay-Doppler domain can be obtained:

[0155]

[0156] where q = 0:MN - 1; I N is the N-order identity matrix; represents the Kronecker product; (q) M represents taking the remainder of q with respect to M points; represents rounding down;

[0157] Step b402, by performing a Fourier transform on the pilot signal in the frequency-Doppler domain and compensating for the phase offset, the interference cancellation signal of the received signal can be obtained:

[0158]

[0159] Step b403, after interference cancellation, the signal containing the data part in the received signal can be expressed as:

[0160] y a = y - H est P DD;

[0161] Among them,

[0162]

[0163] F N is an N-point Fourier transform matrix, and I M is an M-order identity matrix. P represents a set of indication variables of the currently retained valid paths. is to circularly shift the MN-order identity matrix I MN by τ i to obtain step b404, and perform signal detection with the minimum mean square error:

[0164]

[0165] Step b405, the extracted data is used to cancel the interference on the pilot signal, and we get:

[0166]

[0167] Furthermore, the specific manner of step b5 is:

[0168] Let:

[0169]

[0170] Let u1 = u1 + 1, and return to execute step b202 until the difference between H est in two adjacent iteration processes is less than a preset threshold.

[0171] For demodulation and channel decoding are performed to obtain the original transmitted data.

[0172] In this embodiment, we set N = 32, M = 32. The channel model uses a typical tropospheric scattering channel, the time delay is [0, 0.0625, 0.125, 0.1875, 0.25, 0.3125, 0.375, 0.4375, 0.5] μs, the channel fading is [-10, -8, -6, -3, 0, -4, -6, -8, -10] (dB), and the maximum Doppler shift is 3.

[0173] Figure 2 shows the CCDF (Complementary Cumulative Distribution Function) curves of the PAPR (Peak-to-Average Power Ratio) of different schemes. It can be seen from the figure that the PAPR of the proposed frequency Doppler domain superimposed pilot scheme is lower than that of the time delay Doppler domain single superimposed pilot and the time delay Doppler domain sparse superimposed pilot schemes, about 4 dB lower than the time delay Doppler domain single superimposed pilot and 2 dB lower than the time delay Doppler domain sparse superimposed pilot.

[0174] The NMSE is defined as

[0175]

[0176] where H DD is the true value of H est and is the iterative estimated value of H est .

[0177] Figure 3 It can be seen that the NMSE performance of the proposed scheme is always better than that of the time-delay Doppler domain sparse superimposed pilot scheme. When the signal-to-noise ratio (SNR) is 15 dB and 20 dB, the NMSE of the proposed algorithm is about 10 -2 , while the NMSE of the time-delay Doppler domain sparse superimposed pilot and the time-delay Doppler domain single superimposed pilot is about 10 -1 . From Figure 3 it can be seen that the proposed pilot structure has a great improvement in the bit error rate compared with the existing scheme.

[0178] Figure 4 Fig. shows the bit error rate (BER) performance based on the MMSE algorithm, comparing the proposed frequency Doppler domain superimposed pilot scheme, the time-delay Doppler domain single superimposed pilot scheme and the time-delay Doppler domain sparse superimposed pilot scheme. From Figure 4 it can be seen that the proposed scheme is better than the time-delay Doppler domain single superimposed pilot and the time-delay Doppler domain sparse superimposed pilot scheme in terms of BER performance. It can be further seen from the figure that when the proposed frequency Doppler domain superimposed pilot scheme adopts the traditional threshold-based channel estimation method, its performance is almost the same as that of the time-delay Doppler domain single superimposed pilot scheme. However, when the channel estimation method proposed in this paper is adopted, its performance is significantly better than the threshold-based channel estimation method.

[0179] Those skilled in the art will realize that the described embodiments are to help the reader understand the principles of the present invention and should be understood that the protection scope of the present invention is not limited to the described embodiments. For those skilled in the art, various changes and modifications can be made to the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the scope of the claims of the present invention.

Claims

1. A method for low peak-to-average ratio superimposed pilot transmission and channel estimation in a scattering OTFS system, characterized in that, The transmission process includes the following steps: Step a1, the transmitting end obtains the information of the transmitting end; Step a2, the transmitting end performs channel coding and mapping on the information of the transmitting end to form complex data of multi-level quadrature amplitude modulation; Step a3, the transmitting end performs precoding on the complex data to convert it to the frequency Doppler domain, and converts the pilot sequence to the frequency Doppler domain, and then superimposes the two in the frequency Doppler domain; perform serial-to-parallel conversion on the superimposed data to obtain frequency-domain data, perform inverse Fourier transform on the frequency-domain data to finally convert it to the time domain, then add a cyclic prefix and pulse shaping to form a transmitted signal data block, and transmit the transmitted signal data block to the receiving end through the channel; The receiving process includes the following steps: Step b1, the receiving end first performs Fourier transform on the received time-domain signal to obtain the received frequency-domain signal; Step b2, the receiving end performs serial-to-parallel conversion on the received frequency-domain signal to obtain frequency Doppler domain information, and extracts the signal related to the pilot; Step b3, the receiving end uses matching and the dichotomy method to perform channel estimation to obtain the channel estimation result; Step b4, the receiving end converts the received frequency-domain signal to the delay Doppler domain for signal detection and interference cancellation; Step b5, iteratively execute steps b2 to b4 until the difference between the channel estimation results in two adjacent iterations is less than a preset threshold; Step b6, perform demodulation and channel decoding on the data after signal detection to obtain the original transmitted data, and complete the transmission of low peak-to-average ratio superimposed pilot and channel estimation in the scattered OTFS system.

2. The method for low PAPR superimposed pilot transmission and channel estimation in the scattering OTFS system according to claim 1, wherein The specific method in step a3 for the transmitting end to perform precoding on the complex data to convert it to the frequency Doppler domain, convert the pilot sequence to the frequency Doppler domain, and then superimpose the two in the frequency Doppler domain is as follows: The transmitting end performs precoding on the complex data to convert it to the frequency Doppler domain: S d = F M X; where X is complex data, and F M is an M-point Fourier transform matrix, and M is the number of subcarriers in the OTFS system; Convert the pilot sequence to the frequency Doppler domain: Among them, is the pilot sequence; K max is the maximum Doppler shift; N is the number of precoded complex data symbols; Superimpose the two in the frequency Doppler domain: S t = S d + S p .

3. The low PAPR superimposed pilot transmission and channel estimation method in the scattering OTFS system according to claim 2, characterized in that, The specific method of step b2 is as follows: Step b201, the receiving end performs serial-to-parallel conversion on the received frequency-domain signal to obtain frequency Doppler domain information: R t = reshape(R, M, N); Among them, R is the received frequency-domain signal; reshae() represents serial-to-parallel conversion; Step b202, set a variable u1 representing the number of iterations, and let u1 = 1; extract the signal related to the pilot; where L max is the maximum time delay, and K max is the maximum Doppler shift.

4. The method for low PAPR superimposed pilot transmission and channel estimation in the scattered OTFS system according to claim 3, characterized in that The specific method of step b3 is as follows: Step b301, set a variable u2 representing the number of iteration times, and let u2 = 1; for each column of R p , divide it element by element by the original pilot sequence, and then apply the inverse Fourier transform to the result: where \(g = 1:2^K\) max + 1; the superscript \(H\) represents the conjugate transpose; Let: Step b302, for a path with a time delay id r , according to , the position index of the maximum value in it minus K max is denoted as the estimated Doppler frequency shift , where d r is the time delay resolution, and i is a variable indicating the path; if the iteration variable u1 = 1, then i = 1:L max +1; otherwise, i is a variable indicating the valid path currently retained in 1:L max +1; Set the left boundary Set the right boundary Set the midpoint Step b303, construct two matching sequences and where k = -K max :K max ; j represents the imaginary unit; Step b304, let Among them, E p is the power of the pilot sequence P v . Step b305, determine whether the result is less than a preset threshold δ1. If so, let τ i = id r ; And execute step b306 on the path with latency id r ​ Otherwise, make a decision on the path with the delay id r as follows: If then let otherwise Then make and return to execute step b303; Step b306, calculate the channel coefficient: Step b307, set the threshold where σ 2 is the noise power, and E s is the signal power of the complex data X at the transmitting end; Judge whether the iteration variable u2 = 1 holds. If it holds, then: The i-th path that satisfies is denoted as a valid path, and interference cancellation is performed on : Reject the i-th path that does not meet ; If the iteration variable u2 = 1 does not hold, then for the currently retained valid path, perform interference cancellation: Step b308: For the currently retained valid path, let u2 = u2 + 1, and return to execute step b302 until all currently retained valid paths are satisfied From this, the time delay τ i , Doppler frequency shift v i , and channel coefficient h i respectively constitute the time delay τ, Doppler frequency shift v, and channel coefficient h 5. The method for low PAPR superimposed pilot transmission and channel estimation in the scattered OTFS system according to claim 4, wherein The specific method of step b4 is as follows: Step b401, by performing M-point inverse Fourier transform on the received frequency-domain signal R and performing phase compensation, the signal in the delay-Doppler domain can be obtained: where q = 0:MN-1; I N is an N-order identity matrix; denotes the Kronecker product; (q) M denotes taking the remainder of q with respect to M points; Step b402, by performing Fourier transform on the pilot signal in the frequency-Doppler domain and compensating for the phase offset, the interference cancellation signal of the received signal can be obtained: Step b403, after interference cancellation, the signal containing the data part in the received signal can be expressed as: y d = y - H est P DD ; Among them, F N is an N-point Fourier transform matrix, and I M is an M-order identity matrix. P represents a set of indicator variables for the currently retained valid paths, is obtained by circularly shifting the MN-order identity matrix I MN by τ i in a cyclic left shift manner Step b404, perform signal detection with the minimum mean square error: Step b405, the extracted data is used to perform interference cancellation on the pilot signal to obtain:

6. The method for low PAPR superimposed pilot transmission and channel estimation in the scattered OTFS system according to claim 5, wherein The specific method of step b5 is as follows: Let: Let u1 = u1 + 1, and return to execute step b202 until the difference of H in two adjacent iteration processes is less than a preset threshold value. est ​

Images