OFDM System Time-Frequency Synchronization Method, Terminal Device and Storage Medium

Abstract

The present invention discloses a time-frequency synchronization method, a terminal device and a storage medium for an OFDM system, which are divided into two parts: coarse time-frequency estimation and fine time-frequency estimation. First, the symmetric conjugate property of the training sequence is used for coarse timing estimation and coarse frequency offset estimation. Secondly, the CAZAC sequence is used to complete the fine frequency offset estimation and fine timing estimation due to its good autocorrelation property in the frequency domain. The simulation results show that compared with the traditional scheme, the scheme of the present invention can estimate the symbol timing and carrier frequency offset more accurately, and greatly expand the estimation range of the frequency offset.

Description

Technical Field

[0001] The present invention relates to multi - carrier modulation technology, and in particular to a time - frequency synchronization method, a terminal device and a storage medium for an OFDM system. Background Art

[0002] Orthogonal frequency - division multiplexing (OFDM) technology, as a multi - carrier modulation technology, has high spectral efficiency and good anti - multipath interference ability, making OFDM systems a research hotspot in the current communication field. However, OFDM technology is very sensitive to timing errors and carrier frequency offsets. Deviations in system synchronization will affect the orthogonality between sub - carriers, causing inter - carrier interference (ICI) and inter - symbol interference (ISI), resulting in a decline in system performance. At the same time, due to problems such as limited available bandwidth, large signal attenuation, widespread multipath effects and the resulting frequency - selective attenuation in communication channels, it has an impact on the accurate synchronization of communication systems. Therefore, synchronization technology has become one of the main research difficulties of OFDM technology.

[0003] Typical OFDM time - domain synchronization methods are based on specific preamble symbols for synchronization. The most well - known is that M. Schmidl and C. Cox et al. used PN sequences to construct specific training sequences for timing synchronization and frequency - offset synchronization, achieving good results. However, due to the "plateau effect", the accurate starting point of the signal is still ambiguous. Minn et al. improved the timing performance by sliding the timing metric function of the Schmidl method to eliminate the "plateau effect", but there is still the problem of inaccurate timing. Park et al. achieved the pulse characteristics of the timing envelope by designing a centrosymmetric conjugate structure, but due to the existence of the cyclic prefix (CP), there is side - band interference in the method. At the same time, because the peak - to - average power ratio (PAPR) of the preamble composed of pseudo - random (PN) sequences is very large, the nonlinear distortion in signal transmission reduces the performance. Since the constant - amplitude zero - autocorrelation (CAZAC) sequence has a flat power spectrum and zero - autocorrelation performance, which can effectively resist the influence of multipath, some scholars use CAZAC sequences instead of PN sequences to achieve better time - frequency estimation methods. Guangliang Ren et al. constructed a new training sequence using CAZAC sequences, which has a sharp timing synchronization peak and achieves good frequency - offset estimation, but the performance decreases in complex multipath channels. Malik et al. used two conjugate CAZAC sequences as preambles for coarse timing and frequency - offset estimation. Although the carrier frequency offset (CFO) can be accurately estimated, due to the use of two training sequences, the transmission efficiency is low. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a time - frequency synchronization method for an OFDM system in view of the deficiencies of the prior art.

[0005] To solve the above technical problems, the technical solution adopted by the present invention is as follows: A time-frequency synchronization method for an OFDM system, comprising:

[0006] Using the following formula to determine the estimated value of the timing offset:

[0007] Wherein, d is the position of the signal start time, N is the length of the synchronization sequence, P1(d) is the symmetric correlation value of the front and back two parts of the received synchronization sequence, P2(d) is the delay correlation value obtained by dividing the received synchronization sequence into four equal parts, r(d) is the received synchronization signal intercepted with d as the signal start position, r(d + N / 2 - n - 1) represents the received signal sample value of the length of N / 2 - n - 1 after the d position, m s = [m s1 m s2 is the m-sequence for weighting the second half of the synchronization sequence, m s1 and m s2 are respectively the m-sequences for weighting the third part and the fourth part of the synchronization sequence after being divided into four equal parts at the transmitting end, and R(d) is half of the energy of the received signal with a length of N after the d moment.

[0008] The timing offset estimation method of the present invention uses a CAZAC sequence with excellent correlation performance for the design of the synchronization sequence, and at the same time performs an m-sequence weighting operation on the second half of the synchronization sequence. When the receiving end performs timing synchronization, by simultaneously using delay correlation and symmetric correlation to ensure that the timing metric function only obtains a sharp peak at the correct timing point, and no other side lobes appear at other positions. Therefore, the method of the present invention has strong anti-noise and anti-multipath capabilities, which enables the timing offset estimation method to achieve a lower error rate at a lower signal-to-noise ratio.

[0009] Furthermore, the method of the present invention further comprises:

[0010] Using the formula ε = ε1 + ε2 to calculate the normalized frequency offset estimation value ε; wherein, a / π is the fractional multiple frequency offset, and 2z is the integer multiple frequency offset; Wherein, B F is a sequence composed of the even-frequency data after removing the weighting and then performing FFT on the second half of the original synchronization sequence at the transmitting end, g is the cyclic shift amount, B F * (b - g) represents performing a conjugate operation on B F and then performing a cyclic shift of length g, X B is a sequence composed of the even-frequency data after removing the weighting and then performing FFT on the second half of the synchronization sequence at the receiving end, XB (b - z / 2) is the cyclic shift effect of z / 2 on the frequency-domain data of the second half of the received synchronization sequence caused by the existing integer multiple frequency offset z, m * is the deviation between the position of the first path of the received signal and the position of the path with the strongest energy, ω(2b) is the noise effect introduced by the channel when b is the independent variable, and b ∈ [0, N / 4 - 1].

[0011] The carrier frequency offset estimation method of the present invention is designed as a two-stage structure of time domain first and then frequency domain. First, the repeated part in the time domain of the synchronization sequence is used to estimate the fractional frequency offset, and at the same time, the odd integer multiple frequency offset is compensated to an even integer multiple frequency offset. This method has low requirements for the accuracy of timing synchronization. When there are multipaths or other interferences in the channel that cause the timing synchronization to fall within the CP, it will not affect the estimation accuracy of this method. Then, the frequency-domain information of the weighted part in the training sequence is used to perform a fine estimation of the wide-range integer multiple frequency offset, and finally, a wide-range and high-precision carrier frequency offset estimation that can work robustly in a Rayleigh multipath channel is realized.

[0012] The deviation m between the position of the first path of the received signal and the position of the path with the strongest energy * The calculation formula is as follows:

[0013] m * = max(Q2(x) -1 (y > α·max(Q2(x)))) - z / 2;

[0014]

[0015] Wherein, L is the length of the guard interval, x is the cyclic shift amount, y represents the correlation result of Q2(x) when x ∈ [0, L], A F * (b) represents the conjugate operation of A F and the corresponding values of different independent variables b, A F is the sequence composed of the even-frequency data after FFT of the first half of the original synchronization sequence at the transmitting end, X A is the sequence composed of the even-frequency data after FFT of the first half of the synchronization sequence at the receiving end, X A (b - z / 2) is the cyclic shift effect of z / 2 on the frequency-domain data of the first half of the received synchronization sequence caused by the existing integer multiple frequency offset z, and α is the threshold parameter, α ∈ (0, 1).

[0016] The present invention proposes the deviation m between the first arrival path of the signal and the path with the strongest energy with an adaptive threshold by utilizing the frequency-domain characteristics of the training sequence *Estimation method. When the communication signal undergoes different energy attenuations due to unknown interferences when passing through different channels, the method sets the threshold to a decimal multiple of the correlation result corresponding to the path with the strongest energy, so as to achieve the adaptive dynamic change of the threshold under different signal-to-noise ratios, ensure the detection accuracy while reducing the detection threshold as much as possible, and finally accurately estimate the position of the first path of the received signal under low signal-to-noise ratio.

[0017] Further, α takes a value of 0.2 to 0.4.

[0018] Further, α takes a value of 0.3.

[0019] Further, the sequence includes: a cyclic prefix, two CAZAC sequences A with a length of N / 4, two sequences B' generated by conjugate symmetry of sequence A followed by m-sequence weighting, and a cyclic suffix. The present invention adopts CAZAC sequences with excellent correlation performance and is designed as a repetitive structure to resist carrier frequency offset.

[0020] A = [a(0), a(1), …, a(N / 4 - 1)],

[0021] As an inventive concept, the present invention also provides a terminal device, which includes:

[0022] One or more processors; a memory, on which one or more programs are stored. When the one or more programs are executed by the one or more processors, the one or more processors implement the steps of the method of the present invention.

[0023] As an inventive concept, the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the method of the present invention are implemented.

[0024] Compared with the prior art, the beneficial effects of the present invention are: the solution of the present invention can more accurately estimate the symbol timing and carrier frequency offset, and greatly expand the estimation range of the frequency offset. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is the structure diagram of the synchronization sequence adopted by the embodiment of the present invention;

[0026] Figure 2 It is the source diagram of the synchronization sequence CP and CS of the embodiment of the present invention;

[0027] Figure 3 It is the performance comparison diagram of the timing function of the embodiment of the present invention;

[0028] Figure 4 It is the timing capture probability of different α in the embodiment of the present invention;

[0029] Figure 5 is the MSE of timing offset estimation for the embodiments of the present invention;

[0030] Figure 6 is the mean value of frequency offset estimation for the embodiments of the present invention;

[0031] Figures 7(a) and 7(b) are the MSE of frequency offset estimation for the embodiments of the present invention; Figure 7(a) is for a subcarrier spacing of f = 0.3; Figure 7(b) is for a subcarrier spacing of f = 10.3; Detailed implementation manners

[0032] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, rather than all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0033] Embodiment 1

[0034] A complete OFDM system can be divided into two parts: a transmitter and a receiver. In the transmitting part, the time-domain baseband OFDM symbol after IFFT can be expressed as:

[0035]

[0036] where j is the imaginary unit. N is the total number of subcarriers, N u is the number of valid subcarriers, L is the length of the guard interval, including the cyclic prefix and the cyclic suffix. X(k) represents the k-th subcarrier data symbol to be transmitted in the frequency domain, and x(n) represents the time-domain data sampled at continuous time n. Inter-symbol interference is eliminated by inserting a cyclic prefix longer than the channel impulse response in the time domain.

[0037] In the receiving part, due to the influence of reflections at various different interfaces of the signal in the channel, the received signal is a superposition of signals transmitted along different paths with different delays at the receiving end. When there is relative motion between the receiving end and the transmitting end, Doppler frequency shift will occur. Generally, for the same packet of data, if the communication bandwidth is not too wide, the Doppler frequency shift is approximately equal. The normalized carrier frequency offset can convert it into an offset ratio relative to the subcarrier spacing, which is divided into fractional frequency offset and integer frequency offset.

[0038]

[0039] where f offsetrepresents the carrier frequency offset between the transmitting and receiving ends, Δf represents the subcarrier spacing, and ε represents the normalized carrier frequency offset.

[0040] Timing offset is usually modeled as a delay, and frequency offset is modeled as the phase distortion of the received data. Then the final received signal after the signal passes through the channel is:

[0041]

[0042] where p is the number of channel multipaths, h(i) is the channel attenuation coefficient of the i-th path, τ i is the delay of the i-th path of the channel relative to the first path, d is the symbol timing offset, and ω(n) is the additive white Gaussian noise with a mean of 0 and a variance of σ ω 2 of.

[0043] In an OFDM system, the main goal of synchronization is to accurately estimate and compensate for timing offset and frequency offset.

[0044] Compared with traditional PN sequences, CAZAC sequences have better autocorrelation and cross-correlation characteristics, which can improve the synchronization performance in Gaussian channels and multipath fading channels. Its correlation function can be described as:

[0045]

[0046] where N1 is the period of the CAZAC sequence. According to the literature G. Ren, Y. Chang, H. Zhang, and H. Zhang, "Synchronization Method Based on a New Constant Envelop Preamble for OFDM Systems," IEEE Transactions on Broadcasting, vol. 51, no. 1, pp. 139 - 143, 2005. and M. M. U. Gul, X. Ma, and S. Lee, "Timing and frequency synchronization for OFDM downlink transmissions using Zadoff-Chu sequences," IEEE transactions on Wireless Communications, vol. 14, no. 3, pp. 1716 - 1729, 2014., a(k) can be expressed as:

[0047]

[0048] where μ is a positive integer relatively prime to k. When μ takes 1 and N1 takes N / 4, a(k) can be expressed as

[0049]

[0050] The structure of the training sequence adopted in the embodiments of the present invention is as Figure 1 shown, where A is a CAZAC sequence with a length of N / 4, B' is a sequence generated by conjugate symmetry of A and then weighted by an m-sequence, CP is a cyclic prefix, and CS is a cyclic suffix.

[0051] where A is composed of a(k), that is, A = [a(0), a(1),..., a(N / 4 - 1)]. After copying and splicing A, the first half S of the synchronization sequence can be obtained A , that is, S A = [A A]. B is the conjugate symmetric and weighted sequence of A. After copying and splicing, S B can be obtained, that is, S B = [B B]. Weighting can make the timing function curve sharper to improve the timing estimation performance. Then, S B is weighted to obtain the second half S B ' of the synchronization sequence, which can be expressed as:

[0052] S B ' = m s × S B (7)

[0053] where m s is a bipolar m-sequence. In a complex multipath channel, sometimes the path with the strongest received signal energy is not the first path. Therefore, when performing OFDM communication, in addition to adding a CP at the front of the OFDM symbol, a cyclic suffix (CS) is often added at the back to resist ISI. Similarly, the synchronization sequence designed in the embodiments of the present invention also includes both a CP and a CS. The CP is obtained from the end of the first half of the sequence, and the CS is obtained from the front of the second half of the sequence. As Figure 2 shown.

[0054] The embodiments of the present invention adopt the proposed training sequence as the synchronization sequence, and utilize the characteristic that the second half of the sequence is the symmetric conjugate and weighted of the first half. The weighting factor can reduce the influence of the multipath channel on the synchronization at the receiving end, so that the front and back parts of the synchronization signal have better correlation. Therefore, a new timing metric function is defined as:

[0055]

[0056] Here

[0057]

[0058] Among them, P1(d) and P2(d) are the symmetric correlation function value and the delay correlation function value of the received signal respectively. The main function of R(d) is to perform energy normalization. r(d) is the received signal, and m s = [m s1 m s2 is the m-sequence weighted for the second half of the sequence. m s1 and m s2 are used to weight the third and fourth parts of the synchronized sequence after being equally divided into four parts respectively. Through the weighting operation, it is ensured that the timing metric function M(d) will obtain a sharp peak at the correct timing point, and almost zero at other positions, which makes the timing offset estimation have a lower error rate. Therefore, the estimated value of the timing offset is

[0059]

[0060] To better verify the performance of the method in the embodiment of the present invention, Figure 3 it is shown that under an ideal channel, the timing function curves of the method proposed in the embodiment of the present invention are compared with those of the Schmidl method, the Minn method, and the Ren method. Among them, the OFDM symbol length is 1024, the CP length is 128, and the correct timing position is at 0. It can be seen that due to the cyclic prefix, the Schmidl method shows a platform effect near the correct timing point, and its performance is easily affected by interference; the Minn method shows a ramp-shaped metric curve, and there are large sidelobes at ±N / 4 and ±N / 2 of the correct timing position, making it vulnerable to interference; the Ren method has a sharp pulse-shaped metric curve at the correct timing position, but due to the cyclic prefix, sidelobes appear, so its performance is also affected by interference at low signal-to-noise ratios; while the method proposed in the embodiment of the present invention can generate a sharper peak at the correct timing position, and there is no interference from sidelobes, enabling the method to provide a more accurate timing position estimate.

[0061] Generally, the estimation of the carrier frequency offset can be divided into the estimation of the fractional frequency offset and the integer frequency offset. To improve the estimation performance of the carrier frequency offset, the embodiment of the present invention divides the estimation of the carrier frequency offset into two steps: coarse frequency offset estimation and fine frequency offset estimation.

[0062] The coarse frequency offset estimation is performed by using the first half of the received designed training sequence. Its function is to perform FFO estimation while compensating the size of the IFO to an even value, which provides a guarantee for the correct estimation of the IFO fine estimation method in the frequency domain. This method has low requirements for the accuracy of timing synchronization. When there are multipaths or other interferences in the channel that cause the timing synchronization to fall within the CP, it will not affect the estimation accuracy of this method. The specific operation of this method is as follows:

[0063] The theoretical derivation takes the case of a signal passing through an additive white Gaussian noise channel. The received signal r(n) can be expressed as:

[0064] r(n) = x(n)e j2πnε / N + ω(n) (11)

[0065] where x(n) is the transmitted signal at the transmitter, and ω(n) is the noise introduced by the signal passing through the channel. By coarsely timing synchronizing at point d0 and intercepting the synchronization sequence in the received signal, the corresponding method for coarse frequency offset estimation is

[0066]

[0067] As can be seen from Equation (12), at the coarse timing synchronization point d0, the last three terms except the first term have very little correlation, and their calculation results are very small compared to the first term and can be ignored. Then the above equation can be simplified to

[0068]

[0069] Let ε = (a + 2zπ) / π. Then the first term a / π is the fractional frequency offset, and the second part 2z is the integer frequency offset, where |a| < π and z is an integer. Substitute ε into the above equation

[0070]

[0071] Taking the phase of F(d0) gives

[0072]

[0073] Therefore, when z is even, taking the phase of F gives an estimate of a; when z is odd, taking the phase of F gives an estimate of a and simultaneously compensates for z being even. Then the coarse frequency offset synchronization can be described as

[0074]

[0075] The embodiment of the present invention designs an IFO estimation method and a fine timing estimation method adapted to a complex multipath channel. When the path with the strongest energy is not the first path, the traditional method often cannot obtain the accurate starting position of the signal. This method utilizes the weighted characteristics of the second half of the sequence to achieve accurate IFO estimation, and utilizes the first half of the sequence to estimate the starting position of the signal.

[0076] As shown in Equation (11), let the deviation between the position of the first path of the received signal and the position of the path with the strongest energy be m * , and after the coarse frequency offset synchronization, the fractional frequency offset has been compensated. Then the synchronized signal received after the coarse timing synchronization point d0 can be expressed as

[0077]

[0078] For the second half of the sequence, the time-domain signal is the symmetric conjugate sequence of the CAZAC sequence and is weighted. The symmetric conjugate operation does not affect the characteristics of the CAZAC sequence, but the time-domain weighting of the m-sequence makes it necessary to multiply by the weighting factor m before performing the FFT for this part. s To eliminate the influence of weighting, the received signal after removing the influence of the weighting factor can be expressed as

[0079]

[0080] This results in that when there are multiple paths in the received signal, and considering that the coarse timing synchronization position d0 is the position of the strongest energy path, so while eliminating the influence of the weighting factor for the second half of the synchronization sequence, the signals of other paths except the strongest energy path can be regarded as noise due to the mismatch of the weighting factors, and only the signal of the strongest energy path can continue to maintain the characteristics of the CAZAC sequence. After transforming r'(n) to the frequency domain through FFT, it is

[0081]

[0082] As can be seen from Equation (19), the integer multiple frequency offset z is manifested as the cyclic shift of its frequency-domain sequence. After the coarse frequency offset compensation, z has been compensated to an even number, so k - z still keeps the data in the even positions. For the convenience of comparison with the original CAZAC sequence, the even-position data of this sequence are taken out to form a new sequence, that is

[0083]

[0084] Here b ∈ [0, N / 4 - 1], X = [X A X B represents the data of the even frequencies after the FFT of the synchronization sequence, that is, X A and X B respectively represent the sequences composed of the even-frequency data of the first half and the second half after removing the weighting of the synchronization sequence. At this time, for the second half of the sequence

[0085]

[0086] As can be seen from Equation (21), the frequency-domain sequence of the second half of the synchronization sequence has a cyclic shift of z / 2 relative to the original sequence at this position. Then the decision function of the integer multiple frequency offset can be expressed as

[0087]

[0088] Where

[0089]

[0090] Where, B FIt represents the sequence composed of the even-frequency data after FFT of the second half of the original synchronization sequence at the transmitting end. Since CAZAC has good autocorrelation characteristics of the sequence, at the correct frequency offset position, Q1(g) has a sharp peak. Then the finally obtained normalized frequency offset estimate value is ε = ε1 + ε2.

[0091] Since the received signal is the superposition of signals passing through different paths at the receiving end, each path is relatively independent. When correlated with the original sequence, independent peaks will appear in each path. By using the mutual relationship between the signal with the strongest energy in a path and the first-path signal, the deviation between the current synchronization position and the starting position of the signal is obtained, so as to estimate the starting position of the received signal passing through a complex multipath channel.

[0092] At this time, for the first half of the sequence, the formula can be further simplified to obtain

[0093]

[0094] It can be seen from Equation (24) that the first-path component of the received synchronization sequence has a cyclic shift of z / 2 + m * relative to the original sequence. Since the cyclic prefix length is greater than the channel length, by performing a reverse cyclic shift correlation of length L (the cyclic prefix length) between the frequency-domain sequence of the received synchronization signal and the original sequence, sharp peaks will appear after correlation of the received signals of different paths. By setting a threshold, the position of the first path can be found. In order to achieve accurate estimation of the first-path position at low signal-to-noise ratios, the threshold is set to a fractional multiple of the correlation result corresponding to the path with the strongest energy, so as to achieve adaptive dynamic change of the threshold at different signal-to-noise ratios, and the position of the first path of the received signal can still be accurately obtained at low signal-to-noise ratios. Then the decision function of the deviation between the coarse timing synchronization position and the starting moment of the signal can be expressed as

[0095] m * = max(Q2(x) -1 (y > α·max(Q2(x)))) - z / 2 (25)

[0096] where

[0097]

[0098] where y represents the correlation result of Q2, and X A is the sequence composed of the even-frequency data after FFT of the first half of the original synchronization sequence. α ∈ (0, 1) is the threshold parameter. When α is small, it is easily affected by noise, and when α is large, it will cause missed detection of paths with small energy. In order to achieve reliable detection of the first-path position, according to experience, α is in [0.2 - 0.4]

[0099] can achieve the best performance. Here, α = 0.3 is taken. So the starting moment d1 of the signal is

[0100] d1 = d0 - m * (27)

[0101] Simulations are carried out in a multipath fading channel by setting different α values to verify the reliability of the timing fine synchronization method. Among them, the length of the OFDM symbol is 1024, and the length of the guard interval is 128. The channel uses an 8 - tap uniform distribution, with an interval of 16 sampling points between adjacent taps. The channel attenuation satisfies exponential attenuation, and the tap powers of the first path and the last path differ by 13 dB.

[0102] 10000 simulations are carried out for each SNR.

[0103] It can be seen that when only the timing coarse synchronization method is used, the highest timing synchronization capture probability is less than 45%. When the timing fine synchronization method is introduced, the timing capture probability is significantly improved. As Figure 4 shown, when α = 0.1, the timing capture probability is even close to 100% at the highest, but this parameter performs poorly at low SNR. To ensure the timing estimation performance at low SNR, it is more reasonable to use α = 0.3. At this time, the highest synchronization capture probability can reach about 93%, which is more than twice the performance improvement compared to only using the timing coarse synchronization method.

[0104] Compared with the traditional method, the method of the embodiment of the present invention is not only applicable to complex multipath channels, but also only needs one synchronization sequence to achieve accurate and stable time - frequency estimation, and finally obtain the accurate starting position of the signal.

[0105] To verify the effectiveness of this method, the embodiment of the present invention builds an OFDM system simulation based on the method of the embodiment of the present invention in the Matlab environment, and compares the timing capture performance and carrier frequency offset estimation performance in a Gaussian channel and a complex multipath channel under different SNRs. The OFDM system uses 1024 sub - carriers, a 1024 - point IFFT / FFT, and the guard interval length is 128. The multipath fading channel uses an 8 - tap uniform distribution, with an interval of 16 sampling points between adjacent taps. The channel attenuation satisfies exponential attenuation, and the tap powers of the first path and the last path differ by 13 dB. The channel coefficients are time - invariant because the coherence time is much longer than the pulse duration. Without special instructions, 10000 simulations are carried out for each SNR to ensure the reliability of the method.

[0106] To verify the effectiveness of the method proposed in the embodiment of the present invention, the timing and frequency synchronization performance of the method proposed in the embodiment of the present invention is evaluated by the mean square error (MSE).

[0107] By Figure 5It can be seen that the MSE of the method proposed in the embodiment of the present invention is significantly lower than that of the Ren method and the Minn method, and it also verifies that the timing fine synchronization method proposed in the embodiment of the present invention can bring great performance improvement. Since MSE reflects the deviation and variance of the estimation, it can be inferred that the timing estimation performance of the method proposed in the embodiment of the present invention is better than that of the Ren method and the Minn method.

[0108] Figure 6 Figure 4 shows the frequency offset estimation values of the method proposed in the embodiment of the present invention, the Ren method, and the Minn method in an AWGN channel when SNR = 20 dB. We observe that the frequency offset estimation range of the Minn method is only |f| ≤ 2, while the frequency offset estimation range of the method proposed in the embodiment of the present invention is |f| ≤ N / 2, which is the same as the estimation range of the Ren method and much larger than that of the Minn method. Therefore, the method proposed in the embodiment of the present invention has better frequency offset estimation performance.

[0109] Figures 7(a) and 7(b) show the MSE of frequency offset estimation of different methods in a multipath fading channel with different frequency offsets. It can be seen that when the frequency offset is 0.3 sub-carrier spacings (carrier intervals), all three methods have good performance, but the method proposed in the embodiment of the present invention has better performance when the SNR is higher. When the frequency offset is 10.3 sub-carrier spacings, at this time, the performance of the Minn method degrades due to its smaller frequency offset range, while the method in the embodiment of the present invention still maintains good performance. Therefore, the method in the embodiment of the present invention has a larger estimation range without sacrificing accuracy and has better accuracy under the same estimation range.

[0110] The simulation results show that the timing and frequency synchronization performance of the method proposed in the embodiment of the present invention is better than other methods, providing a better choice for the time-frequency synchronization of OFDM systems.

[0111] In an OFDM system, symbol timing and carrier frequency offset will seriously affect the system performance. Based on CAZAC sequences, the embodiment of the present invention designs a new training sequence and proposes a corresponding OFDM symbol timing and frequency offset synchronization method. This method achieves very accurate timing estimation and carrier frequency offset estimation with only one OFDM symbol, overcomes the drawback of excessive timing estimation deviation of traditional methods in complex channels, and at the same time provides a very wide frequency offset estimation range. Therefore, this method is suitable for the time-frequency synchronization of OFDM systems.

[0112] In an OFDM system, symbol timing and carrier frequency offset will seriously affect the system performance. Embodiments of the present invention are based on CAZAC sequences, design a training sequence and propose a corresponding OFDM symbol timing and frequency offset synchronization method. The method of the embodiments of the present invention realizes very accurate timing estimation and carrier frequency offset estimation only with one OFDM symbol, overcomes the drawback of excessive timing estimation deviation of traditional methods in complex channels, and meanwhile provides a very wide frequency offset estimation range. Therefore, this method is suitable for time-frequency synchronization of OFDM systems.

[0113] Embodiment 2

[0114] Embodiment 2 of the present invention provides a terminal device corresponding to Embodiment 1 above. The terminal device can be a processing device for a client, such as a mobile phone, a laptop computer, a tablet computer, a desktop computer, etc., to execute the method of the above embodiment.

[0115] The terminal device of this embodiment includes a memory, a processor, and a computer program stored on the memory; the processor executes the computer program on the memory to implement the steps of the method of Embodiment 1 above.

[0116] In some implementations, the memory can be a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.

[0117] In other implementations, the processor can be various types of general-purpose processors such as a central processing unit (CPU), a digital signal processor (DSP), etc., which are not limited herein.

[0118] Embodiment 3

[0119] Embodiment 3 of the present invention provides a computer-readable storage medium corresponding to Embodiment 1 above, on which a computer program / instructions are stored. When the computer program / instructions are executed by a processor, the steps of the method of Embodiment 1 above are implemented.

[0120] A computer-readable storage medium can be a tangible device that holds and stores instructions used by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination of the above.

[0121] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The solutions in the embodiments of the present application can be implemented using various computer languages. For example, object-oriented programming languages such as Java and interpreted scripting languages such as JavaScript, etc.

[0122] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, such that a series of operation steps are executed on the computer or other programmable devices to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable devices provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0124] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed as including the preferred embodiments as well as all changes and modifications falling within the scope of the present application.

[0125] Obviously, those skilled in the art can make various changes and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these changes and variations.

Claims

1. A time-frequency synchronization method for an OFDM system, characterized in that Including: Determine the estimated value of the timing offset using the following formula: Among them, d is the position at the start time of the signal, N is the length of the synchronization sequence, P1(d) is the symmetric correlation value of the front and back parts of the received synchronization sequence, P2(d) is the delay correlation value obtained by dividing the received synchronization sequence into four equal parts, r(d) is the received synchronization signal intercepted with d as the signal start position, r(d + N / 2 - n - 1) represents the received signal sample value with a length of N / 2 - n - 1 after the d position, m s = [m s1 m s2 is the m-sequence that weights the second half of the synchronization sequence, m s1 and m s2 are the m-sequences that weight the third and fourth parts of the synchronization sequence after being divided into four equal parts at the transmitter end respectively. R(d) is half of the energy of the received signal with a length of N after the d moment; The synchronization sequence includes: a cyclic prefix, two CAZAC sequences A with a length of N / 4, two sequences B' generated by conjugate symmetry of sequence A followed by m-sequence weighting, and a cyclic suffix; The normalized frequency offset estimation value ε is calculated using the formula ε = ε1 + ε2; where a / π is the fractional frequency offset and 2z is the integer frequency offset; Among them, B F is a sequence composed of the even-frequency data after removing the weighting from the second half of the original synchronization sequence at the transmitting end and then performing FFT. g is the cyclic shift amount, and B F * (b - g) represents performing a conjugate operation on B F and then performing a cyclic shift of length g. X B is a sequence composed of the even-frequency data after removing the weighting from the second half of the synchronization sequence at the receiving end and then performing FFT. X B (b - z / 2) is the influence of the existing integer multiple frequency offset z on the frequency-domain data of the second half of the received synchronization sequence, resulting in a cyclic shift of z / 2. m * is the deviation between the position of the first path of the received signal and the position of the path with the strongest energy. ω(2b) is the noise influence introduced by the channel when b is the independent variable, and b ∈ [0, N / 4 - 1].

2. The OFDM system time-frequency synchronization method according to claim 1, characterized in that The deviation m between the position of the first path of the received signal and the position of the path with the strongest energy * is calculated by the following formula: m * = max(Q2(x) -1 (y > α·max(Q2(x)))) - z / 2; Among them, L is the length of the guard interval, x is the cyclic shift amount, y represents the result of Q2(x) when x ∈ [0, L], and A F * (b) represents conjugating A F and then corresponding to different values of the independent variable b. A F is a sequence composed of the even-frequency data after FFT of the first half of the original synchronization sequence at the transmitting end, and X A is a sequence composed of the even-frequency data after FFT of the first half of the synchronization sequence at the receiving end. X A (b - z / 2) represents the cyclic shift effect of z / 2 on the frequency-domain data of the first half of the received synchronization sequence caused by the existing integer multiple frequency offset z, and α is the threshold parameter, where α ∈ (0, 1).

3. The OFDM system time-frequency synchronization method according to claim 2, characterized in that α takes a value of 0.2 to 0.

4.

4. The OFDM system time-frequency synchronization method according to claim 3, characterized in that α takes a value of 0.

3.

5. A terminal device, characterized in that, Including: One or more processors; A memory storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the steps of the method according to any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that It stores a computer program, which when executed by a processor, implements the steps of the method according to any one of claims 1 to 4.

Images