OFDM system synchronization method based on ZC sequence

Abstract

The invention provides a ZC sequence-based OFDM (Orthogonal Frequency Division Multiplexing) system synchronization method, which comprises the following steps of: firstly constructing two ZC sequences ZC1 and ZC2 with different roots, then valuing and arranging the ZC1 sequence and the ZC2 sequence to obtain a training sequence, carrying out IFFT (Inverse Fast Fourier Transform) on the training sequence, adding a cyclic prefix to a data frame, then sending the training sequence by a sending end and transmitting the training sequence through a channel, and calculating the cross correlation between the receiving end data and the local training sequence, finally selecting a symbol synchronization position according to a cross correlation calculation value, and after the synchronization position is determined, estimating the carrier frequency offset. According to the scheme of the invention, accurate symbol synchronization and frequency offset estimation can be carried out in a large-fading and low-signal-to-noise-ratio wired channel environment, the synchronization reliability is enhanced, and the operand is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of OFDM symbol timing synchronization, and particularly to an OFDM system synchronization method based on ZC sequences. Background Art

[0002] In a wired twisted pair channel, due to the influence of various adverse factors during signal transmission, including channel fading and noise interference, a serious symbol out-of-synchronization phenomenon occurs between the receiving end and the transmitting end.

[0003] In traditional symbol synchronization algorithms, the Schmidl algorithm may lead to inaccurate judgment of the starting position of the OFDM symbol due to the peak plateau problem in the timing metric function; the Minn algorithm is prone to starting position detection errors due to the side peak phenomenon, and the main peak of the Park algorithm is not obvious enough. Therefore, the traditional carrier estimation method based on the cyclic prefix has low accuracy, and the frequency offset estimation is inaccurate in the case of low signal-to-noise ratio. Summary of the Invention

[0004] In view of the problem that symbol synchronization is difficult in a traditional OFDM communication system under high noise conditions, the present invention proposes an OFDM system synchronization method based on ZC sequences. This method redesigns the training sequence as an equal-amplitude zero autocorrelation sequence and designs a new training sequence construction method. After passing through the additive white Gaussian noise (AWGN) and the twisted pair channel, by calculating the cross-correlation with the local sequence and according to the characteristics of its own training sequence, the carrier frequency offset can be determined through the calculation of autocorrelation, improving the accuracy of symbol synchronization and carrier frequency offset estimation.

[0005] An OFDM system synchronization method based on ZC sequences includes the following steps:

[0006] Step S1, construct two ZC sequences ZC1 and ZC2 with different roots. The ZC1 and ZC2 sequences are expressed as:

[0007]

[0008] where N represents the number of bits carried by the subcarriers, M1 and M2 represent different prime numbers of N, and M2 = N - M1;

[0009] Step S2, take values and arrange the ZC1 and ZC2 sequences. Name the first N / 2 arrays of the ZC1 sequence as A, name the first N / 2 arrays of the ZC2 sequence as B, and then take the conjugate of B to form the [A B * sequence, and finally form a training sequence X with a total length of N n ;

[0010] Step S3, for the training sequence X nPerform IFFT transformation;

[0011] Step S4, add a cyclic prefix with length Ng to the transmission symbol;

[0012] Step S5, the transmitting end sends out the training sequence after IFFT transformation and transmits it through the channel;

[0013] Step S6, perform cross-correlation calculation between the received data and the local reference sequence;

[0014] Step S7, select the symbol synchronization position according to the cross-correlation calculation value;

[0015] Step S8, according to the maximum value m sync Intercept the sequence at the position to obtain r sync After that, perform carrier frequency offset estimation.

[0016] Furthermore, the autocorrelation expressions of two ZC sequences are as follows:

[0017]

[0018] Among them, k represents the position of the symbol in the sequence, and m represents the time shift amount between the two ZC sequences;

[0019] When M1 and M2 are equal, set M1 = M2 = M, then there is:

[0020]

[0021] When m = 0, the autocorrelation function is R 11 (0): R 11 (0) = N;

[0022] When m ≠ 0, the autocorrelation function R 11 (m) = 0.

[0023] Furthermore, since In the second training sequence, k 2 And the parity of k is the same, then there is The training sequence X n Is transformed through IFFT to obtain x[n]. The first half of x[n], that is, the expression of the first N / 2 is:

[0024]

[0025] The expression of the second half of x[n] is:

[0026]

[0027] Further, in step S5, assuming that the transmitted training sequence is represented as x(n) in the time domain, the channel impulse response is h(n), the data obtained at the receiving end is r(n), and the Gaussian white noise is ω(n), then we have: r(n) = x(n) * h(n) + ω(n).

[0028] Further, in step S6, assuming that y[n] is the local reference sequence, then the cross-correlation function R xy (m) is:[[]]END]]

[0029] Further, in step S7, the delay at the symbol synchronization moment is determined by determining the maximum value of the cross-correlation. Assuming that the offset m corresponding to the maximum value sync is: m sync = arg max m |R xy (m)|, where || represents taking the modulus.

[0030] Further, the carrier frequency offset estimation in step S8 is specifically as follows:

[0031]

[0032] f CFO is the value of the carrier frequency offset, where Ng represents the length of the cyclic prefix, N represents the length of the subcarrier, and ∠(·) represents the phase angle of the complex number.

[0033] The beneficial technical effects of the present invention are:

[0034] The present invention proposes an OFDM system synchronization method based on ZC sequences. By innovatively designing a composite training sequence based on multiple Zadoff-Chu sequences, the accuracy of symbol timing synchronization and the robustness of the system are significantly improved. Due to the excellent auto-correlation and cross-correlation characteristics of the Zadoff-Chu sequences, the proposed training sequence can still achieve efficient and reliable symbol synchronization in a large fading environment;

[0035] Compared with the existing symbol synchronization algorithm, the Schmidl algorithm, the present invention effectively solves the peak platform problem existing in the timing metric function, significantly improves the timing accuracy. Compared with the Minn algorithm, there is no side peak problem. Compared with the Park algorithm, the main peak value is increased, further enhancing the reliability of synchronization;

[0036] The method of the present invention only needs to perform one IFFT operation. First, the sequence is arranged, and then the overall IFFT transformation is performed, rather than first performing IFFT transformation on a predetermined number of sequences and then sorting the sequences, reducing the amount of computation;

[0037] In summary, the novel ZC sequence symbol synchronization method proposed by the present invention can not only clearly capture the effective peak through the cross-correlation operation with the local sequence, but also greatly reduce the computational amount at the receiving end, thereby improving the synchronization accuracy and system efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0039] Figure 1 FIG. is a schematic flowchart of a symbol synchronization method for an OFDM system based on ZC sequences provided by an embodiment of the present invention;

[0040] Figure 2 FIG. is a schematic diagram of the construction of a training sequence provided by an embodiment of the present invention;

[0041] Figure 3 FIG. is an autocorrelation function graph of the ZC combined sequence proposed by the present invention provided by an embodiment of the present invention;

[0042] Figure 4 FIG. is an autocorrelation function graph of the sequences of the traditional Schmidl algorithm, Minn algorithm, and Park algorithm provided by an embodiment of the present invention;

[0043] Figure 5 FIG. is a cross-correlation graph between the ZC sequence and the local sequence of an OFDM system using a twisted pair channel in a 0 dB noise environment provided by an embodiment of the present invention;

[0044] Figure 6 FIG. is a cross-correlation graph between the sequences of the Schmidl algorithm, Minn algorithm, and Park algorithm and the local sequence of an OFDM system using a twisted pair channel in a 0 dB noise environment provided by an embodiment of the present invention;

[0045] Figure 7 FIG. is a graph comparing the accuracy of the training sequence provided by the method of the present invention and the traditional cyclic prefix-based carrier frequency offset estimation method provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. 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.

[0047] An OFDM system synchronization method based on ZC sequences, as Figure 1 shown, includes the following steps:

[0048] Step S1: Construct two ZC sequences ZC1 and ZC2 with different roots. The ZC1 and ZC2 sequences are expressed as:

[0049]

[0050] where N represents the number of bits carried by subcarriers, M1 and M2 represent different prime numbers of N, M2 = N - M1. The following combination is performed on the two ZC sequences with different roots, and the sequence still has good autocorrelation. The autocorrelation formula is as follows:

[0051]

[0052] where k represents the position of the symbol in the sequence, and m represents the time shift amount between the two ZC sequences. Substituting the two ZC sequences into the autocorrelation expression, we get:

[0053]

[0054] After arranging the exponential terms, we get:

[0055]

[0056] When M1 and M2 are equal, set M1 = M2 = M, then we have:

[0057]

[0058] When m = 0, the autocorrelation function is R 11 (0): R 11 (0) = N;

[0059] When m ≠ 0, the autocorrelation function R 11 (m) = 0;

[0060] When M1 and M2 are not equal, the exponential term in the autocorrelation expression contains different frequency terms, which means that the spectra of the two sequences are different. This is because the spectrum of the ZC sequence is determined by the root parameter. Therefore, when the two root parameters are not equal, the spectra of the two ZC sequences do not overlap and can be considered orthogonal, and the autocorrelation at all delays is zero. So the combination of two sequences with different roots does not affect the correlation.

[0061] In the present invention, M1 and M2 selected for the ZC sequence are respectively any prime numbers of N. In the embodiment, it is assumed that N = 512, and M1 and M2 are 3 and 509 respectively. Obviously, R 12 (m) is not 0. So when combining two ZC sequences with different ends, no correlation will be generated between the sequences.

[0062] Step S2: Take values and arrange the ZC1 and ZC2 sequences

[0063] Take the first N / 2 arrays of the ZC1 sequence and name them A, and take the first N / 2 arrays of the ZC2 sequence and name them B. That is, select the first 256 items of ZC1 and ZC2, and then take the conjugate of B to form the [A B * sequence. As Figure 2 shown, finally form a training sequence X with a total length of N n .

[0064] Step S3: Perform IFFT transformation on the training sequence X n to obtain:

[0065]

[0066] Since in the second training sequence k 2 and the parity of k are equal, so now perform IFFT transformation on the X[n] sequence to obtain x[n]. The first half of x[n], that is, the expression of the first N / 2 is:

[0067]

[0068] The second half of x[n], that is, the expression for obtaining x[n + N / 2] is:

[0069]

[0070] Comparing the expressions of the first half and the second half, it can be obtained that after the training sequence passes through IFFT, the first half and the second half differ by (-1) k , multiply the second half by (-1) k, making the first half and the second half equal, which can be used for subsequent carrier frequency offset estimation. And according to the characteristics of the ZC sequence, after the IFFT transformation, the autocorrelation and cross-correlation characteristics of the sequence will not be affected, thus not interfering with its own orthogonality and ideal correlation performance.

[0071] Among them, the first (-1) k is from the second half of the frequency-domain sequence definition (the (-1) k ) introduced after conjugating the sequence with root N - r1), and the second (-1) k is from the (-1) jπk naturally introduced by the IFFT property after shifting the time-domain sequence index by N / 2 points (i.e., e k .

[0072] Step S4: Add a cyclic prefix with length Ng to the transmitted symbol;

[0073] Step S5: The transmitter sends out the training sequence after the IFFT transformation and transmits it through the channel;

[0074] Assume that the transmitted training sequence is represented as x(n) in the time domain, the channel impulse response is h(n), the data obtained at the receiver is n(n), and the Gaussian white noise is ω(n). Then there is: r(n) = x(n) * h(n) + ω(n);

[0075] Step S6: Perform cross-correlation calculation on the receiver data and the local reference sequence

[0076] The cross-correlation function R xy (m) is used to measure the similarity between two signals r[n] and y[n] at different time offsets m. The cross-correlation function gives the "degree of overlap" of the signals within a time window. If the signals r[n] and y[n] are aligned at a certain moment, the cross-correlation value between the two will reach the maximum, indicating the best matching degree of the two signals.

[0077] Assume that y[n] is the local reference sequence, and the cross-correlation function R xy (m) is:

[0078] Step S7: Select the symbol synchronization position according to the cross-correlation calculation value

[0079] Perform cross-correlation calculation on the receiver sequence and the local reference sequence, and determine the delay at the symbol synchronization moment by determining the maximum value of the cross-correlation. The cross-correlation value R xy (m) gives the similarity between the received signal and the local reference sequence at different time offsets. Therefore, the offset m sync corresponding to the maximum value is the most appropriate synchronization position: m sync = arg maxm |R xy (m)|, where || represents taking the modulus.

[0080] Step S8: According to the maximum value m after symbol synchronization sync Intercept the sequence at the position to obtain r sync After that, perform carrier frequency offset estimation, specifically:

[0081]

[0082] where Ng represents the length of the cyclic prefix, N represents the length of the subcarrier, r sync represents the sequence after symbol synchronization, and ∠(·) represents the phase angle (in radians) of the complex number.

[0083] The autocorrelation functions of the ZC sequence and the traditional method sequence proposed by the present invention are respectively as Figure 3 and Figure 4 shown. It can be seen from the figure that after introducing Gaussian white noise, when the signal-to-noise ratio of the signal decreases, the training sequence proposed by the present invention can effectively maintain the autocorrelation performance. Compared with the traditional method, the peak value of the main peak of the method of the present invention is less affected by the decrease of the signal-to-noise ratio.

[0084] The cross-correlation functions of the ZC sequence and the traditional method sequence proposed by the present invention in the OFDM system are respectively as Figure 5 and Figure 6 shown. The subcarrier length is selected as N = 512, the roots M1 and M2 are respectively set to 3 and 5, the carrier frequency is 10 MHz, the sampling frequency is 80 MHz, the signal-to-noise ratio S / N = 0 dB, and it is transmitted through a twisted pair channel. It can be seen from the figure that the method of the present invention is significantly superior to the traditional method in terms of peak performance.

[0085] Figure 7 is the accuracy comparison chart of the training sequence provided by the method of the present invention and the traditional cyclic prefix-based carrier frequency offset estimation method. It can be seen from the figure that after symbol synchronization, when the signal-to-noise ratio of the signal is relatively low, the accuracy of the carrier frequency offset estimation of the method of the present invention is improved. Compared with the traditional method, the frequency offset estimation accuracy of the method of the present invention is less affected by the decrease of the signal-to-noise ratio.

[0086] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for OFDM system synchronization based on ZC sequence, characterized in that: The method comprises: Step S1, construct two ZC sequences ZC1 and ZC2 with different roots, and the ZC1 and ZC2 sequences are expressed as: Wherein, n represents the number of bits carried by the subcarrier, M1 and M2 represent different prime numbers of N, and M2 = N-M1; Step S2, taking values ​​of ZC1 and ZC2 sequences and arranging them, taking the first N / 2 arrays of ZC1 sequence and naming them A, taking the first N / 2 arrays of ZC2 sequence and naming them B, and then conjugating B to form [AB * ] sequence, and finally form a training sequence X with a total length of N n ; Step S3: training sequence X n Perform IFFT transformation; Step S4, adding a cyclic prefix of length Ng to the transmission symbol; Step S5, the transmitting end sends out the training sequence after IFFT transformation and transmits it through the channel; Step S6, performing cross-correlation calculation on the receiving end data and the local reference sequence; Step S7, selecting a symbol synchronization position according to the cross-correlation calculation value; Step S8, based on the maximum value m after symbol synchronization sync Position intercept sequence to get r sync After that, the carrier frequency offset is estimated.

2. The method according to claim 1, characterized in that The autocorrelation expression of two ZC sequences is as follows: Where k represents the position of the symbol in the sequence, and m represents the time shift between two ZC sequences; When M1 and M2 are equal, set M1=M2=M, then: When m = 0, the autocorrelation function is R 11 (0):R 11 (0) = N; When m≠0, the autocorrelation function R 11 (m)=0.

3. The method according to claim 1, characterized in that because In the second training sequence, k 2 and k are equal, then we have The training sequence X n Through IFFT transformation, we get x[n], the first half of x[n], that is, the expression of the first N / 2 is: The expression for the second half of x[n] is:

4. The method according to claim 1, characterized in that In step S5, assume that the transmitted training sequence is represented in the time domain as x(n), the channel impulse response is h(n), the data obtained by the receiving end is r(n), and the Gaussian white noise is ω(n), then: r(n)=x(n)*h(n)+ω(n).

5. The method according to claim 1, characterized in that In step S6, let y[n] be the local reference sequence, then the cross-correlation function R xy (m) is:

6. The method according to claim 1, characterized in that In step S7, the delay of the symbol synchronization time is determined by determining the maximum value of the cross-correlation, and the offset m corresponding to the maximum value is set to sync For: m sync = arg max m |R xy (m)|, where || represents modulo.

7. The method according to claim 1, characterized in that In step S8, the carrier frequency offset estimation is specifically as follows: f CFo That is, the value of the carrier frequency deviation, where Ng represents the length of the cyclic prefix, N represents the length of the subcarrier, and ∠(·) represents the phase angle of the complex number.

Images