CN104052707B - High carrier number OFDM sample frequency fast synchronization methods - Google Patents

Abstract

The invention relates to a fast synchronization method for OFDM sampling frequency with a high number of carriers. It includes the steps: S1. Inserting pilot symbols at the sending end; S2. Using the captured estimated value of the previous symbol to perform sampling frequency offset pre-correction on the current symbol; S3. Obtaining the corresponding pilot phase difference; S4. Using the pilot phase Perform the sampling frequency offset capture value estimation of the current symbol, and accumulate the current symbol capture value and the previous symbol capture estimate value to obtain a new estimate value; S6. When the new capture value is greater than the set capture threshold, based on the frequency domain correction The method uses the capture estimation value of the current symbol to perform sampling frequency offset correction output on the current symbol; when the new capture value is less than the set capture threshold, output the pre-correction signal in step S2; the next symbol is updated to the current symbol and continue to execute S2‑ S4. The invention can perform real-time high-precision sampling frequency offset correction for each OFDM symbol.

Description

Fast Synchronization Method of High Carrier Number OFDM Sampling Frequency

technical field

The present invention relates to the technical field of communications, and more specifically, to a fast synchronization method for OFDM sampling frequency with a high number of carriers.

Background technique

With the rapid development of Internet technology, applications such as high-definition Internet TV, large-scale online games, and online video conferencing are gradually becoming popular. However, the current network speed in my country is far from enough. We need higher-speed access technology to meet the needs of people. For future bandwidth needs. Next Generation Digital Subscriber Line (NG-DSL, Next Generation Digital Subscriber Line) technology will be the most important high-speed wired access technology in the future. The bandwidth of this access technology will exceed 100M, and the transmission rate will exceed 1Gbps, which can well Combined with the optical fiber network, the network speed is greatly improved.

On the physical layer, the NG-DSL technology uses a high-level Orthogonal Frequency Division Multiplexing (OFDM, Orthogonal Frequency Division Multiplexing) technology. OFDM technology is a multi-carrier modulation technology with extremely high frequency band utilization, and it is widely used in wired and wireless access networks. Different from the traditional Digital Subscriber Line (DSL, Digital Subscriber Line) technology, the OFDM modulation method used in NG-DSL technology has two characteristics: high carrier number and high quadrature amplitude modulation (QAM, Quadrature Amplitude Modulation), because these two The big features make NG-DSL technology have high bandwidth and transmission rate, but it is precisely because of these two features that NG-DSL technology has higher requirements for signal synchronization than any previous access technology. Therefore, synchronization technology is particularly important in NG-DSL technology.

In the NG-DSL system, we convert the OFDM baseband signal generated by the D/A converter into an analog signal at the sending end, and convert the analog signal received at the receiving end through the channel transmission through the A/D converter. sampled as a digital signal. The most ideal situation is that the clock frequency of the A/D and D/A devices at the two ends of the transceiver is completely synchronized, and the signal sampled at the receiving end is an unbiased signal, but the clocks of the two devices are often impossible to be exactly the same. There is a certain frequency drift phenomenon in the crystal oscillator, which will cause inconsistency in the signals at the two ends of the transceiver, resulting in inter-carrier interference, thereby reducing the signal-to-noise ratio of the system. We need to maximize the estimation of the deviation of the crystal oscillator frequency at both ends of the transceiver, and correct the signal at the receiving end to improve the output signal-to-noise ratio of the system. Among the wired OFDM signals, the synchronization performance of the system is mainly affected by the sampling frequency synchronization (synchronization also includes symbol timing synchronization, but this kind of synchronization is easier to solve the impact on the system), especially in high-carrier like NG-DSL, In a high-rate OFDM system, the sampling frequency offset is particularly important to the deviation generated by the system. If the sampling frequency offset cannot be well estimated, the high-speed transmission of the system will be out of the question.

Contents of the invention

In order to overcome at least one defect (deficiency) of the above-mentioned prior art, the present invention provides a high-carrier OFDM sampling frequency rapid synchronization method that improves the accuracy of sampling frequency offset estimation.

In order to solve the problems of the technologies described above, the technical solution of the present invention is as follows:

A method for fast synchronization of OFDM sampling frequency with a high number of carriers, comprising the steps of:

S1. Inserting multi-bit pilot symbols at the sending end;

S2. Perform sampling frequency offset pre-correction on the current symbol by using the capture estimated value of the previous symbol;

S3. Obtain the phase information of the corresponding bit pilot after the pre-correction at the receiving end, and calculate the difference with the corresponding bit pilot phase information of the transmitting end to obtain the corresponding bit pilot phase difference;

S4. Estimate the sampling frequency offset of the current symbol by using the pilot phase difference to obtain the capture value of the current symbol, and accumulate the capture value of the current symbol and the capture estimate value of the previous symbol to obtain the capture estimate value of the current symbol;

S5. When the capture value of the current symbol is greater than the set capture threshold, use the capture estimate value of the current symbol to perform sampling frequency offset correction output on the current symbol based on the frequency domain correction method, and update the next symbol to the current symbol to continue to execute S2- S4: Output the pre-correction signal in step S2 when the capture value of the current symbol is smaller than the set capture threshold, and update the next symbol to be the current symbol and continue to execute S2-S4.

In a preferred solution, in step S5, when the capture value of the current symbol is less than the set capture threshold, the next symbol does not perform S2-S4, but performs the tracking process of the sampling frequency offset, specifically as follows:

S6. The next symbol is updated as the current symbol to enter the tracking process, and steps S1-S3 are executed again in the tracking process to obtain the pilot phase difference;

S7. Use the pilot phase difference to estimate the sampling frequency offset tracking value of the current symbol to obtain the tracking value of the current symbol and compare the tracking value of the current symbol with the tracking threshold. When the tracking value of the current symbol exceeds the tracking threshold, the current symbol is The tracking value of the previous symbol is accumulated with the tracking estimated value of the previous symbol as the tracking estimated value of the current symbol, and the current symbol is used to perform frequency domain correction output on the current symbol. When the new tracking value is lower than the tracking threshold, the previous The tracking estimated value of a symbol is used as the tracking estimated value of the current symbol, and the pre-correction signal when step S2 is executed in the output step S6;

S8. The next symbol executes the tracking process.

In a preferred solution, in the step S4, a new sampling frequency offset is estimated by using the pilot phase difference to obtain the capture value of the current symbol, specifically:

Set the pilot inserted at the transmitter Demodulate the corresponding pilot symbols at the receiving end to obtain Then the phase difference of the pilot frequency corresponding to the bit is obtained as:

in Divide the pilot for the corresponding bit:

In formulas (1) and (2), N is the data bit length of the OFDM symbol, Δf represents the normalized sampling frequency offset, L is the length of the cyclic prefix, m represents the mth OFDM symbol, m=1, 2, 3, ..., k represents the kth subcarrier, k=0, 1, 2, 3...N-1, j represents a complex number, X _{m, s} represents the data at the sending end, s has the same meaning as k, and represents the serial number of the carrier number;

The second term in formula (2) is the inter-carrier interference term. When the number of subcarriers is large, the phase value of the inter-carrier interference term approximately obeys a Gaussian distribution with a mean value of 0, so The value is:

According to the values of m and k in formula (3) and the observed To estimate the normalized sampling frequency offset Δf of the system, which is the capture value of the current symbol.

In a preferred solution, the weighted average method with the pilot carrier number k as the weighting factor is also used for data processing when acquiring the current symbol capture value, specifically:

Among them, A _m is the frequency offset estimation constant item, namely:

According to the values of m, k, N, L in formulas (4) and (5) and the observed To estimate the normalized sampling frequency offset Δf of the system, which is the capture value of the current symbol.

In a preferred solution, in step S7, Δf calculated by formula (4) and formula (5) is used as the tracking value of the current symbol.

In a preferred solution, the formula for performing pre-correction in step S2 is:

in is the estimated value of the previous symbol.

6. high carrier number OFDM sampling frequency quick synchronization method according to claim 5, is characterized in that, the tracking threshold in step S7 is set according to prior knowledge:

where Δ is the tracking estimated value, m is the number of symbols, ρ is the tracking threshold coefficient, is the tracking threshold.

In a preferred solution, the pilot signal inserted in the step S1 is 8 bits.

Compared with the prior art, the beneficial effects of the technical solution of the present invention are:

(1) The present invention calculates the pilot phase difference of the corresponding bits of the transmitting end and the receiving end by inserting the frequency domain pilot in the OFDM symbol, and estimates the sampling frequency offset, and the captured estimated value of the previous symbol is used for the current symbol Pre-correct the corresponding bit pilot, add the capture value of the current symbol and the capture estimate value of the previous symbol to update the capture estimate value, and repeat the operation of the current symbol for the pre-correction of the latter symbol until the new capture capture value is greater than up to the set capture threshold. A large number of experiments have proved that generally no more than three symbol estimates are needed to obtain a normalized sampling frequency offset capture value with high precision. After a lot of simulation analysis of NG-DSL system, it is concluded that the approximate relationship between the normalized sampling frequency offset and the number of symbols under the condition of satisfying the NG-DSL output signal-to-noise ratio is mΔf≤10 ^-7 . Therefore, even if the first estimate in the acquisition process is somewhat inaccurate, it will not have a particularly large impact on the system signal-to-noise ratio when the number of symbols is small, and the method of the present invention can reduce the first estimate error It is limited within the range of mΔf≤10 ^-7 , so the present invention can perform real-time high-precision sampling frequency offset correction for each OFDM symbol.

(2) The present invention utilizes an iterative method so that the captured estimated value can quickly converge to near the exact value, meeting the precision requirement of the system. In addition, an optimized data processing scheme is used to reduce the computational complexity of the scheme.

(3) The present invention adds a tracking process after the capture process, and setting the tracking threshold makes the technical solution reach a compromise between complexity and precision, so that the present invention has lower complexity and higher precision, and is suitable for practical use. engineering system.

Description of drawings

FIG. 1 is a flowchart of a specific embodiment of a method for fast synchronization of OFDM sampling frequency with a high number of carriers according to the present invention.

Fig. 2 is a schematic diagram of sampling frequency offset generation in the OFDM system in the present invention.

FIG. 3 is a schematic diagram of the principle of the frequency domain correction method of the present invention.

Fig. 4 is a normalized sampling frequency offset estimation curve of the present invention under different tracking thresholds.

Fig. 5 is a graph comparing the estimated value MSE of the present invention with the estimated value MSE curve of two more accurate traditional estimation methods.

FIG. 6 is a comparison diagram of the estimated velocity and MSE accuracy between the present invention and two more accurate traditional methods.

Fig. 7 is a comparison of constellation diagrams before and after correction of an OFDM symbol in the NG-DSL system using the present invention.

Fig. 8 is a comparison diagram of the original input SNR and the interfered output SNR when an OFDM symbol is interfered by the sampling frequency offset in the NG-DSL system.

Fig. 9 is a comparison diagram of the signal-to-noise ratio of an OFDM symbol before and after the correction of the present invention in the NG-DSL system.

detailed description

The accompanying drawings are for illustrative purposes only and cannot be construed as limiting the patent;

In order to better illustrate this embodiment, some parts in the drawings will be omitted, enlarged or reduced, and do not represent the size of the actual product;

For those skilled in the art, it is understandable that some well-known structures and descriptions thereof may be omitted in the drawings.

In the description of the present invention, it should be understood that the terms "first" and "second" are used for description purposes only, and cannot be interpreted as indicating or implying relative importance or implying the quantity of indicated technical features. Thus, the defined "first" and "second" features may explicitly or implicitly include one or more of these features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

In the description of the present invention, it should be noted that unless otherwise specified and limited, the terms "installation" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral Ground connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediary. It can be said that the internal communication of two components. Those of ordinary skill in the art can understand the specific meanings of the above terms in the present invention in specific situations.

The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

Example 1

As shown in FIG. 1 , it is a flowchart of a specific embodiment of a method for fast synchronization of OFDM sampling frequency with a high number of carriers according to the present invention. Referring to Fig. 1, a kind of high-carrier number OFDM sampling frequency quick synchronization method of the present specific embodiment specifically comprises the following steps:

S101. Insert multi-bit pilot symbols at the sending end; generally, insert multi-bit pilot symbols at the sending end for estimation of sampling frequency offset according to channel estimation feedback on channel state, preferably 8-bit pilot symbols. Insert pilot at the sender Then the corresponding pilot symbols can be demodulated at the receiving end: Wherein, m represents the mth OFDM symbol, m=1, 2, 3,..., k represents the kth subcarrier, k=0, 1, 2, 3...N-1;

S102. Perform sampling frequency offset pre-correction on the current symbol using the capture estimate value of the previous symbol; specifically, for the current symbol, first use the capture estimate value of the previous symbol to perform a pre-correction process on the corresponding bit pilot, and the pre-correction formula is :

among them capture estimate for the previous symbol.

S103. Obtain the phase information of the corresponding bit pilot after pre-correction at the receiving end, and calculate the difference with the corresponding bit pilot phase information of the sending end to obtain the corresponding bit pilot phase difference; specifically, according to the pilot symbol inserted at the sending end and The pilot of the corresponding bit at the receiving end can obtain the phase difference of the pilot of the corresponding bit as:

in Divide the pilot for the corresponding bit:

In formulas (1) and (2), N is the OFDM symbol data bit length, Δf represents the normalized sampling frequency offset, as shown in Figure 2; L is the cyclic prefix length, m represents the mth OFDM symbol, m= 1, 2, 3, ..., k represents the kth subcarrier, k=0, 1, 2, 3...N-1, j represents a complex number, X _{m, s} represents the data at the sending end, s and k have the same meaning, The serial number representing the number of carriers;

The second term in formula (2) is the inter-carrier interference term. When the number of sub-carriers is large, the phase value of the inter-carrier interference term approximately obeys a Gaussian distribution with a mean value of 0. Therefore, when formula (2) is used to calculate the angle, that is, formula (1), only the imaginary digit can get the phase angle at this time, that is, the phase angle in formula (2) item and term, since the second term in the formula (2), that is, the phase of the cumulative term obeys the normal distribution with a mean value of 0, so the angle is taken as that of the previous term Therefore it follows The value is:

S104. Use the pilot phase difference to estimate the acquisition value of the new sampling frequency offset to obtain the acquisition value of the current symbol, and accumulate the acquisition value of the current symbol and the acquisition estimation value of the previous symbol to obtain the acquisition estimation value of the current symbol;

Formula (3) is the number of symbols m, the number of carriers k, the normalized sampling frequency offset Δf and the phase difference The relationship between, due to the use of pilot frequency, so when the values of m and k in the formula (3) are known, the observed To estimate the normalized sampling frequency offset Δf of the system, which is the capture value of the current symbol.

Since the present invention will be used in the NG-DSL system with high input signal-to-noise ratio, the channel noise is less, so the weighted average method with the pilot carrier number k as the weighting factor is adopted to carry out data processing, specifically:

Among them, A _m is the frequency offset estimation constant item, namely:

According to the values of m, k, N, L in formulas (4) and (5) and the observed To estimate the normalized sampling frequency offset Δf of the system, which is the capture value of the current symbol.

In practical applications, the equations (4) and (5) can be used to perform fitting estimation on one OFDM symbol by using 8 pilot symbols to complete the estimation of the first symbol. Since the first symbol does not have its previous symbol, Therefore, the capture value of the first symbol is also its estimated value, so as to obtain the estimated value Δf ₁ of the first symbol, but there is often a certain deviation in the estimated value of the first symbol, so it is necessary to continue to correct the estimated value. When the first coincidence becomes the previous symbol, the second symbol becomes the current symbol;

For the second symbol as the current symbol, first use the capture estimation value of the first symbol to perform a pre-correction process on the corresponding bit pilot, the pre-correction formula is the pre-correction formula in S102, and the pre-correction for the second symbol is obtained Pilot information R′ _m,k , repeat the estimation process of the first symbol with the pilot information R′ _m,k obtained after pre-calibration, and obtain a new normalized sampling frequency offset capture value Δf′ ₂ , that is, use The pilot information obtains the pilot phase difference in S103, and in S104, according to the new pilot phase difference combined with formulas (4) and (5), a new sampling frequency offset capture value estimation can be performed, and a new normalized The normalized sampling frequency offset capture value Δf′ ₂ is the capture value of the second symbol. It can be seen that the second symbol can be obtained by repeating the estimation process of the first symbol after obtaining the pilot information R′ _m,k in step S102 An estimate of the symbol's sampling frequency offset capture value. In step S104, after the estimation of the second symbol is completed, use the estimated value Δf ₁ of the first symbol and the capture value Δf′ ₂ of the second symbol to update the estimated value, that is, to obtain the Capture estimates:

Δf=Δf ₁ +Δf′ ₂ (7).

S105. When the capture value of the current symbol is greater than the set capture threshold, use the capture estimation value of the current symbol to perform sampling frequency offset correction output on the current symbol based on the frequency domain correction method, and update the next symbol to the current symbol to continue to execute S102- S104: Output the pre-correction signal in step S102 when the capture value of the current symbol is smaller than the set capture threshold, and update the next symbol to be the current symbol and continue to execute S102-S104.

In the above steps, the estimated capture value Δf′ ₂ of the current symbol is much smaller than the estimated value Δf ₁ of the previous symbol, which is a correction to the correct value of the captured estimated value of the previous symbol, and the latter symbol repeats the capture value of the current symbol. Action to capture further updates to estimates. A large number of experiments have proved that generally no more than three symbol estimates are needed to obtain a relatively accurate normalized sampling frequency offset capture value. Generally, the estimation process of the first three symbols is called the acquisition process of the normalized sampling frequency offset. After a lot of simulation analysis of NG-DSL system, the approximate relationship between the normalized sampling frequency offset and the number of symbols is obtained under the condition of satisfying the NG-DSL output signal-to-noise ratio.

mΔf≤10 ^-7 (8)

Therefore, even if the first estimate of the acquisition process is somewhat inaccurate, it will not have a particularly large impact on the system signal-to-noise ratio when the number of symbols is small, and the formula (3) in the present invention can be used The first estimation error is limited within the range of formula (8), so the present invention can perform real-time sampling frequency offset correction for each OFDM symbol.

After the capture process is completed, in order to further improve the accuracy and reduce the complexity of calculation, the present invention introduces a tracking process, that is, in step S105, when the capture value of the current symbol is less than the set capture threshold, the pre-correction signal of step S102 is processed For the output of the current symbol, the steps S102-S104 are not performed for the next symbol at this time, but the tracking process of the sampling frequency offset is carried out, as follows:

S106. The next symbol is updated as the current symbol and enters the tracking process. During the tracking process, a multi-bit pilot symbol is inserted at the sending end according to the feedback of the channel state for the estimation of the sampling frequency offset, and the sample estimated by the previous symbol is used. Frequency offset performs sampling frequency offset pre-correction on the current symbol, obtains the phase information of the corresponding bit pilot after pre-correction at the receiving end, and calculates the difference with the corresponding bit pilot phase information at the sending end to obtain the corresponding bit pilot phase difference; here also It is to execute steps S101-S103 again to obtain the pilot phase difference;

S107. Use the pilot phase difference to estimate the sampling frequency offset tracking value of the current symbol to obtain the tracking value of the current symbol, and compare the tracking value of the current symbol with the tracking threshold. When the tracking estimated value of the current symbol exceeds the tracking threshold, set The tracking estimation value of the current symbol and the tracking estimation value of the previous symbol are accumulated as the tracking estimation value of the current symbol, and the frequency domain correction output is performed on the current symbol by using the tracking estimation value of the current symbol, when the new tracking value is lower than the tracking threshold When, the tracking estimated value of the previous symbol is used as the tracking estimated value of the current symbol, and the pre-correction signal when step S102 is executed in the output step S106;

In the specific implementation process, the acquisition method of the tracking value of the current symbol in step S107 is the same as the acquisition method of the acquisition value of the current symbol, that is, formula (4) and formula (5) are also used for calculation, and the difference is that Δf obtained at this time Not as the capture value of the current symbol, but as the tracking value of the current symbol in step S107.

The tracking process will judge the tracking value of the current symbol. When the tracking value of the current symbol exceeds the tracking threshold, the tracking estimated value of the current symbol and the tracking estimated value of the previous symbol are accumulated, and the accumulated value is used to update the tracking estimated value and use The updated tracking estimated value is corrected in the frequency domain. When the new tracking estimated value is lower than the tracking threshold, no update operation is performed on the estimated value and no new frequency domain correction is required. It is only necessary to use the predicted value after performing step S102 again. Correcting the frequency domain signal as the output signal can meet the requirements of the system.

The present invention enters the tracking process after the capture process is completed, and each symbol in the tracking process will also repeat the estimation and update process of the capture process, but the difference is that in the tracking process, it will judge whether to perform regression according to the size of the tracking feedback value. The update of the normalized sampling frequency offset estimate. There are two main tasks in the tracking process: the first is to use the new estimated value to fine-tune the previously estimated deviation; the second is to detect whether there is a new larger deviation, if it occurs, it will be recaptured process. Since the estimated sampling frequency offset tracking value in the tracking process is often small, it will affect the running speed of the system if a small value is adjusted each time, and this small change does not significantly improve the signal-to-noise ratio, so in practice In engineering, we can ignore estimates that are less than a certain threshold. Therefore, in order to reduce the computational complexity of the system, according to the analysis of the synchronization parameters of the NG-DSL system in formula (8), a tracking threshold is set:

where Δ is the tracking estimated value, m is the number of symbols, ρ is the tracking threshold coefficient, is the tracking threshold.

In the tracking process, when the tracking value exceeds the tracking threshold, the frequency domain symbols of the system are corrected with new estimated values; when the tracking value is lower than the tracking threshold, the frequency domain signal of the current symbol does not need to be corrected with new estimated values, Just use the results of the pre-correction stage. The tracking threshold coefficient in formula (9) is determined jointly by precision and complexity, that is, when ρ takes a larger value, the estimation accuracy and complexity of the present invention are higher; when ρ takes a smaller value, the estimation of the present invention Less precision and less complexity. In order to obtain a sampling frequency offset estimation method with lower complexity and higher precision, in practical applications, according to the actual situation, a compromise is made between the estimation accuracy and complexity, so as to achieve the optimal parameter selection suitable for the actual system.

After the estimation of each symbol is completed, it is necessary to use the frequency domain correction method to correct the output frequency domain signal using the estimated value in time. The working principle diagram of the frequency domain correction is shown in FIG. 3 .

The present invention has carried out performance analysis and simulation to above-mentioned method, specifically as follows:

In an NG-DSL system with a line length of 50m, the length of each OFDM data bit is 8192 (FFT/IFFT length), the normalized sampling frequency offset is 3ppm, the sequence number of the pilot carrier is: {505507509511513515517519}, the tracking threshold The coefficients are simulations of new sampling frequency synchronization under three different situations of 4, 8, and 20 respectively. And compared with two traditional methods, the first method is the sampling frequency synchronization method proposed in the document "Estimation and compensation of frequency offset in DAC/ADC clocks in OFDM systems", the second method is the document "OFDM under high-order QAM modulation The sampling frequency synchronization method proposed in "Joint Correction of Carrier and Sampling Frequency Offset".

As shown in FIG. 4 , it is the estimated value curve of the sampling frequency offset under different tracking thresholds simulated by the present invention. can be found in the figure The curve is an estimated curve when the tracking threshold ρ is 4. It can be found that there are many flat estimated values at this time, indicating that in the case of this threshold coefficient value, the present invention does not need to correct too many estimated values. It is enough to estimate and correct the frequency domain signal when several estimated values have large deviations, but at the same time it can also be found that the deviation of the estimated value is the largest in this case, which is equivalent to using the loss of precision in exchange for complex degree of reduction. in the picture The curve is the estimated curve when the tracking threshold ρ is 8, it can be found that the curve at this time does not The curve is so flat, but the accuracy is lower than The curve is higher and is a compromise situation. in the picture The curve is an estimated curve when the tracking threshold ρ is 20. This curve basically needs to be re-estimated and corrected for each symbol. In this case, the estimation accuracy is the highest, but the complexity is also the highest. Therefore, through the above analysis, the tracking threshold coefficient value can be set to 8, which can achieve a compromise between accuracy and complexity.

As shown in Figure 5, it is the MSE curve of the method of the present invention and the two relatively accurate methods introduced above. The MSE curve is an important criterion for judging whether the estimation is accurate. Generally, MSE is defined as:

From Figure 5, it can be found that the MSE value of the method of the present invention is very low, and the tracking estimation value will become more accurate with the increase of the input signal-to-noise ratio, and there is a very accurate estimation in the range of 40dB to 60dB, which can meet the accuracy of the NG-DSL system Require. And the estimation speed of the present invention is also very fast, that is, the most accurate estimation is performed with the least number of symbols, which is also the effect of the iterative calculation method used in the present invention. Fig. 6 is the normalized sampling frequency offset estimation value obtained by different estimated symbol numbers, so as to reflect the estimated speed characteristic of the synchronization invention.

It can be seen from FIG. 6 that the present invention can quickly converge the tracking estimated value to near the exact value, and the MSE obtained by estimating the number of symbols is also smaller than that of the above two methods. Since the simulation environment has a sampling deviation of 0.3ppm, the above figure shows that the MSE is on the order of 1e-17 to 1e-18, so it can be calculated that the jitter value is on the order of 1e-9. In the actual frequency offset estimation, we think this order of magnitude The deviation meets the accuracy requirements of the system for sampling frequency deviation. It shows that the present invention can use fewer estimation symbols to achieve higher estimation accuracy.

When the estimation is completed, it is necessary to use the estimated value to correct the signal in the frequency domain. Using the correction formula of Equation (6) and performing correction according to the frequency domain correction schematic diagram in Figure 3, the corrected frequency domain constellation diagram can be obtained, and the correction The before and after constellation diagrams are compared to judge the correction effect, as shown in Figure 7. It can be found that the uncorrected constellation diagram is very chaotic, the system cannot be demodulated correctly, and a large bit error will be generated.

Fig. 8 shows the original input signal-to-noise ratio and the interfered output signal-to-noise ratio when an OFDM symbol is interfered by sampling frequency offset in the NG-DSL system. It can be seen that due to the sampling frequency offset, the signal-to-noise ratio of the signal is greatly reduced, which seriously affects the transmission of the signal. Fig. 9 is a comparison diagram of the signal-to-noise ratio of an OFDM symbol before and after correction by the present invention in the NG-DSL system. It can be seen that after being corrected by the algorithm of the present invention, the signal-to-noise ratio of the signal has been greatly improved. Comparing Figure 8 and Figure 9, it can be seen that the corrected SNR is basically the same as the input SNR in the low frequency band, and is about 2dB worse than the input SNR in the high frequency band. This is because the frequency domain correction method is difficult to completely eliminate ICI. However, in the NG-DSL system, the adjustment level of the high frequency band is relatively low, and the requirement for the signal-to-noise ratio is not as strict as that of the low frequency band, so the demodulation without error codes can be performed in the simulation.

The same or similar reference numerals correspond to the same or similar components;

The positional relationship described in the drawings is only for illustrative purposes and cannot be construed as a limitation to this patent;

Apparently, the above-mentioned embodiments of the present invention are only examples for clearly illustrating the present invention, rather than limiting the implementation of the present invention. For those of ordinary skill in the art, on the basis of the above description, other changes or changes in different forms can also be made. It is not necessary and impossible to exhaustively list all the implementation manners here. All modifications, equivalent replacements and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the claims of the present invention.

Claims (8)

1. a high carrier number OFDM sampling frequency fast synchronization method, is characterized in that, comprises the steps: S1. Inserting multi-bit pilot symbols at the sending end; S2. Perform sampling frequency offset pre-correction on the current symbol by using the capture estimated value of the previous symbol; S3. Obtain the phase information of the corresponding bit pilot after the pre-correction at the receiving end, and calculate the difference with the corresponding bit pilot phase information of the transmitting end to obtain the corresponding bit pilot phase difference; S4. Estimate the sampling frequency offset of the current symbol by using the pilot phase difference to obtain the capture value of the current symbol, and accumulate the capture value of the current symbol and the capture estimate value of the previous symbol to obtain the capture estimate value of the current symbol; S5. When the capture value of the current symbol is greater than the set capture threshold, use the capture estimate value of the current symbol to perform sampling frequency offset correction output on the current symbol based on the frequency domain correction method, and update the next symbol to the current symbol to continue to execute S2- S4: Output the pre-correction signal in step S2 when the capture value of the current symbol is smaller than the set capture threshold, and update the next symbol to be the current symbol and continue to execute S2-S4. 2. the high-carrier number OFDM sampling frequency fast synchronization method according to claim 1 is characterized in that, in step S5, when the capture value of current symbol is less than the capture threshold value of setting, next symbol does not carry out S2-S4, and carries out The tracking process of the sampling frequency offset is as follows: S6. The next symbol is updated as the current symbol to enter the tracking process, and steps S1-S3 are executed again in the tracking process to obtain the pilot phase difference; S7. Use the pilot phase difference to estimate the sampling frequency offset tracking value of the current symbol to obtain the tracking value of the current symbol and compare the tracking value of the current symbol with the tracking threshold. When the tracking value of the current symbol exceeds the tracking threshold, the current symbol is The tracking value of the previous symbol is accumulated with the tracking estimated value of the previous symbol as the tracking estimated value of the current symbol, and the current symbol is used to perform frequency domain correction output on the current symbol. When the new tracking value is lower than the tracking threshold, the previous The tracking estimated value of a symbol is used as the tracking estimated value of the current symbol, and the pre-correction signal when step S2 is executed in the output step S6; S8. The next symbol executes the tracking process. 3. the high carrier number OFDM sampling frequency fast synchronization method according to claim 2, is characterized in that, utilizes pilot frequency phase difference to carry out new sampling frequency offset estimation and obtains the capture value of current symbol in the described step S4, is specially: Set the pilot inserted at the transmitter Demodulate the corresponding pilot symbols at the receiving end to obtain Then the phase difference of the pilot frequency corresponding to the bit is obtained as: in Divide the pilot for the corresponding bit: $\begin{array}{c}\frac{{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}}{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}\frac{\mathrm{<span\; class="notranslate">\; sin</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&pi;k\&Delta;f</span>}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; sin</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&pi;k\&Delta;f</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; )</span>}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j\&pi;k\&Delta;f</span>}\frac{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; j</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;k\&Delta;f</span>}\frac{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; L</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; )</span>\\ <span\; class="notranslate">\; +</span>\underset{\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; k</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\underset{\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}}\frac{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}}}{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}}\frac{\mathrm{<span\; class="notranslate">\; sin</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; s\&Delta;f</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ]</span>}{\mathrm{<span\; class="notranslate">\; sin</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; s\&Delta;f</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; ]</span>}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; j\&pi;</span>}<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; s\&Delta;f</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ]</span>\\ \mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; j</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; s\&Delta;f</span>}<span\; class="notranslate">\; )</span>\frac{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; L</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}^{<span\; class="notranslate">\; \&prime;</span>}\end{array}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$ In formulas (1) and (2), N is the data bit length of the OFDM symbol, Δf represents the normalized sampling frequency offset, L is the length of the cyclic prefix, m represents the mth OFDM symbol, m=1, 2, 3, ..., k represents the kth subcarrier, k=0, 1, 2, 3...N-1, j represents a complex number, X _{m, s} represents the data at the sending end, s has the same meaning as k, and represents the serial number of the carrier number; The second term in formula (2) is the inter-carrier interference term. When the number of subcarriers is large, the phase value of the inter-carrier interference term approximately obeys a Gaussian distribution with a mean value of 0, so The value is: According to the values of m and k in formula (3) and the observed To estimate the normalized sampling frequency offset Δf of the system, which is the capture value of the current symbol. 4. the high carrier number OFDM sampling frequency fast synchronization method according to claim 3, is characterized in that, also adopts to carry out data processing with the weighted average method of weighting factor with pilot frequency carrier sequence number k when carrying out current symbol acquisition value ,Specifically: Among them, A _m is the frequency offset estimation constant item, namely: ${\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&pi;</span>}\frac{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}\frac{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; L</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$ According to the values of m, k, N, L in formulas (4) and (5) and the observed To estimate the normalized sampling frequency offset Δf of the system, which is the capture value of the current symbol. 5. the high-carrier number OFDM sampling frequency fast synchronization method according to claim 4, is characterized in that, the Δf that adopts formula (4) and formula (5) to calculate obtains as the tracking value of current symbol in step S7. 6. the high carrier number OFDM sampling frequency fast synchronization method according to claim 5, is characterized in that, the formula that described step S2 carries out pre-correction is: ${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; \{</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; j\&pi;k</span>}\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&Delta;f</span>}}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}\frac{\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; N</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; \}</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ;</span>$ in is the estimated value of the previous symbol. 7. high carrier number OFDM sampling frequency quick synchronization method according to claim 6, is characterized in that, the tracking threshold in the step S7 is set according to prior knowledge: $\mathrm{<span\; class="notranslate">\; \&Delta;</span>}<span\; class="notranslate">\; \&le;</span>\frac{{\mathrm{<span\; class="notranslate">\; 10</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 7</span>}}}{\mathrm{<span\; class="notranslate">\; m\&rho;</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 9</span>}<span\; class="notranslate">\; )</span>$ where Δ is the tracking estimated value, m is the number of symbols, ρ is the tracking threshold coefficient, is the tracking threshold. 8. The high-carrier number OFDM sampling frequency fast synchronization method according to any one of claims 1-7, characterized in that the pilot signal inserted in the step S1 is 8 bits.