A method of receiving signal synchronization

Abstract

A signal synchronization method includes sequence interpolation correlation synchronization, detecting the position of the maximum peak and the maximum value R xy (max); selecting correlation sample values on both sides of the maximum peak, linearly fitting the points on both sides, and the intersection of the two fitted straight lines as the predicted optimal synchronization point; selecting points on both sides of the predicted position as target synchronization position candidate points, cutting the target signal for channel estimation, calculating the channel equivalent SNR of different target positions; selecting the position with the maximum SNR as the final synchronization position; the above signal synchronization method is based on the shape of the correlation curve to predict the synchronization position, the sequence interpolation correlation synchronization method finds the approximate position of the peak, the left and right data of the correlation curve are linearly fitted, the intersection of the two fitted straight lines is the predicted optimal synchronization position, and then within a small range on both sides of the predicted position, the best synchronization point is further selected based on the quality of the channel response, which can improve the synchronization effect by 3-4 times.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of communication, in particular to a signal receiving synchronization method. BACKGROUND

[0002] There are two main ideas for the existing high-precision sequence synchronization scheme: one is to realize synchronization through sequence cross-correlation, and the other is to realize synchronization through estimation of frequency domain phase detection.

[0003] Among them, sequence cross-correlation synchronization:

[0004] The original sending sequence is x(t), after a system, the signal is changed through a certain amplitude G and time delay τ, and during the transmission process, a certain interference noise n is introduced, resulting in signal distortion.

[0005] y(t) = G * x(t - τ) + n(t) (1)

[0006] The existing sequence correlation algorithm is generally implemented in two steps: first, the correlation between a segment of length (2N+1) of the received sequence y(t) and the original sequence x(t) is taken out, based on the correlation peak, the coarse synchronization position is located;

[0007]

[0008] Second, based on the initial synchronization position, a part of the local sequence is taken out from the received sequence and the original sequence near the coarse synchronization position, and the sequence cross-correlation is performed again based on the up-sampled sequence, and the fine synchronization position is located based on the correlation peak. Different devices may have different up-sampling schemes, which can be directly implemented based on high-performance ADC (Analog-to-digital converter), or can be implemented through digital signal processing interpolation technology, but the purpose is the same, that is, to increase the digital sampling rate of the signal.

[0009] Another kind of synchronization based on frequency domain phase:

[0010] From the above description, the relationship between the received time domain signal and the sending time domain signal is:

[0011] y(t) = G * x(t - τ) + n(t) (3)

[0012] Taking Fourier transform on both sides, we have

[0013] Y(ω) = G'X(ω)e -jωτ +N(ω) (4)

[0014] From equation (4), if the influence of the noise term N(ω) is ignored, the phase of the ratio of Y(ω) and X(ω) is

[0015]

[0016] The ratio of the time delay of the received signal to the phase of the signal at different frequency points has a fixed relationship

[0017]

[0018] For each frequency point, the relationship between the phase and the frequency can be calculated based on formula (5), and then linear fitting is performed to eliminate the influence of noise, and the slope of the fitting straight line is calculated, and the relative time delay τ between the sequences can be calculated from formula (6).

[0019] The synchronization is realized by the sequence cross-correlation synchronization based on the frequency domain phase, and the calculation amount is small without upsampling processing, but the precision is not as high as that of the sequence cross-correlation synchronization.

[0020] The above two schemes can realize synchronization with certain precision, and meet the needs of most communication scenarios; but for devices with very high synchronization performance requirements, such as communication test instruments, the instruments themselves do not want to introduce additional interference, and theoretically expect to achieve ideal synchronization precision, or the interference signal introduced by synchronization is one or two orders of magnitude lower than the interference introduced during normal communication. Obviously, at this time, the conventional communication synchronization scheme cannot meet the needs of high-precision test scenarios, resulting in limited instrument test result precision and limited application scenarios. SUMMARY

[0021] Therefore, it is necessary to provide a received signal synchronization method with improved precision.

[0022] A received signal synchronization method comprises:

[0023] Obtaining the peak position: based on the sequence interpolation correlation synchronization, detecting the position of the maximum peak and the maximum value R xy (max);

[0024] Selecting sample values: selecting correlation sample values on the correlation curves on the left and right sides of the maximum peak,

[0025] Synchronization position prediction: based on the selected correlation sample values, linear fitting is performed on the left points and linear fitting is performed on the right points, and the intersection point of the two fitting straight lines is taken as the predicted best synchronization point;

[0026] Channel estimation: selecting points on the left and right sides of the predicted position as target synchronization position candidate points, intercepting the target signal for channel estimation, and calculating the channel equivalent SNR of different target positions;

[0027] Signal quality evaluation: selecting the position with the maximum SNR as the final synchronization position.

[0028] In the preferred embodiment, the selection of the sample value in the step of selecting the sample value is based on a threshold: R Thresh = aR xy (max) 0 < a < 1

[0029] For the left correlation sequence, the point of the correlation sample value R xy (t) >= R Thresh is selected;

[0030] For the right correlation sequence, the point of the correlation sample value R xy (t) >= R Thresh is also selected;

[0031] x represents the transmitted reference sequence, y represents the received reference sequence, t represents a specific time point, and R represents the correlation of the transmitted sequence and the received sequence.

[0032] In the preferred embodiment, the channel estimation step further comprises: the channel estimation is based on the time domain implementation: assuming that the transmitted synchronization reference sequence is x(t) and the received sequence is y(t), then

[0033] y(t) = A x(t - τ) + n(t)

[0034] The received signal is the original signal after time delay and amplitude change, and is contaminated by noise n(t) in the transmission process;

[0035] h(t) = y(t) / x(t)

[0036] h(t) has strong correlation in the time domain, while the noise has randomness, and the mean value of the channel response is calculated by sliding average;

[0037]

[0038] The noise is the difference between the channel response and the smoothing result;

[0039]

[0040] The estimated signal-to-noise ratio based on the current synchronization obtained channel can be obtained:

[0041]

[0042] In the preferred embodiment, the synchronization position prediction further comprises: predicting the best synchronization position based on the shape of the correlation waveform.

[0043] In the preferred embodiment, in the channel estimation step, the points selected as the target synchronization position candidate points on the left and right sides of the predicted position are found by traversing the vicinity of the predicted position to find the position with the minimum inter-symbol interference.

[0044] In the preferred embodiment, the channel estimation step includes selecting a synchronization range, and then traversing the positions to select the best position based on the maximum SNR.

[0045] In the preferred embodiment, the selecting sample value step selects the correlation sample value on the left or right side of the maximum peak value greater than 2.

[0046] In the preferred embodiment, the obtaining peak position further includes:

[0047] Coarse synchronization: after the synchronization signal sequence passes through the communication system, coarse synchronization is performed at the receiving side to find the approximate position of the synchronization signal sequence;

[0048] Intercept sequence: a section of the received data is intercepted at the coarse synchronization sequence position as the received sequence for the next signal processing;

[0049] Sampling interpolation: the intercepted received sequence is up-sampled by a high multiple, the local synchronization sequence is up-sampled by the same multiple, and then the cross-correlation of the interpolated sequence is performed to obtain the correlation curve;

[0050] Detecting peak value: the accurate position of the maximum peak value and the maximum R xy (max) are detected according to the correlation curve.

[0051] In the preferred embodiment, the high multiple sampling interpolation in the sampling interpolation step is a high multiple sampling interpolation of more than 8 times.

[0052] The above-mentioned received signal synchronization method is based on a sequence interpolation correlation synchronization method to find the approximate position of the peak value, and the symmetry of the correlation curve is used to linearly fit the data on the left and right sides of the correlation curve respectively. The intersection point of the two fitted straight lines is the predicted best synchronization position, and then based on the quality of the channel response, the best synchronization point is further selected within a small range on the left and right of the predicted position. The best synchronization position should be near the symmetry axis by using the local symmetry of the correlation sequence in shape, and the synchronization accuracy of nearly 100 times relative to the baseband sequence correlation is obtained. Based on the multiple point linear fitting, it is equivalent to least squares processing for large samples, which can suppress the influence of noise and make the scheme applicable to low SNR scenarios. The ultimate goal of communication is to obtain the target signal with the highest SNR, and finally the best synchronization position is selected based on the SNR criterion to obtain the optimal system performance. The received signal synchronization method of the present application is based on the existing signal processing flow, and adds two schemes of morphological prediction and optimal point screening. The accuracy is improved by 3-4 times compared with the traditional sequence interpolation correlation peak detection synchronization scheme based on the prediction result. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1Flow chart of the receiving signal synchronization method of the present application;

[0054] Figure 2 Schematic diagram of the synchronization position prediction of the present application;

[0055] Figure 3 Flow chart of the application of the receiving information synchronization method of the present application in an instrument system;

[0056] Figure 4 Flow chart of the application of the receiving information synchronization method of the present application in a DPD system. DETAILED DESCRIPTION

[0057] The following examples are intended to better illustrate the present application, but are not intended to limit the present application.

[0058] As Figures 1 to 2 shown, a receiving signal synchronization method of the present application comprises:

[0059] Acquiring peak position: based on sequence interpolation correlation synchronization, detecting the position of the maximum peak and the maximum value R xy (max);

[0060] Selecting sample value: selecting correlation sample values on the correlation curve on the left and right sides of the maximum peak,

[0061] Synchronization position prediction: based on the selected correlation sample values, linear fitting is performed on the left points and linear fitting is performed on the right points, and the intersection point of the two fitting straight lines is taken as the predicted optimal synchronization point;

[0062] Channel estimation: selecting points on the left and right sides of the predicted position as target synchronization position candidate points, intercepting the target signal for channel estimation, and calculating the channel equivalent SNR of different target positions;

[0063] Signal quality evaluation: selecting the position with the maximum SNR (Signal to Noise Ratio) as the final synchronization position.

[0064] Further, in the step of selecting sample values of the present embodiment, the selection of the correlation sample values is based on threshold selection: R Thresh =αR xy (max) 0 < α < 1;

[0065] For the left correlation sequence, the points with correlation sample values R xy (t) >= R Thresh (max) are selected;

[0066] For the right correlation sequence, the points with correlation sample values R xy (t) >= R Thresh (max) are also selected;

[0067] x: represents the transmitted reference sequence, y represents the received reference sequence, t represents a certain specific time point, and R represents the correlation of the transmitted sequence and the received sequence.

[0068] Further, the synchronization position prediction of the embodiment further comprises predicting the optimal synchronization position based on the shape of the correlation waveform.

[0069] Further, the channel estimation step of the embodiment further comprises that the channel estimation is based on time domain implementation: assuming that the transmitted synchronization reference sequence is x(t) and the received sequence is y(t), then

[0070] y(t) = A x(t-τ) + n(t)

[0071] The received signal is the signal after the original signal is subjected to time delay and amplitude change, and is contaminated by noise n(t) in the transmission process;

[0072] h(t) = y(t) / x(t)

[0073] h(t) has strong correlation in time domain, while noise has randomness, and the mean value of the channel response is calculated by sliding average;

[0074]

[0075] The noise is the difference between the channel response and the smoothing result;

[0076]

[0077] The estimated signal-to-noise ratio based on the current synchronization obtained channel can be obtained:

[0078]

[0079] Further, in the channel estimation step of the embodiment, the points selected as the target synchronization position candidate points on the left and right sides of the predicted position are traversed in the vicinity of the predicted position to find the position with the minimum inter-symbol interference.

[0080] Further, the channel estimation step of the embodiment comprises selecting a synchronization range, then traversing these positions, and selecting the optimal position based on the maximum SNR.

[0081] Further, in the selecting sample value step of the embodiment, the correlation sample values selected on the correlation curves on the left or right sides of the maximum peak value are greater than 2.

[0082] Further, the obtaining peak position of the embodiment further comprises:

[0083] Coarse synchronization: after the synchronization signal sequence passes through the communication system, coarse synchronization is performed on the receiving side to find the approximate position of the synchronization signal sequence;

[0084] Intercept sequence: intercept a piece of received data at the position of the coarse synchronization sequence as the received sequence for the next signal processing;

[0085] Sample interpolation: do high multiple up-sampling interpolation on the received intercepted sequence, do the same multiple up-sampling on the local synchronization sequence, and then do cross-correlation on the interpolated sequence to obtain a correlation curve;

[0086] Detect peak value: detect the accurate position of the maximum peak value and the maximum value R xy (max) from the correlation curve.

[0087] Further, the high multiple up-sampling in the sample interpolation step of the embodiment is high multiple up-sampling more than 8 times.

[0088] The received signal synchronization method of the application predicts the synchronization position based on the shape of the correlation curve, finds the approximate position of the peak value based on the sequence interpolation correlation synchronization method, uses the symmetry of the correlation curve to do linear fitting on the data on the left and right sides of the correlation curve respectively, the intersection point of the two fitted straight lines is the predicted best synchronization position, and then based on the quality of the channel response, the best synchronization point is further selected within a small range on the left and right of the predicted position. The application does not need additional hardware support, and based on the existing signal processing framework, further signal processing is done through software, and the synchronization effect of 3-4 times higher than the current synchronization scheme can be obtained.

[0089] As shown in Figure 3 the current mainstream single carrier communication systems such as WCDMA, GSM, BlueTooth, etc. are based on the transmitter shaping filter and the receiver matching filter to form a filter bank that meets the Nyquist sampling, and the receiver extracts the signal by finding the optimal sampling position to reduce the inter-symbol interference, the accuracy of the optimal synchronization position determines the size of the inter-symbol interference, and also directly determines the demodulation performance of the final system. Using the received signal synchronization method of the application, the equivalent SNR can be improved from 36-37 dB to more than 50 dB compared with the existing synchronization scheme, which can meet the application requirements of high-precision test instruments and other scenarios.

[0090] The application mainly targets single carrier scenarios, and multi-carrier communication in time domain scenarios can also be used, such as DPD algorithm, delay alignment of received sequence and transmitted sequence, and the smaller the alignment error is, the higher the precision of the pre-distortion algorithm is.

[0091] As shown in Figure 4As shown in the wideband communication system using OFDM (Orthogonal Frequency Division Multiplexing) technology, DPD (Digital Pre-Distortion) is a very important technology, which can effectively improve the efficiency of the power amplifier and bring great social benefits. The performance of the DPD technology is determined by the accuracy of the synchronization between the feedback sequence and the transmission sequence. In order to accurately know the deformation of the transmission sequence after passing through the FEM (Front-end Modules) system, the DPD signal processing can be used to estimate and compensate the deformation. The received signal synchronization method can further improve the performance of the DPD system. In the DPD system, in order to improve the real-time tracking ability of the signal, the morphological prediction result can be directly used to reduce the complexity of the whole system.

[0092] The receiving signal synchronization method of the present application uses the local symmetry of the correlation sequence, and the best synchronization position should be near the symmetry axis, so that the synchronization accuracy of the relative baseband sequence correlation can reach nearly 100 times. Based on the multi-point linear fitting, the least square processing is performed on the large sample, so that the influence of noise can be suppressed, and the scheme is suitable for low SNR scene. The final purpose of communication is to obtain the target signal with the highest SNR, and the best synchronization position is selected based on the SNR criterion, so that the optimal system performance can be obtained. Based on the receiving signal synchronization method of the present application, any additional hardware support is not needed, and the software signal processing is completely realized, so that the existing system can be upgraded through software, and the method has good popularization. The receiving signal synchronization method of the present application is based on the existing signal processing flow, and adds two schemes of morphological prediction and optimal point screening. The two schemes can be used in combination or separately, and the accuracy has been improved by 3-4 times based on the prediction result relative to the traditional sequence interpolation correlation peak detection synchronization scheme. Combined with the target signal quality detection scheme, the optimal performance of the system can be further obtained.

[0093] The sequence correlation synchronization scheme performance is further improved by the receiving signal synchronization method, and the correlation accuracy is improved by more than 100 times statistically relative to the baseband sampling rate, and better synchronization accuracy can still be achieved in a low SNR scene. Based on the sequence interpolation correlation peak search synchronization, the receiving signal synchronization method is based on the shape of the correlation curve to predict the peak value. Compared with the maximum value of the correlation peak, the accuracy is higher (solving the problem of the correlation peak appearing platform, and better resisting the influence of noise), and the optimal position is further found based on the SNR decision. In actual communication systems and instruments, the ultimate goal is to obtain the highest channel SNR. Therefore, in the present application, if the interpolation is 128 times, the point with the maximum SNR of the 128 interpolation positions is obtained, which should be the best communication synchronization position after the original data is increased by 128 times in sampling rate. Obviously, the maximum correlation peak cannot achieve this purpose.

[0094] Based on the above ideal embodiments according to the present application, through the above description, relevant personnel can make various changes and modifications without deviating from the technical idea of the present application. The technical scope of the present application is not limited to the contents of the specification, and the technical scope must be determined according to the scope of the claims.

Claims

1. A method for synchronizing received signals, characterized in that, include: Peak position acquisition: Based on sequence interpolation correlation synchronization, detect the position of the maximum peak and the maximum value R. xy (max); Selecting sample values: Select relevant sample values ​​on the correlation curves on the left and right sides of the maximum peak. Synchronization position prediction: Based on the selected relevant sample values, perform linear fitting on the left point and linear fitting on the right point. The intersection of the two fitted lines is taken as the best synchronization point for prediction. Channel estimation: Select points to the left and right of the predicted optimal synchronization point as candidate points for the target synchronization position, extract the target signal at the candidate points for channel estimation, and calculate the channel equivalent SNR for different candidate points for the target synchronization position. Signal quality assessment: Select the position with the highest SNR as the final synchronization position.

2. The method for synchronizing received signals according to claim 1, characterized in that, In the step of selecting sample values, the selection of relevant sample values ​​is based on a threshold: R Thresh =αR xy (max) 0 < α < 1 For the relevant sequences on the left, select the relevant sample value R. xy (t)>=R Thresh point; For the relevant sequences on the right, the relevant sample value R is also selected. xy (t)>=R Thresh point; x: represents the transmitted reference sequence, y: represents the received reference sequence, t: represents a specific point in time, and R: represents the correlation between the transmitted and received sequences.

3. The method for synchronizing received signals according to claim 1, characterized in that, The channel estimation step further includes: the channel estimation is implemented in the time domain: assuming the transmitted synchronization reference sequence is x(t) and the received sequence is y(t), then the following condition is met: y(t)=A·x(t-τ)+n(t) The received signal is the original signal after time delay τ and amplitude change A. During transmission, it is contaminated by noise n(t), and the channel response h(t) can be expressed as: h(t) = y(t) / x(t) The channel response h(t) exhibits strong correlation in the time domain, while noise is random. The mean of the channel response is calculated using a moving average. Noise is the difference between the channel response and the smoothing result; The signal-to-noise ratio can be estimated based on the channel obtained at the current synchronization:

4. The method for synchronizing received signals according to claim 1, characterized in that, The synchronization position prediction also includes: predicting the position of the optimal synchronization point based on the morphology of the relevant waveform.

5. The method for synchronizing received signals according to claim 1, characterized in that, In the channel estimation step, points are selected as candidate points for the target synchronization position to the left and right of the predicted optimal synchronization point. The system then traverses the vicinity of the optimal synchronization point to find the position with the least inter-symbol interference.

6. The method for synchronizing received signals according to any one of claims 1 to 5, characterized in that, The channel estimation step includes: selecting a synchronization range, then traversing these locations, and selecting the final synchronization location based on the maximum SNR.

7. The method for synchronizing received signals according to any one of claims 1 to 5, characterized in that, In the step of selecting sample values, more than two relevant sample values ​​are selected on the relevant curves to the left or right of the maximum peak.

8. The method for synchronizing received signals according to any one of claims 1 to 5, characterized in that, The method of obtaining the peak position also includes: Coarse synchronization: After the synchronization signal sequence passes through the communication system, coarse synchronization is performed at the receiving side to find the approximate location of the synchronization signal sequence. Truncation Sequence: A segment of received data is truncated from the coarse synchronization sequence position and used as the received sequence for the next signal processing step; Sampling interpolation: The received truncated sequence is upsampled and interpolated at a high multiple, and the local synchronization sequence is upsampled at the same multiple. Then, the interpolated sequences are cross-correlated to obtain the correlation curve. Peak detection: Detect the precise location and maximum value R of the maximum peak based on the correlation curve. xy (max).

9. The method for synchronizing received signals according to claim 8, characterized in that, The high-magnification sampling interpolation step refers to high-magnification sampling interpolation exceeding 8 times.

Images