A cyclic prefix double estimation method based on smoothing correlation

Abstract

The application discloses a cyclic prefix double estimation method based on smooth correlation, solves the problems of low estimation precision, sensitivity to noise and poor adaptability of a traditional cyclic prefix estimation method under low signal-to-noise ratio, uses the method of smooth correlation, obtains a relatively low-precision result through rough estimation, and then uses fine estimation to improve the overall estimation precision of the algorithm. The application still maintains a low estimation error under low signal-to-noise ratio, has low calculation complexity and strong robustness, is very suitable for engineering implementation, and lays a foundation for subsequent subcarrier modulation identification and bit error rate analysis in non-cooperative communication.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of communication signal recognition and detection, and particularly relates to a cyclic prefix double estimation method based on smooth correlation. BACKGROUND

[0002] Orthogonal Frequency Division Multiplexing (OFDM) is a widely used modulation technique in wireless communication systems, especially in modern communication standards such as LTE, 5G, Wi-Fi, etc. OFDM technology divides high-speed data streams into multiple low-speed subcarriers for parallel transmission, greatly improving spectral efficiency and anti-multipath interference capability. However, OFDM signals are usually affected by factors such as multipath effect, frequency offset, noise, etc. in actual transmission, which can introduce inter-symbol interference (ISI) and inter-carrier interference (ICI), thus affecting the performance of the communication system.

[0003] To cope with multipath effect and inter-symbol interference, a cyclic prefix (CP) is usually added before each symbol of the OFDM signal. The cyclic prefix is a copy of the last part of the symbol copied to the front of the symbol, which can effectively eliminate the inter-symbol interference caused by multipath propagation, thus improving the reliability and performance of the system. Traditional cyclic prefix length estimation methods mainly rely on channel estimation, signal correlation analysis or detection based on time-domain signals, but these methods generally have low estimation accuracy, are sensitive to noise and have poor adaptability. Therefore, how to accurately estimate the cyclic prefix length of the OFDM signal under complex wireless channel conditions with low computational complexity and high estimation accuracy is still a technical problem to be solved.

[0004] In related aspects, document [1] increases the IFFT length by each offset during the calculation of the cross-correlation function of the received OFDM signal, and obtains the OFDM signal length according to the periodicity of the obtained function, and then obtains the cyclic prefix. However, this algorithm is less adaptable and too simple, and it is difficult to be used in more complex environments.

[0005] The variable starting point correlation function method mentioned in document [2] has a large fluctuation in the data when the signal-to-noise ratio is low, making it difficult to accurately extract useful information.

[0006] Document [3] uses a cyclic autocorrelation function to obtain the OFDM symbol length, but this method has relatively high complexity and low estimation accuracy at low signal-to-noise ratios. SUMMARY

[0007] The purpose of this invention is to address the problems of low estimation accuracy, sensitivity to noise, and poor adaptability of traditional cyclic prefix estimation methods under low signal-to-noise ratios, and to provide a dual estimation method for cyclic prefixes based on smooth correlation.

[0008] The objective of this invention is achieved through the following technical solution:

[0009] A cyclic prefix dual estimation method based on smooth correlation includes the following steps:

[0010] Step 1: Acquire the baseband OFDM signal and sample it to obtain the baseband discrete OFDM signal s(i);

[0011] Step 2: Perform autocorrelation operation on the baseband discrete OFDM signal s(i), extract the autocorrelation function R(i,m), and take the position m of the second largest value of the autocorrelation function R(i,m) as the OFDM signal IFFT length N, where m is the correlation offset length;

[0012] Step 3: Using the OFDM signal IFFT length N as the smoothing window length, smooth the OFDM signal s(i) in the forward direction along the baseband discrete signal s(i). The local signal obtained after each smoothing by the smoothing window is used as the smoothed signal s. p (i,k), where k is the smoothing step size, is used to calculate the smoothed signal s. p The absolute value R of the cross-correlation function between (i,k) and the original signal s(i) is... pk (i,m);

[0013] Step 4: Continuously increase the smoothing step size k, and extract the absolute value of the cross-correlation function for each smoothing step size k and the correlation offset length m = -(Nk) to obtain the periodic function R. s (i,k)=R pk (i,-(Nk));

[0014] Step 5: Scan R s The method of obtaining R using the zero point position (i,k) yields R. s The period L of the function (i,k) b ; will L b The cyclic prefix length M is roughly estimated as the symbol length of the OFDM signal. b =L b -N;

[0015] Step 6: Using [M] b -3,M b [+3] represents the smoothing range. Smoothing step sizes k are sequentially selected from this range. For each smoothing step size k, the data representing the smoothing length on the right side of the smoothed signal is shifted to the left side to obtain signal s. p2 (i,k), calculate s p2(i, k) is the Pearson correlation coefficient r of the original signal s(i) and the first N sample points of the smoothed signal s(i, k) coef If the Pearson correlation coefficient r of the original signal s(i) and the smoothed signal s(i, k) changes, the step length M is taken as the estimation of the length of the cyclic prefix, and the estimation of the length of the cyclic prefix is completed. a coef a

[0016] Further, the expression of the OFDM signal received at i moment in the step 1 is:

[0017]

[0018] In the formula, a n is the complex signal on the nth subcarrier.

[0019] Further, the autocorrelation function R(i, m) corresponding to the step 2,

[0020]

[0021] In the formula, is the energy of a single sample point, is the energy of the OFDM signal.

[0022] Further, the absolute value R p of the cross-correlation function between the smoothed signal and the original signal is used in the rough estimation method in the step 3, and is obtained by the following formula:

[0023]

[0024] In the formula, is the data overlap length of the offset smoothed signal and the original signal, and the superscript * represents conjugation.

[0025] Further, the length of the cyclic prefix M b = L b -N is obtained by the rough estimation in the step 5, wherein,

[0026] index = find(R s == 0)

[0027] L b = index(2) - index(1)

[0028] In the formula, index is the zero point coordinate of R s .

[0029] Further, the Pearson correlation coefficient r coef of the original signal s(i) and the smoothed signal s(i, k) is obtained in the step 6,

[0030] ​​​

[0031] An electronic device includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described above.

[0032] A computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0033] A computer program product having at least one instruction stored thereon, wherein the at least one instruction, when executed by a processor, implements the steps of the above method.

[0034] The beneficial effects of this invention are as follows:

[0035] This invention utilizes a smoothing correlation method to obtain a relatively low-precision result through coarse estimation, and then uses fine estimation to improve the overall estimation accuracy of the algorithm. This invention maintains a low estimation error even at low signal-to-noise ratios, has low computational complexity and strong robustness, making it very suitable for engineering implementation and laying the groundwork for subsequent subcarrier modulation identification and bit error rate analysis in non-cooperative communication. Attached Figure Description

[0036] Appendix Figure 1 This is a schematic diagram of the coarse estimation method in the method of the present invention.

[0037] Appendix Figure 2 The periodic function R in the coarse estimation method s A schematic diagram.

[0038] Appendix Figure 3 This is a schematic diagram of the precise estimation method in the method of this invention.

[0039] Appendix Figure 4 This is an algorithm flowchart of the method of the present invention.

[0040] Appendix Figure 5 The figures show simulation results of the coarse estimation method with signal-to-noise ratios of 1dB, 3dB, 7dB, and 12dB under Gaussian white noise background in this embodiment of the invention.

[0041] Appendix Figure 6 The figures show the simulation results of the precise estimation method with a signal-to-noise ratio of 1dB and 3dB under a Gaussian white noise background in this embodiment of the invention.

[0042] Appendix Figure 7 This is a comparison chart of the estimation accuracy of the method in this invention and the method in Reference 1 under different signal-to-noise ratios in the embodiments of this invention. Detailed Implementation

[0043] The present invention will now be further described with reference to the accompanying drawings.

[0044] Example 1:

[0045] This invention provides a dual estimation method for cyclic prefixes based on smooth correlation, comprising the following steps:

[0046] Step 1: Acquire the baseband OFDM signal and sample it to obtain the baseband discrete OFDM signal s(i);

[0047] Step 2: Perform autocorrelation operation on the baseband discrete OFDM signal s(i), extract the autocorrelation function R(i,m), and take the position m of the second largest value of the autocorrelation function R(i,m) as the OFDM signal IFFT length N, where m is the correlation offset length;

[0048] Step 3: Using the OFDM signal IFFT length N as the smoothing window length, smooth the OFDM signal s(i) in the forward direction along the baseband discrete signal s(i). The local signal obtained after each smoothing by the smoothing window is used as the smoothed signal s. p (i,k), where k is the smoothing step size, is used to calculate the smoothed signal s. p The absolute value R of the cross-correlation function between (i,k) and the original signal s(i) is... pk (i,m);

[0049] Step 4: Continuously increase the smoothing step size k, and extract the absolute value of the cross-correlation function for each smoothing step size k and the correlation offset length m = -(Nk) to obtain the periodic function R. s (i,k)=R pk (i,-(Nk));

[0050] Step 5: Scan R s The method of obtaining R using the zero point position (i,k) yields R. s The period L of the function (i,k) b ; will L b The cyclic prefix length M is roughly estimated as the symbol length of the OFDM signal. b =L b -N;

[0051] Step 6: Using [M] b -3,M b [+3] represents the smoothing range. Smoothing step sizes k are sequentially selected from this range. For each smoothing step size k, the data representing the smoothing length on the right side of the smoothed signal is shifted to the left side to obtain signal s. p2 (i,k), calculate s p2 The Pearson correlation coefficient r between (i,k) and the first N sampling points of the original signal s(i) is... coef If corresponding to a certain smoothing step size M a r coef If a transition occurs, then M will be... aAs an estimate of the cyclic prefix length, the cyclic prefix estimation is completed.

[0052] In this embodiment, the expression for the OFDM signal received at time i in step 1 is:

[0053]

[0054] In the formula, a n It is a complex signal on the nth subcarrier.

[0055] In this embodiment, step 2 corresponds to the autocorrelation function R(i,m).

[0056]

[0057] In the formula, The energy of a single sampling point, The energy of the OFDM signal.

[0058] In this embodiment, the coarse estimation method in step 3 utilizes the absolute value R of the cross-correlation function between the smoothed signal and the original signal. p It can be obtained from the following formula:

[0059]

[0060] in, This represents the data overlap length between the smoothed signal after offset and the original signal. The superscript * indicates conjugate.

[0061] In this embodiment, step 5 involves a rough estimation of the cyclic prefix length M. b =L b -N, where,

[0062] index = find(R) s ==0)

[0063] L b =index(2)-index(1)

[0064] In the formula, index is R. s The zero point coordinates.

[0065] In this embodiment, the Pearson correlation coefficient r in step 6 is mentioned. coef ,

[0066]

[0067] Example 2:

[0068] The embodiment discloses an electronic device, comprising a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to realize the steps of the above method.

[0069] Embodiment 3

[0070] The embodiment discloses a computer readable storage medium, which stores a computer program / instruction, and the computer program / instruction is executed by a processor to realize the steps of the above method.

[0071] Embodiment 4

[0072] The embodiment discloses a computer program product, which stores at least one instruction, and the at least one instruction is executed by a processor to realize the steps of the above method.

[0073] Embodiment 5

[0074] According to the cyclic prefix double estimation method based on smooth correlation in the embodiment 1, the method is described in combination with the accompanying drawings, and the method comprises the following steps:

[0075] S1: obtaining a baseband OFDM signal through a receiver and sampling the baseband OFDM signal to obtain a baseband discrete OFDM signal s(i).

[0076] S2: performing autocorrelation operation on the received signal, extracting a secondary maximum value position of the autocorrelation function R(i,m) to obtain an IFFT length N of the OFDM signal, and the calculation formula is as follows:

[0077]

[0078] In the formula, m is a correlation offset length, is the energy of a single sampling point, is the energy of the OFDM signal.

[0079] S3: as shown in the accompanying Figure 1 , taking the IFFT length as a smooth window length, then smoothing the received signal in a forward direction, and defining a local signal obtained after each smoothing of the smooth window as a smooth signal s(i,k), k is a smoothing step, and calculating the absolute value R p of the cross-correlation function between the smooth signal and the original signal. The calculation formula is as follows: p

[0080]

[0081] In the formula, is the data overlap length of the offset smooth signal and the original signal.

[0082] S4: as shown in the accompanying Figure 2 ​As shown, the smoothing step is increased constantly, the absolute value of the cross-correlation function at m = -(N-k) after each smoothing is extracted, and the periodic function R is obtained s .

[0083] S5: Scan R s to obtain the function period L by the method of zero point position b . The specific calculation formula is as follows:

[0084] index = find(R s == 0) (3)

[0085] L b = index(2)-index(1) (4)

[0086] In the formula, index is the zero point coordinate of R s .

[0087] S6: The period L b is the symbol length of the OFDM signal, and the length of the cyclic prefix is calculated.

[0088] M b = L b -N (5)

[0089] In the formula, M b is the length of the cyclic prefix obtained by rough estimation.

[0090] S7: As shown in the attached Figure 3 , the smoothing range is [M b -3, M b +3], the data of the right side of the smoothing signal is moved to the left side, and then the Pearson correlation coefficient r coe f of the first N sampling points of the original signal is calculated. The calculation formula is as follows:

[0091]

[0092] S8: Smooth in the range in turn, when r coe f jumps, the smoothing length at this time is the length of the cyclic prefix M a of the OFDM signal;

[0093] S9: Set different signal-to-noise ratios, store the rough estimation result and the precise estimation result of each experiment, and observe the relationship between the algorithm performance and the signal-to-noise ratio;

[0094] S10: Set different signal-to-noise ratios, perform repetitive experiments, store the algorithm estimation result of each experiment, calculate the root mean square error RMSE combined with the actual length of the cyclic prefix, perform error analysis, and compare with other cyclic prefix estimation algorithms to analyze the superiority of the algorithm;

[0095]

[0096] M r The actual length of the cyclic prefix.

[0097] Example 6:

[0098] According to the cyclic prefix double estimation method based on smoothing correlation of the embodiments 1 or 5, the superior performance of the proposed method is analyzed by the following simulation experiment.

[0099] Simulation conditions:

[0100] Sampling frequency: 10MHz

[0101] Carrier frequency: 20kHz

[0102] Number of symbols: 2000

[0103] Number of subcarriers for each modulation mode: 32

[0104] IFFT length: 64

[0105] Cyclic prefix length: 16

[0106] Subcarrier modulation mode: QPSK, 16QAM

[0107] Number of cycles: 1000

[0108] SNR range: [1dB: 12dB]

[0109] As shown in the accompanying drawings, Figure 5 this embodiment gives the simulation results of the coarse estimation method under 1dB, 3dB, 7dB and 12dB. As can be seen, the periodicity of the second largest value obtained by coarse estimation is weak at 1dB, and after 3dB, the periodicity of the second largest value obtained by coarse estimation gradually increases and is relatively stable, with a prominent peak at the end of the period, which is convenient for data processing. The stability of the data period shows that the coarse estimation method has strong robustness and good performance at low SNR.

[0110] As shown in the accompanying drawings, Figure 6 this embodiment gives the results of the fine estimation method under 1dB and 3dB when the coarse estimation result is 18. When the coarse estimation result is 18, the correlation coefficient should change abruptly when the smoothing step length changes from 3 to 4. As can be seen, the fine estimation method can estimate the cyclic prefix length at 1dB, but the abrupt change is not obvious, while at 3dB, the correlation coefficient jumps significantly, and the cyclic prefix length can be accurately estimated.

[0111] The accompanying drawings, Figure 7The estimation accuracy comparison chart of the present method and the method of document 1 under different signal-to-noise ratios is shown. It can be seen that the estimation error of the present method is significantly smaller than that of the method of document 1, and the data of the present method is more stable under low signal-to-noise ratio, which is convenient for processing and extracting information.

[0112] Therefore, the cyclic prefix double estimation method based on smoothing correlation provided by the present application is simple in calculation, strong in robustness and adaptability, and is very suitable for engineering implementation, and lays a foundation for subsequent subcarrier modulation identification and bit error rate analysis in non-cooperative communication.

[0113] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0114] The references in the background art are as follows:

[0115] Document [1] Si Hai-fei, Hu Xing-liu, Shi Zhen, etc. Non-cooperative signal modulation recognition algorithm based on joint feature parameter extraction [J]. Journal of communications, 2020, 41(07): 172-185.

[0116] Document [2] Li Lu-lu. Non-cooperative OFDM signal recognition and parameter estimation in multipath scene [D]. Harbin Engineering University, 2018.

[0117] Document [3] Yan Fu-cheng, Cheng Yu-fan, Lu Xuan-yu, etc. Research on improved OFDM time parameter blind estimation algorithm based on cyclic autocorrelation [J]. Signal processing, 2019, 35(01): 65-74.

Claims

1. A method of double estimation of cyclic prefix based on smoothing correlation, characterized in that, The method comprises the following steps: Step 1: obtaining a baseband OFDM signal and sampling it to obtain a baseband discrete OFDM signal s(i); Step 2: performing autocorrelation operation on the baseband discrete OFDM signal s(i) to extract an autocorrelation function R(i,m), and taking a position m corresponding to a second maximum value of the autocorrelation function R(i,m) as an IFFT length N of the OFDM signal, wherein m is a correlation offset length; Step 3: smoothing the forward of the discrete OFDM signal s(i) with the OFDM signal IFFT length N as the smoothing window length, taking the local signal obtained after each smoothing of the smoothing window as the smoothed signal s p (i,k), k is the smoothing step length, calculating the cross-correlation function of the smoothed signal s p (i,k) and the original signal s(i) pk (i,m); Step 4: Extract the absolute value of the cross-correlation function corresponding to each smoothing step k and the associated offset length m = -(N-k) to obtain the periodic function R s (i,k) = R pk (i,-(N-k)); Step 5: R is obtained by scanning R s (i,k) zero position method R s (i,k) function period L b ; L b As the symbol length of the OFDM signal, the coarse estimate of the cyclic prefix length M b = L b -N; Step 6: take [M b -3, M b +3] as smoothing range, and take smoothing step k from the smoothing range in turn; for each smoothing step k, move the data of right smoothing length in the smoothing signal to the left side to obtain signal s p2 (i, k), calculate the Pearson correlation coefficient r p2 (i, k) between s coef (i, k) and the first N sampling points of the original signal s(i), and if r a (i, k) jumps for a certain smoothing step M coef , then take M a as the estimated value of the cyclic prefix length to complete the estimation of the cyclic prefix.

2. The method of claim 1, wherein the method is based on a smoothing correlation, and wherein the method is a cyclic prefix double estimation method. An expression of the OFDM signal received at the i moment in the step 1 is as follows: where a n is the complex signal on the nth subcarrier.

3. The method of claim 1, wherein the method is characterized by, The autocorrelation function R(i,m) corresponding to the autocorrelation function R(i,m) in the step 2 is as follows: wherein is the energy of a single sample point, is the energy of the OFDM signal.

4. The method of claim 1, wherein the method is characterized by, The rough estimation method in step 3 utilizes the absolute value R of the cross-correlation function between the smoothed signal and the original signal pk is obtained from the following equation: wherein is the data overlap length of the post-offset smoothed signal and the original signal, and the superscript * denotes conjugation.

5. The method of claim 1, wherein the method is based on a smoothing correlation, and wherein the method is a cyclic prefix double estimation method. The step 5 roughly estimates the cyclic prefix length M b = L b -N, wherein, index = find(R s == 0) L b = index(2) - index(1) where index is R s (i, k) zero position.

6. The method of claim 1, wherein the method is based on a smoothing correlation, and The Pearson correlation coefficient r in step 6 coef , 7. An electronic device comprising a memory, a processor, and a computer program stored on the memory, wherein the computer program comprises instructions that, when executed by the processor, cause the electronic device to perform the method of any one of claims 1-6. The processor executes the computer program to realize the steps of the method in any one of claims 1 to 6.

8. A computer-readable storage medium having stored thereon computer program instructions, wherein, The computer program instructions are executed by the processor to realize the steps of the method in any one of claims 1 to 6.

9. A computer program product having stored thereon at least one instruction, characterized in that, The at least one instruction is executed by the processor to realize the steps of the method in any one of claims 1 to 6.

Images