A pilot-based symbol fast recovery method

Abstract

The application discloses a kind of based on pilot optimization symbol fast recovery method, belong to wireless communication technical field, comprising: S1: by analyzing the contact between the theoretical decision straight line of demodulation symbol under no noise interference and the fitting straight line obtained by fitting demodulation symbol decision straight line using pilot symbol under noise interference, obtain the contact between pilot used for fitting straight line and symbol recovery performance, then by creating pilot optimization problem with the optimal symbol recovery performance as target, solve the problem to obtain optimal pilot;S2: the optimal pilot sequence obtained in S1 is inserted in the data symbol of sending end, and after FBMC-OQAM modulation, it is transmitted;S3: the received signal is demodulated at receiving end, and each subcarrier in the demodulated signal is judged respectively, and the data symbol sent by the sending end is recovered.The optimal solution of pilot optimization problem with the optimal symbol recovery performance as target is obtained by the application, so that the optimized pilot symbol can effectively improve the symbol recovery accuracy.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of wireless communication, and more particularly, to a pilot-optimized symbol fast recovery method, a storage medium, and a system. BACKGROUND

[0002] In the 1970s, mobile communication systems appeared in people's field of vision. Mobile communication systems have accompanied the progress and development of human civilization, and have injected more possibilities into the fields of education, medicine, business, and industry. Multi-carrier modulation (MCM) has been widely used in today's wireless communication systems due to its ability to effectively combat fading effects, of which the most representative is the orthogonal frequency division multiplexing technology (OFDM). OFDM achieves parallel transmission of high-speed serial data through frequency division multiplexing and can effectively combat multipath fading effects. However, since the OFDM system uses a rectangular pulse for shaping, the OFDM signal has a high spectral sidelobe, which can cause serious out-of-band leakage. Moreover, the OFDM system requires very strict time-frequency synchronization between the transmitting end and the receiving end, which requires a certain amount of synchronization resources. Therefore, it is necessary to explore more potential multi-carrier modulation techniques to make up for the shortcomings of OFDM.

[0003] Filter bank multicarrier with offset quadrature amplitude modulation (FBMC-OQAM) technology, as an extension of OFDM, is currently favored by researchers. Compared with the OFDM system, the FBMC-OQAM system uses a prototype filter with good time-frequency focusing characteristics, and transmits the real part and the imaginary part of the complex data half a symbol period apart. By designing the prototype filter to meet the inter-carrier orthogonality condition, the FBMC-OQAM system has lower out-of-band leakage and higher spectral efficiency. Unlike the OFDM system, the FBMC-OQAM system only satisfies the orthogonality condition in the real number domain, so the received signal is inevitably affected by the inherent imaginary part interference. Therefore, some traditional schemes suitable for the OFDM system cannot be directly applied to the FBMC-OQAM system, and need to be adjusted according to the characteristics of the FBMC-OQAM system.

[0004] The accuracy and efficiency of symbol recovery at the receiving end of a communication system determine the performance of data transmission of the whole system and are the focus of communication system research. The symbol fast recovery method is a new symbol recovery method based on fitting straight line in the FBMC-OQAM system, and can effectively improve the efficiency of symbol recovery because it does not need to use channel estimation and equalization. The method uses the distribution characteristics of demodulated symbols on the complex plane and combines with the knowledge of graphics, and realizes the fast recovery of symbols by fitting the demarcation line of the decision interval on the complex plane with pilot symbols. Compared with the traditional symbol recovery method in the FBMC-OQAM system, the method has lower computational complexity and still maintains good decision accuracy. Therefore, the symbol fast recovery method based on fitting straight line has more advantages and significance for further research than the traditional symbol recovery method in the FBMC-OQAM system. In the method, the pilot symbols used for fitting straight line determine the performance of symbol recovery, but the current optimization design research on pilot symbols in the symbol recovery method is only limited to the improvement of the performance of channel estimation in the traditional symbol recovery method, such as the imaginary number interference approximation method (IAM-I) and the complex value interference approximation method (IAM-C), which are only for pilot optimization of the IAM method and are not applicable to symbol recovery methods without channel estimation. The optimization research on pilot symbols in the symbol recovery method based on fitting straight line is also blank, so it is very meaningful to study the pilot optimization design in the symbol fast recovery method based on fitting straight line. SUMMARY

[0005] In view of the above defects or improvement needs of the prior art, the present application provides a symbol fast recovery method and device based on pilot optimization, which obtains the optimal solution of the pilot optimization problem in the symbol fast recovery method with the optimal symbol recovery performance as the target, thereby improving the recovery performance of the symbol fast recovery method in the FBMC-OQAM system.

[0006] To achieve the above object, in a first aspect, the present application provides a symbol fast recovery method based on pilot optimization, comprising the following steps:

[0007] S1, by analyzing the relationship between the theoretical decision straight line of demodulated symbols under noiseless interference and the fitting straight line obtained by fitting the demodulated symbol decision straight line under noise interference, the relationship between the pilot used for fitting the straight line and the symbol recovery performance is obtained, and then a pilot optimization problem with the optimal symbol recovery performance as the target is created, and the optimal pilot is obtained by solving the optimization problem;

[0008] S2, inserting the optimal pilot sequence obtained in step S1 into the data symbols at the sending end, and transmitting after FBMC-OQAM modulation;

[0009] S3. After the receiver demodulates the received signal using FBMC-OQAM, it performs a decision operation on each subcarrier in the demodulated signal to recover the data symbols transmitted by the transmitter.

[0010] Preferably, the theoretical decision line for the demodulated symbol under noise-free interference in step S1 is: Among them, H I (k) and H R (k) represents the imaginary and real parts of the channel frequency response H(k), x0∈[-(2 r -2),-(2 r -4),...,+(2 r -2)], and For x k (m) corresponds to the demodulation symbol y k The imaginary and real parts of (m).

[0011] Preferably, the fitted line obtained by fitting the pilot symbol demodulation symbol decision line under noise interference in step S1 is: Where, x0∈[-(2 r -2),-(2 r -4),...,+(2 r -2)],P k (1) = P k (2) = P k These are the two pilot symbols on the k-th subcarrier of the transmitter. For pilot demodulation symbol p k (1) and p k (2) imaginary part, For pilot demodulation symbol p k (1) and p k (2) The real part.

[0012] Preferably, in the fast symbol recovery method described in step S1, the relationship between the pilot signal used for fitting the straight line and the symbol recovery performance is determined by the mean square error of the sine value of the angle between the two straight lines on all subcarriers. And pilot demodulation symbol formula p k (1)=H(k)[P k +jI k (1)]+η k (1) and p k (2)=H(k)[P k +jI k (2)]+η k (2) The condition that needs to be satisfied when the included angle between the two lines is the smallest is: By finding the minimum value, the relationship between the pilot sign and the symbol recovery performance of the fitted straight line can be obtained.

[0013] Preferably, the optimization problem model in step S1 is:

[0014] P1 model:

[0015] Constraint: -(2 r -1)≤p k ≤2 r -1,k∈[0,K-1]

[0016] Where, I k (1) and I k (2) are the first column and the second column of pilot symbols p k suffered from imaginary part interference, A is a constant term, the FBMC-OQAM system is 2 2r QAM modulation, and the number of subcarriers is K.

[0017] Preferably, the decision operation in step S3 using the symbol fast recovery method is to calculate the decision straight line of the kth subcarrier based on the pilot demodulation symbol of the kth subcarrier and the corresponding pilot symbol, to obtain the sending symbol value corresponding to each demodulation symbol by comparing the relative positions of each demodulation symbol and the decision straight line on the kth subcarrier, and to recover the data symbol sent by the sending end.

[0018] According to an embodiment of the present application, the type of the FBMC-OQAM system symbol is any one of 4QAM, 16QAM and 64QAM used in FBMC-OQAM modulation.

[0019] According to a second aspect of the present application, the present application provides a computer readable storage medium, which has a program of a pilot optimization based symbol fast recovery method, and the program is used to implement the above pilot optimization based symbol fast recovery method when executed.

[0020] According to a third aspect of the present application, the present application provides a FBMC-OQAM system, comprising: a sending end and a receiving end.

[0021] The sending end is used to execute the steps S1-S2.

[0022] The receiving end is used to execute the step S3.

[0023] Overall, compared with the prior art, the above technical solutions conceived by the present application mainly have the following technical advantages:

[0024] 1. The method of the present application can obtain the optimal pilot for the purpose of optimal symbol recovery performance in the symbol fast recovery method, and the symbol fast recovery method using the pilot can achieve better symbol recovery performance, thereby improving the quality of wireless communication.

[0025] 2. The preferred symbol decision mode of the present application does not need to use channel estimation and equalization to recover the symbol, has lower calculation complexity, and can also guarantee the accuracy of symbol recovery.

[0026] 3. The method of the present application is suitable for the recovery of different QAM symbols, and the optimal pilot obtained for any kind of QAM symbol can effectively improve the symbol recovery performance of the symbol fast recovery method. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 is a flowchart of a symbol fast recovery method based on pilot optimization disclosed by embodiment 1 of the present application.

[0028] Figure 2 is a bit error rate (BER) comparison diagram of the system using the optimal pilot under 16QAM modulation of embodiment 1 of the present application.

[0029] Figure 3 is a bit error rate (BER) comparison diagram of the system using the optimal pilot under 64QAM modulation of embodiment 1 of the present application. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical scheme and advantages of the present application clearer and more comprehensible, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application. In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as they do not conflict with each other.

[0031] Embodiment 1,

[0032] A symbol fast recovery method based on pilot optimization, as shown in Figure 1 , comprises the following steps:

[0033] S1, by analyzing the relationship between the theoretical decision straight line of the demodulated symbol under noiseless interference and the fitting straight line obtained by fitting the demodulated symbol decision straight line under noise interference, the relationship between the pilot used for fitting the straight line and the symbol recovery performance is obtained, and then by creating a pilot optimization problem with the optimal symbol recovery performance as the target, the optimal pilot is obtained by solving the optimization problem;

[0034] S1.1 The equation of the theoretical decision straight line l0 of demodulation symbols without noise interference is:

[0035]

[0036] wherein, FBMC-OQAM system is denoted as OQAM modulation order 2 2r , r is a positive integer, then x0∈[-(2 r -2), -(2 r -4),..., +(2 r -2)], H I (k) and H R (k) are the imaginary part and the real part of the channel frequency response H(k), and are the imaginary part and the real part of the demodulation symbol y k (m) corresponding to x k (m), the straight line equation l0 divides the complex plane where the demodulation symbol is located into different regions for recovering the symbol, and the decision using the straight line l0 can accurately recover the symbol.

[0037] S1.2 The equation of the fitting decision straight line l1 obtained using two columns of the same pilot symbols under noise interference is:

[0038]

[0039] wherein, x0∈[-(2 r -2), -(2 r -4),..., +(2 r -2)], P k (1) = P k (2) = P k is the two columns of pilot symbols on the kth subcarrier of the sending end, are the imaginary parts of the pilot demodulation symbols p k (1) and p k (2), are the real parts of the pilot demodulation symbols p k (1) and p k (2).

[0040] S1.3 Analysis of the relationship between the theoretical decision line and the fitted line. Since the theoretical decision line is obtained in the ideal case without noise, but when the AWGN noise exists in the system, the demodulation symbol will be offset due to the influence of the AWGN, so the fitted line obtained from the pilot symbol and the theoretical decision line will be deviated. In order to improve the accuracy of symbol recovery, the fitted line should be as close to the theoretical decision line as possible. For the theoretical decision line and the fitted line at x0=0, there is a common intersection point (0,0), for the two lines with intersection points, when the angle between the theoretical decision line and the fitted line is smaller, the fitted line will be closer to the theoretical decision line, at this time the fitted line can more accurately realize symbol recovery.

[0041] S1.4 Further analysis of the relationship between the pilot used for fitting the line in the symbol fast recovery method and the symbol recovery performance;

[0042] Assuming that the number of subcarriers in the system is K, and the number of symbols is N. By calculating the mean square error of the sine value of the angle between the two lines on the subcarrier , the relationship between the fitted line and the symbol recovery performance is obtained. The sine value of the angle θ k between the two lines on the kth subcarrier can be obtained by , wherein k1 and k2 are the slopes of the two lines. The pilot symbol uses the formula p k (1) = H(k) [P k +jI k (1)] + η k (1), and p k (2) = H(k) [P k +jI k (2)] + η k (2) is further simplified, wherein I k (1), I k (2) and η k (1), η k (2) are the imaginary part interference and noise effects on the pilot P k (1) and P k (2). Since the imaginary part interference and noise effects are independent of each other, E(I k (m)η k (m)) = 0. By derivation, it is obtained that is:

[0043]

[0044] Since the channel response H(k) and the noise η k (m) are uncontrollable elements, only the imaginary part interference part in can be optimized. Because the imaginary part interference term (Ik (1)-I k (2)) 2 In the denominator part and the coefficient is positive, the uncontrollable term in the denominator is the square term, which can be considered as a constant term A, and A≥0. So when The smaller the value is, the closer the fitting straight line is to the theoretical decision straight line, and the better the symbol recovery performance is.

[0045] S1.5 creates a pilot optimization problem with the best symbol recovery performance as the goal. Since the imaginary part interference generated in the FBMC-OQAM system is mainly concentrated in the first-order neighborhood and the imaginary part interference values obtained at different index positions are different. Therefore, the interference outside the first-order neighborhood can be ignored first, and the pilot optimization problem under different index conditions is obtained.

[0046] Assume that the imaginary part interference factors of the FBMC-OQAM system are α, β and γ, and α>β, γ. The pilot symbol is p=[p0,p1,...,p K ], the data symbol is a=[a0,a1,...,a K ] and b=[b0,b1,...,b K ]. When the first column of the two-column pilot is under the condition of even index, the objective function can be expressed as:

[0047]

[0048] When the first column of the two-column pilot is under the condition of odd index, the objective function can be expressed as:

[0049]

[0050] According to the above analysis results, the pilot optimization problem can be expressed as:

[0051] P1 model:

[0052] Constraint condition:-(2 r -1)≤p k ≤2 r -1,k∈[0,K-1]

[0053] S1.6 obtains the value boundary condition of the optimal solution of the pilot optimization problem. When only considering the imaginary part interference generated inside the pilot and ignoring the influence of the external environment such as data symbols on the objective function, the objective function of the optimization problem can be divided into two cases: the first column of the pilot is located at even index and the first column of the pilot is located at odd index.

[0054] For the case that the first column of the two-column pilot is located at even index, the objective function G(p0,p1,...,p K-1 ) can be simplified as: For the first column of two-column pilot located at odd index, the objective function G(p0, p1, …, pK-1) can be expressed as: K-1 ) can be expressed as:

[0055] Both of the above cases can be obtained: since and let the first-order partial derivative of the objective function , the stationary point p (0) : p0= p1= …= p K-1 = 0, at this time the objective function is is the maximum value, so the stationary point p0= p1= …= p K-1 = 0 is also the maximum point.

[0056] According to Taylor expansion can be obtained So H G is a semi-negative definite matrix, and the objective function G(p) is a concave function. Therefore, the optimal solution of the optimization problem, that is, the minimum value p * is on the boundary of the problem, so there is at least one k0∈[0, K-1] such that or The original optimization problem is transformed into an M-1 element optimization problem. When the index k of the independent variable p k is within the range [k0-2, k0+2], the variable is a known item, so the partial derivative related to will be reduced, so when calculating the newly generated optimization problem, the Hessian matrix corresponding to the original objective function can be deleted to form an M-1 order matrix related to When the index k of the independent variable p k is not within the range [k0-2, k0+2], the first-order partial derivative and the second-order partial derivative related to p k of the objective function of the M-1 element optimization problem do not change, and all principal minors of the semi-negative definite matrix are non-positive. Therefore, the Hessian matrix of the M-1 element optimization problem is still semi-negative definite, and the objective function is a concave function. The optimal solution of the M-1 element optimization problem obtained is still on the boundary condition, and there is at least one k1∈[0, K-1] such that or In summary, the optimal solution p * of the optimization problem should be on the boundary condition, so for k∈[0, K-1], p k = ±(2 r -1). At this time, the optimization problem can be simplified as:

[0057] P1 model:

[0058] Constraint: pk = ±(2 r -1), k e [0, K-1]

[0059] S1.7 Analysis of the influence of different pilot value patterns on the objective function value. Assume X = 2 r -1, Y = -( r 2-1, then the pilot value pattern P k-1 P k P k+1 There are 8 cases, and the relationship between the objective function and the pilot value pattern under different index conditions can be summarized as shown in Table 1:

[0060] Table 1

[0061]

[0062] According to Table 1, when the pilot value pattern p k-1 p k p k+1 = XYX or YXY, the first column of the two columns of pilots is located at an even index, p k The corresponding objective function value is smaller, which is When the pilot value pattern p k-1 p k p k+1 = XXX or YYY, the first column of the two columns of pilots is located at an odd index, p k The corresponding objective function value is smaller, which is For the pilot value pattern p k-1 p k p k+1 = XYY / YXX and p k-1 p k p k+1 = XXY / YYX, if the objective function value is to be minimized, the pilot value pattern on the odd index and the even index should be XYY / YXX, in which case p k The corresponding objective function value is However, due to the continuity and periodicity of the pilot value pattern, it cannot be guaranteed that all pilot symbols can obtain such objective function values, so further analysis is needed for this case:

[0063] Assume that the pilot symbol p k is located in the value pattern p k-1 p k p k+1 = XYY / YXX, and the corresponding objective function value is When the next pilot p k+2 = X, then p k p k+1 p +k2=YY / X X; when the next pilot p k+2 = Y, then p k p +k 1+ p= k 2 / Y Y. When p k p k+1 p k+2 = YYX or XXY, the pilot symbol p k+1 corresponds to the target function value When p k p k+1 p k+2 = XXX or YYY, the pilot p k+1 corresponds to the target function value At this time, if the next pilot p k+3 ≠ p k+2 , then p k+2 corresponds to the target function value If the values of the pilots thereafter are all equal to the pilot on the previous subcarrier, the pilot sequence will eventually be connected with p k-1 , forming p k-3 p k-2 p k-1 = YYX / XXY, then p k-2 corresponds to the target function value In summary, when a pilot symbol in the pilot sequence corresponds to the target function value , there is at least one pilot in the sequence that corresponds to the target function value Therefore, the number of pilots with the target function value w1 and the number of pilots with the target function value w2 have the relationship

[0064] Conversely, if a pilot symbol p k corresponds to the target function value , then the value of p k should satisfy p k-1 p k p k+1 = YYX / XXY. If the next pilot p k+2 = X, then p k p k+1 p k+2 = YXX / XYX; if the next pilot p k+2 = Y, then p k p k+1 p k+2 = YXY / XYY. When p k p k+1 pk+2 = YXX or XYY, pilot symbol p k+1 The corresponding objective function value is When p k p k+1 p k+2 = XYX or YXY, pilot p k+1 The corresponding objective function value is Under this condition, if the next pilot p k+3 = p k+2 , then p k+ 1p k+2 p k+3 = XYY / YXX, then p k+2 The corresponding objective function value is If the values of the pilots after that are all different from the pilot on the previous subcarrier, due to the cyclical nature of the subcarrier index, the pilot sequence will eventually be connected to p k-1 , forming the value pattern of p k-3 p k-2 p k-1 = XYY / YXX, at which time p k-2 The corresponding objective function value is In summary, when there is a pilot symbol in the pilot sequence whose corresponding objective function value is , there is at least one pilot in the sequence whose corresponding objective function value is Therefore, the number of pilots whose objective function value is w1 and the number of pilots whose objective function value is w2 have the relationship

[0065] From the above analysis, we can see that and we can get Therefore, the number of pilots whose objective function value is in the entire pilot sequence is equal to the number of pilots whose objective function value is .

[0066] Further analysis of the impact of the pilot symbol whose objective function value is on the pilot optimization problem. Since the pilot symbols whose objective function values are and appear in pairs, if we change the objective function values of N1' pilots whose objective function values are to and , the change in the entire system objective function can be represented as:

[0067] ​

[0068] where, G(p) = G(p change It can be known that the target function change quantity generated by N1' part is positive, and the target function change quantity generated by N'2 part can be divided into two cases:

[0069] ① When (α 2 + 4β 2 +A')·(αγ-γ 2 + β 2 )-4β 2 · α 2 ≥ 0, the target function change quantities generated by N1' and N'2 parts are both positive, so G(p change ≥ 0. If you want to get a smaller target function, you need to take the minimum value 0 of N1' and N'2, so there is no pilot symbol in the pilot sequence whose target function value is . Assuming that there are N0 pairs of pilot symbols in the pilot sequence whose target function values are and The total number of subcarriers of the OQAM-FBMC system is K, and the target function value G(p) of the system is:

[0070]

[0071] where, the N0 part of the target function value G(p) is positive, so the target function value G(p) will increase with the increase of N0, and therefore N0 should take the minimum value 0. At this time, the value of the target function is The target function value corresponding to each pilot symbol in the pilot sequence is

[0072] ② When (α 2 + 4β 2 +A')·(αγ-γ 2 + β 2 )-4β 2 · α 2 < 0, the target function change quantity generated by N1' part is positive, so N1' should take the minimum value 0; the target function change quantity generated by N'2 part is negative, and the target function will decrease with the increase of N'2, so N'2 should be as large as possible. Therefore, there is no case that the target function corresponding to the pilot symbol changes from to , but there may be cases that the target function of the pilot symbol changes from and to . Assuming that there are N0 pairs of pilot symbols in the pilot sequence whose target function values are and where N'2 pairs of pilot symbols change their target functions to The system target function value G(p) can be represented as:

[0073]

[0074] Since the target function value generated by N0 part and N'2 part is positive, N0 and N'2 should take the minimum value 0 to make the value of the target function minimum. At this time, N0=N'2=0, the target function value corresponding to each pilot symbol in the pilot sequence is

[0075] Therefore, when each pilot symbol p k in the pilot sequence takes the value , the target function G(p) can take the minimum value, at this time, the pilot sequence needs to satisfy that the first column is at even index and the value mode is XYXY repetition or the first column is at odd index and the value mode is XXXX / YYYY repetition.

[0076] S1.8 obtains the optimal pilot symbol design. The first column at even index and the value mode is XYXY repetition or the first column at odd index and the value mode is XXXX / YYYY repetition can be summarized as three specific cases:

[0077] Table 2

[0078]

[0079]

[0080] Since the data symbol in the system is a random variable, the random variable x cannot directly use the distribution characteristics of the data symbol in the target function when the denominator is in the denominator. Therefore, the Monte Carlo simulation method is used in MATLAB to calculate the target function values of the three cases, and the results are shown in Table 3. By comparing the target function values, the optimal solution of the pilot optimization problem is obtained as follows: when the first column index of the two columns of pilots is 0 and the pilot value mode is p k-1 =-p k =p k+1 =±(2 r -1).

[0081] Table 3

[0082]

[0083] S2, the optimal pilot sequence obtained in step S1 is inserted into the data symbol at the sending end, and after FBMC-OQAM modulation, it is transmitted;

[0084] According to the optimal pilot position and value obtained in step S1, the value sequence p k-1= -p k = p k+1 = ±(2 r -1) optimal pilot sequence, after FBMC-OQAM modulation and transmission;

[0085] S3, the receiving end of the received signal is subjected to FBMC-OQAM demodulation, and each subcarrier in the demodulated signal is subjected to decision operation, thereby recovering the data symbol sent by the sending end;

[0086] Based on the pilot demodulation symbol of the kth subcarrier and the corresponding pilot symbol, the decision straight line L k (x0) is calculated:

[0087]

[0088] Wherein, x0∈[-(2 r -2), -(2 r -4),..., +(2 r -2)], P k (1) = P k (2) = P k The two columns of pilot symbols on the kth subcarrier of the sending end, The imaginary part of the pilot demodulation symbol p k (1) and p k (2), The real part of the pilot demodulation symbol p k (1) and p k (2), And The real part and the imaginary part of the mth demodulation symbol on the kth subcarrier.

[0089] The decision straight line L k (x0) divides the complex plane corresponding to the demodulation symbol into mutually disjoint regions, and by comparing the relative position of the demodulation symbol on the kth subcarrier and the decision straight line L k (x0) in the complex plane, the corresponding sending symbol value of each demodulation symbol is obtained, the demodulation symbols falling into the same interval are judged as the same symbol, and the demodulation symbols falling into different intervals are judged as different symbols.

[0090] Application test

[0091] In order to further illustrate the performance of the pilot optimization based symbol fast recovery method provided in embodiment 1 of the present application, the following will be described in detail in combination with specific experiments:

[0092] In the experiment, the subcarrier number K of the FBMC-OQAM system is set to 1024, the symbol number N of a frame of signals is 30, 16QAM modulation and 64QAM modulation are used, in the SUI3 channel, the filter uses an SRRC filter with a roll-off factor of 0.5. Figure 2 and Figure 3 The bit error rate (BER) performance of the optimal pilot obtained by using the application in 16QAM and 64QAM modulation is respectively shown. Figure 2 and Figure 3 It can be seen that the performance of the optimal pilot obtained by using the application is obviously better than that of other pilot sequences, which shows that the optimal pilot sequence obtained by using the application can effectively improve the symbol recovery performance.

[0093] Those skilled in the art can easily understand that the above description is only a preferred embodiment of the application, and is not used to limit the application, and any modification, equivalent replacement and improvement made within the spirit and principle of the application should be included in the protection scope of the application.

Claims

1. A pilot-based optimization method for fast recovery of FBMC-OQAM system symbols, characterized in that, The method comprises the following steps: S1, obtaining the relationship between the pilot and the symbol recovery performance for the fitting line by analyzing the relationship between the theoretical decision straight line of the demodulation symbol under no noise interference and the fitting straight line obtained by fitting the pilot symbol demodulation symbol decision straight line under noise interference, and then solving the optimization problem by creating a pilot optimization problem with the optimal symbol recovery performance as the target to obtain the optimal pilot; S2, inserting the optimal pilot sequence obtained in step S1 into the data symbol at the sending end, modulating and transmitting after FBMC-OQAM modulation; S3, the receiving end performs FBMC-OQAM demodulation on the received signal, and then performs decision operation on each subcarrier in the demodulation signal to recover the data symbol sent by the sending end; The theoretical decision line of the demodulation symbol without noise interference in step S1 is: wherein and are the imaginary and real parts of the channel frequency response , , and are the imaginary and real parts of the corresponding demodulation symbol , . The fitted line obtained by fitting the pilot symbol demodulation symbol decision line under noise interference in step S1 is: ,in, , For the sending end Two columns of pilot symbols on each subcarrier , Pilot demodulation symbols and The imaginary part, , Pilot demodulation symbols and The real part; The connection between the pilot for fitting straight line and the symbol recovery performance in the symbol fast recovery method described in step S1 is obtained by the mean square error of the sine value of the angle between two straight lines on all subcarriers and the pilot demodulation symbol formula and The condition that needs to be met when the angle between two straight lines is the smallest is obtained as The minimum value is taken, thereby obtaining the connection between the pilot symbol for fitting straight line and the symbol recovery performance; the mean square error is: is the remainder of the denominator of the mean square error, and ; The optimization problem model in step S1 is: ， and for the first and second column of pilot symbols subject to imaginary part interference, FBMC-OQAM systems are modulation, the number of subcarriers is .

2. The method of claim 1, wherein, The type of the FBMC-OQAM system symbol is any one of 4QAM, 16QAM and 64QAM in FBMC-OQAM modulation.

3. A computer readable storage medium having a symbol fast recovery method based on pilot optimization program, the program is executed to implement the symbol fast recovery method based on pilot optimization of the FBMC-OQAM system of any one of claims 1-2.

4. A FBMC-OQAM system, characterized in that, Comprise: The sending end and the receiving end; The sending end is used to execute steps S1-S2 in any one of claims 1-2; The receiving end is used to execute step S3 in any one of claims 1-2.

Images