A frequency synchronization method based on multi-phase frequency locked loop

By using a multi-phase locking ring in the locking ring, using multiple frequency discriminators with different phases and accuracy to perform parallel locking rings, increasing phase tracking, the problem of poor bandwidth and accuracy flexibility of traditional locking rings is solved, and the problem of improved Performance of frequency synchronization.

CN117395767BActive Publication Date: 2024-12-06KEYIDEA SATCOM INFORMATION TECH (NANJING) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311467161.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-11-07
Publication Date
2024-12-06
Estimated Expiration
2043-11-07

AI Technical Summary

Technical Problem

There is only a single frequency discriminator in the traditional frequency lock ring, which leads to poor bandwidth and accuracy flexibility, and the inability to effectively track phases, resulting in limited frequency synchronization performance.

Method used

Multiple frequency discriminators with different phases and accuracy are used, and the parallel frequency locking ring of different frequency discriminators is added through the criterion to make up for the shortcomings of the frequency locking ring.

Benefits of technology

It improves the flexibility and accuracy of frequency synchronization, can track phase more effectively, and enhances frequency synchronization performance under high dynamics, large bandwidth, and low signal-to-noise ratio conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117395767B_ABST
    Figure CN117395767B_ABST
Patent Text Reader

Abstract

The invention relates to the field of wireless communication technology, and discloses a frequency synchronization method based on a multi-phase frequency locked loop. The key points of the technical solution are as follows: according to the frame structure, the load data is eliminated, and the pilot is sent to the differential link alone; the adjacent pilot symbols are averaged, then differentiated in sequence, and finally the differences are averaged; the differential value after averaging the entire pilot segment is obtained by a divider; the real part of the differential value is divided by the imaginary part to obtain the quotient; the arc tangent value of the quotient is obtained by looking up an atan table; according to a switching strategy, it is determined which differential parameter value is used; the value is divided by the differential parameter to obtain a frequency value #imgabs0# estimated by frequency synchronization; the complex amplitude value of the frequency is obtained by looking up a cos table; the input data is stored in a memory, and the data is read out after the complex amplitude value of the frequency is updated; the data is multiplied by the conjugate value of the estimated frequency value, that is, a frequency synchronization is completed; a plurality of discriminators with different phases and precisions are used, and parallel frequency locked loops of different discriminators are used according to the judgment criteria.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and more specifically, to a frequency synchronization method based on a multi-phase frequency locked loop. Background Art

[0002] In modern wireless communication receivers, the frequency synchronization module is extremely important and directly determines the performance of the system. Frequency synchronization of coherent demodulation refers to generating a coherent carrier with the same frequency as the communication carrier at the wireless receiving end. Common frequency synchronization methods include direct methods such as phase-locked loop, M-th power loop, COSTAS loop, and pilot methods such as V&V, Kam, and L&R. Different algorithms have outstanding performance in low signal-to-noise ratio, large bandwidth, low latency, and fast convergence. Among them, the frequency-locked loop is a frequency synchronization technology suitable for high dynamics, large bandwidth, and low signal-to-noise ratio.

[0003] However, there is only a single frequency discriminator in the loop of a traditional frequency-locked loop, and its bandwidth and accuracy can only be controlled by changing the loop parameters, which has poor flexibility.

[0004] Therefore, the present invention provides a frequency synchronization method based on a multi-phase frequency locked loop, which improves the above technical problems. Summary of the invention

[0005] The disclosed embodiment aims to address the deficiencies of the prior art and provide a frequency synchronization method based on a multi-phase frequency-locked loop. The present invention adopts multiple frequency discriminators with different phases and precisions, and uses a parallel frequency-locked loop with different frequency discriminators through a judgment criterion, thereby solving the problem that there is only a single frequency discriminator in the loop of the frequency-locked loop, and its bandwidth and precision can only be changed by the number of loop parameters, and the flexibility is poor. At the same time, a phase tracking part is added to make up for the deficiency that the frequency-locked loop cannot track the phase.

[0006] The above technical purpose of the present invention is achieved through the following technical solutions: A frequency synchronization method based on a multi-phase frequency locked loop, comprising the following steps: S1, according to the frame structure, eliminating the load data, and sending the pilot signal alone to the differential link;

[0007] S2, averaging adjacent pilot symbols, then performing differences in sequence, and finally averaging the differences;

[0008] S3, obtaining the averaged differential value of the entire pilot frequency segment through a divider; dividing the real part of the differential value by the imaginary part to obtain a quotient;

[0009] S4. Obtain the arc tangent value of the quotient by looking up the atan table;

[0010] S5. According to the switching strategy, determine which differential parameter value to use; this value is then divided by the differential parameter to obtain the frequency value of the frequency synchronization estimate.

[0011] S7, obtaining the complex amplitude of the frequency by looking up the cosine table;

[0012] S8, storing the input data into the memory, and reading the data after the complex amplitude value of the frequency is updated;

[0013] S9. Multiply the data by the conjugate value of the estimated frequency value, thereby completing a frequency synchronization.

[0014] As a preferred technical solution of the present invention, the frame structure is composed of a frame header, an inserted pilot and data; the frame header includes a synchronization header segment and a unique code segment;

[0015] The synchronization header includes: a frame header synchronization word, BPSK modulation, a length of S symbols, and can complete bit / code synchronization;

[0016] The unique code segment includes: a frame header unique word, BPSK modulation, a length of U symbols, and completes frame synchronization;

[0017] The inserted pilot includes: a pilot, BPSK modulation, a length of P symbols, defined as 1, an insertion period of every D symbols, inserted after each data block, and completing frequency domain synchronization;

[0018] The data includes: data payload, each segment is D symbols long; protection interval: length is N symbols, no information.

[0019] As a preferred technical solution of the present invention, the process of averaging adjacent pilot symbols, then performing differentials in sequence, and finally averaging the differentials is as follows: assuming that the received wireless communication signal r with frequency deviation is represented as the sum of the communication signal s and Gaussian white noise n, the formula is:

[0020]

[0021] Where A is the complex envelope after modulation, f is the residual frequency deviation, is a random phase, n is a complex additive Gaussian white noise;

[0022] After time domain synchronization, the pilot signal after de-framing is obtained as:

[0023]

[0024] Among them, the complex envelope A of the pilot signal is a constant;

[0025] The discriminator takes the integral of the conjugate product of the data with a fixed phase difference C:

[0026] R(C)=∑ K r(k)×r(k+C) * =∑ K A 2exp(i2πfC)+∑ K N(k)

[0027] Where K is the integration length and N is the sum of the channel noise cross term and the noise square term.

[0028] As a preferred technical solution of the present invention, due to the properties of additive Gaussian white noise n, N can be constructed by two independent Gaussian white noises n, which are additive noises with zero mean but do not satisfy Gaussian distribution, expressed as:

[0029]

[0030] The frequency estimate is then:

[0031]

[0032] Based on the traditional discriminator, the multi-phase discriminator adopts multiple discriminator groups with different phase differences and completes the discriminator together. The multi-level phase difference C is 1, c, c2 symbols respectively, and we can get:

[0033] R(1)=∑ K r(k)×r(k+1) * =∑ K A 2 exp(i2πf)+∑ K N(k)

[0034] R(c)=∑ K r(k)×r(k+c) * =∑ K A 2 exp(i2πfc)+∑ K N(k)

[0035] R(c 2 )=∑ K r(k)×r(k+c 2 ) * =∑ K A 2 exp(i2πfc 2 )+∑ K N(k)

[0036] The estimated values ​​of frequency deviation can be obtained as follows:

[0037]

[0038] As a preferred technical solution of the present invention, according to the switching strategy, it is determined which differential parameter value to use; this value is then divided by the differential parameter to obtain the frequency value of the frequency synchronization estimation. The process is: define Arg[R(1)], Arg[R(c)] and Arg[R(c 2 )] is the discrimination angle a of different levels of formulas on the complex plane; c represents the differential distance; it can be seen that the value interval of the discrimination angle should be [-π, π], otherwise there will be discrimination ambiguity; due to the influence of noise, the actual discrimination angle value interval should be [-π+B, π-B], where B is the noise avoidance angle; in addition, the resolution angle or discrimination ambiguity angle A on the complex plane is defined, which means that under a certain signal-to-noise ratio, when the absolute value a of the discrimination angle calculated by the formula at this level is less than A, the accuracy of its estimated value cannot be further improved, that is, the formula at this level reaches its estimation accuracy or resolution;

[0039] The value of blur angle A is:

[0040] A≤π / c

[0041] Then, the value range of the demodulation angle is [-π+B, -A]∪[A, π-B].

[0042] As a preferred technical solution of the present invention, in practical applications, the received signal signal-to-noise ratio SNR determines the value of A, and the value of A determines the value of C; the lower the signal-to-noise ratio SNR, the larger A, the smaller C, and the slower the convergence of the estimated frequency counter value; therefore, the multi-phase frequency-locked loop determines the C value or the A value according to the signal-to-noise ratio, and comprehensively judges the switching of the formula and the output of the estimated value, as shown in the following formula:

[0043]

[0044] In summary, the present invention has the following beneficial effects:

[0045] First, on the basis of the traditional single discriminator of the frequency-locked loop, multiple discriminators with different phases and accuracies are adopted, and parallel frequency-locked loops with different discriminators are used according to the judgment criteria.

[0046] Secondly, the addition of a phase tracking part can make up for the deficiency of the frequency locked loop that cannot track the phase. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A flow chart of a frequency synchronization method based on a multi-phase frequency locked loop provided in an embodiment of the present invention;

[0048] Figure 2 A framework diagram of a waveform baseband frame structure provided by an embodiment of the present invention;

[0049] Figure 3 A flow chart of a physical layer demodulation process and a frequency synchronization method provided by an embodiment of the present invention;

[0050] Figure 4A schematic diagram of a multiple frequency-locked loop structure provided by an embodiment of the present invention;

[0051] Figure 5 A schematic diagram of a frequency discrimination interval of a multi-frequency frequency locked loop provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The present application is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present application, but do not limit the present application in any form. It should be noted that, for those of ordinary skill in the art, several variations and improvements can also be made without departing from the concept of the present application. These all belong to the protection scope of the present application.

[0053] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0054] It should be noted that, if there is no conflict, the various features in the embodiments of the present application can be combined with each other, all within the scope of protection of the present application. In addition, although the functional module division is performed in the device schematic diagram and the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in a sequence different from the module division in the device or the flow chart. In addition, the words "first", "second", "third", etc. used herein do not limit the data and execution order, but only distinguish the same items or similar items with basically the same functions and effects.

[0055] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used in this specification and in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The term "and / or" used in this specification includes any and all combinations of one or more of the related listed items.

[0056] 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.

[0057] The disclosed embodiment aims to solve the problem that there is only a single discriminator in the loop of the frequency locked loop, and its bandwidth and accuracy can only be changed by the number of loop parameters, and the flexibility is poor. In view of this, the disclosed embodiment proposes a frequency synchronization method based on a multi-phase frequency locked loop, which is based on the single discriminator of the traditional frequency locked loop, and adopts multiple discriminators with different phases and accuracies, and uses parallel frequency locked loops with different discriminators through judgment criteria. At the same time, a phase tracking part is added to make up for the deficiency that the frequency locked loop cannot track the phase.

[0058] Please refer to Figure 1 , Figure 1 The flowchart of the frequency synchronization method based on the multi-phase frequency locked loop described in the embodiment of the present disclosure is shown. The overall process mainly includes the following 9 steps:

[0059] S1. Deframe: according to the frame structure, remove the payload data and send the pilot signal alone to the differential link.

[0060] The wireless communication system uses a pilot-inserted frame structure for communication, which is used for burst non-continuous channels. Its frame structure is mainly divided into three parts: frame header (synchronization header and unique code), inserted pilot, and data, as follows:

[0061] like Figure 2 As shown, the synchronization header segment: frame header synchronization word, BPSK modulation, length S symbols, can complete bit / code synchronization; unique code segment: frame header unique word, BPSK modulation, length U symbols, complete frame synchronization; pilot segment: pilot, BPSK modulation, length is P symbols, defined as 1, insertion period is every D symbols, inserted after each data block, complete frequency domain synchronization; data segment: data payload, each segment length is D symbols; protection interval: length is N symbols, no information.

[0062] like Figure 3 As shown in the figure, frequency synchronization is the synchronization of the receiver in the frequency domain after completing the time domain synchronization (symbol synchronization and frame synchronization). It mainly uses the pilot information inserted at intervals in the data segment to estimate the frequency deviation value in the received signal. In order to take into account the estimation range and accuracy of the frequency deviation, frequency synchronization needs to go through multiple iterations (coarse synchronization, fine estimation, and precise estimation), and adjust the algorithm parameters in real time, so as to obtain a larger estimation range and higher estimation accuracy.

[0063] Among them, coarse frequency estimation is to make a preliminary estimate of the frequency when the frequency deviation is large, generally greater than 10% of the symbol rate. The algorithm of coarse estimation requires a large estimation range and low estimation accuracy. The main algorithms include spectral estimation, Fourier transform, etc.; fine frequency estimation is to make further estimation of the frequency after the rough estimation, generally less than 10% of the symbol rate. The algorithm of fine estimation requires both estimation range and estimation accuracy. The main algorithms include M&M, V&V, Kam algorithm, etc.; fine estimation is the final estimation to meet the demodulation requirements. The fine estimation requires extremely high accuracy. The main algorithms include L&R, frequency-locked loop / phase-locked loop, etc.

[0064] S2, differential, is the main link of the multi-frequency locked loop, which consists of multiple parallel differential links with different parameters. Among them, 1 differential is to differentiate two adjacent pilot symbols and average them after differentiation; c differential is to average c adjacent pilot symbols, then differentiate them in turn, and finally average the differences; c ^ 2 The difference is the same as above. The average value is realized by IIR filter.

[0065] S3, divider, the average difference value of the entire pilot band, the real part is divided by the imaginary part to obtain the quotient.

[0066] S4, atan table, obtain the arc tangent value of the quotient by looking up the table. This value is divided by the difference parameter (1, c, c ^ 2), which is the frequency value estimated by frequency synchronization.

[0067] The multi-phase frequency-locked loop is linked by multiple levels of frequency discriminators. The multi-phase frequency-locked loop has the characteristics of simple implementation, high dynamic, fast convergence, high accuracy and low signal-to-noise ratio requirements. Its structure, such as Figure 4 shown.

[0068] The multi-phase frequency-locked loop is a data-assisted frequency synchronization method. After the input data undergoes symbol (bit) synchronization and frame synchronization, the pilot information in the frame is obtained and sent to multiple frequency discriminators for calculation to obtain an estimated value of the frequency.

[0069] Assume that the received wireless communication signal r with frequency deviation can be simply expressed as the sum of the communication signal s and Gaussian white noise n:

[0070]

[0071] Where A is the complex envelope after modulation, f is the residual frequency deviation, is a random phase, and n is a complex additive Gaussian white noise.

[0072] After time domain synchronization, the pilot signal after de-framing is obtained as:

[0073]

[0074] Among them, the complex envelope A of the pilot signal is a constant.

[0075] The discriminator takes the integral of the conjugate product of the data with a fixed phase difference C:

[0076] R(C)=∑ K r(k)×r(k+C) * =∑ K A 2 exp(i2πfC)+∑k K N(k)

[0077] Where K is the integration length and N is the sum of the channel noise cross term and the noise square term.

[0078] Considering the properties of additive Gaussian white noise n, N can be constructed by two independent Gaussian white noises n, which are additive noises with zero mean but do not satisfy the Gaussian distribution, expressed as:

[0079]

[0080] The frequency estimate is

[0081]

[0082] Based on the traditional discriminator, the multi-phase discriminator adopts multiple discriminator groups with different phase differences and completes the discriminator together. The multi-level phase difference C is 1, c, c2 symbols, etc., and we can get:

[0083] R(1)=∑ K r(k)×r(k+1) * =∑ K A 2 exp(i2πf)+∑ K N(k)

[0084] R(c)=∑ K r(k)×r(k+c) * =∑ K A 2 exp(i2πfc)+∑ K N(k)

[0085] R(c 2 )=∑ K r(k)×r(k+c 2 ) * =∑ K A 2 exp(i2πfc 2 )+∑ K N(k)

[0086] The estimated values ​​of frequency deviation are

[0087]

[0088] In the frequency-locked loop application, the above frequency estimation formulas have different estimation ranges and precisions. From top to bottom, the estimation range is getting smaller and the estimation precision is getting higher. Therefore, the multi-phase frequency-locked loop cannot use the above estimation formulas at the same time, but samples different estimation formulas at different estimation stages. The specific estimation strategy is to use an estimation formula with a large estimation range and low estimation precision at the beginning of the estimation. When the estimation formula reaches its estimation precision, replace it with the next level of estimation formula with a small estimation range and high estimation precision, and so on.

[0089] S5. According to the switching strategy, determine which differential parameter value to use; this value is then divided by the differential parameter to obtain the frequency value of the frequency synchronization estimate.

[0090] The switching strategy for formulas with different estimation ranges and estimation accuracy is: when the current formula reaches its estimation accuracy, switch to the next formula with higher estimation accuracy and narrower estimation range. The estimation accuracy of each formula is related to the signal-to-noise ratio, differential distance c, and integration time length K. When switching, the estimation range of the next formula is required to be slightly larger than the estimation accuracy of the previous level.

[0091] Define Arg[R(1)], Arg[R(c)], and Arg[R(c 2 )] are the discrimination angles a of different levels of formulas on the complex plane. It can be seen that the value interval of the discrimination angle should be [-π, π], otherwise discrimination ambiguity will occur. Taking into account the influence of noise, the actual discrimination angle value interval should be [-π+B, π-B], where B is the noise avoidance angle. In addition, the resolution angle or discrimination ambiguity angle A on the complex plane is defined. Its meaning is that under a certain signal-to-noise ratio, when the absolute value a of the discrimination angle calculated by the formula at this level is less than A, the accuracy of its estimated value cannot be further improved, that is, the formula at this level reaches its estimation accuracy or resolution. Ideally, the ambiguity angle A takes the value:

[0092] A≤π / c

[0093] In summary, the value interval of the demodulation angle should be [-π+B, -A]∪[A, π-B], as shown in Figure 5.

[0094] In practical applications, the received signal signal-to-noise ratio (SNR) determines the value of A, which in turn determines the value of C. The lower the signal-to-noise ratio (SNR), the larger A and the smaller C, and the slower the convergence of the estimated frequency counter value. Therefore, the multi-phase frequency-locked loop must determine the C value or A value based on the signal-to-noise ratio, and make a comprehensive judgment to switch the formula and output the estimated value, as shown in the following formula.

[0095]

[0096] S6. Integrate, accumulate the estimated frequency values ​​according to the sampling points to obtain the phase value of the frequency.

[0097] S7, cosine table, obtain the complex amplitude of the frequency by looking up the table.

[0098] S8. The memory stores the input data (including the payload and the pilot) into a cache, and reads out the data after the complex amplitude value of the frequency is updated.

[0099] In order to obtain higher implementation accuracy, the implementation of the multi-phase frequency-locked loop adopts a multiple-iteration structure and a single-iteration implementation architecture, as shown in Figure 1. Different iterations use differential results of different parameters (1, c, c^2).

[0100] After the frequency synchronization data enters the multi-frequency locked loop, it is divided into two paths. One path enters the memory for delayed storage; the other path is deframed to extract the pilot information. The pilot information is differentiated according to different parameters, and then averaged according to each frame. The amplitude and angle value of the differential signal is obtained through the inverse tangent function, and the estimated value of the frequency is obtained at this time.

[0101] S9, frequency compensation, multiply the data by the conjugate value of the estimated frequency value to complete a frequency synchronization.

[0102] Each time synchronization is performed, the differential parameter becomes larger and larger to improve the estimation accuracy. Therefore, the frequency synchronization process is continuously iterated according to the system accuracy requirements until the system requirements are met.

[0103] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A frequency synchronization method based on a multi-phase frequency locked loop, characterized in that: The method comprises the following steps: S1. According to the frame structure, the payload data is removed and the pilot signal is sent to the differential link alone; S2, averaging adjacent pilot symbols, then performing differences in sequence, and finally averaging the differences; S3, obtaining the averaged differential value of the entire pilot frequency segment through a divider; dividing the real part of the differential value by the imaginary part to obtain a quotient; S4. Obtain the arc tangent value of the quotient by looking up the atan table; S5. According to the switching strategy, determine which differential parameter value to use; this value is then divided by the differential parameter to obtain the frequency value of the frequency synchronization estimate. S7, obtaining the complex amplitude of the frequency by looking up the cosine table; S8, storing the input data into the memory, and reading the data after the complex amplitude value of the frequency is updated; S9, multiplying the data by the conjugate value of the estimated frequency value, thus completing a frequency synchronization; The frame structure consists of a frame header, an inserted pilot and data; the frame header includes a synchronization header segment and a unique code segment; The synchronization header segment includes: a frame header synchronization word, BPSK modulation, a length of S symbols, and can complete bit / code synchronization; The unique code segment includes: a frame header unique word, BPSK modulation, a length of U symbols, and completes frame synchronization; The inserted pilot includes: a pilot, BPSK modulation, a length of P symbols, defined as 1, an insertion period of every D symbols, inserted after each data block, and completing frequency domain synchronization; The data includes: data payload, each segment is D symbols long; guard interval: N symbols long, no information; The process of averaging adjacent pilot symbols, then performing differentiation in sequence, and finally averaging the differences is as follows: assuming that the received wireless communication signal r with frequency deviation is represented as the sum of the communication signal s and Gaussian white noise n, the formula is: Where A is the complex envelope after modulation, f is the residual frequency deviation, is a random phase, n is a complex additive Gaussian white noise; After time domain synchronization, the pilot signal after de-framing is obtained as: Among them, the complex envelope A of the pilot signal is a constant; The discriminator takes the integral of the conjugate product of the data with a fixed phase difference C: R(C)=Σ K r(k)×r(k+C) * =S K A 2 exp(i2πfC)+Σ K N(k) Where K is the integration length, N is the sum of the channel noise cross term and the noise square term; Due to the properties of additive Gaussian white noise n, N can be constructed from two independent Gaussian white noises n, which are additive noises with zero mean but do not satisfy the Gaussian distribution, expressed as: The frequency estimate is then: Based on the traditional discriminator, the multi-phase discriminator adopts multiple discriminator groups with different phase differences and completes the discriminator together. The multi-level phase difference C is 1, c, c2 symbols respectively, and we can get: R(1)=Σ K r(k)×r(k+1) * =∑ K A 2 exp(i2πf)+∑ K N(k) R(c)=Σ K r(k)×r(k+c) * =∑ K A 2 exp(i2πfc)+∑ K N(k) R(c 2 )=∑ K r(k)×r(k+c 2 ) * =∑ K A 2 exp(i2πfc 2 )+∑ K N(k) The estimated values ​​of frequency deviation can be obtained as follows: According to the switching strategy, determine which differential parameter value to use; this value is then divided by the differential parameter to obtain the frequency value of the frequency synchronization estimate. The process is: define Arg[R(1)], Arg[R(c)] and Arg[R(c 2 )] is the discrimination angle a of different levels of formulas on the complex plane; c represents the differential distance; it can be seen that the value interval of the discrimination angle should be [-π, π], otherwise there will be discrimination ambiguity; due to the influence of noise, the actual discrimination angle value interval should be [-π+B, π-B], where B is the noise avoidance angle; in addition, the resolution angle or discrimination ambiguity angle A on the complex plane is defined, which means that under a certain signal-to-noise ratio, when the absolute value of the discrimination angle calculated by the formula at this level |a| is less than A, the accuracy of its estimated value cannot be further improved, that is, the formula at this level reaches its estimation accuracy or resolution; The value of blur angle A is: A≤π / c; Then, the value range of the demodulation angle is [-π+B, -A]∪[A, π-B]; The received signal signal-to-noise ratio (SNR) determines the value of A, which in turn determines the value of C. The lower the signal-to-noise ratio (SNR), the larger A and the smaller C, and the slower the convergence of the estimated frequency counter value. Therefore, the multi-phase frequency-locked loop determines the C value or A value based on the signal-to-noise ratio, and makes a comprehensive judgment to switch the formula and output the estimated value, as shown in the following formula: