Wireless signal blind demodulation method and device

Abstract

The invention belongs to the technical field of signal processing, and provides a blind demodulation method and device for a wireless signal, and the method comprises the steps: screening out a candidate modulation type from a plurality of modulation types according to the time domain features in the wireless signal; for each candidate modulation type, estimating a target parameter of the IQ data in the wireless signal corresponding to the candidate modulation type to obtain a target parameter estimation result; performing symbol synchronization processing on the target parameter estimation result to obtain a symbol synchronization point; calculating EVM (Error Vector Magnitude) corresponding to the candidate modulation type according to the symbol synchronization point, and determining the candidate modulation type corresponding to a minimum value as a target modulation type under the condition that the minimum value in the EVM corresponding to each candidate modulation type is smaller than or equal to an EVM threshold value; and demodulating the wireless signal based on the target modulation type to obtain a demodulation result. According to the method provided by the invention, the blind demodulation efficiency and accuracy of the composite modulation signal in a complex channel or a low signal-to-noise ratio environment are improved.

Description

Technical Field

[0001] The present invention relates to the field of signal processing technology, and in particular to a blind demodulation method and device for wireless signals. Background Technology

[0002] With the continuous development of modern wireless communication technology, the identification and demodulation of digital modulation signals have become key technologies in the fields of non-cooperative communication, spectrum regulation and signal monitoring. The types and complexity of modulation methods have gradually increased, including ASK, PSK, QAM, FSK and other commonly used signals and composite modulation signals; in complex communication environments, how to accurately identify the modulation method and efficiently realize signal demodulation is a major challenge in the current technical field.

[0003] The traditional demodulation method of digital modulation signals relies on prior parameter information, but in actual scenarios, the parameters of the signal are usually unknown and greatly affected by environmental interference. Therefore, the existing technology has emerged through blind demodulation technology, which can realize the identification and demodulation of unknown parameter signals by analyzing and estimating signal characteristics.

[0004] In the related technology, the existing technology extracts time-frequency features from the signal and combines the maximum likelihood estimation to realize the rough estimation of the code rate, completes the blind demodulation of the FSK signal, identifies the signal frequency through the bandwidth scanning strategy, and completes the blind demodulation of the OOK and 2FSK signals by combining the amplitude change and frequency deviation information. The frequency deviation and symbol rate estimation method of the multi-square spectrum and wavelet transform is used for MPSK and MQAM under low signal-to-noise ratio conditions. The above blind demodulation method is for identifying one or two modulation types, and when the universal signal parameters are used to blindly demodulate the composite modulated signal containing two or more modulation modes, the demodulation accuracy and efficiency are low, which makes it difficult to meet the blind demodulation requirements of the composite modulated signal in complex channels or low signal-to-noise ratio environments. SUMMARY OF THE INVENTION

[0005] The present invention provides a blind demodulation method and device for wireless signals, which is used to solve the defects of low accuracy and efficiency in blind demodulation of composite modulated signals by using universal signal parameters, which makes it difficult to meet the blind demodulation requirements of composite modulated signals in complex channels or low signal-to-noise ratio environments, and improves the blind demodulation efficiency and accuracy of composite modulated signals in complex channels or low signal-to-noise ratio environments.

[0006] The present invention provides a blind demodulation method for wireless signals, comprising: Selecting candidate modulation types from a plurality of modulation types according to time domain characteristics in the wireless signal; wherein the candidate modulation types include at least two of 2 / 4ASK, 2 / 4PSK, 4π / DQPSK, 8PSK, OQPSK, π / 4QPSK, 2 / 4 / 8FSK, 8 / 16 / 32 / 64QAM, MSK and 16APSK; For each candidate modulation type, estimate the target parameter of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type to obtain the target parameter estimation result; the target parameter includes at least one of the symbol rate, frequency deviation and phase deviation; Perform symbol synchronization processing on the target parameter estimation result to obtain a symbol synchronization point; calculate the EVM corresponding to the candidate modulation type according to the symbol synchronization point, and when the minimum value of the EVM corresponding to each candidate modulation type is less than or equal to the EVM threshold, determine that the candidate modulation type corresponding to the minimum value is the target modulation type; demodulate the wireless signal based on the target modulation type to obtain a demodulation result.

[0007] According to a blind demodulation method for wireless signals provided by the present invention, the time domain feature includes the envelope flatness of the wireless signal; The step of selecting candidate modulation types from a plurality of modulation types according to the time domain characteristics in the wireless signal includes: Determining the envelope flatness according to the sampling rate and time domain distribution data of the wireless signal; Compare the envelope flatness with the time domain feature decision thresholds corresponding to the multiple modulation types to determine the candidate modulation type.

[0008] According to a blind demodulation method for wireless signals provided by the present invention, the target parameter includes a symbol rate; The target parameter of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type is estimated to obtain the target parameter estimation result, which includes: Perform at least one of differential envelope normalization, instantaneous frequency envelope normalization and quadratic power spectrum processing on the IQ data to obtain a symbol rate estimation result.

[0009] According to a blind demodulation method for wireless signals provided by the present invention, the target parameter further includes frequency deviation; After obtaining the symbol rate estimation result, the method further includes: According to the symbol rate estimation result, the IQ data is processed by at least one of the centroid method, the maximum spectrum value method and the local frequency analysis method to obtain a frequency offset estimation result.

[0010] According to a blind demodulation method for wireless signals provided by the present invention, the target parameter also includes phase deviation; After obtaining the frequency offset estimation result, the method further includes: According to the frequency offset estimation result, the IQ data is processed by at least one of a multi-power summation method and a phase difference method to obtain a phase offset estimation result.

[0011] According to a blind demodulation method for wireless signals provided by the present invention, after obtaining the symbol rate estimation result, the method further includes: When the FFT accuracy corresponding to the symbol rate estimation result is lower than the FFT accuracy threshold, the IQ data is subjected to local discrete Fourier transform DFT processing according to the spectrum position corresponding to the symbol rate estimation result to obtain a refined symbol rate estimation result.

[0012] According to a blind demodulation method for wireless signals provided by the present invention, after obtaining the frequency offset estimation result, the method further includes: When the FFT accuracy corresponding to the frequency offset estimation result is lower than the FFT accuracy threshold, the local refined power spectrum of the IQ data is calculated according to the frequency spectrum position corresponding to the frequency offset estimation result to obtain a refined frequency offset estimation result.

[0013] The present invention also provides a blind demodulation device for wireless signals, comprising: A screening module, used for screening candidate modulation types from a plurality of modulation types according to time domain characteristics in a wireless signal; wherein the candidate modulation types include at least two of 2 / 4ASK, 2 / 4PSK, 4π / DQPSK, 8PSK, OQPSK, π / 4QPSK, 2 / 4 / 8FSK, 8 / 16 / 32 / 64QAM, MSK and 16APSK; A parameter estimation module is used to estimate the target parameters of the in-phase orthogonal IQ data in the wireless signal corresponding to each candidate modulation type, and obtain the target parameter estimation result; the target parameter includes at least one of the symbol rate, frequency deviation and phase deviation; A demodulation module is used to perform symbol synchronization processing on the target parameter estimation result to obtain a symbol synchronization point; calculate the EVM corresponding to the candidate modulation type according to the symbol synchronization point, and determine that the candidate modulation type corresponding to the minimum value is the target modulation type when the minimum value among the EVMs corresponding to the candidate modulation types is less than or equal to the EVM threshold; demodulate the wireless signal based on the target modulation type to obtain a demodulation result.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the above-mentioned blind demodulation methods for wireless signals when executing the computer program.

[0015] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the blind demodulation method of a wireless signal as described in any one of the above is implemented.

[0016] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned blind demodulation methods for wireless signals.

[0017] The blind demodulation method and device of wireless signals provided by the present invention select candidate modulation types from multiple modulation types through the time domain characteristics in the wireless signal, estimate the target parameters of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type for each candidate modulation type, perform symbol synchronization processing on the estimation result and calculate the EVM corresponding to the candidate modulation type, and demodulate the wireless signal according to the candidate modulation type corresponding to the minimum EVM to obtain the demodulation result, thereby improving the blind demodulation efficiency and accuracy of composite modulation signals in complex channels or low signal-to-noise ratio environments. Brief Description of the Figures

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 This is one of the flow charts of the blind demodulation method of wireless signals provided by the present invention.

[0020] Figure 2 This is the second flow chart of the blind demodulation method of wireless signals provided by the present invention.

[0021] Figure 3 This is a schematic diagram of the structure of the blind demodulation device for wireless signals provided by the present invention.

[0022] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Specific implementation method

[0023] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] The following combination Figure 1-Figure 3 Describe the blind demodulation method and device of wireless signals of the present invention.

[0025] Figure 1 is one of the flow charts of the blind demodulation method of wireless signals provided by the present invention, such as Figure 1 As shown, the method includes the following: Step 110: Filter out candidate modulation types from multiple modulation types according to the time domain characteristics of the wireless signal; wherein the candidate modulation types include at least two of 2 / 4ASK, 2 / 4PSK, 4π / DQPSK, 8PSK, OQPSK, π / 4QPSK, 2 / 4 / 8FSK, 8 / 16 / 32 / 64QAM, MSK and 16APSK.

[0026] In this step, the time domain feature may be the envelope flatness (iqR) of the wireless signal.

[0027] In this embodiment, the modulation modes that do not meet the characteristics are screened out from multiple modulation types by calculating the iqR value corresponding to the wireless signal, thereby narrowing the range of possible modulation types; the decision interval of iqR can also be used to classify the candidate modulation types into non-AM, AM and other specific signal types.

[0028] In this embodiment, the candidate modulation types include two or more of 2ASK, 4ASK, 2PSK, 4PSK, 4πQPSK, DQPSK, 8PSK, OQPSK, π / 4QPSK, 2FSK, 4FSK, 8FSK, 8QAM, 16QAM, 32QAM, 64QAM, MSK and 16APSK; when there is only one candidate modulation type obtained by the above-mentioned time domain feature screening, the candidate modulation type can be directly used to demodulate the wireless signal.

[0029] Step 120: For each candidate modulation type, estimate the target parameters of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type to obtain a target parameter estimation result; the target parameter includes at least one of the symbol rate, frequency deviation and phase deviation.

[0030] In this step, the IQ data of the wireless signal may contain multiple unknown parameters, including symbol rate Rs, oversampling multiple, shaping filter used, frequency deviation, phase deviation, modulation mode or constellation diagram, etc., that is, the IQ data cannot be demodulated directly, and it is necessary to blindly estimate the unknown parameters and identify the modulation mode used.

[0031] In this embodiment, for different signal modulation types, one or more target parameters can be used for estimation and identification.

[0032] For example, for different modulation modes, differential envelope normalization (identification of MASK, MFSK, MQAM, 16APSK), instantaneous frequency envelope normalization (identification of MPSK) or quadratic power spectrum (identification of OQPSK, MSK, DQPSK, π / 4QPSK) and other methods are used to estimate the symbol rate of the signal to improve the accuracy of symbol rate estimation.

[0033] For example, for different modulation modes, combined with the characteristics of each modulation type, differentiated frequency offset estimation methods are used, such as the centroid method (identifying MFSK, M is the bit-to-symbol mapping number, M=2,4,8), the spectrum maximum method (identifying MASK, M is the bit-to-symbol mapping number), local frequency analysis (identifying MPSK, MASK, π / 4QPSK, DQPSK, 16APSK and 8 / 16 / 32 / 64QAM), etc., to obtain the frequency offset value of the signal; at the same time, based on the rough estimation, local refined analysis is performed to improve the accuracy of frequency offset estimation.

[0034] For example, for different modulation modes, use phase estimation algorithms suitable for each modulation mode, such as multi-power summation (identifying MSK, MPSK, π / 4QPSK, DQPSK, 16APSK and 8 / 16 / 32 / 64QAM), phase difference (identifying MFSK and MASK), etc., to calculate the phase deviation value of the signal, and then estimate the phase deviation of the signal after frequency deviation compensation according to the initial phase of the modulation mode, and further optimize the phase deviation estimation result.

[0035] Step 130, perform symbol synchronization processing on the target parameter estimation result to obtain a symbol synchronization point; calculate the EVM corresponding to the candidate modulation type according to the symbol synchronization point, and when the minimum value of the EVM corresponding to each candidate modulation type is less than or equal to the EVM threshold, determine the candidate modulation type corresponding to the minimum value as the target modulation type; demodulate the wireless signal based on the target modulation type to obtain a demodulation result.

[0036] In this step, the value of the EVM threshold may be related to the range of parameters such as the signal-to-noise ratio, frequency deviation, and phase deviation of the unknown parameter signal; the EVM threshold may also be dynamically adjusted according to application requirements and parameter ranges, for example, the EVM threshold is 10%.

[0037] In this embodiment, the error vector magnitude (EVM) refers to the difference between a theoretical waveform and an actual received waveform. The value of EVM is the root mean square value of the ratio of the average error vector signal power to the average reference signal power. EVM is used to indicate the quality of the wireless system signal. The calculation formula of EVM is as follows: ; Ierr=Iref-Imeas; Qerr=Qref-Qmeas; Wherein, Ierr is the difference between the in-phase component (I component) of the received signal and the in-phase component of the ideal signal, Qerr is the difference between the quadrature component (Q component) of the received signal and the quadrature component of the ideal signal; Iref and Qref are the reference point coordinates of the constellation diagrams of different modulation modes, N is the number of data constellation points, Imeas and Qmeas are the symbol values ​​obtained by decimating the demodulation results of different modulation methods by sps (up sampling multiple) (sps=fs / Rs), n is the sampling point number or serial number.

[0038] In this embodiment, symbol synchronization is performed based on the EVM values ​​calculated from the oversampling points and the reference constellation points, and the symbol corresponding to the smallest EVM value is selected as the starting synchronization point for demodulation.

[0039] Specifically, after obtaining the symbol rate (code rate), determine whether the decimation multiple sps (sps=fs / Rs) is an integer. If not, resample the IQ data according to the configuration of sps=8 or sps=16 so that the value of the new sampling rate fs_new divided by Rs is an integer sps=8 or sps=16.

[0040] In this embodiment, for modulation modes within the traversal range, when a certain modulation mode is traversed, the corresponding demodulation mode is used to demodulate and obtain the corresponding symbol; for amplitude modulation type modulation, the demodulation result is a code element representing amplitude mapping; for frequency hopping type results, the demodulation result is a code element representing frequency mapping; for phase modulation type results, the demodulation result is a code element representing phase and amplitude mapping.

[0041] In this embodiment, based on the demodulation result, the synchronization point is selected by oversampling multiples and symbol alignment, combined with the error vector magnitude (EVM) characteristics, and the optimal synchronization point is determined according to the minimum EVM value to complete symbol synchronization.

[0042] Specifically, when symbol synchronization is performed on the demodulation result, for the case of sps=8 or sps=16, the 1st to 8th or 1st to 16th data of the demodulation result is used as the starting data, and sps is used as the decimation multiple to extract the demodulation result to obtain the demodulated data after the sps group extraction; for each group of demodulation data, the EVM value of each group is calculated according to the above EVM calculation formula, and the group of data with the smallest EVM value is selected, and its starting point is the best synchronization point.

[0043] In this embodiment, for the modulation modes within the traversal range, each modulation mode can obtain the EVM value corresponding to its best synchronization point, and then select the value with the minimum EVM and compare it with EVM_TH (preset EVM threshold). If the minimum EVM value is less than EVM_TH, the modulation mode corresponding to the minimum EVM value is output as the correct result of the entire blind demodulation process. The output demodulation result includes parameters such as modulation mode, symbol rate, sps, frequency deviation, phase deviation, etc., and the blind demodulation process for an IQ signal with unknown parameters is successfully completed; in addition, if the minimum EVM value is greater than EVM_TH, it means that this blind demodulation process has failed.

[0044] The blind demodulation method for wireless signals provided in the embodiment of the present invention selects candidate modulation types from multiple modulation types through the time domain characteristics in the wireless signal, estimates the target parameters of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type for each candidate modulation type, performs symbol synchronization processing on the estimation result and calculates the EVM corresponding to the candidate modulation type, and demodulates the wireless signal according to the candidate modulation type corresponding to the minimum EVM to obtain a demodulation result, thereby improving the blind demodulation efficiency and accuracy of composite modulation signals in complex channels or low signal-to-noise ratio environments.

[0045] In some embodiments, the time domain feature includes the envelope flatness of the wireless signal; screening candidate modulation types from multiple modulation types according to the time domain feature in the wireless signal includes: (1) Determine the envelope flatness based on the sampling rate and time domain distribution data of the wireless signal.

[0046] In this embodiment, for the in-phase orthogonal IQ data of the wireless signal, the sampling rate and time domain data are known. The signal can be first screened within a certain range according to the characteristics of the time domain data distribution, and then the modulation modes that do not meet the characteristic parameters are eliminated according to the calculated characteristic parameters, and the remaining modulation modes are traversed by blind demodulation; the envelope flatness is calculated as follows: Calculate signal envelope Variance of , the mean is , that is: ; Remove the constant term, and we have: ; Wherein, iqR is the characteristic parameter of the flatness of the signal envelope, is the mean of the envelope square, is the square of the envelope mean, and the minimum value of iqR is 1; the smaller the value of R, the smoother the fluctuation of the signal envelope distribution, and the more likely it is a non-amplitude modulation signal; the larger the value of iqR, the steeper the fluctuation of the signal envelope distribution, and the more likely it is an amplitude modulation signal.

[0047] (2) Compare the envelope flatness with the time domain feature decision thresholds corresponding to various modulation types to determine the candidate modulation type.

[0048] In this embodiment, the time domain feature decision thresholds corresponding to 2 / 4ASK, 2 / 4PSK, 4π / 4PSK, 8PSK, OQPSK, π / 4DQPSK, 2 / 4 / 8 / FSK, 8 / 16 / 32 / 64QAM, MSK and 16APSK are shown in Table 1 below: Table 1. iqR decision interval parameter table

[0049] The blind demodulation method for wireless signals provided by the embodiment of the present invention performs range screening on input signals of unknown modulation types by extracting the envelope flatness of the signal, thereby obtaining a smaller possible range of modulation modes, avoiding blind demodulation of all demodulation modes during blind demodulation, and improving the blind demodulation efficiency of composite modulated signals.

[0050] In some embodiments, the target parameter includes a symbol rate; estimating the target parameter of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type to obtain the target parameter estimation result includes: performing at least one of differential envelope normalization, instantaneous frequency envelope normalization and quadratic power spectrum processing on the IQ data to obtain the symbol rate estimation result.

[0051] ​​​In this embodiment, by performing differential envelope normalization processing on the IQ data, the maximum value of the differential envelope normalized DC-free spectrum is obtained, and the candidate modulation types of MASK, MFSK, MQAM, and 16APSK can be identified; the differential envelope normalization method is specifically implemented by the following steps: (1) Obtaining IQ data from wireless signals, which may contain negative values; (2) To differentiate the IQ data, in order to ensure that the length of the data after differentiation is consistent with the length of the data before differentiation, a point can be added to the front of the data after differentiation. This point can be the first point of the data after differentiation, which is equivalent to the first two points of the data after differentiation being equal in value; (3) Take the absolute value of the differential IQ data, i.e. the envelope; (4) Calculate the ratio of the data sequence containing the symbol rate after taking the absolute value to the maximum value of the data, and normalize the ratio so that the ratio result is in the interval of 0~1; (5) Subtract the mean of the normalized data from the normalized data and perform DC removal to prevent the DC component from being too large and affecting the symbol rate estimation; (6) After performing FFT of the data length points on the DC-removed data to obtain the FFT result, search for the frequency value corresponding to the maximum amplitude on the non-negative frequency axis, which is the roughly estimated symbol rate value.

[0052] In this embodiment, the instantaneous frequency envelope normalization processing is performed on the IQ data to obtain the maximum value of the instantaneous frequency envelope normalized DC-free spectrum, so that the MPSK in the candidate modulation types can be identified; the instantaneous frequency envelope normalization method is specifically implemented by the following steps: Assume that an IQ complex signal of N points with a sampling rate of fs is received, and the arc tangent of the signal in the time domain of the N-point signal is calculated to obtain the discontinuous phase sequence ang in the range [-π, +π] and the difference between the previous and next phases ; Since the process of solving the instantaneous frequency involves two differential (post-subtraction) operations, it can be simplified into one operation when applied to engineering: ; Among them, ; then , and there are: ; Among them, is the instantaneous frequency; the implementation method of the above differential envelope normalization method is the same, the differential operation is replaced by the instantaneous frequency, that is, the instantaneous frequency is obtained by using the phase difference method for the data, and then the above steps (3) to (6) are performed to obtain the maximum value of the instantaneous frequency envelope normalized DC-free spectrum. ​

[0053] In this embodiment, the signal square power spectrum is calculated using IQ data, and OQPSK, MSK, DQPSK and π / 4QPSK among the candidate modulation types can be identified.

[0054] Specifically, for OQPSK recognition, the calculation method is as follows: (1), first calculate the power spectrum information of the signal corresponding to the square of the IQ data; (2) Determine the maximum value of the power spectrum signal of the second power and set the decision threshold to 95% of its maximum value; (3) Search the power spectrum from left to right and from right to left respectively for the first local maximum frequency value that satisfies the decision threshold set in (2) (after finding the first frequency value exceeding the threshold value, the process of searching for the local maximum value within the search window length), denoted as f1 and f2; (4) The rough estimate of the final symbol rate is the average of the absolute values ​​of the two frequency values ​​f1 and f2 found in (3).

[0055] Specifically, for the identification of MSK, DQPSK and π / 4QPSK, the calculation method is as follows: (1) First, find the power spectrum information of the square of IQ data corresponding to each modulation type; (2) Search for the first and second largest frequency values ​​of the power spectrum in (1) (these two maximum values ​​exclude the influence of adjacent points, that is, after finding the first largest frequency value, exclude other larger value points within the search window length, and then search for the "relative" second largest frequency value), denoted as f1 and f2; (3) Take the absolute values ​​of the first and second largest frequency values ​​f1 and f2 found in step (2) and add them together to obtain a rough estimate of the symbol rate.

[0056] The blind demodulation method for wireless signals provided in the embodiment of the present invention obtains a symbol rate estimation result by performing differential envelope normalization, instantaneous frequency envelope normalization and quadratic power spectrum processing on IQ data, thereby improving the accuracy of symbol rate estimation.

[0057] In some embodiments, the target parameter also includes frequency deviation; after obtaining the symbol rate estimation result, the blind demodulation method of the wireless signal also includes: performing at least one of the centroid method, the maximum spectrum value method and the local frequency analysis method on the IQ data according to the symbol rate estimation result to obtain the frequency deviation estimation result.

[0058] In this embodiment, based on the above symbol rate estimation result, the frequency offset of IQ data is estimated by the centroid method, so that MFSK (M=2,4,8) in the candidate modulation types can be identified; the centroid method is specifically implemented by the following steps: Assume that the frequency axis sequence of the power spectrum of the IQ data is [f(0), f(1), ..., f(N-1)], and the corresponding y-axis linear power value sequence (i.e., the result of taking the modulus / N and then squaring it after FFT, and then fftshifting it) is [y(0), y(1), ..., y(N-1)], then the centroid of the power spectrum It can be expressed by the following formula: ; In this embodiment, in order to simplify the calculation, the frequency axis sequence f(n) of the power spectrum can be replaced by the value from 0 to N-1 and brought into the above formula to calculate the centroid. The calculated result is rounded to the nearest integer to determine which index value from 0 to N-1 it is , and then use this index value to correspond to f(n), then the index value corresponds to is .

[0059] In this embodiment, based on the above symbol rate estimation result, the frequency offset estimation is performed on the IQ data by using the maximum spectrum value method or the local frequency analysis method, and the following modulation types in the candidate modulation types can be identified: MASK, MPSK, MASK, π / 4QPSK, DQPSK, 16APSK, 8QAM or 16 / 32 / 64QAM can be identified; the specific implementation is through the following steps: For MASK type signals, the frequency value corresponding to the maximum value of the signal's first power spectrum is its frequency deviation estimation value.

[0060] For MSK signals, first calculate the sum of the maximum values ​​of the positive and negative semi-axes of the signal's square spectrum, and then divide it by (2×2). The obtained frequency value is the frequency deviation estimate.

[0061] For MPSK and OQPSK signals, the frequency value corresponding to the maximum value of the signal's Mth power spectrum is divided by M, and the resulting frequency value is the frequency deviation estimate.

[0062] For π / 4QPSK and DQPSK signals, first calculate the sum of the maximum values ​​of the positive and negative semi-axes of the signal's square spectrum, and then divide it by (2×2). The obtained frequency value is the frequency deviation estimate.

[0063] In this embodiment, the 16APSK signal has two modulation modes and mapping rules, namely 4+12APSK and 8+8APSK. Different modulation modes and mapping rules are processed differently when performing frequency offset estimation. For 4+12APSK signals, the frequency value corresponding to the maximum value of the signal's 12th power spectrum is divided by 12, and the frequency value obtained is the frequency offset estimation value; for 8+8APSK signals, the frequency value corresponding to the maximum value of the signal's 8th power spectrum is divided by 8, and the frequency value obtained is the frequency offset estimation value; for 32QAM signals, the frequency value corresponding to the maximum value of the signal's 12th power spectrum is divided by 12, and the frequency value obtained is the frequency offset estimation value; for 8 / 16 / 64QAM signals, the frequency value corresponding to the maximum value of the signal's 4th power spectrum is divided by 4, and the frequency value obtained is the frequency offset estimation value.

[0064] The blind demodulation method for wireless signals provided in the embodiment of the present invention obtains a frequency offset estimation result by processing IQ data using the centroid method, the maximum spectrum value method and the local frequency analysis method, thereby improving the accuracy of the frequency offset estimation.

[0065] In some embodiments, the target parameter also includes phase deviation; after obtaining the frequency deviation estimation result, the blind demodulation method of the wireless signal further includes: performing at least one of a multi-square sum method and a phase difference method on the IQ data according to the frequency deviation estimation result to obtain the phase deviation estimation result.

[0066] In this embodiment, based on the above-mentioned frequency offset estimation result (the signal after frequency offset compensation), the frequency offset of IQ data is estimated by multi-square summation method and phase difference method, and the following modulation types among the candidate modulation types can be identified: MSK, MPSK, π / 4QPSK, DQPSK, 16APSK or 8 / 16 / 32 / 64QAM; specifically implemented through the following steps: For MFSK (M is the bit-to-symbol mapping number) type signals, since the demodulation process is insensitive to the frequency offset and phase offset, there is no need to estimate the phase offset. The demodulation result can be obtained by performing phase differential to obtain the instantaneous frequency.

[0067] For MASK (M is the binary number for mapping bits to symbols) type signals, the phase is calculated by summing the signal to the first power and then subtracting the initial phase (the initial phase of MASK type signals is 0), which is the estimated value of the phase deviation.

[0068] For MSK signals, the squared sum of the signal is used to calculate the phase, then divided by 2, and then the initial phase (the initial phase of MSK / GMSK signals is 0) is subtracted to obtain the estimated phase deviation.

[0069] For MPSK (M is the bit-to-symbol mapping number) and OQPSK signals, sum the signal to the Mth power to find the phase, divide by M, and then subtract the initial phase (the initial phase of 2 / 8PSK signals is 0, and the initial phase of 4PSK / OQPSK signals is ), which is the estimated value of the phase deviation.

[0070] For π / 4QPSK and DQPSK signals, sum the signal to the eighth power to get the phase, divide by 8, and then subtract the initial phase (the initial phase of π / 4QPSK and DQPSK signals is 0), which is the estimated value of the phase deviation.

[0071] In this embodiment, the 16APSK signal has two modulation modes and mapping rules, namely 4+12APSK and 8+8APSK. Different modulation modes and mapping rules are processed differently when performing phase deviation estimation.

[0072] Specifically, for a 4+12APSK signal, sum the signal to the 12th power to find the phase, divide by 12, and then subtract the initial phase (the initial phase of a 4+12APSK signal is ), which is the estimated value of phase deviation; for 8+8APSK signal, the phase is calculated by summing the signal to the eighth power and then dividing by 8, and then subtracting the initial phase (the initial phase of 8+8APSK signal is ), which is the estimated value of the phase deviation.

[0073] For MQAM ((M is the bit-to-symbol mapping number)) signals, sum the signal to the power of 4 to find the phase, divide by 4, and then subtract the initial phase (the initial phase of the 8QAM signal is , the initial phase of 16 / 32 / 64QAM signal is ), which is the estimated value of the phase deviation.

[0074] In this embodiment, for modulation modes within the traversal range, when a certain modulation mode is traversed, the corresponding phase deviation estimation method is used to estimate the corresponding frequency deviation.

[0075] The blind demodulation method for wireless signals provided by the embodiment of the present invention obtains a phase deviation estimation result by processing IQ data using a multi-square summation method and a phase difference method, thereby improving the accuracy of phase deviation estimation.

[0076] In some embodiments, after obtaining the symbol rate estimation result, the blind demodulation method of the wireless signal further includes: when the FFT accuracy corresponding to the symbol rate estimation result is lower than the FFT accuracy threshold, performing local discrete Fourier transform DFT processing on the IQ data according to the spectrum position corresponding to the symbol rate estimation result to obtain a refined symbol rate estimation result.

[0077] In this embodiment, the symbol rate estimation includes a rough estimation (roughly estimating the symbol rate of the IQ data through the above embodiment to obtain the symbol rate estimation result) and a fine estimation; whether a fine estimation needs to be performed based on the rough estimation is determined as follows: If the FFT accuracy, i.e. the ratio between the sampling rate and the number of FFT points, does not meet the requirement of symbol rate estimation accuracy, it is necessary to supplement the rough estimation with a fine estimation to meet the requirement of symbol rate estimation accuracy.

[0078] In this embodiment, the following method can be used for fine estimation of symbol rate: Based on the rough estimation of symbol rate, local DFT of the signal is performed according to the spectrum position idx of the roughly estimated symbol rate (num points before and after idx (including idx point), i.e. DFT of a total of (2×num+1) points) to improve the estimation accuracy, so as to meet the estimation accuracy requirement.

[0079] Specifically, assuming the sampling rate is fs, the number of FFT points is N, and the rough estimation accuracy is fs / N. After fine estimation, the estimation accuracy is improved to (fs / N) / (2×num+1), and the accuracy is improved by (2×num+1) times; the larger num is, the higher the accuracy of the fine symbol rate estimation result is, and the amount of calculation will also increase greatly. You can choose according to the estimation accuracy requirements and the amount of calculation.

[0080] The blind demodulation method for wireless signals provided in the embodiment of the present invention performs local discrete Fourier transform DFT processing on IQ data through the spectrum position corresponding to the symbol rate estimation result to obtain a refined symbol rate estimation result, thereby further improving the accuracy of the symbol rate estimation result.

[0081] In some embodiments, after obtaining the frequency offset estimation result, the blind demodulation method of the wireless signal further includes: when the FFT accuracy corresponding to the frequency offset estimation result is lower than the FFT accuracy threshold, calculating the local refined power spectrum of the IQ data according to the spectrum position corresponding to the frequency offset estimation result to obtain a refined frequency offset estimation result.

[0082] In this embodiment, the following method can be used for fine estimation of frequency offset: Based on the rough frequency offset estimation result, the power spectrum is locally refined according to the power spectrum position idx where the rough frequency offset estimation result is located (num points before and after idx (including idx point), that is, DFT of a total of (2×num+1) points) to improve the estimation accuracy and meet the estimation accuracy requirements.

[0083] ​Specifically, assuming the sampling rate is fs, the number of FFT points is N, and the rough estimation accuracy is fs / N. After fine estimation, the estimation accuracy is improved to (fs / N) / (2×num+1), and the accuracy is improved by (2×num+1) times; the larger the num, the higher the accuracy of the fine frequency offset estimation result, and the amount of calculation will also increase greatly. You can choose according to the estimation accuracy requirements and the amount of calculation.

[0084] The blind demodulation method for wireless signals provided by the embodiment of the present invention calculates the local refined power spectrum of IQ data through the spectrum position corresponding to the frequency offset estimation result, thereby obtaining a refined frequency offset estimation result, thereby further improving the accuracy of the frequency offset estimation result.

[0085] Figure 2 is a second flow chart of the blind demodulation method of wireless signals provided by the present invention. In Figure 2 In the embodiment shown in FIG. 1 , a blind demodulation method for wireless signals is further implemented by the following steps: (1) Input IQ signal, sampling rate, modulation mode and bit rate; (2) Traverse every possible modulation mode; (3) Taking characteristic parameters in different ways for frequency offset and phase offset estimation; (4) Perform frequency offset estimation and phase offset estimation; (5) When the code rate (corresponding to the symbol rate) is unknown, the code rate of the IQ signal is estimated by using different code rate estimation processes for different modulation modes; when the code rate (corresponding to the symbol rate) is known and the sampling rate / code rate is an integer, the configured filter is used for filtering; when the above sampling rate / code rate is not an integer, the IQ signal is resampled to 8 or 16 times and filtered using the configured filter; (6) When performing FSK modulation, convert the IQ data (corresponding to the IQ signal) into single-channel frequency data; (7) Integrate the above data through some normalization operations and special processing; (8) According to the oversampling multiple sps, a synchronization point within a symbol is selected and the EVM calculation is performed during sampling; (9) Output the EVM corresponding to each modulation mode, and take the EVM that exceeds the threshold and is the smallest as the debugging mode output.

[0086] The blind demodulation device for wireless signals provided by the present invention is described below. The blind demodulation device for wireless signals described below and the blind demodulation method for wireless signals described above can be referred to in correspondence with each other.

[0087] Figure 3 is a schematic diagram of the structure of the blind demodulation device for wireless signals provided by the present invention, such as Figure 3 ​​As shown, the blind demodulation device of the wireless signal includes: a screening module 310, a parameter estimation module 320 and a demodulation module 330.

[0088] The screening module 310 is used to screen candidate modulation types from multiple modulation types according to the time domain characteristics of the wireless signal; wherein the candidate modulation types include at least two of 2 / 4ASK, 2 / 4PSK, 4π / DQPSK, 8PSK, OQPSK, π / 4QPSK, 2 / 4 / 8FSK, 8 / 16 / 32 / 64QAM, MSK and 16APSK; Parameter estimation module 320, for estimating target parameters of in-phase orthogonal IQ data in the wireless signal corresponding to each candidate modulation type, to obtain target parameter estimation results; the target parameters include at least one of symbol rate, frequency deviation and phase deviation; The demodulation module 330 is used to perform symbol synchronization processing on the target parameter estimation result to obtain the symbol synchronization point; calculate the EVM corresponding to the candidate modulation type according to the symbol synchronization point, and when the minimum value of the EVM corresponding to each candidate modulation type is less than or equal to the EVM threshold, determine that the candidate modulation type corresponding to the minimum value is the target modulation type; demodulate the wireless signal based on the target modulation type to obtain a demodulation result.

[0089] The blind demodulation device for wireless signals provided in the embodiment of the present invention selects candidate modulation types from multiple modulation types through the time domain characteristics in the wireless signal, estimates the target parameters of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type for each candidate modulation type, performs symbol synchronization processing on the estimation result and calculates the EVM corresponding to the candidate modulation type, and demodulates the wireless signal according to the candidate modulation type corresponding to the minimum EVM to obtain a demodulation result, thereby improving the blind demodulation efficiency and accuracy of composite modulation signals in complex channels or low signal-to-noise ratio environments.

[0090] Figure 4 An example of a physical structure diagram of an electronic device is shown below, such as Figure 4As shown, the electronic device may include: a processor (processor) 410 , a communication interface (Communications Interface) 420 , a memory (memory) 430 and a communication bus 440 , wherein the processor 410 , the communication interface 420 , and the memory 430 communicate with each other through the communication bus 440 . The processor 410 can call the logic instructions in the memory 430 to execute the blind demodulation method of the wireless signal, which includes: selecting candidate modulation types from multiple modulation types according to the time domain characteristics in the wireless signal; wherein the candidate modulation types include at least two of 2 / 4ASK, 2 / 4PSK, 4π / DQPSK, 8PSK, OQPSK, π / 4QPSK, 2 / 4 / 8FSK, 8 / 16 / 32 / 64QAM, MSK and 16APSK; for each candidate modulation type, estimating the target parameter of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type to obtain the target parameter estimation result; the target parameter includes at least one of the symbol rate, frequency deviation and phase deviation; performing symbol synchronization processing on the target parameter estimation result to obtain the symbol synchronization point; calculating the EVM corresponding to the candidate modulation type according to the symbol synchronization point, and when the minimum value of the EVM corresponding to each candidate modulation type is less than or equal to the EVM threshold, determining the candidate modulation type corresponding to the minimum value as the target modulation type; demodulating the wireless signal based on the target modulation type to obtain the demodulation result.

[0091] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0092] On the other hand, the present invention also provides a computer program product, the computer program product includes a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer can execute the blind demodulation method of the wireless signal provided by the above methods, the method comprising: selecting a candidate modulation type from a plurality of modulation types according to the time domain characteristics in the wireless signal; wherein the candidate modulation types include 2 / 4ASK, 2 / 4PSK, 4π / DQPSK, 8PSK, OQPSK, π / 4QPSK, 2 / 4 / 8FSK, 8 / 16 / 32 / 64QAM, MSK and 16APSK; for each candidate modulation type, estimate the target parameters of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type to obtain the target parameter estimation result; the target parameter includes at least one of the symbol rate, frequency deviation and phase deviation; perform symbol synchronization processing on the target parameter estimation result to obtain the symbol synchronization point; calculate the EVM corresponding to the candidate modulation type according to the symbol synchronization point, and when the minimum value of the EVM corresponding to each candidate modulation type is less than or equal to the EVM threshold, determine the candidate modulation type corresponding to the minimum value as the target modulation type; demodulate the wireless signal based on the target modulation type to obtain the demodulation result.

[0093] On another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the processor executes the blind demodulation method of the wireless signal provided by the above methods, the method comprising: selecting a candidate modulation type from a plurality of modulation types according to the time domain characteristics in the wireless signal; wherein the candidate modulation types include at least two of 2 / 4ASK, 2 / 4PSK, 4π / DQPSK, 8PSK, OQPSK, π / 4QPSK, 2 / 4 / 8FSK, 8 / 16 / 32 / 64QAM, MSK and 16APSK ; For each candidate modulation type, estimate the target parameters of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type to obtain the target parameter estimation result; the target parameter includes at least one of the symbol rate, frequency deviation and phase deviation; perform symbol synchronization processing on the target parameter estimation result to obtain the symbol synchronization point; calculate the EVM corresponding to the candidate modulation type according to the symbol synchronization point, and when the minimum value of the EVM corresponding to each candidate modulation type is less than or equal to the EVM threshold, determine the candidate modulation type corresponding to the minimum value as the target modulation type; demodulate the wireless signal based on the target modulation type to obtain the demodulation result.

[0094] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.

[0095] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, or of course by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A blind demodulation method for a wireless signal, characterized in that: include: Filtering candidate modulation types from a plurality of modulation types according to time domain characteristics in the wireless signal; wherein the candidate modulation types include at least two of 2 / 4ASK, 2 / 4PSK, 4π / DQPSK, 8PSK, OQPSK, π / 4QPSK, 2 / 4 / 8FSK, 8 / 16 / 32 / 64QAM, MSK and 16APSK; For each candidate modulation type, estimate the target parameter of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type to obtain the target parameter estimation result; the target parameter includes at least one of a symbol rate, a frequency offset and a phase offset; Perform symbol synchronization processing on the target parameter estimation result to obtain a symbol synchronization point; calculate the EVM corresponding to the candidate modulation type according to the symbol synchronization point, and when the minimum value of the EVM corresponding to each candidate modulation type is less than or equal to the EVM threshold, determine that the candidate modulation type corresponding to the minimum value is the target modulation type; demodulate the wireless signal based on the target modulation type to obtain a demodulation result.

2. The blind demodulation method for wireless signals according to claim 1, characterized in that: The time domain feature includes the envelope flatness of the wireless signal; The step of selecting a candidate modulation type from a plurality of modulation types according to the time domain characteristics in the wireless signal includes: Determining the envelope flatness according to the sampling rate and time domain distribution data of the wireless signal; The envelope flatness is compared with the time domain feature decision thresholds corresponding to the multiple modulation types to determine the candidate modulation type.

3. The blind demodulation method for wireless signals according to claim 1, characterized in that: The target parameters include symbol rate; The estimating the target parameter of the in-phase orthogonal IQ data in the wireless signal corresponding to the candidate modulation type to obtain the target parameter estimation result includes: The IQ data is processed by at least one of differential envelope normalization, instantaneous frequency envelope normalization and quadratic power spectrum to obtain a symbol rate estimation result.

4. The blind demodulation method for wireless signals according to claim 3, characterized in that: The target parameters also include frequency deviation; After obtaining the symbol rate estimation result, the method further includes: The IQ data is processed according to the symbol rate estimation result using at least one of a centroid method, a maximum spectrum value method and a local frequency analysis method to obtain a frequency offset estimation result.

5. The blind demodulation method for wireless signals according to claim 4, characterized in that: The target parameters also include phase deviation; After obtaining the frequency offset estimation result, the method further includes: At least one of a multi-power sum method and a phase difference method is performed on the IQ data according to the frequency offset estimation result to obtain a phase offset estimation result.

6. The blind demodulation method for wireless signals according to claim 3, characterized in that: After obtaining the symbol rate estimation result, the method further includes: When the FFT accuracy corresponding to the symbol rate estimation result is lower than the FFT accuracy threshold, local discrete Fourier transform DFT processing is performed on the IQ data according to the frequency spectrum position corresponding to the symbol rate estimation result to obtain a refined symbol rate estimation result.

7. The blind demodulation method for wireless signals according to claim 4, characterized in that: After obtaining the frequency offset estimation result, the method further includes: When the FFT accuracy corresponding to the frequency offset estimation result is lower than the FFT accuracy threshold, the local refined power spectrum of the IQ data is calculated according to the frequency spectrum position corresponding to the frequency offset estimation result to obtain a refined frequency offset estimation result.

8. A blind demodulation device for wireless signals, characterized in that: include: A screening module, configured to screen out candidate modulation types from a plurality of modulation types according to time domain characteristics in a wireless signal; wherein the candidate modulation types include at least two of 2 / 4ASK, 2 / 4PSK, 4π / DQPSK, 8PSK, OQPSK, π / 4QPSK, 2 / 4 / 8FSK, 8 / 16 / 32 / 64QAM, MSK and 16APSK; A parameter estimation module, configured to estimate, for each candidate modulation type, a target parameter of in-phase orthogonal IQ data in a wireless signal corresponding to the candidate modulation type, to obtain the target parameter estimation result; the target parameter includes at least one of a symbol rate, a frequency deviation and a phase deviation; A demodulation module is used to perform symbol synchronization processing on the target parameter estimation result to obtain a symbol synchronization point; calculate the EVM corresponding to the candidate modulation type according to the symbol synchronization point, and when the minimum value of the EVM corresponding to each candidate modulation type is less than or equal to the EVM threshold, determine that the candidate modulation type corresponding to the minimum value is the target modulation type; demodulate the wireless signal based on the target modulation type to obtain a demodulation result.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the blind demodulation method for wireless signals according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the blind demodulation method for wireless signals according to any one of claims 1 to 7 is implemented.

Images