Demodulation Method, Device, Computer Equipment and Storage Medium for OFDM Signal

Abstract

The present application relates to a demodulation method, apparatus, computer device, and storage medium for OFDM signals. The method includes: determining a plurality of frequency interval estimated values and a plurality of carrier frequency offset estimated values for each subcarrier of a target signal; for any pair of estimated values, demodulating the target signal based on the pair of estimated values to obtain a baseband signal corresponding to the target signal; for any baseband signal, constructing a constellation diagram corresponding to the baseband signal, obtaining a matrix expression eigenvalue corresponding to the constellation diagram, and determining a target eigenvalue difference according to the matrix expression eigenvalue; and determining a target baseband signal from each baseband signal according to the target eigenvalue differences corresponding to each baseband signal. Using this method can improve the demodulation accuracy of OFDM signals in a non-cooperative communication mode.

Description

Technical Field

[0001] The present application relates to the field of communications, and in particular, to a method and apparatus for demodulating an OFDM signal, a computer device, and a storage medium. Background Art

[0002] OFDM (Orthogonal Frequency Division Multiplexing) technology is a technology for transmitting information through multiple orthogonal subcarriers. The receiving end performs an inverse Fourier transform on the received signal and then demodulates the signal according to the frequency interval between the subcarriers to obtain the information transmitted on each subcarrier. In the scenario of non-cooperative communication, due to the interference of carrier frequency deviation and the inability to know the precise frequency interval between subcarriers, it is difficult to demodulate the OFDM signal.

[0003] In the prior art, the carrier frequency deviation and the frequency interval can be estimated by a blind separation method or by using a cyclic prefix, and then the received signal is demodulated according to the estimated carrier frequency deviation and frequency interval. However, due to the unstable phase difference in the blind separation method and the difficulty in converging the average value of the phase difference, only a rough estimate of the carrier frequency deviation can be obtained; while the methods based on cyclic prefix estimation all assume that the signal follows the IEEE802.16e protocol, but in the non-cooperative mode, the signal may not follow the IEEE-related protocol or the received cyclic prefix is incomplete, resulting in inaccurate estimation and low accuracy of the demodulated signal. Summary of the Invention

[0004] Based on this, it is necessary to provide a method and apparatus for demodulating an OFDM signal, a computer device, and a storage medium for the above technical problems.

[0005] In a first aspect, the present application provides a method for demodulating an OFDM signal. The method includes:

[0006] Determine a plurality of pre-estimated values of the frequency intervals and a plurality of pre-estimated values of the carrier frequency deviations of each subcarrier of the target signal;

[0007] For any pair of pre-estimated values, demodulate the target signal based on the pair of pre-estimated values to obtain a baseband signal corresponding to the target signal, where the pair of pre-estimated values is composed of any one of the pre-estimated values of the frequency interval and any one of the pre-estimated values of the carrier frequency deviation;

[0008] For any one of the baseband signals, construct a constellation diagram corresponding to the baseband signal, obtain a matrix expression eigenvalue corresponding to the constellation diagram, and determine a target eigenvalue difference according to the matrix expression eigenvalue;

[0009] Determine a target baseband signal from each of the baseband signals according to the difference in the target eigenvalue corresponding to each baseband signal.

[0010] In one embodiment, the baseband signal is composed of a plurality of transmission values. Demodulating the target signal based on the estimated value to obtain the baseband signal corresponding to the target signal includes:

[0011] Obtain the received values corresponding to the target signal at a plurality of moments, and obtain the number of subcarriers corresponding to the target signal;

[0012] For any one of the received values, determine the transmission value corresponding to the received value according to the number of subcarriers and the pair of estimated values.

[0013] In one embodiment, constructing a constellation diagram corresponding to the baseband signal and obtaining a matrix expression eigenvalue corresponding to the constellation diagram includes:

[0014] Construct a constellation diagram according to each of the transmission values corresponding to the target signal, and the constellation points in the constellation diagram correspond one-to-one with the transmission values;

[0015] For any two of the constellation points in the constellation diagram, construct the connection edge weight between the two constellation points according to the coordinates of the two constellation points in the constellation diagram;

[0016] Construct a weight matrix corresponding to the constellation diagram according to the connection edge weights between the constellation points;

[0017] Construct a Laplacian matrix corresponding to the constellation diagram according to the weight matrix, and use the eigenvalues of the Laplacian matrix as the matrix expression eigenvalues.

[0018] In one embodiment, constructing the connection edge weight between two constellation points according to the coordinates of the two constellation points in the constellation diagram includes:

[0019] Based on the coordinates of the two constellation points in the constellation diagram and the noise power parameter, construct the connection edge weight between the two constellation points through a Gaussian kernel function.

[0020] In one embodiment, determining the target eigenvalue difference according to the matrix expression eigenvalue includes:

[0021] Sort each of the matrix expression eigenvalues from largest to smallest to obtain an eigenvalue queue;

[0022] Determine the difference between every two adjacent matrix expression eigenvalues in the eigenvalue queue, and use the largest of the differences as the target eigenvalue difference.

[0023] In one embodiment, determining a target baseband signal from each of the baseband signals according to the difference in target eigenvalue corresponding to each of the baseband signals includes:

[0024] Regarding the difference in target eigenvalue, using the baseband signal corresponding to the largest difference in target eigenvalue as the target baseband signal.

[0025] In a second aspect, the present application further provides a demodulation device for OFDM signals. The device includes:

[0026] A first determination module, configured to determine a plurality of estimated frequency intervals and a plurality of estimated carrier frequency offsets for each subcarrier of the target signal;

[0027] A demodulation module, configured to demodulate the target signal based on any pair of the estimated values to obtain a baseband signal corresponding to the target signal, where the pair of estimated values consists of any one of the first estimated values and any one of the second estimated values;

[0028] A construction module, configured to construct a constellation diagram corresponding to any one of the baseband signals, obtain a matrix expression eigenvalue corresponding to the constellation diagram, and determine a difference in target eigenvalue according to the matrix expression eigenvalue;

[0029] A second determination module, configured to determine a target baseband signal from each of the baseband signals according to the difference in target eigenvalue corresponding to each of the baseband signals.

[0030] In one embodiment, the baseband signal consists of multiple transmitted values, and the demodulation module is further configured to:

[0031] Obtain received values corresponding to the target signal at multiple moments and obtain the number of subcarriers corresponding to the target signal;

[0032] Regarding any one of the received values, determine a transmitted value corresponding to the received value according to the number of subcarriers and the pair of estimated values.

[0033] In one embodiment, the construction module is further configured to:

[0034] Construct a constellation diagram according to each of the transmitted values corresponding to the target signal, where the constellation points in the constellation diagram correspond one-to-one to the transmitted values;

[0035] Regarding any two of the constellation points in the constellation diagram, construct a connection edge weight between the two constellation points according to the coordinates of the two constellation points in the constellation diagram;

[0036] Construct a weight matrix corresponding to the constellation diagram according to the connection edge weights between the constellation points.

[0037] Construct a Laplacian matrix corresponding to the constellation diagram according to the weight matrix, and use the eigenvalues of the Laplacian matrix as the matrix expression eigenvalues.

[0038] In one embodiment, the construction module is further configured to:

[0039] Based on the coordinates of two constellation points in the constellation diagram and the noise power parameter, construct the connection edge weight between the two constellation points through a Gaussian kernel function.

[0040] In one embodiment, the construction module is further configured to:

[0041] Sort the matrix expression eigenvalues from largest to smallest to obtain an eigenvalue queue;

[0042] Determine the difference between every two adjacent matrix expression eigenvalues in the eigenvalue queue, and use the largest of the differences as the target eigenvalue difference.

[0043] In one embodiment, the second determination module is further configured to:

[0044] Use the baseband signal corresponding to the largest target eigenvalue difference among the target eigenvalue differences as the target baseband signal.

[0045] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the above-mentioned method is implemented.

[0046] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned method is implemented.

[0047] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the above-mentioned method is implemented.

[0048] The above demodulation method, device, computer device and storage medium of the OFDM signal obtain multiple estimated carrier frequency deviations and frequency intervals, and inversely deduce the baseband signal according to the combination of each possible carrier frequency deviation and frequency interval, obtain the constellation diagram corresponding to the baseband signal, calculate the target eigenvalue difference expressed by the constellation diagram matrix, and then select the target baseband signal according to the target eigenvalue difference. Since the target eigenvalue difference can reflect the aggregation degree of each constellation point cluster in the constellation diagram, and the aggregation degree of the constellation point cluster is related to the accuracy of the estimated value, therefore, selecting the target baseband signal according to the eigenvalue difference can select the target baseband signal with the most conforming constellation point cluster, that is, the one closest to the true value. Therefore, the accuracy of demodulating the OFDM signal can be improved when the carrier frequency deviation and frequency interval are unknown. Brief Description of the Drawings

[0049] Figure 1 It is a schematic flowchart of the demodulation method of the OFDM signal in one embodiment;

[0050] Figure 2 It is a schematic diagram of the constellation diagram in one embodiment;

[0051] Figure 3 It is a schematic flowchart of step 104 in one embodiment;

[0052] Figure 4 It is a schematic flowchart of step 106 in one embodiment;

[0053] Figure 5 It is a schematic flowchart of step 106 in one embodiment;

[0054] Figure 6 It is a schematic diagram of the difference between two adjacent eigenvalues in one embodiment;

[0055] Figure 7 It is a schematic flowchart of the demodulation method of the OFDM signal in one embodiment;

[0056] Figure 8 It is a schematic diagram of the simulation experiment in one embodiment;

[0057] Figure 9 It is a structural block diagram of the demodulation device of the OFDM signal in one embodiment;

[0058] Figure 10 It is an internal structure diagram of the computer device in one embodiment. Detailed Embodiments

[0059] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the 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.

[0060] In one embodiment, as Figure 1 shown, a demodulation method for an OFDM signal is provided. In this embodiment, taking the application of this method to a terminal as an example, it can be understood that this method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0061] Step 102, determine multiple frequency interval pre-estimation values and multiple carrier frequency offset pre-estimation values for each subcarrier of the target signal.

[0062] In the embodiment of the present application, the target signal is the signal received by a terminal (the terminal can be a receiver of an OFDM signal). The OFDM signal is formed by superimposing multiple subcarriers. The frequency difference between each subcarrier is the frequency interval. During the transmission of the OFDM signal, the difference between the carrier frequency of the target signal and the carrier frequency sent by the transmitter due to factors such as interference is the carrier frequency offset.

[0063] Since it is necessary to determine the precise carrier frequency offset to eliminate the influence of the carrier frequency offset on the signal quality when demodulating the OFDM signal, and at the same time, it is also necessary to know the frequency interval between the subcarriers. Therefore, multiple frequency interval estimation values corresponding to the frequency intervals and multiple carrier frequency offset estimation values corresponding to the carrier frequency offsets can be estimated, and then determine which frequency interval estimation value and which carrier frequency offset estimation value are closest to the true value to obtain the most accurate frequency interval and carrier frequency offset. The embodiment of the present application does not specifically limit the method for estimating the frequency interval pre-estimation value and the carrier frequency offset pre-estimation value. Any existing method for estimating the frequency interval pre-estimation value and the carrier frequency offset pre-estimation value is applicable to the embodiment of the present application, such as the blind separation method, the estimation method based on the cyclic prefix, etc.

[0064] It should be noted that when the estimated frequency interval or carrier frequency offset is an interval, multiple values can be taken from the interval as the frequency interval pre-estimation value or the carrier frequency offset pre-estimation value (hereinafter collectively referred to as the pre-estimation value). For example, multiple values can be taken at equal intervals from the interval, or values can be taken from the center of the interval and gradually outward with increasing intervals as the pre-estimation value, etc. The embodiment of the present application does not specifically limit this.

[0065] Step 104, for any pair of pre-estimation values, demodulate the target signal based on the pair of pre-estimation values to obtain the baseband signal corresponding to the target signal. The pair of pre-estimation values is composed of any frequency interval pre-estimation value and any carrier frequency offset pre-estimation value.

[0066] In an embodiment of the present application, any frequency interval estimated value and any carrier frequency deviation estimated value can form an estimated value pair. It should be noted that the frequency interval estimated values or carrier frequency deviation estimated values in different estimated value pairs can be repeated, as long as each estimated value pair is different. For example, the frequency interval estimated value A and the carrier frequency deviation estimated value A can form an estimated value pair, and the frequency interval estimated value A and the carrier frequency deviation estimated value B can also form an estimated value pair.

[0067] For each estimated value pair, the target signal can be demodulated once according to the frequency interval estimated value and the carrier frequency deviation estimated value in the estimated value pair to obtain the baseband signal corresponding to this estimated value pair. After verifying the accuracy of the baseband signal based on the constellation diagram corresponding to the baseband signal, the target baseband signal with the highest accuracy can be obtained from each baseband signal.

[0068] Step 106, for any baseband signal, construct the constellation diagram corresponding to the baseband signal, obtain the matrix expression eigenvalue corresponding to the constellation diagram, and determine the target eigenvalue difference according to the matrix expression eigenvalue.

[0069] In an embodiment of the present application, when the receiver receives the target signal, the target signal is over-sampled using a relatively high sampling frequency. A binary signal in the baseband signal can be obtained from the received value obtained by one sampling, and then the binary signal can be mapped to the complex plane (the I channel and the Q channel are mapped to the horizontal axis and the vertical axis respectively) to obtain a constellation point. After mapping all the binary signals corresponding to the received values, the constellation diagram corresponding to the baseband signal can be obtained.

[0070] The constellation diagram can be represented by a matrix. In order to make the matrix reflect the distance between constellation points, each element in the matrix can represent the distance weight between the constellation point corresponding to the column where the element is located and the constellation point corresponding to the row where the element is located. For example, the element located in the i-th row and j-th column of the matrix represents the distance weight between the i-th constellation point and the j-th constellation point. At the same time, in order to make the distance weight of constellation points closer to each other larger and facilitate subsequent calculations, the distance weight can be inversely proportional to the actual distance between constellation points. The distance weight can be calculated according to the coordinates of the two constellation points in the constellation diagram. For example, the reciprocal of the straight-line distance between the two constellation points can be directly taken as the distance weight, or the noise parameter can also be considered at the same time, and the distance weight can be constructed by the reciprocal of the straight-line distance between the noise parameter and the constellation points. The embodiment of the present application does not make specific limitations on this.

[0071] The eigenvalues of the matrix representation are the eigenvalues corresponding to the matrix of the constellation diagram. The definition of the target eigenvalue difference (eigengap, also known as the eigen-gap) is the largest value among the differences between adjacent eigenvalues after arranging the eigenvalues of the matrix from largest to smallest. In the case where the matrix is a representation of a graph, the target eigenvalue difference can reflect the stability of the clusters formed by the nodes in the graph. Therefore, by calculating the target eigenvalue difference of the matrix of the constellation diagram, a value can be obtained that can characterize the degree of aggregation of the constellation points in the constellation diagram. Furthermore, through this value, it can be determined which constellation diagram has the optimal degree of aggregation of the constellation points, that is, which constellation diagram is the optimal constellation diagram.

[0072] Step 108, determine the target baseband signal from each baseband signal according to the target eigenvalue difference corresponding to each baseband signal.

[0073] In the embodiments of the present application, referring to Figure 2 As shown, in the presence of a carrier frequency deviation, the phases of the constellation points will change, that is, it is equivalent to each point rotating around the center of the constellation diagram; and eliminating the carrier frequency deviation is equivalent to eliminating the rotation of each point and moving each point to its proper position.

[0074] And since, when the modulation method corresponding to the target signal is fixed, the types of received values obtained by sampling the target signal are limited, only a limited type of binary signal combination can be demodulated according to the target signal. Therefore, after eliminating the carrier frequency deviation, multiple binary signals will present the Figure 2 aggregation situation shown in the right figure in the constellation diagram. Since the target eigenvalue difference can reflect the degree of aggregation of the constellation points in the constellation diagram, and the more aggregated the constellation points are, the better the true values of the binary signals are restored, and the closer the carrier frequency deviation and the frequency interval are to the true values. Therefore, the target baseband signal closest to the true value can be determined from each baseband signal according to the target eigenvalue difference.

[0075] The maximum target eigenvalue difference (representing the best clustering of each constellation point in the constellation diagram) can be directly used as the target baseband signal. Alternatively, when the estimated frequency interval and the estimated carrier frequency deviation are in an interval, in order to improve the demodulation accuracy, the estimated value pair corresponding to the maximum target eigenvalue difference can be further obtained, and a plurality of new estimated frequency intervals and a plurality of new estimated carrier frequency deviations can be obtained near the estimated frequency interval and the estimated carrier frequency deviation of the estimated value pair to construct a new estimated value pair (for example, taking the estimated frequency interval as an example, a smaller distance interval can be preset, and centered on the estimated frequency interval, a new estimated frequency interval is taken every distance interval until the number of new estimated frequency intervals meets the requirements. Similar operations can also be performed for the estimated carrier frequency deviation). Then, the above process is repeated for each new estimated value pair, the baseband signal corresponding to each new estimated value pair and the target eigenvalue difference corresponding to the baseband signal are obtained, and the newly obtained target eigenvalue differences are compared with the original maximum target eigenvalue difference, and the maximum target eigenvalue difference among them is reselected; the above process can be repeated multiple times, and the preset distance interval is reduced each time until the number of cycles reaches the threshold, or the maximum target eigenvalue difference in the previous round is still the maximum target eigenvalue difference in this round.

[0076] The OFDM signal demodulation method provided by the embodiments of the present application obtains a plurality of estimated carrier frequency deviations and frequency intervals, and inversely deduces the baseband signal according to each possible combination of the carrier frequency deviation and the frequency interval, obtains the constellation diagram corresponding to the baseband signal, calculates the target eigenvalue difference expressed by the constellation diagram matrix, and then selects the target baseband signal according to the target eigenvalue difference. Since the target eigenvalue difference can reflect the clustering degree of each constellation point cluster in the constellation diagram, and the clustering degree of the constellation point cluster is related to the accuracy of the estimated value, selecting the target baseband signal according to the eigenvalue difference can select the target baseband signal with the most conforming constellation point cluster, that is, the one closest to the true value. Therefore, the demodulation accuracy of the OFDM signal can be improved when the carrier frequency deviation and the frequency interval are unknown.

[0077] In one embodiment, as Figure 3 shown, the baseband signal is composed of multiple transmission values. In step 104, demodulating the target signal based on the estimated value pair to obtain the baseband signal corresponding to the target signal includes:

[0078] Step 302, obtaining the received values corresponding to the target signal at multiple moments and the number of subcarriers corresponding to the target signal.

[0079] Step 304, for any received value, determining the transmission value corresponding to the received value according to the number of subcarriers and the estimated value pair.

[0080] In the embodiments of the present application, the transmitted value is also the binary signal in the foregoing embodiments. A plurality of binary signals are combined to form the information to be transmitted by the transmitter, that is, the baseband signal. When receiving the target signal, the receiver will sample the target signal multiple times, and one sample can obtain a received value. The received value is essentially the result of superposition of a plurality of orthogonal subcarriers. By eliminating the carrier frequency deviation of the received value through the carrier frequency deviation estimate in the estimate pair, and decomposing each subcarrier according to the frequency interval estimate, the transmitted value corresponding to the received value can be obtained.

[0081] Exemplarily, since the received value can be obtained by superposing the values of the transmitted value on each subcarrier, and the value on each subcarrier can be calculated according to the sampling period of the receiver, the channel impulse response of the channel used by the receiver to receive the target signal, the noise parameter (which can be determined by those skilled in the art according to experience), the carrier frequency deviation estimate, and the frequency interval estimate (see Formula (1)), the transmitted value can be deduced from the received value:

[0082]

[0083] where r k denotes the k-th received value collected within one sampling period, h k is the channel impulse response of the channel, M is the number of subcarriers (obtained according to the ratio of the total bandwidth of the target signal to the frequency interval estimate), s k is the transmitted value corresponding to r k e j2π is a representation of an imaginary number, f0 is the carrier frequency deviation estimate, f is the frequency interval estimate, T s is the sampling period, n k is the noise parameter.

[0084] According to Formula (1), the value of s k , that is, the transmitted value, can be deduced. By deducing the transmitted value for all the collected received values, the baseband signal corresponding to the target signal collected within this sampling period can be obtained.

[0085] The demodulation method of the OFDM signal provided by the embodiments of the present application determines the transmitted value corresponding to each received value of the target signal according to the number of subcarriers corresponding to the target signal and the estimate pair, so as to obtain the baseband signal corresponding to the target signal. Furthermore, the constellation diagram corresponding to the baseband signal can be constructed based on each transmitted value, and the target characteristic value difference corresponding to the baseband signal can be obtained. By selecting the target baseband signal according to the target characteristic value difference, the demodulation accuracy of the OFDM signal can be improved when the carrier frequency deviation and the frequency interval are unknown.

[0086] In one embodiment, as Figure 4As shown, in step 106, a constellation diagram corresponding to the baseband signal is constructed, and the matrix expression eigenvalues corresponding to the constellation diagram are obtained, including:

[0087] Step 402, construct a constellation diagram according to each transmission value corresponding to the target signal, and the constellation points in the constellation diagram correspond one-to-one with the transmission values.

[0088] Step 404, for any two constellation points in the constellation diagram, construct the connection edge weight between the two constellation points according to the coordinates of the two constellation points in the constellation diagram.

[0089] Step 406, construct the weight matrix corresponding to the constellation diagram according to the connection edge weights between the constellation points.

[0090] Step 408, construct the Laplacian matrix corresponding to the constellation diagram according to the weight matrix, and use the eigenvalues of the Laplacian matrix as the matrix expression eigenvalues.

[0091] In the embodiments of the present application, the transmission values of the target signal are projected onto the complex plane to obtain constellation points, and the constellation points form a constellation diagram. The constellation diagram can be characterized by the Laplacian matrix. The Laplacian matrix can be calculated according to the weight matrix of the constellation diagram.

[0092] For any two constellation points, the connection edge weight between the constellation points can be constructed according to the coordinates of the constellation points in the constellation diagram (when the constellation points are the same, the connection edge weight can be set to 0, that is, the diagonal elements of the weight matrix are all 0). For example, the reciprocal of the straight-line distance between the coordinates is taken to obtain the connection edge weight, or the distance weight is obtained by other methods of constructing distance weights based on the distance between two points, etc. Set the element in the i-th row and j-th column of the weight matrix as the connection edge weight between the i-th constellation point and the j-th constellation point to obtain the weight matrix, and then the Laplacian matrix of the constellation diagram can be calculated according to the following formula (two):

[0093]

[0094] Among them, L represents the Laplacian matrix, D is a diagonal matrix, the element in its i-th row and i-th column is equal to the sum of all elements in the i-th row of the weight matrix, and W represents the weight matrix.

[0095] Calculate the eigenvalues of the Laplacian matrix to obtain the matrix expression eigenvalues corresponding to the constellation diagram.

[0096] The demodulation method of the OFDM signal provided by the embodiment of the present application constructs a Laplacian matrix corresponding to the constellation diagram and obtains matrix expression eigenvalues based on the Laplacian matrix. Since the Laplacian matrix can accurately express the distances and connection conditions between the constellation points in the constellation diagram, using the Laplacian matrix as the matrix corresponding to the constellation diagram can improve the reflection of the target eigenvalue difference on the clustering degree of the constellation points in the constellation diagram, thereby improving the determination accuracy of the target baseband signal.

[0097] In one embodiment, in step 404, constructing the connection edge weight between two constellation points according to the coordinates of the two constellation points in the constellation diagram includes:

[0098] Based on the coordinates of the two constellation points in the constellation diagram and the noise power parameter, construct the connection edge weight between the two constellation points through the Gaussian kernel function.

[0099] In the embodiment of the present application, the Gaussian kernel function can be used to construct the connection edge weight. The Gaussian kernel function is a monotonically decreasing function with respect to the Euclidean distance between two points, that is, the greater the distance between two points, the smaller the connection edge weight calculated through the Gaussian kernel function; at the same time, the relationship between the Euclidean distance between two points and the connection edge weight can also be controlled by controlling the parameter σ in the Gaussian kernel function. When σ is small, the change in the Euclidean distance between two points has little effect on the change in the connection edge weight, that is, it can achieve the effect that constellation points with a relatively large distance can also be classified into one category, and constellation diagrams with non-tight clustering of constellation points can also have a relatively large target eigenvalue difference, reducing the accuracy requirement for the estimated value pair and improving the speed of obtaining the target baseband signal; when σ is large, the change in the Euclidean distance between two points has a greater effect on the change in the connection edge weight, that is, it can achieve the effect that constellation points with a relatively small distance can be classified into one category, improving the accuracy requirement for the estimated value pair and improving the accuracy of the target baseband signal.

[0100] σ can be set to be positively correlated with the noise power parameter, and the specific relationship between σ and the noise power parameter can be set by those skilled in the art according to experience. It is necessary to select an appropriate value of σ here because when the value of σ is too small, the connection edge weights between points will be relatively similar, making it difficult to distinguish points with different distances from each other; while when the value of σ is too large, the connection edge weight is sensitive to the noise in the constellation diagram, and the noise will cause obvious interference to the value of the connection edge weight. Exemplarily, the Scott rule or the Silverman rule can be used to calculate σ. The method of calculating σ through the Scott rule can be seen in the following formula (III):

[0101]

[0102] Where is the estimator of σ, and K is the total number of received values collected within a sampling period.

[0103] The method for calculating σ by Silverman's rule can be seen in the following formula (4):

[0104]

[0105] Wherein, is the estimator of σ, K is the total number of received values collected within one sampling period, and IQR refers to the interquartile range of the signal.

[0106] Exemplarily, the connection edge weight obtained based on the Euclidean distance of the Gaussian kernel function, the noise power parameter, and the coordinates of the constellation points can be as shown in formula (3):

[0107]

[0108] Wherein, is the constellation point and is the connection edge weight between them, is the constellation point and is the Euclidean distance between them, calculated according to the coordinates of and .

[0109] The demodulation method of the OFDM signal provided by the embodiments of the present application constructs the connection edge weight through the Gaussian kernel function, and sets the parameter σ in the Gaussian kernel function according to the noise power parameter. When the noise power parameter takes different values, the accuracy requirement for the target baseband signal can be adaptively adjusted accordingly, ensuring that the determination speed of the target baseband signal will not be too slow.

[0110] In one embodiment, as shown in Figure 5 , in step 106, determining the target eigenvalue difference according to the matrix expression eigenvalues includes:

[0111] Step 502, sorting each matrix expression eigenvalue from large to small to obtain an eigenvalue queue.

[0112] Step 504, determining the difference between every two adjacent matrix expression eigenvalues in the eigenvalue queue, and taking the largest difference among the differences as the target eigenvalue difference.

[0113] In the embodiments of the present application, the target eigenvalue difference can be obtained by sorting the matrix expression eigenvalues to obtain an eigenvalue queue, and then taking the largest difference among the differences between two adjacent matrix expression eigenvalues in the eigenvalue queue. Refer to Figure 6As shown, it is the result of sorting the eigenvalues of the Laplacian matrix from largest to smallest. It can be seen that the difference between the eigenvalues in the boxed area in the figure is the largest, so this difference is the target eigenvalue difference.

[0114] The OFDM signal demodulation method provided by the embodiments of the present application arranges the eigenvalues from largest to smallest and calculates the maximum value of the differences between two adjacent eigenvalues to obtain the target eigenvalue difference. Since the target eigenvalue difference can reflect the clustering degree of each constellation point cluster in the constellation diagram, and the clustering degree of the constellation point cluster is related to the accuracy of the estimated value, therefore, by selecting the target baseband signal according to the eigenvalue difference, a target baseband signal with the most conforming constellation point cluster, that is, the one closest to the true value, can be selected. Thus, the accuracy of demodulating the OFDM signal can be improved when the carrier frequency offset and the frequency interval are unknown.

[0115] In one embodiment, in step 108, determining the target baseband signal from each baseband signal according to the target eigenvalue difference corresponding to each baseband signal includes:

[0116] Taking the baseband signal corresponding to the largest target eigenvalue difference among the target eigenvalue differences as the target baseband signal.

[0117] In the embodiments of the present application, when the target eigenvalue difference is the largest, it means that the clustering of each constellation point in the constellation diagram is the tightest, the rotation of each constellation point is eliminated, and the constellation point is closest to the true value; therefore, among the baseband signals, the baseband signal corresponding to the largest target eigenvalue difference is the baseband signal closest to the true value. Therefore, the baseband signal corresponding to the largest target eigenvalue difference can be used as the target baseband signal, so as to obtain the result of demodulating the received OFDM signal.

[0118] The OFDM signal demodulation method provided by the embodiments of the present application takes the baseband signal corresponding to the largest target eigenvalue difference as the target baseband signal. Since the baseband signal corresponding to the largest target eigenvalue difference is closest to the true value, the accuracy of determining the target baseband signal can be improved.

[0119] To enable those skilled in the art to better understand the embodiments of the present application, the embodiments of the present application are described below through specific examples.

[0120] Refer to Figure 7 As shown, a flowchart of a method for demodulating an OFDM signal is shown.

[0121] In the embodiments of the present application, all possible frequency intervals and carrier frequency offsets can be traversed. For example, the intervals or possible corresponding values where the frequency interval and carrier frequency offset may be located can be determined, and the estimated values of the frequency interval and carrier frequency offset to be traversed are selected therefrom, and all the selected estimated values of the frequency interval and carrier frequency offset are traversed: for example, an estimated value of a fixed frequency interval can be selected first, and then the estimated value of the carrier frequency offset is selected and combined with the fixed estimated value of the frequency interval to calculate the constellation diagram of the baseband signal demodulated based on this combination, the Laplacian matrix of the constellation diagram, and the eigengap corresponding to the Laplacian matrix. After all the estimated values of the carrier frequency offset are combined with the fixed estimated value of the frequency interval, a new estimated value of the frequency interval is selected, and the above process is repeated until all the estimated values of the frequency interval are traversed. The eigengaps corresponding to all the combinations are compared, and the estimated values of the frequency interval and carrier frequency offset corresponding to the largest eigengap are used as the actual estimated values, and the baseband signal corresponding to the largest eigengap is used as the finally demodulated baseband signal, so that the subcarrier frequency interval, carrier frequency offset, and demodulation result corresponding to the target signal can be obtained. Among them, the method of obtaining the baseband signal, the method of constructing the Laplacian matrix, and the method of calculating the eigengap can be referred to the description of the foregoing embodiments, and the embodiments of the present application will not be elaborated herein.

[0122] The following uses a simulation experiment to illustrate the effects of the embodiments of the present application. Refer to Figure 8 As shown, it is set that the OFDM signal sent by the transmitter has a total of 4 subcarriers, and each subcarrier is modulated by 4QAM (a quadrature amplitude modulation method), and the frequency interval between adjacent subcarriers is 1 MHz. The sampling period of the receiver is set to 1 μs, 1024 samples are taken in each sampling period, the carrier frequency offset is 0.055 MHz, the channel is an AWGN channel, the channel impulse response is 30 dB, and σ in the Gaussian kernel function 2 is 64 times the noise power parameter. By traversing the estimated values of each frequency interval and carrier frequency offset and calculating the eigengaps of various combinations, the results as shown in Figure 8 above can be obtained.

[0123] Select the combination with the largest eigengap therefrom. It can be seen that the estimated value of the carrier frequency offset corresponding to the largest eigengap is 0.055 MHz, which is exactly equal to the carrier frequency offset set for the receiver in the simulation. The constellation diagram before the carrier frequency offset recovery, the difference between two adjacent eigenvalues when the estimated value of the carrier frequency offset is 0.055 MHz, and the constellation diagram after the carrier frequency offset recovery can be referred to Figure 8 as shown below.

[0124] The above method can be applied to the scenario of sensing unlicensed spectrum usage. By analyzing the wireless signals transmitted in the unlicensed spectrum using the above method, information such as the frequency interval and carrier frequency deviation of the unlicensed spectrum can be obtained. Furthermore, it is possible to sense the behavior of unlicensed spectrum usage within the coverage area of the operator's private network, which can help the operator provide services with the ability of proactive sensing security for the wireless networks of private network customers and industrial users, and avoid security threats such as eavesdropping, tampering, and interference to the communication of users within the wireless network.

[0125] The demodulation method of the OFDM signal provided by the embodiments of this application can accurately estimate the frequency interval and carrier frequency deviation of the OFDM system. In theory, as long as a sufficient number of frequency interval estimation values and carrier frequency deviation estimation values are taken, the frequency interval estimation values and carrier frequency deviation estimation values can infinitely approach the true values of the frequency interval and carrier frequency deviation, improve the estimation accuracy of the frequency interval and carrier frequency deviation, and further improve the demodulation accuracy of the OFDM signal in the non-cooperative communication mode.

[0126] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps in other steps.

[0127] Based on the same inventive concept, the embodiments of this application also provide a demodulation device for an OFDM signal for implementing the above-described demodulation method of the OFDM signal. The implementation solutions for solving problems provided by this device are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more embodiments of the demodulation device for the OFDM signal provided below can refer to the limitations on the demodulation method of the OFDM signal in the above text, and will not be repeated here.

[0128] In one embodiment, as Figure 9 shown, a demodulation device 900 for an OFDM signal is provided, including: a first determination module 902, a demodulation module 904, a construction module 906, and a second determination module 908, where:

[0129] The first determination module 902 is configured to determine multiple frequency interval pre-estimation values and multiple carrier frequency deviation pre-estimation values of each sub-carrier of the target signal;

[0130] A demodulation module 904, configured to demodulate the target signal based on any pair of estimated values to obtain a baseband signal corresponding to the target signal, where the pair of estimated values is composed of any one of the first estimated values and any one of the second estimated values;

[0131] A construction module 906, configured to construct a constellation diagram corresponding to the baseband signal for any one of the baseband signals, obtain matrix expression eigenvalues corresponding to the constellation diagram, and determine a target eigenvalue difference according to the matrix expression eigenvalues;

[0132] A second determination module 908, configured to determine a target baseband signal from each of the baseband signals according to the target eigenvalue differences corresponding to each of the baseband signals.

[0133] The OFDM signal demodulation device provided by the embodiments of the present application obtains multiple estimated carrier frequency offsets and frequency intervals, inversely deduces the baseband signal according to each combination of possible carrier frequency offsets and frequency intervals, obtains the constellation diagram corresponding to the baseband signal, calculates the target eigenvalue difference of the matrix expression of the constellation diagram, and then selects the target baseband signal according to the target eigenvalue difference. Since the target eigenvalue difference can reflect the aggregation degree of each constellation point cluster in the constellation diagram, and the aggregation degree of the constellation point cluster is related to the accuracy of the estimated value, the target baseband signal can be selected according to the eigenvalue difference, and the target baseband signal with the most conforming constellation point cluster, that is, the closest to the true value, can be selected. Therefore, the accuracy of demodulating the OFDM signal when the carrier frequency offset and frequency interval are unknown can be improved.

[0134] In one embodiment, the baseband signal is composed of multiple transmitted values, and the demodulation module 904 is further configured to:

[0135] Obtain received values corresponding to the target signal at multiple moments and obtain the number of subcarriers corresponding to the target signal;

[0136] For any one of the received values, determine the transmitted value corresponding to the received value according to the number of subcarriers and the pair of estimated values.

[0137] In one embodiment, the construction module 906 is further configured to:

[0138] Construct a constellation diagram according to each of the transmitted values corresponding to the target signal, where the constellation points in the constellation diagram correspond one-to-one with the transmitted values;

[0139] For any two constellation points in the constellation diagram, construct connection edge weights between the two constellation points according to the coordinates of the two constellation points in the constellation diagram;

[0140] Construct a weight matrix corresponding to the constellation diagram according to the connection edge weights between the respective constellation points;

[0141] Construct a Laplacian matrix corresponding to the constellation diagram according to the weight matrix, and use the eigenvalues of the Laplacian matrix as the matrix expression eigenvalues.

[0142] In one embodiment, the construction module 906 is further configured to:

[0143] Based on the coordinates of two constellation points in the constellation diagram and the noise power parameter, construct the connection edge weight between the two constellation points through a Gaussian kernel function.

[0144] In one embodiment, the construction module 906 is further configured to:

[0145] Sort the matrix expression eigenvalues from largest to smallest to obtain an eigenvalue queue;

[0146] Determine the difference between every two adjacent matrix expression eigenvalues in the eigenvalue queue, and use the largest difference among the differences as the target eigenvalue difference.

[0147] In one embodiment, the second determination module 908 is further configured to:

[0148] Use the baseband signal corresponding to the largest target eigenvalue difference among the target eigenvalue differences as the target baseband signal.

[0149] Each module in the above OFDM signal demodulation device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a memory in a computer device in software form, so that the processor can call and execute the operations corresponding to the above respective modules.

[0150] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in Figure 10 shown. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements an OFDM signal demodulation method.

[0151] Those skilled in the art can understand that Figure 10 The structure shown in Figure 10 is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0152] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0153] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0154] In one embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0155] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data that have been authorized by the user or fully authorized by all parties.

[0156] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0157] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0158] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A demodulation method for OFDM signals, characterized in that, The method includes: Determining a plurality of frequency interval pre - estimates and a plurality of carrier frequency deviation pre - estimates for each sub - carrier of the target signal; For any pair of pre - estimates, demodulating the target signal based on the pair of pre - estimates to obtain the baseband signal corresponding to the target signal, where the pair of pre - estimates consists of any one of the frequency interval pre - estimates and any one of the carrier frequency deviation pre - estimates; For any one of the baseband signals, constructing a constellation diagram corresponding to the baseband signal, obtaining the matrix expression eigenvalues corresponding to the constellation diagram, sorting the matrix expression eigenvalues from large to small to obtain an eigenvalue queue, determining the difference between every two adjacent matrix expression eigenvalues in the eigenvalue queue, and taking the largest of the differences as the target eigenvalue difference; Determining the target baseband signal from the baseband signals according to the target eigenvalue differences corresponding to the baseband signals; 2. The method according to claim 1, wherein The baseband signal consists of a plurality of transmission values. The demodulating the target signal based on the pair of pre - estimates to obtain the baseband signal corresponding to the target signal includes: Obtaining the received values corresponding to the target signal at multiple moments and obtaining the number of sub - carriers corresponding to the target signal; For any one of the received values, determining the transmission value corresponding to the received value according to the number of sub - carriers and the pair of pre - estimates; 3. The method according to claim 2, characterized in that, The constructing the constellation diagram corresponding to the baseband signal and obtaining the matrix expression eigenvalues corresponding to the constellation diagram includes: Constructing a constellation diagram according to each of the transmission values corresponding to the target signal, where the constellation points in the constellation diagram correspond one - to - one with the transmission values; For any two constellation points in the constellation diagram, constructing the connection edge weight between the two constellation points according to the coordinates of the two constellation points in the constellation diagram; Constructing a weight matrix corresponding to the constellation diagram according to the connection edge weights between the constellation points; Constructing a Laplacian matrix corresponding to the constellation diagram according to the weight matrix and taking the eigenvalues of the Laplacian matrix as the matrix expression eigenvalues; 4. The method according to claim 3, characterized in that, The constructing the connection edge weight between the two constellation points according to the coordinates of the two constellation points in the constellation diagram includes: Based on the coordinates of the two constellation points in the constellation diagram and the noise power parameter, constructing the connection edge weight between the two constellation points through a Gaussian kernel function; 5. The method according to claim 1, wherein The determining the target baseband signal from the baseband signals according to the target eigenvalue differences corresponding to the baseband signals includes: Taking the baseband signal corresponding to the largest target eigenvalue difference among the target eigenvalue differences as the target baseband signal; 6. A demodulation device for OFDM signals, characterized in that, The apparatus includes: A first determination module, configured to determine a plurality of frequency interval pre - estimates and a plurality of carrier frequency deviation pre - estimates for each sub - carrier of the target signal; A demodulation module, configured to, for any pair of pre - estimates, demodulate the target signal based on the pair of pre - estimates to obtain the baseband signal corresponding to the target signal, where the pair of pre - estimates consists of any one of the frequency interval pre - estimates and any one of the carrier frequency deviation pre - estimates; A building block for constructing, for any of the baseband signals, a constellation diagram corresponding to the baseband signal, obtaining matrix expression eigenvalues corresponding to the constellation diagram, determining a target eigenvalue difference according to the matrix expression eigenvalues, sorting the matrix expression eigenvalues from largest to smallest to obtain an eigenvalue queue, determining the difference between every two adjacent matrix expression eigenvalues in the eigenvalue queue, and taking the largest of the differences as the target eigenvalue difference; A second determination module for determining a target baseband signal from the baseband signals according to the target eigenvalue differences corresponding to the baseband signals; 7. The device according to claim 6, characterized in that, The baseband signal is composed of multiple transmission values, and the demodulation module is further configured to: Obtain received values corresponding to the target signal at multiple moments and obtain the number of subcarriers corresponding to the target signal; For any of the received values, determine the transmission value corresponding to the received value according to the number of subcarriers and the pre-estimated value pair; 8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented; 9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 5 are implemented; 10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images