Modulation signal processing, devices and electronic equipment

Abstract

The present invention provides a modulation signal processing, device and electronic device. The modulation signal processing method includes: obtaining a modulation signal; splitting the modulation signal into discrete Fourier transform points according to a preset Fourier point rule; the preset Fourier point rule is multiplying an integer power of 2 by an odd number greater than 2; performing a fast Fourier transform on the integer power of 2 in the Fourier points of the modulation signal; performing a discrete Fourier transform on the odd numbers greater than 2 in the Fourier points of the modulation signal; converting the calculation results after the fast Fourier transform and the discrete Fourier transform into a basis transform operation form; and performing parallel calculations on the basis transform operation form based on NEON instructions. The embodiment of the present invention is used to solve the defect that the existing DFT and mixed basis algorithms cannot reduce the algorithm complexity and improve the operation speed when processing the modulation signal.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a modulation signal processing device and electronic equipment. Background Art

[0002] The conversion and analysis of signals in the time domain and frequency domain is an important part of signal processing. For example, the core technology of OFDM (Orthogonal Frequency Division Multiplexing) in 4G and 5G is the Discrete Fourier Transform (DFT) and Inverse Discrete Fourier Transform (IDFT) of the modulated signal, and its algorithm complexity is o(n*n). As the number of transformation points increases, the algorithm complexity increases quadratically, which will greatly challenge the hardware resources when processing signals, resulting in poor signal processing efficiency.

[0003] For the existing Fourier transform:

[0004] 1). The calculation formula of DFT is as follows:

[0005]

[0006] In the formula, X(k) represents the frequency domain part after FFT transformation, and x(n) represents the time domain part before transformation. Represents the rotation factor. Its multiplication complexity and addition complexity are both o(n*n)

[0007] 2). Mixed basis algorithm. Like DFT, it satisfies all points. The algorithm splits the points into irreducible prime numbers step by step. Then it uses the Cooley-Tukey mixed basis algorithm for transformation. For N=N1N2 points, the specific method of the mixed basis algorithm is to rearrange the original signal and then calculate it. The steps are as follows:

[0008] a. Store the signal as a matrix by column

[0009] b. Calculate N2-point DFT for each row

[0010] c. Multiply each item of the matrix by the rotation factor

[0011] d. Calculate N1-point DFT for each column

[0012] e. Read the result array row by row

[0013] Arrange the original data into an array of N1 rows and N2 columns, recorded as x[n1][n2]=x[N1*n1+n2], perform DFT transformation on each row, and record the result as X1[n1][n2]; then multiply X1 by Right now Finally, DFT transform is performed on each column of X2 to obtain X3, and the final result is X[N1*k1+k2]=X3[k1][k2].

[0014] For the number of points of multiple prime number combinations, we can split them up step by step to calculate. n In the case of n -n)).

[0015] The existing DFT and mixed basis algorithms cannot reduce the algorithm complexity and improve the operation speed at the same time when processing modulated signals. Summary of the invention

[0016] The present invention provides a modulation signal processing device and electronic equipment, which are used to solve the defects that the existing DFT and mixed basis algorithms cannot reduce the algorithm complexity and improve the operation speed when processing the modulation signal.

[0017] The present invention provides a modulation signal processing method, comprising:

[0018] Obtaining a modulated signal;

[0019] Splitting the discrete Fourier transform points of the modulated signal according to a preset Fourier point rule; the preset Fourier point rule is multiplying an integer power of 2 by an odd number greater than 2;

[0020] Perform fast Fourier transform on integer powers of 2 in the number of Fourier points of the modulated signal; perform discrete Fourier transform on odd numbers greater than 2 in the number of Fourier points of the modulated signal;

[0021] Convert the calculation results after fast Fourier transform and discrete Fourier transform into basis transform operation form;

[0022] The base transformation operation form is calculated in parallel based on NEON instructions.

[0023] According to a modulation signal processing method provided by the present invention, the discrete Fourier transform points of the modulation signal are split according to a preset Fourier point rule, which is implemented by the following formula:

[0024]

[0025] Wherein, N represents the number of Fourier points of the modulation signal; N1...N nRepresents a prime number greater than 2; r0, r1...r n Indicates 2, N1...N n The corresponding power.

[0026] According to a modulation signal processing method provided by the present invention, the step of performing discrete Fourier transform on an odd number greater than 2 in the Fourier point number of the modulation signal comprises:

[0027] The odd numbers greater than 2 in the Fourier points of the modulation signal are split into prime numbers greater than 2 and multiplied to perform mixed-basis discrete Fourier transform.

[0028] According to a modulation signal processing method provided by the present invention, the conversion of the calculation results after the fast Fourier transform and the discrete Fourier transform into a basis transform operation form comprises:

[0029] The calculation results after fast Fourier transform and discrete Fourier transform are converted into a basis transform operation form in which the transformation matrix is ​​multiplied with the input data.

[0030] According to a modulation signal processing method provided by the present invention, the base transformation operation form is parallelly calculated based on NEON instructions, including:

[0031] The transformation matrix in the form of an integer power of 2 with a base transformation point number is calculated in parallel based on NEON instructions;

[0032] A zero-filling operation is performed on the transformation matrix that does not meet the NEON instruction condition, and parallel calculation is performed based on the NEON instruction based on the transformation matrix that has undergone the zero-filling operation.

[0033] According to a modulation signal processing method provided by the present invention, the transformation matrix that does not meet the NEON instruction condition is padded with 0, and parallel calculation is performed based on the transformation matrix that has undergone the 0-padded operation based on the NEON instruction, including:

[0034] For the transformation matrix corresponding to the prime number greater than 2, the transformation matrix corresponding to the prime number greater than 2 and the input data are padded with 0 until they satisfy the multiple form of 4;

[0035] The transformation matrix and input data corresponding to the prime numbers greater than 2 after the zero-padding operation are calculated in parallel based on NEON instructions.

[0036] The present invention also provides a modulation signal processing device, comprising:

[0037] An acquisition module, used for acquiring a modulated signal;

[0038] A Fourier point splitting module, used for splitting the discrete Fourier transform points of the modulated signal according to a preset Fourier point rule; the preset Fourier point rule is multiplication of an integer power of 2 and an odd number greater than 2;

[0039] A mixed basis transform module, used for performing fast Fourier transform on integer powers of 2 in the number of Fourier points of the modulated signal; performing discrete Fourier transform on odd numbers greater than 2 in the number of Fourier points of the modulated signal;

[0040] A basis transformation operation module is used to convert the calculation results after fast Fourier transformation and discrete Fourier transformation into a basis transformation operation form;

[0041] The NEON instruction parallel computing module is used to perform parallel computing on the base transformation operation form based on the NEON instruction.

[0042] 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 when the processor executes the program, the modulation signal processing method described above is implemented.

[0043] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the modulation signal processing method described in any one of the above is implemented.

[0044] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the modulation signal processing method described above is implemented.

[0045] The modulation signal processing, device and electronic device provided by the present invention split the Fourier points of the modulation signal according to a preset Fourier point rule; the preset Fourier point rule is the multiplication of an integer power of 2 and an odd number greater than 2. In the embodiment of the present invention, the integer power of 2 in the Fourier points of the modulation signal is subjected to a fast Fourier transform; the odd numbers greater than 2 in the Fourier points of the modulation signal are subjected to a discrete Fourier transform; a mixed basis discrete Fourier transform and a fast Fourier transform are combined to perform a modulation signal, thereby improving the operation speed of the Fourier transform of the modulation signal; the embodiment of the present invention converts the calculation results after the fast Fourier transform and the discrete Fourier transform into a basis transform operation form; and the basis transform operation form is parallelly calculated based on the NEON instruction. Since the 16 128-bit NEON registers corresponding to the NEON instruction can realize parallel processing of the basis transform operation form, the algorithm complexity is reduced when processing the modulation signal, so that the embodiment of the present invention reduces the algorithm complexity and improves the operation speed when processing the modulation signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 paying creative work.

[0047] Figure 1 It is one of the flow charts of the modulation signal processing method provided by the present invention;

[0048] Figure 2 This is the second flow chart of the modulation signal processing method provided by the present invention;

[0049] Figure 3 A schematic diagram showing that the present invention stores the values ​​of a 2*2 transformation matrix in an array M and stores the input values ​​in an array N;

[0050] Figure 4 It is a structural schematic diagram of a modulation signal processing method and device provided by the present invention;

[0051] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0052] 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 of 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.

[0053] Combine the following Figure 1-Figure 2 For a description of the modulation signal processing method of the present invention, please refer to Figure 1 , the modulation signal processing method of the embodiment of the present invention includes:

[0054] Step 100: Obtain a modulated signal;

[0055] The electronic device obtains a modulated signal. The modulated signal of the embodiment of the present invention may be a modulated signal of OFDM (Orthogonal Frequency Division Multiplexing) used in 4G and 5G. The modulated signal includes multiple subcarriers. In this embodiment, the modulated signal is a string of data in complex form.

[0056] Step 200, splitting the discrete Fourier transform points of the modulated signal according to a preset Fourier point rule; the preset Fourier point rule is multiplying an integer power of 2 by an odd number greater than 2;

[0057] The electronic device splits the discrete Fourier transform points of the modulated signal according to a preset Fourier point rule. The preset Fourier point rule is an integer power of 2 multiplied by an odd number greater than 2. That is, the electronic device performs a mixed basis operation on the Fourier transform point difference of the modulated signal in the form of an integer power of 2 multiplied by an odd number.

[0058] For example, the modulated signal includes 1200 subcarriers, that is, the number of points of the modulated signal is 1200. After the modulated signal is split into discrete Fourier transform points according to the preset Fourier point rule, 1200=2 4 *75. Electronic devices divide the number of points of the modulated signal into integer powers of 2 4 , and an odd number greater than 2, 75.

[0059] Furthermore, an odd number greater than 2 can be further decomposed into the form of multiplying prime numbers greater than 2.

[0060] In one embodiment, splitting the discrete Fourier transform points of the modulated signal according to a preset Fourier point rule is implemented by the following formula:

[0061]

[0062] Wherein, N represents the number of Fourier points of the modulation signal; N1...N n Represents a prime number greater than 2; r0, r1...r n Indicates 2, N1...N n The corresponding power.

[0063] For example, 1200 = 2 4 *75=2 4 *3*25. Convert 1200-point discrete Fourier transform to a mixed operation of 16-point, 3-point and 25-point discrete Fourier transform.

[0064] Step 300: Perform fast Fourier transform on integer powers of 2 in the number of Fourier points of the modulated signal; perform discrete Fourier transform on odd numbers greater than 2 in the number of Fourier points of the modulated signal;

[0065] The electronic device performs a mixed basis discrete Fourier transform based on the split modulated signal. Specifically, the electronic device performs a fast Fourier transform on the integer power of 2 in the Fourier points of the modulated signal; and performs a discrete Fourier transform on the odd numbers greater than 2 in the Fourier points of the modulated signal.

[0066] Since the integer powers of 2 in the Fourier points of the modulation signal meet the point number conditions of the fast Fourier transform, the fast Fourier transform has the characteristic of fast operation speed. By performing fast Fourier transform on the integer powers of 2 in the Fourier points of the modulation signal; performing discrete Fourier transform on the odd numbers greater than 2 in the Fourier points of the modulation signal, a mixed-basis discrete Fourier transform and a fast Fourier transform are combined on the modulation signal, thereby improving the operation speed of the Fourier transform on the modulation signal.

[0067] In one embodiment, performing discrete Fourier transform on an odd number greater than 2 in the number of Fourier points of the modulated signal comprises:

[0068] The odd numbers greater than 2 in the Fourier points of the modulation signal are split into prime numbers greater than 2 and multiplied to perform mixed-basis discrete Fourier transform.

[0069] For example, it is assumed that the modulated signal includes 1200 Fourier points.

[0070] Step 31: Store the modulated signal in a matrix with a size of 75x16, denoted as X[n1][n2]=x[16*n1+n2].

[0071] Step 32: Perform 16-point DFT transformation on each row of the matrix to obtain X1[n1][n2]. The 16-point DFT transformation can be implemented using a radix-2 fast Fourier transform.

[0072] Step 33, for each item of X1[n1][n2] and the rotation factor Multiply them to get X2[n1][n2], that is

[0073] Step 34: Perform DFT operation on each column of X2[n1][n2] with a point number of 75. Finally, a 1200-point DFT output result is obtained.

[0074] For the 75-point discrete Fourier transform, further split it into 3x25. Perform the operations from step 31 to step 34 again.

[0075] Step 400, converting the calculation results after fast Fourier transform and discrete Fourier transform into a basis transform operation form;

[0076] The electronic device converts the calculation results after the fast Fourier transform and the discrete Fourier transform into a basis transform operation form.

[0077] Specifically, the converting of the calculation results after the fast Fourier transform and the discrete Fourier transform into a basis transform operation form includes:

[0078] The calculation results after fast Fourier transform and discrete Fourier transform are converted into a basis transform operation form in which the transformation matrix is ​​multiplied with the input data.

[0079] For the discrete Fourier transform of the power point, it will eventually be converted into a basis transform, that is, a basis transform of basis 2, basis 3 and basis 5. According to the properties of the discrete Fourier transform, please refer to formula (1). The calculation results after the fast Fourier transform and the discrete Fourier transform can be calculated by the basis transform operation form of matrix multiplication of the transformation matrix and the input data. The right side of formula (1) is the calculation result of the mixed basis Fourier transform of the fast Fourier transform and the discrete Fourier transform of the modulation signal. The left side of formula (1) is the matrix multiplication of the transformation matrix and the input data. After each level of iterative operation, the transformation matrix can be determined.

[0080]

[0081]

[0082] In the embodiment of the present invention, the 1200-point discrete Fourier transform of the modulated signal is decomposed into 75 16-point discrete Fourier transforms, 48 ​​25-point discrete Fourier transforms, and 400 3-point discrete Fourier transforms. For the DFT transform of the power point, it will eventually be converted into a basis transform, namely, basis 2, basis 3, and basis 5. For example, for the basis transform of basis 2, its calculation result is the multiplication of the 2x2 transformation matrix and the 2x1 input data to obtain a 2x1 transformation result.

[0083] Step 500: Perform parallel calculation on the base transformation operation form based on NEON instructions.

[0084] The electronic device performs parallel calculations on the base transformation operation form based on NEON instructions. Since the traditional algorithm mainly uses CPU general registers for serial calculations, the present invention uses 16 128-bit NEON registers provided by the ARM architecture to achieve parallel processing of the base transformation operation form. NEON instructions also reduce the multiplication complexity of the algorithm.

[0085] The modulation signal is split into Fourier points according to a preset Fourier point rule; the preset Fourier point rule is an integer power of 2 multiplied by an odd number greater than 2. The embodiment of the present invention performs a fast Fourier transform on the integer power of 2 in the Fourier points of the modulation signal; performs a discrete Fourier transform on the odd numbers greater than 2 in the Fourier points of the modulation signal; realizes a mixed-basis discrete Fourier transform and a fast Fourier transform on the modulation signal to improve the operation speed of the Fourier transform of the modulation signal; the embodiment of the present invention converts the calculation results after the fast Fourier transform and the discrete Fourier transform into a basis transform operation form; and performs parallel calculations on the basis transform operation form based on NEON instructions. Since the 16 128-bit NEON registers corresponding to the NEON instruction can realize parallel processing of the basis transform operation form, the algorithm complexity is reduced when processing the modulation signal, so that the embodiment of the present invention reduces the algorithm complexity and improves the operation speed when processing the modulation signal.

[0086] For other aspects of the embodiments of the present invention, please refer to Figure 2 Step 500, performing parallel calculation on the base transformation operation form based on NEON instructions, including:

[0087] Step 510, performing parallel calculation on the transformation matrix in the form of an integer power of 2 with a basis transformation point number based on NEON instructions;

[0088] Step 520: Perform a zero-padding operation on the transformation matrix that does not meet the NEON instruction condition, and perform parallel calculation based on the NEON instruction based on the transformation matrix that has undergone the zero-padding operation.

[0089] In most cases, the number of Fourier points of the modulated signal does not satisfy integer powers. That is, the fast Fourier transform cannot meet all point conditions, but the method of padding the points to integer powers can be used to perform fast Fourier transform operations to increase the calculation speed. The disadvantage of this method is that when the number of Fourier points of the modulated signal is a non-integer multiple period, spectrum leakage will occur due to non-integer period truncation.

[0090] The electronic device of the embodiment of the present invention performs parallel calculation based on NEON instructions on the transformation matrix in the form of an integer power of 2 with a base transformation point number; performs a zero-padded operation on the transformation matrix that does not meet the NEON instruction conditions, and performs parallel calculation based on the NEON instructions on the transformation matrix that has undergone the zero-padded operation.

[0091] In one embodiment, step 520, performing a zero-filling operation on the transformation matrix that does not meet the NEON instruction condition, and performing parallel calculation based on the transformation matrix after the zero-filling operation based on the NEON instruction, includes:

[0092] For the transformation matrix corresponding to the prime number greater than 2 whose base transformation points are padded with 0s, the transformation matrix and input data corresponding to the prime number greater than 2 are padded with 0s until they satisfy the multiple form of 4; the transformation matrix and input data corresponding to the prime number greater than 2 after the 0-padded operation are calculated in parallel based on NEON instructions.

[0093] It should be noted that for the fast Fourier transform, because the number of transformation points changes, the transformation result is different from the result of directly performing a discrete Fourier transform; for the transformation matrix, the zero-filling is only to meet the conditions of the NEON instruction for matrix multiplication, and the number of transformation points does not actually change.

[0094] The following example illustrates this:

[0095] Step 501, when the basis transformation is a 2-point discrete Fourier transform: store the value of the 2*2 transformation matrix in array M, the even part stores the real part of the transformation matrix value, and the odd part stores the imaginary part of the transformation matrix value. Store the input value in array N. Figure 3 As shown, Figure 3 A schematic diagram showing that the values ​​of a 2*2 transformation matrix are stored in an array M and the values ​​of input data are stored in an array N according to an embodiment of the present invention.

[0096] Step 502: Arrays M and N are loaded into registers A and B respectively at 8 bits each. This operation is completed by NEON instruction vst1q_f16. Registers A and B are multiplied by complex numbers, and the result is returned to register C. This operation is completed by NEON instruction vcmlaq_f16.

[0097] Step 503: Add the even part and the odd part of the first half and the second half of register C respectively to obtain the result of the basis transformation. This operation is completed by the NEON instruction vgetq_lane_f16.

[0098] Step 504, when the base transformation point number is a prime number n greater than 2, because the NEON instruction can simultaneously satisfy 4 complex numbers for multiplication, the transformation matrix and the input array are padded with 0 to a multiple of 4. For example, when the base transformation is 5, the size of the transformation matrix is ​​changed to 8x8. For example, when the base transformation is 7, the original concept transformation matrix and input data are 7x7 and 7x1, but in order to implement the NEON instruction operation, the transformation matrix and input data are padded with 0 to 8x8 and 8x1. Then the NEON instruction parallel calculation is performed.

[0099] Step 505: store the first row of the n*n transformation matrix into a register at 8 bits per row. The even part stores the real part of the value, and the odd part stores the imaginary part of the value. This operation is done by the NEON instruction vst1q_f16. Perform complex multiplication on registers A and B, and return the result to register C. This operation is done by the NEON instruction vcmlaq_f16. Add all the odd and even parts in register C to get the transformation result.

[0100] Step 506: Perform the operation of step 505 on each row of the n*n transformation matrix. Finally, the output result of the basis transformation is obtained.

[0101] For the transformation matrix that does not meet the NEON instruction condition, the embodiment of the present invention fills the transformation matrix and input data corresponding to the prime number greater than 2 with 0 until they meet the multiple form of 4 for the transformation matrix corresponding to the prime number greater than 2 with the basis transformation point number greater than 2; and performs parallel calculation based on the NEON instruction for the transformation matrix and input data corresponding to the prime number greater than 2 after the 0-filling operation. Compared with the traditional 0-filling fast Fourier algorithm, the spectrum leakage that may be caused by long 0-filling is avoided.

[0102] The embodiment of the present invention combines the mixed-base discrete Fourier transform with the fast Fourier transform, and at the same time, the base transform is realized in high-speed parallel by using 16 128-bit NEON registers provided by ARM corresponding to the NEON instruction, thereby reducing the complexity of the algorithm. For those that do not meet the NEON instruction conditions, a transformation matrix 0-filling method is provided to fully improve the register efficiency, save hardware resources, and improve the operating speed of the modulated signal processing. Compared with the traditional 0-filling fast Fourier transform, the spectrum leakage that may be caused by long 0-filling is avoided.

[0103] The modulation signal processing device provided by the present invention is described below. The modulation signal processing device described below and the modulation signal processing method described above can be referenced to each other.

[0104] Please refer to Figure 4 The present invention also provides a modulation signal processing device, comprising:

[0105] An acquisition module 201 is used to acquire a modulated signal;

[0106] A Fourier point splitting module 202 is used to split the discrete Fourier transform points of the modulated signal according to a preset Fourier point rule; the preset Fourier point rule is the multiplication of an integer power of 2 and an odd number greater than 2;

[0107] A mixed basis transform module 203 is used to perform fast Fourier transform on integer powers of 2 in the number of Fourier points of the modulated signal; and perform discrete Fourier transform on odd numbers greater than 2 in the number of Fourier points of the modulated signal;

[0108] A basis transformation operation module 204, used for converting the calculation results after the fast Fourier transform and the discrete Fourier transform into a basis transformation operation form;

[0109] The NEON instruction parallel computing module 205 is used to perform parallel computing on the base transformation operation form based on the NEON instruction.

[0110] The modulation signal processing device of the embodiment of the present invention splits the Fourier points of the modulation signal according to a preset Fourier point rule; the preset Fourier point rule is the multiplication of an integer power of 2 and an odd number greater than 2. The embodiment of the present invention performs fast Fourier transform on the integer power of 2 in the Fourier points of the modulation signal; performs discrete Fourier transform on the odd numbers greater than 2 in the Fourier points of the modulation signal; realizes the combination of mixed-base discrete Fourier transform and fast Fourier transform on the modulation signal, thereby improving the operation speed of Fourier transform on the modulation signal; the embodiment of the present invention converts the calculation results after fast Fourier transform and discrete Fourier transform into basis transform operation form; and performs parallel calculation on the basis transform operation form based on NEON instructions. Since the 16 128-bit NEON registers corresponding to the NEON instruction can realize parallel processing of basis transform operation form, the algorithm complexity is reduced when processing the modulation signal, so the embodiment of the present invention reduces the algorithm complexity and improves the operation speed when processing the modulation signal.

[0111] According to a modulated signal processing method provided by the present invention, the Fourier point splitting module is implemented by the following formula:

[0112]

[0113] Wherein, N represents the number of Fourier points of the modulation signal; N1...N n Represents a prime number greater than 2; r0, r1...r n Indicates 2, N1...N n The corresponding power.

[0114] According to a modulation signal processing method provided by the present invention, the mixed basis transformation module is also used to split the odd numbers greater than 2 in the Fourier point number of the modulation signal into the form of multiplication of prime numbers greater than 2 to perform mixed basis discrete Fourier transform.

[0115] According to a modulation signal processing method provided by the present invention, the basis transformation operation module is specifically used to convert the calculation results after fast Fourier transformation and discrete Fourier transformation into a basis transformation operation form of matrix multiplication of transformation matrix and input data.

[0116] According to a modulation signal processing method provided by the present invention, the NEON instruction parallel computing module comprises:

[0117] A first computing module, configured to perform parallel computing on the transformation matrix in the form of an integer power of 2 with a base transformation point number based on NEON instructions;

[0118] The second calculation module is used to perform a zero-filling operation on the transformation matrix that does not meet the NEON instruction condition, and perform parallel calculation based on the NEON instruction based on the transformation matrix after the zero-filling operation.

[0119] According to a modulated signal processing method provided by the present invention, the second calculation module includes:

[0120] A zero-filling module is used to fill the transformation matrix corresponding to the prime number greater than 2 and the input data with zeros until the transformation matrix and the input data satisfy the multiple form of 4.

[0121] The actual calculation module is used to perform parallel calculations on the transformation matrix and input data corresponding to the prime numbers greater than 2 after the zero-filling operation based on NEON instructions.

[0122] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520 and the memory 530 communicate with each other through the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the modulation signal processing method, which includes: obtaining a modulation signal; splitting the discrete Fourier transform points of the modulation signal according to a preset Fourier point rule; the preset Fourier point rule is multiplying an integer power of 2 by an odd number greater than 2; performing fast Fourier transform on the integer power of 2 in the Fourier points of the modulation signal; performing discrete Fourier transform on the odd numbers greater than 2 in the Fourier points of the modulation signal; converting the calculation results after fast Fourier transform and discrete Fourier transform into a basis transform operation form; and performing parallel calculation on the basis transform operation form based on NEON instructions.

[0123] In addition, the processor 510 is a processor capable of running NEON instructions. The logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention is essentially 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, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a 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: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk.

[0124] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the modulation signal processing method provided by the above-mentioned methods, and the method includes: obtaining a modulation signal; splitting the discrete Fourier transform points of the modulation signal according to a preset Fourier point rule; the preset Fourier point rule is multiplying an integer power of 2 by an odd number greater than 2; performing a fast Fourier transform on the integer power of 2 in the Fourier points of the modulation signal; performing a discrete Fourier transform on the odd numbers greater than 2 in the Fourier points of the modulation signal; converting the calculation results after the fast Fourier transform and the discrete Fourier transform into a basis transform operation form; and performing parallel calculations on the basis transform operation form based on NEON instructions.

[0125] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the modulation signal processing method provided by the above-mentioned methods, the method comprising: obtaining a modulation signal; splitting the discrete Fourier transform points of the modulation signal according to a preset Fourier point rule; the preset Fourier point rule is multiplying an integer power of 2 by an odd number greater than 2; performing a fast Fourier transform on the integer power of 2 in the Fourier points of the modulation signal; performing a discrete Fourier transform on the odd numbers greater than 2 in the Fourier points of the modulation signal; converting the calculation results after the fast Fourier transform and the discrete Fourier transform into a basis transform operation form; and performing parallel calculations on the basis transform operation form based on NEON instructions.

[0126] 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 displayed as units may or may not be physical units, that is, 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. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0127] 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, and of course, it can also be implemented 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, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including 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.

[0128] 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 modulation signal processing method, characterized in that: include: Obtaining a modulated signal; Splitting the discrete Fourier transform points of the modulated signal according to a preset Fourier point rule; The preset Fourier point rule is the multiplication of an integer power of 2 and an odd number greater than 2; Perform fast Fourier transformation on integer powers of 2 in the number of Fourier points of the modulated signal; perform discrete Fourier transformation on odd numbers greater than 2 in the number of Fourier points of the modulated signal; Convert the calculation results after fast Fourier transform and discrete Fourier transform into basis transform operation form; The base transformation operation form is calculated in parallel based on NEON instructions.

2. The modulation signal processing method according to claim 1, characterized in that: The splitting of the discrete Fourier transform points of the modulated signal according to the preset Fourier point rule is achieved by the following formula: Wherein, N represents the number of Fourier points of the modulation signal; N1...N n Represents a prime number greater than 2; r0, r1...r n Indicates 2, N1...N n The corresponding power.

3. The modulation signal processing method according to claim 1, characterized in that: The step of performing discrete Fourier transform on the odd numbers greater than 2 in the number of Fourier points of the modulated signal comprises: The odd numbers greater than 2 in the Fourier points of the modulation signal are split into prime numbers greater than 2 and multiplied to perform mixed-basis discrete Fourier transform.

4. The modulation signal processing method according to claim 1, characterized in that: The step of converting the calculation results after the fast Fourier transform and the discrete Fourier transform into a basis transform operation form comprises: The calculation results after fast Fourier transform and discrete Fourier transform are converted into a basis transform operation form in which the transformation matrix is ​​multiplied with the input data.

5. The modulation signal processing method according to claim 4, characterized in that: The performing parallel calculation on the base transformation operation form based on NEON instructions includes: The transformation matrix in the form of an integer power of 2 with a base transformation point number is calculated in parallel based on NEON instructions; A zero-filling operation is performed on the transformation matrix that does not meet the NEON instruction condition, and parallel calculation is performed based on the NEON instruction based on the transformation matrix that has undergone the zero-filling operation.

6. The modulation signal processing method according to claim 5, characterized in that: The step of performing a zero-filling operation on the transformation matrix that does not satisfy the NEON instruction condition, and performing parallel calculation based on the NEON instruction based on the transformation matrix that has undergone the zero-filling operation, comprises: For the transformation matrix corresponding to the prime number greater than 2, the transformation matrix corresponding to the prime number greater than 2 and the input data are padded with 0 until they satisfy the multiple form of 4; The transformation matrix and input data corresponding to the prime numbers greater than 2 after the zero-padding operation are calculated in parallel based on NEON instructions.

7. A modulation signal processing device, characterized in that: include: An acquisition module, used for acquiring a modulated signal; A Fourier point splitting module, used for splitting the discrete Fourier transform points of the modulated signal according to a preset Fourier point rule; the preset Fourier point rule is the multiplication of an integer power of 2 and an odd number greater than 2; A mixed basis transform module, used for performing fast Fourier transform on integer powers of 2 in the number of Fourier points of the modulated signal; performing discrete Fourier transform on odd numbers greater than 2 in the number of Fourier points of the modulated signal; A basis transformation operation module is used to convert the calculation results after fast Fourier transformation and discrete Fourier transformation into a basis transformation operation form; The NEON instruction parallel computing module is used to perform parallel computing on the base transformation operation form based on the NEON instruction.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the modulated signal processing method according to any one of claims 1 to 6 is implemented.

9. 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 modulated signal processing method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the modulated signal processing method according to any one of claims 1 to 6 is implemented.

Images