A new low-complexity error vector magnitude calculation method

Abstract

The present invention relates to the field of error vector magnitude calculation, and specifically to a novel low-complexity error vector magnitude calculation method, which simplifies the EVM calculation of QAM modulations of different orders, thereby reducing the redundant processes generated during the judgment process during conventional calculations, improving processing speed, and improving processing efficiency. The technical solution includes: determining the modulation type of the signal; performing normalization processing according to the modulation type of the signal; folding and translating the normalized signal; determining whether the constellation image is equivalent to QPSK modulation, and if so, calculating the EVM of the signal after folding and transformation using the QPSK unit circle calculation method, otherwise continuing to repeat the folding and translation operations. The present invention is suitable for error vector magnitude calculation.

Description

Technical Field

[0001] The present invention relates to the field of error vector magnitude calculation, and in particular to a novel low-complexity error vector magnitude calculation method. Background Art

[0002] Currently, when calculating EVM (Error Vector Magnitude) for different orders of QAM (Quadrature Amplitude Modulation), the common method is to perform a check and calculate EVM at each point. This generates multiple check loops for system links with relatively high sampling rates. Currently, CPUs use branch prediction when processing the if-else check process, requiring pre-ordering of these multiple check processes to improve speed. Since there is no pre-ordering for the demodulated signal, branch prediction is unlikely to improve speed, resulting in relatively slow and inefficient processing. Summary of the Invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a new low-complexity error vector magnitude calculation method, which simplifies the EVM calculation of QAM modulation of different orders, thereby reducing the redundant process generated in the judgment process during conventional calculation, improving the processing speed and improving the processing efficiency.

[0004] The present invention adopts the following technical solutions to achieve the above-mentioned object. The present invention provides a novel low-complexity error vector magnitude calculation method, including:

[0005] S1. Determine the modulation type of the signal;

[0006] S2. Perform normalization processing according to the modulation type of the signal;

[0007] S3, folding and translating the normalized signal;

[0008] S4, determining whether the constellation image is equivalent to QPSK (Quadrature Phase Shift Keying) modulation, if so, proceeding to step S5, otherwise returning to step S3;

[0009] S5. Calculate the EVM of the signal after the folding transformation using the QPSK unit circle calculation method.

[0010] Furthermore, determining the modulation type of the signal specifically includes:

[0011] For test instruments, the modulation type of the signal is determined by the parameters entered by the user.

[0012] Furthermore, the normalization process according to the modulation type of the signal specifically includes:

[0013] Based on the determined modulation type, the coefficients are obtained by comparing the amplitude of the standard constellation diagram with the maximum amplitude of the demodulated constellation diagram, and then the overall normalization is performed as follows:

[0014] Y = e X, X is the constellation point before normalization, e is the coefficient obtained by comparison, and Y is the constellation point after normalization.

[0015] Furthermore, folding the normalized signal specifically includes:

[0016] For higher-order modulations above QPSK, the symbols of the IQ (In-phase and Quadrature) signal are removed and all constellation points are concentrated in the first quadrant.

[0017] Furthermore, performing translation on the normalized signal specifically includes:

[0018] Translate the center point of the first quadrant constellation points to the coordinate origin.

[0019] Furthermore, determining whether the constellation image is equivalent to QPSK modulation specifically includes:

[0020] Whether the constellation image is equivalent to QPSK modulation is determined based on the number of folding times. Specifically, according to the definition of 16QAM and higher-order modulation and the representation of the constellation diagram, for 16n QAM, folding n times means that the corresponding constellation image is equivalent to QPSK modulation, where n is an integer greater than or equal to 1.

[0021] Furthermore, the calculation method of the QPSK unit circle is used to calculate the EVM of the signal after the folding transformation, which specifically includes:

[0022] The constellation diagram after folding transformation is in QPSK state, and the EVM size is calculated by vector as follows:

[0023] err = P2 - P1;

[0024] errPower = sum(err. conj(err));

[0025] sigPower = sum(P2. conj(P2));

[0026] EVM = sqrt(errPower / sigPower) 100;

[0027] Where P1 represents the constellation point of the ideal QPSK, P2 represents the constellation point of the actual received signal, err represents the error vector, conj() represents the conjugate operation, sum() represents the summation operation, errPower represents the error energy, sigPower represents the energy of the actual received signal, and sqrt() represents the square root operation.

[0028] The beneficial effects of the present invention are:

[0029] This invention utilizes a CPU pipeline and vector operations method. Only the modulation type needs to be determined, and subsequent operations are essentially the same for different orders of QAM modulation. This eliminates pipeline idleness caused by the introduction of judgment during processing, ensuring maximum process load and improving processing speed. AVX vector operations also provide a fast method for removing IQ signal symbols and performing signal shifting operations, further improving processing efficiency and accelerating EVM calculations for high-sampling-rate signals while maintaining consistent accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a flow chart of a novel low-complexity error vector magnitude calculation method provided by an embodiment of the present invention;

[0031] Figure 2 It is a schematic diagram of signal folding and translation provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0032] To make the objectives, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0033] The present invention provides a novel low-complexity error vector magnitude calculation method, such as Figure 1 As shown, specifically including:

[0034] S1. Determine the modulation type of the signal;

[0035] S2. Perform normalization processing according to the modulation type of the signal;

[0036] S3, folding and translating the normalized signal;

[0037] S4, determine whether the constellation image is equivalent to QPSK modulation, if so, go to step S5, otherwise return to step S3;

[0038] S5. Calculate the EVM of the signal after the folding transformation using the QPSK unit circle calculation method.

[0039] Specifically, determining the modulation type of the signal includes:

[0040] For test instruments, the modulation type of the signal is determined by the parameters entered by the user.

[0041] Specifically, the normalization process according to the modulation type of the signal includes:

[0042] Based on the determined modulation type, the coefficients are obtained by comparing the amplitude of the standard constellation diagram with the maximum amplitude of the demodulated constellation diagram, and then the overall normalization is performed as follows:

[0043] Y = e X, X is the constellation point before normalization, e is the coefficient obtained by comparison, and Y is the constellation point after normalization.

[0044] Specifically, such as Figure 2 As shown in FIG, folding the normalized signal specifically includes:

[0045] For higher-order modulations above QPSK, the symbols of the IQ signal are removed and all constellation points are concentrated in the first quadrant.

[0046] Specifically, such as Figure 2 As shown, the translation of the normalized signal specifically includes:

[0047] Translate the center point of the first quadrant constellation points to the coordinate origin.

[0048] Specifically, determining whether the constellation image is equivalent to QPSK modulation includes:

[0049] Whether the constellation image is equivalent to QPSK modulation is determined based on the number of folding times. Specifically, according to the definition of 16QAM and higher-order modulation and the constellation diagram representation, for 16n QAM, folding n times means that the corresponding constellation image is equivalent to QPSK modulation, where n is an integer greater than or equal to 1. For example, for 16QAM, folding once is sufficient, and for 32QAM, folding twice is sufficient.

[0050] Specifically, the calculation method of the QPSK unit circle is used to calculate the EVM of the signal after the folding transformation, which specifically includes:

[0051] The constellation diagram after folding transformation is in QPSK state, and the EVM size is calculated by vector as follows:

[0052] err = P2 - P1;

[0053] errPower = sum(err. conj(err));

[0054] sigPower = sum(P2. conj(P2));

[0055] EVM = sqrt(errPower / sigPower) 100;

[0056] Where P1 and P2 are constellation points on the complex plane, P1 represents the constellation point under the ideal QPSK condition, P2 represents the constellation point of the actual received signal, err represents the error vector, conj() represents the conjugate operation, sum() represents the summation operation, errPower represents the error energy, sigPower represents the energy of the actual received signal, and sqrt() represents the square root operation.

[0057] The foregoing description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein and should not be construed as excluding other embodiments. Rather, the present invention can be used in various other combinations, modifications, and environments and can be modified within the scope of the concept described herein through the above teachings or techniques or knowledge in the relevant field. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention are intended to be protected by the appended claims.

Claims

1. A novel low-complexity error vector magnitude calculation method, characterized in that: include: S1. Determine the modulation type of the signal; S2. Perform normalization processing according to the modulation type of the signal; S3, folding and translating the normalized signal; S4, determine whether the constellation image is equivalent to QPSK modulation, if so, go to step S5, otherwise return to step S3; S5. Calculate the EVM of the signal after the folding transformation using the QPSK unit circle calculation method.

2. The novel low-complexity EVM calculation method according to claim 1, characterized in that: The modulation type of the judgment signal specifically includes: For test instruments, the modulation type of the signal is determined by the parameters entered by the user.

3. The novel low-complexity EVM calculation method according to claim 2, wherein: Normalization processing based on the modulation type of the signal specifically includes: Based on the determined modulation type, the coefficients are obtained by comparing the amplitude of the standard constellation diagram with the maximum amplitude of the demodulated constellation diagram, and then the overall normalization is performed as follows: Y = e X, X is the constellation point before normalization, e is the coefficient obtained by comparison, and Y is the constellation point after normalization.

4. The novel low-complexity EVM calculation method according to claim 1, wherein: Folding the normalized signal specifically includes: For higher-order modulations above QPSK, the symbols of the IQ signal are removed and all constellation points are concentrated in the first quadrant.

5. The novel low-complexity EVM calculation method according to claim 1, wherein: The translation of the normalized signal specifically includes: Translate the center point of the first quadrant constellation points to the coordinate origin.

6. The novel low-complexity EVM calculation method according to claim 1, wherein: Determining whether the constellation image is equivalent to QPSK modulation specifically includes: Whether the constellation image is equivalent to QPSK modulation is determined based on the number of folding times. Specifically, according to the definition of 16QAM and higher-order modulation and the representation of the constellation diagram, for 16n QAM, folding n times means that the corresponding constellation image is equivalent to QPSK modulation, where n is an integer greater than or equal to 1.

7. The novel low-complexity EVM calculation method according to claim 1, wherein: The calculation method of the QPSK unit circle is used to calculate the EVM of the signal after the folding transformation, which specifically includes: The constellation diagram after folding transformation is in QPSK state, and the EVM size is calculated by vector as follows: err = P2 - P1; errPower = sum(err. conj(err)); sigPower = sum(P2. conj(P2)); EVM = sqrt(errPower / sigPower) 100; Where P1 represents the constellation point of the ideal QPSK, P2 represents the constellation point of the actual received signal, err represents the error vector, conj() represents the conjugate operation, sum() represents the summation operation, errPower represents the error energy, sigPower represents the energy of the actual received signal, and sqrt() represents the square root operation.

Images