CN104486273B - An Adaptive Direct Quadrature Frequency Conversion Modulation Error Correction Method - Google Patents

Abstract

The invention discloses a self-adaptive direct quadrature frequency modulation error correction method. The method establishes a new quadrature modulation error model and provides a self-adaptive QMC algorithm for correcting the direct quadrature frequency conversion modulation error on the basis of the model. Through antenna coupling and down-conversion frequency mixing, a feedback signal is obtained after intermediate-frequency high-speed sampling, after the feedback signal and a transmitting signal are aligned, an orthogonal modulation error inverse distortion coefficient is estimated, and self-adaptive correction is realized in a digital domain programmable logic device FPGA. The method can efficiently and accurately inhibit local oscillator leakage caused by direct current bias and image interference caused by IQ amplitude phase imbalance, and has important significance for error correction of self-adaptive direct alternating frequency modulation.

Description

An Adaptive Direct Quadrature Frequency Conversion Modulation Error Correction Method

technical field

The invention relates to an adaptive direct quadrature frequency conversion modulation error correction method, which belongs to the technical field of digital signal processing.

Background technique

In a digital communication transmission system, it is necessary to up-convert the baseband signal to a radio frequency, which is beneficial to the efficient transmission of the signal and improves the utilization of spectrum resources. Real signal double or triple frequency conversion converts the digital low-IF signal into an analog low-IF signal, and then performs one or two frequency conversion to the specified frequency point after band-pass filtering. The performance consistency of the multiple frequency conversion system is high, and the local oscillator leakage and image spectrum can be eliminated by filtering. The disadvantage is that it has high requirements for circuit and system design, and the phase noise is poor. In recent years, a new type of direct frequency conversion technology based on quadrature modulation, that is, direct quadrature frequency conversion technology has been developed rapidly. This technology uses the principle of orthogonal cancellation to eliminate unwanted sideband signals and suppress image spectrum. Its outstanding advantage is that it does not need intermediate frequency amplification, filtering and frequency conversion parts, and relaxes the performance requirements for the radio frequency part filter, and even does not need to add a radio frequency filter, thus greatly reducing the volume, weight, power consumption and cost. However, the direct quadrature frequency conversion technology has high requirements on the amplitude-phase balance of the baseband signal and the local oscillator signal. The unbalanced amplitude and phase will reduce the image frequency interference suppression ability of the frequency converter, and the adjacent channel power suppression ratio of the transmitter will decrease. At the same time, due to the inevitable existence of crosstalk, radiation, DC offset modulation and other problems in the circuit, local oscillator leakage is caused, and because the useful signal and the leaked local oscillator signal are very close in frequency spectrum, it is impossible to use a filter to filter remove. On the one hand, the leakage signal of the local oscillator will reduce the efficiency of the transmitter, and on the other hand, it will cause the DC operating point of the receiver to shift, easily causing nonlinear distortion, and even saturation blocking phenomenon.

After searching the existing technology, the Chinese patent application number is CN200810207707, and the technical literature with the publication number CN1707962A discloses an adaptive balance system for IQ amplitude in direct frequency conversion modulation, which can automatically Monitor, track and compensate the IQ amplitude difference of all devices in the entire up-conversion link due to ambient temperature and humidity changes, and the adaptive feedback structure has high precision. The Chinese patent application number is CN200810207707, and the publication number CN 101478287B discloses an adaptive elimination system for carrier leakage in direct conversion modulation, which can automatically monitor, track and compensate all the signals in the entire up-conversion link without interrupting the main signal. The DC component in the device caused by changes in ambient temperature and humidity, and the adaptive feedback structure has high precision.

Therefore, it has become a new technical demand to design a direct quadrature frequency conversion modulation method that can suppress the image interference signal and eliminate the leakage of the local oscillator caused by the DC bias.

Contents of the invention

(1) Technical problems to be solved

The present invention aims at the above-mentioned deficiencies existing in the prior art, by analyzing the cause of the error in the quadrature modulation process, a new error model is established, and a self-adaptive QMC (QuadratureModulation Compensation) algorithm for direct quadrature frequency conversion modulation error correction is proposed, The algorithm can not only suppress the image interference signal caused by the amplitude-phase imbalance, but also eliminate the local oscillator leakage caused by the DC bias.

Establish a mathematical model for the quadrature modulation error in the digital baseband. The ideal quadrature modulation signal is as follows:

Among them, Re represents the real part in the complex number operation, I represents the in-phase signal, Q represents the quadrature signal, j represents the imaginary unit in the complex number, w _c represents the angular frequency, and t represents the continuous time signal. When the baseband signal s(t ) is superimposed with a DC component, then there is:

Among them, I _DC represents the in-phase DC component, and Q _DC represents the quadrature DC component. Assuming that the quadrature modulator is completely orthogonal, then the RF signal obtained after quadrature modulation is:

It can be seen that when the IQ is superimposed with a DC component, the modulated signal will generate an additional carrier leakage signal at the carrier frequency in addition to the desired radio frequency signal.

The problem of IQ imbalance widely occurs in communication systems, and it is manifested in the fact that the real and imaginary parts of the signal are not strictly orthogonal, or the power is unequal during the communication process. In order to simplify the analysis, take the I road as the benchmark, and only calculate the error introduced in the Q road.

When there is only amplitude imbalance, set the amplitude gain of the Q channel as β, which is:

Q'=(1+β)Q (4)

The RF signal obtained after quadrature modulation will appear a double-sideband signal with the carrier as the center frequency:

When there is only phase imbalance, the offset error of the Q channel is θ, and the signal of the Q channel is modulated to the Q' channel, which causes the amplitude change of the actual Q channel and introduces Q channel interference in the I channel:

After the baseband complex signal with phase imbalance is modulated, a double sideband interference signal will be generated:

From equations (2), (4) and (6), the mathematical model expression of quadrature modulation error can be obtained:

Extending formula (8) to a more general situation and expressing it in matrix form, the following error model can be obtained:

Among them, a _i represents the coefficient of QMC. Equation (9) directly describes the unbalanced gain and superimposed DC offset of the IQ signal during transmission, and indirectly describes the phase imbalance of the IQ signal in the form of IQ crosstalk. distortion.

(2) Technical solutions

In the wireless transmission system, the invention couples the antenna and obtains the feedback signal through intermediate frequency high-speed sampling, and after aligning the feedback signal and the transmission signal, estimates the quadrature modulation error inverse distortion QMC coefficient, and realizes the self-distortion in the digital domain chip FPGA. Adaptive correction, the method can efficiently and accurately suppress the local oscillator leakage caused by the DC bias and the image interference caused by the IQ amplitude and phase imbalance, the adaptive direct quadrature frequency conversion modulation error correction method proposed by the present invention, the method includes Follow the steps below:

S1: Initialize the QMC coefficient, and intercept baseband data with a length of L from the input digital baseband signal, and correct the baseband data through a QMC equalizer;

S2: Perform digital-to-analog conversion on the corrected data, and transmit the signal after orthogonal modulation;

S3: Receive the feedback signal through antenna coupling, perform down-conversion mixing and intermediate frequency sampling on the feedback signal, and send the obtained digital signal y(n) to FPGA for quadrature demodulation;

S4: Estimate the delay of the demodulated signal, calculate the delay time, and perform delay alignment processing on the original baseband data z(n);

S5: Estimate QMC coefficients.

Preferably, the error model of the quadrature modulation is shown in formula (9):

Among them, a ₁ , a ₂ , a ₃ , a ₄ , a ₅ , and a ₆ represent QMC coefficients. The model directly describes the unbalanced gain and superimposed DC offset of the IQ signal during transmission, and indirectly uses The form of IQ crosstalk describes the distortion caused by the phase imbalance of the IQ signal. According to the error model, the quadrature modulation error inverse distortion coefficient a _i can be obtained.

Preferably, in S3, the feedback signal is obtained through antenna coupling and intermediate frequency high-speed sampling, and the quadrature modulation error inverse distortion coefficient is estimated after alignment processing is performed on the feedback signal and the transmitted signal.

Preferably, the linear equation of the estimated QMC coefficient in said S5 is as follows:

Z=UA (10)

Among them, Z represents the column vector {z(n), z(n-1), ..., z(n-L+1)} ^T , and A represents the QMC coefficient vector [a ₁ , a ₂ , a ₃ , a ₄ , a ₅ , a ₆ ] ^T , U is the matrix corresponding to coefficient A U=Re{Re(Y), Im(Y), ones(L, 1), j·Re(Y), j·Im(Y ), j·ones(L, 1)}, where Re represents the real part in the complex number operation, Im represents the imaginary part in the complex number operation, ones represents that the elements in the matrix are all 1, and j represents the imaginary number unit in the complex number , the least squares algorithm can be used to solve the linear equation to obtain the estimated value of A, that is, the QMC coefficient.

Preferably, S5 is followed by quadrature modulation error judgment. When the error is less than 10 ^-4 , it can be ignored, and the error range can be determined according to the system accuracy. If the error is greater than 10 ^-4 , the coefficient is passed to the QMC equalizer Continue to calibrate, if the error is less than 10 ^-4 , stop the calibration.

(3) Beneficial effects

It can be seen from the above technical solutions that the self-adaptive direct quadrature frequency conversion modulation error correction method proposed by the present invention can not only suppress the image interference caused by the unbalanced amplitude and phase, but also eliminate the leakage of the local oscillator caused by the DC bias at the same time. It is of great significance to adapt to the error correction of direct quadrature frequency conversion modulation.

Description of drawings

Fig. 1 has shown the functional block diagram of the self-adaptive direct quadrature frequency conversion modulation error correction method that the present invention proposes;

Fig. 2 has shown the system block diagram of the specific embodiment of the present invention;

Fig. 3 shows the QMC algorithm flowchart of the preferred embodiment of the present invention;

Fig. 4 shows a graph of signal processing simulation results of a specific embodiment of the present invention.

detailed description

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in combination with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are exemplary only, and are not intended to limit the scope of the present invention. Also, in the following description, descriptions of well-known structures and techniques are omitted to avoid unnecessarily obscuring the concept of the present invention.

Fig. 1 shows a functional block diagram of an adaptive direct quadrature frequency conversion modulation error correction method proposed by the present invention.

As shown in Figure 1, in the adaptive direct quadrature frequency conversion modulation error correction method proposed by the present invention, after the data is input, it first passes through the QMC equalizer 100, and then performs direct quadrature frequency conversion 101, and then transmits the data after the signal is modulated to a radio frequency. Antenna coupling 102 receives the feedback signal, performs down-conversion 103 frequency mixing and intermediate frequency sampling 104 on the feedback signal, sends the obtained digital signal y(n) to FPGA for demodulation and digital alignment 105, then performs QMC parameter estimation 106, and finally performs Quadrature modulation error decision 107, and decide whether to transmit QMC equalizer coefficients according to the decision result.

Fig. 2 shows a system block diagram of a specific embodiment of the present invention.

As shown in FIG. 2 , in a specific embodiment of the present invention, the encoded and modulated baseband data of a specific standard is input to the FPGA, and quadrature modulation error correction is performed in the FPGA. Initialize the QMC coefficient to 1, and set it to [1, 0, 0, 0, 1, 0]. After the baseband data passes through the QMC201 module, the baseband signal remains unchanged, and is still divided into I and Q signals, and then performs digital-to-analog conversion 202 , convert the digital baseband signal into an analog signal, then perform quadrature modulation 203, modulate the baseband signal to the frequency of the carrier, pass through the amplifier 204 and amplify the signal power, and transmit the radio frequency signal through the antenna 205. The system has a feedback loop, and the feedback data is obtained through antenna coupling, and the feedback signal is subjected to down-conversion and mixing processing through the mixer 206 to obtain an intermediate frequency signal, and then performs analog-to-digital conversion 207 to convert the analog signal into a digital signal, and then performs filtering 208 and demodulation 209 to obtain the error-containing signal y(n), at this time, the signal is expressed as I' and Q', and the delay estimation is performed on the demodulated signal y(n), and the delay time is calculated, and the original The baseband data z(n) is subjected to delay alignment processing 210, and the feedback signal y(n) and the aligned baseband data z(n) are used to estimate the QMC coefficient estimation 211, and the vector of z(n) and y(n) is used to determine whether the positive Intermodulation error 212, if there is an error, the QMC coefficient is passed to the QMC module for correction, and if there is no error, the correction is stopped.

Fig. 3 shows the flowchart of the QMC algorithm of the preferred embodiment of the present invention.

As shown in Figure 3, the QMC algorithm flowchart of the preferred embodiment of the present invention specifically includes the following steps:

S1: QMC coefficient initialization, used to transmit unprecorrected signals;

S2: Intercept the baseband data of length L from the input digital baseband signal, the value range of L is set according to the signal bandwidth and sampling rate, and the general range is [512, 8192];

S3: Pre-correct the data in S2 through the QMC equalizer to offset the local oscillator leakage and image interference of the radio frequency link;

S4: Perform digital-to-analog conversion on the corrected data, and convert the baseband digital signal into an analog signal, which is conducive to high-speed signal transmission;

S5: performing quadrature modulation on the data in S4;

S6: transmitting radio frequency signals through the antenna;

S7: receiving the feedback signal through antenna coupling;

S8: down-convert the received radio frequency signal, and perform intermediate frequency filtering to filter out-of-band noise;

S9: Perform intermediate frequency sampling on the signal of S8, and sample the continuous time signal into a discrete time digital signal;

S10: Perform quadrature demodulation on the digital signal in S9 to obtain two signals of I' and Q' containing errors;

S11: Estimate the delay of the demodulated signal y(n), calculate the delay time, and perform delay alignment processing on the original baseband data z(n);

S12: Estimate the QMC coefficient with the feedback signal y(n) and the aligned baseband data z(n), and solve the linear equation of the QMC coefficient as shown in formula (10):

Z=UA (10)

Among them, Z represents the column vector {z(n), z(n-1), ... z(n-L+1)} ^T , and A represents the QMC coefficient vector [a ₁ , a ₂ , a ₃ , a ₄ , a ₅ , a ₆ ] ^T . U is the matrix corresponding to the coefficient A:

U=Re{Re(Y), Im(Y), ones(L, 1), j·Re(Y), j·Im(Y), j·ones(L, 1)}.

The least squares algorithm can be used to solve the linear equation (10) to obtain the estimated value of A, that is, the QMC coefficient;

S13: Use the vectors of z(n) and y(n) to determine the quadrature modulation error; if there is an error, continue to correct, otherwise stop the correction;

S14: Send the QMC coefficient estimated in S13 to the QMC equalizer, execute S2, and continue to correct.

Fig. 4 shows a graph of signal processing simulation results of a specific embodiment of the present invention.

As shown in Figure 4, the baseband input signal is a single-sided complex multi-tone signal, the carrier frequency is 40MHz, and the estimated sample length L=1024, the RF output result is shown in Figure 4, and the left picture (a) is without QMC correction The schematic diagram of the amplitude-frequency response of the RF signal. The figure (b) on the right is a schematic diagram of the amplitude-frequency response of the RF signal with QMC correction. From the schematic diagram of the amplitude-frequency response of the RF signal, it can be seen that after adding QMC correction, the local oscillator leakage is basically eliminated. , At the same time, the image interference signal generated by the IQ imbalance is obviously suppressed, and the optimization range exceeds 40dB.

In summary, the present invention proposes an adaptive direct quadrature frequency conversion modulation error correction method, which can not only suppress the image interference caused by the unbalanced amplitude and phase, but also eliminate the local oscillator leakage caused by the DC bias at the same time. It is of great significance to the error correction of adaptive direct quadrature frequency conversion modulation.

It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principle of the present invention, and not to limit the present invention. Therefore, any modification, equivalent replacement, improvement, etc. made without departing from the spirit and scope of the present invention shall fall within the protection scope of the present invention. Furthermore, it is intended that the appended claims of the present invention embrace all changes and modifications that come within the scope and metesques of the appended claims, or equivalents of such scope and metes and bounds.

Claims (5)

1. an adaptive direct quadrature frequency conversion modulation error correction method, is characterized in that, described method comprises the steps: S1: Initialize the QMC coefficient, and intercept baseband data with a length of L from the input digital baseband signal, and correct the baseband data through a QMC equalizer; S2: Perform digital-to-analog conversion on the corrected data, and transmit the signal after orthogonal modulation; S3: Receive the feedback signal through antenna coupling, perform down-conversion mixing and intermediate frequency sampling on the feedback signal, and send the obtained digital intermediate frequency signal to FPGA for quadrature demodulation; S4: Estimate the delay of the demodulated signal y(n), calculate the delay time, and perform delay alignment processing with the original baseband data z(n); S5: Estimate QMC coefficients by using the feedback signal and the aligned baseband data; The linear equation for estimating the QMC coefficients is as follows: Z=UA Among them, Z represents the column vector {z(n),z(n-1),...z(n-L+1)} ^T , and A represents the QMC coefficient vector [a ₁ ,a ₂ ,a ₃ ,a ₄ ,a ₅ ,a ₆ ] ^T , U is the matrix corresponding to coefficient A U=Re{Re(Y),Im(Y),ones(L,1),j·Re(Y),j· Im(Y), j ones(L,1)}, where Re represents the real part in the complex number operation, Im represents the imaginary part in the complex number operation, ones represents that the elements in the matrix are all 1, and j represents the complex number The imaginary unit of , the least square algorithm is used to solve the linear equation, and the estimated value of A is obtained, that is, the QMC coefficient. 2. The adaptive direct quadrature frequency conversion modulation error correction method according to claim 1, characterized in that: the value range of L in the S1 is determined according to the signal bandwidth and sampling rate. 3. adaptive direct quadrature frequency conversion modulation error correction method according to claim 1, is characterized in that: the error model of described quadrature modulation is I'＝a ₁ I+a ₂ Q+a ₃ Q'＝a ₄ Q+a ₅ Q+a ₆ Among them, a ₁ , a ₂ , a ₃ , a ₄ , a ₅ , and a ₆ represent QMC coefficients. The model directly describes the unbalanced gain and superimposed DC offset of the IQ signal during transmission, and indirectly uses The form of IQ crosstalk describes the distortion caused by the phase imbalance of the IQ signal. According to the error model, the quadrature modulation error inverse distortion coefficient a _i can be obtained. 4. the self-adaptive direct quadrature frequency conversion modulation error correction method according to claim 1, is characterized in that: in S3, through antenna coupling, and obtain feedback signal by mid-frequency high-speed sampling, after feedback signal and transmission signal are carried out alignment processing, Estimate quadrature modulation error inverse distortion coefficients. 5. The adaptive direct quadrature frequency conversion modulation error correction method according to claim 1, characterized in that: after said S5, the quadrature modulation error judgment is performed, and when the error is less than 10 ⁻⁴ , it is ignored; if the error is greater than 10 ^{− 4} , the coefficient is passed to the QMC equalizer to continue the correction, and if the error is less than 10 ^-4 , the correction is stopped.