Device and method for correcting broadband orthogonal imbalance

Abstract

The invention discloses a broadband orthogonal imbalance correction device and method. The device comprises a reference signal generation module, a broadband signal generation module, a mode selection module, a signal processing module, a compensation module and a correction module. The mode selection module selects an off-line calculation mode or an on-line compensation mode, in an off-line state, a reference analog signal generated by the reference signal generation module is converted into a reference digital signal through the signal processing module, and the compensation module obtains a compensation coefficient according to the reference digital signal. And after the off-line state is switched to the on-line state, a broadband analog signal generated by the broadband signal generation module is converted into a broadband digital signal through the signal processing module, and the correction module corrects the signal in combination with the compensation coefficient to obtain a broadband correction signal. The mode selection module flexibly switches an off-line calculation mode and an on-line compensation mode, so that the mode selection module is high in application flexibility and can adapt to different bandwidths and modulation modes, and compared with a traditional iterative calculation mode, compensation coefficients can be dynamically updated, and frequency-related orthogonal imbalance compensation is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of receiver correction, and in particular to a correction device and method for broadband orthogonal imbalance. Background Art

[0002] In modern wireless communication systems, zero-IF and low-IF architectures are widely used due to their advantages, such as eliminating IF processing steps and simplifying receiver structures. However, errors introduced by local oscillator signals and mismatches in analog link components can cause orthogonal imbalance in the receiver. In narrowband systems, these imbalances primarily manifest as fixed amplitude and phase errors, which can be corrected using simple compensation algorithms.

[0003] However, in broadband systems, this orthogonal imbalance exhibits frequency dependence because the device response varies with frequency. Traditional fixed parameter compensation methods do not have the ability to update online and cannot adapt to environmental changes to compensate for the I / Q mismatch that varies with frequency in broadband systems.

[0004] Therefore, the prior art cannot effectively compensate for frequency-related quadrature imbalance. Summary of the Invention

[0005] In order to overcome the deficiencies of the prior art, the present invention aims to provide a device and method for correcting broadband quadrature imbalance, which corrects broadband signals by dynamically generating compensation coefficients to compensate for frequency-related quadrature imbalance.

[0006] In order to solve the above problems, the present invention is implemented according to the following scheme: Provided is a broadband quadrature imbalance correction device, comprising: a reference signal generation module, a broadband signal generation module, a mode selection module, a signal processing module, a compensation module, and a correction module; the mode selection module is connected to the reference signal generation module, the broadband signal generation module, and the signal processing module; the signal processing module is connected to the compensation module and the correction module; and the compensation module is connected to the correction module; The mode selection module is used to select an offline calculation mode or an online compensation mode. When entering the offline calculation mode, the reference analog signal generated by the reference signal generation module is converted into a reference digital signal by the signal processing module, and the compensation module obtains a compensation coefficient based on the reference digital signal; When entering the online compensation mode from the offline calculation mode, the broadband analog signal generated by the broadband signal generation module is converted into a broadband digital signal by the signal processing module, and the correction module corrects the broadband digital signal according to the broadband digital signal and the compensation coefficient to obtain a broadband correction signal.

[0007] Compared with the prior art, the beneficial effects of the broadband orthogonal imbalance correction device of the present invention are as follows: the mode selection module flexibly switches between offline calculation and online compensation modes, making its application flexible and adaptable to different bandwidths and modulation methods. Compared with the traditional iterative calculation method, it can also dynamically update the compensation coefficient, reduce the calculation complexity, and enhance environmental adaptability to cope with temperature changes, device aging and other problems, and realize compensation for frequency-related orthogonal imbalance.

[0008] Optionally, the signal processing module includes an I-channel processing unit and a Q-channel processing unit with two identical structures; The I-path processing unit and the Q-path processing unit both include a filter and an analog-to-digital converter; the filter is connected to the analog-to-digital converter and the mode selection module, and the analog-to-digital converter is connected to the compensation module and the correction module.

[0009] Optionally, the correction module includes a calculation unit, a delay unit and an output unit; The calculation unit is connected to the output unit, the Q-path processing unit, and the compensation module; the delay unit is connected to the output module and the I-path processing unit.

[0010] A broadband quadrature imbalance correction method is also provided, which is applied to the broadband quadrature imbalance correction device described above, comprising: When the mode selection module selects the offline calculation mode, the reference analog signal generated by the reference signal generation module is converted into a reference digital signal by the signal processing module; The compensation module obtains a compensation coefficient according to the reference digital signal; When the mode selection module switches from the offline calculation mode to the online compensation mode, the broadband analog signal generated by the broadband signal generation module is converted into a broadband digital signal by the signal processing module; The correction module corrects the broadband digital signal according to the broadband digital signal and the compensation coefficient to obtain a broadband correction signal.

[0011] Optionally, the reference analog signal includes sinusoidal reference signals corresponding to multiple target frequency points; The reference signal generating module generates the reference analog signal, including: Obtaining a frequency range of a filter in the signal processing module; Determining a plurality of target frequency points having a generation order according to the frequency range; According to the generation order, sinusoidal reference signals corresponding to a plurality of target frequency points are generated.

[0012] Optionally, the reference digital signal includes an I-path reference signal and a Q-path reference signal corresponding to a plurality of target frequency points; The compensation module obtains a compensation coefficient according to the reference digital signal, including: Summing the I-channel reference signals corresponding to the multiple target frequency points to obtain an I-channel total signal, and summing the Q-channel reference signals corresponding to the multiple target frequency points to obtain a Q-channel total signal; The compensation coefficient is obtained according to the I-channel total signal and the Q-channel total signal.

[0013] Optionally, obtaining the compensation coefficient according to the I-channel total signal and the Q-channel total signal includes: Performing autocorrelation calculation on the Q-path total signal to obtain an autocorrelation vector; Performing cross-correlation calculation on the I-channel total signal and the Q-channel total signal to obtain a cross-correlation vector; The compensation coefficient is obtained according to the autocorrelation vector and the cross-correlation vector.

[0014] Optionally, obtaining the compensation coefficient according to the autocorrelation vector and the cross-correlation vector includes: According to the autocorrelation vector, a Toeplitz matrix is ​​obtained; The compensation coefficient is obtained according to the Toeplitz matrix and the cross-correlation vector.

[0015] Optionally, the broadband digital signal includes an I-channel broadband signal and a Q-channel broadband signal; The correction module corrects the broadband digital signal according to the broadband digital signal and the compensation coefficient to obtain a broadband correction signal, including: Determining a delay to be compensated according to the Q-channel broadband signal and the compensation coefficient; Performing delay processing on the I-channel broadband signal according to the delay to be compensated, to obtain an I-channel correction signal having a synchronous code stream output with the Q-channel broadband signal; The I-channel correction signal and the Q-channel broadband signal constitute the broadband correction signal.

[0016] Optionally, determining a delay to be compensated according to the Q-channel broadband signal and the compensation coefficient includes: A convolution operation is performed on the Q-channel broadband signal and the compensation coefficient, and a symbol delay generated during the operation is used as the delay to be compensated. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a structural block diagram of the correction device of the present invention; Figure 2 is a structural diagram of the correction device of the present invention; Figure 3 This is a hardware control block diagram of the correction device of the present invention; Figure 4 Schematic diagram of coefficients corresponding to compensation filters of different orders of the present invention; Figure 5 The performance test results of the present invention are shown as follows Figure 1 ; Figure 6 The performance test results of the present invention are shown as follows Figure 2 ; Explanation of the accompanying drawings: 1. Reference signal generating module; 2. Broadband signal generating module; 3. Mode switching module; 4. Signal processing module; 401. I-channel processing unit; 402. Q-channel processing unit; 5. Compensation module; 6. Correction module; 601. Calculation unit; 602. Delay unit; 603. Output unit. DETAILED DESCRIPTION

[0018] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0019] When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0020] See also Figure 1-2 As shown, a broadband orthogonal imbalance correction device of the present invention includes: a reference signal generation module 1, a broadband signal generation module 2, a mode selection module 3, a signal processing module 4, a compensation module 5 and a correction module 6; the mode selection module 3 is connected to the reference signal generation module 1, the broadband signal generation module 2, and the signal processing module 4, the signal processing module 4 is connected to the compensation module 5 and the correction module 6, and the compensation module 5 is connected to the correction module 6.

[0021] In one embodiment of the present invention, the signal processing module 4 includes two identical structures, an I-path processing unit 401 and a Q-path processing unit 402; the I-path processing unit 401 and the Q-path processing unit 402 both include a filter and an analog-to-digital converter; the filter is connected to the analog-to-digital converter and the mode selection module 3, and the analog-to-digital converter is connected to the compensation module 5 and the correction module 6.

[0022] In one embodiment of the present invention, the correction module 6 includes a calculation unit 601, a delay unit 602 and an output unit 603; the calculation unit 601 is connected to the output unit 603, the Q-path processing unit 402 and the compensation module 5, and the delay unit 602 is connected to the output module and the I-path processing unit 401.

[0023] In the present invention, the mode selection module 3 is used to select an offline calculation mode or an online compensation mode. When entering the offline calculation mode, the reference analog signal generated by the reference signal generation module 1 is converted into a reference digital signal by the signal processing module 4, and the compensation module 5 obtains a compensation coefficient based on the reference digital signal; when entering the online compensation mode from the offline calculation mode, the broadband analog signal generated by the broadband signal generation module 2 is converted into a broadband digital signal by the signal processing module 4, and the correction module 6 corrects the broadband digital signal according to the broadband digital signal and the compensation coefficient to obtain a broadband correction signal.

[0024] Combine Figure 2 The working process of the correction device of the present invention is described as follows: When the mode selection module 3 is se1=00, the correction device enters the offline calculation mode. At this time, the reference signal generation module 1 generates a reference analog signal Ref signal and transmits it to the signal processing module 4. The reference analog signal Ref signal is converted into a reference digital signal and output to the compensation module 5. The compensation module 5 calculates the compensation coefficient according to the reference digital signal. Among them, the reference analog signal generated by the reference signal generation module 1 is a series of reference single-tone signals (sinusoidal reference signals), the frequency distribution is determined by the filter bandwidth in the I-way processing unit 401 / Q-way processing unit 402, and the signal length is 50 samples. The specific calculation method is as follows: FreN = linspace(0,BW,16).' Ref signal = real(exp(i.*2.*pi.*(FreN).*t)) Wherein, FreN is the frequency of the reference single tone signal, BW is the filter bandwidth, and Ref signal is the reference single tone signal generated according to the frequency distribution determined by the filter bandwidth.

[0025] When the mode selection module 3 is se1=01, the correction device enters the online compensation mode from the offline calculation mode. At this time, the broadband signal generation module 2 generates a broadband analog signal Chirp / QPSK signal_in and transmits it to the I-way processing module and the Q-way processing module respectively, and converts it into an I-way broadband signal and a Q-way broadband signal (broadband digital signal). The calculation module calculates the Q-way broadband signal and the compensation coefficient. , determine the delay to be compensated for compensating the I-channel broadband signal, and the delay module performs delay processing on the I-channel broadband signal according to the delay to be compensated, ensuring that the I-channel correction signal and the Q-channel correction signal (broadband correction signal) output by the output module have synchronized code stream outputs.

[0026] A broadband quadrature imbalance correction method of the present invention is applied to the broadband quadrature imbalance correction device described above, comprising: S1: When the mode selection module 3 selects the offline calculation mode, the reference analog signal generated by the reference signal generation module 1 is converted into a reference digital signal by the signal processing module 4; when the mode selection module 3 is se1=00, the correction device enters the offline calculation mode.

[0027] In one embodiment of the present invention, the reference analog signal includes a plurality of sinusoidal reference signals corresponding to target frequency points; the reference signal generation module 1 generates the reference analog signal, including: obtaining the frequency range of the filter in the signal processing module 4; determining a plurality of target frequency points having a generation order according to the frequency range; and generating a plurality of sinusoidal reference signals corresponding to the target frequency points according to the generation order.

[0028] When entering the offline calculation mode, the reference signal generation module 1 generates a reference analog signal Ref signal and transmits it to the signal processing module 4, which converts the reference analog signal Ref signal into a reference digital signal and outputs it to the compensation module 5. The compensation module 5 calculates the compensation coefficient according to the reference digital signal. Among them, the reference analog signal generated by the reference signal generation module 1 is a series of reference single-tone signals (sinusoidal reference signals corresponding to multiple target frequency points), the frequency distribution is determined by the filter bandwidth in the I-way processing unit 401 / Q-way processing unit 402, and the signal length is 50 samples. The specific calculation method is as follows: FreN = linspace(0,BW,16).' Ref signal = real(exp(i.*2.*pi.*(FreN).*t)) Wherein, FreN is the frequency of the reference single tone signal, BW is the filter bandwidth, and Ref signal is the reference single tone signal generated according to the frequency distribution determined by the filter bandwidth.

[0029] S2: The compensation module 5 obtains a compensation coefficient according to the reference digital signal; the compensation coefficient is a multi-order coefficient.

[0030] In one embodiment of the present invention, the reference digital signal includes an I-channel reference signal and a Q-channel reference signal corresponding to multiple target frequency points; the compensation module 5 obtains a compensation coefficient based on the reference digital signal, including: summing the I-channel reference signals corresponding to the multiple target frequency points to obtain an I-channel total signal, and summing the Q-channel reference signals corresponding to the multiple target frequency points to obtain a Q-channel total signal; and obtaining the compensation coefficient based on the I-channel total signal and the Q-channel total signal.

[0031] In one embodiment of the present invention, a compensation coefficient is obtained based on the I-channel total signal and the Q-channel total signal, including: performing autocorrelation calculation on the Q-channel total signal to obtain an autocorrelation vector; performing cross-correlation calculation on the I-channel total signal and the Q-channel total signal to obtain a cross-correlation vector; and obtaining the compensation coefficient based on the autocorrelation vector and the cross-correlation vector.

[0032] In one embodiment of the present invention, obtaining the compensation coefficient according to the autocorrelation vector and the cross-correlation vector includes: obtaining a Toeplitz matrix according to the autocorrelation vector; and obtaining the compensation coefficient according to the Toeplitz matrix and the cross-correlation vector.

[0033] S3: When the mode selection module 3 switches from the offline calculation mode to the online compensation mode, the broadband analog signal generated by the broadband signal generation module 2 is converted into a broadband digital signal by the signal processing module 4.

[0034] S4: The correction module 6 corrects the broadband digital signal according to the broadband digital signal and the compensation coefficient to obtain a broadband correction signal.

[0035] In one embodiment of the present invention, the broadband digital signal includes an I-channel broadband signal and a Q-channel broadband signal; the correction module 6 corrects the broadband digital signal according to the broadband digital signal and the compensation coefficient to obtain a broadband correction signal, including: determining a delay to be compensated according to the Q-channel broadband signal and the compensation coefficient; performing delay processing on the I-channel broadband signal according to the delay to be compensated to obtain an I-channel correction signal having a synchronous code stream output with the Q-channel broadband signal; the I-channel correction signal and the Q-channel broadband signal constitute the broadband correction signal.

[0036] In one embodiment of the present invention, determining the delay to be compensated based on the Q-channel broadband signal and the compensation coefficient includes: performing a convolution operation on the Q-channel broadband signal and the compensation coefficient, and using the symbol delay generated during the operation as the delay to be compensated.

[0037] See also Figure 3 The following is a hardware control block diagram of the correction device of the present invention. Figure 3The correction device and correction method of the present invention are explained: Figure 3 The ANALOG SIGNAL GEN in the figure is a reference signal generation module 1, which includes a PLL phase-locked loop controlled by a register to generate a sinusoidal reference signal (reference single tone signal) of a specific frequency as a reference analog signal (corresponding to Figure 3 The sine_wave register is used to control the frequency of the reference analog signal. It is written to each frequency point one by one according to the service bandwidth range. If the filter bandwidth is 20M, the scanning frequency points are 1Mhz, 2Mhz, ..., 19Mhz, 20Mhz, a total of 20 frequency points.

[0038] Figure 3 Wq_mode_se1 in is used to output a mode selection signal to the mode selection module 3, which controls the mode selection module 3 to select the offline calculation module or the online compensation mode. When it is 0, it enters the offline calculation mode, and when it is 1, it enters the online compensation mode from the offline calculation mode.

[0039] When the correction device enters the online compensation mode from the offline calculation mode, Figure 3 The ANALOG SIGNAL GEN in the DAC is connected to the FRONT END ATTENUATOR. At this time, the ANALOG SIGNAL GEN and the FRONT END ATTENUATOR together constitute a broadband signal generation module 2 for generating a broadband analog signal.

[0040] Figure 3 The LPF-ADC in the figure is the signal processing module 4, which will simultaneously output the I-channel signal and the Q-channel signal. Specifically, it converts the reference analog signal into the reference digital signal datapath_I of the I-channel and the reference digital signal datapath_Q of the Q-channel, and converts the broadband analog signal into the broadband digital signal RX_IDATA_SSI (I-channel broadband signal) of the I-channel and the broadband digital signal RX_QDATA_SSI (Q-channel broadband signal) of the Q-channel.

[0041] Figure 3 The WB QEC CALC in the compensation module is 5, and the specific process of obtaining the compensation coefficient is as follows: The reference single-tone signal (reference signal corresponding to a single frequency point) is stored in a register after analog-to-digital conversion by the signal processing module 4 and added to the reference single-tone signal (sinusoidal reference signal) at the next frequency point after conversion. The result of the addition is updated to the original storage register until the reference single-tone signals (sinusoidal reference signals) at all frequency points are received and summed by the signal processor for analog-to-digital conversion. The above process is performed separately for the I / Q channels, and finally the storage register obtains two groups of numbers, namely the total signal Sum_dataI of the I channel and the total signal Sum_dataQ of the Q channel.

[0042] After obtaining the total signal Sum_dataI of the I channel and the total signal Sum_dataQ of the Q channel, the autocorrelation vector r and the cross-correlation vector p are constructed through autocorrelation and cross-correlation. The lengths of the two vectors are controlled by the wq_comp_order register in the figure. After the autocorrelation vector r is calculated, the autocorrelation vector r is used to construct the Toeplitz matrix R, and then the Gaussian elimination method is used to obtain the compensation filter coefficients. , the compensation coefficient order n obtained will be the same as the wq_comp_order register.

[0043] When WB QEC CALC (compensation module 5) calculates the compensation coefficient After that, the wq_calc_done signal is set to 1, and the offline calculation module can enter the online compensation mode.

[0044] Figure 3 WB QEC COMP is the calculation unit 601, DELAY is the delay unit 602, and DATA PORT is the output unit 603. The specific process of compensating and correcting the broadband digital signal is as follows: The calculation unit 601 (WB QEC COMP) calculates the compensation coefficient sent by the Q-channel broadband signal RX_QDATA_SSI and the compensation module 5 (WB QECCALC) A convolution operation is performed, and a symbol delay of half the coefficient length is generated during the convolution operation. The symbol delay is determined as the delay to be compensated. The delay unit 602 performs delay compensation on the I-channel broadband signal RX_IDATA_SSI according to the delay to be compensated, and determines the I / Q two-channel signal (corresponding to the output module) output by the output module. Figure 3 The two signals RX½_IDATA_OUT± and RX½_QDATA_OUT± in the 12-bit DAC have synchronized bit stream outputs, enabling dynamic compensation of frequency-related quadrature imbalance.

[0045] In one embodiment of the present invention, the calculation unit 601 includes a compensation filter for performing a convolution operation on the compensation coefficient obtained by the compensation module 5 and the received signal (Q-channel broadband signal) to achieve real-time correction. The present invention uses a matrix approximation mathematical method to model and compensate for orthogonal imbalance. The specific process is as follows: For broadband zero-IF / low-IF receivers, in addition to the errors introduced by the local oscillator signal, the system responses of components such as the I / Q filters and analog-to-digital converters vary within the receiver bandwidth. This difference causes the amplitude and phase errors of the I / Q channels to vary with frequency, resulting in frequency-dependent orthogonal imbalance.

[0046] The frequency response of the I / Q paths can be expressed as and Ideally, the I / Q frequency responses are identical, namely:

[0047] However, in the actual signal transmission process, there are differences in the frequency responses of the I / Q channels, that is,

[0048] in, represents the frequency-dependent amplitude error, Represents the frequency-dependent phase error.

[0049] From above, Figure 3 WB QEC CALC (compensation module 5) is used to find the compensation coefficient , so that the output signal of the Q channel after filtering approaches the output signal of the I channel, making the following equation true:

[0050] in, is the frequency response of the compensation filter.

[0051] Assume that the compensation filter is a finite-length FIR filter with an order of , then establish the matrix form, compensation coefficient It can be expressed as a vector as follows:

[0052] Signal The output generated by this compensation filter It can be expressed in the following matrix form:

[0053] in is the input signal The vector form contains And its historical values:

[0054] At this time, the minimum mean square error lms can be set as the cost function ,set up is the target signal and constructs the gradient equation, which is expressed as follows:

[0055] Gradient derivative, its expression is as follows:

[0056] By minimizing the criterion and setting the gradient equation to 0, the Wiener-Hopf equation can be constructed. The matrix form is:

[0057] in,

[0058]

[0059] For input signal The autocorrelation function The matrix is ​​specifically a Toeplitz matrix constructed by performing autocorrelation calculation on the I-channel total signal and the Q-channel total signal to obtain the autocorrelation vector r. Its expression is as follows:

[0060] For input signal With target signal The cross-correlation function of dimensional vector, specifically, the cross-correlation vector p is obtained by cross-correlation calculation of the I-channel total signal and the Q-channel total signal, and its expression is as follows:

[0061] Then use the matrix inversion , and use Gaussian elimination method to solve the Wiener-Hopf linear equations. Here is a brief description of the Gaussian elimination steps: Step 1: The matrix and vector Merge into augmented matrix [ ]; Step 2: In the loop, gradually eliminate the lower elements of each column to make the matrix become upper triangular; Step 3: Back-substitute, starting from the last row and back-substituting upwards, to calculate the value of each unknown number.

[0062] In one embodiment of the present invention, the calculation unit 601 includes a compensation filter for performing a convolution operation on the compensation coefficient obtained by the compensation module 5 and the received signal (Q-channel broadband signal) to achieve real-time correction of the broadband analog signal.

[0063] See also Figure 4 As shown in the figure, it is a schematic diagram of coefficients corresponding to compensation filters of different orders of the present invention, showing the coefficient values ​​of a 25-order compensation filter. These coefficients are calculated by the above-mentioned matrix approximation algorithm and correspond to the parameters of each order of the filter.

[0064] See also Figure 5-6 The results are the performance test results of the present invention, where Figure 5 The schematic diagram of the signal before and after compensation of 16 scanning frequency points is shown. Figure 6 A schematic diagram of compensation before and after 32 scanning frequency points is shown. It can be seen that for broadband service signals (broadband analog signals), the present invention can achieve image suppression of >60dC when M=25 order. The image suppression ratio is negatively correlated with orthogonal imbalance. The smaller the image suppression ratio, the more severe the imbalance. The image suppression at the signal receiving end improves with the increase of the reference single tone signal, achieving higher frequency compensation accuracy. The actual correction effect depends on the accuracy of the compensation coefficient fixed point and the frequency resolution of the reference single tone signal.

[0065] Next, the compensation effect achievable by the present invention is described by applying the correction device of the present invention to a 4G LTE 20M receiver and a satellite communication receiver. 1) 4G LTE 20M receiver, operating frequency is 30.72MHz, signal bandwidth is 20MHz: ①System configuration: Compensation filter order of compensation module 5: 16th order Calibration signal: 16 sinusoidal reference signals evenly distributed within a 20MHz bandwidth (reference single tone signal) Sampling rate: 184.32MS / s ②The calibration process of executing the calibration method for compensation: After power-on, it automatically enters the calibration state and enters the offline calculation mode; Send 16 sinusoidal reference signals (reference single-tone signals) in sequence; Measure the I / Q response difference at each frequency point; Calculate and store compensation coefficients; ③Compensation effect: The measured image suppression reaches 61.2dBc; EVM improvement dropped from 8.3% to 1.5%; Correction time <5ms.

[0066] 2) Satellite communication receiver, operating in Ku band, with a signal bandwidth of 100MHz: ①System configuration: Compensation filter order of compensation module 5: 16th order Calibration signal: 8 sinusoidal reference signals (reference single tone signals) evenly distributed within a 100MHz bandwidth ②Compensation effect: Image suppression reaches 58.7dBc; Performance fluctuation is <1dB when the temperature changes; Power consumption increased by <5%.

[0067] The present invention achieves complete coverage of the receiver frequency band through multi-frequency point correction, without the need to adopt traditional orthogonal correction methods, and without the need to introduce special training sequences to evaluate parameterization and analyze frame structures, resulting in limited correction scenarios. At the same time, through the matrix approximation method, there is no need to adopt traditional iterative calculation methods, which reduces the resource consumption of hardware design, greatly reduces the requirements for timing, and can achieve faster correction speed. In addition, through the two working modes of offline calculation mode and online compensation mode, extremely high robustness is achieved, and a high image rejection ratio can still be maintained in complex environments, thereby optimizing orthogonal imbalance.

[0068] The above are only preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A broadband quadrature imbalance correction device, characterized in that: include: A reference signal generation module, a broadband signal generation module, a mode selection module, a signal processing module, a compensation module, and a correction module; the mode selection module is connected to the reference signal generation module, the broadband signal generation module, and the signal processing module; the signal processing module is connected to the compensation module and the correction module; and the compensation module is connected to the correction module; The mode selection module is used to select an offline calculation mode or an online compensation mode. When entering the offline calculation mode, the reference analog signal generated by the reference signal generation module is converted into a reference digital signal by the signal processing module, and the compensation module obtains a compensation coefficient based on the reference digital signal; When entering the online compensation mode from the offline calculation mode, the broadband analog signal generated by the broadband signal generation module is converted into a broadband digital signal by the signal processing module, and the correction module corrects the broadband digital signal according to the broadband digital signal and the compensation coefficient to obtain a broadband correction signal.

2. The broadband quadrature imbalance correction device according to claim 1, characterized in that: The signal processing module includes an I-channel processing unit and a Q-channel processing unit with two identical structures; The I-path processing unit and the Q-path processing unit both include a filter and an analog-to-digital converter; The filter is connected to the analog-to-digital converter and the mode selection module, and the analog-to-digital converter is connected to the compensation module and the correction module.

3. The broadband quadrature imbalance correction device according to claim 2, characterized in that: The correction module includes a calculation unit, a delay unit and an output unit; The calculation unit is connected to the output unit, the Q-path processing unit, and the compensation module; the delay unit is connected to the output module and the I-path processing unit.

4. A broadband quadrature imbalance correction method, applied to a broadband quadrature imbalance correction device according to any one of claims 1 to 3, characterized in that: include: When the mode selection module selects the offline calculation mode, the reference analog signal generated by the reference signal generation module is converted into a reference digital signal by the signal processing module; The compensation module obtains a compensation coefficient according to the reference digital signal; When the mode selection module switches from the offline calculation mode to the online compensation mode, the broadband analog signal generated by the broadband signal generation module is converted into a broadband digital signal by the signal processing module; The correction module corrects the broadband digital signal according to the broadband digital signal and the compensation coefficient to obtain a broadband correction signal.

5. The method for correcting broadband quadrature imbalance according to claim 4, wherein: The reference analog signal includes a plurality of sinusoidal reference signals corresponding to target frequency points; The reference signal generating module generates the reference analog signal, including: Obtaining a frequency range of a filter in the signal processing module; Determining a plurality of target frequency points having a generation order according to the frequency range; According to the generation order, sinusoidal reference signals corresponding to a plurality of target frequency points are generated.

6. The method for correcting broadband quadrature imbalance according to claim 5, wherein: The reference digital signal includes an I-path reference signal and a Q-path reference signal corresponding to a plurality of target frequency points; The compensation module obtains a compensation coefficient according to the reference digital signal, including: Summing the I-channel reference signals corresponding to the multiple target frequency points to obtain an I-channel total signal, and summing the Q-channel reference signals corresponding to the multiple target frequency points to obtain a Q-channel total signal; The compensation coefficient is obtained according to the I-channel total signal and the Q-channel total signal.

7. The method for correcting broadband quadrature imbalance according to claim 6, wherein: Obtaining the compensation coefficient according to the I-channel total signal and the Q-channel total signal includes: Performing autocorrelation calculation on the Q-path total signal to obtain an autocorrelation vector; Performing cross-correlation calculation on the I-channel total signal and the Q-channel total signal to obtain a cross-correlation vector; The compensation coefficient is obtained according to the autocorrelation vector and the cross-correlation vector.

8. The method for correcting broadband quadrature imbalance according to claim 7, wherein: Obtaining the compensation coefficient according to the autocorrelation vector and the cross-correlation vector includes: According to the autocorrelation vector, a Toeplitz matrix is ​​obtained; The compensation coefficient is obtained according to the Toeplitz matrix and the cross-correlation vector.

9. The method for correcting broadband quadrature imbalance according to claim 4, wherein: The broadband digital signal includes an I-channel broadband signal and a Q-channel broadband signal; The correction module corrects the broadband digital signal according to the broadband digital signal and the compensation coefficient to obtain a broadband correction signal, including: Determining a delay to be compensated according to the Q-channel broadband signal and the compensation coefficient; Performing delay processing on the I-channel broadband signal according to the delay to be compensated, to obtain an I-channel correction signal having a synchronous code stream output with the Q-channel broadband signal; The I-channel correction signal and the Q-channel broadband signal constitute the broadband correction signal.

10. The method for correcting broadband quadrature imbalance according to claim 9, wherein: Determining a delay to be compensated according to the Q-channel broadband signal and the compensation coefficient includes: A convolution operation is performed on the Q-channel broadband signal and the compensation coefficient, and a symbol delay generated during the operation is used as the delay to be compensated.

Images