Broadband compensation method and device for IQ phase imbalance and computer equipment

Abstract

The invention relates to an IQ phase imbalance broadband compensation method and device and computer equipment. The method comprises the following steps: determining the phase response of each channel; phase equalization all-pass filters of the corresponding channels are obtained based on the phase responses; carrying out cascade connection on the all-pass filter and the corresponding channel; and compensating each frequency point of the channel after each cascade all-pass filter. According to the method, the channels are equalized, and then the frequency points of the cascaded channels are compensated, so that the purpose of broadband calibration compensation between IQ is achieved, and IQ phase imbalance broadband compensation can be accurately carried out.

Description

Technical Field

[0001] The present application relates to the field of chip technology, and particularly to an IQ phase imbalance broadband compensation method, apparatus, and computer device. Background Art

[0002] The digital tester comes with an Arbitrary Waveform Generator / Source (hereinafter referred to as AWG) and a Digitizer / Capture (hereinafter referred to as DGT), which can be used for various types of chip tests. With the rapid development of the electronic information field, the structure of electronic information systems has become increasingly complex, especially with higher bandwidth requirements. This means that higher sampling rates and bandwidths of AWG and DGT can be applicable to more complex chip test scenarios. For AC circuit test scenarios, both AWG and DGT include IQ functions, and two channels implement a set of IQ signals. Keeping IQ orthogonal is a prerequisite for correct signal demodulation. Therefore, the phase-frequency characteristics between hardware channels should be consistent. However, due to the imperfections of devices, even if the components between IQ channels have exactly the same layout and routing, their respective phase-frequency characteristics will still have differences. This difference will affect the chip test results. Therefore, it is necessary to calibrate the different phase differences between different frequency points between IQ channels.

[0003] The existing IQ phase imbalance compensation technology is based on the calibration of single-tone signals. After establishing a mathematical model of IQ offset for single-tone signals and estimating the IQ amplitude and phase imbalance factors, a correction matrix is calculated to compensate for IQ.

[0004] However, the current calibration based on single-tone signals can only estimate the phase imbalance factors at single frequency points for compensation. Due to the imperfections of the hardware of the analog-digital hybrid board card channels, the phase imbalance factors corresponding to the same frequency points between IQ channels are not equal within the entire bandwidth. For AC tests, if the imbalance factors at each frequency point within the frequency band vary too much, even in the case of taking their average value, the imbalance will deteriorate after compensation. Summary of the Invention

[0005] Based on this, in view of the above technical problems, it is necessary to provide an IQ phase imbalance broadband compensation method, apparatus, and computer device that can accurately perform IQ phase imbalance broadband compensation.

[0006] In a first aspect, the present application provides an IQ phase imbalance broadband compensation method, the method including: Determine the phase response of each channel; Respectively obtain the phase equalization all-pass filters corresponding to the channels based on the phase response; Cascade the all-pass filter with the corresponding channel; Compensate each frequency point of the channels after each cascaded all-pass filter.

[0007] In one embodiment, the determining the phase response of each channel includes: Collect the discrete time-domain data of each channel, and respectively convert each of the discrete time-domain data into frequency-domain data; Respectively obtain each initial phase response based on each of the frequency-domain data, and perform phase unwrapping on each of the initial phase responses to obtain the phase response of each channel.

[0008] In one embodiment, the respectively obtaining the phase equalization all-pass filter corresponding to the channel based on the phase response includes: Perform interpolation on the phase response to obtain an interpolated phase response; Determine the group delay corresponding to each frequency point of the channel based on the interpolated phase response; Obtain the phase equalization all-pass filter of the channel based on the group delay corresponding to each frequency point of the channel.

[0009] In one embodiment, the obtaining the phase equalization all-pass filter of the channel based on the group delay corresponding to each frequency point of the channel includes: Based on the group delay corresponding to each frequency point of the channel, obtain the expected group delay under the sampling point unit; Obtain a complex cepstrum sequence based on the expected group delay under the sampling point unit; Obtain the unwrapped phase and the complex cepstrum even sequence based on the complex cepstrum sequence; Obtain the magnitude of the Fourier transform of the minimum-phase signal based on the complex cepstrum even sequence; Determine the phase equalization all-pass filter of the channel based on the magnitude of the Fourier transform of the minimum-phase signal and the unwrapped phase.

[0010] In one embodiment, the compensating each frequency point of the channels after each cascaded all-pass filter includes: Determine the first group delay corresponding to each frequency point of the I channel and the second group delay corresponding to each frequency point of the Q channel after each cascaded all-pass filter, and based on the first group delay and the second group delay, obtain the group delay difference between the channels; Based on the group delay difference, perform a time-domain data point coordinate drift operation on all data points of the lagging channel to compensate each frequency point of the lagging channel.

[0011] In one embodiment, the based on the group delay difference, performing a time-domain data point coordinate drift operation on all data points of the lagging channel to compensate each frequency point of the lagging channel includes: Obtain an average delay difference based on the group delay difference; Shift the time-domain coordinates of all data points in the lag channel to the left by the average delay difference number of sampling points to compensate for each frequency point in the lag channel.

[0012] In one embodiment, the shifting the time-domain coordinates of all data points in the lag channel to the left by the average delay difference number of sampling points to compensate for each frequency point in the lag channel includes: Interpolate the time-domain data point coordinates of the lag channel; Shift the coordinates of each interpolated data point to the left by the average delay difference number of sampling points to compensate for each frequency point in the lag channel.

[0013] In one embodiment, the determining the first group delay corresponding to each frequency point of the I channel after each cascaded all-pass filter and the second group delay corresponding to each frequency point of the Q channel includes: Determine the first target phase response of the I channel and the second target phase response of the Q channel after each cascaded all-pass filter; Interpolate the first target phase response to obtain the first target interpolated phase response of the I channel, and interpolate the second target phase response to obtain the second target interpolated phase response of the Q channel; Based on the first target interpolated phase response, determine the first group delay corresponding to each frequency point of the I channel after each cascaded all-pass filter, and based on the second target interpolated phase response, determine the second group delay corresponding to each frequency point of the Q channel after each cascaded all-pass filter.

[0014] In a second aspect, the present application further provides an IQ phase imbalance broadband compensation device, the device includes: A phase response determination module, configured to determine the phase response of each channel; An all-pass filter determination module, configured to respectively obtain the phase equalization all-pass filter of the corresponding channel based on the phase response; A cascading module, configured to cascade the all-pass filter with the corresponding channel; A compensation module, configured to compensate for each frequency point of the channel after each cascaded all-pass filter.

[0015] In a third aspect, the present application further provides a computer device, including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the above method are implemented.

[0016] The above IQ phase imbalance broadband compensation method, device and computer device first determine the phase responses of each channel, respectively obtain the phase equalization all-pass filters corresponding to the channels based on the phase responses, cascade the all-pass filters with the corresponding channels, and compensate each frequency point of the channels after each cascaded all-pass filter. In this way, each channel is equalized first, and then the frequency points of the cascaded channels are compensated to achieve the purpose of broadband calibration compensation between IQ, and accurate IQ phase imbalance broadband compensation can be performed. Brief Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a schematic flowchart of the IQ phase imbalance broadband compensation method in one embodiment; Figure 2 It is a principle framework diagram of the IQ phase imbalance broadband compensation method in the present application; Figure 3 It is a flowchart of the determination steps of the all-pass filter in one embodiment; Figure 4 It is a comparison effect diagram of calculating the group delay by using the cubic polynomial interpolation of the channel phase response and then using the difference method and directly using the difference method to calculate the group delay in one embodiment; Figure 5 It is a schematic diagram of the comparison between the single-frequency point and broadband calibration of the phase imbalance calibration compensation in one embodiment; Figure 6 It is the broadband calibration compensation effect before and after interpolation in one embodiment; Figure 7 It is a flowchart of the IQ channel phase imbalance bandwidth compensation method in one embodiment; Figure 8 It is a structural block diagram of the IQ phase imbalance broadband compensation device in one embodiment; Figure 9 It is an internal structure diagram of a computer device in one embodiment. Detailed Description of the Embodiments

[0019] In order to make the purpose, technical solutions and advantages of the present application clearer, the following will further describe the present application in detail in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application and are not used to limit the present application.

[0020] In one embodiment, as Figure 1 shown, an IQ phase imbalance broadband compensation method is provided. In this embodiment, this method is exemplified by being applied to an analog-digital hybrid board. It can be understood that this method can also be applied to a host computer, an industrial control computer, or a communication intelligent device, and can also be applied to include calibration hardware. In this embodiment, the method includes the following steps: S102: Determine the phase response of each channel.

[0021] The phase response of the channel is obtained by sampling the channel to obtain discrete time-domain data, and then converting the discrete time-domain data into frequency-domain data and based on the frequency-domain data.

[0022] The channels in this application include the I channel and the Q channel in the analog-digital hybrid board. Each I channel and each Q channel in the analog-digital hybrid board form a group. An analog-digital hybrid board may include at least one group of the above channels. When compensating for the IQ phase imbalance in each group of the analog-digital hybrid board, each group of the above channels can be compensated serially or in parallel. No specific limitation is made here. The compensation method for each group of the above channels can adopt the method in this application. In addition, the calculation of the phase equalization all-pass filter for the I channel and the Q channel in each group can be parallel calculation or serial calculation, which will not be elaborated here. In practical applications, since the analog-digital hybrid board includes multiple groups of the above channels, when compensating, each group of the above channels is compensated in parallel, and during the compensation calculation process of each group of the above channels, the calculation of the phase equalization all-pass filter for the I channel and the Q channel adopts parallel calculation. Subsequently, the I channel and the Q channel are respectively cascaded with the corresponding phase equalization all-pass filter, and the group delay of the I channel and the Q channel is calculated in parallel, and then the group delay difference of each group is calculated. Finally, based on the group delay difference, each frequency point of the lagging channel is compensated.

[0023] S104: Respectively obtain the phase equalization all-pass filter corresponding to the channel based on the phase response.

[0024] Among them, the group delay corresponding to the channel can be obtained based on the phase response of the channel, the group delay corresponding to the all-pass filter can be obtained based on the group delay of the channel, and then the phase response corresponding to the all-pass filter can be obtained based on the group delay corresponding to the all-pass filter. Furthermore, the phase equalization all-pass filter corresponding to the channel can be obtained based on the phase response corresponding to the all-pass filter.

[0025] The group delay is a quantity used to describe how fast the phase change is with respect to the frequency change.

[0026] In some alternative embodiments, the group delay of a corresponding channel can be obtained based on the phase response of the channel. The group delay can be approximated by using the difference method to calculate the derivative of the phase response, and the negative of the derivative is the group delay. However, since the frequency point interval of the current sampling is large, directly using the finite difference method to calculate the group delay results in too low precision and inaccurate calculation. Therefore, the phase response of the channel is first interpolated, for example, the interpolated phase response is obtained through cubic polynomial interpolation, and then the group delay of the channel is obtained based on the interpolated phase response.

[0027] Based on the group delay of the channel, the group delay corresponding to the all-pass filter can be obtained, that is, the group delay in the all-pass form on 0~2π is obtained, which can reduce the influence of the mutation of the group delay on the mutation point.

[0028] In this application, the design of the digital all-pass filter is based on the discrete Hilbert transform relationship. Its basic principle is to determine the phase of the pole part of the desired all-pass filter according to the given group delay; then the amplitude spectrum of the pole part is obtained from the phase through the discrete Hilbert transform relationship. In other embodiments, the design of the all-pass filter can also adopt other methods, which are not specifically limited here. In this embodiment, the complex cepstrum sequence can be obtained first based on the group delay corresponding to the all-pass filter, and then the phase response corresponding to the all-pass filter is obtained based on the complex cepstrum sequence.

[0029] Finally, the denominator polynomial coefficients and numerator coefficients corresponding to the all-pass filter are obtained by solving based on the phase response corresponding to the all-pass filter, so that the expression of the all-pass filter can be obtained, that is, the all-pass filter is obtained.

[0030] S106: Cascade the all-pass filter with the corresponding channel.

[0031] Among them, in combination with Figure 2 shown, Figure 2 is the principle framework diagram of the IQ phase imbalance broadband compensation method in this application. Figure 2 Only the principle framework of one group of channels is shown therein. Those skilled in the art can know that the principle frameworks of other groups of channels are the same as that in Figure 2 and will not be elaborated here. In this embodiment, each channel is cascaded with at least one all-pass filter to first equalize each channel, and then the frequency points of the cascaded channels are compensated for the time delay difference of each channel through data position shifting.

[0032] Among them, for the sake of easy understanding, the principle corresponding to the principle framework in Figure 2 is described: Let the transfer function of the all-pass filter be , and the transfer function of the channel system be , then the cascaded system is:

[0033] The frequency response of the cascaded system is as follows:

[0034] Among them, the parameter carrying "ap" is used to represent the parameter corresponding to the all-pass filter, and the parameter carrying "d" is used to represent the parameter of the channel.

[0035] The phase relationship of the cascaded system is as follows:

[0036] From the group delay it can be obtained that the group delay of the cascaded system is:

[0037] Among them is the group delay of the all-pass system, is the group delay of the channel system. Therefore, the all-pass filter can be designed according to the group delay of the channel, so that the passband of the cascaded system satisfies:

[0038] is a constant.

[0039] What this application involves is the IQ channel of the analog-digital hybrid board. Therefore, this application designs the I channel and the Q channel. Among them, the group delay after the I channel is cascaded with the all-pass filter APF1 is , and the group delay after the Q channel is cascaded with the all-pass filter APF2. At this time, the group delay is independent of frequency, and are certain constants, and the phase difference between the IQ channels is

[0040] According to the time-shift property of the Fourier transform, the delay in the time domain is manifested as a phase shift in the frequency domain, that is, for a time-domain signal with a delay time , this signal generates phase shift in the frequency domain. Therefore, after passing through the all-pass filter, only the time-domain signal of the leading channel needs to be delayed , so that the phases of the two channels can be aligned, that is, channel equalization is achieved.

[0041] Based on this, the principle of the determination method of the all-pass filter in step S104 can include: For an M-order all-pass filter, the transfer function is:

[0042] Among them . Then the frequency response of the M-order all-pass filter is:

[0043] From the above equation, it can be obtained that the group delay response of is related to the group delay of its denominator polynomial as follows:

[0044] The same relationship also holds for the average value of the group delay:

[0045] and

[0046] The discrete Hilbert transform relationship correlates the log magnitude and phase of the Fourier transform of a minimum-phase signal. That is:

[0047] where and correspond to the Fourier transform and the unwrapped phase respectively, is the cepstral coefficient of . To obtain a stable system, the denominator polynomial coefficients must be minimum-phase signals. Therefore, the steps of the all-pass filter design algorithm based on the discrete Hilbert transform relationship are first to obtain the coefficients of the denominator polynomial using equations and, and then from equation it can be seen that the numerator coefficients are the flipped sequence of the denominator coefficients, and finally the coefficients of the desired all-pass filter are obtained.

[0048] S108: Compensate each frequency point of the channels after cascading all-pass filters.

[0049] The determined all-pass filters are cascaded into the corresponding channels to align the phases of each channel, that is, to achieve channel equalization. However, there are still differences in the group delay between IQ, and these differences result in equal delay differences at each frequency point between IQ. The delay differences should be compensated for. Therefore, each frequency point of the channels after cascading all-pass filters is determined and then compensated.

[0050] The above IQ phase imbalance broadband compensation method first determines the phase response of each channel; respectively obtains the phase equalization all-pass filters corresponding to the channels based on the phase response; cascades the all-pass filters with the corresponding channels; compensates each frequency point of the channels after cascading all-pass filters. In this way, each channel is equalized first, and then the frequency points of the cascaded channels are compensated to achieve the purpose of broadband calibration compensation between IQ, and can accurately perform IQ phase imbalance broadband compensation.

[0051] In one optional embodiment, determining the phase response of each channel includes: collecting the discrete time-domain data of each channel and respectively converting the discrete time-domain data into frequency-domain data; respectively obtaining the initial phase response of each channel based on the frequency-domain data of each channel, and performing phase unwrapping on each initial phase response to obtain the phase response of each channel.

[0052] Among them, a multi-tone signal is sent to the channel and the time-domain data is collected . The multi-tone frequency vector is set according to the allowed bandwidth and sampling rate of the channel , and is a frequency vector with uniform intervals, that is , the frequency point interval can divide , that is, the number of points of the subsequent FFT operation. In this embodiment, .

[0053] The determination method of the phase response of each channel includes: first, ADC sampling to obtain discrete time-domain data , performing FFT (Fourier transform) operation on the discrete time-domain data to obtain frequency-domain data ; then using the four-quadrant arctangent function to obtain the phase response, that is ; in order to eliminate the jumps in the phase signal, finally, the unwrap algorithm is used to perform phase unwrapping on it to obtain . It should be noted that other methods can also be used for the method of obtaining frequency-domain data based on discrete time-domain data, and other methods can also be used for the phase unwrapping method, which is not specifically limited here.

[0054] In one optional embodiment, respectively obtaining the phase equalization all-pass filter corresponding to the channel based on the phase response includes: performing interpolation on the phase response to obtain the interpolated phase response; determining the group delay corresponding to each frequency point of the channel based on the interpolated phase response; obtaining the phase equalization all-pass filter of the channel based on the group delay corresponding to each frequency point of the channel.

[0055] Among them, as shown in Figure 3 , Figure 3 is a flowchart of the determination steps of the all-pass filter in an embodiment. In this embodiment, first calculate the group delay corresponding to the channel .

[0056] Among them, in this application, the difference method is used to approximately calculate the derivative of the phase response , and its negative value is the group delay. However, because the frequency point interval of the current sampling is very large, directly using the finite difference method to calculate the group delay has too low accuracy and leads to inaccurate calculation. First, perform interpolation on the phase response Interpolation is performed, for example, cubic polynomial interpolation is used to obtain , that is, the phase-frequency curve is fitted by the interpolation method, where the interpolation step size is . Generally, the smaller the step size, the higher the approximation accuracy. However, if the step size is too small, the calculation result may be inaccurate due to numerical rounding errors. Therefore, it is necessary to balance the calculation accuracy and stability to select an appropriate step size . After interpolation, the derivative of the phase response is calculated using the difference method :

[0057] The group delay is the negative of the derivative, that is, the group delay of the channel :

[0058] Among them, in this application, the group delay is calculated by first interpolating and then using the difference method, which can improve the smoothness of the group delay curve, ensure the continuity of the first-order derivative of the interpolated phase response at the original frequency points, so as to obtain the instantaneous slope to approximate the group delay response of the IQ channel, and the calculation accuracy is high. Therefore, the imbalance between IQs is compensated after broadband calibration.

[0059] Specifically, as shown in Figures 4 to 6 , Figure 4 is the comparison effect of calculating the group delay by cubic polynomial interpolation of the channel phase response and then using the difference method and directly using the difference method in an embodiment. Figure 5 is a schematic diagram of the comparison between single-frequency point and broadband calibration of phase imbalance calibration compensation in an embodiment. Figure 6 is the broadband calibration compensation effect before and after interpolation in an embodiment.

[0060] Overall, the algorithm of this application has the following advantages: If the difference calculation is directly performed without interpolation, due to the too large frequency point interval, the calculation accuracy is very low, and it deteriorates after calibration. The curve of calculating the group delay after phase interpolation is smoother and the calculation accuracy is high. Therefore, the imbalance between IQs is compensated after broadband calibration; Secondly, for the scenario where the phase imbalance of different frequency points between IQ channels is different, the traditional single-frequency calibration method for IQ imbalance cannot play a role, while using the broadband filtering calibration method, the phase in a single channel is equalized using an all-pass filter and then the time delay difference between channels is compensated, so that different phase imbalances of IQ can be compensated.

[0061] In one of the optional embodiments, a phase equalization all-pass filter for a channel is obtained based on the group delay corresponding to each frequency point of the channel, including: obtaining an expected group delay in sampling point units based on the group delay corresponding to each frequency point of the channel; obtaining a complex cepstrum sequence based on the expected group delay in sampling point units; obtaining an unwrapped phase and a complex cepstrum even sequence based on the complex cepstrum sequence; obtaining the magnitude of the Fourier transform of the minimum-phase signal based on the complex cepstrum even sequence; and determining the phase equalization all-pass filter for the channel based on the magnitude of the Fourier transform of the minimum-phase signal and the unwrapped phase.

[0062] Among them, the method for determining the phase equalization all-pass filter for each channel in this application is the same. Only the general principle is described in this application. For the method of determining the phase equalization all-pass filter for each channel, the corresponding parameters can be replaced with channel parameters, which will not be elaborated here.

[0063] Obtaining an expected group delay in sampling point units based on the group delay corresponding to each frequency point of the channel includes: calculating the expected group delay of the all-pass filter to be designed , and this expected group delay of the all-pass filter is used to deduce the characterization function of the all-pass filter. To reduce the influence of the sudden change of the group delay on the mutation point, let the group delay of the all-pass filter be in the all-pass form on. Then (symmetric about ), the group delay on is:

[0064] The expected group delay of the all-pass filter to be designed , where .

[0065] Subsequently, will be converted from time units to sampling point units to obtain , which is the expected group delay in sampling point units.

[0066] Among them, obtaining a complex cepstrum sequence based on the expected group delay in sampling point units includes: performing an N-point IFFT on to obtain a sequence , and then calculating the complex cepstrum sequence using the following formula:

[0067] Obtaining an unwrapped phase and a complex cepstrum even sequence based on the complex cepstrum sequence includes: Calculating the unwrapped phase by performing a Fourier transform on the complex cepstrum sequence and then finding its imaginary part:

[0068] Among them, Imag represents taking the imaginary part, and FFT represents the Fourier transform.

[0069] Calculate the complex cepstrum even sequence . Among them

[0070] Perform an IFFT operation on to obtain the magnitude of the Fourier transform of the minimum-phase signal , that is the real part of, where |.| represents taking the modulus.

[0071] Based on the magnitude of the Fourier transform of the minimum-phase signal and the unwrapped phase, determine the phase equalization all-pass filter for the channel, including: Calculate the denominator polynomial coefficients of the all-pass filter . It should be noted that as increases the magnitude of decreases, and when , , this is because is the minimum-phase sequence.

[0072] According to the order of the required all-pass filter, obtain the denominator coefficients of the all-pass filter:

[0073] Based on the denominator coefficients of the all-pass filter, obtain the numerator coefficients of the transfer function of the all-pass filter, .

[0074] In this way, based on the denominator coefficients and numerator coefficients of the all-pass filter, the all-pass filter for the corresponding channel can be obtained. Among them, for the phase equalization all-pass filter of the I channel, the coefficients of the numerator and denominator of its transfer function are respectively ; for the phase equalization all-pass filter of the Q channel, the coefficients of the numerator and denominator of its transfer function are respectively .

[0075] In the above embodiment, after obtaining the phase of the sampled IQ multi-tone signal, first interpolate the phase response, and then use the difference method to calculate the group delay. This method ensures that the first derivative of the interpolated phase response is continuous at the original frequency points, so as to obtain the instantaneous slope to approximate the group delay response of the IQ channels.

[0076] In one alternative embodiment, compensating each frequency point of the channels after each cascaded all-pass filter includes: determining a first group delay corresponding to each frequency point of the I channel and a second group delay corresponding to each frequency point of the Q channel after each cascaded all-pass filter, and obtaining a group delay difference between the channels based on the first group delay and the second group delay; based on the group delay difference, performing a time-domain data point coordinate drift operation on all data points of the lagging channel to compensate each frequency point of the lagging channel.

[0077] The first group delay and the second group delay are determined based on the channels after cascading the corresponding all-pass filters. In this embodiment, the group delay corresponding to each channel can be determined separately. It should be noted that the method for determining the first group delay and the second group delay in this embodiment can be the difference method. And to ensure accuracy, first interpolate the phase response of the channels after cascading the corresponding all-pass filters, and then calculate the corresponding group delay based on the interpolated phase response.

[0078] In one alternative embodiment, determining a first group delay corresponding to each frequency point of the I channel and a second group delay corresponding to each frequency point of the Q channel after each cascaded all-pass filter includes: determining a first target phase response of the I channel and a second target phase response of the Q channel after each cascaded all-pass filter; interpolating the first target phase response to obtain a first target interpolated phase response of the I channel, and interpolating the second target phase response to obtain a second target interpolated phase response of the Q channel; determining a first group delay corresponding to each frequency point of the I channel after each cascaded all-pass filter based on the first target interpolated phase response, and determining a second group delay corresponding to each frequency point of the Q channel after each cascaded all-pass filter based on the second target interpolated phase response.

[0079] For convenience, the determination method of the first group delay of the I channel is taken as an example for illustration. The determination method of the second group delay of the Q channel is similar and will not be elaborated here.

[0080] Among them, after cascading the corresponding all-pass filter for the I channel, discrete time-domain data of the I channel is sampled, and then the discrete time-domain data of the I channel after cascading the corresponding all-pass filter is processed, such as performing a Fourier transform to obtain frequency-domain data. Subsequently, the phase response of the I channel after cascading the corresponding all-pass filter is obtained by using the four-quadrant arctangent function. Finally, to eliminate the jumps in the phase signal, phase unwrapping is performed on it to obtain the first target phase response of the I channel after cascading the corresponding all-pass filter.

[0081] Then interpolate the first target phase response to obtain a first target interpolated phase response of the I channel. For example, cubic polynomial interpolation is used for interpolation. The determination of the interpolation step size can refer to the above text and will not be elaborated here.

[0082] Finally, based on the first target interpolation phase response, the first group delay corresponding to each frequency point of the I channel after each cascaded all-pass filter is determined. Specifically, the first group delay corresponding to each frequency point of the I channel after the cascaded all-pass filter is determined by the difference method based on the first target interpolation phase response.

[0083] Among them, the group delay difference between the two channels , because the phase difference between the channels is compensated, so is relatively flat across the entire frequency band. Then, the method of shifting the time-domain data point coordinates is used to achieve the purpose of wideband calibration compensation between IQ, and at this time, the channel itself also basically has a linear phase.

[0084] In one optional embodiment, based on the group delay difference, a time-domain data point coordinate drift operation is performed on all data points of the lagging channel to compensate for each frequency point of the lagging channel, including: obtaining the average delay difference based on the group delay difference; shifting the time-domain coordinates of all data points of the lagging channel to the left by the average delay difference number of sampling points to compensate for each frequency point of the lagging channel.

[0085] Among them, the calculation method of the average delay difference is: , where n is the number of frequency points.

[0086] After determining the average delay difference, shift the time-domain coordinates of all data points of the lagging channel to the left by sampling points, where the lagging channel is the phase-lagging channel.

[0087] In one optional embodiment, shifting the time-domain coordinates of all data points of the lagging channel to the left by the average delay difference number of sampling points to compensate for each frequency point of the lagging channel includes: interpolating the time-domain data point coordinates of the lagging channel; shifting the coordinates of each interpolated data point to the left by the average delay difference number of sampling points to compensate for each frequency point of the lagging channel.

[0088] Among them, in some embodiments, is not an integer sampling point and cannot be directly shifted. In this application, the sinc interpolation method (other methods can be used in other embodiments) is used to shift all the time-domain data coordinates of the channel. The interpolation formula is:

[0089] The original coordinate position m of the time-domain data to be shifted is , and the value corresponding to the integer sampling point , represents the convolution kernel, t is the interpolation coordinate point, is the value calculated by interpolation, and all the time-domain data points of the channel are delayed by The coordinate position after sampling points is 。Specifically, for a certain new coordinate point

[0090] represents the length of the sinc convolution kernel, which is theoretically an infinite number of points. However, in actual calculation, the convolution kernel cannot be infinitely long and truncation will occur. It is necessary to select an appropriate according to the accuracy requirements.

[0091] In this embodiment, by shifting the data position, the group delay difference of each frequency point in the IQ channels is compensated. After the delay, the phase imbalance between the I channel and the Q channel will be compensated, and there will be no IQ mismatch influence from the channel itself on the broadband signal.

[0092] Specifically, for the convenience of understanding, in combination with Figure 7 shown in Figure 7 is a flowchart of the IQ channel phase imbalance bandwidth compensation method in an embodiment. In this embodiment, the parameters corresponding to the I channel carry i, and the parameters corresponding to the Q channel carry q to distinguish the parameters of the I channel and the Q channel.

[0093] Collect multi-tone data for the I channel and the Q channel and perform subsequent processing. Taking the I channel in Figure 7 as an example, calculate the phase corresponding to each frequency point of the I channel, calculate the group delay between the I channels using cubic polynomial interpolation and the difference method, calculate the expected group delay of the all-pass filter to be designed for the I channel , convert the group delay from the time unit to the sampling point unit, and then calculate the complex cepstrum sequence , perform Fourier transform on the complex cepstrum sequence and calculate the unwrapped phase , calculate the even sequence of the complex cepstrum , perform FFT transform on the even sequence of the complex cepstrum and take the real number to obtain the amplitude of the Fourier transform of the minimum phase signal, calculate the denominator polynomial , obtain the denominator coefficients of the all-pass filter according to the required filter order , calculate the numerator coefficients of the all-pass filter , cascade the I channel with the all-pass filter for filtering, and calculate the first group delay of the cascaded I channel. Similarly, the second group delay of the Q channel can be calculated in the above manner, and then calculate the group delay difference between the I channel and the Q channel, and calculate the average value of the group delay difference between the IQ channels, and convert it to the sampling point unit. Use the sinc interpolation algorithm to perform time-domain coordinate shift on all data points of the channel, and the broadband calibration and compensation of the phase imbalance between the IQ channels are completed.

[0094] In the above embodiments, after obtaining the phase of the sampled IQ multi-tone signal, the phase response is first interpolated by a cubic polynomial, and then the group delay is calculated using the difference method. This method ensures that the first derivative of the interpolated phase response is continuous at the original frequency points, so as to obtain the instantaneous slope to approximate the group delay response of the IQ channels. After equalizing the single-channel phase based on the all-pass filter, the method of shifting the coordinates of the time-domain data points is used to achieve the purpose of broadband calibration and compensation between IQ, and at this time, the channel itself also basically has a linear phase.

[0095] It should be understood that although the steps in the flowcharts involved in the above embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or steps in other steps.

[0096] Based on the same inventive concept, an embodiment of the present application also provides an IQ phase imbalance broadband compensation device for implementing the IQ phase imbalance broadband compensation method described above. The implementation solutions provided by this device to solve problems are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more embodiments of the IQ phase imbalance broadband compensation device provided below can refer to the limitations on the IQ phase imbalance broadband compensation method in the above text, and will not be repeated here.

[0097] In an exemplary embodiment, as Figure 8 shown, an IQ phase imbalance broadband compensation device is provided, including: a phase response determination module 801, an all-pass filter determination module 802, a cascading module 803, and a compensation module 804, where: The phase response determination module 801 is used to determine the phase response of each channel; The all-pass filter determination module 802 is used to obtain the phase equalization all-pass filter of the corresponding channel based on the phase response respectively; The cascading module 803 is used to cascade the all-pass filter with the corresponding channel; The compensation module 804 is used to compensate each frequency point of the channel after cascading each all-pass filter.

[0098] In one optional embodiment, the above-mentioned phase response determination module 801 is specifically configured to collect discrete time-domain data of each channel, and convert the discrete time-domain data of each channel into frequency-domain data respectively; obtain initial phase responses of each channel based on the frequency-domain data respectively, and perform phase unwrapping on the initial phase responses to obtain the phase responses of each channel.

[0099] In one optional embodiment, the above-mentioned all-pass filter determination module 802 is specifically configured to perform interpolation on the phase response to obtain an interpolated phase response; determine the group delay corresponding to each frequency point of the channel based on the interpolated phase response; and obtain the phase equalization all-pass filter of the channel based on the group delay corresponding to each frequency point of the channel.

[0100] In one optional embodiment, the above-mentioned all-pass filter determination module 802 is specifically configured to obtain the expected group delay in units of sampling points based on the group delay corresponding to each frequency point of the channel; obtain a cepstrum sequence based on the expected group delay in units of sampling points; obtain an unwrapped phase and a cepstrum even sequence based on the cepstrum sequence; obtain the magnitude of the Fourier transform of the minimum-phase signal based on the cepstrum even sequence; and determine the phase equalization all-pass filter of the channel based on the magnitude of the Fourier transform of the minimum-phase signal and the unwrapped phase.

[0101] In one optional embodiment, the above-mentioned compensation module 804 is specifically configured to determine a first group delay corresponding to each frequency point of the I channel after each cascaded all-pass filter and a second group delay corresponding to each frequency point of the Q channel, and obtain the group delay difference between channels based on the first group delay and the second group delay; and perform a time-domain data point coordinate drift operation on all data points of the lagging channel based on the group delay difference to compensate each frequency point of the lagging channel.

[0102] In one optional embodiment, the above-mentioned compensation module 804 is specifically configured to obtain an average delay difference based on the group delay difference; shift the time-domain coordinates of all data points of the lagging channel to the left by the average delay difference number of sampling points to compensate each frequency point of the lagging channel.

[0103] In one optional embodiment, the above-mentioned compensation module 804 is specifically configured to perform interpolation on the time-domain data point coordinates of the lagging channel; shift the coordinates of each interpolated data point to the left by the average delay difference number of sampling points to compensate each frequency point of the lagging channel.

[0104] In one of the alternative embodiments, the compensation module 804 is specifically configured to determine a first target phase response of the I channel and a second target phase response of the Q channel after each cascaded all-pass filter; perform interpolation on the first target phase response to obtain a first target interpolated phase response of the I channel, and perform interpolation on the second target phase response to obtain a second target interpolated phase response of the Q channel; determine a first group delay corresponding to each frequency point of the I channel after each cascaded all-pass filter based on the first target interpolated phase response, and determine a second group delay corresponding to each frequency point of the Q channel after each cascaded all-pass filter based on the second target interpolated phase response.

[0105] Each module in the above IQ phase imbalance broadband compensation device can be implemented in whole or in part by software, hardware, and their combination. Each of the above modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0106] In an exemplary embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 9 shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used for the processor to exchange information with external devices. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC), or other technologies. The computer program, when executed by the processor, implements an IQ phase imbalance broadband compensation method. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0107] Those skilled in the art can understand, Figure 9The structure shown is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0108] In one embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0109] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0110] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0111] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in this application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., without limitation.

[0112] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this application.

[0113] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A broadband IQ phase imbalance compensation method, characterized in that: The method comprises: Determine the phase response of each channel; Obtaining phase-balanced all-pass filters of corresponding channels based on the phase responses respectively; Cascading the all-pass filter with the corresponding channel; Compensate for each frequency point of the channel after each level of cascaded all-pass filters.

2. The method according to claim 1, characterized in that: The determining of the phase response of each channel comprises: Collecting discrete time domain data of each channel, and converting each discrete time domain data into frequency domain data; Each initial phase response is obtained based on each frequency domain data, and each initial phase response is phase unwrapped to obtain a phase response of each channel.

3. The method according to claim 1, characterized in that The phase-balanced all-pass filters of the corresponding channels are obtained based on the phase responses respectively, including: interpolating the phase response to obtain an interpolated phase response; Determine the group delay corresponding to each frequency point of the channel based on the interpolated phase response; A phase-balanced all-pass filter of the channel is obtained based on the group delay corresponding to each frequency point of the channel.

4. The method according to claim 3, characterized in that The phase-balanced all-pass filter of the channel is obtained based on the group delay corresponding to each frequency point of the channel, including: Based on the group delay corresponding to each frequency point of the channel, the expected group delay in sampling point units is obtained; Obtaining a complex cepstrum sequence based on the expected group delay in sampling point units; Obtaining an unwrapped phase and a complex cepstrum even sequence based on the complex cepstrum sequence; Obtaining the amplitude of the minimum phase signal Fourier transform based on the complex cepstrum even sequence; A phase-equalized all-pass filter of a channel is determined based on the amplitude of the Fourier transform of the minimum phase signal and the unwrapped phase.

5. The method according to any one of claims 1 to 4, characterized in that: The compensating each frequency point of the channel after each stage of cascaded all-pass filters includes: Determine a first group delay corresponding to each frequency point of the I channel after each stage of the cascaded all-pass filter and a second group delay corresponding to each frequency point of the Q channel, and obtain a group delay time difference between the channels based on the first group delay and the second group delay; Based on the group delay time difference, a time domain data point coordinate drift operation is performed on all data points of the lagging channel to compensate for each frequency point of the lagging channel.

6. The method according to claim 5, characterized in that The step of performing a time domain data point coordinate drift operation on all data points of the lagging channel based on the group delay time difference to compensate for each frequency point of the lagging channel includes: Obtaining an average delay difference based on the group delay difference; The time domain coordinates of all data points of the lagging channel are shifted leftward by the average delay difference sampling points to compensate for each frequency point of the lagging channel.

7. The method according to claim 6, characterized in that The time domain coordinates of all data points of the lagging channel are shifted to the left by the average delay difference sampling points to compensate for each frequency point of the lagging channel, including: interpolating the time domain data point coordinates of the lag channel; The coordinates of each interpolated data point are shifted leftward by the average delay difference sampling points to compensate for each frequency point of the lagging channel.

8. The method according to claim 5, characterized in that The determining of the first group delay corresponding to each frequency point of the I channel after each stage of the cascaded all-pass filter and the second group delay corresponding to each frequency point of the Q channel includes: Determine a first target phase response of the I channel and a second target phase response of the Q channel after each stage of the cascaded all-pass filter; Interpolating the first target phase response to obtain a first target interpolation phase response of the I channel, and interpolating the second target phase response to obtain a second target interpolation phase response of the Q channel; A first group delay corresponding to each frequency point of the I channel after each stage of the cascaded all-pass filter is determined based on the first target interpolation phase response, and a second group delay corresponding to each frequency point of the Q channel after each stage of the cascaded all-pass filter is determined based on the second target interpolation phase response.

9. An IQ phase imbalance broadband compensation device, characterized in that: The device comprises: A phase response determination module, used to determine the phase response of each channel; An all-pass filter determination module, used to obtain phase-balanced all-pass filters of corresponding channels based on the phase responses respectively; A cascade module, used for cascading the all-pass filter with the corresponding channel; The compensation module is used to compensate for each frequency point of the channel after each stage of the cascaded all-pass filter.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images