A method and apparatus for fast calibration of a large digital array

Abstract

The application provides a large digital array fast calibration method and device, the method comprises the following steps: S1: simultaneously sending different random sequences to multiple transmitting channels, setting the length of the random sequences to be the same and performing cyclic transmission in each channel of the antenna array; S2: a signal collector is arranged at a far-field receiving end of the antenna array, radio frequency signals transmitted by the multiple transmitting channels are received by the signal collector, the starting time point of the transmitting signals is analyzed, and then the random sequences in each transmitting channel and the digital IQ data received by the signal collector are cross-correlated; S3: according to the obtained cross-correlation value information, a channel estimation algorithm is used to calculate the multiple hardware circuit characteristics for calibration, and the characteristics are compensated to each channel for calibration. The application solves the problems of slow calibration of the existing large antenna array and the need for another public channel to realize calibration.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of large-scale digital array calibration, and more particularly to a fast calibration method and device for a large-scale digital array. BACKGROUND

[0002] The accuracy of antenna array beam output is highly dependent on the accuracy and consistency of the time delay, gain and phase of each channel. When the three indicators of each channel differ greatly, the radio frequency signal cannot form the expected beam in the air. The calibration of the gain and phase of each channel is divided into online calibration and offline calibration. Offline calibration, also known as ground calibration, mainly relies on instruments. Online calibration is divided into open-loop calibration and closed-loop calibration. Open-loop calibration relies on the compensation table of the transmit-receive channel, and closed-loop calibration relies on the feedback circuit and amplitude and phase judgment circuit provided in each channel of the array. For a digital antenna array, which usually refers to an array in which the transmit and receive signals of the antenna are digitized through a short analog circuit, the phase and amplitude changes are completed by digital processing. The commonly used method for calibrating the multiple transmit channels or multiple receive channels of a digital array is to prepare an additional common radio frequency channel. Because the interface visible to the outside of a digital array is usually a baseband on one end and a radio frequency antenna output on the other end, multiple channel calibration usually requires the use of a multi-channel phase reference instrument such as a vector network analyzer. Therefore, an additional common radio frequency channel is needed to apply both radio frequency ports in the phase reference instrument such as a vector network analyzer for calibration. In addition, the current channel calibration scheme usually calibrates each channel sequentially in time. Although this calibration performance is good, it is very time-consuming when calibrating a super large array.

[0003] Therefore, in order to solve the problem of multi-channel amplitude and phase calibration of a digital array, in the present application, when calibrating a digital array, the relative values of the time delay, amplitude and phase between multiple channels are found to achieve fast calibration of multiple channels without the need for an additional common channel. SUMMARY

[0004] The present application aims to overcome at least one of the above-mentioned defects (shortcomings) of the prior art, and provides a fast calibration method and device for a large-scale digital array, which solves the problems of slow calibration of a large-scale antenna array and the need for an additional common channel for calibration.

[0005] The technical solution adopted by the present application is a fast calibration method for a large-scale digital array, which comprises the following steps:

[0006] S1: simultaneously sending different random sequences to multiple transmit channels, and setting the lengths of the random sequences to be the same and performing cyclic transmission in each channel of the antenna array;

[0007] S2: a signal collector is arranged at a far-field receiving end of the antenna array, and radio frequency signals transmitted by the multiple transmitting channels are received by the signal collector, and a starting time point of the transmitted signal is analyzed, and then cross-correlation operations are performed on the random sequences in the respective transmitting channels and the digital IQ data received by the signal collector;

[0008] S3: multiple hardware circuit characteristics used for calibration are calculated by using a channel estimation algorithm according to the obtained cross-correlation value information, and the multiple hardware circuit characteristics are compensated into the respective channels for calibration.

[0009] In the present application, the cross-correlation performance can be effectively improved by simultaneously sending different random sequences to the multiple transmitting channels, and the synchronization accuracy and robustness of the system are enhanced. The cross-correlation operations are performed on the random sequences in the respective transmitting channels and the digital IQ data received by the signal collector through the signal collector, so that the time delay, amplitude, phase and other hardware circuit characteristics can be calculated by using a channel estimation algorithm according to the cross-correlation value, and the calibration can be realized by compensating the multiple hardware circuit characteristics into the respective channels. Therefore, a public channel is not needed, and the calibration of the multiple channels can be completed by using the transmitting and receiving channels themselves. Since the signal characteristics collected are close to the signal characteristics actually used by the antenna array, the calibration process is very fast, and the calibration of a large digital array can be completed in seconds, and the calibration accuracy fully meets the actual use.

[0010] Preferably, the step S2 comprises: setting a cross-correlation operation function according to the transmitted signal and the received signal, and the cross-correlation operation function formula is:

[0011]

[0012] wherein x[n] is the transmitted signal, y[n] is the received signal, t is the time delay, indicating the time offset of the signal y[n] relative to the signal x[n], and n is the time index.

[0013] The signal similarity between different channels is analyzed by using the cross-correlation function. Since the random sequences sent by each channel are known, the hardware circuit characteristics can be estimated by the receiving end by calculating the cross-correlation function between the received signal and the known sequence. Based on the result of the cross-correlation analysis, the calibration parameters of each channel can be calculated, and the calibration parameters are compensated into the respective channels for calibration.

[0014] Preferably, the multiple hardware circuit characteristics in the step S3 at least comprise:

[0015] The cross-correlation value R xyThe maximum value of [t] is obtained to get the best delay, so as to determine the alignment moment of the signal to realize signal synchronization; the phase difference and the amplitude difference are calculated according to the obtained cross-correlation value, and the formulas are as follows: phase difference Φ = angle (Rxy); amplitude difference A = |Rxy|.

[0016] Further preferably, the multiple hardware circuit characteristics further include an in-band fluctuation h of the channel obtained through channel estimation, and the formula is as follows:

[0017]

[0018] Wherein, F represents Fourier transform; F -1 represents inverse Fourier transform.

[0019] Therefore, by using the cross-correlation value calculated by the cross-correlation function, the time delay, amplitude, phase and in-band fluctuation and other multiple different hardware circuit characteristics can be calculated through channel estimation, so as to apply these hardware circuit characteristic compensation quantities to each transmitting channel to realize channel calibration, which can effectively improve the signal quality and help reduce the bit error rate, thereby improving the overall performance of the wireless communication system. For time delay compensation: according to the time delay estimation, the signals of each channel are time-aligned to make the starting points of the signals consistent; for amplitude compensation: by adjusting the gain of each channel, the signal amplitudes are unified; for phase compensation: by adjusting the phase of each channel, all signal phases are aligned; for in-band fluctuation compensation: the gain fluctuation and phase fluctuation of the channel are compensated to make the received signal as close as possible to the state of the transmitted signal, thereby reducing the bit error rate.

[0020] Preferably, the method further comprises interpolating the cross-correlation function to obtain a more accurate time delay. Specifically, first, the preliminary time delay t0 is calculated, that is, the maximum value of the cross-correlation value is obtained, then the interpolation interval is determined, several time points near t0 are selected for interpolation, then linear, quadratic or cubic interpolation method is used for interpolation calculation, and according to the above several points, a curve is fitted to obtain a more accurate time delay estimation, finally, the obtained more accurate time delay estimation value is used in the subsequent compensation process, further improving the calibration accuracy.

[0021] Preferably, in the step S3 of compensating the multiple hardware circuit characteristics into each channel for calibration, it further comprises monitoring whether the multiple transmitting channels maintain a distinguishable autocorrelation level to set the number of transmitting channels for simultaneous calibration, wherein the distinguishable level of each hardware circuit characteristic can be set according to the calibration requirement level of the antenna array.

[0022] Since the number of channels that can work depends on the influence of the correlation characteristics of each channel when multiple signals work together, especially when multiple signals superimpose different amplitude, phase, time delay, passband amplitude fluctuation, passband phase fluctuation and other characteristics, therefore, by judging whether the multiple channels still maintain the distinguishable autocorrelation level, the system can flexibly determine which transmitting channels are necessary under the current calibration requirement, and which can be combined or temporarily not needed for calibration. By dynamically optimizing the number of transmitting channels that need to be calibrated at the same time, the calibration process can be more flexible, effectively reducing the computational burden of the system, reducing resource consumption, and improving the overall efficiency of the calibration process.

[0023] Preferably, in the step S1, the transmission time of the sequence of part of the channels is staggered by several symbols from the transmission time of the other channels, or part of the channels use sequences of different lengths for transmission, so as to obtain better correlation performance.

[0024] Preferably, in the step S3, when calibrating, the entire channel or part of the channel of the antenna array can be selected according to the specification of the antenna array, when the multiple hardware circuit characteristics are compensated to the entire channel of the antenna array, the entire antenna array is calibrated at one time; when the multiple hardware circuit characteristics are compensated to part of the channel of the antenna array, by using the characteristics that there is a certain repeated coverage between the local arrays and the channel state at the repeated place is unchanged, the global calibration data of the entire array can be obtained by multiple local tests.

[0025] For full channel calibration: when the hardware circuit characteristics of the antenna array are compensated to the entire array, the calibration of all channels at one time can complete the overall calibration in a short time, ensuring the performance consistency and precision of the entire array, and being suitable for the scene that needs fast and full coverage calibration; for partial channel calibration: in the case of compensation only for part of the channel, by the repeated coverage characteristics of the local array, the repeated area can be used as a reference to ensure the stability and consistency of the calibration result, and the calibration can be performed in stages, without the need to process all channels at one time, which can significantly reduce the computational complexity and time consumption of each calibration, especially in the case of large-scale array, reducing the complexity of global calibration and making the calibration process more flexible.

[0026] Preferably, the method further comprises: calibrating the receiving channel, only one sequence with good autocorrelation characteristics needs to be generated, the sequence is sent to multiple receiving channels by a signal source in the far field, after the sequence is collected, it is sent to the channel estimation module for channel characteristic calculation, and the hardware circuit characteristics of the multiple channels are obtained to realize fast calibration.

[0027] The calibration of the multiple channels is performed by the same autocorrelation signal sequence, which ensures that the hardware characteristics of all receiving channels are consistent and can remain consistent in subsequent signal receiving processes, reduces errors caused by differences in hardware characteristics between different channels, and realizes rapid calibration.

[0028] In another aspect, the application also provides a rapid calibration device for a large digital array, which comprises at least a plurality of transmitting channels, a channel characteristic calculation module and a signal collector arranged in the far field of the antenna array, wherein the transmitting channels comprise at least a characteristic compensation module, a DAC+up-modulation module, a power amplifier module and an antenna unit.

[0029] The signal collector is connected to the channel characteristic calculation module, and is used to receive the radio frequency signals transmitted by the multiple transmitting channels, analyze the starting time point of the transmitted signals, then perform cross-correlation operation on the random sequence in each transmitting channel and the digital IQ data received by the signal collector, and send the obtained cross-correlation value to the channel characteristic calculation module for channel characteristic calculation to obtain the multiple hardware circuit characteristics for calibration.

[0030] One end of the channel characteristic calculation module is connected to the incident end of each transmitting channel, and is used to receive the random sequence in each channel; the other end is connected to the characteristic compensation unit of each transmitting channel, and is used to compensate the multiple hardware circuit characteristics for calibration into each transmitting channel for calibration.

[0031] Compared with the prior art, the application has the following beneficial effects: in the present application, the cross-correlation performance can be effectively improved by simultaneously sending different random sequences to multiple transmitting channels, the synchronization accuracy and robustness of the system are enhanced, and the cross-correlation operation is performed on the random sequence in each transmitting channel and the digital IQ data received by the signal collector through the signal collector, so that the time delay, amplitude, phase and other hardware circuit characteristics can be calculated by the channel estimation algorithm through the cross-correlation value, and the calibration can be realized by compensating them into each channel, so that the calibration of multiple channels can be completed through the transmitting and receiving channels themselves without the need for a public channel, and since the collected signal characteristics are close to the actual signal characteristics used by the antenna array, the calibration process is very fast, and the calibration of a large digital array can be completed in seconds, and the calibration accuracy fully meets the actual use. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1 The method flowchart provided for the embodiment is shown.

[0033] Figure 2 The transmitting channel calibration device structure diagram provided for the embodiment is shown.

[0034] Figure 3 A time-domain superposition diagram of a multi-channel calibration signal is provided for the embodiment.

[0035] Figure 4 A structure diagram of a receiving channel calibration device is provided for the embodiment.

[0036] Figure 5 Another structure diagram of a transmitting channel calibration device is provided for the embodiment. DETAILED DESCRIPTION

[0037] The drawings of the present application are only used for illustrative purposes and cannot be understood as a limitation of the present application. In order to better illustrate the following embodiments, some components in the drawings may be omitted, enlarged or reduced, and do not represent the actual product size; it is understandable for those skilled in the art that some well-known structures and their descriptions in the drawings may be omitted.

[0038] Embodiment 1

[0039] As shown in Figure 1 , the embodiment provides a fast calibration method for a large digital array, which comprises:

[0040] Step S1: simultaneously sending different random sequences to multiple transmitting channels, setting the lengths of the random sequences to be the same and performing cyclic transmission in each channel of the antenna array;

[0041] Among them, the transmitted random sequences can achieve a very high autocorrelation value and a very low cross-correlation value, and the delay, amplitude and phase of the autocorrelation are very stable, and even if the transmitter and the receiver have no time synchronization condition, the starting time point of the transmitted signal can still be quickly analyzed in the received signal, as Figure 2 To better the cross-correlation performance, multiple channels send different random sequences, and the calibration signal 1 can use but is not limited to ZC sequence, the calibration signal 2 can use but is not limited to m sequence, and the calibration signal N can use but is not limited to PN sequence, etc. The type of sequence sent by each channel can be selected according to actual needs, thereby effectively reducing interference and improving the accuracy of signal detection. In addition, by cyclically transmitting the random sequence, the signal transmitted by each channel can be repeated periodically, thereby ensuring that the receiving end accumulates enough signal samples in multiple transmissions, increasing the accuracy of calibration.

[0042] Preferably, as Figure 3As shown in the step S1, the sequence sending time of part of the channels can also be set to be staggered by several symbols from the sending time of other channels, or part of the channels can use sequences of different lengths for sending, so as to obtain better cross-correlation performance. The selection of the above channels can include but is not limited to the positions in the array that are most susceptible to interference, and can also be given according to empirical values.

[0043] Step S2: As shown in the step S2, a signal collector is arranged at a far-field receiving end of the antenna array, and is used to receive the radio frequency signals transmitted by the multiple transmitting channels, analyze the starting time point of the transmitted signals, and then perform cross-correlation operation on the random sequence in each transmitting channel and the digital IQ data received by the signal collector. Figure 2

[0044] Preferably, the step S2 includes: setting a cross-correlation operation function according to the transmitted signals and the received signals, and the cross-correlation operation function formula is:

[0045]

[0046] Wherein, x[n] is the transmitted signal; y[n] is the received signal; t is the time delay, which represents the time offset of the signal y[n] relative to the signal x[n]; and n is the time index.

[0047] By using the cross-correlation function to analyze the signal similarity between different channels, since the random sequence transmitted by each channel is known, the receiving end can estimate the hardware circuit characteristics by calculating the cross-correlation function between the received signal and the known sequence, so that based on the result of the cross-correlation analysis, the calibration parameters of each channel can be calculated, and the calibration parameters are compensated into each channel for calibration.

[0048] Step S3: The multiple hardware circuit characteristics for calibration are calculated by using a channel estimation algorithm according to the obtained cross-correlation value information, and are compensated into each channel for calibration.

[0049] Specifically, the multiple hardware circuit characteristics in the step S3 at least include:

[0050] The optimal time delay is obtained by obtaining the maximum value of the cross-correlation value R xy [t], so as to determine the alignment time of the signals and realize signal synchronization; if the value of R xy [t] is large, it means that the coincidence degree of the two signals is high under the delay t, and there is strong correlation, so that the optimal alignment time can be obtained to realize signal synchronization.

[0051] ​According to the obtained cross-correlation values, phase difference and amplitude difference are calculated, and the formulas are: phase difference Φ = angle (Rxy); amplitude difference A = |Rxy|. According to the calculated phase difference, the phase change of the signal in the process of signal transmission due to factors such as multipath effect, reflection or phase rotation is reflected, and the phase of the signal is adjusted, so that the phases of the signals are consistent. And the amplitude difference reflects the amplitude change of the signal in the process of signal transmission due to factors such as attenuation and path loss, and by gain compensation of the signal of each channel, the amplitudes of the signals of all channels are consistent.

[0052] Further preferably, the multiple hardware circuit characteristics further include the in-band fluctuation h of the channel obtained by channel estimation, and the formula is:

[0053]

[0054] Wherein, F represents Fourier transform; F -1 represents inverse Fourier transform.

[0055] The in-band fluctuation refers to the change of the channel within the transmission bandwidth, and the existence of these fluctuations will directly affect the quality of the received signal, increase the bit error rate and reduce the communication performance of the system. Therefore, it is also important to calibrate and compensate these fluctuations to ensure the reliability of the communication system.

[0056] Therefore, in this embodiment, by using the cross-correlation values calculated by the cross-correlation function, time delay, amplitude, phase and in-band fluctuation and other different hardware circuit characteristics can be calculated by channel estimation, so that these hardware circuit characteristic compensation quantities are applied to each transmitting channel to realize channel calibration, which can effectively improve the signal quality, also helps to reduce the bit error rate, thereby improving the overall performance of the wireless communication system. For time delay compensation: according to the time delay estimation, the time alignment of the signal of each channel is performed to make the starting points of the signals consistent; for amplitude compensation: by adjusting the gain of each channel, the signal amplitude is unified; for phase compensation: by adjusting the phase of each channel, all signal phases are aligned; for in-band fluctuation compensation: compensate the gain fluctuation and phase fluctuation of the channel, so that the received signal is restored to the state as close as possible to the transmitted signal, thereby reducing the bit error rate.

[0057] Preferably, the method further includes interpolating the cross-correlation function to obtain a more accurate time delay. Specifically, firstly, the initial time delay t0 is calculated, i.e., the maximum value of the cross-correlation is obtained. Then, the interpolation interval is determined, and several time points near t0 are selected for interpolation. Next, linear, quadratic, or cubic interpolation methods are used for interpolation calculation. Based on the above points, a curve is fitted to obtain a more accurate time delay estimate. Through interpolation, the time delay accuracy can be improved to the nanosecond level, thereby improving the system's ability to compensate for time delay. Finally, the obtained more accurate time delay estimate is used in subsequent compensation processes to further improve the calibration accuracy.

[0058] Preferably, in step S3, when compensating multiple hardware circuit characteristics to each path for calibration, the method further includes: monitoring whether the multiple transmission channels maintain a distinguishable autocorrelation level to set the number of transmission channels to be calibrated simultaneously, wherein the distinguishability level of each hardware circuit characteristic can be set according to the calibration requirements of the antenna array.

[0059] like Figure 2 As shown, in this embodiment, m channels are combined for transmission, and one channel is used for unified reception, i.e., m transmits and one receives. Therefore, the actual transmitted signal S and the actual received signal T are:

[0060] S = x1 + x2 + x3 + ... + x m T = y1 + y2 + y3 + ... + y m

[0061] When the cross-correlation characteristics of the transmitted sequences of each channel are good enough, the correlation result between the random sequence x transmitted by each channel and the actual received signals T of all channels is approximately equal to the correlation result when that channel transmits and receives alone. That is...

[0062]

[0063] Meanwhile, since how many channels can work depends on the influence of the correlation characteristics of each channel when multiple signals work together, especially after the multiple signals superimpose different amplitude, phase, time delay, passband amplitude fluctuation, passband phase fluctuation and other characteristics, therefore, by judging whether the multiple channels still maintain the distinguishable autocorrelation level, the system can flexibly determine which transmitting channels are necessary under the current calibration requirement, and which can be combined or temporarily not needed to calibrate. By dynamically optimizing and adjusting the number of transmitting channels that need to be calibrated at the same time, the calibration process can be more flexible, effectively reducing the computational burden of the system, reducing resource consumption, and improving the overall efficiency of the calibration process. In the embodiment, the distinguishable level can be set according to the calibration requirement level of the antenna array, for example, the amplitude distinguishable level can be set to 0.2dB, and the phase distinguishable level can be set to 2dB, and the specific indicators are set according to the actual project requirements. Therefore, according to the set distinguishable level of each hardware circuit characteristic, the number of channels that can be calibrated at the same time can be increased until the calibration resolution no longer meets the index requirements.

[0064] Preferably, when calibrating in the step S3, the calibration of all channels or partial channels of the antenna array can be selected according to the specification of the antenna array, when the multiple hardware circuit characteristics are compensated to all channels of the antenna array, the entire antenna array is calibrated at one time; when the antenna array is too large to make far-field testing difficult, only a part of the array surface can be calibrated, by compensating the multiple hardware circuit characteristics to the partial channels of the antenna array, by using the characteristics that there is a certain repeated coverage between the local arrays and the channel state of the repeated part is unchanged, the global calibration data of the entire array surface can be obtained by multiple local tests.

[0065] For all-channel calibration: when the hardware circuit characteristics of the antenna array are compensated to the entire array, the calibration of all channels at one time can complete the overall calibration in a short time, ensuring the performance consistency and accuracy of the entire array, which is suitable for scenarios that require fast and full-coverage calibration; for partial-channel calibration: in the case of compensation only for partial channels, through the repeated coverage characteristics of the local array, the repeated area can be used as a reference to ensure the stability and consistency of the calibration results, and the calibration can be performed in stages, without the need to process all channels at one time, which can significantly reduce the computational complexity and time consumption of each calibration, especially in the case of large-scale array, reducing the complexity of global calibration and making the calibration process more flexible.

[0066] Therefore, in this embodiment, simultaneously sending different random sequences to multiple transmission channels can effectively improve cross-correlation performance, enhance the synchronization accuracy and robustness of the system, and by setting up a signal acquisition device to perform cross-correlation calculations on the random sequences in each transmission channel and the digital IQ data received by the signal acquisition device, the hardware circuit characteristics such as time delay, amplitude, and phase can be calculated using the channel estimation algorithm based on the cross-correlation value. These characteristics can then be compensated into each path to achieve calibration, eliminating the need to prepare an additional common channel. The calibration of multiple channels can be completed through the transmit and receive channels themselves. Furthermore, since the acquired signal characteristics are close to the actual signal characteristics used in the antenna array, the simultaneous transmission or reception of signals through multiple channels makes the calibration process very fast, completing the calibration of large digital arrays in seconds, and the calibration accuracy fully meets the requirements of actual use.

[0067] Preferably, the method further includes: calibrating the receiving path from... Figure 4 As can be seen, it is an antenna array with one channel for transmission and m channels for reception, i.e., one transmit and m receive. Therefore, it is only necessary to generate one sequence with good autocorrelation characteristics, and send the sequence to multiple receiving channels in the mid-to-far field using a signal source. After the sequence is acquired, it is sent to the channel estimation module for channel feature calculation, so as to obtain the hardware circuit characteristics of multiple channels and realize rapid calibration.

[0068] By using the same autocorrelation signal sequence for multi-channel calibration, it is ensured that the hardware characteristics of all receiving channels are consistent and can maintain consistency in subsequent signal reception, reducing errors caused by differences in hardware characteristics between different channels and achieving rapid calibration.

[0069] Example 2

[0070] like Figure 2 As shown, this embodiment provides a rapid calibration device for a large digital array. The device includes at least several transmission channels, a channel feature calculation module, and a signal acquisition device set in the far field of the antenna array. The transmission channel includes at least a feature compensation module, a DAC+ up-modulation module, a power amplifier module, and an antenna unit, and the feature compensation module, the DAC+ up-modulation module, the power amplifier module, and the antenna unit are connected in sequence.

[0071] The signal acquisition unit is connected to the channel feature calculation module and is used to receive radio frequency signals transmitted from multiple transmission channels, analyze the start time of the transmission signal, and then perform cross-correlation calculation with the random sequence in each transmission channel and the digital IQ data received by the signal acquisition unit. The obtained cross-correlation value is sent to the channel feature calculation module to calculate the channel feature and obtain the multiple hardware circuit characteristics used for calibration.

[0072] One end of the channel feature calculation module is connected to the incident end of each transmitting channel for receiving the random sequence in each channel, and the other end is connected to the feature compensation unit of each transmitting channel for compensating the multiple hardware circuit features used for calibration into each transmitting channel for calibration.

[0073] The large digital array fast calibration device provided by the embodiment of the present scheme is used for executing the large digital array fast calibration method described in Embodiment 1, the implementation manners of which are consistent with the implementation manners of the large digital array fast calibration method provided by Embodiment 1 of the present scheme, and the same beneficial effects can be achieved, which will not be described here again.

[0074] Embodiment 3

[0075] As shown in Figure 5 In order to further calibrate the large digital array, reduce the bit error rate, and improve the quality of the signal, an interference estimation module is further added in the present embodiment to offset the influence of the downlink received signal intermodulation on the uplink transmitted signal, so that the anti-interference performance of the antenna array is further improved by combining the two architectures, and the quality of the signal is improved, as shown in Figure 5 One end of the interference estimation module is connected behind the feature compensation module of the transmitting channel for collecting data from the transmitting channel that needs to be interfered, and the other end is connected to the signal collector arranged at the far-field receiving end for collecting data from the receiving channel.

[0076] Then, according to the collected data, the interference estimation module is optimized to obtain the cancellation coefficient of the receiving channel and the uplink transmission interference, and the cancellation coefficient is used to construct an anti-interference model. In the anti-interference model, the cancellation coefficient and the downlink received signal are combined to obtain a cancellation signal that is equal in size and opposite in phase to the uplink transmission interference. The generated cancellation signal is added or subtracted to the uplink transmission signal to affect the phase, thereby effectively solving the out-of-band interference of the transmission to the reception in the antenna array.

[0077] Specifically, the optimization processing includes:

[0078] Step S41: According to the collected data, the downlink received signal and the power amplifier intermodulation model are used to construct a fitting intermodulation signal, and the fitting intermodulation signal is uplink shaped filtered to obtain a fitting intermodulation signal containing only the uplink passband part;

[0079] Preferably, in the present embodiment, the power amplifier intermodulation model selects the Volterra series model, and the actual used power amplifier intermodulation model is obtained for the component of the signal falling in the uplink passband, and the formula is:

[0080] Y'(k) = y(k)G(t)

[0081] Wherein, y(k) is Volterra series model output; G(t) is the output of the uplink shaping filter; Y'(k) represents the output of the power amplifier intermodulation model, which is the fitting intermodulation signal containing only the uplink passband part.

[0082] And since the power amplifier model is centered on the downlink received signal, and the uplink shaping filter is centered on the uplink transmitted signal, the power amplifier intermodulation model also needs to be band-limited to align the frequency points of the power amplifier intermodulation model and the uplink shaping filter.

[0083] Step S42: Perform cross-correlation operation on the filtered fitting intermodulation signal in step S41 and the actual uplink transmitted interference signal, and compare the cross-correlation operation result with a preset threshold value. If the cross-correlation result is greater than the threshold value, it indicates that the intermodulation of the downlink receiving channel has an impact on the reception of the uplink transmitted signal.

[0084] The cross-correlation operation formula is:

[0085]

[0086] Wherein, x * (k-t) represents the transmitted signal; Y'(k) represents the output of the power amplifier intermodulation model; r yx (t) represents the cross-correlation result.

[0087] Step S43: If the cross-correlation result is greater than the threshold value, use an optimization algorithm to estimate the cancellation coefficient of the downlink receiving channel and the uplink transmitted interference, and use the cancellation coefficient to construct an anti-interference model. In the anti-interference model, the cancellation coefficient and the downlink received signal are combined to obtain a cancellation signal that is equal in size and opposite in phase to the uplink transmitted interference. The generated cancellation signal is added or subtracted from the uplink transmitted signal to affect the phase, thereby eliminating the interference falling within the uplink band.

[0088] Preferably, in the step S43, the use of the optimization algorithm to estimate the cancellation coefficient of the downlink receiving channel and the uplink transmitted interference specifically includes:

[0089] S51: Use the actual uplink transmitted interference signal x(k) to subtract the output Y'(k) of the power amplifier intermodulation model to obtain the error e(k) of the anti-interference model;

[0090] e(k) = x(k) - Y'(k)

[0091] Use an optimization algorithm to find a suitable cancellation coefficient h such that

[0092]

[0093] Since the overdetermined equations do not have a unique solution, the least squares method is used to find an approximate solution for the optimal offset coefficient, thus obtaining a series of offset coefficients, the formula of which is:

[0094] H=(y*y H ) -1 y H x

[0095] Where y represents the downlink received signal; x represents the uplink transmitted signal; and H represents the set of cancellation coefficients.

[0096] If the series of cancellation coefficients H obtained from the above steps are all insufficient to reduce the interference signal below the ground noise level, then the following operation is performed:

[0097] S52: Select the set of a cancellation coefficients with the best cancellation effect to form a new initial set, denoted as initial set C = [H1H2H3…H ... a ];

[0098] S53: Optimize the cancellation coefficients in the initial set C until the obtained cancellation coefficients can cancel the interference signal below the ground noise.

[0099] Specifically, the optimization process in step S53 includes:

[0100] S531: Select the two sets of offset coefficients with the best offset effect from the initial set C, then randomly swap a of the offset coefficients, and calculate the offset error of the two new offset coefficient sets. If the offset error of the new offset coefficient set is more than 1 dB better than the offset error of the worst performing set in the initial set C, then add the newly generated offset coefficient sets H′1 and H′2 to the initial set C to form a new set C1.

[0101] S532: Randomly select two sets of offset coefficients from set C1 and randomly swap a coefficients in them. Calculate the offset error of the two new offset coefficient sets. If the offset error of the new offset coefficient set is more than 1 dB better than the offset error of the worst performing group in the initial set C, then add the newly generated offset coefficient sets H″1 and H″2 to set C1 to form a new set C2.

[0102] S533: Randomly select a set of offset coefficients from set C2, and then randomly select from this set of coefficients... There are several offsetting coefficients, where b is the number of offsetting coefficients in this set. This indicates rounding up; then the amplitude of these selected offset coefficients is changed to random values ​​in the range [0.95A, 1.05A] until the offset error of the new offset coefficient set is more than 1dB better than the offset error of the worst performing group in the initial set C. Then the newly generated offset coefficient set is added to the initial set C to form a new set C3.

[0103] S534: judging the number of the cancellation coefficient sets in the set C3, if greater than 4b, deleting some sets of the cancellation coefficients with the worst performance until the number of the elements of the set is less than or equal to 4b;

[0104] S535: repeating the steps S531-S534 until the obtained cancellation coefficients can cancel the interference signal to below the noise floor.

[0105] Therefore, in the embodiment, in the case that the existing large antenna array calibration is slow and needs to additionally prepare a public channel to realize calibration, the problems are effectively solved in the embodiments 1 and 2, and further combined with the digital interference cancellation mode to further solve the out-of-band interference of the transmission to the reception in the antenna array, so that the large digital array fast calibration is realized by combining the two architectures, and the anti-interference performance of the antenna array is improved.

[0106] Obviously, the above embodiments of the present application are only examples for clearly illustrating the technical solutions of the present application, and are not intended to limit the specific embodiments of the present application. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for fast calibration of a large digital array, characterized in that, The method includes: S1: Simultaneously transmit different random sequences to multiple transmission channels, setting the random sequences to have the same length and transmitting them cyclically in each channel of the antenna array; S2: A signal acquisition device is set at the far-field receiver in the antenna array, and it is used to receive the radio frequency signals transmitted from multiple transmission channels, analyze the start time of the transmission signal, and then perform cross-correlation calculations on the random sequence in each transmission channel and the digital IQ data received by the signal acquisition device. S3: Based on the obtained cross-correlation information, the channel estimation algorithm is used to calculate the multiple hardware circuit characteristics used for calibration, and these characteristics are compensated into each path for calibration. The transmission channel includes a feature compensation module; An interference estimation module is also connected after the feature compensation module and the signal acquisition device. After the interference estimation module collects data from the feature compensation module and the signal acquisition unit, it obtains a cancellation coefficient based on the collected data to establish an anti-interference model; and obtains a cancellation signal through the interference model to act on the out-of-band interference of the transmitter to the receiver in the antenna array. The cancellation coefficient is obtained based on the collected data, specifically by obtaining the cancellation coefficient for the receiving channel and uplink transmission interference based on the collected data, including the following steps: S41. Construct a fitted intermodulation signal using the downlink received signal and the power amplifier intermodulation model, and perform uplink shaping filtering on it to obtain a fitted intermodulation signal that only contains the uplink passband portion. S42. Perform cross-correlation calculation between the fitted intermodulation signal and the actual uplink transmitted interference signal: S43. If the cross-correlation result is greater than the threshold, use an optimization algorithm to estimate the cancellation coefficient of downlink receive channel and uplink transmit interference; In step S43, an optimization algorithm is used to estimate the cancellation coefficient of the downlink receive channel and the uplink transmit interference, specifically including the following steps: S51: Obtain the error of the anti-interference model by subtracting the output of the power amplifier intermodulation model from the actual uplink transmitted interference signal; Use an optimization algorithm to find a suitable offset coefficient h that minimizes the expected value of the square of the error; The least squares method is used to find an approximate solution for the optimal offset coefficient, thus obtaining the set of offset coefficients, the formula of which is: Where y represents the downlink received signal; x represents the uplink transmitted signal; and H represents the set of cancellation coefficients.

2. The method of claim 1, wherein, Step S2 includes: setting a cross-correlation function based on the transmitted signal and the received signal, wherein the formula for the cross-correlation function is: in, To transmit signals; For receiving the signal; t is the time delay, representing the signal... Relative to signal The time offset; n is the time index.

3. The method of claim 2, wherein, The multiple hardware circuit features mentioned in step S3 include at least the following: By obtaining cross-correlation values The optimal time delay is obtained by maximizing the cross-correlation value, thereby determining the signal alignment time and achieving signal synchronization; the phase difference and amplitude difference are calculated based on the obtained cross-correlation value, and their formulas are as follows: Phase difference ; Amplitude difference .

4. The method of claim 2, wherein, The multiple hardware circuit characteristics also include obtaining the in-band ripple h of the channel through channel estimation, the formula of which is: wherein denotes the Fourier transform; denotes the inverse Fourier transform.

5. The method of claim 3, wherein, The method also includes interpolating the cross-correlation function to obtain a more accurate time delay.

6. A method of fast calibration of a large digital array according to any of claims 1-5, characterized in that, In step S3, when compensating multiple hardware circuit characteristics to each path for calibration, the method further includes: monitoring whether the multiple transmission channels maintain a distinguishable autocorrelation level to set the number of transmission channels to be calibrated simultaneously, wherein the distinguishability level of each hardware circuit characteristic can be set according to the calibration requirements of the antenna array.

7. A method of fast calibration of a large digital array according to any of claims 1-5, characterized in that, In the step S1, the transmission time of the sequence of part of the channels can be staggered with other channels by several symbols, or part of the channels use sequences of different lengths for transmission, so as to obtain better cross-correlation performance.

8. A method for fast calibration of a large digital array according to any of claims 1-5, characterized in that, In the step S3, when calibration is performed, the calibration of all channels or part of the channels of the antenna array can be selected according to the specification of the antenna array, when the multiple hardware circuit characteristics are compensated into all channels of the antenna array, the entire antenna array is calibrated at one time; when the multiple hardware circuit characteristics are compensated into part of the channels of the antenna array, by virtue of the characteristics that there is a certain repeated coverage between local arrays and the state of the channels at the repeated positions is unchanged, the global calibration data of all array surfaces can be obtained by performing multiple local tests.

9. The method of claim 7, wherein, The method further comprises: calibrating the receiving channel, only one sequence with good autocorrelation characteristics needs to be generated, the sequence is transmitted into multiple receiving channels by using a signal source in the middle and far field, after the sequence is collected, the sequence is sent into a channel estimation module for channel characteristic calculation, so as to obtain the hardware circuit characteristics of the multiple channels and realize fast calibration.

10. A fast calibration device for a large digital array according to the method of any one of claims 1-9, characterized in that, The device at least comprises a plurality of transmitting channels, a channel characteristic calculation module and a signal collector arranged in the middle and far field of the antenna array, wherein the transmitting channel at least comprises a characteristic compensation module, a DAC+up-modulation module, a power amplifier module and an antenna unit. The signal collector is connected with the channel characteristic calculation module, is used for receiving the radio frequency signals transmitted by the multiple transmitting channels, analyzing the starting time point of the transmitted signals, then performing cross-correlation operation on the random sequence in each transmitting channel and the digital IQ data received by the signal collector, and sending the obtained cross-correlation value to the channel characteristic calculation module for channel characteristic calculation to obtain the multiple hardware circuit characteristics used for calibration. One end of the channel characteristic calculation module is connected to the incident end of each transmitting channel, is used for receiving the random sequence in each channel, and the other end is connected with the characteristic compensation unit of each transmitting channel, is used for compensating the multiple hardware circuit characteristics used for calibration into each transmitting channel for calibration.

Images