An open-loop method and device for removing out-of-band nonlinear interference to a receiving array

Abstract

The present invention relates to the technical field of antenna array transceiver interference removal, and more specifically, to an open-loop method and apparatus for removing out-of-band nonlinear interference from a receiving array. The method comprises: S1: when the array is not operating, using a search algorithm to find the transmit predistortion coefficients for each channel that minimize the interference of the transmit array on the receiving array; S2: when the array is operating, directly convolving the predistortion coefficients obtained in step S1 with the downlink transmit signal, thereby applying them to each transmit channel to destroy the coherence of the transmit nonlinear signal. The present invention effectively resolves out-of-band interference from transmission to reception in an antenna array, significantly reducing the need for filters for out-of-band suppression in the transmit channel, reducing the array's power consumption, volume, and weight, and improving the array's efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of antenna array transmission and reception interference removal, and more particularly to an open-loop method and device for removing out-of-band nonlinear interference to a receiving array. Background Art

[0002] In a coplanar antenna array with coplanar transmit and receive signals, when the antenna operates in FDD mode, the transmitted signal generates a nonlinear spectrum due to the nonlinear characteristics of the transmit channel. This spectrum typically falls within the receive frequency band, and even after isolation of the transmit and receive antennas and suppression by the transmit out-of-band filter, some energy still remains.

[0003] In non-array systems, such as a one-receiver-one-transmitter FDD RF system, the common practice is to assume that as long as the energy of the transmitted nonlinear spectrum falling into the receiving channel is lower than a certain threshold of the air noise floor at the input port of the receiving path, the impact on the receiving channel is considered negligible. Therefore, the out-of-band suppression of the filter at the transmit output port is usually increased to suppress interference with the receiving channel.

[0004] However, in an array system, even if the nonlinear spectrum emitted by each pair of transmitting and receiving antennas falls into the receiving channel below a certain threshold of the air noise floor, the coherence of the main signals transmitted through the multi-path causes their nonlinear spectra to also be coherent, which in turn causes the spectra falling into the receiving path to also be coherent. Therefore, these coherent signals will be amplitude- and phase-weighted by the receiving array, thus being higher than the thermal noise combined from the entire receiving array, thereby affecting the signal-to-noise ratio of the receiving link.

[0005] Typically, energy outside the transmission band is suppressed by adding a filter after the last amplifier stage. However, increasing the filter's out-of-band suppression will result in a decrease in the gain within the transmission band and a corresponding increase in the filter's size and weight, thus affecting the efficiency, power consumption, size, and weight of the transmission link.

[0006] To minimize the efficiency, power consumption, and size of the transmit link, a method is proposed to address the out-of-band interference problem in the antenna array by pre-distorting the signals in each transmit path without imposing excessively stringent requirements on the transmit filter. Summary of the Invention

[0007] The present invention aims to overcome at least one defect (shortcoming) of the above-mentioned prior art and provides an open-loop method and device for removing out-of-band nonlinear interference of the transmission on the receiving array, which effectively solves the out-of-band interference of the transmission on the reception in the antenna array, greatly reduces the demand for out-of-band suppression of the filter in the transmission channel, reduces the power consumption, volume and weight of the array, and improves the efficiency of the array.

[0008] In a first aspect, the technical solution adopted by the present invention is an open-loop method for removing interference of out-of-band nonlinearity on a receiving array, the method comprising:

[0009] S1: When the array is not in operation, a search algorithm is used to find the transmit predistortion coefficients of each channel that minimize the interference of the transmit array on the receive array.

[0010] S2: When the array is in operation, the predistortion coefficients obtained in step S1 are directly convolved with the downlink transmission signal, and are thus applied to each transmission channel to destroy the coherence of the transmission nonlinear signal.

[0011] In this application, digital pre-distortion is applied to the array. There is no need to suppress the out-of-band nonlinearity of each transmitting power amplifier. Instead, the transmitting antenna array normally forms a data beam, and the receiving channel receives the transmitted out-of-band leakage signal. The transmit pre-distortion coefficient of each channel with the least interference from the transmitting array to the receiving array is applied to each transmitting channel, thereby minimizing the transmit nonlinearity of the receiving channel. The pre-distortion coefficient is only for the out-of-band spectrum of the transmitting signal, and the coherence of the normal in-band signal is not destroyed, which does not affect the transmit and receive beams of the antenna array. Moreover, the pre-distortion coefficient is not only for a single power amplifier, but can be optimized based on the entire receiving array, thereby effectively solving the out-of-band interference of the transmission on the reception in the antenna array, greatly reducing the demand for out-of-band suppression of the filter by the transmitting channel, reducing the power consumption, volume and weight of the array, and improving the efficiency of the array.

[0012] Preferably, the search algorithm in step S1 is a genetic algorithm.

[0013] By optimizing the pre-distortion coefficients of the transmitting array through a genetic algorithm, the interference to the receiving array is minimized, which can improve the signal quality of the communication system, reduce the interference caused by nonlinear distortion, and improve the spectrum efficiency. It can also adapt to different channel conditions and interference environments, ensuring that good optimization effects are maintained in a dynamically changing communication environment.

[0014] Preferably, in step S1, a genetic algorithm is used to find the transmission predistortion coefficients of each channel that minimize the interference of the transmitting array on the receiving array, including:

[0015] S31: Initialize the population and generate several different transmit predistortion coefficient combinations d = [d1, d2, ..., d m ], where m is the number of transmit channels and d is the transmit predistortion coefficient;

[0016] S32: Establishing a fitness function and calculating the fitness corresponding to each transmit predistortion coefficient combination;

[0017] S33: Generate a new transmit predistortion coefficient combination through iterative optimization through selection, crossover and mutation operations, then repeat the operation of step S32, recalculate the fitness corresponding to each new transmit predistortion coefficient combination, and output the optimal transmit predistortion coefficient combination when the termination condition is met.

[0018] By continuously iteratively updating the genetic algorithm, the optimal predistortion coefficient with minimum interference is finally obtained. Applying it to each transmitting channel can prevent the nonlinear spectrum generated by the transmitting channel from falling into the receiving frequency band and affecting the quality of the received signal, effectively avoiding interference with the received signal.

[0019] Preferably, in step S32, the fitness function is composed of several performance indicators, and can be optimized by adjusting the value of a weight coefficient according to different application requirements, wherein the value of the weight coefficient is greater than 0.

[0020] The more performance indicators included in the fitness function, and the importance of different performance indicators is adjusted by adjusting the value of the weight coefficient according to different application requirements, so that the obtained fitness value reflects a better effect of the set of pre-distortion coefficients on interference and signal quality optimization, and ultimately makes the obtained pre-distortion coefficients more ideal, making the antenna array more anti-interference.

[0021] Preferably, the fitness function includes at least the performance indicators of maximum signal-to-noise ratio, minimum intermodulation distortion, minimum interference power and minimum mean square error. The obtained fitness function formula is:

[0022]

[0023] Among them, P signal is the power of the signal; P noise is the power of noise; IMD is the minimum intermodulation distortion; P interference is to minimize the interference power; MSE is to minimize the mean square error; ω1, ω2, ω3, and ω4 are the weight coefficients of the above performance indicators respectively.

[0024] In the fitness function formula, these indicators can be balanced by setting different weight coefficients. For example, if SNR is more important, the weight of ω1 can be increased; if IMD is more important, the weight of ω2 can be increased accordingly, thereby outputting the optimal solution and obtaining the optimal combination of transmit predistortion coefficients, thereby minimizing the interference of the transmit array on the receive array and optimizing system performance.

[0025] Preferably, step S33 includes: selecting a transmit predistortion coefficient combination with higher fitness as a parent individual, then performing a crossover operation to generate a new transmit predistortion coefficient combination, then randomly mutating some of the transmit predistortion coefficients in the new transmit predistortion coefficient combination to introduce new genetic information, and finally re-iterating the operations of steps S32-S33 until the optimal transmit predistortion coefficient combination is obtained.

[0026] The above operation can introduce diversity in the transmit predistortion coefficients, prevent the algorithm from falling into a local optimal solution, and make the obtained transmit predistortion coefficient combination close to the global optimal solution, so that it can be applied to various transmit channels to effectively improve the performance of the antenna array.

[0027] Preferably, in step S2, the predistortion coefficient obtained in step S1 is directly convolved with the downlink transmission signal, thereby being applied to each transmission channel to destroy the coherence of the transmission nonlinear signal. The processing formula is:

[0028]

[0029] in, is the downlink transmission signal after predistortion processing; d i is the transmit predistortion coefficient of the i-th transmit channel; U i is the i-th downlink transmission signal.

[0030] By using the obtained optimal transmit predistortion coefficient and applying it to the transmit channel through the above formula, the interference of nonlinear distortion on the receiving array is effectively reduced, the spectrum efficiency is improved, and the signal quality of the communication system is improved.

[0031] Preferably, the termination condition is that the fitness reaches a preset threshold or reaches a maximum number of iterations. The termination condition can be selected according to actual needs to obtain the desired optimized transmit predistortion coefficients and reduce the impact of interference in the receiving array.

[0032] On the other hand, the present invention also provides an open-loop device for removing interference of nonlinearities outside the transmission band on a receiving array, the device comprising at least:

[0033] A parameter processing unit, used for finding the transmission predistortion coefficient of each channel with the minimum interference of the transmitting array to the receiving array by using a genetic algorithm when the array is not in operation;

[0034] The signal control unit is used to directly convolve the optimized pre-distortion coefficient with the downlink transmission signal when the array is in working state, thereby applying the coefficient to each transmission channel to realize signal pre-distortion processing.

[0035] In this device, the transmission predistortion coefficient of each channel is obtained through the parameter processing unit, and the digital predistortion method is applied to the array. There is no need to suppress the out-of-band nonlinearity of each transmitting power amplifier. Instead, the transmitting antenna array normally sends a data beam, and the receiving channel receives the transmitted out-of-band leakage signal. The transmission predistortion coefficient of each channel with the minimum interference of the transmitting array on the receiving array is applied to each transmitting channel, thereby minimizing the transmission nonlinearity of the receiving channel. The predistortion coefficient is only for the out-of-band spectrum of the transmitting signal, and the coherence of the normal in-band signal is not destroyed, which does not affect the transmit and receive beams of the antenna array. In addition, the predistortion coefficient is not only for a single power amplifier, but can be optimized based on the entire receiving array, thereby effectively solving the out-of-band interference of the transmission on the reception in the antenna array, greatly reducing the demand for the out-of-band suppression of the filter by the transmitting channel, reducing the power consumption, volume and weight of the array, and improving the efficiency of the array.

[0036] Preferably, the parameter processing unit includes:

[0037] Initialization component: Initialize the population and generate several different transmit predistortion coefficient combinations d = [d1, d2, ..., d m ], where m is the number of transmit channels and d is the transmit predistortion coefficient;

[0038] Fitness calculation component: establishes a fitness function and calculates the fitness corresponding to each transmit predistortion coefficient combination;

[0039] Optimization Iteration Component: Generates a new transmit predistortion coefficient combination through iterative optimization through selection, crossover, and mutation operations. Then, it re-enters the fitness calculation component for fitness calculation. When the termination condition is met, it outputs the optimal transmit predistortion coefficient combination.

[0040] Among them, in the fitness calculation component, the fitness function formula is:

[0041]

[0042] Among them, P signal is the power of the signal; P noise is the power of noise; IMD is the minimum intermodulation distortion; P interference is to minimize the interference power; MSE is to minimize the mean square error; ω1, ω2, ω3, and ω4 are the weight coefficients of the above performance indicators respectively.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention applies digital predistortion to an array. It does not require suppressing the out-of-band nonlinearity of each transmitting power amplifier. Instead, the transmitting antenna array normally forms a data beam, and the receiving channel receives the transmitted out-of-band leakage signal. The transmit predistortion coefficients of each channel with the least interference from the transmitting array to the receiving array are applied to each transmitting channel, thereby minimizing the transmit nonlinearity of the receiving channel. The predistortion coefficients are only targeted at the out-of-band spectrum of the transmitted signal, and the coherence of the normal in-band signal is not destroyed, which does not affect the transmit and receive beams of the antenna array. In addition, the predistortion coefficients are not only targeted at a single power amplifier, but can be optimized based on the entire receiving array, thereby effectively solving the out-of-band interference of transmission on reception in the antenna array, greatly reducing the transmit channel's demand for out-of-band suppression of the filter, reducing the array's power consumption, volume, and weight, and improving the array's efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flow chart of the method provided in this embodiment.

[0046] Figure 2 This is a schematic diagram of the device structure provided in this embodiment.

[0047] Figure 3 This is a schematic diagram of another device structure provided in this embodiment. DETAILED DESCRIPTION

[0048] The accompanying drawings are for illustrative purposes only and are not to be construed as limiting the present invention. To better illustrate the following embodiments, some components in the accompanying drawings may be omitted, enlarged, or reduced in size, and do not represent actual product dimensions. Those skilled in the art will appreciate that some well-known structures and their descriptions may be omitted from the accompanying drawings.

[0049] Example 1

[0050] like Figure 1 As shown, the technical solution adopted by this embodiment of the present invention is an open-loop method for removing interference of nonlinearity outside the transmission band on the receiving array, and the method includes:

[0051] S1: When the array is not in operation, a search algorithm is used to find the transmit predistortion coefficients of each channel that minimize the interference of the transmit array on the receive array.

[0052] S2: When the array is in operation, the predistortion coefficients obtained in step S1 are directly convolved with the downlink transmission signal, and are thus applied to each transmission channel to destroy the coherence of the transmission nonlinear signal.

[0053] In this application, digital pre-distortion is applied to the array. There is no need to suppress the out-of-band nonlinearity of each transmitting power amplifier. Instead, the transmitting antenna array normally forms a data beam, and the receiving channel receives the transmitted out-of-band leakage signal. The transmit pre-distortion coefficient of each channel with the least interference from the transmitting array to the receiving array is applied to each transmitting channel, thereby minimizing the transmit nonlinearity of the receiving channel. The pre-distortion coefficient is only for the out-of-band spectrum of the transmitting signal, and the coherence of the normal in-band signal is not destroyed, which does not affect the transmit and receive beams of the antenna array. Moreover, the pre-distortion coefficient is not only for a single power amplifier, but can be optimized based on the entire receiving array, thereby effectively solving the out-of-band interference of the transmission on the reception in the antenna array, greatly reducing the demand for out-of-band suppression of the filter by the transmitting channel, reducing the power consumption, volume and weight of the array, and improving the efficiency of the array.

[0054] In this embodiment, a typical microstrip coplanar antenna pattern is provided, such as Figure 2 As shown, the blue is the transmitting microstrip antenna, the light orange is the receiving microstrip antenna, the transmitting and receiving are coplanar antennas, on the same PCB stack, sharing the same antenna interface, and the typical transmit-receive isolation is around 20dB. For example, if the maximum output peak power of the final power amplifier of the transmitting antenna can reach 39.5dBm, and the fallback uses an average power of 30dBm, the spectrum leakage of the transmission in the receiving frequency band is mainly the nonlinear intermodulation product of the transmitting signal. Due to different power amplifier characteristics, the nonlinear products are also different. For example, a reasonable nonlinear value of 60dBc for the 7th order nonlinearity is taken, that is, the average power of the transmission falling into the receiving band is 30dBm-60dBm. c = -30dBm. For a 20MHz OFDM signal, its power spectral density is -30dBm-73dB = -103dBm / Hz. Typically, an S-segment dielectric cavity filter has a suppression factor of 60 to 70dBc with 60MHz transmit / receive isolation. For example, if the array gain of a large receiving antenna is 38dB, the total transmitted interference received by the array is -103dBm / Hz-70dBc-20dBc+38dB = -155dBm / Hz. This value is far greater than the noise floor of the receiving array, which is a spectral density of approximately -174dBm / Hz.

[0055] Secondly, in addition to considering the interference within the same group of transmit and receive antennas, it is also necessary to consider the interference between the transmit and receive antennas of the entire array. Considering the coupling of all transmitting antennas to the same receiving antenna unit, this superposition is still coherent superposition. After simulation and testing, it was found that the signal received by a single receiving antenna in the middle position is not only the previously estimated -103dBm / Hz - 70dB - 20dB = -193dBm / Hz, but is increased by 10dB to -183dBm / Hz. The total interference received by the entire receiving array will also increase further, that is, not just -155dBm / Hz, but close to -145dBm / Hz.

[0056] As can be seen from the above discussion, the interference caused by out-of-band transmission on reception is coherent. Therefore, the higher the gain of the receiving antenna array, the greater the nonlinearity of the received transmission leakage. Therefore, in the present invention, a digital pre-distortion method is adopted. When the array is not in operation, a genetic algorithm is used to find the transmission pre-distortion coefficients of each path that minimize the interference of the transmitting array on the receiving array. After recording them, these pre-distortion coefficients are applied to each transmission channel when the antenna is in operation, thereby destroying the coherence of the transmitted nonlinear signal.

[0057] Preferably, the search algorithm in step S1 is a genetic algorithm.

[0058] By optimizing the pre-distortion coefficients of the transmitting array through a genetic algorithm, the interference to the receiving array is minimized, which can improve the signal quality of the communication system, reduce the interference caused by nonlinear distortion, and improve the spectrum efficiency. It can also adapt to different channel conditions and interference environments, ensuring that good optimization effects are maintained in a dynamically changing communication environment.

[0059] Preferably, in step S1, a genetic algorithm is used to find the transmission predistortion coefficients of each channel that minimize the interference of the transmitting array on the receiving array, including:

[0060] S31: Initialize the population and generate several different transmit predistortion coefficient combinations d = [d1, d2, ..., d m ], where m is the number of transmit channels and d is the transmit predistortion coefficient;

[0061] S32: Establishing a fitness function and calculating the fitness corresponding to each transmit predistortion coefficient combination;

[0062] S33: Generate a new transmit predistortion coefficient combination through iterative optimization through selection, crossover and mutation operations, then repeat the operation of step S32, recalculate the fitness corresponding to each new transmit predistortion coefficient combination, and output the optimal transmit predistortion coefficient combination when the termination condition is met.

[0063] By continuously iteratively updating the genetic algorithm, the optimal predistortion coefficient with minimum interference is finally obtained. Applying it to each transmitting channel can prevent the nonlinear spectrum generated by the transmitting channel from falling into the receiving frequency band and affecting the quality of the received signal, effectively avoiding interference with the received signal.

[0064] Preferably, in step S32, the fitness function is composed of several performance indicators, and can be optimized by adjusting the value of a weight coefficient according to different application requirements, wherein the value of the weight coefficient is greater than 0.

[0065] The more performance indicators included in the fitness function, and the importance of different performance indicators is adjusted by adjusting the value of the weight coefficient according to different application requirements, so that the obtained fitness value reflects a better effect of the set of pre-distortion coefficients on interference and signal quality optimization, and ultimately makes the obtained pre-distortion coefficients more ideal, making the antenna array more anti-interference.

[0066] Preferably, the fitness function includes at least the performance indicators of maximum signal-to-noise ratio, minimum intermodulation distortion, minimum interference power and minimum mean square error. The obtained fitness function formula is:

[0067]

[0068] Among them, P signal is the power of the signal; P noise is the power of noise; IMD is the minimum intermodulation distortion; P interference is to minimize the interference power; MSE is to minimize the mean square error; ω1, ω2, ω3, and ω4 are the weight coefficients of the above performance indicators respectively.

[0069] Among them, in the above formula The goal is to maximize the signal-to-noise ratio (SNR). A higher SNR indicates a higher fitness. Intermodulation distortion (IMD), interference power, and mean square error (MSE) are typically minimized, so their weights should be negative. Lower IMD, interference, and error indicate better system performance and higher fitness.

[0070] In the fitness function formula, the relative importance of these indicators in the fitness function can be emphasized by increasing their weight coefficient values ​​or adjusting the weight ratio between them. If a certain indicator is more critical to the optimization, it can be given a larger weight. For example, if SNR is more important, the weight of ω1 can be increased; if IMD is more important, the weight of ω2 can be increased accordingly, so as to output the optimal solution and obtain the optimal transmit predistortion coefficient combination, thereby minimizing the interference of the transmit array on the receive array and optimizing the system performance.

[0071] Preferably, the fitness function formula may optionally further include performance indicators such as the signal-to-interference ratio (SIR) and the out-of-band power (OOBP) to determine the ideality of the transmit predistortion coefficient. The resulting formula is as follows:

[0072]

[0073] For SNR (signal-to-noise ratio), a higher SNR is generally preferred, as it represents signal quality and reception clarity. For IMD (intermodulation distortion), a lower IMD is preferred, as it affects signal quality. For interference power, a lower MSE is preferred, as reducing interference improves communication quality. For mean square error (MSE), a lower MSE is preferred, as it reflects the system's accuracy and error level. For SIR (signal-to-interference ratio), a higher SIR is preferred, as it represents the ratio of signal to interference strength, and a higher SIR helps improve the system's anti-interference capability. For out-of-band power (OOBP), a lower OOBP is preferred, as high OOBP can cause spectrum pollution. The weights assigned to each objective can be adjusted based on actual application requirements and optimization goals.

[0074] Preferably, step S33 includes: selecting a transmit predistortion coefficient combination with higher fitness as a parent individual, then performing a crossover operation to generate a new transmit predistortion coefficient combination, then randomly mutating some of the transmit predistortion coefficients in the new transmit predistortion coefficient combination to introduce new genetic information, and finally re-iterating the operations of steps S32-S33 until the optimal transmit predistortion coefficient combination is obtained.

[0075] The above operation can introduce diversity in the transmit predistortion coefficients, prevent the algorithm from falling into a local optimal solution, and make the obtained transmit predistortion coefficient combination close to the global optimal solution, so that it can be applied to various transmit channels to effectively improve the performance of the antenna array.

[0076] Preferably, in step S2, the predistortion coefficient obtained in step S1 is directly convolved with the downlink transmission signal, thereby being applied to each transmission channel to destroy the coherence of the transmission nonlinear signal. The processing formula is:

[0077]

[0078] in, is the downlink transmission signal after predistortion processing; d i is the transmit predistortion coefficient of the i-th transmit channel; U i is the i-th downlink transmission signal.

[0079] By using the obtained optimal transmit predistortion coefficient and applying it to the transmit channel through the above formula, the interference of nonlinear distortion on the receiving array is effectively reduced, the spectrum efficiency is improved, and the signal quality of the communication system is improved.

[0080] Preferably, the termination condition is that the fitness reaches a preset threshold or reaches a maximum number of iterations. The termination condition can be selected according to actual needs to obtain the desired optimized transmit predistortion coefficients and reduce the impact of interference in the receiving array.

[0081] Example 2

[0082] like Figure 2 As shown, this embodiment provides an open-loop device for removing interference of nonlinearities outside the transmission band on a receiving array, the device comprising at least:

[0083] A parameter processing unit, used for finding the transmission predistortion coefficient of each channel with the minimum interference of the transmitting array to the receiving array by using a genetic algorithm when the array is not in operation;

[0084] The signal control unit is used to directly convolve the optimized pre-distortion coefficient with the downlink transmission signal when the array is in working state, so as to apply it to each transmission channel to realize signal pre-distortion processing.

[0085] The parameter processing unit is connected between the receiving channel and the transmitting channel, and the signal control unit is arranged between the DBF module and the DAC+ modulation module of the transmitting channel. Specifically, one end of the parameter processing unit is connected to the receiving DBF module, and the other end is connected to the signal control unit. In the non-working state, a genetic algorithm is used to find the transmission predistortion coefficients of each channel with the minimum interference of the transmitting array on the receiving array, and then send them to the signal control unit to perform predistortion processing on the transmitting channel when the array is in the working state, thereby reducing nonlinear interference in the transmitting array.

[0086] In this device, the transmission predistortion coefficient of each channel is obtained through the parameter processing unit, and the digital predistortion method is applied to the array. There is no need to suppress the out-of-band nonlinearity of each transmitting power amplifier. Instead, the transmitting antenna array normally sends a data beam, and the receiving channel receives the transmitted out-of-band leakage signal. The transmission predistortion coefficient of each channel with the minimum interference of the transmitting array on the receiving array is applied to each transmitting channel, thereby minimizing the transmission nonlinearity of the receiving channel. The predistortion coefficient is only for the out-of-band spectrum of the transmitting signal, and the coherence of the normal in-band signal is not destroyed, which does not affect the transmit and receive beams of the antenna array. In addition, the predistortion coefficient is not only for a single power amplifier, but can be optimized based on the entire receiving array, thereby effectively solving the out-of-band interference of the transmission on the reception in the antenna array, greatly reducing the demand for the out-of-band suppression of the filter by the transmitting channel, reducing the power consumption, volume and weight of the array, and improving the efficiency of the array.

[0087] Preferably, the parameter processing unit includes:

[0088] Initialization component: Initialize the population and generate several different transmit predistortion coefficient combinations d = [d1, d2, ..., d m ], where m is the number of transmit channels and d is the transmit predistortion coefficient;

[0089] Fitness calculation component: establishes a fitness function and calculates the fitness corresponding to each transmit predistortion coefficient combination;

[0090] Optimization Iteration Component: Generates a new transmit predistortion coefficient combination through iterative optimization through selection, crossover, and mutation operations. Then, it re-enters the fitness calculation component for fitness calculation. When the termination condition is met, it outputs the optimal transmit predistortion coefficient combination.

[0091] Among them, in the fitness calculation component, the fitness function formula is:

[0092]

[0093] Among them, P signal is the power of the signal; P noise is the power of noise; IMD is the minimum intermodulation distortion; P interference is to minimize the interference power; MSE is to minimize the mean square error; ω1, ω2, ω3, and ω4 are the weight coefficients of the above performance indicators respectively.

[0094] Example 3

[0095] On the basis of embodiments 1 and 2, an interference estimation module is added in this embodiment to offset the influence of downlink intermodulation on uplink reception, thereby further improving the anti-interference performance of the antenna array by combining the two architectures. Specifically, Figure 3As shown, one end of the interference estimation module is connected to the channel control unit of the transmitting channel, and is used to collect data from the transmitting channel that needs to perform interference cancellation; one end is connected between the DAC+ down-modulation module and the DBF module of the receiving channel, and is used to collect data from all receiving channels.

[0096] Then, based on the collected data, optimization processing is performed in the interference estimation module to obtain the cancellation coefficients of the downlink channel and uplink interference. The cancellation coefficients are used to construct an anti-interference model. In the anti-interference model, the cancellation coefficients are combined with the downlink signal to obtain a cancellation signal that is equal to the uplink interference in magnitude and opposite in phase. The generated cancellation signal is added or subtracted from the uplink signal to affect the phase, thereby eliminating the interference falling within the uplink band.

[0097] Specifically, the optimization process includes:

[0098] Step S41: constructing a fitting intermodulation signal based on the collected data using the downlink signal and the power amplifier intermodulation model, and performing uplink shaping filtering on the fitting intermodulation signal to obtain a fitting intermodulation signal containing only the uplink passband portion;

[0099] Preferably, the power amplifier intermodulation model in this embodiment uses a Volterra series model, and the power amplifier intermodulation model actually used is obtained for the component of the signal falling in the uplink passband, and its formula is:

[0100] Y′(k)=y(k)G(t)

[0101] Where y(k) is the output of the Volterra series model; G(t) is the output of the uplink shaping filter; and Y′(k) represents the output of the power amplifier intermodulation model, which is a fitted intermodulation signal containing only the uplink passband portion.

[0102] And because the power amplifier model is centered on the downlink signal, and the uplink shaping filter is centered on the uplink signal, it is also necessary to perform band-limiting on the power amplifier intermodulation model to align the frequencies of the power amplifier intermodulation model and the uplink shaping filter.

[0103] Step S42: Cross-correlation is performed on the fitted intermodulation signal filtered in step S41 and the actual uplink interference signal, and the cross-correlation result is compared with a preset threshold. If the cross-correlation result is greater than the threshold, it indicates that the intermodulation of the downlink channel has an impact on the uplink reception. Then, the set of all downlink channels that may interfere with the uplink reception is recorded as B.

[0104] Wherein, the cross-correlation calculation formula is:

[0105]

[0106] Among them, x *(kt) represents the uplink signal; Y′(k) represents the output of the power amplifier intermodulation model; r yx (t) represents the cross-correlation result.

[0107] Step S43: Use an optimization algorithm to estimate the cancellation coefficient of the downlink channel set B and the uplink interference, and use the cancellation coefficient to construct an anti-interference model. In the anti-interference model, the cancellation coefficient is combined with the downlink signal to obtain a cancellation signal that is equal to the uplink interference in magnitude and opposite in phase. The generated cancellation signal is added or subtracted from the uplink signal to affect the phase, thereby eliminating the interference falling within the uplink band.

[0108] Preferably, the using of an optimization algorithm to estimate the cancellation coefficient of the downlink channel set B and the uplink interference specifically includes:

[0109] S51: Subtract the output Y′(k) of the power amplifier intermodulation model from the actual uplink interference signal x(k) to obtain the error e(k) of the anti-interference model, and obtain several groups of cancellation coefficients based on the error;

[0110] S52: Select n cancellation coefficient groups with the best cancellation effects from a number of cancellation coefficient groups to form a new initial set, recorded as initial set C = [H1H2H3...H n ];

[0111] S53: Optimizing the cancellation coefficients in the initial set C until the obtained cancellation coefficients can cancel the interference signal below the noise floor.

[0112] Further preferably, in step S51, obtaining a plurality of offset coefficient groups according to the error includes:

[0113] S511: Use optimization algorithm to find a suitable offset coefficient h so that

[0114]

[0115] S512: Since the overdetermined equation has no unique solution, the least squares method is used to find an approximate solution for the optimal offset coefficient, and several offset coefficient groups H are obtained, whose formula is:

[0116] H=(U*U H ) -1 U H X

[0117] Among them, U represents the downlink transmission signal; X represents the uplink reception signal.

[0118] Further preferably, in step S53, the optimization process includes:

[0119] S531: Select the two best cancellation coefficient sets from the initial set C, then randomly swap a coefficients between them and calculate the cancellation error between the two new cancellation coefficient sets. If the cancellation error of the new cancellation coefficient set is at least 1 dB better than the cancellation error of the worst-performing set in the initial set C, then add the newly generated cancellation coefficient set to the initial set C to form a new set C1.

[0120] S532: Randomly select two sets of cancellation coefficients from set C1 and randomly swap a coefficients therein. Calculate the cancellation error of the two new sets of cancellation coefficients. If the cancellation error of the new cancellation coefficients is at least 1 dB better than the cancellation error of the worst-performing set in the initial set C, then add the newly generated cancellation coefficient set to set C1 to form a new set C2.

[0121] S533: Randomly select a set of offset coefficients from set C2, and then randomly select The number of offset coefficients is m, represents rounding up; then the amplitude A of the selected cancellation coefficients is changed to a random value in the interval [0.95A, 1.05A] until the cancellation error of the new cancellation coefficient is better than the cancellation error of the worst-performing group in the initial set C by more than 1 dB. The newly generated cancellation coefficient group is then added to the initial set C to form a new set C3;

[0122] S534: Determine the number of cancellation coefficient groups in set C3. If the number is greater than 4m, delete the worst performing groups of cancellation coefficients until the number of elements in the set is less than or equal to 4m.

[0123] S535: Repeat steps S531-S534 until the obtained cancellation coefficient can cancel the interference signal below the noise floor.

[0124] Therefore, in this embodiment, while embodiments 1 and 2 effectively solve the out-of-band interference of transmission to reception in the antenna array by adopting digital predistortion, digital interference cancellation is further combined to eliminate interference falling within the uplink band. The two architectures are combined to further improve the anti-interference performance of the antenna array, thereby greatly reducing the demand for out-of-band suppression of the filter in the transmission channel, reducing the power consumption, volume, and weight of the array, and improving the efficiency of the array.

[0125] Obviously, the above embodiments of the present invention are merely examples for the purpose of clearly illustrating the technical solutions of the present invention, and are not intended to limit the specific implementation methods of the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the claims of the present invention shall be included within the scope of protection of the claims of the present invention.

Claims

1. An open-loop method for removing out-of-band nonlinear interference to a receiving array, characterized in that: The method comprises: S1: When the array is not in operation, a search algorithm is used to find the transmit predistortion coefficients of each channel that minimize the interference of the transmit array on the receive array. S2: When the array is in operation, the predistortion coefficients obtained in step S1 are directly convolved with the downlink transmission signal, thereby being applied to each transmission channel to destroy the coherence of the transmission nonlinear signal; In step S2, the predistortion coefficient obtained in step S1 is directly convolved with the downlink transmission signal, and is applied to each transmission channel to destroy the coherence of the transmission nonlinear signal. The processing formula is: in, It is the downlink transmission signal after pre-distortion processing; is the transmit predistortion coefficient of the i-th transmit channel; is the i-th downlink transmission signal; In step S1, the search algorithm is a genetic algorithm; In step S1, a genetic algorithm is used to find the transmission predistortion coefficients of each channel that minimize the interference of the transmitting array on the receiving array, including: S31: Initialize the population and generate several different transmit predistortion coefficient combinations d=[d1,d2,…,d m ], where m is the number of transmit channels and d is the transmit predistortion coefficient; S32: Establishing a fitness function and calculating the fitness corresponding to each transmit predistortion coefficient combination; S33: Generate a new transmit predistortion coefficient combination through iterative optimization through selection, crossover, and mutation operations, then repeat the operation of step S32, recalculate the fitness corresponding to each new transmit predistortion coefficient combination, and output the optimal transmit predistortion coefficient combination when the termination condition is met; In step S32, the fitness function is composed of several performance indicators, and can be optimized by adjusting the value of the weight coefficient according to different application requirements, wherein the value of the weight coefficient is greater than 0; The fitness function includes at least the performance indicators of maximum signal-to-noise ratio, minimum intermodulation distortion, minimum interference power and minimum mean square error. The obtained fitness function formula is: in, is the power of the signal; is the power of noise; IMD is the minimum intermodulation distortion; is to minimize the interference power; MSE is to minimize the mean square error; are the weight coefficients of the above performance indicators respectively.

2. The open-loop method for removing out-of-band nonlinear interference to a receiving array according to claim 1, characterized in that: Step S33 includes: selecting a transmit pre-distortion coefficient combination with a higher fitness as a parent individual, then performing a crossover operation to generate a new transmit pre-distortion coefficient combination, then randomly mutating some of the transmit pre-distortion coefficients in the new transmit pre-distortion coefficient combination to introduce new genetic information, and finally re-iterating the operations of steps S32-S33 until the optimal transmit pre-distortion coefficient combination is obtained.

3. The open-loop method for removing out-of-band nonlinear interference to a receiving array according to any one of claims 1 to 2, characterized in that: The termination condition is that the fitness reaches a preset threshold or reaches a maximum number of iterations.

4. An open-loop device for removing out-of-band nonlinear interference to a receiving array, characterized in that: The device at least comprises: A parameter processing unit, used for finding the transmission predistortion coefficient of each channel with the minimum interference of the transmitting array to the receiving array by using a genetic algorithm when the array is not in operation; A signal control unit is used to directly convolve the optimized pre-distortion coefficient with the downlink transmission signal when the array is in operation, thereby applying it to each transmission channel to implement signal pre-distortion processing; The parameter processing unit includes: Initialization component: Initialize the population and generate several different transmit predistortion coefficient combinations d=[d1,d2,…,d m ], where m is the number of transmit channels and d is the transmit predistortion coefficient; Fitness calculation component: establishes a fitness function and calculates the fitness corresponding to each transmit predistortion coefficient combination; Optimization Iteration Component: Generates a new transmit predistortion coefficient combination through iterative optimization through selection, crossover, and mutation operations. Then, it re-enters the fitness calculation component for fitness calculation. When the termination condition is met, it outputs the optimal transmit predistortion coefficient combination. Among them, in the fitness calculation component, the fitness function formula is: in, is the power of the signal; is the power of noise; IMD is the minimum intermodulation distortion; is to minimize the interference power; MSE is to minimize the mean square error; are the weight coefficients of the above performance indicators respectively.

Images