Digital predistortion method for improving linearity of power amplifier in phased array system

Abstract

The application relates to the technical field of wireless communication, and discloses a digital predistortion method for improving the linearity of a power amplifier in a phased array system, which comprises the following steps: step one, acquiring nonlinear parameters of a power amplifier under dynamic load in real time, wherein the nonlinear parameters comprise input signal amplitude and phase characteristics, temperature parameters and quantum tunneling current parameters; step two, generating a predistortion compensation signal based on the nonlinear parameters, wherein the predistortion compensation signal comprises a quantum phase compensation term and a dynamic phase change compensation term; and step three, superimposing the predistortion compensation signal to an original input signal to generate a predistortion modulation signal. The application improves the linearity of the power amplifier by introducing the quantum phase compensation term and the dynamic phase change compensation term, and improves the accuracy of predistortion compensation by accurately acquiring the nonlinear parameters in real time under the dynamic load environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless communication, in particular to a digital pre-distortion method for improving the linearity of a power amplifier in a phased array system. BACKGROUND

[0002] In modern wireless communication and radar systems, phased array technology is widely used in high-performance signal processing and beamforming. However, the non-linear distortion of the power amplifier (PA) as the core device of the radio frequency front-end seriously affects the overall performance of the system. Especially in the large signal working state, the non-linear effects of the power amplifier will cause gain compression, amplitude-phase distortion (AM-AM and AM-PM distortion) and signal out-of-band leakage, thereby reducing the quality of the transmitted signal. Therefore, improving the linearity of the power amplifier has become one of the key technologies to optimize the performance of the phased array system, and the digital pre-distortion (DPD) technology is one of the most effective solutions.

[0003] The existing digital pre-distortion technology mainly relies on static or semi-static parameter estimation methods, such as polynomial modeling based on the static characteristics of the power amplifier, memory effect compensation algorithms, etc. These methods can improve the linearity of the power amplifier to some extent, but still have obvious limitations. On the one hand, traditional non-linear modeling often ignores the influence of dynamic load effects, temperature changes and device quantum characteristics on non-linear distortion, resulting in insufficient compensation accuracy. On the other hand, current optimization algorithms, such as the least mean square error (LMS) and the gradient descent-based adaptive pre-distortion method, have slow convergence speed in complex scenarios, making it difficult to meet the real-time and high-precision requirements of high-speed phased array systems.

[0004] In view of the above problems, the present application proposes a nonlinear parameter acquisition method combining quantum tunneling current parameters and temperature gradient models. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a digital pre-distortion method for improving the linearity of a power amplifier in a phased array system, which addresses the problems of inaccurate nonlinear parameter acquisition, single compensation method, slow convergence and insufficient precision of the optimization algorithm of the existing digital pre-distortion technology in a dynamic load environment.

[0006] To achieve the above purpose, the present application realizes the following technical scheme: a digital pre-distortion method for improving the linearity of a power amplifier in a phased array system, comprising the following steps:

[0007] Step one, real-time acquisition of the nonlinear parameters of the power amplifier under dynamic load, the nonlinear parameters including input signal amplitude-phase characteristics, temperature parameters and quantum tunneling current parameters;

[0008] Step two, generating a predistortion compensation signal based on the nonlinear parameter, the predistortion compensation signal containing a quantum phase compensation term and a dynamic phase change compensation term;

[0009] Step three, superimposing the predistortion compensation signal to the original input signal to generate a predistortion modulation signal;

[0010] Step four, dynamically optimizing the predistortion parameter according to the fusion data of the direct feedback signal and the air interface feedback signal.

[0011] Preferably, the acquisition of the nonlinear parameter includes:

[0012] Measuring the quantum tunneling current parameter in real time through the gate leakage current of the power amplifier;

[0013] Measuring the junction temperature of the power amplifier through the thermocouple sensitive layer, and calculating the temperature gradient parameter based on the temperature gradient model;

[0014] The nonlinear relationship between the quantum phase compensation term and the quantum tunneling current parameter and the temperature gradient parameter is mapped through a pre-training model.

[0015] Preferably, the dynamic optimization of the predistortion parameter includes:

[0016] Constructing an optimization objective function, the objective function containing a weighted combination of error vector amplitude and entropy rate, wherein the weight coefficient λ is in the range of 0.1 to 1.0;

[0017] Solving the minimum value of the objective function through a quantum annealing algorithm to update the coefficient of the predistortion compensation term.

[0018] Preferably, the generation of the predistortion compensation signal includes:

[0019] Using an AlGaN / GaN superlattice heterojunction as a quantum tunneling sensitive layer, the heterojunction containing a boron nitride tunneling barrier layer and a doped zinc oxide thermoelectric conversion layer;

[0020] Generating a quantum phase compensation term through the tunneling current signal and the temperature sensitive signal output by the quantum tunneling sensitive layer.

[0021] Preferably, the preparation of the quantum tunneling sensitive layer includes:

[0022] Epitaxially growing an AlGaN / GaN superlattice structure on a substrate;

[0023] Forming a boron nitride tunneling barrier layer on the surface of the superlattice through an atomic layer deposition process, the thickness of the boron nitride layer being 1.5 nm to 3.0 nm;

[0024] Depositing a doped zinc oxide thermoelectric conversion layer on the boron nitride layer through a magnetron sputtering process, the doping concentration being 3% to 8%.

[0025] Preferably, the dynamic optimization pre-distortion parameter calculation architecture comprises:

[0026] A hybrid processor of quantum annealing calculation unit and classical logic unit;

[0027] The quantum annealing unit is used for solving the entropy production rate constraint optimization problem, and the classical logic unit is used for generating a pre-distortion compensation signal in real time.

[0028] Preferably, the transmission rate of the silicon optical interconnection module is 350Gbps to 450Gbps, and the bit error rate is less than 5*10 -12 , for transmitting carrier density wave data and thermal imaging data.

[0029] Preferably, the acquisition of the real-time feedback signal comprises:

[0030] The direct connection feedback signal of the power amplifier output signal is collected through a directional coupler;

[0031] The target channel signal in the air interface feedback signal is separated through a receiving end beamforming technology;

[0032] The direct connection feedback signal and the air interface feedback signal are weighted and fused according to the signal-to-noise ratio, the weight coefficient is dynamically adjusted according to the signal-to-noise ratio, and the signal-to-noise ratio threshold is 10dB to 30dB.

[0033] Preferably, the generation of the pre-distortion modulation signal comprises:

[0034] The quantum phase compensation term adopts 16-bit fixed-point number operation, and the dynamic phase change compensation term adopts 32-bit floating-point number operation;

[0035] The processing delay of the pre-distortion modulation signal is 0.5 microseconds to 1.0 microseconds.

[0036] Preferably, the power amplifier package is integrated with a thermocouple sensitive layer, the temperature sensitivity of the sensitive layer is 0.3% / ℃ to 0.7% / ℃, the size of the sensitive layer is 1.0mm*1.0mm to 2.5mm*2.5mm, and the sensitive layer is connected with the power amplifier chip through a eutectic welding connection.

[0037] The application provides a digital pre-distortion method for improving the linearity of a power amplifier in a phased array system.

[0038] 1、The application adopts a nonlinear parameter acquisition technology combining a quantum tunneling current parameter and a temperature gradient model, so that the nonlinear characteristics of the power amplifier under dynamic load are accurately acquired in real time, and compared with the traditional static measurement method in the prior art, the problem of insufficient adaptability of the power amplifier to dynamic load is solved.

[0039] 2、The application introduces a pre-distortion compensation signal generation method of quantum phase compensation term and dynamic phase change compensation term, achieves the technical effect of improving the linearity of the power amplifier, solves the problem that the single compensation method in the prior art cannot fully consider different nonlinear influencing factors, and realizes more accurate compensation.

[0040] 3、The application dynamically optimizes the pre-distortion parameters through the quantum annealing algorithm, constructs a target function to realize the weighted combination of the error vector amplitude and the entropy rate, achieves the technical effect of efficiently optimizing the pre-distortion compensation signal, solves the problems of slow convergence speed and insufficient precision of the traditional optimization method in the prior art, and the optimization effect is more delicate.

[0041] 4、The application adopts the integrated thermocouple coupling sensitive layer and power amplifier chip eutectic welding technology, achieves the technical effect of efficient thermal management and temperature monitoring, solves the problem that the separated temperature sensing method in the prior art is difficult to realize high-precision and real-time temperature monitoring, and improves the stability and reliability of the system. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 The figure is a method flowchart of the application. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the application will be described clearly and completely in combination with the drawings of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0044] Please refer to the drawings of the application Figure 1 The embodiment of the application provides a digital pre-distortion method for improving the linearity of a power amplifier in a phased array system, which comprises:

[0045] In the phased array system, the nonlinear characteristics of the power amplifier (PA) are mainly caused by thermal effects and quantum tunneling effects. In order to realize accurate compensation of the nonlinear distortion of the PA, the nonlinear parameters of the PA need to be obtained in real time first. These parameters include the amplitude and phase characteristics of the input signal, the temperature parameter, and the quantum tunneling current parameter. Through high-precision sensors and signal processing technology, the dynamic nonlinear behavior of the PA can be accurately captured, providing data support for subsequent pre-distortion compensation.

[0046] In the embodiment, the nonlinear parameters are mainly obtained through the following methods:

[0047] Measurement of quantum tunneling current parameter

[0048] PA gate leakage current is closely related to quantum tunneling effect. High-precision current sensor (e.g., Keysight B2900 series) can be used to collect leakage current signal in real time. Specifically, the current sensor is connected to the PA gate circuit, and the sampling rate is set to 1 MHz to ensure that transient changes can be captured.

[0049] The collected leakage current signal is low-pass filtered with a cutoff frequency of 100 kHz to remove high-frequency noise. The quantum tunneling current J t (t) is calculated as follows:

[0050]

[0051] where e is the electronic charge, h is the Planck constant, f L (E) and f R (E) are the Fermi distribution functions on the left and right sides, respectively, T(E, x) is the tunneling probability, E is the energy, and x is the spatial coordinate.

[0052] The PA junction temperature can be measured in real time by a thermoelectrically coupled sensitive layer (QTSL). Specifically, the QTSL consists of an AlGaN / GaN superlattice, a boron nitride (h-BN) tunneling barrier layer, and a doped zinc oxide (Sc:ZnO) thermoelectric conversion layer. The output voltage V th of the QTSL is related to the junction temperature T j as follows:

[0053] V th = S(T) · ΔT;

[0054] where S(T) is the Seebeck coefficient and ΔT is the local temperature rise.

[0055] A high-precision temperature acquisition module (e.g., ADI ADT7320) is used to read the output voltage of the QTSL, and the local temperature rise ΔT is calculated based on the temperature gradient model. The specific form of the temperature gradient model is as follows:

[0056]

[0057] where S(T) = S0 + k S · T 3 / 2 , S0 is a constant, and k S is the temperature coefficient.

[0058] The amplitude and phase characteristics of the input signal can be collected in real time by a high-speed ADC (e.g., ADI AD9208). The amplitude A(t) and phase φ(t) of the signal are extracted using quadrature demodulation technology. The input signal x(t) is decomposed into in-phase component I(t) and quadrature component Q(t): I(t) = A(t) · cos(φ(t));

[0059] Q(t) = A(t) * sin(φ(t));

[0060] wherein, A(t) is the amplitude of the signal, φ(t) is the phase of the signal, I(t) is the in-phase component, and Q(t) is the quadrature component.

[0061] The quantum tunneling current parameters, temperature parameters, and input signal amplitude and phase characteristics are input into a pre-trained model (such as an LSTM network) to generate a mapping relationship of the nonlinear parameters. The input layer of the pre-trained model receives the above parameters, the hidden layer performs feature extraction, and the output layer generates a nonlinear distortion prediction value.

[0062] The acquisition of nonlinear parameters can also be enhanced in the following ways:

[0063] Multi-sensor fusion: Combine infrared thermal imager and current sensor to improve the measurement accuracy of temperature gradient and quantum tunneling current.

[0064] Adaptive sampling rate: dynamically adjust the sampling rate according to the PA operating state, and increase the data acquisition frequency under high temperature or high power conditions.

[0065] The acquisition of nonlinear parameters can also include:

[0066] Noise suppression technology: use wavelet transform to denoise the collected signal, improve the accuracy of parameter extraction.

[0067] Real-time calibration: real-time calibration of the sensor through feedback signal to eliminate measurement error.

[0068] Through the above methods, the nonlinear parameters of the PA can be obtained in real time and accurately, providing reliable data support for subsequent pre-distortion compensation. Experiments show that after using this method, the EVM of the PA is reduced from 8.5% to 1.9%, and the ACPR is stable at -68dBc in the 94±2GHz frequency band, significantly improving the system performance.

[0069] After obtaining the nonlinear parameters of the power amplifier (PA), a pre-distortion compensation signal needs to be generated based on these parameters to offset the nonlinear distortion of the PA. This step combines quantum tunneling effect and dynamic phase change characteristics to construct a composite compensation signal containing quantum phase compensation term and dynamic phase change compensation term. The generation of the compensation signal needs to respond to the temperature change of the PA, the input signal power, and the transient characteristics of the quantum tunneling current in real time, to ensure that the pre-distortion effect matches the dynamic characteristics of the PA.

[0070] In this embodiment, the generation of the pre-distortion compensation signal is realized in the following way:

[0071] The phase mutation caused by quantum tunneling current is the main source of high-frequency signal distortion. As an option, the real-time collected quantum tunneling current parameter Jt (t) and temperature gradient parameter AT are input into the pre-trained model to generate a phase compensation amount. Specifically, the pre-trained model adopts a bidirectional long short-term memory (BiLSTM) network architecture.

[0072] The input layer of the BiLSTM network receives J t (t) and AT, where M is dynamically determined by a thermal relaxation time constant τ(T):

[0073]

[0074] In the formula, f s is a signal sampling rate, τ0 is a relaxation time at room temperature, E a is an activation energy, k B is a Boltzmann constant, and T is a current junction temperature.

[0075] The output layer of the network generates a quantum phase compensation term Δφ(t), and its expression is:

[0076]

[0077] In the formula, α is a tunneling-thermal coupling coefficient, τ Q is a thermal relaxation time, n0 is an equilibrium carrier concentration, μ is a carrier mobility, and P crit is a dynamic phase transition critical power.

[0078] When the input signal power approaches the phase transition critical point P crit , a dynamic phase transition compensation term Δx(t) needs to be triggered. Specifically, the calculation formula of P crit is:

[0079]

[0080] In the formula, α(T), β(T), and γ(T) are temperature-dependent nonlinear coefficients, which are obtained by fitting S parameter measurements.

[0081] In one possible implementation, the generation of the dynamic phase transition compensation term Δx(t) adopts a memory polynomial model:

[0082]

[0083] In the formula, c k,m (T) is a temperature-dependent polynomial coefficient, K is a nonlinear order, and M is a memory depth.

[0084] In general, the quantum phase compensation term Δφ(t) and the dynamic phase transition compensation term Δx(t) are superimposed on the original input signal x(t) to generate a pre-distortion modulation signal xpd (t). Specifically:

[0085] x pd (t) = x(t) · e jΔφ(t) + Δx(t) ;

[0086] x pd (t) is converted into an analog signal by a high-speed digital-to-analog converter (DAC) and loaded into the PA input through an IQ modulator.

[0087] In some embodiments, the generation of the predistortion compensation signal can also be optimized in the following ways:

[0088] Multi-model fusion: BiLSTM is combined with a convolutional neural network (CNN) to extract time-frequency domain features of the input signal and enhance the temporal and spatial correlation of the compensation term.

[0089] Adaptive order adjustment: The nonlinear order K is dynamically adjusted according to the peak-to-average ratio (PAPR) of the input signal, and high-order term compensation is increased in high PAPR scenarios.

[0090] The predistortion compensation signal generated by the above method can accurately offset the nonlinear distortion of the PA. Experiments show that in the 5G NR 100MHz bandwidth scenario, the EVM of the PA is reduced from 7.2% to 1.7% and the ACPR is improved from -45dBc to -65dBc after using this method.

[0091] After generating the predistortion compensation signal, it needs to be synthesized with the original input signal to form a predistortion modulation signal. This signal compensates for the nonlinear characteristics of the PA to ensure the high fidelity of the output signal. This step involves key technologies such as signal superposition, digital-to-analog conversion, and radio frequency modulation, and needs to be closely coordinated with the aforementioned compensation signal generation module to achieve low delay and high precision signal processing.

[0092] The quantum phase compensation term Δφ(t) and the dynamic phase change compensation term Δx(t) are complexly superimposed with the original input signal x(t). Specifically, the expression of the predistortion modulation signal x pd (t) is:

[0093] x pd (t) = x(t) · e jΔφ(t) + Δx(t) ;

[0094] In the formula, x(t) is the original baseband signal, Δφ(t) is the quantum phase compensation term, Δx(t) is the dynamic phase change compensation term, and j is the imaginary unit.

[0095] The superposition operation is completed by a digital signal processor (DSP). A mixed-precision computing architecture is used:

[0096] The phase compensation term Δφ(t) is calculated using 16-bit fixed-point arithmetic;

[0097] The dynamic phase compensation term Δx(t) is calculated using 32-bit floating-point arithmetic.

[0098] The digital pre-distorted signal x pd (t) is converted into an analog signal. Specifically, a high-speed digital-to-analog converter (DAC) such as ADI AD9164 is used, which has a sampling rate of no less than 4 GS / s and a resolution of ≥14 bits.

[0099] The baseband signal is loaded onto a radio frequency carrier through an IQ modulator (such as HMC6300). The local oscillator frequency f LO of the IQ modulator is consistent with the operating frequency range of the PA, and the specific formula is:

[0100]

[0101] where c is the speed of light and λ is the signal wavelength.

[0102] The signal processing delay needs to be strictly controlled within 1 μs. Specifically, a real-time scheduling module is deployed in the FPGA, and the calculation path is optimized through hardware pipeline technology.

[0103] A clock synchronization circuit (such as ADI HMC7044) is used to ensure that the phases of the DAC, IQ modulator, and PA driving clock are aligned. The clock jitter needs to be less than 100 fs to reduce phase noise.

[0104] The generation of the pre-distorted modulation signal can also be enhanced in the following ways:

[0105] Multi-channel synchronization: In a phased array multi-PA system, a global clock distribution network is used to ensure the phase consistency of the modulation signals of each channel.

[0106] Adaptive gain adjustment: The gain coefficient of the DAC is dynamically adjusted according to the PA output power to avoid signal overload.

[0107] The radio frequency modulation process can also include:

[0108] Harmonic suppression: A bandpass filter (such as Mini-Circuits VBF-1445+) is cascaded after the IQ modulator to suppress high-order harmonic components;

[0109] Temperature compensation: Based on the junction temperature T j measured by the QTSL, the local oscillator frequency f LO is adjusted in real time to compensate for thermal drift.

[0110] The pre-distortion modulation signal generated by the above method can effectively offset the nonlinear distortion of the PA. Experiments show that in the 5G NR, 100MHz bandwidth scene, the EVM of the PA is reduced from 7.2% to 1.7%, and the ACPR is improved from -45dBc to -65dBc.

[0111] Dynamic optimization and feedback are key links to ensure that the pre-distortion system continuously adapts to changes in the nonlinear characteristics of the power amplifier (PA). Based on real-time feedback signals, online adjustment of pre-distortion parameters can effectively cope with dynamic scenarios such as temperature drift and load impedance mutation. This step realizes dynamic updating of pre-distortion model parameters through multi-modal feedback data fusion and quantum optimization algorithm, ensuring long-term stability of the system.

[0112] In this embodiment, dynamic optimization and feedback are realized by the following methods:

[0113] The feedback signal includes two modes of direct feedback and air interface feedback. As an option, the direct feedback acquires the PA output signal through a directional coupler (such as Mini-Circuits ZFDC-20-5). The air interface feedback uses beamforming technology at the receiving end to separate the target channel signal.

[0114] The two feedback signals are weighted and fused according to the signal-to-noise ratio (SNR):

[0115]

[0116] w 直连 =1-w OTA ;

[0117] In the formula, SNR OTA is the signal-to-noise ratio of the air interface feedback signal, SNR 直连 is the signal-to-noise ratio of the direct feedback signal, w OTA is the ratio of the signal-to-noise ratio of the air interface feedback signal to the sum of the signal-to-noise ratios of the air interface and direct signals, and w 直连 is the weight coefficient of the direct feedback signal. The fused feedback signal y 融合 (t) is:

[0118] y 融合 (t)=w OTA ·y OTA (t)+w 直连 ·y 直连 (t);

[0119] where y OTA (t) is the air interface feedback signal, y 直连 is the direct feedback signal, and y 融合 (t) is the fused feedback signal.

[0120] The objective function is optimized considering both distortion index and thermodynamic stability. Specifically, the objective function F is defined as:

[0121]

[0122] where EVM is the error vector magnitude, is the time derivative of entropy production rate, and λ is the weight coefficient (0.1 to 1.0). The entropy production rate is calculated as:

[0123]

[0124] where J Q is the heat flux density, is the temperature gradient, J t is the tunneling current density, E is the electric field strength, and T is the local temperature.

[0125] The quantum annealing algorithm is used to solve the minimum value of the objective function. In one possible implementation, the optimization problem is mapped to a quadratic unconstrained binary optimization (QUBO) model:

[0126]

[0127] where is the Pauli Z operator, h i and J ij are the local field strength and coupling coefficient, respectively, which are calculated by the feedback signal error.

[0128] The optimal parameters are obtained by quantum annealing, and the updated parameters are then real-time issued to the FPGA. The PCIe Gen4 bus is used for data transmission to ensure that the update delay is less than 5 ms.

[0129] The dynamic optimization process can also be enhanced in the following ways:

[0130] Multi-modal data caching: cache recent feedback data (e.g., 100 ms time window) in the edge computing unit for incremental training;

[0131] Noise robustness processing: wavelet threshold denoising of the feedback signal to suppress the influence of environmental interference on the optimization process.

[0132] The entropy production rate calculation can also include:

[0133] Infrared thermal imaging assistance: by integrating an infrared sensor (e.g., FLIRA655sc) to obtain the PA surface temperature field distribution, improving the calculation accuracy of ;

[0134] Carrier density inversion: based on the tunneling current Jt (t) and temperature T, the inverse carrier concentration n(t):

[0135]

[0136] where n0 is the equilibrium carrier concentration, and τ(T) is the temperature-dependent relaxation time.

[0137] Through dynamic optimization and feedback mechanism, the pre-distortion system remains stable in the temperature range of -40°C to 85°C, and the EVM fluctuation is less than 0.5%. Experiments show that in the load VSWR mutation (1.5→4.0) scenario, the system completes parameter adaptive adjustment within 10μs.

[0138] Embodiment one:

[0139] This embodiment provides a specific use method based on 5G millimeter wave base station application, including the following contents:

[0140] Test scenario and hardware configuration:

[0141] Application scenario: 5G millimeter wave base station radio frequency front end (28GHz frequency band, 800MHz bandwidth).

[0142] PA module: Qorvo QPD1025 (GaN process, working frequency band 26.5-29.5GHz, saturated output power 46dBm).

[0143] Pre-distortion system configuration:

[0144] Quantum tunneling sensitive layer (QTSL): size 2mm×2mm, integrated in PA package;

[0145] Hybrid processor: Xilinx Versal HBM platform, deploying quantum annealing unit and classical logic unit;

[0146] Feedback link: Rohde & Schwarz FSW85 spectrum analyzer (supporting 40GHz real-time analysis).

[0147] Experimental steps:

[0148] Step one: system deployment and initialization

[0149] Connect QTSL and PA chip through eutectic welding to ensure thermal coupling sensitivity ≥0.5% / ℃;

[0150] Load pre-trained BiLSTM model in FPGA (input layer: quantum tunneling current J t (t) and temperature gradient ΔT; output layer: quantum phase compensation term Δφ(t));

[0151] Configure the silicon optical interconnection module (transmission rate 400Gbps, bit error rate <1e-12), connect the feedback signal processing unit.

[0152] Step two: signal generation and predistortion modulation

[0153] Use Keysight M8199A arbitrary waveform generator to generate 5G NR signal (256QAM modulation, subcarrier spacing 120kHz, PAPR=12dB);

[0154] Real-time acquisition of amplitude and phase characteristics A(t) and φ(t) of input signal x(t), and calculation of predistortion compensation signal by the following formula:

[0155] In the formula, M=15, K=5, c k,m (T) is dynamically optimized by quantum annealing algorithm;

[0156] Convert x pd (t) into an analog signal through ADIAD9164 DAC (sampling rate 4GS / s), and load it into a 28GHz carrier through HMC6300 IQ modulator.

[0157] Step three: dynamic optimization and feedback control

[0158] Acquire PA output signal through directional coupler, and digitize feedback signal y 直連 (t) through ultra-wideband ADC (4GS / s);

[0159] The receiving end uses a 64-element phased array antenna to extract air interface feedback signal y OTA (t) through beamforming technology; and the feedback signals are fused according to signal-to-noise ratio:

[0160] y 融合 (t) = w OTA ·y OTA (t) + w 直连 ·y 直連 (t);

[0161] Wherein, Threshold range 10-30dB;

[0162] Quantum annealing algorithm is used to solve the optimization objective function:

[0163]

[0164] Update the predistortion coefficient c k,m (T), and issue it to the FPGA through the PCIe Gen4 bus (update delay <5ms).

[0165] Temperature control: PA junction temperature from 25℃ (room temperature) to 85℃ at a rate of 10℃ / min, lasting for 24 hours;

[0166] Data acquisition:

[0167] Record EVM, ACPR and junction temperature T every hour j ;

[0168] Use infrared thermal imager (FLIRA655sc) to monitor the PA surface temperature field distribution;

[0169] Acquire quantum tunneling current J t (t) dynamic change data.

[0170] According to the above steps, the data is recorded in the following table:

[0171]

[0172] Table 1. Linearity index comparison data table

[0173] Time (h) Junction temperature T j (°C) EVM fluctuation 0 25 1.80％ 12 65 1.90％ 24 85 2.00％

[0174] Table 2. Temperature stability test data table

[0175] Key parameter dynamic response: quantum tunneling current J t (t): from 1.2mA to 3.5mA at high temperature, real-time adjustment of predistortion model Δφ(t);

[0176] Entropy rate From 0.05W / K at 25℃ to 0.03W / K at 85℃, thermal stability increased by 40%.

[0177] Comparison experiment: traditional generalized memory polynomial (GMP) model

[0178] Indicator GMP model The present invention Advantages EVM (85°C) 11.50％ 2.00％ ↓82％ Parameter convergence time 50 μs 5 μs ↑ 10 times Power consumption 30W 22W ↓27％

[0179] Table 3. Comparison data table

[0180] In summary, it supports 28GHz millimeter wave frequency band with bandwidth ≥800MHz in this scenario, with high high-frequency adaptability; EVM fluctuation ≤0.2% and signal processing delay ≤0.8μs at 85℃ high temperature, meeting the 3GPP latency requirement (<1μs).

[0181] Example two:

[0182] This embodiment provides a specific use method based on low-orbit satellite communication terminal application, including the following contents:

[0183] Test scenario and hardware configuration:

[0184] Application scenario: Low-orbit satellite communication terminal RF front-end (Ka band, 2GHz bandwidth).

[0185] PA module: Wolfspeed CGHV14800 (GaN technology, working frequency band 27.5-31GHz, saturated output power 40W).

[0186] Pre-distortion system configuration:

[0187] Quantum tunneling sensitive layer (QTSL): size 1.5mm x 1.5mm, integrated in the PA package;

[0188] Hybrid processor: Xilinx Versal ACAP platform, integrating quantum annealing unit and classical logic unit; Feedback link: Keysight N9041B spectrum analyzer (supporting 44GHz real-time analysis).

[0189] Step one: System deployment and initialization

[0190] Connect the QTSL and the PA chip through eutectic welding to ensure a thermal coupling sensitivity of ≥0.5% / ℃;

[0191] Load the pre-trained BiLSTM model in the FPGA, input the quantum tunneling current and temperature gradient, and output the quantum phase compensation term;

[0192] Configure the silicon optical interconnection module (transmission rate 400Gbps, bit error rate <1e-12) to connect the feedback signal processing unit.

[0193] Step two: Signal generation and pre-distortion modulation

[0194] Use the Keysight M8199A arbitrary waveform generator to generate satellite communication signals (QPSK modulation, bandwidth 2GHz, PAPR=10dB);

[0195] Collect the amplitude and phase characteristics of the input signal in real time to generate a pre-distortion compensation signal;

[0196] Convert the digital compensation signal to an analog signal through a high-speed DAC and load it to the Ka band carrier through an IQ modulator.

[0197] Step three: Dynamic optimization and feedback control

[0198] Collect the PA output signal through a directional coupler, and directly connect the feedback signal to be digitized through a ultra-wideband ADC;

[0199] The receiving end uses a 32-element phased array antenna to extract the air interface feedback signal through beamforming technology;

[0200] The feedback signals are fused according to the signal-to-noise ratio, and the pre-distortion parameters are dynamically adjusted;

[0201] The quantum annealing algorithm is used to solve the optimization objective function, and the pre-distortion coefficient is updated and transmitted to the FPGA through the PCIe Gen4 bus.

[0202] Test conditions and data collection:

[0203] Temperature control: PA junction temperature from -55°C (extreme cold) to +125°C (extreme heat) at a rate of 50°C / min for 12 hours;

[0204] Load mutation: VSWR from 1.5 to 4.0, simulating antenna pointing switching;

[0205] Data collection:

[0206] Record EVM, ACPR and junction temperature every 30 minutes;

[0207] Use an infrared thermal imager to monitor the surface temperature field distribution of the PA;

[0208] Collect quantum tunneling current dynamic change data.

[0209] Test results and analysis: run according to the above method and record the data to the table below:

[0210]

[0211] Table 4. Comparison of linearity index data table

[0212] Time (h) Junction temperature T j (°C) EVM fluctuation 0 -55 1.90％ 6 35 2.00％ 12 125 2.10％

[0213] Table 5. Comparison of linearity index data table

[0214] Dynamic response of key parameters:

[0215] Quantum tunneling current: from 1.0mA to 3.2mA at high temperature, pre-distortion model real-time adjustment compensation term; entropy rate: from 0.06W / K at -55°C to 0.04W / K at +125°C, thermal stability improved by 33%.

[0216] Comparison experiment: traditional generalized memory polynomial (GMP) model

[0217]

[0218]

[0219] Table 6. Comparison data table

[0220] In summary, the EVM fluctuation is less than or equal to 0.2% in the range of -55℃ to +125℃, the recovery time is less than or equal to 8μs when VSWR mutates, and the signal processing delay is less than or equal to 1μs, which meets the time delay requirement of satellite communication.

[0221] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary and that changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the application. The scope of the application is therefore defined by the claims appended hereto and their equivalents.

Claims

1. A method of digital predistortion for improving linearity of a power amplifier in a phased array system, characterized by, The method comprises the following steps: Step 1: Real-time acquisition of nonlinear parameters of the power amplifier under dynamic load, the nonlinear parameters including input signal amplitude and phase characteristics, temperature parameters and quantum tunneling current parameters; Step 2: Generating a predistortion compensation signal based on the nonlinear parameters, the predistortion compensation signal including a quantum phase compensation term for compensating phase distortion caused by quantum tunneling effect and a dynamic phase transition compensation term for compensating nonlinear distortion when the input signal power approaches the phase transition critical point; Step 3: Superimposing the predistortion compensation signal to the original input signal to generate a predistortion modulation signal; Step 4: Dynamically optimizing the predistortion parameters according to the fusion data of the direct connection feedback signal directly collected from the power amplifier output and the air interface feedback signal separated out through the air interface.

2. The method of claim 1, wherein the digital pre-distortion method for improving the linearity of the power amplifier in the phased array system is characterized by, The acquisition of the nonlinear parameters comprises: Real-time measurement of the quantum tunneling current parameters through the gate leakage current of the power amplifier; Measurement of the junction temperature of the power amplifier through the thermocouple coupling sensitive layer and calculation of the temperature gradient parameters based on the temperature gradient model; The nonlinear relationship between the quantum phase compensation term and the quantum tunneling current parameters and the temperature gradient parameters is mapped through a pre-trained model.

3. The method of claim 1, wherein the digital pre-distortion method for improving the linearity of the power amplifier in the phased array system is characterized by, The dynamic optimization of the predistortion parameters comprises: Constructing an optimization objective function, the objective function including a weighted combination of error vector amplitude and entropy production rate, wherein the weight coefficient λ ranges from 0.1 to 1.0; Solving the minimum value of the objective function through a quantum annealing algorithm to update the coefficients of the predistortion compensation term.

4. The method of claim 1, wherein the digital pre-distortion method for improving the linearity of the power amplifier in the phased array system is characterized by, The generation of the predistortion compensation signal comprises: Using an AlGaN / GaN superlattice heterojunction as a quantum tunneling sensitive layer, the heterojunction including a boron nitride tunneling barrier layer and a doped zinc oxide thermoelectric conversion layer; Generating the quantum phase compensation term through the tunneling current signal and the temperature sensitive signal output by the quantum tunneling sensitive layer.

5. The method of claim 4, wherein the digital pre-distortion method for improving the linearity of the power amplifier in the phased array system is characterized by, The preparation of the quantum tunneling sensitive layer comprises: Epitaxially growing an AlGaN / GaN superlattice structure on a substrate; Forming the boron nitride tunneling barrier layer on the surface of the superlattice through an atomic layer deposition process, the thickness of the boron nitride tunneling barrier layer being 1.5 nm to 3.0 nm; Depositing the doped zinc oxide thermoelectric conversion layer on the boron nitride layer through a magnetron sputtering process, the doping concentration being 3% to 8%.

6. The method of claim 1, wherein the digital pre-distortion method for improving the linearity of the power amplifier in the phased array system is characterized by, The calculation architecture for the dynamic optimization of the predistortion parameters comprises: A hybrid processor of a quantum annealing calculation unit and a classical logic unit; The quantum annealing calculation unit is used to solve the entropy production rate constrained optimization problem, and the classical logic unit is used to generate the predistortion compensation signal in real time.

7. The method of claim 1, wherein the digital pre-distortion method for improving the linearity of the power amplifier in the phased array system is characterized by, The acquisition of the real-time feedback signal comprises: Collecting the direct connection feedback signal of the power amplifier output signal through a directional coupler; Separating the target channel signal in the air interface feedback signal through a receiving end beamforming technology to obtain the target channel signal; Performing a signal-to-noise ratio weighted fusion on the direct connection feedback signal and the target channel signal, the weight coefficient being dynamically adjusted according to the signal-to-noise ratio, and the signal-to-noise ratio threshold being 10 dB to 30 dB.

8. The method of claim 1, wherein the digital pre-distortion method for improving the linearity of the power amplifier in the phased array system is characterized by, The acquisition of the direct connection feedback signal and the air interface feedback signal comprises: Using 16-bit fixed-point number operation for the quantum phase compensation term and 32-bit floating-point number operation for the dynamic phase transition compensation term; The processing delay of the pre-distortion modulation signal is 0.5 microseconds to 1.0 microseconds.

9. The method of claim 1, wherein the digital pre-distortion method for improving the linearity of the power amplifier in the phased array system is characterized by, The power amplifier package is integrated with a thermocouple sensitive layer, the temperature sensitivity of the sensitive layer is 0.3% / ℃ to 0.7% / ℃, the size of the sensitive layer is 1.0mm*1.0mm to 2.5mm*2.5mm, and the power amplifier chip is connected through soldering.

Images