A power optimization method for high-power pulse amplification based on digital preprocessing

Abstract

This invention relates to the field of digital signal processing and power amplification technology, and discloses a power optimization method for high-power pulse amplification based on digital preprocessing. The method involves sampling and normalizing the pulse voltage into a discrete pulse signal with unit energy. Under a unified energy reference, a frequency domain transformation is performed to extract the amplitude and phase of each frequency point, forming a spectral sequence. A frequency response model is established based on the power amplifier's cutoff frequency to obtain the amplitude gain coefficient. A pre-compensation function, including an amplitude pre-compensation factor and a phase compensation amount varying with frequency, is constructed and applied to the spectral sequence, undergoing inverse transformation to form a pre-compensated time-domain signal. This pre-compensated time-domain signal is then input into the power amplifier, reconstructing the output spectrum and undergoing inverse transformation to obtain the output time-domain signal. The peak power and spectral energy fidelity are calculated. When the fidelity meets a threshold, power normalization is performed, and the signal is then connected to the transmission system. This solves the problems of difficulty in unifying compensation and spectral distortion, achieving the target power output while ensuring spectral fidelity.

Description

Technical Field

[0001] This invention relates to the field of digital signal processing and power amplification technology, specifically to a power optimization method for high-power pulse amplification based on digital preprocessing. Background Technology

[0002] In applications such as radar, satellite communication, and pulsed power device driving, high-power pulse signals typically need to be amplified by a power amplifier before being sent to the load. Because high-power amplifiers generally exhibit frequency-dependent gain compression and phase nonlinearity under wideband operating conditions, the amplitude of the high-frequency portion of the pulse signal's spectrum is easily reduced and the phase is distorted after passing through the amplifier. This results in a broadened time-domain waveform, a slower rise time, and a decrease in peak voltage, thereby reducing peak power and energy density, affecting the system's detection range, measurement accuracy, or energy transfer efficiency.

[0003] In existing technologies, methods such as pre-distortion, feedback linearization, and envelope tracking are commonly used to suppress nonlinear distortion in power amplifiers. For example, digital pre-distortion techniques are mostly based on behavioral models, fitting amplitude-amplitude and amplitude-phase characteristics at baseband or intermediate frequency, and adjusting the amplitude and phase of the input signal through lookup tables, polynomials, or memory polynomials to make the amplifier output as close to linear as possible. However, these methods are mostly designed around communication modulation signals, assuming that the signal has stationary statistical characteristics, and focusing more on the average error vector amplitude and adjacent channel leakage ratio. They are not sensitive to single or few repetitions of high-peak pulse signals, and have limited ability to maintain the instantaneous peak amplitude of the pulse and control the shape of the pulse envelope. In addition, existing pre-distortion schemes mostly construct the input-output mapping relationship directly in the time domain or envelope domain. Their pre-distortion function is obtained by fitting a large amount of test data, relying on empirical parameters and online adaptive adjustment, and lacking direct and calculable control over the frequency domain energy distribution and phase distribution. In high-power broadband pulse scenarios, the gain at the high-frequency end of the amplifier typically drops rapidly. Existing technologies often ensure peak output by increasing amplifier margin, raising supply voltage, or using multi-stage amplification. This increases system size, losses, and cost, and still struggles to avoid excessive compression of high-frequency components relative to low-frequency components, resulting in a sacrifice of power efficiency and bandwidth efficiency to guarantee peak output power. Current solutions generally lack a systematic approach that starts from the digital domain and reverse-engineers the input spectrum structure around the "target output spectrum." They cannot achieve synchronous constraints on the magnitude and phase of the output spectrum through calculable and verifiable preprocessing under given amplifier frequency response characteristics, thus making it difficult to significantly improve pulse peak power without changing the hardware structure.

[0004] Therefore, this paper aims to propose a power optimization method for high-power pulse amplification based on digital preprocessing. By performing energy normalization, frequency domain analysis, amplifier characteristic modeling and pre-compensation on the input analog pulse signal, and then time domain reconstruction and power normalization, the peak power output of the pulse signal amplifier is systematically improved while ensuring the fidelity of the spectral structure. This overcomes the problems of non-uniform gain and phase distortion in traditional amplifiers at different frequency components. Summary of the Invention

[0005] This invention provides a power optimization method for high-power pulse amplification based on digital preprocessing, which helps to solve the problems mentioned in the background art.

[0006] This invention provides the following technical solution: a power optimization method for high-power pulse amplification based on digital preprocessing, comprising:

[0007] The analog pulse voltage is sampled to generate a discrete voltage sequence covering a predetermined time. The energy index is calculated and normalized to obtain a discrete pulse signal with unit energy.

[0008] Discrete frequency domain transformation is performed on a unit energy discrete pulse signal to establish a set of discrete frequency points that match the sampling parameters. Complex spectral values ​​are obtained according to the frequency points, and the amplitude and phase are calculated to form a spectral amplitude sequence and a spectral phase sequence, which are then combined into a complex spectral value sequence.

[0009] A frequency response model of the power amplifier is established based on the cutoff frequency parameter of the power amplifier. The amplitude gain coefficient is obtained at each frequency point, and the amplitude distribution of the output spectrum after frequency response compression is calculated.

[0010] Based on the power amplifier frequency response model, a complex spectrum pre-compensation function is constructed. An amplitude pre-compensation factor and a frequency-varying advance phase compensation amount are set at each frequency point to form a complex pre-compensation function that simultaneously contains amplitude and phase compensation.

[0011] By applying a complex spectrum pre-compensation function to a complex spectrum value sequence, a pre-compensated spectrum value sequence is obtained. Then, a discrete inverse transform is performed on the sequence to reconstruct the pre-compensated time-domain discrete pulse signal.

[0012] The pre-compensated time-domain discrete pulse signal is input into the power amplifier, the output spectrum is reconstructed based on the frequency response model, the output amplitude and phase are calculated according to the frequency points, different frequency bands are segmented and represented, and combined into a complex output spectrum value sequence.

[0013] Perform a discrete inverse transform on the complex output spectrum value sequence to obtain the time-domain discrete signal of the power amplifier output, calculate the peak power index and construct the spectrum energy fidelity index to evaluate the peak power improvement effect;

[0014] The average power index is calculated based on the discrete time-domain signal output by the power amplifier. When the spectral energy fidelity index meets the preset constraints, the power is normalized to obtain the normalized output signal, and the normalized output signal is connected to the input port of the transmission system.

[0015] Optionally, the step of sampling the analog pulse voltage, generating a discrete voltage sequence covering a predetermined time, calculating and normalizing the energy index to obtain a discrete pulse signal with unit energy specifically includes:

[0016] Set the analog pulse voltage output by the signal source as a real-value continuous-time signal, and set the total sampling time length of the real-value continuous-time signal and the fixed time interval between adjacent samples;

[0017] Based on the total sampling time and the sampling time interval, calculate the total number of discrete sample points that can be obtained during the entire sampling time, and establish a time-domain discrete sample index for each sample point in chronological order.

[0018] According to the sampling time interval, the corresponding analog pulse voltage values ​​are collected at each continuous time moment to construct a discrete voltage sequence composed of the original discrete voltage sample values ​​at each sampling moment;

[0019] The energy index of the discrete voltage sequence is obtained by squaring and summing all the original discrete voltage sample values ​​in the discrete voltage sequence.

[0020] When the energy index is positive, the ratio of each original discrete voltage sample value to the square root of the energy index is used as the normalized discrete voltage sample value to construct a unit energy discrete pulse signal; when the energy index is zero, all normalized discrete voltage sample values ​​are set to zero.

[0021] Optionally, the step of performing discrete frequency domain transformation on the unit energy discrete pulse signal, establishing a set of discrete frequency points matching the sampling parameters, obtaining complex spectrum values ​​according to the frequency points, calculating the amplitude and phase, forming a spectrum amplitude sequence and a spectrum phase sequence, and combining them into a complex spectrum value sequence specifically includes:

[0022] The sampling frequency corresponding to the discrete sampling is obtained based on the sampling time interval;

[0023] Construct a set of discrete frequency points according to the sampling frequency and the total number of discrete sample points, and establish a one-to-one correspondence between each discrete frequency point and the frequency domain discrete frequency index;

[0024] Taking a discrete pulse signal of unit energy as input, and according to the discrete frequency domain transformation rules, the corresponding complex spectrum value is calculated at each discrete frequency point. The complex spectrum value includes the real part and the imaginary part.

[0025] Based on the complex spectrum values ​​at each discrete frequency point, the spectral amplitude at each frequency point is calculated. The spectral amplitude is obtained by squaring the real part and the imaginary part of the complex spectrum, summing them, and taking the square root.

[0026] Based on the complex spectral values ​​of each discrete frequency point, the spectral phase of each frequency point is obtained by using a two-parameter arctangent operation, forming a spectral amplitude sequence and a spectral phase sequence that correspond one-to-one with the discrete frequency index in the frequency domain.

[0027] Optionally, the step of establishing a power amplifier frequency response model based on the power amplifier's cutoff frequency parameters, obtaining the amplitude gain coefficient at each frequency point, and calculating the output spectrum amplitude distribution after frequency response compression specifically includes:

[0028] Obtain the cutoff frequency parameters of the power amplifier to be modeled;

[0029] Based on the cutoff frequency parameter and the set of discrete frequency points, a corresponding amplitude gain coefficient is established at each discrete frequency point. The amplitude gain coefficient varies with the frequency in the frequency range below or equal to the cutoff frequency parameter, and converges to zero in the frequency range above the cutoff frequency parameter.

[0030] Using the spectral amplitude sequence as input, the corresponding spectral amplitude and amplitude gain coefficient are multiplied at each discrete frequency point to obtain the output spectral amplitude distribution after frequency response compression by the power amplifier.

[0031] Optionally, the construction of a complex spectral pre-compensation function based on the power amplifier frequency response model, setting an amplitude pre-compensation factor and a frequency-varying advance phase compensation amount at each frequency point, forms a complex pre-compensation function that simultaneously includes amplitude and phase compensation, specifically including:

[0032] At each discrete frequency point, the reciprocal of the amplitude gain coefficient is calculated for the frequency points where the amplitude gain coefficient is greater than zero to obtain the amplitude pre-compensation factor. For the frequency points where the amplitude gain coefficient is zero, the amplitude pre-compensation factor is set to zero.

[0033] Based on the cutoff frequency parameter and the preset high-frequency weighting coefficient, the advance phase compensation amount, which changes monotonically with the increase of frequency, is calculated at each discrete frequency point. The advance phase compensation amount takes a larger value in the high-frequency region and a smaller value in the low-frequency region.

[0034] At each discrete frequency point, the amplitude pre-compensation factor is multiplied by a complex exponential factor consisting of the advance phase compensation amount to obtain a sequence of complex spectral pre-compensation function values ​​that simultaneously contain amplitude compensation components and phase compensation components.

[0035] Optionally, the step of applying a complex spectral pre-compensation function to a complex spectral value sequence to obtain a pre-compensated spectral value sequence, performing a discrete inverse transform on it, and reconstructing the pre-compensated time-domain discrete pulse signal specifically includes:

[0036] At each discrete frequency point, read the complex spectrum value from the pre-compensated spectrum value sequence;

[0037] Perform a discrete inverse transform operation on the pre-compensated spectral value sequence, calculate the corresponding discrete voltage sample value at each discrete sampling time in the time domain, and reconstruct the pre-compensated discrete pulse signal in the time domain.

[0038] Optionally, the step of inputting the pre-compensated time-domain discrete pulse signal into the power amplifier, reconstructing the output spectrum based on the frequency response model, calculating the output amplitude and phase at frequency points, representing different frequency bands in segments, and combining them into a complex output spectrum value sequence specifically includes:

[0039] The pre-compensated time-domain discrete pulse signal is input into the power amplifier. Using the power amplifier frequency response model, the complex output spectrum value sequence of the pre-compensated time-domain discrete pulse signal after being processed by the power amplifier is calculated at each discrete frequency point.

[0040] The complex output spectrum value sequence is segmented according to whether the amplitude gain coefficient is greater than zero: at frequency points where the amplitude gain coefficient is greater than zero, the output spectrum amplitude is the product of the pre-compensated spectrum amplitude and the amplitude gain coefficient, and the output spectrum phase is the pre-compensated spectrum phase; at frequency points where the amplitude gain coefficient is equal to zero, the output spectrum amplitude and output spectrum phase are set to zero.

[0041] Optionally, the step of performing a discrete inverse transform on the complex output spectral value sequence to obtain the power amplifier output time-domain discrete signal, calculating the peak power index and constructing a spectral energy fidelity index to evaluate the peak power improvement effect specifically includes:

[0042] Perform a discrete inverse transform operation on the complex output spectrum value sequence, calculate the discrete time-domain voltage sample value of the power amplifier output at each discrete sampling time in the time domain, and construct the discrete time-domain signal of the power amplifier output;

[0043] The peak power index of the power amplifier output time-domain discrete signal is constructed by squaring all discrete voltage sample values ​​in the power amplifier output time-domain discrete signal and selecting the maximum value at all time-domain discrete sampling times.

[0044] The amplitudes of each discrete frequency point in the complex output spectral value sequence and the complex spectral value sequence of the uncompensated unit energy discrete pulse signal are squared and summed. When the sum of squares of the amplitudes of the complex output spectral value sequence is greater than zero, the ratio of the sum of squares of the amplitudes of the complex output spectral value sequence to the sum of squares of the amplitudes of the complex spectral value sequence of the unit energy discrete pulse signal is used as the spectral energy fidelity index. When the sum of squares of the amplitudes of the complex output spectral value sequence is equal to zero, the spectral energy fidelity index is set to zero.

[0045] Set the spectral energy fidelity threshold to 0.01 and the ideal value to 1. When the absolute value of the difference between the spectral energy fidelity index and the ideal value is less than or equal to the spectral energy fidelity threshold, the peak power enhancement effect is deemed effective. When the absolute value of the difference between the spectral energy fidelity index and the ideal value is greater than the spectral energy fidelity threshold, the peak power enhancement effect is deemed ineffective.

[0046] Optionally, the step of calculating the average power index based on the discrete-time signal output by the power amplifier, normalizing the power when the spectral energy fidelity index meets a preset constraint to obtain a normalized output signal, and connecting the normalized output signal to the input port of the transmission system, specifically includes:

[0047] The average power index of the output signal within one sampling period is calculated by squaring all discrete voltage sample values ​​in the discrete time-domain output signal of the power amplifier and taking the arithmetic mean at each discrete sampling time in the time domain.

[0048] When the average power index is greater than zero and the deviation between the spectral energy fidelity index and the ideal value is less than or equal to the spectral energy fidelity threshold, the amplitude adjustment coefficient is calculated based on the ratio of the system target power index to the average power index. Each discrete voltage sample value in the time-domain discrete signal output by the power amplifier is multiplied by the amplitude adjustment coefficient to obtain the output discrete voltage sample value sequence after power normalization.

[0049] When the average power index is greater than zero and the deviation between the spectral energy fidelity index and the ideal value is greater than the spectral energy fidelity threshold, the discrete time-domain signal output by the power amplifier is directly used as the normalized output signal; when the average power index is equal to zero, all output discrete voltage sample values ​​are set to zero to form the normalized output signal.

[0050] Connect the normalized output signal to the input port of the transmitting system.

[0051] The present invention has the following beneficial effects:

[0052] 1. The continuous analog pulse voltage signal is converted into a discrete voltage sequence based on a preset sampling interval. After calculating the sequence energy index, it is normalized to obtain a discrete pulse signal with unit energy. Power balancing across the entire signal band is pre-processed to ensure that subsequent frequency domain transformations and compensation operations operate on the same energy reference, making the compensation coefficients for each frequency band component more stable and reliable. Compared to traditional methods of directly processing the original signal spectrum, energy normalization prevents over- or under-compensation of frequency components under different amplifier gain characteristics due to initial energy differences. Simultaneously, this step eliminates the impact of signal amplitude fluctuations on subsequent frequency domain analysis, providing a solid foundation for constructing accurate spectral amplitude and phase sequences, and improving the system's anti-interference capability and implementation stability. In practical engineering, the power distribution of each frequency band after amplification can be precisely controlled, effectively reducing distortion and aberration.

[0053] 2. For the normalized discrete-time signal, a discrete-frequency-domain transform is performed to obtain complete spectral amplitude and phase information, which is then organized into a sequence of complex spectral values. A unified frequency-domain discrete transform rule and frequency point index mapping are employed, combined with a two-parameter arctangent operation to accurately extract the phase, ensuring the complete revelation of the spectral structure. Unlike traditional approaches that only acquire amplitude information or simplify phase processing, this scheme emphasizes simultaneously capturing amplitude and phase, combining them into a sequence of complex spectral values. This provides a directly usable data format for subsequent complex pre-compensation functions, avoiding error accumulation caused by parameter conversion and quadratic interpolation. Furthermore, the constructed spectral amplitude and phase sequences are incorporated into the frequency response model and pre-compensation function design, achieving a systematic improvement in compensation accuracy. In practical applications, this step enables the system to comprehensively control even minor distortions in the signal spectrum, providing core data support for high-fidelity amplification.

[0054] 3. Based on the cutoff frequency parameter of the power amplifier, a concise yet accurate frequency response model was established, and the amplitude gain coefficient at each discrete frequency point was calculated. The gain function was designed with the cutoff frequency as the core, allowing it to gradually change according to physical characteristics below the cutoff frequency and instantaneously return to zero above the cutoff frequency, thus realistically simulating the bandwidth limitation of the amplifier. This targeted modeling compensates for the deviation between traditional ideal gain assumptions and the dynamic characteristics of the real amplifier, making subsequent pre-compensation design more targeted. Through the discretization calculation of the gain coefficient, accurate compression amplitude indicators were provided for the construction of the frequency domain pre-compensation function and subsequent piecewise representation, helping to accurately offset the amplifier's energy compression in different frequency bands during compensation. In engineering implementation, this method simplifies the model construction process while balancing the computational efficiency of the algorithm with the system's adaptability to real hardware characteristics, effectively reducing parameter tuning costs.

[0055] 4. Based on the power amplifier frequency response model, a complex spectral pre-compensation function that simultaneously incorporates amplitude and phase compensation was designed. The amplitude compensation factor is constructed using the reciprocal of the gain coefficient, perfectly offsetting the energy compression caused by the amplifier. The phase compensation amount is designed based on frequency weights, incorporating high-frequency advance and low-frequency mitigation to achieve phase pre-adjustment. By unifying amplitude and phase compensation into a single complex function and applying it simultaneously to the spectrum via complex multiplication, a one-time dual compensation is achieved. This overcomes the phase and amplitude interference problem inherent in traditional step-by-step compensation. This strategy can precisely adjust each frequency component in the frequency domain, ensuring that the compensated time-domain signal possesses ideal energy and phase distribution at the amplifier input.

[0056] 5. The generated complex spectrum pre-compensation function is applied to the complex spectrum value sequence. After obtaining the pre-compensated spectrum value sequence, a discrete inverse transform is performed to reconstruct the time-domain pulse signal. This tightly integrates frequency domain compensation and time-domain reconstruction, avoiding synchronization errors caused by separate processing in the traditional intermediate frequency and time domains. By directly performing a one-time inverse transform on the complex spectrum value sequence, the reconstructed pre-compensated time-domain signal inherently possesses both amplitude and phase adjustment effects, ensuring that the signal input to the power amplifier already has optimized characteristics in the time domain. This method not only simplifies the data conversion process but also ensures consistent compensation accuracy, providing the optimal signal shape for subsequent amplifier input.

[0057] 6. After inputting the pre-compensated time-domain signal into the power amplifier, the output spectrum is reconstructed based on the frequency response model. The output amplitude and phase are calculated at discrete frequency points, and the response characteristics of different frequency bands are represented in segments, which are then combined into a complex output spectrum value sequence. A segmented representation strategy is introduced to independently statistically analyze and calibrate key segments such as low and high frequencies, and passband and stopband. Compared to overall linear response analysis, this segmented method can more precisely distinguish the nonlinear distortion and phase shift between frequency bands, helping engineers to adjust compensation parameters or hardware design in a targeted manner. The complex output spectrum value sequence generated after segmentation reflects both the overall amplification characteristics and highlights the details of each frequency band, providing rich data dimensions for subsequent peak power assessment and fidelity calculation, enabling refined monitoring and optimization of amplifier performance.

[0058] 7. Perform a discrete inverse transform on the complex output spectral value sequence to recover the power amplifier output time-domain signal and calculate the peak power index. Simultaneously, construct a spectral energy fidelity index to quantify the spectral compensation effect. Combining peak power and spectral fidelity—two important performance parameters—forms a multi-dimensional evaluation system. Unlike traditional isolated evaluations that only focus on peak power improvement or spectral consistency, this method considers both simultaneously, enabling the monitoring of spectral distortion while optimizing peak power, thus achieving the optimal trade-off in system design and parameter tuning. Through this comprehensive evaluation, engineers can clearly see the relationship between power improvement and spectral distortion, providing a clear direction for further iterative compensation design and improving the overall system performance and controllability.

[0059] 8. Based on the quantitative determination of spectral fidelity and average power, the output time-domain signal is normalized, and the final signal is connected to the input port of the transmitting system. This solves the contradiction that traditional high-power amplifiers often struggle to balance target power output and spectral consistency. This solution pre-calculates the normalization coefficients to ensure that the output signal strictly matches the system's target power, and flexibly selects compensation or direct output within the fidelity threshold range, ensuring that a signal that meets power requirements while maintaining spectral integrity can be obtained under different operating scenarios. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] Example, refer to Figure 1 A power optimization method for high-power pulse amplification based on digital preprocessing includes:

[0063] The analog pulse voltage is sampled to generate a discrete voltage sequence covering a predetermined time. The energy index is calculated and normalized to obtain a discrete pulse signal with unit energy.

[0064] Discrete frequency domain transformation is performed on a unit energy discrete pulse signal to establish a set of discrete frequency points that match the sampling parameters. Complex spectral values ​​are obtained according to the frequency points, and the amplitude and phase are calculated to form a spectral amplitude sequence and a spectral phase sequence, which are then combined into a complex spectral value sequence.

[0065] A frequency response model of the power amplifier is established based on the cutoff frequency parameter of the power amplifier. The amplitude gain coefficient is obtained at each frequency point, and the amplitude distribution of the output spectrum after frequency response compression is calculated.

[0066] Based on the power amplifier frequency response model, a complex spectrum pre-compensation function is constructed. An amplitude pre-compensation factor and a frequency-varying advance phase compensation amount are set at each frequency point to form a complex pre-compensation function that simultaneously contains amplitude and phase compensation.

[0067] By applying a complex spectrum pre-compensation function to a complex spectrum value sequence, a pre-compensated spectrum value sequence is obtained. Then, a discrete inverse transform is performed on the sequence to reconstruct the pre-compensated time-domain discrete pulse signal.

[0068] The pre-compensated time-domain discrete pulse signal is input into the power amplifier, the output spectrum is reconstructed based on the frequency response model, the output amplitude and phase are calculated according to the frequency points, different frequency bands are segmented and represented, and combined into a complex output spectrum value sequence.

[0069] Perform a discrete inverse transform on the complex output spectrum value sequence to obtain the time-domain discrete signal of the power amplifier output, calculate the peak power index and construct the spectrum energy fidelity index to evaluate the peak power improvement effect;

[0070] The average power index is calculated based on the discrete time-domain signal output by the power amplifier. When the spectral energy fidelity index meets the preset constraints, the power is normalized to obtain the normalized output signal, and the normalized output signal is connected to the input port of the transmission system.

[0071] First, the analog pulse voltage signal is discretely sampled and its energy normalized, solving the problem of inconsistent compensation coefficients caused by varying signal amplitudes in traditional methods. This allows subsequent frequency domain processing to be performed under the same energy reference. Next, a frequency domain transformation is performed under a unified reference, accurately extracting amplitude and phase information and combining them into a complex spectral value sequence. This fills the gap in existing technologies that often only focus on amplitude while ignoring phase, laying the data foundation for dual compensation. A realistic amplifier frequency response model is constructed, incorporating the cutoff frequency parameter into the gain coefficient calculation. This simulates the amplifier's gain normalization and distortion across different frequency bands from the source, making compensation more targeted. Subsequently, a complex pre-compensation function is designed, merging the amplitude compensation factor and phase pre-adjustment. This multiplication simultaneously cancels the amplifier's energy compression and phase distortion of the signal, solving the problems of mutual interference and convergence difficulties in traditional separate compensation. The pre-compensation function is then applied to the complex spectral value sequence and inversely transformed back to the time domain, further bridging the frequency and time domain processing stages and avoiding distortion caused by double interpolation or resampling. The pre-compensated time-domain signal is then input to the amplifier. Guided by the same frequency response model, the output spectrum is reconstructed and represented in segments. Unlike existing technologies that only provide the overall response, this scheme can finely demonstrate the amplification effect of each frequency band, facilitating subsequent quantitative evaluation. The peak power and spectral energy fidelity of the reconstructed output signal are jointly evaluated, overcoming the drawback of focusing only on peak output while ignoring spectral distortion, thus forming a scientific measurement system with two indicators. Finally, under the premise of meeting the fidelity threshold, the output signal is power normalized, ensuring not only the target power output but also spectral fidelity, guaranteeing that the signal has sufficient power without significant spectral distortion.

[0072] The process of sampling the analog pulse voltage, generating a discrete voltage sequence covering a predetermined time, calculating and normalizing the energy index to obtain a discrete pulse signal with unit energy specifically includes:

[0073] Set the analog pulse voltage output by the signal source as a real-value continuous-time signal, and set the total sampling time length of the real-value continuous-time signal and the fixed time interval between adjacent samples;

[0074] Based on the total sampling time and the sampling time interval, calculate the total number of discrete sample points that can be obtained during the entire sampling time, and establish a time-domain discrete sample index for each sample point in chronological order.

[0075] According to the sampling time interval, the corresponding analog pulse voltage values ​​are collected at each continuous time moment to construct a discrete voltage sequence composed of the original discrete voltage sample values ​​at each sampling moment;

[0076] The energy index of the discrete voltage sequence is obtained by squaring and summing all the original discrete voltage sample values ​​in the discrete voltage sequence.

[0077] When the energy index is positive, the ratio of each original discrete voltage sample value to the square root of the energy index is used as the normalized discrete voltage sample value to construct a unit energy discrete pulse signal; when the energy index is zero, all normalized discrete voltage sample values ​​are set to zero.

[0078] Set the signal source output analog pulse voltage to a real function. The entire sampling process lasted for a duration of 100 minutes. The fixed time interval between two adjacent samples is ;in, It is a continuous-time variable;

[0079] Calculate the total number of sample points obtained after discrete sampling of a continuous signal. ;

[0080] No. The continuous time intervals corresponding to each sampling point are: , ;in, This is an index for discrete samples in the time domain;

[0081] Constructing a discrete signal sequence after sampling: , ;in, In the first Each sampling time The original discrete voltage sample values ​​obtained at the location;

[0082] Calculate the original discrete signal sequence Energy index ;

[0083] when At that time, normalization is performed to construct a unit energy signal:

[0084] ;in, These are the normalized discrete voltage sample values;

[0085] when At that time, take , .

[0086] The process of performing discrete frequency domain transformation on a unit energy discrete pulse signal, establishing a set of discrete frequency points matching the sampling parameters, obtaining complex spectral values ​​at each frequency point, calculating amplitude and phase, forming a spectral amplitude sequence and a spectral phase sequence, and combining them into a complex spectral value sequence, specifically includes:

[0087] The sampling frequency corresponding to the discrete sampling is obtained based on the sampling time interval;

[0088] Construct a set of discrete frequency points according to the sampling frequency and the total number of discrete sample points, and establish a one-to-one correspondence between each discrete frequency point and the frequency domain discrete frequency index;

[0089] Taking a discrete pulse signal of unit energy as input, and according to the discrete frequency domain transformation rules, the corresponding complex spectrum value is calculated at each discrete frequency point. The complex spectrum value includes the real part and the imaginary part.

[0090] Based on the complex spectrum values ​​at each discrete frequency point, the spectral amplitude at each frequency point is calculated. The spectral amplitude is obtained by squaring the real part and the imaginary part of the complex spectrum, summing them, and taking the square root.

[0091] Based on the complex spectral values ​​of each discrete frequency point, the spectral phase of each frequency point is obtained by using a two-parameter arctangent operation, forming a spectral amplitude sequence and a spectral phase sequence that correspond one-to-one with the discrete frequency index in the frequency domain.

[0092] Calculate the sampling frequency corresponding to discrete sampling ;

[0093] Construct a set of discrete frequencies: , ;in, In order to be with the first The actual frequency value corresponding to each frequency index; For the index of discrete frequency points in the frequency domain;

[0094] Calculate normalized discrete signals In the Complex spectral values ​​at each frequency point Specifically:

[0095] , ;in, The imaginary unit; It is an exponential function;

[0096] Calculate the first The spectral amplitude corresponding to each frequency point Specifically:

[0097] , ;in, For complex numbers The real part; For complex numbers The imaginary part;

[0098] Extract the first The spectral phase corresponding to each frequency point Specifically:

[0099] , ;in, It is a two-parameter arctangent function.

[0100] The process of establishing a power amplifier frequency response model based on the power amplifier's cutoff frequency parameters, obtaining the amplitude gain coefficient at each frequency point, and calculating the output spectrum amplitude distribution after frequency response compression specifically includes:

[0101] Obtain the cutoff frequency parameters of the power amplifier to be modeled;

[0102] Based on the cutoff frequency parameter and the set of discrete frequency points, a corresponding amplitude gain coefficient is established at each discrete frequency point. The amplitude gain coefficient varies with the frequency in the frequency range below or equal to the cutoff frequency parameter, and converges to zero in the frequency range above the cutoff frequency parameter.

[0103] Using the spectral amplitude sequence as input, the corresponding spectral amplitude and amplitude gain coefficient are multiplied at each discrete frequency point to obtain the output spectral amplitude distribution after frequency response compression by the power amplifier.

[0104] Obtain the amplifier's cutoff frequency, denoted as... ;

[0105] The amplifier gain function is constructed as follows:

[0106] , ;in, For the amplifier at the 1st Amplitude gain coefficient at each frequency point;

[0107] The compressed output spectral amplitude is:

[0108] , ;in, The signal is compressed by the amplifier frequency response at the 1st... Amplitude indicators at each frequency point.

[0109] The complex spectral pre-compensation function is constructed based on the power amplifier frequency response model. An amplitude pre-compensation factor and a frequency-varying advance phase compensation amount are set at each frequency point, forming a complex pre-compensation function that simultaneously includes amplitude and phase compensation. Specifically, this includes:

[0110] At each discrete frequency point, the reciprocal of the amplitude gain coefficient is calculated for the frequency points where the amplitude gain coefficient is greater than zero to obtain the amplitude pre-compensation factor, and the amplitude pre-compensation factor is set to zero for the frequency points where the amplitude gain coefficient is zero.

[0111] Based on the cutoff frequency parameter and the preset high-frequency weighting coefficient, the advance phase compensation amount, which changes monotonically with the increase of frequency, is calculated at each discrete frequency point. The advance phase compensation amount takes a larger value in the high-frequency region and a smaller value in the low-frequency region.

[0112] At each discrete frequency point, the amplitude pre-compensation factor is multiplied by a complex exponential factor consisting of the advance phase compensation amount to obtain a sequence of complex spectral pre-compensation function values ​​that simultaneously contain amplitude compensation components and phase compensation components.

[0113] The amplitude pre-compensation factor is constructed as follows:

[0114] , ;in, This represents the amplitude pre-compensation factor constructed to counteract amplifier gain compression.

[0115] The high-frequency advance phase compensation function is constructed as follows:

[0116] , ;in, In the first The amount of phase precompensation applied at each frequency point;

[0117] Calculate in the first Complex spectrum precompensation function values ​​at each frequency point Specifically:

[0118] , .

[0119] The process of applying a complex spectral pre-compensation function to a complex spectral value sequence to obtain a pre-compensated spectral value sequence, performing a discrete inverse transform on the sequence, and reconstructing the pre-compensated time-domain discrete pulse signal specifically includes:

[0120] At each discrete frequency point, read the complex spectrum value from the pre-compensated spectrum value sequence;

[0121] Perform a discrete inverse transform operation on the pre-compensated spectral value sequence, calculate the corresponding discrete voltage sample value at each discrete sampling time in the time domain, and reconstruct the pre-compensated discrete pulse signal in the time domain.

[0122] Construct the spectrum-compensated signal: , ;in, To the original spectrum Apply pre-compensation function The second one obtained spectral values ​​at each frequency point;

[0123] Reconstructing the time-domain signal:

[0124] , ;in, For the pre-compensated spectrum The discrete time-domain sample values ​​obtained after inverse transformation.

[0125] The feature is that the step of inputting the pre-compensated time-domain discrete pulse signal into the power amplifier, reconstructing the output spectrum based on the frequency response model, calculating the output amplitude and phase at frequency points, representing different frequency bands in segments, and combining them into a complex output spectrum value sequence specifically includes:

[0126] The pre-compensated time-domain discrete pulse signal is input into the power amplifier. Using the power amplifier frequency response model, the complex output spectrum value sequence of the pre-compensated time-domain discrete pulse signal after being processed by the power amplifier is calculated at each discrete frequency point.

[0127] The complex output spectrum value sequence is segmented according to whether the amplitude gain coefficient is greater than zero: at frequency points where the amplitude gain coefficient is greater than zero, the output spectrum amplitude is the product of the pre-compensated spectrum amplitude and the amplitude gain coefficient, and the output spectrum phase is the pre-compensated spectrum phase; at frequency points where the amplitude gain coefficient is equal to zero, the output spectrum amplitude and output spectrum phase are set to zero.

[0128] The pre-compensated signal is calculated after being amplified at the [number]th [time]. The output spectrum values ​​at each frequency point are as follows:

[0129] , ;in, The pre-compensated signal is amplified and then... Output spectrum values ​​at each frequency point;

[0130] right Segmented representation:

[0131] , ;

[0132] The output amplitude is represented in segments as follows:

[0133] , ;in, For complex numbers The modulus of the output spectrum, i.e., the output spectrum at the th... The amplitude at each frequency point;

[0134] The output phase segmentation is represented as follows:

[0135] , ;in, The phase angle is a complex number.

[0136] The process of performing a discrete inverse transform on the complex output spectral value sequence to obtain the discrete time-domain signal of the power amplifier output, calculating the peak power index and constructing a spectral energy fidelity index, and evaluating the peak power improvement effect specifically includes:

[0137] Perform a discrete inverse transform operation on the complex output spectrum value sequence, calculate the discrete time-domain voltage sample value of the power amplifier output at each discrete sampling time in the time domain, and construct the discrete time-domain signal of the power amplifier output;

[0138] The peak power index of the power amplifier output time-domain discrete signal is constructed by squaring all discrete voltage sample values ​​in the power amplifier output time-domain discrete signal and selecting the maximum value at all time-domain discrete sampling times.

[0139] The amplitudes of each discrete frequency point in the complex output spectral value sequence and the complex spectral value sequence of the uncompensated unit energy discrete pulse signal are squared and summed. When the sum of squares of the amplitudes of the complex output spectral value sequence is greater than zero, the ratio of the sum of squares of the amplitudes of the complex output spectral value sequence to the sum of squares of the amplitudes of the complex spectral value sequence of the unit energy discrete pulse signal is used as the spectral energy fidelity index. When the sum of squares of the amplitudes of the complex output spectral value sequence is equal to zero, the spectral energy fidelity index is set to zero.

[0140] Set the spectral energy fidelity threshold to 0.01 and the ideal value to 1. When the absolute value of the difference between the spectral energy fidelity index and the ideal value is less than or equal to the spectral energy fidelity threshold, the peak power enhancement effect is deemed effective. When the absolute value of the difference between the spectral energy fidelity index and the ideal value is greater than the spectral energy fidelity threshold, the peak power enhancement effect is deemed ineffective.

[0141] The computational amplifier outputs a time-domain signal, specifically:

[0142] , ;in, For the amplifier output signal at the 1st The discrete voltage values ​​in the time domain corresponding to each sampling time;

[0143] The peak power index is constructed as follows: ;in, To output discrete signal Peak power index;

[0144] Perform steps S701 and S702 to construct the spectral energy fidelity index. Specifically:

[0145] S701, when hour,

[0146] make ;

[0147] S702, when season ;

[0148] Set the spectral energy fidelity threshold. ;

[0149] when When the compensation effect is deemed effective, it will be judged as valid.

[0150] when In such cases, the compensation effect will be deemed invalid.

[0151] The process of calculating the average power index based on the discrete-time signal output by the power amplifier, normalizing the power when the spectral energy fidelity index meets a preset constraint, obtaining a normalized output signal, and connecting the normalized output signal to the input port of the transmission system specifically includes:

[0152] The average power index of the output signal within one sampling period is calculated by squaring all discrete voltage sample values ​​in the discrete time-domain output signal of the power amplifier and taking the arithmetic mean at each discrete sampling time in the time domain.

[0153] When the average power index is greater than zero and the deviation between the spectral energy fidelity index and the ideal value is less than or equal to the spectral energy fidelity threshold, the amplitude adjustment coefficient is calculated based on the ratio of the system target power index to the average power index. Each discrete voltage sample value in the time-domain discrete signal output by the power amplifier is multiplied by the amplitude adjustment coefficient to obtain the output discrete voltage sample value sequence after power normalization.

[0154] When the average power index is greater than zero and the deviation between the spectral energy fidelity index and the ideal value is greater than the spectral energy fidelity threshold, the discrete time-domain signal output by the power amplifier is directly used as the normalized output signal; when the average power index is equal to zero, all output discrete voltage sample values ​​are set to zero to form the normalized output signal.

[0155] Connect the normalized output signal to the input port of the transmitting system.

[0156] The average power index of the output signal is calculated as follows:

[0157] ;in, For output signal Average power parameters over a sampling period;

[0158] Steps S801 to S803 are performed to perform amplitude normalization, specifically as follows:

[0159] S801, when and At that time, let the normalized signal be:

[0160] , ;in, These are the discrete output voltage sample values ​​after power normalization. The target power index is preset for the system;

[0161] S802, when and At that time, take:

[0162] , ;

[0163] S803, when At that time, take: , ;

[0164] Will Connect to the launch system.

[0165] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0166] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A power optimization method for high-power pulse amplification based on digital preprocessing, characterized in that, include: The analog pulse voltage is sampled to generate a discrete voltage sequence covering a predetermined time. The energy index is calculated and normalized to obtain a discrete pulse signal with unit energy. Discrete frequency domain transformation is performed on a unit energy discrete pulse signal to establish a set of discrete frequency points that match the sampling parameters. Complex spectral values ​​are obtained according to the frequency points, and the amplitude and phase are calculated to form a spectral amplitude sequence and a spectral phase sequence, which are then combined into a complex spectral value sequence. Based on the cutoff frequency parameter of the power amplifier, a frequency response model of the power amplifier is established. The amplitude gain coefficient is obtained at each frequency point, and the amplitude distribution of the output spectrum after frequency response compression is calculated. Specifically, this includes: Obtain the cutoff frequency parameters of the power amplifier to be modeled; Based on the cutoff frequency parameter and the set of discrete frequency points, a corresponding amplitude gain coefficient is established at each discrete frequency point. The amplitude gain coefficient varies with the frequency in the frequency range below or equal to the cutoff frequency parameter, and converges to zero in the frequency range above the cutoff frequency parameter. Using the spectral amplitude sequence as input, the corresponding spectral amplitude and amplitude gain coefficient are multiplied at each discrete frequency point to obtain the output spectral amplitude distribution after frequency response compression by the power amplifier; Based on the power amplifier frequency response model, a complex spectrum pre-compensation function is constructed. An amplitude pre-compensation factor and a frequency-varying advance phase compensation amount are set at each frequency point to form a complex pre-compensation function that simultaneously contains amplitude and phase compensation. By applying a complex spectrum pre-compensation function to a complex spectrum value sequence, a pre-compensated spectrum value sequence is obtained. Then, a discrete inverse transform is performed on the sequence to reconstruct the pre-compensated time-domain discrete pulse signal. The pre-compensated time-domain discrete pulse signal is input into the power amplifier, the output spectrum is reconstructed based on the frequency response model, the output amplitude and phase are calculated according to the frequency points, different frequency bands are segmented and represented, and combined into a complex output spectrum value sequence. Perform a discrete inverse transform on the complex output spectrum value sequence to obtain the time-domain discrete signal of the power amplifier output, calculate the peak power index and construct the spectrum energy fidelity index to evaluate the peak power improvement effect; The average power index is calculated based on the discrete time-domain signal output by the power amplifier. When the spectral energy fidelity index meets the preset constraints, the power is normalized to obtain the normalized output signal, and the normalized output signal is connected to the input port of the transmission system.

2. The power optimization method for high-power pulse amplification based on digital preprocessing according to claim 1, characterized in that, The process of sampling the analog pulse voltage, generating a discrete voltage sequence covering a predetermined time, calculating and normalizing the energy index to obtain a discrete pulse signal with unit energy specifically includes: Set the analog pulse voltage output by the signal source as a real-value continuous-time signal, and set the total sampling time length of the real-value continuous-time signal and the fixed time interval between adjacent samples; Based on the total sampling time and the sampling time interval, calculate the total number of discrete sample points that can be obtained during the entire sampling time, and establish a time-domain discrete sample index for each sample point in chronological order. According to the sampling time interval, the corresponding analog pulse voltage values ​​are collected at each continuous time moment to construct a discrete voltage sequence composed of the original discrete voltage sample values ​​at each sampling moment; The energy index of the discrete voltage sequence is obtained by squaring and summing all the original discrete voltage sample values ​​in the discrete voltage sequence. When the energy index is positive, the ratio of each original discrete voltage sample value to the square root of the energy index is used as the normalized discrete voltage sample value to construct a unit energy discrete pulse signal; when the energy index is zero, all normalized discrete voltage sample values ​​are set to zero.

3. The power optimization method for high-power pulse amplification based on digital preprocessing according to claim 2, characterized in that, The process of performing discrete frequency domain transformation on a unit energy discrete pulse signal, establishing a set of discrete frequency points matching the sampling parameters, obtaining complex spectral values ​​at each frequency point, calculating amplitude and phase, forming a spectral amplitude sequence and a spectral phase sequence, and combining them into a complex spectral value sequence, specifically includes: The sampling frequency corresponding to the discrete sampling is obtained based on the sampling time interval; Construct a set of discrete frequency points according to the sampling frequency and the total number of discrete sample points, and establish a one-to-one correspondence between each discrete frequency point and the frequency domain discrete frequency index; Taking a discrete pulse signal of unit energy as input, and according to the discrete frequency domain transformation rules, the corresponding complex spectrum value is calculated at each discrete frequency point. The complex spectrum value includes the real part and the imaginary part. Based on the complex spectrum values ​​at each discrete frequency point, the spectral amplitude at each frequency point is calculated. The spectral amplitude is obtained by squaring the real part and the imaginary part of the complex spectrum, summing them, and taking the square root. Based on the complex spectral values ​​of each discrete frequency point, the spectral phase of each frequency point is obtained by using a two-parameter arctangent operation, forming a spectral amplitude sequence and a spectral phase sequence that correspond one-to-one with the discrete frequency index in the frequency domain.

4. The power optimization method for high-power pulse amplification based on digital preprocessing according to claim 3, characterized in that, The complex spectral pre-compensation function is constructed based on the power amplifier frequency response model. An amplitude pre-compensation factor and a frequency-varying advance phase compensation amount are set at each frequency point, forming a complex pre-compensation function that simultaneously includes amplitude and phase compensation. Specifically, this includes: At each discrete frequency point, the reciprocal of the amplitude gain coefficient is calculated for the frequency points where the amplitude gain coefficient is greater than zero to obtain the amplitude pre-compensation factor, and the amplitude pre-compensation factor is set to zero for the frequency points where the amplitude gain coefficient is zero. Based on the cutoff frequency parameter and the preset high-frequency weighting coefficient, the advance phase compensation amount, which changes monotonically with the increase of frequency, is calculated at each discrete frequency point. At each discrete frequency point, the amplitude pre-compensation factor is multiplied by a complex exponential factor consisting of the advance phase compensation amount to obtain a sequence of complex spectral pre-compensation function values ​​that simultaneously contain amplitude compensation components and phase compensation components.

5. The power optimization method for high-power pulse amplification based on digital preprocessing according to claim 4, characterized in that, The process of applying a complex spectral pre-compensation function to a complex spectral value sequence to obtain a pre-compensated spectral value sequence, performing a discrete inverse transform on the sequence, and reconstructing the pre-compensated time-domain discrete pulse signal specifically includes: At each discrete frequency point, read the complex spectrum value from the pre-compensated spectrum value sequence; Perform a discrete inverse transform operation on the pre-compensated spectral value sequence, calculate the corresponding discrete voltage sample value at each discrete sampling time in the time domain, and reconstruct the pre-compensated discrete pulse signal in the time domain.

6. The power optimization method for high-power pulse amplification based on digital preprocessing according to claim 5, characterized in that, The process involves inputting a pre-compensated time-domain discrete pulse signal into a power amplifier, reconstructing the output spectrum based on a frequency response model, calculating the output amplitude and phase at frequency points, representing different frequency bands in segments, and combining them into a complex output spectrum value sequence. Specifically, this includes: The pre-compensated time-domain discrete pulse signal is input into the power amplifier. Using the power amplifier frequency response model, the complex output spectrum value sequence of the pre-compensated time-domain discrete pulse signal after being processed by the power amplifier is calculated at each discrete frequency point. The complex output spectrum value sequence is segmented according to whether the amplitude gain coefficient is greater than zero: at frequency points where the amplitude gain coefficient is greater than zero, the output spectrum amplitude is the product of the pre-compensated spectrum amplitude and the amplitude gain coefficient, and the output spectrum phase is the pre-compensated spectrum phase; at frequency points where the amplitude gain coefficient is equal to zero, the output spectrum amplitude and output spectrum phase are set to zero.

7. The power optimization method for high-power pulse amplification based on digital preprocessing according to claim 6, characterized in that, The process of performing a discrete inverse transform on the complex output spectral value sequence to obtain the discrete time-domain signal of the power amplifier output, calculating the peak power index and constructing a spectral energy fidelity index, and evaluating the peak power improvement effect specifically includes: Perform a discrete inverse transform operation on the complex output spectrum value sequence, calculate the discrete time-domain voltage sample value of the power amplifier output at each discrete sampling time in the time domain, and construct the discrete time-domain signal of the power amplifier output; The peak power index of the power amplifier output time-domain discrete signal is constructed by squaring all discrete voltage sample values ​​in the power amplifier output time-domain discrete signal and selecting the maximum value at all time-domain discrete sampling times. The amplitudes of each discrete frequency point in the complex output spectral value sequence and the complex spectral value sequence of the uncompensated unit energy discrete pulse signal are squared and summed. When the sum of squares of the amplitudes of the complex output spectral value sequence is greater than zero, the ratio of the sum of squares of the amplitudes of the complex output spectral value sequence to the sum of squares of the amplitudes of the complex spectral value sequence of the unit energy discrete pulse signal is used as the spectral energy fidelity index. When the sum of squares of the amplitudes of the complex output spectral value sequence is equal to zero, the spectral energy fidelity index is set to zero. Set the spectral energy fidelity threshold to 0.01 and the ideal value to 1. When the absolute value of the difference between the spectral energy fidelity index and the ideal value is less than or equal to the spectral energy fidelity threshold, the peak power enhancement effect is deemed effective. When the absolute value of the difference between the spectral energy fidelity index and the ideal value is greater than the spectral energy fidelity threshold, the peak power enhancement effect is deemed ineffective.

8. The power optimization method for high-power pulse amplification based on digital preprocessing according to claim 7, characterized in that, The process of calculating the average power index based on the discrete-time signal output by the power amplifier, normalizing the power when the spectral energy fidelity index meets a preset constraint, obtaining a normalized output signal, and connecting the normalized output signal to the input port of the transmission system specifically includes: The average power index of the output signal within one sampling period is calculated by squaring all discrete voltage sample values ​​in the discrete time-domain output signal of the power amplifier and taking the arithmetic mean at each discrete sampling time in the time domain. When the average power index is greater than zero and the deviation between the spectral energy fidelity index and the ideal value is less than or equal to the spectral energy fidelity threshold, the amplitude adjustment coefficient is calculated based on the ratio of the system target power index to the average power index. Each discrete voltage sample value in the time-domain discrete signal output by the power amplifier is multiplied by the amplitude adjustment coefficient to obtain the output discrete voltage sample value sequence after power normalization. When the average power index is greater than zero and the deviation between the spectral energy fidelity index and the ideal value is greater than the spectral energy fidelity threshold, the discrete time-domain signal output by the power amplifier is directly used as the normalized output signal; when the average power index is equal to zero, all output discrete voltage sample values ​​are set to zero to form the normalized output signal. Connect the normalized output signal to the input port of the transmitting system.

Images