WO2018090417A1 - 基于双级预失真的超宽带复杂格式矢量调制误差修正方法 - Google Patents

基于双级预失真的超宽带复杂格式矢量调制误差修正方法 Download PDF

Info

Publication number
WO2018090417A1
WO2018090417A1 PCT/CN2016/109890 CN2016109890W WO2018090417A1 WO 2018090417 A1 WO2018090417 A1 WO 2018090417A1 CN 2016109890 W CN2016109890 W CN 2016109890W WO 2018090417 A1 WO2018090417 A1 WO 2018090417A1
Authority
WO
WIPO (PCT)
Prior art keywords
error
parameter
predistortion
modulation
vector
Prior art date
Application number
PCT/CN2016/109890
Other languages
English (en)
French (fr)
Inventor
台鑫
刘亮
樊晓腾
左永锋
薛晓楠
徐明哲
李增红
Original Assignee
中国电子科技集团公司第四十一研究所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国电子科技集团公司第四十一研究所 filed Critical 中国电子科技集团公司第四十一研究所
Publication of WO2018090417A1 publication Critical patent/WO2018090417A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03C—MODULATION
    • H03C7/00—Modulating electromagnetic waves
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/32—Carrier systems characterised by combinations of two or more of the types covered by groups H04L27/02, H04L27/10, H04L27/18 or H04L27/26
    • H04L27/34—Amplitude- and phase-modulated carrier systems, e.g. quadrature-amplitude modulated carrier systems
    • H04L27/36—Modulator circuits; Transmitter circuits
    • H04L27/366—Arrangements for compensating undesirable properties of the transmission path between the modulator and the demodulator

Definitions

  • the invention relates to the technical field of testing, in particular to a method for correcting vector modulation error of an ultra-wideband complex format based on two-stage predistortion.
  • the signals generated by the vector modulation function of the vector signal generator are mainly classified into two types, a multi-tone modulated signal and a complex modulated signal (for example, QPSK, FSK, QAM, etc.).
  • a multi-tone modulated signal for example, QPSK, FSK, QAM, etc.
  • a complex modulated signal for example, QPSK, FSK, QAM, etc.
  • the I/Q vector modulator has been significantly improved in performance over the years.
  • the modulation errors of current mainstream I/Q vector modulators are still obvious and cannot be ignored.
  • the amplitude-frequency response characteristics of the vector modulation error are obviously fluctuating, and multiple modulation errors overlap and mutually couple at each frequency point, which makes the high-precision vector modulation face great challenges.
  • Figure 1 shows a typical vector modulation link diagram, which consists of two parts: the transmission channel and the vector modulator.
  • V i and V q represent I/Q two-way input baseband signals, respectively
  • V m represents a modulated output signal.
  • the three dominant modulation errors include: I/Q two-baseband signal transmission channel gain imbalance error (the gain in the figure is represented by g i and g q respectively), DC offset Carrier leakage caused by component (the DC offset components in the figure are represented by a i and a q respectively) and carrier orthogonality error .
  • the more general method is the analog compensation correction method, that is, the special compensation circuit is used to perform error compensation correction on the vector modulation circuit under specific working conditions. Then, the obtained compensation value is used for the normal working state to realize the approximate compensation of the vector modulation error.
  • the working principle of the method is described in detail below.
  • the calibration compensation loop is mainly composed of a calibration source, a transmission channel, a compensation circuit, a vector modulator, and a compensation controller.
  • the main working principle is as follows:
  • the calibration source outputs (1,0), (0,1), Four values of (-1,0), (0,-1), set the phase shifter adjustment parameters one by one, traverse the orthogonality error compensation pre-stored calibration list data, select the IQ four-way four values, the output signal
  • the calibration value with the smallest mean value of the envelope voltage is used as the orthogonality error calibration parameter value.
  • the above three steps of calibration compensation are performed for each frequency sampling point, thereby establishing a calibration list of the full-band vector modulation error.
  • the input signal is switched to the baseband signal input mode, the calibration source needs continuous excitation, and the vector calibration error is approximated and calibrated using the obtained calibration list.
  • the present invention proposes an ultra-wideband complex format vector modulation error correction method based on two-stage predistortion.
  • An ultra-wideband complex format vector modulation error correction method based on two-stage predistortion includes: a standard signal generation module, a class I predistortion error compensator, a class II predistortion error compensator, an origin offset compensator, Transmission channel and vector modulator;
  • the Class I predistortion error compensator reverses the amplitude and phase response of the error in the form of pre-distortion of the shaping filter; the Class II predistortion error compensator has broadband characteristics through Class I predistortion equalization in the entire operating band.
  • the modulation error, ie, the I/Q gain imbalance and the gain imbalance positive angle error are corrected; the origin offset compensator corrects the narrowband characteristic modulation error, ie, carrier leakage, by means of an analog device.
  • the method for correcting an ultra-wideband complex format vector modulation error based on the two-stage predistortion includes the following steps:
  • Step (1) wideband vector modulation error amplitude and phase response characteristics measurement:
  • Threshold value after completing the parameter adjustment of each sampling point in the full working band, a set of different values of ⁇ and ⁇ are obtained, and the high and low fluctuation characteristics of the values are the amplitude-frequency response characteristics of the signal wide-band vector modulation error;
  • Step (2) shaping filter pre-distortion
  • the ideal amplitude-frequency response characteristic of the compensation filter is obtained; Then, after a fixed order of fft transform, a set of compensation filter parameters with the same order of the shaping filter are obtained;
  • the pre-distorted filter coefficients are embedded into the baseband signal generation module to complete the equalization of the amplitude-frequency response characteristics of the working band modulation error;
  • Step (3) modulation error correction based on digital-to-analog fusion:
  • step (31) the transmission channel and the vector modulator are regarded as a whole, and the input vector is written as:
  • the output modulation signal envelope V m is written as
  • V m M(V c +A), (2)
  • g i and g q are the I/Q baseband signal transmission channel gains, and a i and a q are DC offset components.
  • V c CV o +B, (4)
  • step (22) the identification process of parameters B and C is performed:
  • V e envelope detector made minimum Due to the characteristics of the detector diode, the detected value Ve has a convex function property near the minimum value, expressed as:
  • ⁇ e is the inherent detection offset of the detection diode
  • G e is the gain of the detection feedback circuit, both of which are fixed values
  • d and f represent parameters to be identified, and ⁇ 1 and ⁇ 2 represent tunable parameters
  • the input quantity V o is taken as a value:
  • ⁇ means that the search is stopped when the error is less than the predetermined range
  • search methods of parameters d and f are as follows:
  • ⁇ is a step parameter, where ⁇ G e is a positive value with a small value to ensure convergence of equation (10);
  • Step (23) after obtaining the parameter estimates of B and C, the value of B is sent to the DAC device of the origin offset compensator to achieve suppression of carrier leakage; the value of C is substituted into the value of the baseband signal generating module.
  • the digital values of the I/Q two channels are pre-distorted by direct addition and multiplication, thereby completing the overall correction of the modulation error of the full working band.
  • parameter adjustment is performed by means of traversal search.
  • the two-stage predistortion method achieves the equalization and synchronization reduction of the overall error of the working band, and the broadband vector modulation.
  • the error has a significant correction effect
  • a multi-parameter editable embedded sweep excitation signal model is proposed to ensure the consistency of the test state and working state of the measured vector signal generator.
  • Figure 1 is a schematic diagram of a typical vector modulation link
  • Figure 2 is a schematic diagram of the measured results of wideband multitone modulation (uncorrected error).
  • FIG. 3 is a schematic diagram of a conventional vector modulation error simulation calibration method
  • FIG. 4 is a schematic diagram of a two-stage predistortion error correction method according to the present invention.
  • Figure 5 (a, b, c) is a schematic diagram of the two-stage predistortion error correction effect
  • Figure 6 (a, b) is a schematic diagram of the measurement of the amplitude and phase response of the signal link modulation error.
  • the electromagnetic signals used in various high-end frequency devices today have a rapid development trend of broadband and complexity.
  • One of the core determinants of the development is the leap-forward improvement of complex vector modulation accuracy under large modulation bandwidth conditions.
  • the current high-end vector signal generator products all have broadband vector modulation function.
  • the vector modulation accuracy deteriorates rapidly with the increase of modulation bandwidth.
  • the symbol rate is greater than 40Msym/s, that is, the modulation bandwidth is greater than 80MHz
  • the EVM is often greater than 5%, and in many cases, the actual test requirements cannot be met.
  • most of the vector signal generators adopt the design scheme of fundamental wave vector modulation plus vector up-conversion.
  • This scheme is limited by the following three aspects in the compensation and correction process of wide-band vector modulation error due to its essential characteristics: (1) Both the analog device-based correction method imposes bandwidth constraints on the wideband error correction; (2) the wideband long-signal link modulation error amplitude-phase frequency response test is difficult; (3) under complex modulation, the modulation information is full of the whole work. In the frequency band, the signal model based on the frequency point cannot be obtained, and the modulation error modeling is difficult.
  • the invention proposes an ultra-wideband complex format vector modulation error correction method based on two-stage predistortion, and proposes vector modulation error correction based on digital-analog fusion, broadband multi-parameter editable multi-tone modulation baseband signal embedded modeling, and Wideband complex modulation error correction based on pre-distortion of shaping filter overcomes the above three problems and realizes three variations of vector error correction methods from analog to digital, from single frequency point to wide band, from multi-tone modulation to complex modulation. It effectively improves the accuracy of complex format vector modulation under large modulation bandwidth conditions.
  • the error correction method proposed by the present invention includes a standard signal generation module, a class I predistortion error compensator, a class II predistortion error compensator, an origin offset compensator, a transmission channel, and a vector.
  • the transmission channel and the vector modulator are the main generating parts of the modulation error. Therefore, the following three dominant errors are mainly concerned: the I/Q two-baseband signal transmission channel gain imbalance error (the gain in Figure 4 is used separately) i and g q represent), carrier leakage caused by DC bias component (DC offset component in Figure 4 is represented by a i and a q respectively) and carrier orthogonality error .
  • the amplitude-frequency response characteristics of various errors in the working band are obviously ups and downs, as shown in Fig. 5a.
  • the pre-distortion mode of the I-stage predistortion error compensator shaping filter, the amplitude-frequency response fluctuation characteristic of the reverse cancellation error makes the error balance in the whole working band approximate a straight line, as shown in Fig. 5b.
  • the Class II predistortion error compensator corrects the modulation error with wideband characteristics through I-level predistortion equalization in the entire working band: I/Q gain imbalance and gain imbalance positive angle error; the origin offset compensator passes the simulation The way the device corrects the narrowband characteristic modulation error: carrier leakage. While solving various modulation error coupling problems, the loss of DAC resolution is significantly reduced. The final correction effect is shown in Figure 5c. Shown.
  • Step (1) wideband vector modulation error amplitude and phase response characteristics measurement:
  • the invention firstly proposes a multi-parameter editable embedded swept excitation signal model (1) according to the functional characteristics of the vector signal generator itself:
  • Equation (1) is a standard baseband signal model of monophonic modulation. Due to the introduction of non-ideal characteristics, the image noise is uneven at different frequency points, that is, when the parameter k is different.
  • the parameter adjustment can be completed by means of traversal search. After the parameter adjustment of each sampling point in the full working band (ie, the actual modulation bandwidth) is completed, a set of different values of ⁇ and ⁇ are obtained, and the high and low fluctuation characteristics of these values are the broadband phase modulation error amplitude and phase response. The characteristics, as shown in Figure 5a, complete the corresponding tests.
  • Step (2) shaping filter pre-distortion
  • This step is a process for realizing the class I predistortion error compensator of FIG.
  • the present invention proposes a wide-band modulation error amplitude-frequency corresponding characteristic equalization method based on fixed-order forming filter pre-distortion.
  • the gain parameter is passed.
  • the ideal amplitude and phase response characteristics of the compensation filter can be obtained.
  • a fixed order of fft transform can obtain a set of compensation filter parameters that are the same as the shaped filter order.
  • the fitting error is negligible.
  • the above-described shaping filter and supplemental filter are all included in the class I predistortion error compensator.
  • the obtained compensation filter coefficient is multiplied by the original molding filter coefficient point to complete the pre-distortion of the shaping filter.
  • the pre-distorted filter coefficients are embedded in the baseband signal generation module to complete the equalization of the amplitude-frequency response characteristics of the operating band modulation error, as shown in Figure 5b.
  • Step (3) modulation error correction based on digital-to-analog fusion:
  • the invention proposes a modulation error correction method based on digital-analog fusion for the characteristics of three dominant modulation errors.
  • the I/Q gain imbalance and the I/Q positive angle error with wideband characteristics are corrected by predistortion, and the origin offset error with narrowband characteristics is corrected by means of analog device compensation, thereby solving various errors. While superimposing the coupling problem, the loss of the DAC resolution is effectively reduced, and the quantization noise of the generated signal is reduced.
  • step (2) the modulation error of the entire working band has been approximated to a straight line, that is, the modulation error in the working band is similar to the modulation error of the carrier point, so in the design of the class II predistortion compensator, only Consider the error correction of the carrier point.
  • the specific process is as follows:
  • Step (31) the main generation part transmission channel of the vector modulation error and the vector modulator are regarded as a whole, and the input vector can be written as: Considering the three main error distortions, the output modulation signal envelope V m can be written as
  • V m M(V c +A), (2)
  • V c CV o +B, (4)
  • Step (32), performing the identification process of parameters B and C as follows: First, the identification of the DC offset error compensation parameter B is performed.
  • ⁇ e is the inherent detection bias of the detector diode
  • G e is the gain of the detection feedback circuit.
  • the parameter C is related to the compensation of the I/Q gain imbalance error and the carrier orthogonality error. According to the structural characteristics of the equation (5), the parameter C is re-established. defined as:
  • d and f represent parameters to be identified, and ⁇ 1 and ⁇ 2 represent tunable parameters.
  • ⁇ is the step parameter, where ⁇ G e is a positive value with a small value to ensure the convergence of equation (10). According to the actual verification, about 8 iterations are needed.
  • Step (33) after obtaining the parameter estimates of B and C, the value of B is sent to the DAC device of the origin offset compensator to achieve suppression of carrier leakage; and the value of C is substituted into the value of the baseband signal generating module.
  • the digital values of the I/Q two channels are pre-distorted by direct addition and multiplication, thereby completing the overall correction of the modulation error of the full working band.
  • the invention provides a method for correcting vector modulation error compensation of ultra-wideband complex format based on two-stage predistortion, and corrects the broadband characteristic modulation error by using digital predistortion, which breaks through the prior art method which completely depends on the analog device.
  • the drawbacks effectively break the bandwidth characteristics of the analog device itself.
  • the error correction method proposed by the invention comprehensively considers the amplitude and frequency response of the signal link of the full working band
  • the influence of sex on various modulation errors, through the two-stage predistortion method, the equalization and synchronization reduction of the overall error of the working band is achieved, which has significant correction effect on the wideband vector modulation error; and the prior art method only applies to the single frequency. Point or narrowband modulation is effective, and there is nothing to do with wideband modulation errors.
  • the error correction method proposed by the invention adopts the pre-distortion method to the greatest extent, and saves 80% of the analog correction circuit compared with the prior art method, and obviously eliminates the hidden danger of individual performance of the simulation device itself, and improves the error correction. Precision.
  • the invention proposes a multi-parameter editable embedded swept excitation signal model, which ensures the consistency of the test state and the working state of the measured vector signal generator; avoids the prior art method relying on the calibration source alone excitation test, normal working state
  • the actual baseband signal is quite different from the calibration state, and only the drawback of approximate calibration can be achieved.

Landscapes

  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Digital Transmission Methods That Use Modulated Carrier Waves (AREA)
  • Transmitters (AREA)

Abstract

一种基于双级预失真的超宽带复杂格式矢量调制误差修正方法，其环路包括：标准信号生成模块、I级预失真误差补偿器、II级预失真误差补偿器、原点偏移补偿器、传输通道和矢量调制器；I级预失真误差补偿器以成型滤波器预失真的方式，反向抵消宽带矢量调制误差的幅相频响起伏特性；II级预失真误差补偿器将整个工作波段中经过I级预失真均衡的具有宽带特性的调制误差即I/Q增益不平衡和增益不平衡正角度误差予以修正；原点偏移补偿器则通过模拟器件的方式修正窄带特性调制误差即载波泄露。通过对于宽带特性矢量调制误差采用数字预失真的方式予以修正，对宽带复杂格式矢量调制误差做出了有效修正。

Description

基于双级预失真的超宽带复杂格式矢量调制误差修正方法 技术领域
本发明涉及测试技术领域，特别涉及一种基于双级预失真的超宽带复杂格式矢量调制误差修正方法。
背景技术
通过矢量信号发生器的矢量调制功能发生的信号主要分为两类，多音调制信号和复杂调制信号(例如QPSK、FSK、QAM等)。I/Q矢量调制器作为实现矢量调制的关键部件，经过多年的发展，其性能得到了显著提升。尽管如此，目前主流I/Q矢量调制器的调制误差仍然很明显，不容忽视。尤其是在大调制带宽条件下，矢量调制误差的幅相频率响应特性高低起伏明显，各频点多种调制误差交叠互耦，使高精度矢量调制面临巨大挑战。
首先介绍单频点矢量调制误差特性。图1给出了典型的矢量调制链路示意图，主要由传输通道和矢量调制器两部分组成。图1中，Vi和Vq分别表示I/Q两路输入基带信号，Vm表示调制输出信号。影响矢量调制精度的误差因素较多，其中3种占主导地位的调制误差包括：I/Q两路基带信号传输通道增益不平衡误差(图中增益分别用gi和gq表示)、直流偏置分量造成的载波泄漏(图中直流偏置分量分别用ai和aq表示)和载波正交度误差
Figure PCTCN2016109890-appb-000001
。一个经过严谨设计的矢量调制电路，在进行误差修正之前，这3种主要矢量调制误差的典型值大约是：增益不平衡度2-3％、载波正交度误差2-3°、载波泄漏大约占总量程的2-3％。在此量级误差的影响下，信号镜像噪声抑制仅为30dB左右，造成的交调失真大约是0.3dB，无法满足调制精度要求，必须对基带信号进行误差补偿修正。
对于宽带矢量调制，由于工作波段内各频点的调制误差相互耦合、叠加，使各种误差的解析表达异常复杂。以多音调制为例，如图2所示，由于上述3种主要误差在各频点的影响程度各异，导致各音的镜像噪声幅值差异明显，使交调噪声杂乱分布，最终造成多音调制信号严重失真，无法使用。对于宽带复杂调制格式，带来的影响是EVM显著恶化，严重超标。
目前，比较通用的方法是模拟补偿修正方法，即利用专用模拟电路，在特定的工作条件下，对矢量调制电路进行误差补偿修正。进而将得到的补偿值用于正常工作状态，来实现矢量调制误差的近似补偿，下面详细介绍此种方法的工作原理。
图3是矢量调制误差模拟校准方法的示意图，校准补偿回路主要由校准源、传输通道、补偿电路、矢量调制器和补偿控制器等部分组成，其主要工作原理为：
首先将输入端切换到校准源模式，设定待校准频率点，然后进行以下三个步骤：
(1)利用校准源将IQ两路信号置零，在射频输出端通过包络检测器监测输出电压，根据DSP中的预存校准数据列表，逐一设置补偿电路中的偏置电压补偿参数，遍历所有数据列表后，选择包络检测器输出电压最小的参数取值作为此频率点的直流偏置补偿值；
(2)在直流偏置补偿值生效的前提下，利用校准源将IQ两路信号分别设置为(0，1)、(1，0)，与第一步类似，逐一设置衰减器参数，遍历IQ增益平衡补偿预存校准列表数据后，选择IQ两路上述两种取值下输出信号包络电压差值最小的补偿参数取值，作为工作状态下的增益不平衡校准补偿参数值；
(3)在前两组校准参数生效的前提下，校准源分别输出(1，0)、(0，1)、 (-1，0)、(0，-1)四个值，逐一设置移相器调整参数，遍历正交度误差补偿预存校准列表数据后，选择IQ两路上述四种取值下，输出信号包络电压两两差值均值最小的校准值作为正交度误差校准参数值。
然后，对每个频率采样点进行以上三步校准补偿，从而建立全频段矢量调制误差的校准列表。在正常工作状态下，将输入信号切换到基带信号输入模式，校准源需要持续激励，利用得到的校准列表对矢量调制误差进行近似校准。
现有技术方案存在以下不足：
(1)由于上述方法完全基于模拟器件，仅对载波频点误差进行修正，进而将此频点的补偿数据用于全工作波段，不考虑误差的幅相频响的影响，因此仅对单频点或者窄带调制误差的修正有效，无法对宽带调制误差做出有效修正；
(2)由于模拟器件自身带宽特性的束缚，会限制整机的工作带宽的扩展，对大调制带宽信号会带来严重失真；
(3)上述方法需要一系列复杂的模拟电路，包括：校准源、衰减器、偏置电压产生电路和移相器等，由于模拟器件自身的性能存在个体差异，校准精度受到限制；
(4)通过校准过程建立的补偿参数列表是在校准源单独激励下得到的，正常工作状态下的实际基带信号与校准状态差别较大，只能做到近似校准。
发明内容
为解决上述现有技术中的不足，本发明提出一种基于双级预失真的超宽带复杂格式矢量调制误差修正方法。
本发明的技术方案是这样实现的：
一种基于双级预失真的超宽带复杂格式矢量调制误差修正方法，其环路包括：标准信号生成模块、I级预失真误差补偿器、II级预失真误差补偿器、原点偏移补偿器、传输通道和矢量调制器；
I级预失真误差补偿器以成型滤波器预失真的方式，反向抵消误差的幅相频响起伏特性；II级预失真误差补偿器将整个工作波段中经过I级预失真均衡的具有宽带特性的调制误差即I/Q增益不平衡和增益不平衡正角度误差予以修正；原点偏移补偿器则通过模拟器件的方式修正窄带特性调制误差即载波泄露。
可选地，上述基于双级预失真的超宽带复杂格式矢量调制误差修正方法，具体包括以下步骤：
步骤(1)、宽带矢量调制误差幅相频响特性测定：
首先，提出多参数可编辑嵌入式扫频激励信号模型，如式(1)：
Figure PCTCN2016109890-appb-000002
在此模型中，
Figure PCTCN2016109890-appb-000003
和
Figure PCTCN2016109890-appb-000004
分别表示I/Q两路注入激励信号，λ和ω分别代表了扫频信号的幅度增益和群时延；当λ＝1且ω＝0时，式(1)是单音调制的标准基带信号模型，将此模型式(1)中参数设为λ＝1 ω＝0，嵌入到被测矢量信号发生器的基带发生模块中，通过调节λ和ω两个参数，逐渐改善镜像噪声直至达到预定阈值；在完成全工作波段内的各取样点的参数调节之后，得到一组λ和ω的不同取值，这些取值的高低起伏特性即为信号宽带矢量调制误差幅相频响特性；
步骤(2)、成型滤波器预失真：
首先，根据步骤(1)中测定的宽带矢量调制幅相频响特性，经过增益参数取倒数、群时延参数取反的操作后，得到补偿滤波器的理想幅相频响特性；然 后，经过固定阶数的fft变换得到一组与成型滤波器阶数相同的补偿滤波器参数；
然后，将得到的补偿滤波器系数与原成型滤波器系数点乘，完成成型滤波器的预失真；
最后，将预失真后的滤波器系数嵌入到基带信号发生模块，完成工作波段调制误差幅相频响特性的均衡；
步骤(3)、基于数模融合的调制误差修正：
步骤(31)，将传输通道和矢量调制器看成一个整体，输入矢量写为：
Figure PCTCN2016109890-appb-000005
输出调制信号包络Vm写为
Vm＝M(Vc+A)，           (2)
其中，
Figure PCTCN2016109890-appb-000006
式中，gi和gq为I/Q两路基带信号传输通道增益，ai和aq为直流偏置分量，
Figure PCTCN2016109890-appb-000007
为载波正交度误差；
采取以下误差补偿器，
Vc＝CVo+B，                (4)
其中，
Figure PCTCN2016109890-appb-000008
Figure PCTCN2016109890-appb-000009
步骤(22)，进行参数B和C的辨识过程：
首先，进行直流偏置误差补偿参数B的辨识，采取最小二乘辨识准则：先将基带信号置零，即Vo＝0；通过调整参数B使包络检测器的检测值Ve取得最小值；由于检波二极管的特性，检测值Ve在最小值附近呈现凸函数性质，表示为：
Ve(B)≈Ge||B+A||2+Γe，         (7)
其中，Γe是检波二极管的固有检测偏置，Ge是检波反馈电路增益，这两个值均为固定值；
接下来，将参数B取一组特殊值，计算出相应的一组Ve的取值，进而将此组B和Ve的取值作为已知量，参数A作为未知量，通过最小二乘的方式求得参数A的最优估计值
Figure PCTCN2016109890-appb-000010
最后，根据
Figure PCTCN2016109890-appb-000011
的关系，求得参数B的最优估计值
Figure PCTCN2016109890-appb-000012
即矢量调制误差的直流偏置误差；
在确定直流偏置参数
Figure PCTCN2016109890-appb-000013
的取值之后，接下来做误差补偿器中参数C的辨识：
根据式(5)的结构特点，将参数C重新定义为：
Figure PCTCN2016109890-appb-000014
其中，d和f表示待辨识参数，ε1和ε2表示可调参数；
将输入量Vo分别取值：
Figure PCTCN2016109890-appb-000015
从而得到一组检波反馈取值：
Figure PCTCN2016109890-appb-000016
通过调节参数d和f，使
Figure PCTCN2016109890-appb-000017
实现参数辨识，“≈”表示当误差小于既定范围停止搜索，参数d和f的搜寻方法如下：
经级数展开得到以下关系式：
Figure PCTCN2016109890-appb-000018
进而得到
Figure PCTCN2016109890-appb-000019
其中δ是步进参数，这里δGeΛ取值数值小的正数，以保证式(10)的收敛；
步骤(23)，在得到B和C的参数估计值之后，将B的数值送入原点偏移补偿器的DAC器件中，实现对载波泄露的抑制；将C的数值代入基带信号发生模块的数值计算过程中，将I/Q两路的数字值通过直接的加和乘运算实现预失真，从而完成对全工作波段的调制误差的整体修正。
可选地，在所述多参数可编辑嵌入式扫频激励信号模型中，采用遍历搜索的方式完成参数调节。
本发明的有益效果是：
(1)综合考虑了全工作波段信号链路幅相频响特性对各种调制误差的影响，通过双级预失真的方式做到了工作波段整体误差的均衡归一和同步降低，对于宽带矢量调制误差具有显著的修正作用；
(2)最大程度的采用了预失真方式实现，与现有技术方法比较，节省了80％的模拟修正电路，明显消除了模拟器件自身的性能存在个体差异隐患，提高了误差修正精度；
(3)提出了多参数可编辑嵌入式扫频激励信号模型，保证了被测矢量信号发生器的测试状态和工作状态的一致性。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为典型的矢量调制链路示意图；
图2为宽带多音调制实测结果(误差未修正)示意图；
图3为现有的矢量调制误差模拟校准方法原理图；
图4为本发明的双级预失真误差修正方法原理图；
图5(a，b，c)是双级预失真误差修正效果示意图；
图6(a，b)是信号链路调制误差幅相频响测定示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
当今各种高端用频设备所用的电磁信号呈现宽带化和复杂化的快速发展趋势，其发展的核心决定因素之一是大调制带宽条件下复杂矢量调制精度的跨越式提升。目前的高端矢量信号发生器产品均具有宽带矢量调制功能，然而由于宽带矢量调制误差修正技术发展的滞后，导致了矢量调制精度随调制带宽的增加而迅速恶化。当码元速率大于40Msym/s，即调制带宽大于80MHz时，EVM往往大于5％，很多情况下无法满足实际的测试需求。目前矢量信号发生器整机大都采用基波矢量调制加矢量上变频的设计方案，此方案由于其本质特性决定了在宽带矢量调制误差的补偿修正过程中受到了以下三方面难题的制约：(1)既有基于模拟器件的修正方式对宽带误差修正造成带宽束缚；(2)宽带长信号链路调制误差幅相频率响应测试困难；(3)复杂调制下，调制信息充满整个工作 频段，无法得到基于频点的信号模型，调制误差建模困难。
本发明提出了一种基于双级预失真的超宽带复杂格式矢量调制误差修正方法，提出基于数模融合的矢量调制误差修正、宽带多参数可编辑的多音调制基带信号嵌入式建模、以及基于成型滤波器预失真的宽带复杂调制误差修正，克服了上述三方面的难题，实现了矢量误差修正方法从模拟到数字、从单频点到宽带、从多音调制到复杂调制的三个跨越，有效提高了大调制带宽条件下复杂格式矢量调制的精度。
下面结合说明书附图对本发明的方法进行详细说明。
如图4所示，本发明提出的误差修正方法，其环路包括：标准信号生成模块、I级预失真误差补偿器、II级预失真误差补偿器、原点偏移补偿器、传输通道和矢量调制器。其中，传输通道和矢量调制器是调制误差主要的产生部位，因此，主要关注以下三种占主导地位的误差：I/Q两路基带信号传输通道增益不平衡误差(图4中增益分别用gi和gq表示)、直流偏置分量造成的载波泄漏(图4中直流偏置分量分别用ai和aq表示)和载波正交度误差
Figure PCTCN2016109890-appb-000020
。
在大调制带宽条件下，由于矢量信号发生器信号长链路各级组件幅频频响的影响，工作波段内的各种误差的幅相频响特性跌宕起伏明显，如图5a所示。I级预失真误差补偿器成型滤波器预失真的方式，反向抵消误差的幅相频响起伏特性，使整个工作波段内的误差均衡近似为一条直线，如图5b所示。II级预失真误差补偿器将整个工作波段中经过I级预失真均衡的具有宽带特性的调制误差：I/Q增益不平衡和增益不平衡正角度误差予以修正；原点偏移补偿器则通过模拟器件的方式修正窄带特性调制误差：载波泄露。在解决了各种调制误差耦合问题的同时，显著减少了DAC分辨率的损失，最终的修正效果示意图如图5c 所示。
下面详细介绍本发明误差修正方法的各个步骤。
步骤(1)、宽带矢量调制误差幅相频响特性测定：
本发明首先根据矢量信号发生器本身的功能特点，提出一种多参数可编辑嵌入式扫频激励信号模型式(1)：
Figure PCTCN2016109890-appb-000021
在此模型中，
Figure PCTCN2016109890-appb-000022
和
Figure PCTCN2016109890-appb-000023
分别表示I/Q两路注入激励信号，λ和ω分别代表了扫频信号的幅度增益和群时延。当λ＝1且ω＝0时，式(1)是单音调制的标准基带信号模型，由于非理想特性的引入，导致在不同频点，即参数k不同取值时，镜像噪声高低不平。将此模型式(1)中参数设为λ＝1 ω＝0，嵌入到被测矢量信号发生器的基带发生模块中，可以激励产生如图6a所示的结果。通过调节λ和ω两个参数，会逐渐改善镜像噪声直至达到预定阈值，如图6b所示。工程中可采用遍历搜索的方式完成参数调节。在完成全工作波段(即实际调制带宽)内的各取样点的参数调节之后，会得到一组λ和ω的不同取值，这些取值的高低起伏特性即为宽带矢量调制误差幅相频响特性，如图5a所示，从而完成相应的测试。
步骤(2)、成型滤波器预失真：
本步骤是实现图4中I级预失真误差补偿器的过程。矢量信号发生器中基带信号发生模块常用的设计架构，程序滤波器往往采取固定阶数、参数时变的方式。根据此特点，本发明提出了基于固定阶数成型滤波器预失真的宽带调制误差幅频相应特性均衡方法。
首先，根据步骤(1)中测定的宽带矢量调制幅相频响特性，经过增益参数 取倒数，群时延参数取反的操作后，可得到补偿滤波器的理想幅相频响特性。然后，经过固定阶数的fft变换可以得到一组与成型滤波器阶数相同的补偿滤波器参数。此补偿滤波器的幅相频响与理想的补偿频响存在一定的拟合误差，但由于矢量信号发生器的幅相频响特性起伏缓慢，因此这里的拟合误差可忽略。上述成型滤波器、补充滤波器均包含在I级预失真误差补偿器中。
其次，将得到的补偿滤波器系数与原成型滤波器系数点乘，即可完成成型滤波器的预失真。
最后，将预失真后的滤波器系数嵌入到基带信号发生模块，即可完成工作波段调制误差幅相频响特性的均衡，如图5b所示。
步骤（3）、基于数模融合的调制误差修正：
本发明针对3种主导调制误差的自身特性，提出一种基于数模融合的调制误差修正方法。将具有宽带特性的I/Q增益不平衡和I/Q正角度误差通过预失真的方式修正，将具有窄带特性的原点偏移误差利用模拟器件补偿的方式予以修正，从而在解决了多种误差叠加耦合问题的同时，有效降低的DAC分辨率的损失，降低了所产生信号的量化噪声。
在步骤（2）中，已将整个工作波段的调制误差均衡近似为一条直线，即工作波段内的调制误差均与载波点的调制误差近似，因此在II级预失真补偿器的设计中，仅考虑载波点的误差修正即可，具体过程如下：
步骤（31），如图4所示，将矢量调制误差的主要产生部位传输通道和矢量调制器看成一个整体，输入矢量可写为：
Figure PCTCN2016109890-appb-000024
再考虑三种主要误差失真的作用下，输出调制信号包络Vm可写为
Vm＝M(Vc+A)，           （2）
其中，
Figure PCTCN2016109890-appb-000025
为补偿三种误差，达到输出调制信号包络与原始基带信号相等，即Vm＝Vo，可采取以下误差补偿器，
Vc＝CVo+B，             (4)
其中，
Figure PCTCN2016109890-appb-000026
Figure PCTCN2016109890-appb-000027
步骤(32)，进行参数B和C的辨识过程，具体如下：首先进行直流偏置误差补偿参数B的辨识，这里采取的是最小二乘辨识准则：先将基带信号置零，即：Vo＝0；通过调整参数B使包络检测器的检测值Ve取得最小值；由于检波二极管的特性，检测值Ve在最小值附近呈现凸函数性质，可表示为：
Ve(B)≈Ge||B+A||2+Γe，           (7)
其中，Γe是检波二极管的固有检测偏置，Ge是检波反馈电路增益，这两个值均为固定值，不影响参数辨识；接下来，将B取一组特殊值，计算出相应的一组Ve的取值，进而将此组B和Ve的取值作为已知量，参数A作为未知量，通过最小二乘的方式求的参数A的最优估计值
Figure PCTCN2016109890-appb-000028
最后，根据
Figure PCTCN2016109890-appb-000029
的关系，求得参数B的最优估计值
Figure PCTCN2016109890-appb-000030
即矢量调制误差的直流偏置误差。因为检波电路噪声的影响，这里需要大约8个B的特殊取值，才能得到最终的辨识值。
在确定直流偏置参数
Figure PCTCN2016109890-appb-000031
的取值之后，接下来做误差补偿器中参数C的辨识：参数C关系到I/Q增益不平衡误差和载波正交度误差的补偿，根据式(5)的结构特点，将参数C重新定义为：
Figure PCTCN2016109890-appb-000032
其中，d和f表示待辨识参数，ε1和ε2表示可调参数。
总体思路是将输入量Vo分别取值：
Figure PCTCN2016109890-appb-000033
从而得到一组检波反馈取值：
Figure PCTCN2016109890-appb-000034
通过调节参数d和f，使
Figure PCTCN2016109890-appb-000035
实现参数辨识。此处使用“≈”，表示当误差小于既定范围即可停止搜索。参数d和f的搜寻方法如下：
经级数展开可以得到以下关系式：
Figure PCTCN2016109890-appb-000036
进而得到
Figure PCTCN2016109890-appb-000037
其中δ是步进参数，这里δGeΛ取值数值小的正数，以保证式(10)的收敛，根据实际验证，大约需要8次迭代。
步骤(33)，在得到B和C的参数估计值之后，将B的数值送入原点偏移补偿器的DAC器件中，实现对载波泄露的抑制；将C的数值代入基带信号发生模块的数值计算过程中，将I/Q两路的数字值通过直接的加和乘运算实现预失真，从而完成对全工作波段的调制误差的整体修正。
本发明提出的一种基于双级预失真的超宽带复杂格式矢量调制误差补偿修正方法，对于宽带特性调制误差采用数字预失真的方式予以修正，突破了现有技术方法中完全依赖于模拟器件的弊端，有效打破了模拟器件自身的带宽特性束缚。
本发明提出的误差修正方法，综合考虑了全工作波段信号链路幅相频响特 性对各种调制误差的影响，通过双级预失真的方式做到了工作波段整体误差的均衡归一和同步降低，对于宽带矢量调制误差具有显著的修正作用；而现有技术方法仅对单频点或窄带调制有效，对宽带调制误差无能为力。
本发明提出的误差修正方法最大程度的采用了预失真方式实现，与现有技术方法比较，节省了80％的模拟修正电路，明显消除了模拟器件自身的性能存在个体差异隐患，提高了误差修正精度。
本发明提出了多参数可编辑嵌入式扫频激励信号模型，保证了被测矢量信号发生器的测试状态和工作状态的一致性；避免了现有技术方法依靠校准源单独激励测试，正常工作状态下的实际基带信号与校准状态差别较大，只能做到近似校准的弊端。
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (3)

  1. 一种基于双级预失真的超宽带复杂格式矢量调制误差修正方法，其特征在于，其环路包括：标准信号生成模块、I级预失真误差补偿器、II级预失真误差补偿器、原点偏移补偿器、传输通道和矢量调制器；
    I级预失真误差补偿器以成型滤波器预失真的方式，反向抵消误差的幅相频响起伏特性；II级预失真误差补偿器将整个工作波段中经过I级预失真均衡的具有宽带特性的调制误差即I/Q增益不平衡和增益不平衡正角度误差予以修正；原点偏移补偿器则通过模拟器件的方式修正窄带特性调制误差即载波泄露。
  2. 如权利要求1所述的一种基于双级预失真的超宽带复杂格式矢量调制误差修正方法，其特征在于，具体包括以下步骤：
    步骤(1)、宽带矢量调制误差幅相频响特性测定：
    首先，提出多参数可编辑嵌入式扫频激励信号模型，如式(1)：
    Vi c＝λ(k)sin(2πkft+ω(k))
    Vq c＝cos(2πkft)                        (1)
    k＝1,2,…,N
    在此模型中，Vi c和Vq c分别表示I/Q两路注入激励信号，λ和ω分别代表了扫频信号的幅度增益和群时延；当λ＝1且ω＝0时，式(1)是单音调制的标准基带信号模型，将此模型式(1)中参数设为λ＝1ω＝0，嵌入到被测矢量信号发生器的基带发生模块中，通过调节λ和ω两个参数，逐渐改善镜像噪声直至达到预定阈值；在完成全工作波段内的各取样点的参数调节之后，得到一组λ和ω的不同取值，这些取值的高低起伏特性即为信号宽带矢量调制误差幅相频响特性；
    步骤(2)、成型滤波器预失真：
    首先，根据步骤(1)中测定的宽带矢量调制幅相频响特性，经过增益参数取倒数、群时延参数取反的操作后，得到补偿滤波器的理想幅相频响特性；然 后，经过固定阶数的fft变换得到一组与成型滤波器阶数相同的补偿滤波器参数；
    然后，将得到的补偿滤波器系数与原成型滤波器系数点乘，完成成型滤波器的预失真；
    最后，将预失真后的滤波器系数嵌入到基带信号发生模块，完成工作波段调制误差幅相频响特性的均衡；
    步骤(3)、基于数模融合的调制误差修正：
    步骤(31)，将传输通道和矢量调制器看成一个整体，输入矢量写为：
    Figure PCTCN2016109890-appb-100001
    输出调制信号包络Vm写为
    Vm＝M(Vc+A)，                (2)
    其中，
    Figure PCTCN2016109890-appb-100002
    式中，gi和gq为I/Q两路基带信号传输通道增益，ai和aq为直流偏置分量，
    Figure PCTCN2016109890-appb-100003
    为载波正交度误差；
    采取以下误差补偿器，
    Vc＝CVo+B，           (4)
    其中，
    Figure PCTCN2016109890-appb-100004
    Figure PCTCN2016109890-appb-100005
    步骤(22)，进行参数B和C的辨识过程：
    首先，进行直流偏置误差补偿参数B的辨识，采取最小二乘辨识准则：先将基带信号置零，即Vo＝0；通过调整参数B使包络检测器的检测值Ve取得最小值；由于检波二极管的特性，检测值Ve在最小值附近呈现凸函数性质，表示为：
    Ve(B)≈Ge||B+A||2+Γe，          (7)
    其中，Γe是检波二极管的固有检测偏置，Ge是检波反馈电路增益，这两个值均为固定值；
    接下来，将参数B取一组特殊值，计算出相应的一组Ve的取值，进而将此组B和Ve的取值作为已知量，参数A作为未知量，通过最小二乘的方式求得参数A的最优估计值
    Figure PCTCN2016109890-appb-100006
    最后，根据
    Figure PCTCN2016109890-appb-100007
    的关系，求得参数B的最优估计值
    Figure PCTCN2016109890-appb-100008
    即矢量调制误差的直流偏置误差；
    在确定直流偏置参数
    Figure PCTCN2016109890-appb-100009
    的取值之后，接下来做误差补偿器中参数C的辨识：
    根据式(5)的结构特点，将参数C重新定义为：
    Figure PCTCN2016109890-appb-100010
    其中，d和f表示待辨识参数，ε1和ε2表示可调参数；
    将输入量Vo分别取值：
    Figure PCTCN2016109890-appb-100011
    从而得到一组检波反馈取值：
    Figure PCTCN2016109890-appb-100012
    通过调节参数d和f，使
    Figure PCTCN2016109890-appb-100013
    实现参数辨识，“≈”表示当误差小于既定范围停止搜索，参数d和f的搜寻方法如下：
    经级数展开得到以下关系式：
    Figure PCTCN2016109890-appb-100014
    进而得到
    Figure PCTCN2016109890-appb-100015
    其中δ是步进参数，这里δGeΛ取值数值小的正数，以保证式(10)的收敛；
    步骤(23)，在得到B和C的参数估计值之后，将B的数值送入原点偏移补偿器的DAC器件中，实现对载波泄露的抑制；将C的数值代入基带信号发生模块的数值计算过程中，将I/Q两路的数字值通过直接的加和乘运算实现预失真，从而完成对全工作波段的调制误差的整体修正。
  3. 如权利要求2所述的一种基于双级预失真的超宽带复杂格式矢量调制误差修正方法，其特征在于，在所述多参数可编辑嵌入式扫频激励信号模型中，采用遍历搜索的方式完成参数调节。
PCT/CN2016/109890 2016-11-15 2016-12-14 基于双级预失真的超宽带复杂格式矢量调制误差修正方法 WO2018090417A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201611036502.5A CN106534038B (zh) 2016-11-15 2016-11-15 基于双级预失真的超宽带复杂格式矢量调制误差修正方法
CN2016110365025 2016-11-15

Publications (1)

Publication Number Publication Date
WO2018090417A1 true WO2018090417A1 (zh) 2018-05-24

Family

ID=58356263

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/109890 WO2018090417A1 (zh) 2016-11-15 2016-12-14 基于双级预失真的超宽带复杂格式矢量调制误差修正方法

Country Status (2)

Country Link
CN (1) CN106534038B (zh)
WO (1) WO2018090417A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109995335A (zh) * 2019-04-30 2019-07-09 厦门理工学院 一种可高效硬件实现的分段数字预失真方法
CN113037168A (zh) * 2019-12-25 2021-06-25 大族激光科技产业集团股份有限公司 一种永磁同步电机辨识电感电阻参数的方法
CN113468842A (zh) * 2021-08-16 2021-10-01 重庆大学 一种基于矢量量化的宽带数字预失真算法
CN114465677A (zh) * 2020-11-10 2022-05-10 晶晨半导体(上海)股份有限公司 校正宽带系统i/q不平衡的方法、宽带系统和介质
CN117172163A (zh) * 2023-08-15 2023-12-05 重庆西南集成电路设计有限责任公司 幅相控制电路的幅相二维优化方法、系统、介质及电子设备

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109257311B (zh) * 2017-07-14 2021-04-16 北京中科晶上科技股份有限公司 确定误差矢量幅度的方法及系统
CN110212997B (zh) * 2019-06-04 2022-02-22 成都德芯数字科技股份有限公司 一种调制误码率的获取方法及装置
CN111162782B (zh) * 2019-12-31 2021-09-21 京信网络系统股份有限公司 直流校准方法、系统、设备和存储介质
CN113259286B (zh) * 2021-07-07 2021-10-15 易兆微电子(杭州)股份有限公司 一种发射机及其iq失衡和直流偏置的补偿方法和装置

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101694986A (zh) * 2009-11-02 2010-04-14 中国电子科技集团公司第四十一研究所 一种基于模拟基带修正的矢量调制误差补偿方法
CN103023829A (zh) * 2012-11-26 2013-04-03 电信科学技术研究院 一种调制精度的估计方法及装置
CN104935535A (zh) * 2015-04-17 2015-09-23 中国电子科技集团公司第四十一研究所 一种自适应大调制带宽i/q调制误差数字补偿方法及系统

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7154939B2 (en) * 2002-02-05 2006-12-26 Motorola, Inc. Adaptive vector modulation method and system
JP5572662B2 (ja) * 2011-05-23 2014-08-13 アンリツ株式会社 直交変調器及び信号発生装置並びに直交変調方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101694986A (zh) * 2009-11-02 2010-04-14 中国电子科技集团公司第四十一研究所 一种基于模拟基带修正的矢量调制误差补偿方法
CN103023829A (zh) * 2012-11-26 2013-04-03 电信科学技术研究院 一种调制精度的估计方法及装置
CN104935535A (zh) * 2015-04-17 2015-09-23 中国电子科技集团公司第四十一研究所 一种自适应大调制带宽i/q调制误差数字补偿方法及系统

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
XIAO PENG ET AL.: "Calibration and Compensation of Wideband Digital Vector Modulation Channel", MODERN ELECTRONIC TECHNIQUE, no. 7, 31 July 2009 (2009-07-31) *
ZHOU JUNJIE ET AL.: "A vector Modulation Error Compensation method of I/Q Modulator", CHINESE JOURNAL OF SCIENTIFIC INSTRUMENT, vol. 31, no. 8, 31 August 2010 (2010-08-31), pages 299 - 301 *

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109995335A (zh) * 2019-04-30 2019-07-09 厦门理工学院 一种可高效硬件实现的分段数字预失真方法
CN109995335B (zh) * 2019-04-30 2022-11-15 厦门理工学院 一种可高效硬件实现的分段数字预失真方法
CN113037168A (zh) * 2019-12-25 2021-06-25 大族激光科技产业集团股份有限公司 一种永磁同步电机辨识电感电阻参数的方法
CN113037168B (zh) * 2019-12-25 2023-05-09 大族激光科技产业集团股份有限公司 一种永磁同步电机辨识电感电阻参数的方法
CN114465677A (zh) * 2020-11-10 2022-05-10 晶晨半导体(上海)股份有限公司 校正宽带系统i/q不平衡的方法、宽带系统和介质
CN113468842A (zh) * 2021-08-16 2021-10-01 重庆大学 一种基于矢量量化的宽带数字预失真算法
CN117172163A (zh) * 2023-08-15 2023-12-05 重庆西南集成电路设计有限责任公司 幅相控制电路的幅相二维优化方法、系统、介质及电子设备
CN117172163B (zh) * 2023-08-15 2024-04-12 重庆西南集成电路设计有限责任公司 幅相控制电路的幅相二维优化方法、系统、介质及电子设备

Also Published As

Publication number Publication date
CN106534038A (zh) 2017-03-22
CN106534038B (zh) 2020-02-14

Similar Documents

Publication Publication Date Title
WO2018090417A1 (zh) 基于双级预失真的超宽带复杂格式矢量调制误差修正方法
US6771709B2 (en) System and method for direct transmitter self-calibration
JP5432374B2 (ja) 無線通信のための高効率送信機
JP5275973B2 (ja) 複合増幅器の信号歪みを補償するための方法
JP4801079B2 (ja) 任意波形のプリディストーション・テーブルの生成
US20100074367A1 (en) Adaptive combiner error calibration algorithms in all-digital outphasing transmitter
US9568521B2 (en) Distortion estimation apparatus and method
KR101255561B1 (ko) 멱급수형 디지털 프리디스토터와 그 왜곡 보상 제어 방법
US8953711B2 (en) Configurable pre-emphasis component for transmission circuitry
EP1345319A2 (en) Method and system for reducing non-linearities
US11025339B2 (en) Method for compensating channel distortions by pre-distortion of Mach-Zehnder modulators, based on symmetric imbalance
TWI442697B (zh) 射頻功率放大器之前饋線性化
JP6340207B2 (ja) 非線形歪み検出装置及び歪み補償電力増幅器
CN109740225A (zh) 一种用于宽带行波管计算评估的方法
Gao Linearization of Wideband Wi-Fi Power Amplifiers Using RF Analog Memory Predistortion
CN113259286A (zh) 一种发射机及其iq失衡和直流偏置的补偿方法和装置
Li et al. Nonideal effects of reconstruction filter and I/Q imbalance in digital predistortion
TWI810056B (zh) Rf通信組件執行的方法和rf通信組件
Zhao et al. Joint compensation of power amplifier nonlinearity and I/Q impairments for broadband and dual-band transmitters
CN112702148B (zh) 基于时域波形误差负反馈实现信号evm校正的方法、系统、装置、处理器及其存储介质
US8989250B1 (en) Method and apparatus for equalizing distortion in a digital pre-distortion observation path
Yang et al. Calibration of transmitter IQ imbalance with 1-bit feedback
CN113114122B (zh) 改进rascal算法数字预失真设计方法、系统及应用
Madhuwantha et al. Novel Calibration Technique for Wideband Transmitters using Constellation Mapping
Zhu et al. An iterative calibration technique for LINC transmitter

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16921518

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16921518

Country of ref document: EP

Kind code of ref document: A1