CN111147027B - Digital predistortion method and device based on broadband wireless communication system - Google Patents
Digital predistortion method and device based on broadband wireless communication system Download PDFInfo
- Publication number
- CN111147027B CN111147027B CN202010039064.8A CN202010039064A CN111147027B CN 111147027 B CN111147027 B CN 111147027B CN 202010039064 A CN202010039064 A CN 202010039064A CN 111147027 B CN111147027 B CN 111147027B
- Authority
- CN
- China
- Prior art keywords
- tone signal
- power amplifier
- training
- adjusting
- sequence
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
- H03F1/32—Modifications of amplifiers to reduce non-linear distortion
- H03F1/3241—Modifications of amplifiers to reduce non-linear distortion using predistortion circuits
- H03F1/3258—Modifications of amplifiers to reduce non-linear distortion using predistortion circuits based on polynomial terms
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
- H03F1/32—Modifications of amplifiers to reduce non-linear distortion
- H03F1/3241—Modifications of amplifiers to reduce non-linear distortion using predistortion circuits
- H03F1/3264—Modifications of amplifiers to reduce non-linear distortion using predistortion circuits in audio amplifiers
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
- H03F1/32—Modifications of amplifiers to reduce non-linear distortion
- H03F1/3241—Modifications of amplifiers to reduce non-linear distortion using predistortion circuits
- H03F1/3282—Acting on the phase and the amplitude of the input signal
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/38—Synchronous or start-stop systems, e.g. for Baudot code
- H04L25/40—Transmitting circuits; Receiving circuits
- H04L25/49—Transmitting circuits; Receiving circuits using code conversion at the transmitter; using predistortion; using insertion of idle bits for obtaining a desired frequency spectrum; using three or more amplitude levels ; Baseband coding techniques specific to data transmission systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Power Engineering (AREA)
- Nonlinear Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Algebra (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Amplifiers (AREA)
Abstract
The invention discloses a method and a device for digital predistortion based on a broadband wireless communication system, which comprises the following steps: generating a training sequence, wherein a gap is reserved between two adjacent repeated bursts in the training sequence, and the repeated bursts comprise a synchronous sequence and a single tone signal; determining the start position of the single tone signal of each repeated burst according to the synchronization sequence; fixing the frequency of a single-tone signal of a training sequence, adjusting the power of the single-tone signal, training to obtain the nonlinear distortion characteristic of the power amplifier, and adjusting the nonlinear predistortion parameters in the digital predistorter; and fixing the power of the single-tone signal of the training sequence in a linear interval of the power amplifier, adjusting the frequency of the single-tone signal, training to obtain the memory distortion characteristic of the power amplifier, and adjusting the memory predistortion parameters in the digital predistorter. The invention only needs to train the parameters in the predistorter in the initial stage, so that the adaptive algorithm is degenerated into a training algorithm.
Description
Technical Field
The invention relates to the technical field of digital predistortion, in particular to a method and a device for digital predistortion based on a broadband wireless communication system.
Background
With the development of wireless communication technology, limited spectrum resources are required to carry larger and larger data, so that the spectrum resources are increasingly in short supply. In a wireless communication system, high-order modulation technology such as Quadrature Amplitude Modulation (QAM) and broadband transmission technology such as Orthogonal Frequency Division Multiplexing (OFDM) are combined, so that the spectrum utilization rate is improved.
OFDM is also widely known as one of the key technologies in Fourth Generation Mobile Communication (4G) as an important technology in Fifth Generation Mobile Communication (5G). However, the Peak-to-Average Power Ratio (PAPR) of the OFDM signal is high, and when the OFDM signal passes through a Power Amplifier (PA), signal distortion is more easily caused by the nonlinear characteristic of the Power Amplifier. Nonlinear distortion can be divided into in-band signal distortion and out-of-band spectral spreading. In-band signal distortion can bring amplitude distortion and phase distortion of signals; the out-of-band spectrum spreading causes a large Adjacent Channel Leakage Ratio (ACLR), which disturbs the Adjacent channels and reduces the reliability of the whole system. In addition, in the broadband wireless communication system, the broadband signal is transmitted, so the memory distortion is not negligible. Most of the traditional power amplifier linearization schemes are only suitable for narrow-band systems, or the power amplifier efficiency is sacrificed to achieve linearized input and output. Therefore, the OFDM technology is better used in the broadband wireless communication system, and the nonlinear distortion and the memory distortion of the broadband power amplifier must be simultaneously processed.
1) Distortion characteristics of power amplifier
a) Power amplifier nonlinear distortion
In the power amplifier, due to the existence of the nonlinear device, a signal passes through the power amplifier to generate nonlinear distortion. Non-linear distortion is manifested in the time domain in that when the amplitude of the input signal reaches a certain degree, the amplitude of the output signal and the amplitude of the input signal will no longer be in a linear relationship. Which in the frequency domain, appears as spectral regrowth, i.e. a new spectrum is generated in the output signal. If the regenerated spectrum is within the signal bandwidth, the regenerated spectrum can generate in-band interference on the signal; if the regenerated spectrum is outside the signal bandwidth, the regenerated spectrum may interfere with the near channel.
b) Power amplifier memory distortion
The actual power amplifier is influenced by temperature, bias voltage, voltage change and self aging during operation, so that the actual power amplifier has more or less memory characteristics and generates memory distortion. The memory distortion is shown in a time domain that the output of a power amplifier at a certain moment is not only related to the current input but also related to the historical input. The memory distortion is represented in a frequency domain as different amplification factors of signals with different input frequencies by a power amplifier, namely frequency selective amplification.
c) Power amplifier model
In the case where the signal bandwidth is sufficiently small relative to the channel bandwidth, the signal passes through a power amplifier with negligible memory distortion. However, in the broadband wireless communication system, the bandwidth of the OFDM modulated signal is large, so that the memory distortion of the power amplifier cannot be ignored. The signal has nonlinear distortion and memory distortion through the power amplifier. In addition, in an actual system, a certain degree of memory characteristics exist, so that it is necessary to consider a power amplifier model with memory.
The following are several typical memorable power amplifier models.
i. Polynomial model
In the polynomial model with memory power amplifier, the relation between input and output,
wherein x (n) is PA input, y (n) is PA output, the order of K polynomial, M is the depth of power amplifier memory, akmIs the coefficient corresponding to each term of the polynomial.
Wiener model
Wiener (H-N) model: a non-linear system showing a linear Filter (FIR) with memory connected in series with a non-memory; the schematic diagram of the Wiener model is shown in fig. 1, wherein H denotes a linear filter FIR and N denotes a nonlinear system.
Hammerstein model
Hammerstein (N-H) model: a memoryless nonlinear system is connected with a linear Filter (FIR) with memorability in series; the schematic diagram of the Hammerstein model is shown in FIG. 2, where H denotes the linear filter FIR and N denotes the nonlinear system.
2) Digital predistortion technique
In order to solve the problem of nonlinear distortion of the power amplifier, domestic and foreign scholars propose various different power amplifier linearization methods through research. The final purpose of the methods is to improve the power efficiency of the power amplifier as much as possible on the basis of keeping the linear input/output of the power amplifier. The power back-off technique is applied at the earliest, and negative feedback technique, feed-forward technique and predistortion technique which is applied more generally at present are proposed later.
Among the above conventional linearization methods, the predistortion technique is a relatively economical and practical one. With the rapid development of the DSP technology, the Digital predistortion technology is combined with the DSP to form adaptive Digital Pre-Distortion (DPD).
As shown in the basic block diagram of the baseband adaptive DPD shown in fig. 3, an input signal x (n) is subjected to distortion compensation by a predistorter to obtain y (n); y (n) is subjected to D/A conversion, an out-of-band signal is filtered by a filter, then amplified by a power amplifier after up-conversion quadrature modulation, and finally z (n) is output; taking the coupling data of z (n) as the data of a return branch, and obtaining u (n) through down-conversion quadrature modulation, filtering and A/D conversion; and taking the input x (n) loop delay signal and feedback processed data u (n) as input data of an adaptive algorithm so as to update parameters in the digital filter.
In the actual design of hardware, two points need to be noted: one is that a part of signals of x (n) is used as the input of the self-adaptive algorithm, loop estimation delay needs to be added, and the loop estimation delay needs to correspond to the signals u (n); but orthogonal modulation of up-down frequency conversion, and the local oscillators of the up-down frequency conversion and the orthogonal modulation have to be designed to be consistent, otherwise, the synchronization cannot be realized.
The digital signals are stored in the predistorter in the DPD, and when the power amplifier works, the changes of the environments such as temperature, humidity and the like or the changes of input signals can cause the changes of nonlinearity and memory of the power amplifier, so that an adaptive algorithm needs to be added to modify the parameters in the predistorter in real time, and the reliable operation of the whole system can be guaranteed.
When the power amplifier distortion affects the phase distortion, the phase factor needs to be considered in the predistortion algorithm; at this time, the input signal x (n) and the feedback processed data u (n) must be synchronized, so that the phase-related training can be performed to obtain the parameters of the predistorter. In the existing baseband predistortion technology, generally, before an input signal x (n) enters a training algorithm, loop delay estimation is performed, and the loop delay estimation estimates the delay of a loop, so that the input signal x (n) and feedback-processed data u (n) are ensured to be synchronous.
Disclosure of Invention
In view of the above-mentioned drawbacks, the present invention provides a method and apparatus for digital predistortion based on a broadband wireless communication system.
The invention discloses a digital predistortion method based on a broadband wireless communication system, which comprises the following steps:
generating a training sequence; wherein the training sequence comprises a plurality of repeated bursts, a gap is left between two adjacent repeated bursts, and each repeated burst comprises a synchronization sequence and a tone signal;
performing feedback processing on each single-tone signal in the training sequence to obtain a feedback processing signal;
determining the starting position of the single tone signal of each repeated burst according to the synchronous sequence of the training sequence;
fixing the frequency of the single-tone signal of the training sequence, adjusting the power of the single-tone signal, training to obtain the nonlinear distortion characteristic of the power amplifier, and adjusting the nonlinear predistortion parameters in the digital predistorter;
and fixing the power of the single-tone signal of the training sequence in a linear interval of the power amplifier, adjusting the frequency of the single-tone signal, training to obtain the memory distortion characteristic of the power amplifier, and adjusting the memory predistortion parameters in the digital predistorter.
As a further improvement of the present invention, there is a correlation between the synchronization sequences;
from the sync sequence, a single tone signal start position for each repeated burst is determined using a cross-correlation method.
As a further development of the invention, the synchronization sequence is a pseudo-random sequence.
As a further improvement of the present invention, the frequency of the single-tone signal in the training sequence is fixed, the power of the single-tone signal is adjusted, the nonlinear distortion characteristic of the power amplifier is obtained by training, and the nonlinear predistortion parameter in the digital predistorter is adjusted; the method comprises the following steps:
fixing the single tone signal frequency in the training sequence;
according to the input power range of the power amplifier, adjusting the power of the single-tone signal and traversing from small to large;
establishing an amplitude value mapping relation before and after the same single-tone signal feedback processing, and training to obtain the memorability distortion characteristic of the power amplifier;
and adjusting nonlinear predistortion parameters in the digital predistorter by adopting a lookup table method according to the memorability distortion characteristic of the power amplifier.
As a further improvement of the present invention, the power of the single-tone signal of the fixed training sequence is within a linear interval of the power amplifier, the frequency of the single-tone signal is adjusted, the memory distortion characteristic of the power amplifier is obtained by training, and the memory predistortion parameters in the digital predistorter are adjusted; the method comprises the following steps:
fixing the single tone signal power of the training sequence in a linear interval of a power amplifier;
adjusting the frequency of the single-tone signal and traversing from small to large;
training the amplitude and the phase before and after the feedback processing of the same single-tone signal by adopting an FIR filter method to obtain the memory distortion characteristic of the power amplifier;
and adjusting the memory predistortion parameters in the digital predistorter according to the memory distortion characteristics of the power amplifier.
The invention also discloses a device for digital predistortion based on the broadband wireless communication system, which comprises:
a training sequence generator to:
generating a training sequence; wherein the training sequence comprises a plurality of repeated bursts, a gap is left between two adjacent repeated bursts, and each repeated burst comprises a synchronization sequence and a tone signal;
a feedback processing module to:
performing feedback processing on each single-tone signal in the training sequence to obtain a feedback processing signal;
a training module to:
determining the starting position of the single tone signal of each repeated burst according to the synchronous sequence of the training sequence;
fixing the frequency of the single-tone signal of the training sequence, adjusting the power of the single-tone signal, training to obtain the nonlinear distortion characteristic of the power amplifier, and adjusting the nonlinear predistortion parameters in the digital predistorter;
and fixing the power of the single-tone signal of the training sequence in a linear interval of the power amplifier, adjusting the frequency of the single-tone signal, training to obtain the memory distortion characteristic of the power amplifier, and adjusting the memory predistortion parameters in the digital predistorter.
As a further improvement of the present invention, there is a correlation between the synchronization sequences;
from the sync sequence, a single tone signal start position for each repeated burst is determined using a cross-correlation method.
As a further development of the invention, the synchronization sequence is a pseudo-random sequence.
As a further improvement of the present invention, the frequency of the single-tone signal in the training sequence is fixed, the power of the single-tone signal is adjusted, the nonlinear distortion characteristic of the power amplifier is obtained by training, and the nonlinear predistortion parameter in the digital predistorter is adjusted; the method comprises the following steps:
fixing the single tone signal frequency in the training sequence;
according to the input power range of the power amplifier, adjusting the power of the single-tone signal and traversing from small to large;
establishing an amplitude value mapping relation before and after the same single-tone signal feedback processing, and training to obtain the memorability distortion characteristic of the power amplifier;
and adjusting nonlinear predistortion parameters in the digital predistorter by adopting a lookup table method according to the memorability distortion characteristic of the power amplifier.
As a further improvement of the present invention, the power of the single-tone signal of the fixed training sequence is within a linear interval of the power amplifier, the frequency of the single-tone signal is adjusted, the memory distortion characteristic of the power amplifier is obtained by training, and the memory predistortion parameters in the digital predistorter are adjusted; the method comprises the following steps:
fixing the single tone signal power of the training sequence in a linear interval of a power amplifier;
adjusting the frequency of the single-tone signal and traversing from small to large;
training the amplitude and the phase before and after the feedback processing of the same single-tone signal by adopting an FIR filter method to obtain the memory distortion characteristic of the power amplifier;
and adjusting the memory predistortion parameters in the digital predistorter according to the memory distortion characteristics of the power amplifier.
Compared with the prior art, the invention has the beneficial effects that:
the digital predistortion method and the device only need to train parameters in the predistorter at an initial stage to ensure that the adaptive algorithm is degenerated into a training algorithm; the changes of the nonlinearity and the memorability of the power amplifier do not need to be considered, and the parameters in the predistorter do not need to be modified in real time; therefore, the loop time delay estimation can be avoided, the predistortion algorithm can be simplified, and the accuracy is higher.
Drawings
FIG. 1 is a schematic diagram of a Wiener model; FIG. 2 is a schematic diagram of the Hammerstein model;
FIG. 3 is a basic block diagram of the baseband adaptive DPD;
FIG. 4 is a diagram of a training sequence according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a digital predistortion method and apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
The invention is described in further detail below with reference to the attached drawing figures:
as shown in fig. 5, the present invention provides a method for digital predistortion based on a broadband wireless communication system, which can avoid using loop delay estimation, but generate a set of specific training sequences, so that a training algorithm can obtain the timing sequence of an input signal x (n), so as to ensure that the input signal x (n) and a feedback signal u (n) are synchronized, and perform phase-related training to obtain parameters of a predistorter; the method specifically comprises the following steps:
step 1, generating a training sequence; wherein:
the training sequence can be generated in advance and stored in a memory; temporary generation is avoided, and time is saved;
as shown in fig. 4, the training sequence includes a plurality of repeated bursts, and a gap is left between two adjacent repeated bursts, so as to allow enough time for the training algorithm to perform training;
each repetitive burst comprises a sync sequence and a mono signal, the sync sequence needs to have good correlation, for example, a pseudo-random sequence (1/-1); based on the correlation of the sync sequences, the single tone signal start position of each repeated burst is determined using a cross-correlation method.
the feedback processing of the single-tone signal comprises:
D/A conversion is carried out on the single-tone signal through a DAC, an out-of-band signal is filtered through a filter, then the single-tone signal is amplified through a power amplifier after being modulated through RF, and z (n) is finally output; taking the coupling data of z (n) as the data of a return branch, and obtaining a feedback processing signal u (n) through RF demodulation, filtering and A/D conversion; due to the nonlinear distortion and the memory distortion of the power amplifier, the amplitude, the phase and the like of the single-tone signal before and after the feedback processing are changed; the invention utilizes the parameters of the single-tone signal before and after the feedback processing to train, and obtains the nonlinear distortion characteristic and the memorability distortion characteristic of the power amplifier, thereby adjusting the nonlinear predistortion parameter and the memorability predistortion parameter in the digital predistorter.
And 3, finding the start position of the single tone signal in each repeated burst by using a cross-correlation method by using the synchronous sequence in each repeated burst in the training sequence by using a training algorithm.
Step 4, fixing the frequency of the single-tone signal of the training sequence by a training algorithm, adjusting the power of the single-tone signal, training to obtain the nonlinear distortion characteristic of the power amplifier, and adjusting the nonlinear predistortion parameter in the digital predistorter; wherein, specifically include:
1) fixing the frequency of the single tone signal in the training sequence; the single-tone signal with fixed frequency is not influenced by the memory distortion of the power amplifier, so the nonlinear distortion characteristic is considered independently;
2) adjusting the power of the single-tone signal according to the input power range of the power amplifier and traversing from small to large;
3) establishing an amplitude value mapping relation before and after the same single-tone signal feedback processing, and training to obtain the memorability distortion characteristic of the power amplifier;
4) and adjusting nonlinear predistortion parameters in the digital predistorter by adopting a lookup table method according to the memorability distortion characteristic of the power amplifier.
Step 5, the training algorithm fixes the single-tone signal power of the training sequence in the linear interval of the power amplifier, adjusts the frequency of the single-tone signal, trains to obtain the memory distortion characteristic of the power amplifier, and adjusts the memory predistortion parameters in the digital predistorter; wherein, specifically include:
1) the power of the single tone signal of the fixed training sequence is in a linear interval of the power amplifier;
2) adjusting the frequency of the single-tone signal and traversing from small to large;
3) training the amplitude and the phase before and after the feedback processing of the same single-tone signal by adopting an FIR filter method to obtain the memory distortion characteristic of the power amplifier;
4) and adjusting the memory predistortion parameters in the digital predistorter according to the memory predistortion characteristics of the power amplifier.
For step 4 and step 5, a specific embodiment description is set forth below by taking a 40MHz bandwidth OFDM system in WIFI 11n as an example:
assuming that the sampling rate of the DAC/ADC is 40M in a 40MHz bandwidth OFDM system in WIFI 11n, the number of sampling points of the single-tone signal is at least 128 points to ensure that the single-tone signal has at least one complete period. The single-tone signal is s (t) ═ a × exp (j × 2 pi × fc × t). Synchronous training also uses a 128-point pseudorandom sequence.
1. Training nonlinear predistortion parameters:
in this embodiment, the non-linear pre-stageThe distortion adopts a lookup table method. Assume a nonlinear predistorted input signal is Sin＝xa*exp(j*xp) Then the nonlinear predistortion output signal is Spd＝Axa*xa*exp(j*(xp+Pxa)). In this embodiment, the power amplifier nonlinear distortion phase distortion is assumed to be 0, so P can be setxa＝0。
The tone signal frequency is fixed to fc 312.5 KHz.
In the training sequence, the amplitude of each tone signal is traversed from small to large.
Assuming that the maximum input amplitude of the power amplifier is 1, we set the amplitude traversal range to
In the present embodiment, N is 100.
Correspondingly, in the training algorithm in the feedback loop, the detected amplitude values are [ x (0), x (1), …, x (N-1) ]. Then the non-linear predistortion look-up table is set as follows:
2. training memory predistortion parameters:
a) in this embodiment, the memorable predistortion is implemented using an FIR filter method with a coefficient dw. If the coefficient of the FIR filter of the memory distortion model of the power amplifier is w, dw needs to ensure the following equation in order to eliminate the memory distortion of the power amplifier:
b) And in the FIR filter coefficient training, the training signal must ensure that the amplitude of the single tone signal is small enough, thereby ensuring that the amplitude of the input signal of the power amplifier is in the linear working interval of the power amplifier.
c) The frequency of each tone signal is traversed from small to large in the training sequence.
d) In this embodiment, the wideband bandwidth is 40M, and we set the sweep frequency fc ═ Δ f × M, M ∈ [ -64, -63, …, -1,1,2, …,63], where Δ f ═ 312.5KHz is the subcarrier spacing.
e) The influence of the memory distortion on the amplitude and phase for each frequency point is as follows,
T(t)＝S(t)*Hk*exp(j*Qk)＝P*Hk*exp(j*2*π*Δf*k*t+j*Qk) Where k is the subcarrier index.
f) Define gamma, ak＝Hk*exp(j*Qk) Then the frequency domain response vector w is:
w＝[γ-64,γ-63,…,γ-1,1,γ1,…,γ63]
g) then the memorability predistortion FIR filter coefficient is
Where IFFT128(.) refers to an inverse fast fourier transform of 128 lengths.
As shown in fig. 5, the present invention provides a digital predistortion apparatus based on a broadband wireless communication system, which can avoid using loop delay estimation, but generate a set of specific training sequences, so that a training algorithm can obtain the timing sequence of an input signal x (n), so as to ensure that the input signal x (n) and a feedback signal u (n) are synchronized, and perform phase-related training to obtain parameters of a predistorter; the method specifically comprises the following steps:
a training sequence generator for generating a training sequence; wherein:
the training sequence can be generated in advance and stored in a memory; temporary generation is avoided, and time is saved;
as shown in fig. 4, the training sequence includes a plurality of repeated bursts, and a gap is left between two adjacent repeated bursts, so as to allow enough time for the training algorithm to perform training;
each repetitive burst comprises a sync sequence and a mono signal, the sync sequence needs to have good correlation, for example, a pseudo-random sequence (1/-1); based on the correlation of the sync sequences, the single tone signal start position of each repeated burst is determined using a cross-correlation method.
The feedback processing module is used for carrying out feedback processing on each single-tone signal in the training sequence to obtain a feedback processing signal; wherein:
the feedback processing module comprises a DAC (digital-to-analog converter), an RF (radio frequency) modulation module, a power amplifier, a coupler, an RF demodulation module and an ADC (analog-to-digital converter), the single-tone signal is subjected to D/A (digital-to-analog) conversion through the DAC, an out-of-band signal is filtered through a filter, then the out-of-band signal is amplified through the power amplifier after being subjected to RF modulation, and z (n; taking the coupling data of z (n) as the data of a return branch, and obtaining a feedback processing signal u (n) through RF demodulation, filtering and A/D conversion; due to the nonlinear distortion and the memory distortion of the power amplifier, the amplitude, the phase and the like of the single-tone signal before and after the feedback processing are changed; the invention utilizes the parameters of the single-tone signal before and after the feedback processing to train, and obtains the nonlinear distortion characteristic and the memorability distortion characteristic of the power amplifier, thereby adjusting the nonlinear predistortion parameter and the memorability predistortion parameter in the digital predistorter.
A training module to:
the training algorithm uses the sync sequence in each repeated burst in the training sequence to find the start position of the tone signal in each repeated burst using a cross-correlation method.
The training algorithm fixes the single-tone signal frequency of the training sequence, adjusts the power of the single-tone signal, trains to obtain the nonlinear distortion characteristic of the power amplifier, and adjusts the nonlinear predistortion parameters in the digital predistorter; wherein, specifically include:
1) fixing the frequency of the single tone signal in the training sequence; the single-tone signal with fixed frequency is not influenced by the memory distortion of the power amplifier, so the nonlinear distortion characteristic is considered independently;
2) adjusting the power of the single-tone signal according to the input power range of the power amplifier and traversing from small to large;
3) establishing an amplitude value mapping relation before and after the same single-tone signal feedback processing, and training to obtain the memorability distortion characteristic of the power amplifier;
4) and adjusting nonlinear predistortion parameters in the digital predistorter by adopting a lookup table method according to the memorability distortion characteristic of the power amplifier.
The training algorithm fixes the single-tone signal power of the training sequence in the linear interval of the power amplifier, adjusts the frequency of the single-tone signal, trains to obtain the memory distortion characteristic of the power amplifier, and adjusts the memory predistortion parameters in the digital predistorter; wherein, specifically include:
1) the power of the single tone signal of the fixed training sequence is in a linear interval of the power amplifier;
2) adjusting the frequency of the single-tone signal and traversing from small to large;
3) training the amplitude and the phase before and after the feedback processing of the same single-tone signal by adopting an FIR filter method to obtain the memory distortion characteristic of the power amplifier;
4) and adjusting the memory predistortion parameters in the digital predistorter according to the memory predistortion characteristics of the power amplifier.
For the training module, a specific embodiment description is set forth below by taking a 40MHz bandwidth OFDM system in WIFI 11n as an example:
assuming that the sampling rate of the DAC/ADC is 40M in a 40MHz bandwidth OFDM system in WIFI 11n, the number of sampling points of the single-tone signal is at least 128 points to ensure that the single-tone signal has at least one complete period. The single-tone signal is s (t) ═ a × exp (j × 2 pi × fc × t). Synchronous training also uses a 128-point pseudorandom sequence.
1. Training nonlinear predistortion parameters:
in this embodiment, the nonlinear predistortion uses a look-up table approach. Assume a nonlinear predistorted input signal is Sin＝xa*exp(j*xp) Then the nonlinear predistortion output signal is Spd＝Axa*xa*exp(j*(xp+Pxa)). In the bookIn the embodiment, the nonlinear distortion phase distortion of the power amplifier is assumed to be 0, so that P can be setxa＝0。
The tone signal frequency is fixed to fc 312.5 KHz.
In the training sequence, the amplitude of each tone signal is traversed from small to large.
Assuming that the maximum input amplitude of the power amplifier is 1, we set the amplitude traversal range to
In the present embodiment, N is 100.
Correspondingly, in the training algorithm in the feedback loop, the detected amplitude values are [ x (0), x (1), …, x (N-1) ]. Then the non-linear predistortion look-up table is set as follows:
2. training memory predistortion parameters:
a) in this embodiment, the memorable predistortion is implemented using an FIR filter method with a coefficient dw. If the coefficient of the FIR filter of the memory distortion model of the power amplifier is w, dw needs to ensure the following equation in order to eliminate the memory distortion of the power amplifier:
b) And in the FIR filter coefficient training, the training signal must ensure that the amplitude of the single tone signal is small enough, thereby ensuring that the amplitude of the input signal of the power amplifier is in the linear working interval of the power amplifier.
c) The frequency of each tone signal is traversed from small to large in the training sequence.
d) In this embodiment, the wideband bandwidth is 40M, and we set the sweep frequency fc ═ Δ f × M, M ∈ [ -64, -63, …, -1,1,2, …,63], where Δ f ═ 312.5KHz is the subcarrier spacing.
e) The influence of the memory distortion on the amplitude and phase for each frequency point is as follows,
T(t)＝S(t)*Hk*exp(j*Qk)＝P*Hk*exp(j*2*π*Δf*k*t+j*Qk) Where k is the subcarrier index.
f) Define gamma, ak＝Hk*exp(j*Qk) Then the frequency domain response vector w is:
w＝[γ-64,γ-63,…,γ-1,1,γ1,…,γ63]
g) then the memorability predistortion FIR filter coefficient is
Where IFFT128(.) refers to an inverse fast fourier transform of 128 lengths.
The invention has the advantages that:
the digital predistortion method and the device only need to train parameters in the predistorter at an initial stage to ensure that the adaptive algorithm is degenerated into a training algorithm; the changes of the nonlinearity and the memorability of the power amplifier do not need to be considered, and the parameters in the predistorter do not need to be modified in real time; therefore, the loop time delay estimation can be avoided, the predistortion algorithm can be simplified, and the accuracy is higher.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. A method for digital predistortion based on a wideband wireless communication system, comprising:
generating a training sequence; wherein the training sequence comprises a plurality of repeated bursts, a gap is left between two adjacent repeated bursts, and each repeated burst comprises a synchronization sequence and a tone signal;
performing feedback processing on each single-tone signal in the training sequence to obtain a feedback processing signal;
determining the starting position of the single tone signal of each repeated burst according to the synchronous sequence of the training sequence;
fixing the frequency of the single-tone signal of the training sequence, adjusting the power of the single-tone signal, training to obtain the nonlinear distortion characteristic of the power amplifier, and adjusting the nonlinear predistortion parameters in the digital predistorter;
fixing the single-tone signal power of the training sequence in a linear interval of the power amplifier, adjusting the frequency of the single-tone signal, training to obtain the memory distortion characteristic of the power amplifier, and adjusting the memory predistortion parameters in the digital predistorter;
fixing the frequency of the single-tone signal in the training sequence, adjusting the power of the single-tone signal, training to obtain the nonlinear distortion characteristic of the power amplifier, and adjusting the nonlinear predistortion parameters in the digital predistorter; the method comprises the following steps:
fixing the single tone signal frequency in the training sequence;
according to the input power range of the power amplifier, adjusting the power of the single-tone signal and traversing from small to large;
establishing an amplitude value mapping relation before and after the same single-tone signal feedback processing, and training to obtain the memorability distortion characteristic of the power amplifier;
adjusting nonlinear predistortion parameters in the digital predistorter by adopting a lookup table method according to the memorability distortion characteristic of the power amplifier;
the single-tone signal power of the fixed training sequence is in a linear interval of the power amplifier, the frequency of the single-tone signal is adjusted, the memory distortion characteristic of the power amplifier is obtained through training, and the memory predistortion parameters in the digital predistorter are adjusted; the method comprises the following steps:
fixing the single tone signal power of the training sequence in a linear interval of a power amplifier;
adjusting the frequency of the single-tone signal and traversing from small to large;
training the amplitude and the phase before and after the feedback processing of the same single-tone signal by adopting an FIR filter method to obtain the memory distortion characteristic of the power amplifier;
and adjusting the memory predistortion parameters in the digital predistorter according to the memory distortion characteristics of the power amplifier.
2. The method of digital predistortion according to claim 1, wherein there is a correlation between the synchronization sequences;
from the sync sequence, a single tone signal start position for each repeated burst is determined using a cross-correlation method.
3. The method of digital predistortion according to claim 2, wherein the synchronization sequence is a pseudo-random sequence.
4. An apparatus for digital predistortion based on a broadband wireless communication system, comprising:
a training sequence generator to:
generating a training sequence; wherein the training sequence comprises a plurality of repeated bursts, a gap is left between two adjacent repeated bursts, and each repeated burst comprises a synchronization sequence and a tone signal;
a feedback processing module to:
performing feedback processing on each single-tone signal in the training sequence to obtain a feedback processing signal;
a training module to:
determining the starting position of the single tone signal of each repeated burst according to the synchronous sequence of the training sequence;
fixing the frequency of the single-tone signal of the training sequence, adjusting the power of the single-tone signal, training to obtain the nonlinear distortion characteristic of the power amplifier, and adjusting the nonlinear predistortion parameters in the digital predistorter;
fixing the single-tone signal power of the training sequence in a linear interval of the power amplifier, adjusting the frequency of the single-tone signal, training to obtain the memory distortion characteristic of the power amplifier, and adjusting the memory predistortion parameters in the digital predistorter;
fixing the frequency of the single-tone signal in the training sequence, adjusting the power of the single-tone signal, training to obtain the nonlinear distortion characteristic of the power amplifier, and adjusting the nonlinear predistortion parameters in the digital predistorter; the method comprises the following steps:
fixing the single tone signal frequency in the training sequence;
according to the input power range of the power amplifier, adjusting the power of the single-tone signal and traversing from small to large;
establishing an amplitude value mapping relation before and after the same single-tone signal feedback processing, and training to obtain the memorability distortion characteristic of the power amplifier;
adjusting nonlinear predistortion parameters in the digital predistorter by adopting a lookup table method according to the memorability distortion characteristic of the power amplifier;
the single-tone signal power of the fixed training sequence is in a linear interval of the power amplifier, the frequency of the single-tone signal is adjusted, the memory distortion characteristic of the power amplifier is obtained through training, and the memory predistortion parameters in the digital predistorter are adjusted; the method comprises the following steps:
fixing the single tone signal power of the training sequence in a linear interval of a power amplifier;
adjusting the frequency of the single-tone signal and traversing from small to large;
training the amplitude and the phase before and after the feedback processing of the same single-tone signal by adopting an FIR filter method to obtain the memory distortion characteristic of the power amplifier;
and adjusting the memory predistortion parameters in the digital predistorter according to the memory distortion characteristics of the power amplifier.
5. The apparatus for digital predistortion according to claim 4, wherein there is a correlation between the synchronization sequences;
from the sync sequence, a single tone signal start position for each repeated burst is determined using a cross-correlation method.
6. The apparatus for digital predistortion according to claim 5, wherein the synchronization sequence is a pseudo-random sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010039064.8A CN111147027B (en) | 2020-01-14 | 2020-01-14 | Digital predistortion method and device based on broadband wireless communication system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010039064.8A CN111147027B (en) | 2020-01-14 | 2020-01-14 | Digital predistortion method and device based on broadband wireless communication system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111147027A CN111147027A (en) | 2020-05-12 |
| CN111147027B true CN111147027B (en) | 2021-02-19 |
Family
ID=70525044
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010039064.8A Active CN111147027B (en) | 2020-01-14 | 2020-01-14 | Digital predistortion method and device based on broadband wireless communication system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111147027B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023184483A1 (en) * | 2022-04-01 | 2023-10-05 | Zte Corporation | Wireless communication and sensing method and device thereof |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103326973A (en) * | 2013-07-03 | 2013-09-25 | 苏州工业园区昶辰通信科技有限公司 | Distortion characteristic estimation method predistortion device and method thereof |
| WO2015161522A1 (en) * | 2014-04-26 | 2015-10-29 | 华为技术有限公司 | Coefficient training device, method and system and base station |
| CN105229927A (en) * | 2013-05-22 | 2016-01-06 | 瑞典爱立信有限公司 | For the linearisation of the intermodulation frequency band of concurrent dual-band power amplifier |
| CN107359864A (en) * | 2017-07-25 | 2017-11-17 | 中国工程物理研究院电子工程研究所 | The adaptive agile digital pre-distortion method of frequency agility power amplifier |
-
2020
- 2020-01-14 CN CN202010039064.8A patent/CN111147027B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105229927A (en) * | 2013-05-22 | 2016-01-06 | 瑞典爱立信有限公司 | For the linearisation of the intermodulation frequency band of concurrent dual-band power amplifier |
| CN103326973A (en) * | 2013-07-03 | 2013-09-25 | 苏州工业园区昶辰通信科技有限公司 | Distortion characteristic estimation method predistortion device and method thereof |
| WO2015161522A1 (en) * | 2014-04-26 | 2015-10-29 | 华为技术有限公司 | Coefficient training device, method and system and base station |
| CN107359864A (en) * | 2017-07-25 | 2017-11-17 | 中国工程物理研究院电子工程研究所 | The adaptive agile digital pre-distortion method of frequency agility power amplifier |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111147027A (en) | 2020-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6072364A (en) | Adaptive digital predistortion for power amplifiers with real time modeling of memoryless complex gains | |
| Kang et al. | On compensating nonlinear distortions of an OFDM system using an efficient adaptive predistorter | |
| Anttila et al. | Joint mitigation of power amplifier and I/Q modulator impairments in broadband direct-conversion transmitters | |
| US7203247B2 (en) | Digital predistortion technique for WCDMA wireless communication system and method of operation thereof | |
| JP4261367B2 (en) | Method and apparatus for improving transmitter performance using baseband conversion | |
| EP1204216B1 (en) | Method and apparatus for distortion compensation of radio device | |
| US8472556B2 (en) | Amplifier device and predistortion control method | |
| US8081710B2 (en) | System and method for corrected modulation with nonlinear power amplification | |
| CN101662821B (en) | Signal processing method and communication system | |
| KR20070046779A (en) | A pre-distorter for orthogonal frequency division multiplexing systems and method of operating the same | |
| EP1949635A2 (en) | Peak-to-average power reduction | |
| Helaoui et al. | Power and efficiency enhancement of 3G multicarrier amplifiers using digital signal processing with experimental validation | |
| US8548404B2 (en) | Linearization of power amplifiers through digital in-band predistortion followed by analog predistortion | |
| JP4905551B2 (en) | OFDM modulated wave output device and distortion compensation method | |
| Jeong et al. | Wideband envelope tracking power amplifier with reduced bandwidth power supply waveform | |
| JP3451947B2 (en) | OFDM modulator | |
| Rahati Belabad et al. | A novel generalized parallel two-box structure for behavior modeling and digital predistortion of RF power amplifiers at LTE applications | |
| CN111147027B (en) | Digital predistortion method and device based on broadband wireless communication system | |
| Chiu et al. | Predistorter based on frequency domain estimation for compensation of nonlinear distortion in OFDM systems | |
| Rodrigues et al. | IMD reduction with SLM and PTS to improve the error-probability performance of nonlinearly distorted OFDM signals | |
| CN111988254B (en) | Low-complexity peak-to-average ratio compression and predistortion joint optimization method | |
| WO2007000923A1 (en) | Multicarrier transmitting apparatus and multicarrier transmitting method | |
| Molina et al. | Generation of optimum excitation waveforms for mobile radio channel sounding | |
| Abdelaziz et al. | PAPR reduction and digital predistortion for non-contiguous waveforms with well-localized spectrum | |
| Zhu et al. | A blind AM/PM estimation method for power amplifier linearization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |