CN109889166A - A kind of concurrent double frequency-band digital pre-distortion method of single feedback loop based on time-interleaved sampling - Google Patents
A kind of concurrent double frequency-band digital pre-distortion method of single feedback loop based on time-interleaved sampling Download PDFInfo
- Publication number
- CN109889166A CN109889166A CN201910184466.4A CN201910184466A CN109889166A CN 109889166 A CN109889166 A CN 109889166A CN 201910184466 A CN201910184466 A CN 201910184466A CN 109889166 A CN109889166 A CN 109889166A
- Authority
- CN
- China
- Prior art keywords
- signal
- matrix
- band
- double frequency
- frequency
- 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.)
- Granted
Links
Landscapes
- Amplifiers (AREA)
Abstract
The invention discloses a kind of concurrent double frequency-band digital pre-distortion methods of single feedback loop based on time-interleaved sampling, belong to wireless communication field.Calculated concurrent double frequency-band input signal x (n) is successively passed through into frequency mixer first, after digital analog converter and power amplifier, amplified signal is acquired in the feedback loop, after down coversion and low-pass filter, obtains filtered analog signal y (t).Then in t=0 and t=τ, it is utilized respectively undersampling factor M1And M2Analog-digital converter y (t) is sampled, obtain signal y1(n) and y2(n) it and stores.By y1(n) and y2(n) signal is decomposed into using time MT as the sampling interval respectivelyWithAnd Fourier transformation, restructuring matrix are carried out respectivelyWithUtilize restructuring matrixWithStructural matrix uL(f), useful signal y is recovered after matrixing and conversionr(n).From yr(n) it is isolated again in and original signal xa(n) and xb(n) the signal y in corresponding feedback loopra(n) and yrb(n), by xa(n), xb(n) and yra(n), yrb(n) it compares to obtain digital pre-distortion block coefficient by least square method.This invention ensures that digital predistortion linear effect, does not increase the complexity and other hardware costs of system circuit, predistortion architecture has more flexibility.
Description
Technical field
The invention belongs to wireless communication field, specifically a kind of concurrent double frequency of single feedback loop based on time-interleaved sampling
Band digital pre-distortion method.
Background technique
Power amplifier (Power Amplifier, PA) is the important devices in wireless communication system transmitters, it can handle
Low level signal amplification is the very big frequency microwave signal of energy.And existing frequently-used orthogonal frequency division multiplexing (Orthogonal
Frequency Division Multiplexing, OFDM) technology possesses the subcarrier of multiple same magnitudes and modulation system,
Subcarrier distribution is over different frequencies and mutually orthogonal, and which results in very high peak-to-average force ratio (Peak-to-Average Power
Ratio,PAPR).High peak-to-average power ratio will lead to signal and increase in the distortion of amplitude and phase Jing Guo power amplifier, on the one hand
The distortion in band is caused, the neighboring trace leakage of frequency spectrum on the other hand can be also generated, the information on other channels is interfered.Cause
This, for the concurrent data flow of the superelevation in future broadband systems, the linearisation research of power amplifier has great meaning
Justice.
With the increase and diversified development of communication service, the problem that frequency spectrum resource and power resource are limited is increasingly tight
Weight.In order to improve frequency spectrum resource utilization rate, high-order modulating is widely applied.High power amplifier output power needs
A certain amount of rollback is carried out to guarantee the requirement of a degree of linearity, can but reduce the work of power amplifier to a certain extent
Make efficiency.
Adaptive digital pre-distortion technology is that compensation non-linearity of power amplifier is distorted one of the best way.In recent years,
As multiband, high-bandwidth signals transmission demand are continuously increased, digital pre-distortion sampling circuit bandwidth is faced with great challenge,
The digital pre-distortion technology that can be used for concurrent multiband occurs therewith.Traditional two-dimensional digital pre-distortion technology (2D-DPD) knot
Structure completes the independent acquisition of the dual band signal of power amplifier output by double-feedback-loop, to obtain respective frequency band
Pre-distortion coefficients.And double-feedback-loop is needed to realize the compensation of double frequency-band digital pre-distortion respectively, to increase digital pre-
It is distorted hardware implementation cost.
In order to optimize feedback channel structure, there is researcher to propose the concurrent double frequency-band skill of single feedback loop based on lack sampling
Sampling rate in feedback loop can be effectively reduced in art, and signal occurs still can effectively recognize in the case where aliasing after lack sampling
Pre-distortion coefficients.Although its used method can reduce the hardware requirement of system, since useful signal occurs after lack sampling
Aliasing, system noise can be also superimposed, therefore the valuable information in relation to peak signal power may be lost, and with peak
In the communication system of equal power ratio (PAPR), if without accurately capturing peak signal power, linear properties will deteriorate.
Summary of the invention
The present invention is to solve in the following high bandwidth multi-band communication digital pre-distortion feedback loop, and required sampling rate is increasingly
The problem of increase, proposes a kind of concurrent double frequency-band digital pre-distortion method of single feedback loop based on time-interleaved sampling；It is logical
It crosses the multiple relatively prime sampling in single feedback loop and obtains time-interleaved signal, to accurately extract pre-distortion coefficients；It is protecting
Under the premise of demonstrate,proving power amplifier non-linear distortion effect of optimization, sample rate is effectively reduced for feedback loop and proposes a kind of new solution think of
Road.
Specific steps include:
Step 1: calculating concurrent double frequency-band input letter using two baseband signals of concurrent double frequency-band as original signal
Number x (n)；
Concurrent double frequency-band input signal formula are as follows:
Wherein xa(n) and xb(n) be respectively concurrent double frequency-band to be launched two baseband signals；Wherein, w0=2 π f0, f0
For the carrier frequency of signal.
Step 2: double frequency-band input signal x (n) successively passes through frequency mixer, after digital analog converter and power amplifier, anti-
It is fed back in road and acquires amplified signal, after down coversion, harmonic distortion and the band in amplified signal are filtered out using low-pass filter
Outer crosstalk obtains filtered analog signal y (t).
τ is the time interval of signal source repeat its transmission.
Step 3: utilizing undersampling factor M in initial transmissions time t=01Analog-digital converter to filtered letter
Number y (t) is sampled, and signal y is obtained1(n) it and stores；
Wherein T is fully sampled rate unit time interval；
Step 4: utilizing undersampling factor M in launch time t=τ2Analog-digital converter to filtered signal y
(t) it is sampled, obtains signal y2(n) it and stores.
Step 5: by signal y1(n) and y2(n) signal is decomposed into using time MT as the sampling interval respectivelyWith
Wherein, M=M1·M2；
Step 6: to the signal of each decompositionWithFourier transformation, restructuring matrix are carried out respectively
With
For signalAfter carrying out Fourier transformation, periodic function is obtained
F is the frequency of signal；
By periodic functionLeft and right is all multiplied by exp (- j2 π M1PTf), restructuring matrix is obtained
It is one-dimensional matrix:
yM(f) be a M row matrix: yM(f)={ Y (f)；Y(f+1/MT)；…；Y(f+(M-1)/MT)}；
Similarly obtain restructuring matrix
Step 7: utilizing restructuring matrixWithStructural matrix uL(f), extensive after matrixing and conversion
Appear again useful signal yr(n)。
Firstly, structural matrix uL(f) are as follows:
Then, matrixing is carried out, by uL(f) it indicates are as follows: uL(f)=ΦL,MyM(f)
By to matrix yM(f) the non-zero row vector in is recombinated, and new matrix is obtainedΦ simultaneouslyL,MColumn
Vector is reduced to S accordingly, is denoted as ΦL,S。
In S≤L,
Finally, solving matrix
It willReconfigure row vector Yr(f), to Yr(f) it carries out Fourier inversion and obtains useful signal yr
(n)。
YrIt (f) include the spectrum information of concurrent double frequency-band present position in Y (f) in.
Step 8: from the useful signal y after reconstructr(n) it is isolated again in and original signal xa(n) and xb(n) corresponding
Signal y in feedback loopa(n) and yb(n)。
Firstly, from the useful signal y after reconstructr(n) two signal y in concurrent double frequency-band are isolated inra(n) and yrb
(n)
Wherein,For the IDFT matrix of premultiplication；
HaAnd HbFor ideal filter matrix, indicate are as follows:
Wherein IlAnd IrIt is all the unit matrix of (NS/2M) × (NS/2M)；
Wherein w0=2 π (fa-fb)/2/fs, similarly obtain yrb(n)。
In conclusion being successfully separated out signal yra(n) and yrb(n) signal y (n) is regarded as after ideal bandpass filter
The signal for two frequency bands that down coversion obtains, then yra(n) and yrb(n) with original signal xa(n)、xb(n) meet such as institute in following formula
The 2D-DPD model shown.
WhereinWithIt is model coefficient,Set constitute one-dimensional vector C(a),Set constitute it is one-dimensional
Vector C(b), K is non-linear order, and M is memory depth.
Model coefficient vector C is extracted by least square method(a)And C(b), therefore digital pre-distortion block coefficient is able to more
Newly.
The present invention has the advantages that
1, the concurrent double frequency-band digital pre-distortion method of a kind of single feedback loop based on time-interleaved sampling is guaranteeing number
Under the premise of predistortion linear effect, effectively reduce in digital pre-distortion feedback loop needed for required analog-digital converter (ADC)
Sample rate.
2, single-lens reflex camera is used only in the concurrent double frequency-band digital pre-distortion method of a kind of single feedback loop based on time-interleaved sampling
It is fed back to road singly adjustable ADC, does not increase the complexity and other hardware costs of system circuit, such as: frequency mixer, filter etc..
3, the concurrent double frequency-band digital pre-distortion method of a kind of single feedback loop based on time-interleaved sampling, using recovery
Algorithm restores the signal of any carrier position in a certain range, so that feedback loop is no longer limited by frequency mixer, predistortion architecture
With more flexibility.
Detailed description of the invention
Fig. 1 is in a kind of concurrent double frequency-band digital pre-distortion method of single feedback loop based on time-interleaved sampling of the invention
Schematic diagram；
Fig. 2 is a kind of concurrent double frequency-band digital pre-distortion method of single feedback loop based on time-interleaved sampling of the invention
Flow chart；
Fig. 3 is a kind of concurrent double frequency-band digital pre-distortion method of single feedback loop based on time-interleaved sampling of the invention
Schematic diagram；
Fig. 4 be the specific embodiment of the invention use synchronize relatively prime sampling instance schematic diagram；
Fig. 5 is power spectral density comparison diagram of the present invention using different DPD frameworks.
Specific embodiment
Below in conjunction with attached drawing and example, the present invention is described in further detail.
The present invention loses to reduce the information as caused by lack sampling, improves the linearisation degree of power amplifier, mentions
A kind of concurrent double frequency-band digital pre-distortion method (Time interleaved of single feedback loop based on time-interleaved sampling is gone out
sampling based digital predistortion technique with a single feedback loop
For dual-band DPD), time-interleaved signal is obtained by the sampling relatively prime twice to power amplifier output signal, is fed back to
Single low sampling rate ADC need to be only used in road, can be achieved with the acquisition of power amplifier output useful signal.Then, indirect learning is utilized
Structure extracts the pre-distortion coefficients of each frequency band respectively.
Based on time-interleaved concurrent two-band list feedback loop principle as shown in Figure 1, by two frequency band x to be launcheda
(n) and frequency band xb(n) it inputs in DPD model, passes through digital analog converter DAC-a and DAC-b respectively；Each self-bonding own signal pair
The carrier frequency f answeredaAnd fb, two signals couple small signal by coupler after PA and enter in feedback loop, feeding back to
It is (f that centre frequency is first passed around in roada+fbSignals revivification to base band is then passed through a low pass filtered by the down conversion module of)/2
Wave device, then double sampling is carried out in time t=0 and t=τ respectively by the adjustable ADC of sampling rate and obtains signal y1(n) and
y2(n), by y obtained1(n) and y2(n) signal y is reconstructed into using recovery algorithmsr(n), it is carried out using the signal of recovery adaptive
Predistortion is answered, solves to obtain pre-distortion parameters using least square method afterwards.
Specific steps are as shown in Figures 2 and 3, comprising:
Step 1: calculating concurrent double frequency-band input letter using two baseband signals of concurrent double frequency-band as original signal
Number x (n)；
Concurrent double frequency-band input signal formula are as follows:
Wherein xa(n) and xb(n) be respectively concurrent double frequency-band to be launched two baseband signals；Wherein, w0=2 π f0, f0
For the carrier frequency of signal.Frequency band xa(n) and frequency band xb(n) corresponding pre-distortion parameters are C(a)And C(b), initial value C is set(a)
=C(b)=[1,0,0 ..., 0]T, the carrier frequency for emitting signal is respectively faAnd fb, the spread bandwidth after power amplifier is respectively
BaAnd Bb.Regard the output of two-band power amplifier as a complete broadband signal, the carrier frequency of this signal is between two carrier waves
Centre frequency, i.e. (fa+fb)/2；Setting launch time is t1=0 and t2=τ, while the undersampling factor of double sampling is set
Respectively M1And M2。
Step 2: double frequency-band input signal x (n) successively passes through frequency mixer, after digital analog converter and power amplifier, anti-
It is fed back in road and acquires amplified signal, after down coversion, the harmonic distortion in feedback loop signals is filtered out using low-pass filter
With with outer crosstalk, filtered analog signal y (t) is obtained.
Power amplifier output broadband signal is down-converted to base band in the feedback loop, down coversion local oscillator is also between two carrier waves
Centre frequency, the out-of-band distortion in frequency spectrum is eliminated after down coversion by a low-pass filter, is obtained analog signal y (t),
Filter output signal y (t) are as follows:
It will be in t=0 and t=τ respectively by being M with sample rate1And M2Lack sampling ADC, wherein T be Nai Kuisi
Special fully sampled rate unit time interval, then y (t) should meet
Step 3: the time interval τ according to signal source repeat its transmission carries out signal twice to filtered signal y (t)
Transmitting and sampling, obtain signal y1(n) and y2(n) it and stores；
In initial transmissions time t=0 and t=τ, undersampling factor M is utilized1And M2Analog-digital converter adopted respectively
Sample obtains signal y1(n) and y2(n)；
M1With M2For two relatively prime integers, most useful signals can be just sampled.The signal of double sampling is all saved
For the reconstruct of signal, the relationship of double sampling is as shown in Fig. 4 top half in time-domain signal, in every time interval MT,
DAC-a is with time interval M1T is sampled, and can collect M altogether2It is a, and DAC-b is then with time interval M2T is sampled,
M can be collected altogether1A point.After knowing double sampling, y (t) shares M within the time of MT1+M2- 1 point can be sampled
It arrives.
Step 4: by signal y1(n) and y2(n), it is decomposed into signalWith
Wherein, M=M1·M2；As shown in the lower half portion Fig. 4, by y1(n) and y2(n) it is decomposed into L, (L=M1+M2- 1) a to adopt
Sample time interval is M1M2The signal of T carries out phase recovery and Fourier transformation to signal after each decomposition, will pass through Fourier
Transformed signal reconstruction is a matrix.Useful signal y can be recovered after matrixing and conversionr(n)。
Step 5: to the signal of each decompositionWithSignal is carried out respectively by the Fourier of time domain to frequency domain
Transformation, restructuring matrixWith
For signalAfter carrying out Fourier transformation, the periodic function using 1/MT as the period is obtained
It is added after the frequency displacement of the second frequency domain part elder generation of line subrepresentation M.F is the frequency of signal；f∈Θ0, Θ0=[0,1/
MT)。
By periodic functionLeft and right is all multiplied by exp (- j2 π M1PTf), it is rewritten as the form of following two matrixes:
WhereinIt is one-dimensional matrix:
yM(f) be a M row matrix, M row vector be cut by Y (f) M leaf length be 1/MT spectral slice
Composition:
yM(f)={ Y (f)；Y(f+1/MT)；…；Y(f+(M-1)/MT)} (9)
SettingFormula (7) is write again:
Similarly obtain restructuring matrix
Step 6: utilizing restructuring matrixWithStructural matrix uL(f), extensive after matrixing and conversion
Appear again useful signal yr(n)。
Firstly, structural matrix uL(f) are as follows:
Then, it brings formula (10) (11) into formula (12) and carries out matrixing, by uL(f) it indicates are as follows:
uL(f)=ΦL,MyM(f) (13)
L=M1+M2- 1,
Due to yM(f) useful information for restoring signal is contained in, so needing to restore y by formula (13)M(f), formula
It (13) is a underdetermined equation, it can not direct solution.But the y known to formula (9)MIt (f) include the spectrum information of different location in,
And include asemantic protection interval in concurrent double frequency-band, i.e. yM(f) the row vector part in is zero, so yM(f) row to
Amount can reduce as S (S=λ M), by matrix yM(f) the non-zero row vector in is recombinated, and new matrix is obtainedΦ simultaneouslyL,MColumn vector be reduced to S accordingly, be denoted as ΦL,S。
In S≤L, formula (13) rewrites the form for making an overdetermined equation:
It is obtained finally, being solved by (14)
It willRow vector Y is reconfigured according to the position in formula (9)r(f), YrIt (f) include concurrent double in Y (f) in
The spectrum information of frequency band present position, to Yr(f) it carries out Fourier inversion and obtains useful signal yr(n)。
Step 7: from the useful signal y after reconstructr(n) it is isolated again in and original signal xa(n) and xb(n) corresponding
Signal y in feedback loopa(n) and yb(n)。
Firstly, from the useful signal y after reconstructr(n) two signal y in concurrent double frequency-band are isolated inra(n) and yrb
(n)
Wherein,For the IDFT matrix of premultiplication；
HaAnd HbFor ideal filter matrix, indicate are as follows:
Wherein IlAnd IrIt is all the unit matrix of (NS/2M) × (NS/2M)；Formula (17) (18) is brought into formula (16), it can be close
Seemingly it is expressed as
Wherein w0=2 π (fa-fb)/2/fs, similarly obtain yrb(n)。
In conclusion signal y can be successfully separated out by above-mentioned sample modera(n) and yrb(n), obtained letter is reconstructed
Number yra(n) and yrb(n) it may be regarded as the letter for two frequency bands that signal y (n) down coversion after ideal bandpass filter obtains
Number, then yra(n) and yrb(n) with original signal xa(n)、xb(n) meet the 2D-DPD model as shown in following formula.
WhereinWithIt is model coefficient, K is non-linear order, and M is memory depth.It is extracted by least square method
DPD model coefficient is updated, and obtains the signal y in feedback loopa(n) and yb(n).By comparing renewable pre-distortion parameters
waWith wb。
WhereinWithIt is model coefficient, as shown in figure 3,Set constitute an one-dimensional vector C(a),'s
Set constitutes an one-dimensional vector C(b), K is non-linear order, and M is memory depth.
Model coefficient vector C is extracted by least square method(a)And C(b), therefore digital pre-distortion block coefficient is able to more
Newly.
A kind of novel double-frequency DPD structure with single feedback control loop of the present invention, it improves list using time-interleaved signal
The validity that signal is acquired in feedback control loop reduces the sample rate of required single ADC simultaneously.As shown in figure 5, the present invention utilizes
Orthogonal frequency division multiplexing (OFDM) signal of two 15MHz is tested, and two ofdm signals point are modulated by up-converter module
To at 19.75GHz and 19.95GHz, then emitted by power amplifier.In the feedback loop, conventional digital predistortion
If being sampled merely with a single feedback loop to power amplifier output signal in scheme and analyzing extraction pre-distortion coefficients
When, sampling bandwidth need to reach 300MHz, i.e., required ADC sampling rate need to reach 600MSPS, and method proposed by the invention only needs
Wanting sampling rate is respectively that the small-sized ADC of two time synchronizations of 100MSPS and 120MSPS can correctly extract predistortion module
Coefficient.The experimental results showed that three curves show original OFDM signal respectively, merely through power amplifier signal with
And the spectrogram of the signal Jing Guo two modules of digital pre-distortion block and power amplifier, original signal and process number of the present invention
Spread spectrum situation does not all occur for the signal of word predistortion module, and adjacent signal power ratio (ACPR) can be reduced to -52dBc, and
- 28dBc is then up to without the ACPR of the signal of predistortion.Experiments have shown that sampling of the structure proposed in required single ADC
When rate reduces 5 times, still pre-distortion coefficients can be correctly extracted to compensate the non-thread of power amplifier for digital pre-distortion block
Property.To sum up, the present invention is able to satisfy the linearization performance of power amplifier, reduces feedback loop system cost and scale.
Claims (3)
1. a kind of concurrent double frequency-band digital pre-distortion method of single feedback loop based on time-interleaved sampling, which is characterized in that tool
Body step includes:
Step 1: calculating concurrent double frequency-band input signal x using two baseband signals of concurrent double frequency-band as original signal
(n)；
Step 2: double frequency-band input signal x (n) successively passes through frequency mixer, after digital analog converter and power amplifier, feeding back to
Amplified signal is acquired in road, after down coversion, filters out harmonic distortion in amplified signal using low-pass filter and with diplomacy
Distortion is adjusted, filtered analog signal y (t) is obtained；
τ is the time interval of signal source repeat its transmission；
Step 3: utilizing undersampling factor M in initial transmissions time t=01Analog-digital converter to filtered signal y (t)
It is sampled, obtains signal y1(n) it and stores；
Wherein T is fully sampled rate unit time interval；
Step 4: utilizing undersampling factor M in launch time t=τ2Analog-digital converter to filtered signal y (t) carry out
Sampling, obtains signal y2(n) it and stores；
Step 5: by signal y1(n) and y2(n) signal is decomposed into using time MT as the sampling interval respectivelyWith
Wherein, M=M1·M2；
Step 6: to the signal of each decompositionWithFourier transformation, restructuring matrix are carried out respectivelyWith
For signalAfter carrying out Fourier transformation, periodic function Y is obtained1 p(f):
F is the frequency of signal；
By periodic function Y1 p(f) left and right is all multiplied by exp (- j2 π M1PTf), restructuring matrix is obtained
It is one-dimensional matrix:
yM(f) be a M row matrix: yM(f)={ Y (f)；Y(f+1/MT)；...；Y(f+(M-1)/MT)}；
Similarly obtain restructuring matrix
Step 7: utilizing restructuring matrixWithStructural matrix uL(f), it is recovered after matrixing and conversion
Useful signal yr(n)；
Firstly, structural matrix uL(f) are as follows:
Then, matrixing is carried out, by uL(f) it indicates are as follows: uL(f)=ΦL,MyM(f)
L=M1+M2- 1,
By to matrix yM(f) the non-zero row vector in is recombinated, and new matrix is obtainedΦ simultaneouslyL,MColumn vector
It is reduced to S accordingly, is denoted as ΦL,S；
In S≤L,
Finally, solving matrix
It willReconfigure row vector Yr(f), to Yr(f) it carries out Fourier inversion and obtains useful signal yr(n)；
YrIt (f) include the spectrum information of concurrent double frequency-band present position in Y (f) in；
Step 8: from the useful signal y after reconstructr(n) it is isolated again in and original signal xa(n) and xb(n) corresponding feedback
Signal y in circuita(n) and yb(n)。
2. a kind of concurrent double frequency-band digital pre-distortion side of single feedback loop based on time-interleaved sampling as described in claim 1
Method, which is characterized in that concurrent double frequency-band input signal formula described in step 1 are as follows:
Wherein xa(n) and xb(n) be respectively concurrent double frequency-band to be launched two baseband signals；Wherein, w0=2 π f0, f0For letter
Number carrier frequency.
3. a kind of concurrent double frequency-band digital pre-distortion side of single feedback loop based on time-interleaved sampling as described in claim 1
Method, which is characterized in that the step eight specifically:
Firstly, from the useful signal y after reconstructr(n) two signal y in concurrent double frequency-band are isolated inra(n) and yrb(n)
Wherein,For the IDFT matrix of premultiplication；
HaAnd HbFor ideal filter matrix, indicate are as follows:
Wherein IlAnd IrIt is all the unit matrix of (NS/2M) × (NS/2M)；
Wherein w0=2 π (fa-fb)/2/fs, similarly obtain yrb(n)；
In conclusion being successfully separated out signal yra(n) and yrb(n) it is regarded as signal y (n) down coversion after ideal bandpass filter
The signal of two obtained frequency bands, then yra(n) and yrb(n) with original signal xa(n)、xb(n) meet the 2D- as shown in following formula
DPD model；
WhereinWithIt is model coefficient,Set constitute one-dimensional vector C(a),Set constitute one-dimensional vector
C(b), K is non-linear order, and M is memory depth；
Model coefficient vector C is extracted by least square method(a)And C(b), therefore digital pre-distortion block coefficient is updated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910184466.4A CN109889166B (en) | 2019-03-12 | 2019-03-12 | Single feedback loop concurrent dual-band digital predistortion method based on time interleaved sampling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910184466.4A CN109889166B (en) | 2019-03-12 | 2019-03-12 | Single feedback loop concurrent dual-band digital predistortion method based on time interleaved sampling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109889166A true CN109889166A (en) | 2019-06-14 |
| CN109889166B CN109889166B (en) | 2021-01-01 |
Family
ID=66931745
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910184466.4A Active CN109889166B (en) | 2019-03-12 | 2019-03-12 | Single feedback loop concurrent dual-band digital predistortion method based on time interleaved sampling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109889166B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021169997A1 (en) * | 2020-02-24 | 2021-09-02 | 华为技术有限公司 | Digital pre-distortion circuit, digital pre-distortion coefficient acquisition method and related device |
| CN113691267A (en) * | 2021-09-06 | 2021-11-23 | 中国工程物理研究院电子工程研究所 | Self-interference suppression method for carrier aggregation receiving end in simultaneous dual-band |
| WO2024051520A1 (en) * | 2022-09-08 | 2024-03-14 | 中兴通讯股份有限公司 | Digital predistortion circuit and method, and radio frequency chip and electronic device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101902258A (en) * | 2009-05-31 | 2010-12-01 | 大唐移动通信设备有限公司 | Method and device for acquiring digital pre-distortion processing parameter |
| US20160191087A1 (en) * | 2011-09-22 | 2016-06-30 | Dali Systems Co. Ltd. | System and method for increasing bandwidth for digital predistortion in multi-channel wideband communication systems |
| CN106301631A (en) * | 2016-10-18 | 2017-01-04 | 天津大学 | A kind of coprime lack sampling frequency spectrum sensing method based on Subspace Decomposition and device thereof |
| CN106685368A (en) * | 2017-01-03 | 2017-05-17 | 电子科技大学 | Digital predistortion system and method achieving undersampling feedback |
| CN107612856A (en) * | 2017-10-10 | 2018-01-19 | 京信通信系统（中国）有限公司 | A kind of digital pre-distortion processing method and device |
-
2019
- 2019-03-12 CN CN201910184466.4A patent/CN109889166B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101902258A (en) * | 2009-05-31 | 2010-12-01 | 大唐移动通信设备有限公司 | Method and device for acquiring digital pre-distortion processing parameter |
| US20160191087A1 (en) * | 2011-09-22 | 2016-06-30 | Dali Systems Co. Ltd. | System and method for increasing bandwidth for digital predistortion in multi-channel wideband communication systems |
| CN106301631A (en) * | 2016-10-18 | 2017-01-04 | 天津大学 | A kind of coprime lack sampling frequency spectrum sensing method based on Subspace Decomposition and device thereof |
| CN106685368A (en) * | 2017-01-03 | 2017-05-17 | 电子科技大学 | Digital predistortion system and method achieving undersampling feedback |
| CN107612856A (en) * | 2017-10-10 | 2018-01-19 | 京信通信系统（中国）有限公司 | A kind of digital pre-distortion processing method and device |
Non-Patent Citations (2)
| Title |
|---|
| YIJIU ZHAO等: "Sparse Multiband Signal Acquisition Receiver With Co-Prime Sampling", 《IEEE ACCESS》 * |
| YOUJIANG LIU等: "Concurrent Dual-Band Digital Predistortion With a Single Feedback Loop", 《IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021169997A1 (en) * | 2020-02-24 | 2021-09-02 | 华为技术有限公司 | Digital pre-distortion circuit, digital pre-distortion coefficient acquisition method and related device |
| CN113691267A (en) * | 2021-09-06 | 2021-11-23 | 中国工程物理研究院电子工程研究所 | Self-interference suppression method for carrier aggregation receiving end in simultaneous dual-band |
| CN113691267B (en) * | 2021-09-06 | 2023-03-10 | 中国工程物理研究院电子工程研究所 | Self-interference suppression method for carrier aggregation receiving end in simultaneous dual-band |
| WO2024051520A1 (en) * | 2022-09-08 | 2024-03-14 | 中兴通讯股份有限公司 | Digital predistortion circuit and method, and radio frequency chip and electronic device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109889166B (en) | 2021-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109889166A (en) | A kind of concurrent double frequency-band digital pre-distortion method of single feedback loop based on time-interleaved sampling | |
| EP1702445B1 (en) | Modulation and demodulation of ofdm signals | |
| US6294956B1 (en) | System and method for producing amplified signal(s) or version(s) thereof | |
| EP2200244B1 (en) | Method and apparatus for multi-carrier frequency division multiplexing transmission | |
| CN108134756B (en) | Wireless communication system based on vortex electromagnetic wave and orthogonal frequency division multiplexing | |
| CN101232488B (en) | Communication method capable of reducing peak average power ratio of OFDM system | |
| CN108881083B (en) | Envelope-assisted RF/IF digital predistortion technique for wideband ROF system | |
| CN112956166B (en) | Systems and methods for multi-carrier or multi-band low-PAPR precoding handling cyclic prefixes | |
| EP1867123A2 (en) | Signal transmitter for wideband wireless communication | |
| KR20090113915A (en) | Radio communication system | |
| CN110661546A (en) | Method and system for eliminating harmonic interference digital domain in 5G terminal based on NSA architecture | |
| CN105763495A (en) | Digital pre-distortion method and device | |
| CN105684338A (en) | Crest factor reduction of inter-band carrier aggregated signals | |
| CN102751950A (en) | Digital predistortion method of multirange multimode radio frequency power amplifier | |
| Hu et al. | A novel single feedback architecture with time-interleaved sampling for multi-band DPD | |
| CN109714142A (en) | Efficient OTDM transmitting method and system | |
| Zayani et al. | Experimental testbed of post-OFDM waveforms toward future wireless networks | |
| WO2009046627A1 (en) | A method for correcting signals in radio frequency links and the pre-distortion unit thereof | |
| CN101267415B (en) | Flexible uplink multi-address transmission device based on filter group and its method | |
| CN106464273A (en) | Signal processing method, transmitter and compressive sampling receiver | |
| Shaheen et al. | Absolute exponential companding to reduced PAPR for FBMC/OQAM | |
| Yadav et al. | Nonlinearity effect of Power Amplifiers in wireless communication systems | |
| CN101729134B (en) | Generalized multiple carrier frequency division multiple access transmission equipment based on orthogonal transformation processing | |
| Suma et al. | Orthogonal frequency division multiplexing peak‐to‐average power ratio reduction by best tree selection using coded discrete cosine harmonic wavelet packet transform | |
| CN104301269B (en) | Equivalent sampling pre-distortion system and its method |
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 |