CN108023844A - A kind of digital pre-distortion system of real signal lack sampling - Google Patents
A kind of digital pre-distortion system of real signal lack sampling Download PDFInfo
- Publication number
- CN108023844A CN108023844A CN201710439383.6A CN201710439383A CN108023844A CN 108023844 A CN108023844 A CN 108023844A CN 201710439383 A CN201710439383 A CN 201710439383A CN 108023844 A CN108023844 A CN 108023844A
- Authority
- CN
- China
- Prior art keywords
- mrow
- signal
- msub
- mover
- time delay
- 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
Classifications
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/0202—Channel estimation
- H04L25/024—Channel estimation channel estimation algorithms
- H04L25/0242—Channel estimation channel estimation algorithms using matrix methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mathematical Physics (AREA)
- Power Engineering (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Amplifiers (AREA)
Abstract
The present invention is a kind of digital pre-distortion system of real signal lack sampling, and step is as follows：1) lack sampling is carried out to the one-component of power amplifier output complex baseband signal with single channel analog-digital converter, obtains feedback signal；2) time delay between time delay estimation and compensating unit estimation and compensation original complex base band input signal and feedback signal；3) phase deviation between original complex base band input signal and feedback signal through delay compensation of phase estimation and compensating unit compensation estimation and compensation；4) pre-distortion coefficients estimation unit is using through the original complex base band input signal through time delay and phase compensation and feedback signal, as input, iteration updates the coefficient vector of predistorter.The present invention is only needed to power amplifier output complex baseband signal with a progress lack sampling mutually and in quadrature component, substantially reduce analog-digital converter sampling rate is required while, the number of analog-digital converter is reduced to from two-way all the way, reduces the hardware cost and power consumption of digital pre-distortion.
Description
Technical field
The present invention relates to field of signal processing, and in particular to a kind of digital pre-distortion system of real signal lack sampling.
Background technology
Power amplifier is the important component of wireless communication system transmitting terminal, at the same be also transmitting terminal it is the most expensive with
The component of energy consumption.In order to improve the power-efficient of power amplifier with lifted transmitting end system entirety efficiency ratio, power amplification
Device is operated in the state close to saturation output power mostly, causes signal to produce serious non-linear distortion.Non-linear distortion is removed
The quality of transmitting signal can be damaged, reduces the bit error rate performance of receiving terminal, can also make signal that spread spectrum occur, to neighbouring letter
Road produces interference.On the other hand, since the day of Radio Spectrum Resource is becoming tight, the modulated signal of all kinds of spectral efficients, example
Such as quadrature amplitude modulation (QAM), Amplitude-Phase keying (APSK) and Orthogonal Frequency Division Multiplexing (OFDM) are in wireless communication system
In be widely adopted.Compared to constant enveloped modulation signals such as minimum phase keyings (MSK), the modulated signal of spectral efficient is often
With higher peak value-average power (PAPR) so that under the conditions of identical average output power, the tune of spectral efficient
Signal processed is more sensitive for the non-linear distortion of power amplifier.In order to while power amplifier power supply efficiency is improved,
The problems such as solving transmission signal quality decline caused by the non-linear distortion of power amplifier and adjacent channel interference, industry phase
After the linearization technique for proposing each power-like amplifier, including Power back, feed forward approach, feedback transmitter, analog predistortion and
Digital pre-distortion.Wherein digital pre-distortion is because of the validity of its nonlinear distortion compensation, the adaptivity of digital processing and technology
The advantages that flexibility of deployment, is paid close attention to be subject to industry, in wireless communication system, the particularly third generation and forth generation mobile communication
It is widely used in system.
But the evolution of wireless communication system also brings many new challenges to digital pre-distortion technology, one is exactly pair
The support of broadband signal.Digital pre-distortion technology need by one be known as feedback control loop special receiver circuit, to
The power amplifier output signal of non-linear distortion is sampled.Since non-linear distortion can cause the extension of signal spectrum, pass
On system, the bandwidth that feedback control loop should possess will substantially exceed the bandwidth of power amplifier original input signal, one in Practical Project
As take more than 5 times of power amplifier original input signal bandwidth, this proposes the analog-digital converter (ADC) in feedback control loop
Very high requirement.By taking forth generation mobile communication system as an example, according to the carrier aggregation mode of five component carriers, each component
The bandwidth of carrier wave is 20MHz, then maximum signal bandwidth is 100MHz, is sampled with 5 times of signal bandwidth, then ADC is adopted
Sample speed is up to 500MSPS (zero intermediate frequency sample, two-way ADC) or 1000MSPS (if sampling, all the way ADC).And in studying
The 5th Generation Mobile Communication System in, signal bandwidth is up to 2.5GSPS up to more than 500MHz, corresponding ADC sampling rates
Or 5GSPS.So high sampling rate so that the type selecting of ADC and the engineering design of feedback control loop are all difficult, and at a high speed
The price and high power consumption of ADC costlinesses can greatly increase the cost and power consumption of digital pre-distortion in itself, largely counteract and adopt
The beneficial effect caused by digital pre-distortion.
In order to solve under broadband application scene, the problem of digital pre-distortion feedback control loop sampling rate is excessive, it has been proposed that
Many new digital pre-distortion methods for reducing feedback control loop sampling rates, wherein comparing typical method has frequency spectrum extrapolation
With the method modeled based on lack sampling power amplifier.The basic principle of frequency spectrum extrapolation is that band logical filter is added before feedback control loop
The bandwidth of ripple device limiting power amplifier output signal is to reduce sampling rate, then by the power in bandpass filter passband
Correlation inside and outside amplifier output signal and passband between power amplifier output signal, thus it is speculated that go out complete work(on frequency spectrum
Rate amplifier exports signal, on this basis by the coefficient of traditional pre-distortion coefficients discrimination method estimation predistorter.To the greatest extent
Pipe frequency spectrum extrapolation on the premise of original predistortion effect is kept, can be effectively reduced the sampling rate of feedback control loop ADC,
But the level identical with power amplifier original input signal bandwidth can only be at most reduced to, otherwise the result of frequency spectrum extrapolation is by nothing
Method is guaranteed, and limits application of the frequency spectrum extrapolation in future broadband wireless communication systems.Built based on lack sampling power amplifier
The method of mould, then be that the output signal directly to power amplifier carries out low speed sampling, and without anti-aliasing filter, although by
This obtained signal will be inevitably generated aliasing, but can accomplish to believe in sample point and the output of real power amplifier
It is number consistent.Again because the identification of power amplifier model, it is only necessary to sufficient amount of power amplifier input-output signal, and
It is whether continuous unrelated in time with these signals, therefore the power amplifier input-output of low sampling rate can be utilized to believe
Power amplifier model number is directly recognized, recycles the power amplifier model that identification obtains to obtain power continuous in time and puts
Big device output signal, the coefficient of the traditional pre-distortion coefficients discrimination method estimation predistorter of last resort.
In theory, the method based on the modeling of lack sampling power amplifier, its performance is only related with the number of samples that sampling obtains,
And it is unrelated with the sampling rate of ADC, therefore the sampling rate of ADC can be arbitrarily low, is the number that a kind of comparison is adapted to broadband signal
Word pre-distortion method.But this method is only applicable to the situation of complex baseband signal lack sampling, therefore although the sampling rate of ADC very
It is low, it is still desirable to which that two-way ADC distinguishes the same phase and quadrature component of sampled power amplifier output signal, is unfavorable for reducing numeral
The cost of implementation and power consumption of pre-distortion system, and this method needs to obtain power amplifier model first, then the base in model
The coefficient of predistorter is estimated on plinth, computation complexity is higher；On the other hand, sampling complex baseband signal is intended to using just
Hand over demodulator to separate the same phase and quadrature component of amplifier output signal, can inevitably be exported in power amplifier
Amplitude imbalance and quadrature error are introduced in signal, so as to influence the performance of digital pre-distortion system.
The content of the invention
The present invention provides a kind of digital pre-distortion system of real signal lack sampling, to overcome existing lack sampling digital pre-distortion
Technology need to sample complex baseband signal, and the shortcomings that need first to carry out power amplifier model identification.
The present invention is achieved through the following technical solutions：
A kind of digital pre-distortion system of real signal lack sampling, including time delay estimation and compensating unit, phase estimation and benefit
Unit and predistorter coefficient estimation unit are repaid, wherein：
The time delay estimation and compensating unit, the lack sampling reality base with complex baseband input signal S1 and from analog-digital converter
Band signal S12 is input, when estimating the time-delay value of signal S12 delay signals S1, and signal S1 being carried out according to the time delay
Prolong compensation, generate the complex baseband input signal S13 after delay compensation；
The phase estimation and compensating unit, turn with the complex baseband input signal S13 after delay compensation and from modulus
The lack sampling real baseband signal S12 of parallel operation is input, estimates phase places of the signal S12 relative to signal S13, and according to the phase
Position rotation carries out phase compensation to signal S13, generates the complex baseband input signal S14 after phase compensation；
The predistorter coefficient estimation unit, with the complex baseband signal S14 after phase compensation and from analog-to-digital conversion
The lack sampling real baseband signal S12 of device is input, in an iterative manner, estimates the renewal vector of predistorter coefficient vector, and
Coefficient after renewal is sent to predistorter.
Beneficial effects of the present invention include：
First, digital pre-distortion method of the present invention is only needed to power amplifier output complex baseband signal with mutually and orthogonal
One in component progress lack sampling, substantially reduce analog-digital converter sampling rate is required while, by analog-digital converter
Number the hardware cost and power consumption for all the way, significantly reducing digital pre-distortion are reduced to from two-way；Secondly, number of the present invention
Word pre-distortion method can directly obtain the coefficient of predistorter, be put without as existing lack sampling method, first carrying out power
The identification of big device model, saves the expense on calculating；Finally, due to need to only the same phase of complex baseband signal be exported to power amplifier
A progress lack sampling with quadrature component, is not related to the mutually matching problem with quadrature component two paths of signals, from tradition
The influence of amplitude imbalance and quadrature error in feedback control loop.
Brief description of the drawings
Fig. 1 is the structure chart of conventional digital pre-distortion technology.
Fig. 2 is the structure chart of the predistortion based on the modeling of lack sampling power amplifier.
Fig. 3 is pre-distortion technology structure chart proposed by the invention.
Embodiment
Develop simultaneously embodiment below in conjunction with attached drawing, the present invention will be described in detail.
Conventional digital pre-distortion technology structure is as shown in Figure 1.The original input signal S1 of complex radical band forms is sent to pre- mistake
True device 11, the predistorted pre-distorted signals S2 for being treated as complex radical band forms, and changed by two-way digital analog converter (DAC) 12
To simulation complex baseband signal S3, then it is orthogonally modulated device 13 and is modulated to radio frequency, form 4 exciting power amplifier of RF signal S
14.The output signal of power amplifier 14 is S5, and a part therein passes through directional coupler 15, is fed to quadrature demodulator
21, it is demodulated to form simulation complex baseband signal S6, and complex baseband signal S7 is converted to by two-way analog-digital converter (ADC).In advance
Difference between distorter coefficient estimation unit 41 comparison signal S1 and signal S7, and adjust the coefficient of predistorter 11 so that
Difference between signal S1 and signal S7 it is minimum (loop time delay will cause signal S1 after the time S7 signals S7, it is necessary to using
Time delay is estimated and compensation method is with cancellation loop time delay.But because traditional time delay estimation and compensation method are known, and
And will not explanation of the invention have an impact, therefore traditional time delay estimation and compensation method be comprised in attached drawing 1
In attached drawing 2).The carrier signal of quadrature modulator 13 and quadrature demodulator 21 is inputted to be provided by local oscillator 1, digital-to-analogue conversion
The clock signal of device 12 and analog-digital converter 23 is then provided by clock generator 2, and the switching rate of digital analog converter 12 and
The sampling rate of analog-digital converter 23 is FS。
The problem of conventional digital predistortion architecture, is that its requirement to analog-digital converter 23 is excessive, and power is caught to be accurate
The non-linear distortion of amplifier 14 causes the spread spectrum of signal S5, and how the sampling rate of analog-digital converter 23 is at least signal S1
More than 5 times of Qwest's frequency.In order to solve this problem, it is proposed that the predistortion side based on the modeling of lack sampling power amplifier
Method [2], its structure are as shown in Figure 2.It is different from conventional digital predistortion architecture, D points of the clock signal approach of analog-digital converter 23
Frequency device 3, therefore it is F that analog-digital converter 23, which is operated in sampling rate,SUnder the lack sampling state of/D, take a message by the complex radical of lack sampling
Number it is S8.On the other hand, the power amplifier original input signal S1 of complex radical band forms passes through D times of extracting unit 31, is formed and owed
The original input signal S9 of sampling.Power amplifier model parameter estimation unit 32 is determined by comparison signal S8 and signal S9
The parameter of power amplifier model 32, then for power amplifier model 33 using signal S1 as input, generation power amplifier is actual
Export the estimation S10 of signal.Finally, the difference between predistorter coefficient estimation unit 41 comparison signal S1 and signal S10, and
Adjust the coefficient of predistorter 11 so that the difference between signal S1 and signal S10 is minimum.Since power amplifier model 32 is joined
Several determines, it is only necessary to the sampling point of sufficient amount signal S8 and signal S9, and it is unrelated with the sampling rate of signal S8 and signal S9,
Therefore D values can be arbitrary integer in theory, i.e. analog-digital converter 23 can be operated under arbitrarily low sampling rate.
This method is only applicable to the situation of complex baseband signal lack sampling, that is, as traditional predistortion architecture, it is necessary to
Using two-way analog-digital converter, the increase of cost and power consumption thus will be brought；Secondly, this method can not directly obtain predistorter
Coefficient, and must first carry out the identification of power amplifier model, computation complexity is higher；Finally, complex baseband signal meaning is sampled
Taste, which, to be needed to separate the same phase and quadrature component of amplifier output signal using quadrature demodulator, this can unavoidably exist
Amplitude imbalance and quadrature error are introduced in power amplifier output signal, so as to influence the performance of digital pre-distortion system.For
The disadvantages mentioned above is avoided to be influenced caused by predistortion performance and cost, the numeral that the present invention provides a kind of real signal lack sampling is pre-
Anti-aliasing techniques.
Fig. 3 is the structure chart according to the method for the present invention embodiment.The power amplifier output fed back to from directional coupler
Signal, downconverts to base band through frequency mixer 22 and obtains simulation real baseband signal S11, the analog-digital converter 23 of single channel is with FS/ D is to adopt
Sample speed carries out lack sampling to S11, obtains the real baseband signal S12 of lack sampling.Time delay is estimated and compensating unit 43 is with complex base band
The real baseband signal S12 of input signal S1 and lack sampling is input, is estimated and the time delay between this, and carries out time delay to signal S1
Compensation, the signal after compensating is S13；Phase estimation and compensating unit 44 are with the complex baseband input signal after delay compensation
The real baseband signal S12 of S13 and lack sampling is input, estimates phase difference therebetween, and carry out phase benefit to signal S13
Repay, thermal compensation signal S14.Predistorter real coefficient estimation unit 45 in an iterative manner, is estimated using signal S12 and S14 as input
The renewal vector of 11 coefficient vector of predistorter is counted, and the coefficient after renewal is sent to predistorter 11.
The time delay estimation and compensating unit 44, the purpose is to eliminate the time delay between signal S1 and signal S12, make the two
Align in time.Signal S1 is by predistorter 11, digital analog converter 12, quadrature modulator 13, power amplifier 14, orientation
Coupler 15, frequency mixer 22 and analog-digital converter 23, are converted into signal S12, and above-mentioned all devices can introduce time delay, cause letter
Number S12 lags behind signal S1 in time.Traditionally generally using time delay between cross-correlation method two signals of estimation, if participating in phase
Two complex signals closed are a (n) and b (n), then the time delay estimation between two signalsFor
Wherein ar(n),ai(n),br(n),bi(n) real and imaginary parts of signal a (n) and b (n), N are represented respectivelytTo be mutual
The length of computing is closed,Represent modulo operation, j is imaginary unit.It is but signal S12 is the real signal of lack sampling, its is direct
Carrying out computing cross-correlation with S1 obviously can not obtain correct time delay estimated result.
The time delay estimation and compensating unit 44, estimate the simultaneously time delay between thermal compensation signal S1 and S12 with the following method：
Step 101), if signal S1=z (n), real part and imaginary part are respectively zr(n) and zi(n), signal S12=y (Dn), limit
Fixed response time estimate τ is integer, in section [- τmax,τmax] in change τ values, and find the τ for making maximum on the right side of following formula
Value, is denoted as integral multiple sampling point time delayI.e.
Step 102), defines Time delay Estimation Accuracy Δ τ, in section [- 1,1], changes τ values with step delta τ, and searching makes
Maximum τ values, are denoted as small several times sampling point time delay on the right side of following formulaI.e.
Step 103), signal S1 is postponedA sampled point, wherein integral multiple sampling point time delayDelay uses displacement
Register realization, small several times sampling point time delayRealized using Farrow wave filters.Signal is S13=x after delay compensation
(n)。
The phase estimation and compensating unit 45, the purpose is to eliminate the phase deviation between signal S13 and signal S12,
Ensure that follow-up pre-distortion coefficients renewal process is normally carried out.Although exported from local oscillator to quadrature modulator 3 and frequency mixer
22 frequency of carrier signal is completely the same, but there may be phase deviation therebetween.If export to the carrier wave of quadrature modulator
Signal initial phase is 0, and real part and the imaginary part of phase deviation θ, power amplifier output signal S5 are respectively yr(t) and yi(t),
Then the simulation real baseband signal S11 after the mixing of frequency mixer 22 is:
WhereinRepresent low-pass filtering operation.After lack sampling, signal S12 is
If the real part of signal S13 and imaginary part are respectively xr(n) and xi(n), to xr(n) and xi(n) carry out D times to extract, obtain
xr(Dn) and xi(Dn), it is preferred that phase deviation θ can be estimated by following equation
WhereinWithRepresentation signal y (Dn) and signal x respectivelyrAnd signal y (Dn) and signal x (Dn),i(Dn) it
Between cross-correlation coefficient.It can be substituted when actually realizing with the inner product operation between the signal of certain length, it is preferred that phase
Position deviation θ can carry out approximate evaluation by following equation
Wherein
Signal y (Dn) and signal x is represented respectivelyrAnd signal y (Dn) and signal x (Dn),i(Dn) inner product operation between, NpTo participate in
The signal length of inner product operation.After phase estimation is completed, phase compensation is carried out to signal S1, this process is by the way that signal S1 is multiplied
WithRealize, the signal after compensating is signal S14=u (n).
The predistorter coefficient estimation unit 45, it is the iteration renewal using signal S14 and signal S12 as input that it, which is acted on,
The coefficient of predistorter 11.To illustrate the operating procedure of predistorter coefficient estimation unit 45, the first work to predistorter 11
It is briefly described as principle.
The input signal of predistorter 11 is S1=x (n), and output signal is the line of multiple nonlinear functions of input signal
Property combination, by taking common memory polynomial model as an example, it exports signal S2=xpd(n) relation with input signal is represented by
Wherein K, Q represent the non-linear order of the highest of memory polynomial model and memory depth, a respectivelyk,qRepresent that memory is more
The coefficient of item formula model, then K=5, the memory polynomial model predistorter of Q=1, its basic function collection are
{x(n),x(n)|x(n)|2,x(n)|x(n)|4,x(n-1),x(n-1)|x(n-1)|2,x(n-1)|x(n-1)|4}；
The input/output relation of predistorter 11 can also be write as matrix form
xpd=Xa,
Wherein xpdFor the column vector of the output signal composition of predistorter 11, X is observing matrix, and a is predistorter 11
Coefficient vector, is defined as follows
xpd=[xpd(n) xpd(n-1) … xpd(n-N)]T,
A=[a1,0 a2,0 … a3,1]T；
It is pointed out that here with K=5, the work of predistorter 11 is illustrated exemplified by the memory polynomial model of Q=1
Principle, but the operation principle is also suitable for the predistorter model of other any basic function collections.
The predistorter coefficient estimation unit 45 is with 11 coefficient vector a of following steps renewal predistorter：
Step 301) initializes the coefficient vector a of predistorter 11, makes the coefficient that basic function x (n) is corresponded in coefficient vector a
For 1, remaining coefficient all 0, this make it that the output signal S2 and input signal S1 of predistorter 11 are completely the same；
Step 302), collects the signal S14 of continuous DN sampling point and the signal S12 of continuous N number of sampling point, forms column vector
U=[u (0), u (1) ..., u (DN-1)]T,
Y=[y (0), y (D) ..., y (DN-D)]T
WhereinRepresenting matrix transposition；
Step 303), based on column vector u, the basic function collection generator matrix U according to predistorter 11；
Step 304), matrix U are a complex matrixs, and it is U to make its real partr, imaginary part Ui, matrix is combined into the following manner
Step 305), from matrixThe first row start, extract a line out per D rows, form matrixSuch as assume D=5,
MatrixOne shares 1000 rows, thenMatrix is included from top to bottom{ 1,6,11 ..., 996 } OK；
Step 306), calculates auxiliary vector w
WhereinRepresenting matrix is inverted, urFor the real part of column vector u, G is desired power amplifier linearity gain.
Step 307), updates the coefficient vector a of predistorter 11
anew=aold+[I jI]w
Wherein aoldFor the coefficient vector a, a of last round of iterationnewIt is unit matrix for the coefficient vector a, I of renewal, it is arranged
Number is identical with the element number of coefficient vector a；
Step 308), repeat step 302) to step 307), until reaching certain iterations, or predistorter 11
Coefficient vector a tend to restrain.
In conclusion the foregoing is merely a prefered embodiment of the invention, it is not intended to limit the scope of the present invention.
Within the spirit and principles of the invention, any modification, equivalent replacement, improvement and so on, should be included in the present invention's
Within protection domain.
Claims (3)
1. a kind of digital pre-distortion system of real signal lack sampling, it is characterised in that including time delay estimation and compensating unit, phase
Estimation and compensating unit and predistorter coefficient estimation unit, wherein：
The time delay estimation and compensating unit, the lack sampling reality base band letter with complex baseband input signal S1 and from analog-digital converter
Number S12 is input, the time-delay value of estimation signal S12 delay signals S1, and carries out time delay benefit to signal S1 according to the time delay
Repay, generate the complex baseband input signal S13 after delay compensation；
The phase estimation and compensating unit, with the complex baseband input signal S13 after delay compensation and from analog-digital converter
Lack sampling real baseband signal S12 for input, estimation signal S12 relative to signal S13 phase place, and according to the phase revolve
Turn to carry out phase compensation to signal S13, generate the complex baseband input signal S14 after phase compensation；
The predistorter coefficient estimation unit, with the complex baseband signal S14 after phase compensation and from analog-digital converter
Lack sampling real baseband signal S12 is input, in an iterative manner, estimates the renewal vector of predistorter coefficient vector, and will more
Coefficient after new is sent to predistorter.
2. a kind of digital pre-distortion system of real signal lack sampling as claimed in claim 1, it is characterised in that the time delay is estimated
Meter and compensating unit, estimate the simultaneously time delay between thermal compensation signal S1 and S12 with the following method：
Step 101), if signal S1=z (n), real part and imaginary part are respectively zr(n) and zi(n), signal S12=y (Dn), during restriction
It is integer to prolong estimate τ, in section [- τmax,τmax] in change τ values, and find and make maximum τ values on the right side of following formula, remember
For integral multiple sampling point time delayI.e.
<mrow>
<msub>
<mover>
<mi>&tau;</mi>
<mo>^</mo>
</mover>
<mi>int</mi>
</msub>
<mo>=</mo>
<munder>
<mi>argmax</mi>
<mi>&tau;</mi>
</munder>
<mo>|</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>n</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<msub>
<mi>N</mi>
<mi>t</mi>
</msub>
</munderover>
<mi>y</mi>
<mrow>
<mo>(</mo>
<mi>D</mi>
<mi>n</mi>
<mo>)</mo>
</mrow>
<msub>
<mi>z</mi>
<mi>r</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>D</mi>
<mi>n</mi>
<mo>-</mo>
<mi>&tau;</mi>
<mo>)</mo>
</mrow>
<mo>|</mo>
<mo>+</mo>
<mo>|</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>n</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<msub>
<mi>N</mi>
<mi>t</mi>
</msub>
</munderover>
<mi>y</mi>
<mrow>
<mo>(</mo>
<mi>D</mi>
<mi>n</mi>
<mo>)</mo>
</mrow>
<msub>
<mi>z</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>D</mi>
<mi>n</mi>
<mo>-</mo>
<mi>&tau;</mi>
<mo>)</mo>
</mrow>
<mo>|</mo>
</mrow>
Step 102), defines Time delay Estimation Accuracy Δ τ, in section [- 1,1], changes τ values with step delta τ, and searching make it is following
Maximum τ values, are denoted as small several times sampling point time delay on the right side of expression formulaI.e.
<mrow>
<msub>
<mover>
<mi>&tau;</mi>
<mo>^</mo>
</mover>
<mrow>
<mi>f</mi>
<mi>r</mi>
<mi>a</mi>
<mi>c</mi>
</mrow>
</msub>
<mo>=</mo>
<munder>
<mrow>
<mi>arg</mi>
<mi>max</mi>
</mrow>
<mi>&tau;</mi>
</munder>
<mo>|</mo>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>n</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<msub>
<mi>N</mi>
<mi>t</mi>
</msub>
</munderover>
<mi>y</mi>
<mrow>
<mo>(</mo>
<mrow>
<mi>D</mi>
<mi>n</mi>
</mrow>
<mo>)</mo>
</mrow>
<msub>
<mi>z</mi>
<mi>r</mi>
</msub>
<mrow>
<mo>(</mo>
<mrow>
<mi>D</mi>
<mi>n</mi>
<mo>-</mo>
<msub>
<mover>
<mi>&tau;</mi>
<mo>^</mo>
</mover>
<mi>int</mi>
</msub>
<mo>-</mo>
<mi>&tau;</mi>
</mrow>
<mo>)</mo>
</mrow>
<mo>|</mo>
<mo>+</mo>
<mo>|</mo>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>n</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<msub>
<mi>N</mi>
<mi>t</mi>
</msub>
</munderover>
<mi>y</mi>
<mrow>
<mo>(</mo>
<mrow>
<mi>D</mi>
<mi>n</mi>
</mrow>
<mo>)</mo>
</mrow>
<msub>
<mi>z</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mrow>
<mi>D</mi>
<mi>n</mi>
<mo>-</mo>
<msub>
<mover>
<mi>&tau;</mi>
<mo>^</mo>
</mover>
<mi>int</mi>
</msub>
<mo>-</mo>
<mi>&tau;</mi>
</mrow>
<mo>)</mo>
</mrow>
<mo>|</mo>
</mrow>
Step 103), signal S1 is postponedA sampled point, wherein integral multiple sampling point time delayDelay uses shift LD
Device realization, small several times sampling point time delayRealized using Farrow wave filters, signal is S13=x (n) after delay compensation.
3. a kind of digital pre-distortion system of real signal lack sampling as claimed in claim 1 or 2, it is characterised in that described pre-
Distorter coefficient estimation unit is with the coefficient vector a of following steps renewal predistorter：
Step 301) initializes the coefficient vector a of predistorter, and the coefficient for making to correspond to basic function x (n) in coefficient vector a is 1, its
Remaining coefficient all 0, this make it that the output signal S2 and input signal S1 of predistorter 11 are completely the same；
Step 302), collects the signal S14 of continuous DN sampling point and the signal S12 of continuous N number of sampling point, forms column vector
U=[u (0), u (1) ..., u (DN-1)]T,
Y=[y (0), y (D) ..., y (DN-D)]T
Wherein ()TRepresenting matrix transposition；
Step 303), based on column vector u, the basic function collection generator matrix U according to predistorter；
Step 304), matrix U are a complex matrixs, and it is U to make its real partr, imaginary part Ui, matrix is combined into the following manner
<mrow>
<mover>
<mi>U</mi>
<mo>~</mo>
</mover>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<msub>
<mi>U</mi>
<mi>r</mi>
</msub>
</mtd>
<mtd>
<mrow>
<mo>-</mo>
<msub>
<mi>U</mi>
<mi>i</mi>
</msub>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
Step 305), from matrixThe first row start, extract a line out per D rows, form matrix
Step 306), calculates auxiliary vector w
<mrow>
<mi>w</mi>
<mo>=</mo>
<msup>
<mrow>
<mo>&lsqb;</mo>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mover>
<mi>U</mi>
<mo>~</mo>
</mover>
<mi>D</mi>
</msub>
<mo>)</mo>
</mrow>
<mi>T</mi>
</msup>
<msub>
<mover>
<mi>U</mi>
<mo>~</mo>
</mover>
<mi>D</mi>
</msub>
<mo>&rsqb;</mo>
</mrow>
<mrow>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msup>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mover>
<mi>U</mi>
<mo>~</mo>
</mover>
<mi>D</mi>
</msub>
<mo>)</mo>
</mrow>
<mi>T</mi>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>u</mi>
<mi>r</mi>
</msub>
<mo>-</mo>
<mi>y</mi>
<mo>/</mo>
<mi>G</mi>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
Wherein ()-1Representing matrix is inverted, urFor the real part of column vector u, G is desired power amplifier linearity gain；
Step 307), updates the coefficient vector a of predistorter
anew=aold+[I jI]w
Wherein aoldFor the coefficient vector a, a of last round of iterationnewCoefficient vector a, I for renewal are unit matrix, its columns with
The element number of coefficient vector a is identical；
Step 308), repeat step 302) to step 307), until reaching certain iterations, or the coefficient of predistorter
Vectorial a tends to restrain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710439383.6A CN108023844B (en) | 2017-06-12 | 2017-06-12 | Digital predistortion system for real signal undersampling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710439383.6A CN108023844B (en) | 2017-06-12 | 2017-06-12 | Digital predistortion system for real signal undersampling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108023844A true CN108023844A (en) | 2018-05-11 |
| CN108023844B CN108023844B (en) | 2020-07-10 |
Family
ID=62079228
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710439383.6A Active CN108023844B (en) | 2017-06-12 | 2017-06-12 | Digital predistortion system for real signal undersampling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108023844B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109474553A (en) * | 2018-12-06 | 2019-03-15 | 上海移芯通信科技有限公司 | A kind of estimation method and system of the distorted signals parameter for terminal transmission access |
| CN112202695A (en) * | 2020-08-05 | 2021-01-08 | 重庆大学 | Under-sampling digital predistortion method and system based on Landweber iterative algorithm |
| CN112803972A (en) * | 2021-02-01 | 2021-05-14 | 西安电子科技大学 | Digital predistortion method and system suitable for MIMO transmitter full loop distortion compensation |
| CN113676426A (en) * | 2021-08-24 | 2021-11-19 | 东南大学 | Intelligent digital predistortion system and method for dynamic transmission |
| CN114745239A (en) * | 2022-03-31 | 2022-07-12 | 宁波大学 | Digital predistortion method based on single-path feedback |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002103897A1 (en) * | 2001-06-14 | 2002-12-27 | Telefonaktiebolaget Lm Ericsson | Pre-distortion using a by-pass signal |
| CN102969987A (en) * | 2012-12-17 | 2013-03-13 | 东南大学 | Undersampling-based broadband power-amplifier pre-distortion method |
| CN104283829A (en) * | 2013-07-12 | 2015-01-14 | 亚德诺半导体技术公司 | Digital pre-distortion systems in transmitters |
| CN105958952A (en) * | 2016-04-25 | 2016-09-21 | 中国科学技术大学 | Single-bit DPD (Digital Pre-Distortion) method for power amplifier |
| CN106685368A (en) * | 2017-01-03 | 2017-05-17 | 电子科技大学 | Digital predistortion system and method achieving undersampling feedback |
-
2017
- 2017-06-12 CN CN201710439383.6A patent/CN108023844B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002103897A1 (en) * | 2001-06-14 | 2002-12-27 | Telefonaktiebolaget Lm Ericsson | Pre-distortion using a by-pass signal |
| CN102969987A (en) * | 2012-12-17 | 2013-03-13 | 东南大学 | Undersampling-based broadband power-amplifier pre-distortion method |
| CN104283829A (en) * | 2013-07-12 | 2015-01-14 | 亚德诺半导体技术公司 | Digital pre-distortion systems in transmitters |
| CN105958952A (en) * | 2016-04-25 | 2016-09-21 | 中国科学技术大学 | Single-bit DPD (Digital Pre-Distortion) method for power amplifier |
| CN106685368A (en) * | 2017-01-03 | 2017-05-17 | 电子科技大学 | Digital predistortion system and method achieving undersampling feedback |
Non-Patent Citations (5)
| Title |
|---|
| 余平: "宽带射频功率放大器反馈低采样率数字预失真关键技术", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 刘发林等: "欠采样数字预失真", 《微波学报》 * |
| 吴晓芳等: "基于正交模型辨识辅助的欠采样数字预失真方法", 《***工程与电子技术》 * |
| 杨德伟等: "基于嵌入式Linux数字基带预失真控制***的设计与实现", 《实验室研究与探索》 * |
| 王华等: "解调环路非理想特性对预失真***的影响", 《北京理工大学学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109474553A (en) * | 2018-12-06 | 2019-03-15 | 上海移芯通信科技有限公司 | A kind of estimation method and system of the distorted signals parameter for terminal transmission access |
| CN112202695A (en) * | 2020-08-05 | 2021-01-08 | 重庆大学 | Under-sampling digital predistortion method and system based on Landweber iterative algorithm |
| CN112803972A (en) * | 2021-02-01 | 2021-05-14 | 西安电子科技大学 | Digital predistortion method and system suitable for MIMO transmitter full loop distortion compensation |
| CN112803972B (en) * | 2021-02-01 | 2022-04-01 | 西安电子科技大学 | Digital predistortion method and system suitable for MIMO transmitter full loop distortion compensation |
| CN113676426A (en) * | 2021-08-24 | 2021-11-19 | 东南大学 | Intelligent digital predistortion system and method for dynamic transmission |
| CN114745239A (en) * | 2022-03-31 | 2022-07-12 | 宁波大学 | Digital predistortion method based on single-path feedback |
| CN114745239B (en) * | 2022-03-31 | 2023-08-08 | 宁波大学 | Digital predistortion method based on one-way feedback |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108023844B (en) | 2020-07-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108023844A (en) | A kind of digital pre-distortion system of real signal lack sampling | |
| EP1402700B1 (en) | Adaptive predistorter based on the probability distribution function of the output amplitude | |
| US7606322B2 (en) | Digital pre-distortion technique using nonlinear filters | |
| CN102948071B (en) | Modulate unknowable digital hybrid mode power amplifier system and method | |
| Anttila et al. | Joint mitigation of power amplifier and I/Q modulator impairments in broadband direct-conversion transmitters | |
| CN102055411B (en) | Power amplifier linearization correcting circuit and method based on multi-channel feedback | |
| JP6542120B2 (en) | Method and system for matching widely spaced signals for wideband digital predistortion in a wireless communication system | |
| CN103312640B (en) | A kind of method of joint channel estimation and IQ imbalance compensation | |
| FI107212B (en) | I / Q modulator DC offset correction | |
| CN102882818B (en) | Amending method and amending system directing at unbalanced zero intermediate frequency feedback | |
| WO1999057806A1 (en) | Linearization method for amplifier, and amplifier arrangement | |
| CN106506417A (en) | A kind of digital pre-distortion system of narrowband feedback and method | |
| CN102427336A (en) | Radio frequency power amplification system with function of adaptive digital predistortion linearization | |
| CN102939716B (en) | Multi-band wideband power amplifier digital predistortion system and method | |
| CN102437822A (en) | Self-adaptive digital pre-distortion linear system of radio frequency power amplifier | |
| CN102291154A (en) | Polar coordinate transmitter | |
| CN107395294A (en) | A kind of emitter radio frequency non-ideal characteristic antidote | |
| CN107124141A (en) | The method for the digital pre-distortion verified based on complex-field matrix numerical solution adaptive error | |
| CN106161304A (en) | A kind of transmitting terminal IQ imbalance compensation method of joint channel estimation | |
| Anttila et al. | Recursive learning-based joint digital predistorter for power amplifier and I/Q modulator impairments | |
| CN113868994A (en) | Threshold value optimization method suitable for power amplifier model based on CPWL function | |
| CN105680919B (en) | It can compensate for the staggered form mimo system pre-distortion method of IQ nonequilibrium effect | |
| CN103346792A (en) | Method and device for eliminating clock jitter in analog-to-digital conversion and digital pre-distortion method | |
| CN109740225A (en) | A method of it calculates and assesses for broad-band TWT | |
| CN201947373U (en) | High efficiency digital television transmitting device |
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 |