CN110649895A - Intermediate frequency digital predistortion method using local peak correction memory polynomial - Google Patents
Intermediate frequency digital predistortion method using local peak correction memory polynomial Download PDFInfo
- Publication number
- CN110649895A CN110649895A CN201910929798.0A CN201910929798A CN110649895A CN 110649895 A CN110649895 A CN 110649895A CN 201910929798 A CN201910929798 A CN 201910929798A CN 110649895 A CN110649895 A CN 110649895A
- Authority
- CN
- China
- Prior art keywords
- local peak
- intermediate frequency
- signal
- memory polynomial
- peak value
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 18
- 238000012937 correction Methods 0.000 title claims abstract description 14
- 230000000737 periodic effect Effects 0.000 claims abstract description 30
- 238000005070 sampling Methods 0.000 claims abstract description 27
- 238000012545 processing Methods 0.000 claims description 6
- 230000010355 oscillation Effects 0.000 claims description 3
- 230000003446 memory effect Effects 0.000 abstract description 6
- 241000668842 Lepidosaphes gloverii Species 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 239000011159 matrix material Substances 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 238000009795 derivation Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000005693 optoelectronics Effects 0.000 description 2
- 238000013475 authorization Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
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
Landscapes
- Physics & Mathematics (AREA)
- Algebra (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Nonlinear Science (AREA)
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Amplifiers (AREA)
Abstract
The intermediate frequency digital predistortion method of utilizing the local peak value to correct the memory polynomial, utilize two mixers to mix the input, output signal of the ultra-wideband system into the intermediate frequency signal; sampling the intermediate frequency signal, and then deriving the sampling signal to obtain a periodic local peak value; correcting the memory polynomial model by the periodic local peak value, and modeling the ultra-wideband system by using the corrected memory polynomial model to obtain a predistortion parameter; and then, carrying out nonlinear correction on the ultra-wideband system by using the predistortion parameters. Compared with the prior art, the invention has the beneficial effects that: the invention introduces the periodic local peak value of the intermediate frequency signal to correct the memory polynomial model, and because the periodic local peak value just falls on the signal envelope, the memory effect of a long scale can be well expressed only by a few local peak values during modeling, thereby effectively reducing the model coefficient and reducing the algorithm complexity.
Description
Technical Field
The invention relates to a digital predistortion technology, in particular to an intermediate frequency digital predistortion method using a local peak correction memory polynomial.
Background
Ultra-wideband (UWB) communication has a huge data transmission rate advantage, and Radio Over Fiber (ROF) based UWB communication is a popular technology for achieving high-speed seamless coverage in a small range. FCC has allocated 3.1-10.6 GHz frequency band without authorization for ultra-wideband communication.
However, like conventional radio frequency transmission systems, the ROF-based ultra-wideband communication (UWB) system is also an analog transmission system, and signal distortion may be caused by optical subcarrier modulation and transmission nonlinearity due to various nonlinear optoelectronic devices (optical modulator, optical receiver, radio frequency amplifier, etc.). Because of its ultra-wideband nature, the nonlinear distortion is even more severe and must be mitigated by linearization techniques.
The analog predistortion technology has the characteristics of high bandwidth, low cost and the like. However, the characteristic parameters of the optoelectronic device in the ROF system may change with the change of time and temperature, and the adaptability and high stability of the system cannot be realized by using the analog predistortion. With the rapid development of DSP technology, digital predistortion technology becomes the mainstream of linearization technology.
The traditional baseband digital predistorter is developed more mature, but the sampling bandwidth is limited, so that the traditional baseband digital predistorter is not suitable for an ultra-wideband system. For a communication system with dual-band or tri-band signals coexisting, researchers have proposed models such as 2D-DPD and 3D-DPD. However, as the band increases, the coefficients of the model will increase rapidly, and the algorithm complexity increases greatly. Therefore, for a communication system with more bands coexisting and other ultra-wideband systems, an intermediate frequency/radio frequency digital predistortion method is recently proposed, and a memory polynomial is used for nonlinear modeling of the ultra-wideband system.
The nonlinearity of the ultra-wideband signal, which is modulated, transmitted and received by an ROF system, contains obvious memory effect, and when the intermediate frequency is sampled, the sampling time interval is still very small, the memory effect of the signal envelope time scale is very deep compared with the intermediate frequency, and the long-scale memory effect is shown. This makes the memory depth of the digital predistortion model based on the memory polynomial too long, the coefficients of the model too many, and the algorithm too complicated. Therefore, how to reduce the model coefficients is an important issue.
Disclosure of Invention
The invention aims to provide an intermediate frequency digital predistortion method using a local peak correction memory polynomial, and a long-scale memory effect can be expressed only by a few periodic local peaks during modeling.
The intermediate frequency digital predistortion method using local peak correction memory polynomial includes the following steps:
step A: mixing an input signal and an output signal of the ultra-wideband system into an intermediate frequency signal by using a first mixer and a second mixer respectively;
and B: sampling the two intermediate frequency signals to obtain sampling signals;
and C: carrying out digital signal processing on the sampling signal to obtain a periodic local peak value of the sampling signal;
step D: correcting a memory polynomial model by using the periodic local peak value, and modeling the ultra-wideband system by using the corrected memory polynomial model to obtain a predistortion parameter;
the memory polynomial model is as follows:
wherein, x (n) is the intermediate frequency sampling signal, w (n) is the periodic local peak value thereof, J and P are the nonlinear order and the memory depth of the intermediate frequency signal, K and Q are the nonlinear order and the memory depth related to the periodic local peak value, a is the central signal, and b is the periodic local peak value;
step E: and then, correcting the nonlinearity of the ultra-wideband system by using the predistortion parameters.
In the foregoing technical solution, further, in step a, the local oscillation frequencies of the first mixer and the second mixer are between 3000MHz and 12000 MHz.
In the above technical solution, further, in the step a, the frequency of the input signal is between 3000MHz and 12000 MHz; the frequency of the output signal is between 3000MHz and 12000 MHz.
In the above technical solution, further, in step a or B, the intermediate frequency signal frequency is between 500MHz and 1500 MHz.
In the above technical solution, further, in the step B, an oscilloscope is used to sample the intermediate frequency signal.
In the above technical solution, further, in step C, the digital signal processing means that a derivative is obtained from the digital sequence of the sampling signal, and in a process of changing the derivative from positive to negative, a point where an absolute value of the derivative is the smallest is the periodic local peak.
Compared with the prior art, the invention has the beneficial effects that: the invention introduces the periodic local peak value of the intermediate frequency signal to correct the memory polynomial model, and because the periodic local peak value just falls on the signal envelope, the memory effect of a long scale can be well expressed only by a few local peak values during modeling, thereby effectively reducing the model coefficient and reducing the algorithm complexity.
Drawings
Fig. 1 is a graph of a sampled signal according to the present invention versus periodic local peaks.
FIG. 2 is a schematic diagram of the numerical predictive true modeling of the present invention.
Fig. 3 is a schematic diagram of the digital predictive true modeled ultra-wideband system correction according to the present invention.
FIG. 4 is a schematic of the sampled data and its derivatives according to the present invention.
FIG. 5 is an enlarged view of a schematic of the sampled data and its derivatives according to the present invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying examples.
As shown in fig. 1-5, the intermediate frequency digital predistortion method using local peak correction memory polynomial includes the following steps:
step A: mixing an input signal and an output signal of the ultra-wideband system into an intermediate frequency signal by using a first mixer and a second mixer respectively;
and B: sampling the two intermediate frequency signals to obtain sampling signals;
and C: carrying out digital signal processing on the sampling signal to obtain a periodic local peak value of the sampling signal;
step D: correcting a memory polynomial model by using the periodic local peak value, and modeling the ultra-wideband system by using the corrected memory polynomial model to obtain a predistortion parameter;
the memory polynomial model is as follows:
wherein, x (n) is the intermediate frequency sampling signal, w (n) is the periodic local peak value thereof, J and P are the nonlinear order and the memory depth of the intermediate frequency signal, K and Q are the nonlinear order and the memory depth related to the periodic local peak value, a is the central signal, and b is the periodic local peak value;
step E: and then, correcting the nonlinearity of the ultra-wideband system by using the predistortion parameters.
In step a, the local oscillation frequencies of the first mixer and the second mixer are between 3000MHz and 12000 MHz. The input signal of the ultra-wideband system is provided by a signal generator.
In the step A, the frequency of the input signal is between 3000MHz and 12000 MHz; the frequency of the output signal is 3000MHz to 12000 MHz.
In step A or B, the intermediate frequency signal frequency is between 500MHz and 1500 MHz.
And in the step B, sampling the intermediate frequency signal by adopting an oscilloscope.
In step C, the digital signal processing means performing derivation on the digital sequence of the sampling signal, and a maximum value is the periodic local peak value. Specifically, the derivative is zero, and the sample is not necessarily sampled exactly due to the interval of sampling, and when the derivative is derived for the array of sampled signals, the derivative is greater than zero first and less than zero later, that is, the periodic local peak in this embodiment.
During modeling, a signal generator is used for directly inputting an original signal into an ultra-wideband system, the input and output signals of the ultra-wideband system are respectively subjected to down-conversion to intermediate frequency through a first mixer and a second mixer, then an oscilloscope samples the signals into sampling signals (digital signals), the sampling signals are sent to a matlab software for a computer to perform a digital signal processor in an off-line manner, namely, derivation is performed on a digital sequence of the sampling signals, and an inflection point, which is larger than zero firstly and smaller than zero later, of the derivative is the periodic local peak value;
as can be seen from the enlarged partial view, at the local peak, the derivative thereof is positive but small, close to 0, and at the next point, the derivative becomes negative, as shown in fig. 5. Therefore, the position of the peak value can be determined by searching the turning points of positive and negative numbers in the derivative curve;
then, correcting a memory polynomial model by using the periodic local peak value, modeling the ultra-wideband system by using the corrected memory polynomial model, and determining a predistortion parameter (parameter matrix); specifically, after the local peak value is obtained, modeling is performed by using a memory polynomial. In the modeling process, the memory polynomial can be simplified and expressed
The non-linear order and memory depth J, P of the intermediate frequency signal can be respectively 5 and 3, and the non-linear order and memory depth K, Q related to the periodic local peak value can also be respectively 5 and 3. While higher order non-linearity orders and memory depth are negligible. Thus, the memory polynomial can be simplified to
The modeling process implemented using matlab software is as follows:
1. constructing an input matrix:
b1 ═ x; I/V/X is an input signal
b2＝abs(x).*x；
b3＝abs(x).^2.*x；
b4＝abs(x).^3.*x；
b5＝abs(x).^4.*x；
c1＝w；
c2＝abs(w).*w；
c3＝abs(w).^2.*w；
c4＝abs(w).^3.*w；
c5＝abs(w).^4.*w；
b＝[b1'；b2'；b3'；b4'；b5'；0 b1(1:(n-1))'；0 b2(1:(n-1))'；0 b3(1:(n-1))'；0 b4(1:(n-1))'；0 b5(1:(n-1))'；0 0 b1(1:(n-2))'；0 0 b2(1:(n-2))'；0 0 b3(1:(n-2))'；0 0 b4(1:(n-2))'；0 0 b5(1:(n-2))']'；
c＝[c1'；c2'；c3'；c4'；c5'；0 c1(1:(n-1))'；0 c2(1:(n-1))'；0 c3(1:(n-1))'；0 c4(1:(n-1))'；0 c5(1:(n-1))'；0 0 c1(1:(n-2))'；0 0 c2(1:(n-2))'；0 0 c3(1:(n-2))'；0 0 c4(1:(n-2))'；0 0 c5(1:(n-2))']'；
d ═ b; c ]; //// input matrix, the simplified modified memory polynomial may then be expressed as: out is the output signal,
2. calculating a parameter matrix
h pinv (d) out; the inverse matrix is denoted by h, which is the parameter matrix to be obtained, the first half being the parameter a and the second half being the parameter b.
In step E, in the on-line correction stage, the rf signal output by the signal generator is first converted into an if signal by the down converter, the if signal is sampled by the digital predistorter, and is predistorted by the parameter matrix obtained during modeling, and then output to the up converter, and is converted back to the rf signal again, and then input to the ultra-wideband system. The down converter is the same as the mixer in modeling, while the up converter has an input signal frequency between 500MHz and 1500MHz and an output signal frequency between 3000MHz and 12000 MHz.
The present invention is not limited to the above-described embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the spirit of the present invention.
Claims (6)
1. The intermediate frequency digital predistortion method using the local peak correction memory polynomial is characterized by comprising the following steps:
step A: mixing an input signal and an output signal of the ultra-wideband system into an intermediate frequency signal by using a first mixer and a second mixer respectively;
and B: sampling the two intermediate frequency signals to obtain sampling signals;
and C: carrying out digital signal processing on the sampling signal to obtain a periodic local peak value of the sampling signal;
step D: correcting a memory polynomial model by using the periodic local peak value, and modeling the ultra-wideband system by using the corrected memory polynomial model to obtain a predistortion parameter;
the memory polynomial model is as follows:
wherein, x (n) is the intermediate frequency sampling signal, w (n) is the periodic local peak value thereof, J and P are the nonlinear order and the memory depth of the intermediate frequency signal, K and Q are the nonlinear order and the memory depth related to the periodic local peak value, a is the central signal, and b is the periodic local peak value;
step E: and then, the predistortion parameters are utilized to correct the nonlinearity of the ultra-wideband system.
2. The method for digital predistortion of an intermediate frequency using a local peak correction memory polynomial according to claim 1, wherein in step a, the local oscillation frequencies of the first mixer and the second mixer are between 3000MHz and 12000 MHz.
3. The intermediate frequency digital predistortion method using local peak correction memory polynomial as set forth in claim 1, characterized in that in step a, the frequency of the input signal is between 3000 MHz-12000 MHz; the frequency of the output signal is between 3000MHz and 12000 MHz.
4. The method for digital predistortion of an intermediate frequency using a local peak correction memory polynomial as set forth in claim 1, wherein in step a or B, the frequency of the intermediate frequency signal is between 500MHz and 1500 MHz.
5. The intermediate frequency digital predistortion method using local peak correction memory polynomial as set forth in claim 1, wherein in step B, an oscilloscope is used for sampling the intermediate frequency signal.
6. The method of claim 1, wherein in step C, the digital signal processing is to derive the digital sequence of the sampled signals, and during the change of the derivative from positive to negative, the point where the absolute value of the derivative is the smallest is the periodic local peak.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910929798.0A CN110649895A (en) | 2019-09-27 | 2019-09-27 | Intermediate frequency digital predistortion method using local peak correction memory polynomial |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910929798.0A CN110649895A (en) | 2019-09-27 | 2019-09-27 | Intermediate frequency digital predistortion method using local peak correction memory polynomial |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110649895A true CN110649895A (en) | 2020-01-03 |
Family
ID=69011823
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910929798.0A Pending CN110649895A (en) | 2019-09-27 | 2019-09-27 | Intermediate frequency digital predistortion method using local peak correction memory polynomial |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110649895A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060232332A1 (en) * | 2005-04-13 | 2006-10-19 | Braithwaite Richard N | Adaptive predistortion linearized amplifier system employing selective sampling |
| CN101267187A (en) * | 2008-05-07 | 2008-09-17 | 北京北方烽火科技有限公司 | A self-adapted pre-distortion method and system for broadband linear power amplifier |
| CN108881083A (en) * | 2018-06-27 | 2018-11-23 | 云南大学 | Broadband ROF system envelope assists RF/IF digital pre-distortion technology |
-
2019
- 2019-09-27 CN CN201910929798.0A patent/CN110649895A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060232332A1 (en) * | 2005-04-13 | 2006-10-19 | Braithwaite Richard N | Adaptive predistortion linearized amplifier system employing selective sampling |
| CN101267187A (en) * | 2008-05-07 | 2008-09-17 | 北京北方烽火科技有限公司 | A self-adapted pre-distortion method and system for broadband linear power amplifier |
| CN108881083A (en) * | 2018-06-27 | 2018-11-23 | 云南大学 | Broadband ROF system envelope assists RF/IF digital pre-distortion technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9735741B2 (en) | Receivers for digital predistortion | |
| CN102143107B (en) | Method and device for realizing pre-distortion of digital baseband | |
| CN1288341A (en) | Method and device for reducing adjacent channel power in radio communication system | |
| CN107359864B (en) | Self-adaptive agile digital predistortion method for frequency agile power amplifier | |
| CN104219182B (en) | RF power amplification linearized system based on wideband modulator | |
| CN207706212U (en) | A kind of short range digital pre-distortion processing device | |
| CN105656834A (en) | Digital correction method for IQ channel mismatch of novel broadband receiver | |
| CN104883140A (en) | Digital predistortion device based on broadband radio frequency power amplifier | |
| CN105305980A (en) | Feedback type pre-distortion linearization method | |
| CN107592083A (en) | The digital pre-distortion system and its method of millimeter wave broadband power amplifier | |
| CN108199690B (en) | Power amplifier digital predistortion device and method based on cubic spline and with limited DDR function model | |
| US20090247095A1 (en) | Method and system to prevent harmonics from causing distortion in a communications system | |
| WO2021174463A1 (en) | Digital predistortion device and transmitter | |
| Kumar et al. | Bandlimited DPD adapted APD for 5G communication | |
| CN110649895A (en) | Intermediate frequency digital predistortion method using local peak correction memory polynomial | |
| CN109039969B (en) | Method for implementing broadband digital predistorter | |
| CN110266276A (en) | 5G distributed power amplifier low-speed digital pre-distortion method | |
| CN102904846B (en) | A kind of digital pre-distortion processing method adapting to fast changed signal | |
| CN202261371U (en) | Mixed-mode radio-frequency zooming system | |
| CN109565482A (en) | Transmitter and digital pre-distortion calibration method | |
| CN113612453A (en) | Low-sampling-rate feedback digital predistortion correction method and device | |
| Vu et al. | A LUT-based scheme for LNA linearization in direct RF sampling receivers | |
| CN106712728A (en) | Method for compensating power amplifier predistortion in satellite communication | |
| CN101414804B (en) | Power amplifier and nonlinearity correction method, apparatus thereof | |
| CN113364473A (en) | Digital predistortion device and transmitter |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200103 |