CN102938638B - Cross coupling modeling method of concurrency multiband nonlinear system and linear device - Google Patents
Cross coupling modeling method of concurrency multiband nonlinear system and linear device Download PDFInfo
- Publication number
- CN102938638B CN102938638B CN201210416973.4A CN201210416973A CN102938638B CN 102938638 B CN102938638 B CN 102938638B CN 201210416973 A CN201210416973 A CN 201210416973A CN 102938638 B CN102938638 B CN 102938638B
- Authority
- CN
- China
- Prior art keywords
- concurrent
- mutual coupling
- wave band
- multiband
- type multi
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Amplifiers (AREA)
Abstract
The invention discloses a cross coupling modeling method of a concurrency multiband nonlinear system and a linear device. The cross coupling modeling method comprises the following steps of: dividing a concurrency multiband signal into baseband data sequences of each wave band through a cross coupling type multi-input digital pre-distorter for digital pre-distortion; synchronously inputting the baseband data sequences of the wave bands into corresponding baseband variable radio frequency modules; furthermore, capturing distorted baseband data of each wave band in a concurrency multiband distorted radio frequency signal of the output end of a concurrency multiband non-linear system through the concurrency multiband non-linear property; and directly inputting distorted base band data of each wave band and original base band data of each wave band in the concurrency multiband signal into a multi-input cross coupling digital pre-distortion parameter trainer for extracting a pre-distorted parameter. The cross coupling modeling method has the advantages that the requirement of a rate of a digital-to-analogue converter (DA) in the baseband variable radio frequency module can be reduced; the quantity of digital signal processing equipment can be reduced; and the resource cost of a digital signal processing device can be saved.
Description
Technical field
The present invention relates to modeling method and the digital pre-distortion technology of non linear system, especially relate to a kind of mutual coupling modeling method and linearization device of concurrent multiband nonlinear system.
Background technology
Nonlinear device is ubiquitous, and most of active device and equipment all exist non-linear, just nonlinear strong and weak different.If the non-linear comparison of device is weak, be called almost linear device, what can be existed non-linearly ignores.And for strong nonlinearity device (as power amplifier, low noise amplifier and photo-detector etc.), it is non-linear becomes very important problem.This nonlinear characteristic of strong nonlinearity device can cause the introducing of harmonic wave and intermodulation composition, not only reduces signal to noise ratio like this, and can make mutually to disturb between channel, causes the distortion distortion of signal.But be most of equipment in communication system or radar system be all non-linear equipment, wherein part active equipment is non-linear very strong, must consider the nonlinear characteristic of whole system, therefore there is nonlinear system for these, we are referred to as non linear system.But traditional non linear system is generally the non linear system of arrowband, in order to can compatible multiband, improve the overall utilization rate of system, concurrent multi-band system receives much concern.For wireless communication system, it is a typical non linear system, and because wireless communication system standards is numerous, and these standards are by long-term co-existence.But different standards uses different frequency ranges, some standards in 3G and 4G network in addition, a standard uses multiple different frequency range.Standard is more, and frequency range is more, and different frequency range needs different non-linear equipments (as radio-frequency (RF) receiving and transmission module) at present, make a base station there is many complete equipments, finally cause base station volume excessive, addressing difficulty, energy consumption is too high, and operation cost such as to increase greatly at the problems.In order to address these problems, support that the concurrent multi-band communication device of multiple standards and multiple frequency range receives much concern, support that the multiband multimode radio-frequency power amplifier being applied to communication system of at least two frequency ranges is by successful design at present.But the non-linear of radio-frequency power amplifier is the main source causing communication system signal distortion, when concurrent multiband radio-frequency power amplifier is applied to the communication system that multiple network coexists, multiple frequency band signals is loaded on radio-frequency power amplifier simultaneously, the distorted signals of communication system will be more complicated and serious, therefore proposes more harsh requirement to the modeling of concurrent multi-band communication systems and linearisation.Because concurrent multi-wave signal comprises the signal of multiple wave band, if adopt conventional method to carry out modeling and linearisation to concurrent multiband radio-frequency (RF) power amplification, propose very high requirement by the switching rate of analog to digital converter and digital to analog converter with the disposal ability of Digital Signal Processing (FPGA or the DSP) device realizing predistorter, the requirement of the speed proposed in some situations and computing capability aspect is present stage, cannot be met in the quite a while even from now on all at all.Even if sometimes can barely meet proposed requirement, system cost also can be caused greatly to rise, and system power dissipation increases greatly, and device fabrication and operation cost sharply rise.In addition, when Nonlinear Modeling and pre-distortion parameters extraction are carried out to the base band data of concurrent multi-wave signal, adopting conventional method to have very strong digital signal processing capability by needing signal handling equipment, taking a large amount of digital signal processor resources, design is complicated, and processing speed is slower.Therefore, modeling is carried out to concurrent multiband nonlinear system in the urgent need to finding new method and nonlinear model and realizes digital pre-distortion linearization.
Summary of the invention
Technical problem to be solved by this invention is to provide and a kind ofly can adopts the analog to digital converter and digital to analog converter that conversion rate is lower, and it is less to take digital signal processor resource, arithmetic speed is the mutual coupling modeling method of concurrent multiband nonlinear system and linearization device faster.
The present invention solves the problems of the technologies described above adopted technical scheme: a kind of mutual coupling modeling method of concurrent multiband nonlinear system, comprises the following steps:
(1) the raw baseband data sequence of each wave band in concurrent multiband primary signal is respectively inputted in a base band varying firing rate module, each base band varying firing rate module exports after the raw baseband data sequence of corresponding wave band is converted to radiofrequency signal, and the radiofrequency signal synchronism output of each wave band;
(2) radiofrequency signal of each wave band of synchronism output is carried out power combing, obtain concurrent multiband original radio frequency signal;
(3) concurrent multiband original radio frequency signal inputted in concurrent multiband nonlinear system, drive concurrent multiband nonlinear system, concurrent multiband nonlinear system exports concurrent multiband distortion radiofrequency signal;
(4) the distortion baseband data sequence of concurrent multiband distortion radiofrequency signal at each wave band of concurrent multiband nonlinear system output is extracted;
(5) using the input as mutual coupling type multi input nonlinear model of the real part of the raw baseband data sequence of each wave band in concurrent multiband primary signal and imaginary part, using concurrent multiband distortion radiofrequency signal in the real part of the distortion baseband data sequence of each wave band and imaginary part as the output of mutual coupling type multi input nonlinear model, in parameter training device, utilize parameter extraction Algorithm for Solving to obtain mutual coupling type multi input nonlinear model shape parameter;
(6) by mutual coupling type multi input nonlinear model Parameter transfer in mutual coupling type multi input nonlinear model;
(7) be input in mutual coupling type multi input nonlinear model by the real part of the raw baseband data sequence of each wave band in concurrent multiband primary signal and imaginary part, mutual coupling type multi input nonlinear model will export real part and the imaginary part of the emulation baseband data sequence of each wave band in concurrent multiband distorted signal;
(8) the emulation baseband data sequence of each wave band in concurrent multiband distorted signal is respectively inputted in a base band varying firing rate module, each base band varying firing rate module exports after the emulation baseband data sequence of corresponding wave band being converted to emulation radiofrequency signal, and the emulation radiofrequency signal synchronism output of each wave band;
(9) the emulation radiofrequency signal of each wave band of synchronism output is carried out power combing, obtain concurrent multiband emulation radiofrequency signal;
(10) concurrent multiband emulation radiofrequency signal and concurrent multiband distortion radiofrequency signal are compared, the accuracy of checking mutual coupling type multi input nonlinear model, if this concurrent multiband emulation radiofrequency signal reaches predetermined simulation nicety, the mutual coupling type multi input nonlinear model shape parameter that then step (5) obtains is effective nonlinear model shape parameter, and modeling process terminates; If this concurrent multiband emulation radiofrequency signal can not meet model accuracy requirement, then return the extraction that step (1) repeats mutual coupling type multi input nonlinear model shape parameter.
A kind of mutual coupling linearization device of concurrent multiband nonlinear system, comprise mutual coupling type multi input digital predistorter, multiple base band varying firing rate module, mixer, concurrent multiband nonlinear characteristic extraction element and mutual coupling type multi input digit pre-distortion parameters training aids, the described quantity of base band varying firing rate module and the wave band quantity Matching of concurrent multi-wave signal, described mutual coupling type multi input digital predistorter is provided with multipair input/output terminal, often pair of input/output terminal of described mutual coupling type multi input digital predistorter is for accessing the primary signal of a wave band in concurrent multi-wave signal and exporting the predistortion base band data of this wave band, the input of each base band varying firing rate module respectively accesses the predistortion base band data of a wave band in concurrent multi-wave signal, the output of each base band varying firing rate module is all connected with the input of described mixer, the output of described mixer is used for being connected with the input of concurrent multiband nonlinear system, the input of described concurrent multiband nonlinear characteristic extraction element is for connecting the output of concurrent multiband nonlinear system, the output of described concurrent multiband nonlinear characteristic extraction element is connected with described mutual coupling type multi input digit pre-distortion parameters training aids, described mutual coupling type multi input digit pre-distortion parameters training aids accesses the raw baseband data sequence of concurrent each wave band of multiband primary signal, described mutual coupling type multi input digit pre-distortion parameters training aids passes to described mutual coupling type multi input digital predistorter by extracting the pre-distortion parameters obtained.
Described mutual coupling type multi input digit pre-distortion parameters training aids is neural net digit pre-distortion parameters training aids, based on the pre-distortion parameters training aids of singular value decomposition or recursion lowest mean square pre-distortion parameters training aids, the type of described mutual coupling type multi input digit pre-distortion parameters training aids is mated with described mutual coupling type multi input digital predistorter.
Compared with prior art, the invention has the advantages that mutual coupling modeling method is by corresponding for the raw baseband data sequence inputting of each wave band synchronous base band varying firing rate module, and mutual coupling linearization device carries out predistortion by mutual coupling type multi input digital predistorter to concurrent multiband primary signal, and by corresponding for the input of the predistortion baseband data sequence of each wave band synchronous base band varying firing rate module, which decrease the requirement to digital to analog converter (DA) speed in base band varying firing rate module, in addition, the base band data of each wave band in the concurrent multiband distortion radiofrequency signal of concurrent multiband nonlinear system output is caught by concurrent multiband nonlinear characteristic extraction element, and the distortion base band data of each wave band and the raw baseband data of each wave band are directly inputted in multi input Coupling Model parameter training device or pre-distortion parameters training aids carry out parameter extraction, not only greatly reduce the requirement of the sample rate to the analog to digital converter of catching distortion base band data (AD), and the demand significantly reduced digital signal processing capability, save digital signal processor resource overhead.
Accompanying drawing explanation
Fig. 1 is the theory diagram of mutual coupling modeling method of the present invention;
Fig. 2 is the theory diagram of mutual coupling modeling method embodiment of the present invention;
Fig. 3 is the frequency spectrum comparison diagram of 1001 type CDMA2000 modulated radio signals of concurrent two waveband emulation radiofrequency signal and concurrent two waveband distortion radiofrequency signal medium and low frequency wave band in mutual coupling modeling method embodiment of the present invention;
Fig. 4 is the frequency spectrum comparison diagram of the single carrier WCDMA modulated radio signal of concurrent two waveband emulation radiofrequency signal and concurrent two waveband distortion radiofrequency signal medium-high frequency wave band in the embodiment of mutual coupling modeling method of the present invention;
Fig. 5 is mutual coupling linearization device theory diagram of the present invention;
Fig. 6 is the theory diagram of mutual coupling linearization device embodiment of the present invention.
Embodiment
Below in conjunction with accompanying drawing embodiment, the present invention is described in further detail.
As shown in Figure 1, the invention provides a kind of mutual coupling modeling method of concurrent multiband nonlinear system, comprise the following steps:
(1) the raw baseband data sequence of each wave band in concurrent multiband primary signal is respectively inputted in a base band varying firing rate module, each base band varying firing rate module exports after the raw baseband data sequence of corresponding wave band is converted to radiofrequency signal, and the radiofrequency signal synchronism output of each wave band;
(2) radiofrequency signal of each wave band of synchronism output is carried out power combing, obtain concurrent multiband original radio frequency signal;
(3) concurrent multiband radiofrequency signal inputted in concurrent multiband nonlinear system, drive concurrent multiband nonlinear system, concurrent multiband nonlinear system exports concurrent multiband distortion radiofrequency signal;
(4) the distortion baseband data sequence of concurrent multiband distortion radiofrequency signal at each wave band of concurrent multiband nonlinear system output is extracted;
(5) real part of the raw baseband data sequence of each wave band in concurrent multiband primary signal and imaginary part and concurrent multiband distortion radiofrequency signal are input in mutual coupling type multi input nonlinear model parameter training device at the real part of the distortion baseband data sequence of each wave band and imaginary part carry out model parameter training, utilize parameter extraction Algorithm for Solving to obtain mutual coupling type multi input nonlinear model shape parameter;
(6) by mutual coupling type multi input nonlinear model Parameter transfer in mutual coupling type multi input nonlinear model;
(7) be input in mutual coupling type multi input nonlinear model by the real part of the raw baseband data sequence of each wave band in concurrent multiband primary signal and imaginary part, mutual coupling type multi input nonlinear model will export real part and the imaginary part of the emulation baseband data sequence of each wave band in concurrent multiband distorted signal;
(8) the emulation baseband data sequence of each wave band in concurrent multiband distorted signal is respectively inputted in a base band varying firing rate module, each base band varying firing rate module exports after the emulation baseband data sequence of corresponding wave band is converted to model radiofrequency signal, and the model radiofrequency signal synchronism output of each wave band;
(9) the model radiofrequency signal of each wave band of synchronism output is carried out power combing, obtain concurrent multiband emulation radiofrequency signal;
(10) concurrent multiband emulation radiofrequency signal and concurrent multiband distortion radiofrequency signal are compared, the accuracy of checking mutual coupling type multi input nonlinear model, if this concurrent multiband emulation radiofrequency signal reaches predetermined simulation nicety, the mutual coupling type multi input nonlinear model shape parameter that then step (5) obtains is effective nonlinear model shape parameter, and modeling process terminates; If this concurrent multiband emulation radiofrequency signal can not meet model accuracy requirement, then return the extraction that step (1) repeats mutual coupling type multi input nonlinear model shape parameter.
In modeling method of the present invention, raw baseband data sequence due to each wave band in concurrent multiband primary signal respectively inputs in a base band varying firing rate module, each base band varying firing rate module exports after the raw baseband data sequence of corresponding wave band is converted to radiofrequency signal, because process respectively each wave band in concurrent multiband primary signal, thus, lower to the requirement of digital to analog converter (DA) in base band varying firing rate module, the lower and lower-cost digital to analog converter of speed can be adopted to realize this programme.In addition in the present invention, the raw baseband data sequence of each wave band in concurrent multiband primary signal and concurrent multiband distortion radiofrequency signal are directly inputted in mutual coupling type multi input nonlinear model parameter training device at the real part of the distortion baseband data sequence of each wave band and imaginary part and carry out model parameter training and obtain mutual coupling type multi input nonlinear model shape parameter; Process need not be carried out by digital signal processing appts to each wave band base band data and obtain equivalent concurrent multiband base band data, save digital signal processor resource overhead.
Embodiment: choose the checking of concurrent two waveband non linear system and carry out mutual coupling modeling method of the present invention, wherein in concurrent two waveband primary signal, the 1st wave band is 1001CDMA2000 signal (the CDMA2000 signal of 4 carrier waves of 1900MHz, and middle two carrier wave vacancies), 2nd wave band is the single carrier WCDMA signal of 1960MHz, concurrent two waveband non linear system is power bands is export the Doherty power amplifier that peak power is 51dBm, and the power bands of this power amplifier is 1880-1980MHz.
As shown in Figure 2, adopt mutual coupling modeling method of the present invention to carry out modeling to concurrent two waveband non linear system, specifically comprise the following steps:
(1) by raw baseband data sequence inputting the 1st base band varying firing rate module of the 1st wave band in concurrent two waveband primary signal, by in raw baseband data sequence inputting the 2nd base band varying firing rate module of the 2nd wave band in concurrent two waveband primary signal, the original radio frequency signal of the 1st wave band exports by the 1st base band varying firing rate module, the original radio frequency signal of the 2nd wave band exports by the 2nd base band varying firing rate module, and the original radio frequency signal of the original radio frequency signal of the 1st wave band and the 2nd wave band keeps synchronous;
(2) original radio frequency signal of the original radio frequency signal of the 1st wave band and the 2nd wave band be synchronously input in the 1st mixer and carry out power combing, the 1st mixer exports concurrent two waveband original radio frequency signal corresponding to concurrent two waveband primary signal;
(3) concurrent two waveband original radio frequency signal inputted in concurrent two waveband non linear system, drive concurrent two waveband non linear system, concurrent two waveband non linear system exports concurrent two waveband distortion radiofrequency signal; Concurrent two waveband distortion radiofrequency signal is that concurrent two waveband original radio frequency signal is by producing the signal after distortion after concurrent two waveband non linear system;
(4) the non-linear extraction element of concurrent multiband is adopted to extract the distortion baseband data sequence of concurrent two waveband distortion radiofrequency signal at the 1st wave band and the distortion baseband data sequence of the 2nd wave band of concurrent two waveband non linear system output, the name that wherein the non-linear extraction element of concurrent multiband adopts inventor to propose is called a kind of multiband multimode radio-frequency non-linearity of power amplifier feature extraction device, it applies for patent of invention, the applying date is on April 27th, 2012, and application number is 201210129607.0;
(5) by the real part I of the raw baseband data sequence of the 1st wave band in concurrent two waveband primary signal
in_1(n) and imaginary part Q
in_1the real part I of the raw baseband data sequence of the 2nd wave band in (n), concurrent two waveband primary signal
in_2(n) and imaginary part Q
in_2n (), concurrent two waveband distortion radiofrequency signal are at the real part I of the distortion baseband data sequence of the 1st wave band
out_1 (n) and imaginary part Q
out_1(n) and the real part I of concurrent two waveband distortion radiofrequency signal in the distortion baseband data sequence of the 2nd wave band
out_2(n) and imaginary part Q
out_2n () is input in mutual coupling type multi input nonlinear model parameter training device and carries out model parameter training, obtain mutual coupling type multi input nonlinear model shape parameter, wherein mutual coupling type multi input nonlinear model parameter training device adopts real number Time-Delay Recurrent RBF Neural Networks model parameter training aids, wherein n ∈ [1......N], N is the length of training data sequence here;
(6) by mutual coupling type multi input nonlinear model Parameter transfer in mutual coupling type multi input nonlinear model, wherein mutual coupling type multi input nonlinear model adopts real number Time-Delay Recurrent RBF Neural Networks model;
(7) by the real part I of the raw baseband data sequence of the 1st wave band in concurrent two waveband primary signal
in_1(n) and imaginary part Q
in_1the real part I of the raw baseband data sequence of the 2nd wave band in (n) and concurrent multiband primary signal
in_2(n) and imaginary part Q
in_2n () is input in mutual coupling type multi input nonlinear model, mutual coupling type multi input nonlinear model exports the emulation baseband data sequence of the 2nd wave band in the emulation baseband data sequence of the 1st wave band in concurrent two waveband distorted signal and concurrent two waveband distorted signal;
(8) the emulation baseband data sequence of the 1st wave band in concurrent two waveband distorted signal is inputted in the 3rd base band varying firing rate module, the emulation baseband data sequence of the 2nd wave band in concurrent two waveband distorted signal is synchronously inputted in the 4th base band varying firing rate module, export after the emulation baseband data sequence of the 1st wave band is converted to the model radiofrequency signal of the 1st wave band by the 3rd base band varying firing rate module, export after the emulation baseband data sequence of the 2nd wave band is converted to the emulation radiofrequency signal of the 2nd wave band by the 4th base band varying firing rate module, the emulation radiofrequency signal of the emulation radiofrequency signal of the 1st wave band and the 2nd wave band keeps synchronous,
(9) the emulation radiofrequency signal of the emulation radiofrequency signal of the 1st wave band and the 2nd wave band synchronously inputted in the 2nd mixer and carry out power combing, the 2nd mixer exports concurrent two waveband emulation radiofrequency signal;
(10) concurrent two waveband emulation radiofrequency signal and concurrent two waveband distortion radiofrequency signal are compared, concurrent two waveband model radiofrequency signal meets model accuracy requirement, modeling process terminates, and the mutual coupling type multi input nonlinear model shape parameter obtained in step (5) is modeling parameters.
In the present embodiment, digital to analog converter (DA) in 1st base band varying firing rate module, the 2nd base band varying firing rate module, the 3rd base band varying firing rate module and the 4th base band varying firing rate module can adopt speed to be the 3-11 rank intermodulation component that 60-110MSa/S(considers wave band 1 and the maximum useful signal of wave band 2) digital to analog converter, reduce the requirement to the digital to analog converter in base band varying firing rate module, cost is lower.
As the frequency spectrum comparison diagram that Fig. 3 is concurrent two waveband emulation radiofrequency signal and concurrent two waveband distortion radiofrequency signal medium and low frequency wave band 1001 CDMA2000 in the present embodiment; Fig. 4 is the frequency spectrum comparison diagram of concurrent two waveband emulation radiofrequency signal and concurrent two waveband distortion radiofrequency signal medium-high frequency wave band single carrier WCDMA in the embodiment of the present invention; From Fig. 3 and Fig. 4, we can find out no matter at wave band 1 or wave band 2, the strong nonlinearity characteristic of concurrent two waveband non linear system can be described accurately by the modeling method of this concurrent multiband nonlinear system, thus this modeling method known is while guarantee modeling accuracy, greatly reduce the speed of analog to digital converter and digital to analog converter, cost is reduced greatly, and the computational resource that extraction of model parameters is consumed reduce greatly.
As shown in Figure 5, present invention also offers a kind of mutual coupling linearization device of concurrent multiband nonlinear system, comprise mutual coupling type multi input digital predistorter, multiple base band varying firing rate module, mixer, concurrent multiband nonlinear characteristic extraction element and mutual coupling type multi input digit pre-distortion parameters training aids, the quantity of base band varying firing rate module and the wave band quantity Matching of concurrent multi-wave signal, mutual coupling type multi input digital predistorter is provided with multipair input/output terminal, often pair of input/output terminal of mutual coupling type multi input digital predistorter is for accessing the primary signal of a wave band in concurrent multi-wave signal and exporting the predistortion baseband data sequence of this wave band, the input of each base band varying firing rate module respectively accesses the predistortion baseband data sequence of a wave band in concurrent multi-wave signal, the output of each base band varying firing rate module is all connected with the input of mixer, the output of mixer is used for being connected with the input of concurrent multiband nonlinear system, the input of concurrent multiband nonlinear characteristic extraction element is for connecting the output of concurrent multiband nonlinear system, the output of concurrent multiband nonlinear characteristic extraction element is connected with mutual coupling type multi input digit pre-distortion parameters training aids, mutual coupling type multi input digit pre-distortion parameters training aids accesses the raw baseband data sequence of concurrent each wave band of multiband primary signal, mutual coupling type multi input digit pre-distortion parameters training aids passes to described mutual coupling type multi input digital predistorter by extracting the pre-distortion parameters obtained.
In mutual coupling linearization device of the present invention, owing to adopting the base band varying firing rate module with the quantity Matching of concurrent multiband primary signal, each base band varying firing rate module only needs the primary signal of a conversion wave band, thus, lower to the requirement of digital to analog converter (DA) in base band varying firing rate module, the lower and lower-cost digital to analog converter of speed can be adopted to realize this programme.The output base band data of concurrent multiband nonlinear characteristic extraction element is directly trained in multi input digit pre-distortion parameters training aids in addition, decreases digital signal processing appts, saves DSP resource overhead.
Embodiment: choose the checking of concurrent two waveband non linear system and carry out mutual coupling linearization device of the present invention.As shown in Figure 6, a kind of mutual coupling linearization device of concurrent two waveband non linear system, comprise mutual coupling type multi input digital predistorter, two base band varying firing rate modules, mixer, concurrent multiband nonlinear characteristic extraction element and mutual coupling type multi input digit pre-distortion parameters training aids, mutual coupling type multi input digital predistorter is provided with two pairs of input/output terminals, in the parallel two waveband signal of 1st pair of input/output terminal access of mutual coupling type multi input digital predistorter the 1st wave band primary signal and export the predistortion baseband data sequence of this wave band, in the parallel two waveband signal of 2nd pair of input/output terminal access of mutual coupling type multi input digital predistorter 2 wave bands in the end pre-distorted signal and export the predistortion baseband data sequence of this wave band, the input of the 1st base band varying firing rate module accesses the predistortion baseband data sequence of the 1st wave band in concurrent multi-wave signal, the input of the 2nd base band varying firing rate module accesses the predistortion baseband data sequence of the 2nd wave band in concurrent multi-wave signal, the output of the 1st base band varying firing rate module is all connected with the input of mixer with the output of the 2nd base band varying firing rate module, the output of mixer is connected with the input of concurrent two waveband non linear system, being connected with the input of output with concurrent multiband nonlinear characteristic extraction element of concurrent two waveband non linear system, the output of concurrent multiband nonlinear characteristic extraction element is connected with mutual coupling type multi input digit pre-distortion parameters training aids, the raw baseband data sequence of the 1st wave band and the raw baseband data sequence of the 2nd wave band in the parallel two waveband signal of mutual coupling type multi input digit pre-distortion parameters training aids access, mutual coupling type multi input digit pre-distortion parameters training aids passes to mutual coupling type multi input digital predistorter by extracting the pre-distortion parameters obtained.
In the present embodiment, mutual coupling type multi input digit pre-distortion parameters training aids can be function neural network predistortion parameter training device, based on pre-distortion parameters training aids or recursion lowest mean square (RLS) the pre-distortion parameters training aids of singular value decomposition (SVD), the type of mutual coupling type multi input digit pre-distortion parameters training aids is mated with described mutual coupling type multi input digital predistorter, i.e. the type of mutual coupling type multi input digit pre-distortion parameters training aids specifically depend on mutual coupling type multi input digital predistorter based on the type of nonlinear model.
Claims (3)
1. a mutual coupling modeling method for concurrent multiband nonlinear system, is characterized in that comprising the following steps:
(1) the raw baseband data sequence of each wave band in concurrent multiband primary signal is respectively inputted in a base band varying firing rate module, each base band varying firing rate module exports after the raw baseband data sequence of corresponding wave band is converted to radiofrequency signal, and the radiofrequency signal synchronism output of each wave band;
(2) radiofrequency signal of each wave band of synchronism output is carried out power combing, obtain concurrent multiband original radio frequency signal;
(3) concurrent multiband original radio frequency signal inputted in concurrent multiband nonlinear system, drive concurrent multiband nonlinear system, concurrent multiband nonlinear system exports concurrent multiband distortion radiofrequency signal;
(4) the distortion baseband data sequence of concurrent multiband distortion radiofrequency signal at each wave band of concurrent multiband nonlinear system output is extracted;
(5) real part of the raw baseband data sequence of each wave band in concurrent multiband primary signal and imaginary part, concurrent multiband distortion radiofrequency signal are input in mutual coupling type multi input nonlinear model parameter training device at the real part of the distortion baseband data sequence of each wave band and imaginary part carry out model parameter training, obtain mutual coupling type multi input nonlinear model shape parameter;
(6) by mutual coupling type multi input nonlinear model Parameter transfer in mutual coupling type multi input nonlinear model;
(7) be input in mutual coupling type multi input nonlinear model by the real part of the raw baseband data sequence of each wave band in concurrent multiband primary signal and imaginary part, mutual coupling type multi input nonlinear model exports real part and the imaginary part of the model baseband data sequence of each wave band in concurrent multiband distorted signal;
(8) the model baseband data sequence of each wave band in concurrent multiband distorted signal is respectively inputted in a base band varying firing rate module, each base band varying firing rate module exports after the model baseband data sequence of corresponding wave band is converted to model radiofrequency signal, and the model radiofrequency signal synchronism output of each wave band;
(9) the model radiofrequency signal of each wave band of synchronism output is carried out power combing, obtain concurrent multiband model radiofrequency signal;
(10) concurrent multiband model radiofrequency signal and concurrent multiband distortion radiofrequency signal are compared, the accuracy of checking mutual coupling type multi input nonlinear model, if this concurrent multiband emulation radiofrequency signal reaches predetermined simulation nicety, the mutual coupling type multi input nonlinear model shape parameter that then step (5) obtains is modeling parameters, and modeling process terminates; If this concurrent multiband model radiofrequency signal is not up to standard, then return the extraction that step (1) repeats mutual coupling type multi input nonlinear model shape parameter.
2. the mutual coupling linearization device of a concurrent multiband nonlinear system, it is characterized in that comprising mutual coupling type multi input digital predistorter, multiple base band varying firing rate module, mixer, concurrent multiband nonlinear characteristic extraction element and mutual coupling type multi input digit pre-distortion parameters training aids, the described quantity of base band varying firing rate module and the wave band quantity Matching of concurrent multi-wave signal, described mutual coupling type multi input digital predistorter is provided with multipair input/output terminal, often pair of input/output terminal of described mutual coupling type multi input digital predistorter is for accessing the primary signal of a wave band in concurrent multi-wave signal and exporting the predistortion base band data of this wave band, the input of each base band varying firing rate module respectively accesses the predistortion base band data of a wave band in concurrent multi-wave signal, the output of each base band varying firing rate module is all connected with the input of described mixer, the output of described mixer is used for being connected with the input of concurrent multiband nonlinear system, the input of described concurrent multiband nonlinear characteristic extraction element is for connecting the output of concurrent multiband nonlinear system, the output of described concurrent multiband nonlinear characteristic extraction element is connected with described mutual coupling type multi input digit pre-distortion parameters training aids, described mutual coupling type multi input digit pre-distortion parameters training aids accesses the raw baseband data of each wave band in concurrent multiband primary signal, described mutual coupling type multi input digit pre-distortion parameters training aids passes to described mutual coupling type multi input digital predistorter by extracting the pre-distortion parameters obtained.
3. the mutual coupling linearization device of a kind of concurrent multiband nonlinear system according to claim 2, it is characterized in that described mutual coupling type multi input digit pre-distortion parameters training aids is neural net digit pre-distortion parameters training aids, based on the pre-distortion parameters training aids of singular value decomposition or recursion lowest mean square pre-distortion parameters training aids, the type of described mutual coupling type multi input digit pre-distortion parameters training aids is mated with described mutual coupling type multi input digital predistorter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210416973.4A CN102938638B (en) | 2012-10-26 | 2012-10-26 | Cross coupling modeling method of concurrency multiband nonlinear system and linear device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210416973.4A CN102938638B (en) | 2012-10-26 | 2012-10-26 | Cross coupling modeling method of concurrency multiband nonlinear system and linear device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102938638A CN102938638A (en) | 2013-02-20 |
| CN102938638B true CN102938638B (en) | 2015-04-22 |
Family
ID=47697518
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210416973.4A Active CN102938638B (en) | 2012-10-26 | 2012-10-26 | Cross coupling modeling method of concurrency multiband nonlinear system and linear device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102938638B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107231133A (en) * | 2017-04-25 | 2017-10-03 | 宁波大学 | A kind of shortwave power amplifier harmonic non-linear catching method |
| WO2019043434A1 (en) * | 2017-08-30 | 2019-03-07 | Telefonaktiebolaget Lm Ericsson (Publ) | Correction of specific intermodulation products in a concurrent multi-band system |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105956291B (en) * | 2016-05-09 | 2018-08-21 | 南阳师范学院 | A kind of non-linear modeling method of high field MRI high power RF power amplifier |
| CN106027047B (en) * | 2016-05-18 | 2019-01-22 | 中国电子科技集团公司第四十一研究所 | The Linearization Method of non-linear analog/digital conversion based on FPGA control |
| CN107453717B (en) * | 2017-07-07 | 2020-09-22 | 东莞理工学院 | System for constructing string-type multi-compartment depth memory nonlinear behavior model |
| CN110765720A (en) * | 2019-09-12 | 2020-02-07 | 重庆大学 | Power amplifier predistortion method of complex value assembly line recurrent neural network model |
| CN114114191B (en) * | 2021-11-25 | 2024-04-23 | 上海无线电设备研究所 | Dual-band echo signal preprocessing method based on FPGA+DSP architecture |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102088427A (en) * | 2011-01-21 | 2011-06-08 | 上海交通大学 | Digital predistortion device and method |
| CN102361476A (en) * | 2011-07-29 | 2012-02-22 | 上海交通大学 | Apparatus and method for digital predistortion of broadband multi-band power amplifier |
| CN102647188A (en) * | 2012-04-27 | 2012-08-22 | 宁波大学 | Extracting device of nonlinear characteristics of multi-band multi-mode radio frequency power amplifier |
| CN102751950A (en) * | 2012-06-29 | 2012-10-24 | 宁波大学 | Digital predistortion method of multirange multimode radio frequency power amplifier |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8606116B2 (en) * | 2011-01-13 | 2013-12-10 | Applied Optoelectronics, Inc. | System and method for distortion compensation in response to frequency detection |
-
2012
- 2012-10-26 CN CN201210416973.4A patent/CN102938638B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102088427A (en) * | 2011-01-21 | 2011-06-08 | 上海交通大学 | Digital predistortion device and method |
| CN102361476A (en) * | 2011-07-29 | 2012-02-22 | 上海交通大学 | Apparatus and method for digital predistortion of broadband multi-band power amplifier |
| CN102647188A (en) * | 2012-04-27 | 2012-08-22 | 宁波大学 | Extracting device of nonlinear characteristics of multi-band multi-mode radio frequency power amplifier |
| CN102751950A (en) * | 2012-06-29 | 2012-10-24 | 宁波大学 | Digital predistortion method of multirange multimode radio frequency power amplifier |
Non-Patent Citations (1)
| Title |
|---|
| 并发双波段射频功放数字预失真同步性研究;惠明等;《微波学报》;20121031;第28卷(第5期);第69-71页 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107231133A (en) * | 2017-04-25 | 2017-10-03 | 宁波大学 | A kind of shortwave power amplifier harmonic non-linear catching method |
| CN107231133B (en) * | 2017-04-25 | 2020-06-16 | 宁波大学 | Harmonic nonlinear capturing method for short-wave power amplifier |
| WO2019043434A1 (en) * | 2017-08-30 | 2019-03-07 | Telefonaktiebolaget Lm Ericsson (Publ) | Correction of specific intermodulation products in a concurrent multi-band system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102938638A (en) | 2013-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102938638B (en) | Cross coupling modeling method of concurrency multiband nonlinear system and linear device | |
| CN101286963B (en) | Wideband adaptive digital predistortion engine apparatus based on programmable device | |
| CN107592083B (en) | Digital predistortion system and method for millimeter wave broadband power amplifier | |
| CN103051574A (en) | Digital predistortion processing method and digital predistortion processing system | |
| CN102751950B (en) | Digital predistortion method of multirange multimode radio frequency power amplifier | |
| CN104883140A (en) | Digital predistortion device based on broadband radio frequency power amplifier | |
| WO2016095528A1 (en) | Digital predistortion method and apparatus, and computer storage medium | |
| CN103715992A (en) | Power-amplifier pre-distortion device and method based on simplified Volterra series | |
| CN107659273B (en) | Reconfigurable digital predistortion system and method for millimeter wave broadband power amplifier | |
| CN102843108B (en) | A kind of efficient linear radio frequency power amplifying device and method | |
| CN102291347B (en) | DPD (Digital Pre-Distortion) processing method and equipment based on multiband spectrum | |
| CN202818232U (en) | Doherty power amplifying circuit | |
| CN105227507B (en) | Nonlinear systematic distortion correction device and method | |
| Suryasarman et al. | Optimizing the identification of digital predistorters for improved power amplifier linearization performance | |
| Jaraut et al. | Digital predistortion technique for low resource consumption using carrier aggregated 4G/5G signals | |
| CN102769589B (en) | A kind of method and system improving digital pre-distortion performance | |
| CN104301269B (en) | Equivalent sampling pre-distortion system and its method | |
| CN102983820B (en) | Nonlinear injection type linearization system and digital predistortion method | |
| CN102271106B (en) | Pre-distortion processing method and device | |
| CN102904846B (en) | A kind of digital pre-distortion processing method adapting to fast changed signal | |
| CN202261371U (en) | Mixed-mode radio-frequency zooming system | |
| CN104580043A (en) | Digital pre-distortion system and method thereof | |
| CN103384142A (en) | Predistortion device | |
| CN203800956U (en) | Digital pre-distortion system of off-line estimation pre-distortion coefficient | |
| CN102647188B (en) | Extracting device of nonlinear characteristics of multi-band multi-mode radio frequency power amplifier |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |