CN103580617A - Method and device for training predistorter - Google Patents

Method and device for training predistorter Download PDF

Info

Publication number
CN103580617A
CN103580617A CN201210253669.2A CN201210253669A CN103580617A CN 103580617 A CN103580617 A CN 103580617A CN 201210253669 A CN201210253669 A CN 201210253669A CN 103580617 A CN103580617 A CN 103580617A
Authority
CN
China
Prior art keywords
predistorter
signal
data
training
coefficient
Prior art date
Application number
CN201210253669.2A
Other languages
Chinese (zh)
Inventor
施展
李辉
陈培
周建民
Original Assignee
富士通株式会社
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 富士通株式会社 filed Critical 富士通株式会社
Priority to CN201210253669.2A priority Critical patent/CN103580617A/en
Publication of CN103580617A publication Critical patent/CN103580617A/en

Links

Abstract

The invention discloses a method and a device for training a predistorter. The method for training the predistorter includes the steps: pre-storing effective data of a period of time; multiplying the pre-stored effective data to generate training data; training the predistorter based on the training data to obtain the coefficient of the predistorter.

Description

The training method of predistorter and device

Technical field

The training method of the predistorter of relate generally to non linear system of the present invention and device.More particularly, the present invention relates to training method and the device for the predistorter of nonlinear power amplifier.

Background technology

As the Typical Representative of non linear system, power amplifier is the important component part of a lot of electronic equipments, and it can amplify the faint signal of telecommunication, to meet the needs of transmission.Wherein, the energy of amplification comes from DC power supply, and power amplifier can be converted into AC signal by DC energy, and the power level of AC signal is satisfied the demand.The ratio that power amplifier is converted to AC energy by DC energy is known as the efficiency of power amplifier.

According to the physical characteristic of power amplifier, along with the power of input signal is ascending, the characteristic curve of the reflection input signal of power amplifier and the power relation of output signal can be divided into linear zone, inelastic region and saturation region.Fig. 1 illustrates the non-linear input/output signal characteristic of power amplifier.At input signal V iNthe region that amplitude is less, the output V of power amplifier PA oUTalmost input signal V iNlinearity amplify, but along with input signal V iNthe increase of amplitude, the nonlinear characteristic of power amplifier is obviously to the last saturated gradually.This non-linear behavior is in time domain, and output signal is not that the ideal of input signal is amplified; This non-linear behavior is on frequency domain, and due to intermodulation effect, video stretching (secondary lobe rising) appears outward in the band of the signal being exaggerated, and occurs as shown in Figure 2, having shown distortion (main lobe distortion) by the caused video stretching of nonlinear power amplifier in band.Video stretching can affect the normal operation of the miscellaneous equipment that works in adjacent channel.

Ideally, wish that power amplifier only plays linear effect of amplifying, output signal is the simple linear amplification of input signal, therefore allows power amplifier be operated in linear zone.But the efficiency that now power amplifier is converted to AC signal by direct current signal is very low, causes the waste of large energy, and need to increase extra heat dissipation equipment.Due to physically, when the envelope fluctuation of input signal is deeply to inelastic region, the situation that the efficiency of power amplifier will be when only fluctuating in linear zone.And along with the appearance of new-type modulation system, the dynamic range of signal envelope is increasing, therefore there is nonlinear distortion inevitable.

Therefore, on the one hand in order to improve the efficiency of power amplifier, on the other hand because the signal in a lot of Modern Communication System has very large dynamic range (peak-to-average power ratio), so power amplifier often need to be in inelastic region work, thereby cause the distortion of output signal (on frequency domain, in band, there is distortion, the outer video stretching that occurs of band).In power amplifier field, conventionally this distortion is called to the nonlinear characteristic of power amplifier.

In numerous methods that overcome non-linearity of power amplifier characteristic, base band predistortion is a kind of method receiving much concern.As shown in Figure 3, illustrate for offsetting the schematic diagram of input/output signal characteristic of the predistorter for power amplifier of the nonlinear characteristic shown in Fig. 1.Base band predistortion technology, by using the contrary characteristic of predistorter PD simulated power amplifier, makes original input signal V iNpredistortion occurred before being input to power amplifier, thereby compensating power amplifier is non-linear, and obtains and there is no the amplification output signal V that distorts at the output of power amplifier oUT.The input/output signal characteristic with the power amplifier of pre-distortion function all shows extraordinary linear characteristic before until approach saturation region, as shown in Figure 4, the schematic diagram of the input/output signal characteristic of the power amplifier that is provided with predistorter is shown.

Visible, predistorter can overcome the non-linear of power amplifier effectively.Yet, in order to make predistorter normal operation, need to predistorter, train in advance, to obtain the coefficient of suitable predistorter.And after predistorter comes into operation, due to the variation of environmental factor of job site etc., the characteristic of power amplifier also there will be variation.In this case, need to re-start training to predistorter, to upgrade the coefficient of predistorter.For ease of explanation, the coefficient obtaining is in advance called to the initial value of the coefficient of predistorter, the coefficient obtaining during by renewal is called the renewal value of the coefficient of predistorter.The initial value of coefficient and the process of renewal value that it should be noted that acquisition predistorter are all the processes that predistorter is trained.

Training process is mainly using power amplifier output signal as feedback signal, according to it, adjust the coefficient of predistorter, by the continuous adjustment of coefficient being reached to predetermined condition (as training algorithm convergence, power amplifier output signal meet certain spectrum requirement etc.), obtain suitable coefficient.

According to the character of feedback signal, training method can be divided into vector method and scalar method.Vector method requires complete I, the Q two paths of signals of feedback power amplifier output, and compares with primary signal.Scalar method requires the power of the signal of feedback power amplifier output.Scalar method is lower to the requirement of feedback signal, and computational speed is very fast.The present invention relates generally to scalar method.

In prior art, there are the following problems: the training process of predistorter is subject to the adverse effect of training data fluctuation, and traditional processing method especially can cause the training time long for the obtaining of the initial value of the coefficient of predistorter.

Fig. 5 shows the topology example of traditional scalar method training predistorter.

In Fig. 5, signal source 501 is input to predistorter 502 by primary signal x (t).Predistorter 502 has factor alpha i, predistortion function Q α i{ }, processes through predistorter 502, obtains pre-distorted signals z (t).Pre-distorted signals z (t) carries out through digital to analog converter 504 that quadrature modulation is carried out in digital-to-analogue conversion (two digital to analog converters 504 are processed respectively I, Q two paths of signals), quadrature modulator 503, upconverter 505 is input to power amplifier 506 after carrying out up-conversion.Power amplifier 506 is processed according to function P{}, obtains output signal y (t).Output signal y (t) outputs to antenna through coupler 507, simultaneously through feeding back as feedback signal.Upgrade judging unit 508 and according to the character of feedback signal, judge whether to upgrade the coefficient of predistorter.In the training stage, feedback signal is provided to cost function generation unit 509.Cost function generation unit 509 calculates according to feedback signal, obtains cost function CF.Coefficient update unit 510 generates new predistorter factor alpha i and offers predistorter 502 according to cost function CF.Predistorter 502 carries out work according to new coefficient.Control unit 511 is controlled.While adopting scalar method, according to the power calculation cost function of the high fdrequency component of above-mentioned feedback signal.By minimization cost function, obtain the coefficient of predistorter.Under ideal state, through training after a while, adaptive training algorithm is restrained gradually, thereby obtains predistorter coefficient.

Formula 1 as follows has exemplarily provided the computational methods of CF.

CF ( n ) = Σ t = t n t n + m - 1 | f { y ( t ) } | 2 / m - - - ( 1 )

Wherein, y (t) is as the output signal of feedback signal, power amplifier.F{} represents to obtain the filter of the high fdrequency component in feedback signal.M represents to get the m data of long a period of time, t nand t n+m-1it is the starting and ending time during this period of time.Due to the randomness of signal x (t) itself, may cause different n values corresponding to different CF values, produce fluctuation.

That is, cost function CF changes according to the power of the high fdrequency component of feedback signal.Because primary signal x (t) itself has randomness, therefore, even if the coefficient of predistorter is constant, different x (t) data segment also can cause different CF.In fact, the coefficient of predistorter is also changing, and therefore, the fluctuation between x (t) data segment produces and disturbs the training of predistorter.

In other words, if primary signal x (t) is constant, the coefficient of predistorter constantly changes, and through adjustment after a while, coefficient can converge to suitable value.Yet, the data of supposing the very first time section of primary signal x (t) are transfused to, and obtain accordingly feedback signal 1, according to feedback signal 1, obtain new coefficient 1, however after new coefficient 1 enables, processing be the data of the second time period of primary signal x (t), because the data of the second time period data with respect to very first time section fluctuate, therefore,, even if new coefficient meets the data of very first time section very much, likely do not meet very much the data of the second time period yet.Therefore, the data of the second time period may cause coefficient towards other incorrect trend adjustment, and on the basis of coefficient 1, along former trend, further do not adjust.

Visible, the fluctuation of input signal can have a negative impact to the training of predistorter.Especially in the process of coefficient initial value that solves predistorter, compare with the coefficient update process after predistorter work a period of time, due to coefficient indiscipline, therefore need to obtain for more time suitable coefficient initial value.

For above-mentioned situation, traditional way is the time (being the m value in formula 1) that lengthens input signal, thereby reduces the fluctuation of the different time sections of input signal.This way can cause lengthen measuring the time of feedback signal power, because must wait for that the time that m grows just can carry out the calculating of a formula 1, obviously this can cause the training time to lengthen at double.

Therefore, wish training method and the corresponding trainer of a kind of predistorter of design, the adverse effect that its fluctuation that can overcome input signal is brought, can not cause the training time to extend simultaneously.

Summary of the invention

Provided hereinafter about brief overview of the present invention, to the basic comprehension about some aspect of the present invention is provided.Should be appreciated that this general introduction is not about exhaustive general introduction of the present invention.It is not that intention is determined key of the present invention or pith, and nor is it intended to limit the scope of the present invention.Its object is only that the form of simplifying provides some concept, usings this as the preorder in greater detail of discussing after a while.

The training method and the corresponding trainer that the object of this invention is to provide a kind of predistorter, the adverse effect that its fluctuation that can overcome input signal is brought, can not cause the training time to extend simultaneously.

To achieve these goals, according to an aspect of the present invention, provide a kind of training method of predistorter, comprising: the valid data of pre-stored a period of time; By prestored valid data are repeatedly repeated to generating training data; And based on described training data, predistorter is trained, to obtain the coefficient of predistorter.

According to a further aspect in the invention, provide a kind of power-magnifying method, comprising: by predistorter, original input signal is carried out to predistortion, output predistorted input signal; By digital to analog converter, predistorted input signal is converted to analog signal; By upconverter, analog signal is up-converted to radiofrequency signal; And by power amplifier, radiofrequency signal is carried out to signal after power amplification power output are amplified as output signal; Wherein, described predistorter adopts training method training as above.

In accordance with a further aspect of the present invention, provide a kind of trainer of predistorter, having comprised: training data generation unit, has been configured to the valid data of pre-stored a period of time and by prestored valid data are repeatedly repeated to generating training data; And control unit, be configured to based on described training data, predistorter be trained to obtain the coefficient of predistorter.

In accordance with a further aspect of the present invention, provide a kind of power amplification device, having comprised: according to the trainer of predistorter as above, for predistorter is trained; Predistorter, is configured to original input signal to carry out predistortion, output predistorted input signal; Digital to analog converter, is configured to predistorted input signal to be converted to analog signal; Upconverter, is configured to analog signal to up-convert to radiofrequency signal; And power amplifier, be configured to radiofrequency signal to carry out signal after power amplification power output are amplified as output signal.

In addition, according to a further aspect in the invention, also provide a kind of storage medium.Described storage medium comprises machine-readable program code, and when carrying out described program code on messaging device, described program code is carried out according to said method of the present invention described messaging device.

In addition, in accordance with a further aspect of the present invention, also provide a kind of program product.Described program product comprises the executable instruction of machine, and when carrying out described instruction on messaging device, described instruction is carried out according to said method of the present invention described messaging device.

In specification part below, provide other aspects of the present invention, wherein, describe in detail for disclosing fully the preferred embodiments of the present invention, and it is not applied to restriction.

Accompanying drawing explanation

Below with reference to the accompanying drawings illustrate embodiments of the invention, can understand more easily above and other objects, features and advantages of the present invention.Parts in accompanying drawing are just in order to illustrate principle of the present invention.In the accompanying drawings, same or similar technical characterictic or parts will adopt same or similar Reference numeral to represent.In accompanying drawing:

Fig. 1 is the schematic diagram of nonlinear characteristic that the input/output signal of nonlinear power amplifier is shown;

Fig. 2 is the spectrum diagram illustrating by nonlinear power amplifier institute amplifying signal;

Fig. 3 illustrates for offsetting the schematic diagram of input/output signal characteristic of the power amplifier pre-distortion device of the nonlinear characteristic shown in Fig. 1;

Fig. 4 is the schematic diagram that the input/output signal characteristic of the power amplifier that is provided with predistorter is shown;

Fig. 5 is the figure that the topology example of traditional scalar method training predistorter is shown;

Fig. 6 is the figure that the trainer of predistorter according to an embodiment of the invention is shown;

Fig. 7 is the flow chart that the training method of predistorter according to an embodiment of the invention is shown;

Fig. 8 is the flow chart that the training method of predistorter is according to another embodiment of the invention shown;

Fig. 9 is the flow chart illustrating according to power-magnifying method of the present invention;

Figure 10 is the figure illustrating according to power amplification device of the present invention;

Figure 11 is the figure illustrating according to the detailed configuration example of power amplification device of the present invention; And

Figure 12 is the block diagram that the exemplary configurations of personal computer is shown.

Embodiment

In connection with accompanying drawing, one exemplary embodiment of the present invention is described in detail hereinafter.All features of actual execution mode are not described for clarity and conciseness, in specification.Yet, should understand, in the process of any this practical embodiments of exploitation, must make a lot of decisions specific to execution mode, to realize developer's objectives, for example, meet those restrictive conditions with system and traffic aided, and these restrictive conditions may change to some extent along with the difference of execution mode.In addition,, although will also be appreciated that development is likely very complicated and time-consuming, concerning having benefited from those skilled in the art of present disclosure, this development is only routine task.

At this, also it should be noted is that, for fear of the details because of unnecessary fuzzy the present invention, only show in the accompanying drawings with according to the closely-related apparatus structure of the solution of the present invention and/or treatment step, and omitted other details little with relation of the present invention.In addition, also it is pointed out that element and the feature in an accompanying drawing of the present invention or a kind of execution mode, described can combine with element and feature shown in one or more other accompanying drawing or execution mode.

As mentioned above, in conventional art, the fluctuation of input signal has adversely affected the training process of predistorter, and relevant measure has extended coefficient value, the especially time of coefficient initial value that obtains predistorter.

The present inventor recognizes the fluctuation of the input signal that has its source in of the problems referred to above.Therefore,, if make input signal relatively steady, can overcome the problems referred to above.Meanwhile, the m value that traditional method has strengthened in formula 1 makes input signal relatively steady, but causes the acquisition time of computing formula 1 desired data elongated because m value adds conference, therefore, greatly extends the training time.Therefore inventor is designed to by repeating input signal, input signal be become steadily dexterously, meanwhile, because means are repeating datas, the data segment being repeated does not need very long, therefore, has avoided the prolongation of training time.

Based on above-mentioned thought, design has realized training method and the device of predistorter of the present invention.Embodiments of the invention are described with reference to the accompanying drawings.

Fig. 7 shows the training method of predistorter according to an embodiment of the invention.This training method comprises: the valid data of pre-stored a period of time (step S701); By prestored valid data repeatedly being repeated to generating training data (step S702); And based on described training data, predistorter is trained, to obtain the coefficient (step S703) of predistorter.

In step S701, the valid data of pre-stored a period of time.

Described valid data can be to capture out the data that send in real time from signal source, can be also the pre-prepd data for training.

The length that it should be noted that above-mentioned a period of time is the m in formula 1.M is time spans corresponding to the data of actual Reusability.For fear of data, cut mistake, in practical operation, can capture the valid data of the time span that is greater than m, and therefrom cut the data in the middle of being positioned at.For example, capture the valid data of 10 milliseconds, can store the data of the 2-9 millisecond in these 10 milliseconds of data as reusable data.As long as guarantee that the long valid data of m are repeated.

In step S702, by prestored valid data are repeatedly repeated to generating training data.

It should be noted that the data that are repeated should be complete, each data length repeating is m.

In step S703, the training data based on generated, trains predistorter, to obtain the coefficient of predistorter.

Because of the present invention, set about being a little mainly to train the data this respect of use, therefore, the training method of predistorter applicatory can be the method for any scalar method training predistorter in prior art.Therefore the present invention has good applicability.

Fig. 8 shows the training method of predistorter according to another embodiment of the invention.This training method comprises: the valid data of pre-stored a period of time (step S801); By prestored valid data repeatedly being repeated to generating training data (step S802); Described training data is carried out to low-pass filtering, and using filtering result as training data (step S804); And based on described training data, predistorter is trained, to obtain the coefficient (step S803) of predistorter.

Wherein step S801-S803 is consistent with the step S701-S703 in a upper embodiment.While being repeated due to data segment, there is discontinuity in head and the tail directly amalgamation meeting, therefore, increases step S804, can suppress the secondary lobe that between two segment datas that are repeated, discontinuity causes.

It is worth mentioning that, according to the training method of predistorter of the present invention, due to the adverse effect of having eliminated the fluctuation of input data and causing, can avoid lengthening the training time, therefore, be suitable for very much obtaining the initial value of the coefficient of predistorter simultaneously.Meanwhile, owing to calculating initial value and the renewal value of the coefficient of predistorter, be all the training process that carries out predistorter, therefore, the present invention is equally applicable to obtain the renewal value of the coefficient of predistorter.

Therefore, can carry out according to training method of the present invention, to obtain the coefficient initial value of predistorter.And in the time need to upgrading the coefficient of predistorter, can repeat according to training method of the present invention, to obtain the renewal value of the coefficient of predistorter.

In addition, due to the coefficient update value to predistorter, calculating is to have the adjustment of the enterprising row coefficient in basis of existing coefficient, and therefore, with respect to the ground design factor initial value that grows out of nothing, the spent time is relatively less.Therefore in the time need to upgrading the coefficient of predistorter, the data of the long period section of the non-repeatability that signal source can be sent are in real time as training data, and the training data based on such, predistorter is trained to obtain to the renewal value of the coefficient of predistorter.Although may cause the time lengthening of each computing formula 1 due to the m in increase formula 1, because iterations is relatively less, so can not spend the renewal value that the long time obtains the coefficient of predistorter.

Because predistorter is to design in order to overcome the nonlinear characteristic of power amplifier, therefore, in power amplifier passing in time, for example, be subject to the impact of surrounding environment, and during the change of occurrence features, need to upgrade the coefficient of predistorter.Therefore, can set suitable frequency spectrum condition, as frequency band concentration degree etc.In the situation that the output signal of power amplifier does not meet predetermined frequency spectrum condition, be judged as the coefficient that need to upgrade predistorter.By carrying out, according to training method of the present invention, again train predistorter, obtain the renewal value of the coefficient of predistorter.This adaptive adjustment mode is very suitable for power amplifier and is arranged on the situation in the violent environment of the climate changes such as high mountain, snowfield.In such environment, day and night temperature is large, and daytime and evening may differ tens degree, larger to the properties influence of power amplifier.No matter be to arrange power amplifier in this class environment, the coefficient initial value that carries out predistorter calculates, or the frequent coefficient update value of training to obtain predistorter repeatedly, and the present invention is extremely applicable.

In addition, as mentioned above, valid data can be to capture out the data that send in real time from signal source.Due to the training need time, and signal source is sent new data continuously, therefore, can be when carrying out described training method, the data buffering that signal source is sent is in memory.

When training is complete, memory sends the data of buffering on the one hand to rear class, continues on the other hand to receive the data of newly sending, and forms by the buffering of data flow.

Therefore, those skilled in the art can be according to actual needs, with reference to time of training need and data rate etc. because of the size of reasonable design memory usually.

In addition,, in the situation that agreement allows, also the discardable training sending from signal source continues the valid data that send from signal source with valid data and at training period.After training, the new valid data that send from signal source can directly send to predistorter and carry out predistortion and subsequent treatment.

Correspondingly, can design a kind of power-magnifying method.

Fig. 9 shows the flow chart according to power-magnifying method of the present invention.

Power-magnifying method according to the present invention comprises: by predistorter, original input signal is carried out to predistortion, output predistorted input signal (step S901); By digital to analog converter, predistorted input signal is converted to analog signal (step S902); By upconverter, analog signal is up-converted to radiofrequency signal (step S903); And by power amplifier, radiofrequency signal is carried out to signal after power amplification power output are amplified as output signal (step S904); Wherein, described predistorter adopts according to training method training of the present invention.

Below, with reference to figure 6,10,11, describe according to the trainer of predistorter of the present invention and power amplification device.

Fig. 6 shows the trainer of predistorter according to an embodiment of the invention.

As shown in Figure 6, the trainer 60 of predistorter comprises training data generation unit 61 and control unit 62.Training data generation unit 61 is configured to the valid data of pre-stored a period of time and by prestored valid data are repeatedly repeated to generating training data.Control unit 62 is configured to based on described training data, predistorter be trained to obtain the coefficient of predistorter.

Trainer 60 also can comprise low pass filter 63, and its training data that is configured to that training data generation unit 61 is generated carries out low-pass filtering and using filtering result as training data.

As mentioned above, trainer 60 both can be used for predistorter to train, to obtain the initial value of the coefficient of predistorter, be also used in the situation of the coefficient that need to upgrade predistorter predistorter is trained, to obtain the renewal value of the coefficient of predistorter.

The data that training data generation unit 61 can send in real time from signal source, capture the valid data of described a period of time.Correspondingly, trainer 60 also can comprise memory 64, and it is configured to the data that buffering signals source sends, thereby prevents loss of data.

Trainer 60 also can comprise renewal judging unit 65, and it is configured in the situation that the output signal of power amplifier does not meet predetermined frequency spectrum condition, is judged as the coefficient that need to upgrade predistorter.Now, the data of the long period section of the non-repeatability that training data generation unit 61 can in real time send signal source are as training data, and the training data of control unit 62 based on such trains to obtain the renewal value of the coefficient of predistorter to predistorter.

Trainer 60 also can comprise switch unit 66, for suitable data are exported to predistorter.Particularly, switch unit 66 is under the control of control unit 62, when predistorter is trained, output is from the training data of training data generation unit 61 or low pass filter 63, the data that the data through memory 64 bufferings that send according to the situation signal source output of agreement when predistorter works or signal source send in real time, do not export data in other cases.

Trainer 60 also can comprise cost function generation unit 67 and coefficient update unit 68.Cost function generation unit 67 generates cost function according to the output of power amplifier.The cost function that coefficient update unit 68 provides according to cost function generation unit 67 generates new predistorter coefficient and offers predistorter.

Figure 10 shows according to power amplification device of the present invention.

As shown in figure 10, power amplification device 100 comprises the trainer 101 according to predistorter of the present invention, for predistorter is trained; Predistorter 102, is configured to original input signal to carry out predistortion, output predistorted input signal; Digital to analog converter 103, is configured to predistorted input signal to be converted to analog signal; Upconverter 104, is configured to analog signal to up-convert to radiofrequency signal; And power amplifier 105, be configured to radiofrequency signal to carry out signal after power amplification power output are amplified as output signal.Power amplification device 100 also comprises quadrature modulator 106 alternatively, and its analog signal that is configured to 103 outputs of logarithmic mode transducer is carried out quadrature modulation, and the analog signal output after quadrature modulation is arrived to upconverter 104.

Figure 11 shows the detailed configuration example according to power amplification device of the present invention.

For the relation between the unit of the unit of predistorter trainer and predistorter, power amplification device is shown better, drawn Figure 11.

Signal source 1101 in Figure 11, quadrature modulator 1103, coupler 1107 are distinguished corresponding consistent with signal source 501, quadrature modulator 503, coupler 507 in Fig. 5.Quadrature modulator 1103 is option means.Coupler 1107 can be replaced by having other unit of function along separate routes.

Predistorter 1102 in Figure 11, digital to analog converter 1104, upconverter 1105, power amplifier 1106 are distinguished corresponding consistent with predistorter 102, digital to analog converter 103, upconverter 104, power amplifier 105 in Figure 10.

Renewal judging unit 1108 in Figure 11, cost function generation unit 1109, coefficient update unit 1110, control unit 1111, training data generation unit 1112, switch unit 1113, low pass filter 1114, memory 1115 are distinguished corresponding consistent with renewal judging unit 65, cost function generation unit 67, coefficient update unit 68, control unit 62, training data generation unit 61, switch unit 66, low pass filter 63, memory 64 in Fig. 6.

In addition, should also be noted that all modules in said apparatus, unit can be configured by the mode of software, firmware, hardware or its combination.Configure spendable concrete means or mode and be well known to those skilled in the art, do not repeat them here.In the situation that realizing by software and/or firmware, from storage medium or network to the computer with specialized hardware structure, example general purpose personal computer 1200 is as shown in figure 12 installed the program that forms this software, and this computer, when various program is installed, can be carried out various functions etc.

In Figure 12, CPU (CPU) 1201 carries out various processing according to the program of storage in read-only memory (ROM) 1202 or from the program that storage area 1208 is loaded into random access memory (RAM) 1203.In RAM1203, also store as required data required when CPU1201 carries out various processing etc.

CPU1201, ROM1202 and RAM1203 are connected to each other via bus 1204.Input/output interface 1205 is also connected to bus 1204.

Following parts are connected to input/output interface 1205: importation 1206, comprises keyboard, mouse etc.; Output 1207, comprises display, such as cathode ray tube (CRT), liquid crystal display (LCD) etc., and loud speaker etc.; Storage area 1208, comprises hard disk etc.; With communications portion 1209, comprise that network interface unit is such as LAN card, modulator-demodulator etc.Communications portion 1209 via network such as internet executive communication is processed.

As required, driver 1210 is also connected to input/output interface 1205.Detachable media 1211, such as disk, CD, magneto optical disk, semiconductor memory etc. are installed on driver 1210 as required, is installed in storage area 1208 computer program of therefrom reading as required.

In the situation that realizing above-mentioned series of processes by software, from network such as internet or storage medium are such as detachable media 1211 is installed the program that forms softwares.

It will be understood by those of skill in the art that this storage medium is not limited to wherein having program stored therein shown in Figure 12, distributes separately to user, to provide the detachable media 1211 of program with equipment.The example of detachable media 1211 comprises disk (comprising floppy disk (registered trade mark)), CD (comprising compact disc read-only memory (CD-ROM) and digital universal disc (DVD)), magneto optical disk (comprising mini-disk (MD) (registered trade mark)) and semiconductor memory.Or storage medium can be hard disk comprising in ROM1202, storage area 1208 etc., computer program stored wherein, and be distributed to user together with the equipment that comprises them.

The present invention also proposes a kind of program product that stores the instruction code that machine readable gets.When described instruction code is read and carried out by machine, can carry out above-mentioned according to the method for the embodiment of the present invention.

Correspondingly, for carrying the above-mentioned storage medium that stores the program product of the instruction code that machine readable gets, be also included within of the present invention open.Described storage medium includes but not limited to floppy disk, CD, magneto optical disk, storage card, memory stick etc.

In the above in the description of the specific embodiment of the invention, the feature of describing and/or illustrating for a kind of execution mode can be used in same or similar mode in one or more other execution mode, combined with the feature in other execution mode, or substitute the feature in other execution mode.

Should emphasize, term " comprises/comprises " existence that refers to feature, key element, step or assembly while using herein, but does not get rid of the existence of one or more further feature, key element, step or assembly or add.

In addition, the time sequencing of describing during method of the present invention is not limited to is to specifications carried out, also can be according to other time sequencing ground, carry out concurrently or independently.The execution sequence of the method for therefore, describing in this specification is not construed as limiting technical scope of the present invention.

Although described the present invention and advantage thereof in detail, be to be understood that in the situation that do not depart from the spirit and scope of the present invention that limited by appended claim and can carry out various changes, alternative and conversion.And, the application's term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thereby the process, method, article or the equipment that make to comprise a series of key elements not only comprise those key elements, but also comprise other key elements of clearly not listing, or be also included as the intrinsic key element of this process, method, article or equipment.The in the situation that of more restrictions not, the key element being limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment that comprises described key element and also have other identical element.

remarks

1. a training method for predistorter, comprises the steps:

The valid data of pre-stored a period of time;

By prestored valid data are repeatedly repeated to generating training data; And

Based on described training data, predistorter is trained, to obtain the coefficient of predistorter.

2. the method as described in remarks 1, also comprises: described training data is carried out to low-pass filtering, and using filtering result as training data.

3. the method as described in one of remarks 1-2, wherein, carries out described training method, to obtain the coefficient initial value of predistorter.

4. the method as described in one of remarks 1-2, wherein, in the time need to upgrading the coefficient of predistorter, repeats described training method, to obtain the renewal value of the coefficient of predistorter.

5. the method as described in one of remarks 1-2, wherein, in the time need to upgrading the coefficient of predistorter, the data of the long period section of the non-repeatability that signal source is sent are in real time as training data, and based on described training data, predistorter is trained to obtain to the renewal value of the coefficient of predistorter.

6. the method as described in one of remarks 1-2, wherein, when carrying out described training method, the data buffering that signal source is sent is in memory.

7. the method as described in one of remarks 1-2, wherein, described valid data are to capture out the data that send in real time from signal source.

8. the method as described in one of remarks 1-2, wherein, in the situation that the output signal of power amplifier does not meet predetermined frequency spectrum condition, is judged as the coefficient that need to upgrade predistorter.

9. a power-magnifying method, comprising:

By predistorter, original input signal is carried out to predistortion, output predistorted input signal;

By digital to analog converter, predistorted input signal is converted to analog signal;

By upconverter, analog signal is up-converted to radiofrequency signal; And

By power amplifier, radiofrequency signal is carried out to signal after power amplification power output are amplified as output signal;

Wherein, described predistorter adopts according to the training method training one of remarks 1-8 Suo Shu.

10. a trainer for predistorter, comprising:

Training data generation unit, is configured to the valid data of pre-stored a period of time and by prestored valid data are repeatedly repeated to generating training data; And

Control unit, is configured to based on described training data, predistorter be trained to obtain the coefficient of predistorter.

11. trainers as described in remarks 10, also comprise:

Low pass filter, the training data that is configured to that training data generation unit is generated carries out low-pass filtering and using filtering result as training data.

12. trainers as described in one of remarks 10-11, wherein, described trainer is for training predistorter, to obtain the initial value of the coefficient of predistorter.

13. trainers as described in one of remarks 10-11, wherein, described trainer is in the situation that need to upgrade the coefficient of predistorter predistorter is trained, to obtain the renewal value of the coefficient of predistorter.

14. trainers as described in one of remarks 10-11, wherein, described training data generation unit be further configured in the situation that the data of the long period section of the non-repeatability that the coefficient that need to upgrade predistorter sends signal source in real time as training data;

Described control unit is further configured to trains to obtain the renewal value of the coefficient of predistorter to predistorter based on described training data.

15. trainers as described in one of remarks 10-11, also comprise:

Memory, is configured to the data that buffering signals source sends.

16. trainers as described in one of remarks 10-11, wherein, described training data generation unit, is further configured to the valid data that capture described a period of time the data that send in real time from signal source.

17. trainers as described in one of remarks 10-11, also comprise:

Upgrade judging unit, be configured to, in the situation that the output signal of power amplifier does not meet predetermined frequency spectrum condition, be judged as the coefficient that need to upgrade predistorter.

18. trainers as described in one of remarks 10-11, also comprise:

Switch unit, be configured under the control of control unit, when predistorter is trained, output is from the training data of training data generation unit or low pass filter, when predistorter works, output, from the data of the signal source transmission of memory, is not exported data in other cases.

19. trainers as described in one of remarks 10-11, also comprise:

Cost function generation unit, for generating cost function according to the output of power amplifier;

Coefficient update unit, generates new predistorter coefficient and offers predistorter for the cost function providing according to cost function generation unit.

20. 1 kinds of power amplification devices, comprising:

According to the trainer of the predistorter one of remarks 10-19 Suo Shu, for predistorter is trained;

Predistorter, is configured to original input signal to carry out predistortion, output predistorted input signal;

Digital to analog converter, is configured to predistorted input signal to be converted to analog signal;

Upconverter, is configured to analog signal to up-convert to radiofrequency signal; And

Power amplifier, is configured to radiofrequency signal to carry out signal after power amplification power output are amplified as output signal.

Claims (10)

1. a training method for predistorter, comprises the steps:
The valid data of pre-stored a period of time;
By prestored valid data are repeatedly repeated to generating training data; And
Based on described training data, predistorter is trained, to obtain the coefficient of predistorter.
2. the method for claim 1, also comprises: described training data is carried out to low-pass filtering, and using filtering result as training data.
3. the method as described in one of claim 1-2, wherein, in the time need to upgrading the coefficient of predistorter, repeats described training method, to obtain the renewal value of the coefficient of predistorter.
4. the method as described in one of claim 1-2, wherein, in the time need to upgrading the coefficient of predistorter, the data of the long period section of the non-repeatability that signal source is sent are in real time as training data, and based on described training data, predistorter is trained to obtain to the renewal value of the coefficient of predistorter.
5. the method as described in one of claim 1-2, wherein, when carrying out described training method, the data buffering that signal source is sent is in memory.
6. a power-magnifying method, comprising:
By predistorter, original input signal is carried out to predistortion, output predistorted input signal;
By digital to analog converter, predistorted input signal is converted to analog signal;
By upconverter, analog signal is up-converted to radiofrequency signal; And
By power amplifier, radiofrequency signal is carried out to signal after power amplification power output are amplified as output signal;
Wherein, described predistorter adopts according to the training method training one of claim 1-5 Suo Shu.
7. a trainer for predistorter, comprising:
Training data generation unit, is configured to the valid data of pre-stored a period of time and by prestored valid data are repeatedly repeated to generating training data; And
Control unit, is configured to based on described training data, predistorter be trained to obtain the coefficient of predistorter.
8. trainer as claimed in claim 7, also comprises:
Low pass filter, the training data that is configured to that training data generation unit is generated carries out low-pass filtering and using filtering result as training data.
9. the trainer as described in one of claim 7-8, also comprises:
Memory, is configured to the data that buffering signals source sends.
10. a power amplification device, comprising:
According to the trainer of the predistorter one of claim 7-9 Suo Shu, for predistorter is trained;
Predistorter, is configured to original input signal to carry out predistortion, output predistorted input signal;
Digital to analog converter, is configured to predistorted input signal to be converted to analog signal;
Upconverter, is configured to analog signal to up-convert to radiofrequency signal; And
Power amplifier, is configured to radiofrequency signal to carry out signal after power amplification power output are amplified as output signal.
CN201210253669.2A 2012-07-20 2012-07-20 Method and device for training predistorter CN103580617A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210253669.2A CN103580617A (en) 2012-07-20 2012-07-20 Method and device for training predistorter

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210253669.2A CN103580617A (en) 2012-07-20 2012-07-20 Method and device for training predistorter

Publications (1)

Publication Number Publication Date
CN103580617A true CN103580617A (en) 2014-02-12

Family

ID=50051698

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210253669.2A CN103580617A (en) 2012-07-20 2012-07-20 Method and device for training predistorter

Country Status (1)

Country Link
CN (1) CN103580617A (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1593031A (en) * 2000-11-30 2005-03-09 阿雷伊通讯有限公司 Training sequence for a radio communications system
CN101527544A (en) * 2008-03-05 2009-09-09 富士通株式会社 Device and method for identifying inverse characteristic of nonlinear system, power amplifier and predistorter thereof
CN201409180Y (en) * 2009-01-05 2010-02-17 成都凯腾四方数字广播电视设备有限公司 Self-adapting baseband linearization device of digital television transmitter
CN102082752A (en) * 2010-02-25 2011-06-01 大唐移动通信设备有限公司 Digital predistortion processing method and equipment
CN102480450A (en) * 2010-11-30 2012-05-30 富士通株式会社 Predistorter control device and method as well as power control state detection method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1593031A (en) * 2000-11-30 2005-03-09 阿雷伊通讯有限公司 Training sequence for a radio communications system
CN101527544A (en) * 2008-03-05 2009-09-09 富士通株式会社 Device and method for identifying inverse characteristic of nonlinear system, power amplifier and predistorter thereof
CN201409180Y (en) * 2009-01-05 2010-02-17 成都凯腾四方数字广播电视设备有限公司 Self-adapting baseband linearization device of digital television transmitter
CN102082752A (en) * 2010-02-25 2011-06-01 大唐移动通信设备有限公司 Digital predistortion processing method and equipment
CN102480450A (en) * 2010-11-30 2012-05-30 富士通株式会社 Predistorter control device and method as well as power control state detection method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王勇等: ""基于非迭代算法和非直接学习结构的查询表TWTA预失真器"", 《通信学报》 *

Similar Documents

Publication Publication Date Title
US9913194B2 (en) Method and system for baseband predistortion linearization in multi-channel wideband communication systems
Roblin et al. Frequency-selective predistortion linearization of RF power amplifiers
JP4394409B2 (en) Predistortion type amplifier with distortion compensation function
US8049560B2 (en) Increasing the density of larger magnitude amplifier output samples for estimating a model of digital pre-distortion to compensate for amplifier non-linearity
Mkadem et al. Physically inspired neural network model for RF power amplifier behavioral modeling and digital predistortion
JP4786644B2 (en) Distortion compensation device
CN100533956C (en) Uncorrelated self-adaptive predistorter
JP5516269B2 (en) Amplifying apparatus and predistortion control method
CN1972525B (en) Ultra directional speaker system and signal processing method thereof
CN1833418B (en) Digital predistortion system and method for correcting memory effects within an RF power amplifier
CN100492888C (en) Predistorter for phase modulated signals with low peak to average ratios
KR100789125B1 (en) Wideband enhanced digital injection predistortion system and method
DE69728651T2 (en) Transmitter with linearized amplifier
EP1705801B1 (en) Distortion compensation apparatus
US7333562B2 (en) Nonlinear distortion compensating circuit
Schreurs et al. RF power amplifier behavioral modeling
CN100566133C (en) Be used to amplify the equipment and the method for input signal with input signal power
CN103975564A (en) Processor having instruction set with user-defined non-linear functions for digital pre-distortion (DPD) and other non-linear applications
JP3590571B2 (en) Distortion compensator
EP2521260A1 (en) Digital predistortion system and method for high efficiency transmitters
CN100483949C (en) Method and apparatus for generating a pulse width modulated signal
EP2263355B1 (en) High resolution digital modulator by switching between discrete PWM or PPM values
Raich et al. Orthogonal polynomials for power amplifier modeling and predistorter design
CN101061633B (en) Model based distortion reduction for power amplifiers
JP4669513B2 (en) Transmission circuit and communication device

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
AD01 Patent right deemed abandoned

Effective date of abandoning: 20180814

AD01 Patent right deemed abandoned