(19)
(11)EP 1 048 176 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
07.03.2012 Bulletin 2012/10

(21)Application number: 99935163.8

(22)Date of filing:  05.08.1999
(51)International Patent Classification (IPC): 
H04Q 1/00(2006.01)
(86)International application number:
PCT/KR1999/000435
(87)International publication number:
WO 2000/008870 (17.02.2000 Gazette  2000/07)

(54)

APPARATUS AND METHOD OF LINEARIZING A POWER AMPLIFIER IN A MOBILE RADIO COMMUNICATION SYSTEM

VORRICHTUNG UND VERFAHREN ZUR LINEARISIERUNG EINES LEISTUNGSVERSTÄRKERS IN EINEM MOBILEN FUNKKOMMUNIKATIONSSYSTEM

APPAREIL ET PROCEDE DE LINEARISATION D'UN AMPLIFICATEUR DE PUISSANCE DANS UN SYSTEME DE RADIOCOMMUNICATION MOBILE


(84)Designated Contracting States:
AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE

(30)Priority: 06.08.1998 KR 9832029

(43)Date of publication of application:
02.11.2000 Bulletin 2000/44

(73)Proprietor: SAMSUNG ELECTRONICS CO., LTD.
Suwon City, Kyungki-do 442-370 (KR)

(72)Inventor:
  • HA, Ji-Won
    Seoul 151-061 (KR)

(74)Representative: Grünecker, Kinkeldey, Stockmair & Schwanhäusser 
Anwaltssozietät Leopoldstrasse 4
80802 München
80802 München (DE)


(56)References cited: : 
WO-A1-95/06354
US-A- 5 524 285
  
      
    Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


    Description

    BACKGROUND OF THE INVENTION


    1. Field of the Invention



    [0001] The present invention relates generally to a mobile radio communication system, and more particularly to an apparatus and method of linearizing the characteristic of a power amplifier in the mobile radio communication system by compensating for the non-linear characteristic of active elements included in a transmitting stage of the system.

    2. Description of the Related Art



    [0002] A high-power amplifier which is used for transmitting analog data or digital data in mobile radio communication systems requires a high spectrum efficiency as well as a high power efficiency in order to construct a low-power consuming system in a limited frequency band. In order to meet such general requirements in the system, baseband data modulation methods such as QPSK and QAM having a high spectrum efficiency have been used. Also, a high-efficiency power amplifier such as a class C amplifier has been used to improve the power efficiency of a transmitter in the system. Such a high-efficiency power amplifier generally has strong non-linear characteristics, consequently producing a distortion phenomenon, such as a sidelobe reproduction, in its output spectrum. This phenomenon is especially prevalent in the case where a modulated signal such as a QPSK or QAM signal, which does not have a constant envelope characteristic, passes through the power amplifier with the non-linear characteristic.

    [0003] Various methods have been proposed for preventing the distortion of the output spectrum of the power amplifier resulting from the non-linear characteristics of the power amplifier. One among them is a method of compensating for the non-linear characteristic of the high-power amplifier by adaptively tracking the non-linear characteristic of the power amplifier, predistorting the baseband data in a manner opposite to the distortion caused by the non-linear characteristic of the power amplifier.

    [0004] FIG. 1 is a block diagram illustrating the construction of a conventional power amplifier employing the above-described adaptive predistortion method.

    [0005] Referring to FIG. 1, K-bit data encoded by an encoder (not illustrated) is inputted to a shift register 10 and a modulation select read only memory (ROM) 52. The shift register 10 has the length of an L-symbol span, its output has a size of LK bits. At this time, if there is no linear distortion caused by the filtering operation of afinter (not illustrated) existing in the system, the length of one symbol is enough for the length L of the shift register 10, while if a linear distortion due to the system filtering exists, the length of the shift register 10 should be longer than one symbol.

    [0006] The LK-bit output of the shift register 10 is inputted to a predistort RAM 12. This predistort RAM 12 is stored with predistortion data mapped for data outputted from the shift register 10. The predistortion data is updated in accordance with input error data. Upon receiving data from the shift register 10, the predistort RAM 12 outputs predistortion data corresponding to the received data. That is, the predistort RAM 12 predistorts data outputted from the shift register 10 using error data having a phase opposite to a distortion of the transmission signal so that a radio frequency (RF) transmission section detects the distortion of the transmission signal and compensates for the detected distortion. The output of the predistort RAM 12 is converted to an analog signal for transmission by an I-channel digital-to-analog (D/A) converter 14 and a Q-channel digital-to-analog (D/A) converter 16, respectively. The analog signals converted by the respective D/A converters 14 and 16 are low-pass-filtered through low-pass filters (LPFs) 18 and 20, and then inputted to a quadrature modulator 22. The analog signals inputted to the quadrature modulator 22 are mixed with an output of a first oscillator 32, and then modulated to an intermediate frequency signal in the quadrature modulator 22. The intermediate frequency signal modulated by the quadrature modulator 22 is determined by the first intermediate frequency (IF) oscillator 32, and is mixed with an output of a second oscillator 28 by a mixer 24 to be converted to a final radio frequency (RF) transmission frequency. The RF frequency signal outputted from the mixer 24 is finally amplified by a power amplifier 26, and then transmitted through an antenna.

    [0007] A portion of the output of the power amplifier 26 is fed back to a mixer 30 by a signal coupler 54, and the fed-back signal is mixed with the output of the second oscillator 28 by the mixer 30 to be converted to a first IF signal. The converted IF signal is then converted to baseband data by a quadrature demodulator 34 using a local oscillation signal outputted from an IF oscillator 32.

    [0008] The baseband signal converted by the quadrature demodulator 34 is compared with each output signal of D/A converters 48 and 50, which is used as a reference signal for generating an error signal, by analog adders 40 and 42, respectively. Here, output signals of the modulation select ROM 52 are inputted to the D/A converters 48 and 50, and used as reference signals for comparing with the signals fed back to update the value of the predistort RAM 12. The reference signals added in the analog adders 40 and 42 and the fed-back signals are respectively converted to digital signals, and then added to the digital signals of the predistort RAM 12 by digital adders 36 and 46 to update the value of the predistort RAM 12. Error data outputted from the digital adders 3 6 and 46 are inputted to the predistort RAM 12 via a data bus, and then stored in addresses determined by the shift register 10 to complete the predistortion process with respect to the baseband data.

    [0009] However, there are some disadvantages according to the conventional predistortion method shown in FIG.1. First, the shift register 10 for generating addresses and the modulation select ROM 52 for obtaining the reference signal required for generating the error signal must be employed. Also, the high-accuracy adders 40 and 42 for obtaining the error signal must be employed to update the value of the predistort RAM 12. Constructing such high-accuracy analog adders is difficult, and is highly dependant on accuracy. In addition, it is generally known that the performance obtained by the predistortion method is lower than that obtained by a feedforward method.

    [0010] WO 95/06354 discloses a radio transmitter having a power amplifier and a linearizer arrangement. A training signal is applied to the amplifier and the linearizer arrangement is adjusted during a training mode of operation. A look-up table is provided for storing predetermined operating condition adjustment parameters. According to an operating condition input control means select operating condition adjustment parameters during the training mode and adjust the training signal to compensate for loop gain variations during the training mode.

    [0011] US-A-5524285 discloses a radio transmitter with power amplifier and linearization wherein the radio transmitter includes a power amplifier, a linearizer for maintaining linearity in the power amplifier, and a feedback path for feeding a signal from an output of the power amplifier to the linear riser. Adjustable loop linearization parameters, for example phase and gain are provided. Also a look-up table is used to store predetermined loop linearization parameters and this table cooperates with a microprocessor having an operatic condition input. US-5 699 383 discloses a prior art transmitter.

    SUMMARY OF THE INVENTION



    [0012] It is an object of the present invention to provide an apparatus and a method of linearizing the characteristics of a power amplifier in a mobile radio communication system by compensating for the non-linear characteristics of active elements included in a transmitting stage of the system wherein a predistortion data is directly generated in accordance with size of the input digital data.

    [0013] This object is solved by the features of the independent claims. Advantageous embodiments are discloses by the sub claims.

    BRIEF DESCRIPTION OF THE DRAWINGS



    [0014] The foregoing and other objects, features and advantages of the present invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which:

    FIG.1 is a block diagram illustrating the construction of a prior art power amplifier; and

    FIG. 2 is a block diagram of the linearizing apparatus of a power amplifier in accordance with a preferred embodiment of the present invention.


    DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS



    [0015] The preferred embodiment of the present invention will now be described in greater detail with reference to the drawings, in which the same or similar elements are denoted by the same reference numerals in different drawings. In the following description of the present invention, a detailed description of known functions and configurations will be omitted when it may make the subject matter of the present invention unclear.

    [0016] The apparatus and method of linearizing the characteristics of a power amplifier in a mobile radio communication system according to a preferred embodiment of the present invention utilize a predistortion algorithm for the baseband signal and a feedforward method for an RF-modulated signal to compensate for the non-linear characteristic of active elements, such as the power amplifier included in a transmitting stage of the system, and thus reduce a distortion phenomenon of an output spectrum of the RF signal.

    [0017] The address of a lookup table for generating predistortion data is directly generated in accordance with the size of input digital data. The digital data inputted for the generation of reference signals which is to be compared with fed-back signals is synchronized with fed-back received signals by adjusting a delay time using an algorithmic delay of the digital data. The modulated signal is processed by the feedforward method, and thus an additional performance improvement can be obtained as well as the performance improvement obtained by the predistortion algorithm. In other words, the performance of the system can be further improved by applying the feedforward method to the modulated signal, in comparison to the method utilizing only the baseband data predistortion algorithm.

    [0018] FIG. 2 is a block diagram of the linearizing apparatus of a power amplifier according to a preferred embodiment of the present invention.

    [0019] Referring to FIG. 2, a digital shaping filter 100 receives an input signal and produces I-channel and Q-channel signals of a baseband. Multipliers 102 and 104 receive the I-channel and Q-channel signals outputted from the digital shaping filter 100, respectively, while receiving predistortion data corresponding to the I-channel and Q-channel from a predistortion lookup table 170. The multipliers 102 and 104 output products obtained by multiplying the I-channel and Q-channel signals by the predistortion data, respectively. The D/A converters 106 and 108 convert the digital data outputted from the multipliers 102 and 104, respectively, to analog signals. Low-pass filters 110 and 112 remove unnecessary components from the output signals of the D/A converters 106 and 108, respectively. A quadrature modulator 114 performs a quadrature-modulation with respect to the output signals of the low-pass filters 110 and 112 using a local oscillation frequency outputted from an oscillator 132. The oscillator 132 outputs the local oscillation frequency to the quadrature modulator 114 and the quadrature demodulator 144.

    [0020] A first directional coupler 116 couples a portion of a modulated signal outputted from the quadrature modulator 114 to apply a coupled signal to a phase shifter 126, and provides a portion of its output to a band-pass filter 118 for passing only the signal of the transmission band. For example, when the power of the modulated signal outputted from the quadrature modulator 114 is 10 dBm, the 9 dBm component of the modulated signal is sent to the band-pass filter while the remaining I dBm component is sent to the pahse shifter 126. The band-pass filter 118 passes therethrough only the transmission band signal derived from the output signals of the first directional coupler 116. The output of the band-pass filter 118 is phase-matched with a feedforward circuit section by a delay 176, and then inputted to a pre-amplifier 174. The pre-amplifier 174 power-amplifies the output of the delay 176. A second directional coupler 120 couples an output of the pre-amplifier 174 to an output of an automatic gain control amplifier 130 as is explained later. A power amplifier 122 finally power- amplifies an output of the second directional coupler 120.

    [0021] A third directional coupler 124 feeds back a portion of an output of the power amplifier 122, similarly to the first directional coupler 120. An attenuator 142 attenuates an output of the third directional coupler 124 to a desired level. A quadrature demodulator 144 performs a quadrature-demodulation with respect to an output of the attenuator 142, based on the output of the oscillator 132 received thereto, thereby I-channel and Q-channel signals. Low-pass filters 146 and 148 low-pass-filter the I-channel and Q-channel signals from the quadrature demodulator 144, respectively. A/D converters 150 and 152 convert analog outputs of the low-pass filters 146 and 148, respectively, to digital signals.

    [0022] A delay 158 delays the I-channel and Q-channel signals from the digital shaping filter 100. Comparators 154 and 156 receive the I-channel and Q-channel signals of the input signal outputted from the delay 158, respectively, while receiving I-channel and Q-channel signals of the fed-back output signal from the A/D converters 150 and 152, respectively. The comparators 154 and 156 compare the received signals, thereby error signals, respectively. Multipliers 160 and 162 multiply corresponding outputs of the comparators 154 and 156 by adaptation constants µi and µq, respectively. Adders 164 and 166 add outputs of the corresponding multipliers 160 and 162 to I-channel and Q-channel signals of the predistortion lookup table 170, and output added signals to the predistortion lookup table 170. An address generator 168 generates addresses of the predistortion lookup table 170 using the respective I-channel and Q-channel signals outputted from the digital shaping filter 100. The predistortion lookup table 170 is stored with predistortion data for input data therein. The predistortion lookup table 170 also receives an address outputted from the address generator 168 and stores outputs from the adders 164 and 166 in the received address. That is, the predistortion lookup table 170 updates the predistort data corresponding to the address generated from the address generator 168. When the outputs from the adder 164 and 166 have a value of 0, no predistortion data in the predistortion lookup table 170 is updated.

    [0023] A square block 140 squares and adds the outputs of the comparators 154 and 156. A normalizer block 172 normalizes an output of the square block 140 to adjust the size of the output signal of the square block 140 to an address range of the feedforward lookup table 138, as is explained later. The feedforward lookup table 138 is stored with feedforward compensation data for the output from the normalizer block 172. Upon receiving a signal from the normalizer block 172, the feedforward lookup table 138 outputs feedforward compensation data corresponding to the received signal. A D/A converter 136 converts the compensation data outputted from the feedforward lookup table 138 to an analog signal. A level shifter 134 receives and shifts an output of the D/A converter 136 to a gain-control voltage range of the automatic gain control amplifier 130. A band-rejection filter (BRF) 128 removes a signal of the transmission band from the output signal of the phase shifter 126. The automatic gain control amplifier 130 receives and amplifies an output of the band-rejection filter 128 with a gain determined by a gain control voltage from a level shifter 134, and outputs an amplified signal to the second directional coupler 120.

    [0024] The operation of the linearization apparatus of the power amplifier in accordance with the above embodiment of the present invention is explained below in detail with reference to FIG. 2. The digital shaping filter 100 performs a pulse-shaping of the respective I-channel and Q-channel digital signals of the baseband. The signals pulse-shaped by the digital shaping filter 100 are inputted to the multipliers 102 and 104 to be respectively multiplied by the corresponding outputs of the predistortion lookup table 170. The output signals of the digital shaping filter 100 are inputted to the address generator 168 to be used for generating the addresses of the predistortion lookup table 170.

    [0025] The outputs of the digital shaping filter 100 are used for producing the reference signals which are compared with the fed-back signals. At this time, in order to apply the predistortion algorithm to the received data which is the same as the transmitted data, the time point of comparing the transmitted data and the received data must coincide along with a proper delay time for compensating the delay generated at the transmission and the reception stages. The delay 158 delays the output signals of the digital shaping filter 100. If a digital signal processor is employed in the system, such delay operation can be implemented by an algorithm.

    [0026] In order to update the value of the predistortion lookup table 170, the I-channel and Q-channel signals outputted from the delay 158 are respectively compared with the outputs of the A/D converters 150 and 152 to calculate the error signals. The error signals are then multiplied by the adaptation constants µi and µq for determining the convergence speed and stability of the algorithm in the respective multipliers 160 and 162. Thereafter, the signals are added to the values stored in the predistortion lookup table 170 by the adders 164 and 166, and the added signals are then stored in the address position of the predistortion lookup table 170, as determined by the address generator 168. Since the signals multiplied by the multipliers 102 and 104 are the values prior to the update of the predistortion lookup table 170, the predistortion amount to be applied to following data will be determined by the property of the input data prior to one sample. The output signals of the multipliers 102 and 104 are converted to analog signals by the D/A converters 106 and 108, and the analog signals are inputted to the quadrature modulator 114 through the low-pass filters 110 and 112 for removing the unnecessary high-frequency components.

    [0027] A portion of the output of the quadrature modulator 114, after passing through the first directional coupler 116, is inputted to the band-pass filter 118, while the other portion thereof is inputted to the phase shifter 126. At this time, the phase shifter 126 serves to shift the phase of the modulated signal by 180 degrees, and this is necessary to produce the predistortion signal used in the feedforward method. The phase-reversed signal outputted from the phase shifter 126 is filtered by the band-pass filter 128, so that the signal of the transmission band is removed, but the signal excepting the transmission band passes through the band-pass filter 128 to be inputted to the automatic gain control amplifier 130. The output of the first directional coupler 116 is filtered by the band-pass filter 118 for passing only the signal of the transmission band, delayed by the delay 176 for compensating for the delay caused by the feedforward circuit, and then amplified by the pre-amplifier 174. The output of the pre-amplifier 174 is added to the output of the automatic gain control amplifier 130 by the second directional coupler 120, and the added signal is finally inputted to the power amplifier 122.

    [0028] A portion of the output of the power amplifier 122 is coupled by the third directional coupler 124, attenuated by the attenuator 142, and then demodulated by the quadrature demodulator 144. The demodulated signal is filtered through the respective low-pass filters 146 and 148, respectively, and the filtered signals are converted to digital signals by the A/D converters 150 and 152. The outputs of the A/D converters 150 and 152 are compared with the delayed input signals outputted from the delay 158 by the comparators 154 and 156 to produce the error signals. The error signals are used not only for updating the values of the predistortion lookup table 170, but also as the input signals of the feedforward circuit according to the present invention. Specifically, the outputs of the comparators 154 and 156 are squared and then added by the square block 140, and the output of the square block 140 is inputted to the normalizer block 172, so that the size of the output of the square block 140 is adjusted to be within the address range of the feedforward lookup table 138. The normalized signal is then applied to the address of the feedforward lookup table 138.

    [0029] The output of the feedforward lookup table 138 is converted to an analog signal by the D/A converter 136 to be inputted to the level shifter 134. This is required to match the output voltage of the D/A converter 136 with the gain control voltage range of the automatic gain control amplifier 130. The gain of the automatic gain control amplifier 130 is adjusted by the control voltage of the automatic gain control amplifier 130, which is the output of the level shifter 134, and this causes the size of the transmission signal to be inversely amplified according to the distortion amount of the fed-back signal. Specifically, if the error between the transmission signal and the fed-back signal is large, the distortion signal also becomes large. In this case, the size of the reverse-phased signal which exists outside the transmission band becomes larger by increasing the gain of the automatic gain control amplifier 130. If the error between the transmission signal and the fed-back signal is small, the distortion signal also becomes small. In this case, the size of the reverse-phased signal which exists outside the transmission band becomes smaller by decreasing the gain of the automatic gain control amplifier 130.

    [0030] According to the above-described method, the phase-reversed unnecessary signal existing outside the transmission band, which is outputted from the automatic gain control amplifier 130, is coupled to the original transmission signal by the second directional coupler 120, and then finally radiated through the power amplifier 122. The system according to the embodiment of the present invention compensates for the non-linear characteristic of the power amplifier 122 by predistorting the data of the baseband, and then further compensates for the spectrum distortion phenomenon using the feedforward method by utilizing the modulated signal as well. Accordingly, the system performance can be improved in comparison to the system for linearizing the power amplifier only using the data predistortion algorithm.

    [0031] As described above, according to the linearization apparatus of the power amplifier according to the embodiment of the present invention, the address of the lookup table for generating the predistortion data is directly generated in accordance with the size of the input digital data. The digital data inputted for the generation of the reference signals, which is to be compared with fed-back signals, is synchronized with fed-back received signals by adjusting the delay time utilizing the algorithmic delay of the digital data. Also, the modulated signal is processed by the feedforward method, and thus an additional performance improvement can be obtained as well as the performance improvement obtained by the predistortion algorithm. In other words, the performance of the system can be further improved by applying the feedforward method to the modulated signal in comparison to the method utilizing only the baseband data predistortion algorithm.

    [0032] While this invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiment, but, on the contrary, it is intended to cover various modifications within the scope of the invention as described in the appended claims.


    Claims

    1. An apparatus for linearizing a power amplifier in a mobile radio communication system comprising:

    an error detector (158; 124, 142, 144, 146, 148, 150, 152; 154, 156; 160, 162) for detecting an error between input and power-amplified output signals of the power amplifier by comparing the input signal with a portion of the output signal fed back to obtain an error signal;

    a predistortion lookup table (170) for storing predistortion data;

    a predistortion lookup table controller (168; 164, 166) for updating the data of the predistortion lookup table which corresponds to a position of a present input data by adding the error signal from the error detector to an output of the predistortion lookup table;

    a feedforward lookup table (138) for storing feedforward control data;

    a feedforward lookup table controller (140, 172) for outputting corresponding feedforward control data of the feedforward lookup table by detecting a size of the error signal;

    a linearizer (126, 128, 130) for distorting the input signal in accordance with the predistortion data, controlling a gain of the input signal in accordance with the feedforward control data, and coupling a gain-controlled output to the input signal of the power amplifier (120).


     
    2. The linearizing apparatus as recited in claim 1, wherein the error detector comprises:

    a delay (158) for delaying the input signal;

    a feedback block (124, 142, 144, 146, 148, 150, 152) for feeding back and attenuating the output of the power amplifier;

    a comparator (154, 156) for detecting the error signal by comparing an output of the feedback block and an output of the delay; and

    a multiplier (160, 162) for producing the error signal by multiplying an output of the comparator by an adaption constant.


     
    3. The linearizing apparatus as recited in claim 1, wherein the predistortion lookup table controller comprises:

    an address generator (168) for generating an address of the predistortion lookup table using the input signal; and

    an adder (164, 166) for adding the error signal to the output of the predistortion lookup table to store updated predistortion lookup data corresponding to the address in the predistortion lookup table corresponding to the address.


     
    4. The linearizing apparatus as recited in claim 1, wherein the feedforward lookup table controller comprises:

    a square block (140) for obtaining a size of the error signal by squaring the error signal; and

    a normalizer block (172) for normalizing the output of the square block and outputting a normalized output to the feedforward lookup table.


     
    5. The linearizing apparatus as claimed in claim 1, wherein the linearizer comprises:

    a multiplier for predistorting the input signal by multiplying the input signal by the predistortion data, and outputting the predistorted signal to the power amplifier;

    a phase shifter (126) for phase-shifting an output of the multiplier;

    a band-rejection filter (128) for removing a signal band from an output of the band-rejection filter; and

    an automatic gain control amplifier (130) for controlling a gain of an output of the band-rejection filter in accordance with the feedforward control data, and coupling the gain-controlled output to the input signal of the power amplifier.


     
    6. The linearizing apparatus according to claim 1 comprising:

    an address generator (168) for generating an address of the predistortion lookup table using an input signal;

    means for distorting the input signal by the predistortion data;

    a quadrature modulator (114) for modulating a predistorted signal;

    a delay (176) for delaying a modulated signal;

    said linearizer(126, 130) for removing a signal band by coupling a delayed signal, gain controlling a filtered signal in accordance with the feedforward control data, and then coupling a gain-controlled signal to an output of the modulator; and

    a quadrature demodulator (144) for demodulating a power-amplified signal by coupling.


     
    7. The linearizing apparatus as recited in claim 6, wherein the error detector comprises:

    a delay (158) for delaying the input signal;

    a comparator (154, 156) for detecting the error signal by comparing an output of the power amplifier and an output of the delay; and

    a multiplier (160, 162) for producing the error signal by multiplying an output of the comparator by an adaption constant.


     
    8. The linearizing apparatus as recited in claim 6, wherein the feedforward lookup table controller comprises:

    a square block (140) for obtaining a size of the error signal by squaring the error signal; and

    a normalizer block (172) for normalizing the output of the square block and outputting a normalized output to the feedforward lookup table.


     
    9. A method of linearizing a power amplifier in a mobile radio communication system having a lookup table (170) for storing predistortion data and feedforward control data, comprising the steps of:

    linearizing an input signal by distorting the input signal using the stored predistortion data, controlling a gain of the predistorted input signal in accordance with the stored feedforward control data and coupling a gain-controlled output to the input signal of a power-amplifier;

    power-amplifying and outputting the linearized input signal;

    producing an error signal by comparing a portion of signal resulting from a feed-back of the output signal with the delayed input signal; and

    updating the predistortion lookup table by adding the error signal to the predistortion data and storing the added data in the predistortion lookup table determined by the input data, and applying the error signal to an address of the feedforward lookup table.


     
    10. The linearizing method as recited in claim 10, wherein the linearizing step comprises the steps of:

    producing a predistorted signal by multiplying the input signal by the predistortion data;

    phase-shifting the predistorted signal;

    removing a signal band from the phase-shifted signal; and

    controlling a gain of the band-rejected signal in accordance with the feedforward control data and coupling the gain-controlled signal to the predistorted signal.


     
    11. The linearizing method as in claim 10, wherein the error signal producing step comprises the steps of:

    delaying the input signal;

    detecting the error signal by comparing the output signal with the delayed signal; and

    producing the error signal by multiplying the detected error signal by an adaption constant.


     
    12. The linearizing method as recited in claim 9, wherein the predistortion data updating step further comprises the step of storing the predistortion data updated by adding the error signal to the output of the lookup table in the predistortion lookup table corresponding to the address.
     
    13. The linearizing method as recited in claim 9, wherein the feedforward control data producing step comprises the steps of:

    obtaining a size of the error signal by squaring the error signal; and

    normalizing the size of the error signal and outputting the normalized error signal to the feedforward lookup table.


     


    Ansprüche

    1. Vorrichtung zum Linearisieren eines Leistungsverstärkers in einem Mobilfunk-Kommunikationssystem, die umfasst:

    eine Abweichungs-Erfassungseinrichtung (158; 124, 142, 144, 146, 148, 150, 152, 154, 156, 160, 162), mit der eine Abweichung zwischen einem Eingangs-Signal und einem Leistungsverstärkung unterzogenen Ausgangs-Signal des Leistungsverstärkers erfasst wird, indem das Eingangs-Signal mit einem Teil des Ausgangs-Signals verglichen wird, der zurückgekoppelt wird, um ein Abweichungs-Signal zu gewinnen;

    eine Vorverzerrungs-Verweistabelle (170) zum Speichern von Vorzerrungs-Daten;

    eine Steuereinrichtung (168; 164, 166) für die Vorverzerrungs-Verweistabelle, mit der die Daten der Vorverzerrungs-Verweistabelle aktualisiert werden, die einer Position aktueller Eingangs-Daten entsprechen, indem das Abweichungs-Signal von der Abweichungs-Erfassungseinrichtung zu einem Ausgang der Vorverzerrungs-Verweistabelle addiert wird;

    eine Vorwärtsregelungs-Verweistabelle (138) zum Speichern von Vorwärtsregelungs-Daten;

    eine Steuereinrichtung (140, 172) für die Vorwärtsregelungs-Verweistabelle, mit der entsprechende Vorwärtsregelungs-Daten der Vorwärtsregelungs-Verweistabelle ausgegeben werden, indem eine Größe des Abweichungs-Signals erfasst wird;

    eine Linearisierungseinrichtung (126, 128, 130), mit der das Eingangs-Signal entsprechend den Vorverzerrungs-Daten verzerrt wird, eine Verstärkung des Eingangs-Signals entsprechend den Vorwärtsregelungs-Daten geregelt wird und ein Verstärkungsregelung unterzogener Ausgang mit dem Eingangs-Signal des Leistungsverstärkers (120) gekoppelt wird.


     
    2. Linearisierungsvorrichtung nach Anspruch 1, wobei die Abweichungs-Erfassungseinrichtung umfasst:

    ein Verzögerungsglied (158) zum Verzögern des Eingangs-Signals;

    einen Rückkoppelungs-Block (124, 142, 144, 146, 148, 150, 152), mit dem der Ausgang des Leistungsverstärkers zurückgekoppelt und gedämpft wird;

    einen Komparator (154, 156), mit dem das Abweichungs-Signal erfasst wird, indem ein Ausgang des Rückkoppelungs-Blocks und ein Ausgang des Verzögerungsgliedes verglichen werden; und

    einen Multiplizierer (160, 162), mit dem das Abweichungs-Signal erzeugt wird, indem ein Ausgang des Komparators mit einer Adaptions-Konstante multipliziert wird.


     
    3. Linearisierungsvorrichtung nach Anspruch 1, wobei die Steuereinrichtung der Vorzerrungs-Verweistabelle umfasst:

    einen Adressgenerator (168), mit dem eine Adresse der Vorverzerrungs-Verweistabelle unter Verwendung des Eingangs-Signals erzeugt wird; und

    einen Addierer (164, 166), mit dem das Abweichungs-Signal zu dem Ausgang der Vorverzerrungs-Verweistabelle addiert wird, um aktualisierte Vorverzerrungs-Verweisdaten, die der Adresse entsprechen, in der Vorverzerrungs-Verweistabelle zu speichern, die der Adresse entspricht.


     
    4. Linearisierungsvorrichtung nach Anspruch 1, wobei die Steuereinrichtung der Vorwärtsregelungs-Verweistabelle umfasst:

    einen Quadrier-Block (140), mit dem eine Größe des Abweichungs-Signals ermittelt wird, indem das Abweichungs-Signal quadriert wird; und

    einen Normalisier-Block (172), mit dem der Ausgang des Quadrier-Blocks normalisiert wird und ein normalisierter Ausgang an die Vorwärtsregelungs-Verweistabelle ausgegeben wird.


     
    5. Linearisierungsvorrichtung nach Anspruch 1, wobei die Linearisierungseinrichtung umfasst:

    einen Multiplizierer, mit dem das Eingangs-Signal vorverzerrt wird, indem das Eingangs-Signal mit den Vorverzerrungs-Daten multipliziert wird, und das vorverzerrte Signal an den Leistungsverstärker ausgegeben wird;

    einen Phasenschieber (126), mit dem eine Phase des Ausgangs des Multiplizierers verschoben wird;

    ein Sperrfilter (128), mit dem ein Signalband aus einem Ausgang des Sperrfilters entfernt wird; und

    einen automatischen Regelverstärker (130), mit dem eine Verstärkung eines Ausgangs des Sperrfilters entsprechend den Vorwärtsregelungs-Daten geregelt wird und der verstärkungsgeregelte Ausgang mit dem Eingangs-Signal des Leistungsverstärkers gekoppelt wird.


     
    6. Linearisierungsvorrichtung nach Anspruch 1, die umfasst:

    einen Adressgenerator (168) zum Erzeugen einer Adresse der Vorverzerrungs-Verweistabelle unter Verwendung eines Eingangs-Signals;

    eine Einrichtung zum Verzerren des Eingangs-Signals mittels der Vorverzerrungs-Daten;

    eine Quadraturmodulations-Einrichtung (114) zum Modulieren eines vorverzerrten Signals;

    ein Verzögerungsglied (176) zum Verzögern eines modulierten Signals;

    die Linearisierungseinrichtung (126, 130), mit der ein Signalband durch Koppeln eines verzögerten Signals entfernt wird, Verstärkungsregelung eines gefilterten Signals entsprechend den Vorwärtsregelungs-Daten durchgeführt wird und dann ein Verstärkungsregelung unterzogenes Signal zu einem Ausgang der Modulationseinrichtung gekoppelt wird; und

    eine Quadraturdemodulations-Einrichtung (144) zum Demodulieren eines Leistungsverstärkung unterzogenen Signals durch Koppeln.


     
    7. Linearisierungsvorrichtung nach Anspruch 6, wobei die Abweichungs-Erfassungseinrichtung umfasst:

    ein Verzögerungsglied (158) zum Verzögern des Eingangs-Signals;

    einen Komparator (154, 156), mit dem das Abweichungs-Signal erfasst wird, indem ein Ausgang des Leistungsverstärkers und ein Ausgang des Verzögerungsgliedes verglichen werden; und

    einen Multiplizierer (160, 162), mit dem das Fehler-Signal erzeugt wird, indem ein Ausgang des Komparators mit einer Adaptions-Konstante multipliziert wird.


     
    8. Linearisierungsvorrichtung nach Anspruch 6, wobei die Steuereinrichtung der Vorwärtsregelungs-Verweistabelle umfasst:

    einen Quadrier-Block (140), mit dem eine Größe des Abweichungs-Signals ermittelt wird, indem das Abweichungs-Signal quadriert wird; und

    einen Normalisier-Block (172), mit dem der Ausgang des Quadrier-Blocks normalisiert wird und ein normalisierter Ausgang an die Vorwärtsregelungs-Verweistabelle ausgegeben wird.


     
    9. Verfahren zum Linearisieren eines Leistungsverstärkers in einem Mobilfunk-Kommunikationssystem mit einer Verweistabelle (170) zum Speichern von Vorverzerrungs-Daten und Vorwärtsregelungs-Daten, das die folgenden Schritte umfasst:

    Linearisieren eines Eingangs-Signals durch Verzerren des Eingangs-Signals unter Verwendung der gespeicherten Vorverzerrungs-Daten, Regeln einer Verstärkung des vorverzerrten Eingangs-Signals entsprechend den gespeicherten Vorwärtsregelungs-Daten und Koppeln eines Verstärkungsregelung unterzogenen Ausgangs mit dem Eingangs-Signal eines Leistungsverstärkers;

    Leistungsverstärken und Ausgeben des linearisierten Eingangs-Signals;

    Erzeugen eines Abweichungs-Signals durch Vergleichen eines Teils eines Signals, das aus einer Rückkopplung des Ausgangs-Signals resultiert, mit dem verzögerten Eingangs-Signal; und

    Aktualisieren der Vorverzerrungs-Verweistabelle durch Addieren des Abweichungs-Signals zu den Vorverzerrungs-Daten und Speichern der addierten Daten in der Vorverzerrungs-Verweistabelle, die durch die Eingangs-Daten bestimmt wird, sowie Anlegen des Abweichungs-Signals an eine Adresse der Vorwärtsregelungs-Verweistabelle.


     
    10. Linearisierungsverfahren nach Anspruch 9, wobei der Linearisierungsschritt die folgenden Schritte umfasst:

    Erzeugen eines vorverzerrten Signals durch Multiplizieren des Eingangs-Signals mit den Vorverzerrungs-Daten;

    Verschieben der Phase des vorverzerrten Signals;

    Entfernen eines Signalbandes aus dem Phasenverschiebung unterzogenen Signal; und

    Regeln einer Verstärkung des Sperrfiltern unterzogenen Signals entsprechend den Vorwärtsregelungs-Daten und Koppeln des Verstärkungsregelung unterzogenen Signals mit dem vorverzerrten Signal.


     
    11. Linearisierungsverfahren nach Anspruch 10, wobei der Schritt des Erzeugens eines Abweichungs-Signals die folgenden Schritte umfasst:

    Verzögern des Eingangs-Signals;

    Erfassen des Abweichungs-Signals durch Vergleichen des Ausgangs-Signals mit dem verzögerten Signal; und

    Erzeugen des Abweichungs-Signals durch Multiplizieren des erfassten Abweichungs-Signals mit einer Adaptions-Konstante.


     
    12. Linearisierungsverfahren nach Anspruch 9, wobei der Schritt des Aktualisierens der Vorverzerrungs-Daten den Schritt umfasst, in dem die durch Addieren des Abweichungs-Signals zu dem Ausgang der Verweistabelle aktualisierten Vorverzerrungs-Daten in der Vorverzerrungs-Verweistabelle gespeichert werden, die der Adresse entspricht.
     
    13. Linearisierungsverfahren nach Anspruch 9, wobei der Schritt des Erzeugens von Vorwärtsregelungs-Daten die folgenden Schritte umfasst:

    Ermitteln einer Größe des Abweichungs-Signals durch Quadrieren des Abweichungs-Signals; und

    Normalisieren der Größe des Abweichungs-Signals und Ausgeben des normalisierten Abweichungs-Signals an die Vorwärtsregelungs-Verweistabelle.


     


    Revendications

    1. Appareil pour linéariser un amplificateur de puissance dans un système de radiocommunications mobiles, comprenant :

    un détecteur d'erreur (158 ; 124, 142, 144, 146, 148, 150, 152 ; 154, 156 ; 160, 162) pour détecter une erreur entre un signal d'entrée et un signal de sortie amplifié en puissance de l'amplificateur de puissance en comparant le signal d'entrée à une partie du signal de sortie en retour afin d'obtenir un signal d'erreur,

    une table de consultation de prédistorsion (170) pour mémoriser des données de prédistorsion,

    un contrôleur de table de consultation de prédistorsion (168 ; 164, 166) pour mettre à jour les données de la table de consultation de prédistorsion qui correspondent à une position de données actuelles en entrée en ajoutant le signal d'erreur issu du détecteur d'erreur à une sortie de la table de consultation de prédistorsion,

    une table de consultation de correction aval (138) pour mémoriser des données de commande de correction aval,

    un contrôleur de table de consultation de correction aval (140, 172) pour fournir en sortie des données de commande de correction aval correspondantes de la table de consultation de correction aval en détectant une taille du signal d'erreur,

    un dispositif de linéarisation (126, 128, 130) pour effectuer une distorsion du signal d'entrée en fonction des données de prédistorsion, commander un gain du signal d'entrée en fonction des données de commande de correction aval, et coupler une sortie commandée en gain sur le signal d'entrée de l'amplificateur de puissance (120).


     
    2. Appareil de linéarisation selon la revendication 1, dans lequel le détecteur d'erreur comprend :

    un circuit de retard (158) pour retarder le signal d'entrée,

    un bloc de rétroaction (124, 142, 144, 146, 148, 150, 152) pour retourner et atténuer la sortie de l'amplificateur de puissance,

    un comparateur (154, 156) pour détecter le signal d'erreur en comparant une sortie du bloc de rétroaction et une sortie du circuit de retard, et

    un multiplieur (160, 162) pour générer le signal d'erreur en multipliant la sortie du comparateur par une constante d'adaptation.


     
    3. Appareil de linéarisation selon la revendication 1, dans lequel le contrôleur de table de consultation de prédistorsion comprend :

    un générateur d'adresse (168) pour générer une adresse de la table de consultation de prédistorsion en utilisant le signal d'entrée, et

    un additionneur (164, 166) pour ajouter le signal d'erreur à la sortie de la table de consultation de prédistorsion afin de mémoriser des données de consultation de prédistorsion mises à jour correspondant à l'adresse dans la table de consultation de prédistorsion correspondant à l'adresse.


     
    4. Appareil de linéarisation selon la revendication 1, dans lequel le contrôleur de table de consultation de correction aval comprend :

    un bloc d'élévation au carré (140) pour obtenir une taille du signal d'erreur en élevant le signal d'erreur au carré, et

    un bloc de normalisation (172) pour normaliser la sortie du bloc d'élévation au carré et fournir une sortie normalisée à la table de consultation de correction aval.


     
    5. Appareil de linéarisation selon la revendication 1, dans lequel le dispositif de linéarisation comprend :

    un multiplieur pour effectuer la prédistorsion du signal d'entrée en multipliant le signal d'entrée par les données de prédistorsion, et fournir en sortie le signal prédistordu à l'amplificateur de puissance,

    un déphaseur (126) pour déphaser la sortie du multiplieur,

    un filtre de réjection de bande (128) pour supprimer une bande de signal d'une sortie du filtre de réjection de bande, et

    un amplificateur de commande de gain automatique (130) pour commander un gain d'une sortie du filtre de réjection de bande en fonction des données de commande de correction aval, et coupler la sortie commandée en gain au signal d'entrée de l'amplificateur de puissance.


     
    6. Appareil de linéarisation selon la revendication 1, comprenant :

    un générateur d'adresse (168) pour générer une adresse de la table de consultation de prédistorsion en utilisant un signal d'entrée,

    un moyen pour effectuer une distorsion du signal d'entrée grâce aux données de prédistorsion,

    un modulateur en quadrature (114) pour moduler le signal prédistordu,

    un circuit de retard (176) pour retarder le signal modulé,

    ledit dispositif de linéarisation (126, 130) permettant de supprimer une bande du signal en couplant un signal retardé, de commander le gain du signal filtré en fonction des données de commande de correction aval, puis de coupler le signal commandé en gain à la sortie du modulateur, et

    un démodulateur en quadrature (144) pour démoduler un signal amplifié en puissance par couplage.


     
    7. Appareil de linéarisation selon la revendication 6, dans lequel le détecteur d'erreur comprend :

    un circuit de retard (158) pour retarder le signal d'entrée,

    un comparateur (154, 156) pour détecter le signal d'erreur en comparant une sortie de l'amplificateur de puissance à une sortie du circuit de retard, et

    un multiplieur (160, 162) pour générer le signal d'erreur en multipliant une sortie du comparateur par une constante d'adaptation.


     
    8. Appareil de linéarisation selon la revendication 6, dans lequel le contrôleur de table de consultation de correction aval comprend :

    un bloc d'élévation au carré (140) pour obtenir une taille du signal d'erreur en élevant le signal erreur au carré, et

    un bloc de normalisation (172) pour normaliser la sortie du bloc d'élévation au carré et fournir une sortie normalisée à la table de consultation de correction aval.


     
    9. Procédé de linéarisation d'un amplificateur de puissance dans un système de radiocommunications mobiles comportant une table de consultation (170) pour mémoriser des données de prédistorsion et des données de commande de correction aval, comprenant les étapes consistant à:

    linéariser un signal d'entrée en effectuant une distorsion du signal d'entrée en utilisant les données mémorisées de prédistorsion, commander un gain du signal d'entrée prédistordu en fonction des données mémorisées de commande de correction aval, et coupler une sortie commandée en gain au signal d'entrée d'un amplificateur de puissance,

    amplifier en puissance et fournir en sortie le signal d'entrée linéarisé,

    générer un signal d'erreur en comparant une partie du signal résultant d'une rétroaction du signal de sortie au signal d'entrée retardé, et

    mettre à jour la table de consultation de prédistorsion en ajoutant le signal d'erreur aux données de prédistorsion et en mémorisant les données ajoutées dans la table de consultation de prédistorsion déterminées par les données d'entrée, et appliquer le signal d'erreur à une adresse de la table de consultation de correction aval.


     
    10. Procédé de linéarisation selon la revendication 10, dans lequel l'étape de linéarisation comprend les étapes consistant à :

    générer un signal prédistordu en multipliant le signal d'entrée par les données de prédistorsion,

    déphaser le signal prédistordu,

    supprimer une bande de signal du signal déphasé, et

    commander un gain du signal à bande rejetée en fonction des données de commande de correction aval et coupler le signal commandé en gain au signal prédistordu.


     
    11. Procédé de linéarisation selon la revendication 10, dans lequel l'étape de génération du signal d'erreur comprend les étapesconsistant à :

    appliquer un retard au signal d'entrée,

    détecter le signal d'erreur en comparant le signal de sortie au signal retardé, et

    générer le signal d'erreur en multipliant le signal d'erreur détecté par une constante d'adaptation.


     
    12. Procédé de linéarisation selon la revendication 9, dans lequel l'étape de mise à jour des données de prédistorsion comprend en outre l'étape de mémorisation des données de prédistorsion mises à jour en ajoutant le signal d'erreur à la sortie de la table de consultation dans la table de consultation de prédistorsion correspondant à l'adresse.
     
    13. Procédé de linéarisation selon la revendication 9, dans lequel l'étape de génération des données de commande de correction aval comprend les étapes consistant à :

    récupérer une taille du signal d'erreur en élevant le signal d'erreur au carré, et

    normaliser la taille du signal d'erreur et fournir en sortie le signal d'erreur normalisé à la table de consultation de correction aval.


     




    Drawing











    Cited references

    REFERENCES CITED IN THE DESCRIPTION



    This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

    Patent documents cited in the description