Method and device for training an rf amplifier linearization device, and mobile terminal incorporating same

Translation of description of equivalent WO 2004045066 A1

METHOD AND DEVICE FOR LEARNING LINEARIZATION DEVICE AMPLIFIER RF, AND MOBILE TERMINAL INCLUDING SUCH DEVICE

The present invention relates to linearization of RF power amplifiers (RF). It finds applications in particular in the RF transmitters of the mobile terminals of digital radio communication systems. It can also be applied in RF transmitter base stations especially during the first start of such a station.

In current digital radio systems, it is desired to transmit information with a maximum throughput in an RF frequency band that is assigned to a transmission channel (hereinafter radio channel). To do this, the modulations used in recent years include a phase or frequency modulation component and an amplitude modulation component.

In addition, radio channels coexist in a specific frequency band allocated to the system. Each radio channel is subdivided into logical channels by time division. In each time period ( "Time Slot" in English), it is issued a symbol group called burst or package ( "Burst" in English).

It is necessary to ensure that, at every moment, the power level emitted in each radio channel does not interfere communications in an adjacent radio channel. Thus, specifications require that the power level of an RF signal transmitted in a radio channel is determined in an adjacent radio channel, for example lower than 60 dB (decibels), the RF signal power level transmitted in said channel determined radio.

It is therefore necessary that the spectrum of the signal to be transmitted, resulting in particular of the type of modulation used and the bit rate is not distorted by the RF transmitter. In particular, it is necessary that the RF transmitter has an output power characteristic as a function of the input power, that is linear. However, radio frequency power amplifier (hereinafter RF amplifier) ​​present in the RF transmitter has a linear characteristic with low power output nonlinear but as soon as the power exceeds a certain threshold. We also know that the performance of the RF amplifier is all the better that we work in an area close to saturation that is to say in the nonlinear regime. Thus the need for linearity and the need for high efficiency (to save battery) require the use of linearization techniques to correct for non-linearities of the RF amplifier. Two of the most commonly used techniques are the adaptive predistortion baseband and the Cartesian loop baseband.

In the technique of the Cartesian loop, the signal to be transmitted is generated in baseband format I and Q. In addition, a coupler followed by a demodulator possible to sample a portion of the transmitted RF signal and to transpose the band of base (downconversion), format I and Q. the baseband signal is compared with the baseband signal to be transmitted. An error signal resulting from this comparison drives a modulator, which transposes to the radiofrequency domain (upconversion). The modulator output signal is amplified by an RF amplifier which delivers the transmitted RF signal. In the art of adaptive predistortion in baseband, the signal to be transmitted is generated in baseband in I and Q format, predistorted via a predistortion device. Then, this signal is transposed to the RF domain through an RF modulator. Then, it is amplified in an RF amplifier. A coupler followed by an RF demodulator used to remove part of the transmitted RF signal and translate the baseband, size I, Q. signal demodulated baseband is digitized and compared with the baseband signal to emit. An adaptation of the pre-distortion coefficients, performed during a learning phase of the predistortion device, allows to converge the format signal to demodulated I and Q signal in I and Q format to be output.

In both techniques, a portion of the transmitted signal is tapped at the output of the RF amplifier to compare it with the signal to be transmitted. The result that the linearity is not obtained immediately but only after a certain time necessary for the convergence of the linearization device. Learning the linearization device requires the transmission of a particular data sequence or training sequence. This observation certainly applies to adaptive predistortion than the Cartesian loop, even if it requires to ensure its stability, initial adjustment phase and amplitude levels akin to an apprenticeship.

The learning method disclosed in WO 94/10765, and is based on the issuance by the issuers of the system of particular sequences, called linearization training sequences, during linearization training phases. More particularly, training sequences are transmitted in isolation in time intervals forming a particular logical channel of the radio channels, which is only dedicated to linearization. However, this method has several disadvantages. First, it requires a preliminary synchronization of all transmitters so that they emit their respective linearization training sequence in the logical channel dedicated to linearization. Moreover, no data transmission can take place in the time intervals of this logical channel. In addition, at the beginning of each issue or when changing radio channel, the issuer is obliged to wait for the next time interval of the logical channel dedicated to linearization, unless considerably complicate the system. This is why the temporal spacing between two time intervals of said logical channel can not exceed the second, to guarantee a certain quality of service (QoS). This technique is very detrimental to the spectral efficiency of the radio communication system.

In general, there are radio communication systems in which the frame structure is not suitable for transmission of a training sequence, for example where any specific time interval has been provided for this purpose during the definition of the frame structure. In order to overcome all or part of the disadvantages of the aforementioned prior art, a first aspect of the invention relates to a learning method of a linearization device of a radiofrequency amplifier which is included within a radiofrequency transmitter of a terminal movable from a radio communication system comprising a fixed network and mobile terminals, which transmitter is adapted for transmitting bursts according to a determined frame structure, each burst comprising symbols belonging to a determined alphabet of symbols. The method comprises the steps of: a) generating a linearization training sequence comprising a determined number N of symbols, where N is a determined integer; b) transmitting the linearization training sequence by means of the radiofrequency transmitter, in at least some bursts transmitted by the latter; c) comparing the linearization training sequence transmitted with the linearization training sequence generated so as to drive said linearization device.

Advantageously, in step b), the linearization training sequence is included in a sequence of symbols further provided to allow adjustment of parameters of the transmission chain between said first equipment and a second equipment of the radiocommunication system with wherein said first equipment communicates.

Transmission chain means all the components participating in two-way communication between a first and a second device, typically a mobile terminal and base station with which it communicates.

Preferably, the symbol sequence provided to enable the setting of parameters is a sequence of symbols provided to allow the dynamic control of the gain of an amplifier of a variable gain radio frequency receiver of a second radio communication system of the equipment with which the first equipment communicates. In other words, the training sequence is transmitted in step b) within a time interval reserved in the frame structure for transmitting a sequence of AGC (Automatic Gain Control), and it ensures at the same time the role of such a sequence CAG. Is thus used for transmission of the training sequence of transmission time for a sequence of symbols necessary for other purposes, namely an AGC sequence transmitted to allow the dynamic control of the power of issue handheld reception.

According to an advantage, the value of the symbols of the AGC sequence is not subject to any constraints (the AGC sequence must simply be known fixed network). There are thus free to select the symbols of the sequence, or at least a part of symbols of the sequence, so that these symbols form a satisfactory training sequence.

According to another advantage, the recurrence of the AGC sequence is adapted to the needs of learning of the RF amplifier linearization device. In fact, the AGC sequence is generally emitted in the top frame and during a change of logical channel when an RF frequency change and / or during a change of power level. Now it is substantially at such times as the linearization training sequence needs to be issued.

A second aspect of the invention relates to a learning device of a linearization device of a radiofrequency amplifier which is included within a radiofrequency transmitter of a first equipment of a radiocommunication system, which transmitter is adapted to transmit bursts according to a determined frame structure, each burst comprising symbols belonging to a determined alphabet of symbols. The device comprises: a) means for generating a linearization training sequence comprising a determined number N of symbols, where N is a determined integer; b) means for transmitting the linearization training sequence by means of the transmitter in at least certain of the bursts transmitted by the latter; c) means for comparing the linearization training sequence transmitted with the linearization training sequence generated so as to drive said linearization device.

Advantageously, the linearization training sequence is included in a sequence of symbols further provided to allow adjustment of parameters of the transmission chain between said first equipment and a second equipment of the radiocommunication system with which said first equipment communicates.

Preferably, the symbol sequence provided to enable the setting of parameters is a sequence of symbols provided to allow the dynamic control of the gain of an amplifier of a variable gain radio frequency receiver of a second radio communication system of the equipment with which the first equipment communicates.

Stated otherwise, said means for transmitting are adapted to transmit the training sequence within a time interval reserved in the frame structure for transmitting a sequence of CAG, and the training sequence ensures at the same time the role of such a sequence CAG.

A third aspect of the invention relates to a mobile terminal of a radiocommunication system comprising a radiofrequency transmitter

'Having a radiofrequency amplifier and a radio frequency amplifier linearization device, which further comprises a device for training the linearization device of the second aspect.

A fourth aspect of the invention relates to a base station of a radiocommunication system comprising a radiofrequency transmitter having a radiofrequency amplifier and a radio frequency amplifier linearization device, which further comprises a learning device of the device of linearization of the third aspect. Other features and advantages of the invention will appear on reading the following description. This is purely illustrative and should be read with reference to the accompanying drawings in which:

- Figure 1 is a block diagram of an exemplary mobile terminal according to the invention;

- Figure 2 is a diagram illustrating a first example of the bursts transmitted by the mobile terminal, without AGC sequence;

- Figure 3 is a diagram illustrating a second example of the bursts transmitted by the mobile terminal, with an AGC sequence which according to the invention comprises a linearization training sequence; and,

- Figure 4 is a diagram illustrating the implementation of an AGC method between a first and a second device, and vice versa.

In Figure 1, shows schematically how an exemplary mobile terminal of the invention. Such a mobile terminal belongs for example to a radio communication system which further comprises a fixed network with base stations.

The terminal includes a transmit subsystem 100, a receive chain 200, a control unit 300, and a permanent memory 400, and a device 500 for automatic gain control (AGC) associated with an RF receiver the receive chain 200.

The transmission system 100 comprises a source of useful data 10, for example a speech coder delivering voice coding data. The source 10 is coupled to an M-ary data modulator 20 which modulates the data in baseband transmit according to a modulation with M distinct states, where M is an integer determined. Binary data it receives from the source 10 are reflected by the modulator 20 into symbols belonging to an M-ary alphabet that is to say comprising M distinct symbols. The output of modulator 20 is coupled to the input of a radio frequency transmitter 30. From the sequence of received symbols, the transmitter 30 produces an RF signal suitable for radio transmission via an antenna or a cable. The output of transmitter 30 is coupled to an antenna transmitter / receiver 40 via a switch 41. Thus, the RF signal produced by the transmitter is transmitted on the radio channel associated with the transmitter.

The receive chain 200 comprises a radio frequency receiver 50 which is coupled to the antenna 40 via the switch 41, for receiving an RF signal. The receiver 50 transposes the RF area to the baseband (downconversion). The receive chain 200 also includes a data demodulator M-ary 60, coupled to the receiver 50. The data demodulator 60 provides the baseband demodulation of received signal data that is to say the inverse of that provided by the modulator 20. Finally, the receive chain 200 includes a data-consuming device 70, such as a speech decoder coupled to the demodulator 60. This device receives as input the binary data output from the demodulator 60 .

The unit 300 is for example a microprocessor or microcontroller that manages the mobile terminal. In particular, it controls the data modulator 20, the data demodulator 60, the transmitter 30 and the switch 41. It also generates signaling data which are supplied to modulator 20 to be transmitted in the logical channels of appropriate signaling. Conversely, the unit 300 receives the data demodulator 60 signaling data sent by the fixed network into logical channels suitable signaling including timing information and operation commands.

The memory 400 is for example a ROM ( "Read Only

Memory "), EPROM (" Electrically Programmable ROM ") or Flash EPROM in which are stored data that are used for operating the mobile terminal. These data include a linearization training sequence to which we will return later.

We will now describe in detail an embodiment of the transmitter 30. In this example, the transmitter 30 includes a radio frequency power amplifier 31, an RF modulator 32 which transposes the baseband to the radiofrequency field ( conversion uplink), a linearization device 33, a learning module 34 associated with the linearization device.

The output of the power amplifier 31 outputs the RF signal to be transmitted. That is why it is coupled to the antenna 40 via the switch 41. The input of the power amplifier 31 receives a radio frequency signal delivered by the output of the RF modulator 32. The input thereof is coupled to the output of the data modulator 20 for receiving the sequence of symbols forming the baseband signal to be transmitted, through the linearization device 33. the latter comprises for example a predistortion device comprising a pallet ( "look-up table" in English) which translates each value of the signal to be transmitted into a predistorted value. Alternatively or additionally, the device 33 may also include amplitude control means the signal output of the transmitter 30.

The learning unit 34 performs learning of the linearization device 33 according to an input signal which reflects the RF signal delivered by the output of the power amplifier 31. To this end, the module 34 receives a part of the RF signal, which is tapped at the output of the power amplifier 31 by a coupler 36. as necessary, the module 34 ensures the return baseband RF signal thus taken. Although being shown completely within the transmitter 30, the module 34 may, at least in part, be implemented by means belonging to the control unit 300, including software means.

Finally, the automatic gain control device 500 allows the controller 300 to dynamically vary the gain of the variable gain amplifier 59 of the RF receiver 50, according to information which is received from the base station with which the terminal communicates, by a method known in itself. Under this method, the base station transmits at determined instants a specific sequence, called AGC sequence. This sequence is known and recognizable by the mobile terminal. It allows him to measure the signal strength received from the base station and to derive a gain control amplifier 59. This method is implemented in the mobile terminal by the device 500 under the control of the unit 300.

The principle of such a method will be described below with reference to the diagram of Figure 4. We will now describe the operation of the mobile terminal during a learning phase, by the device 34, the linearization device 33. Although this is not mentioned in each case in the following, it is understood that the terms "learning phase" and the terms "training sequence" refer to learning the linearization device 33 performed by the device learning 34 in the control unit 300.

The learning process of the device 33 comprises a step of generating a training sequence comprising a determined number N of symbols, where N is an integer. This step is performed by the data modulator 20 under control of the control unit 300. For this purpose, the unit 300 reads a corresponding bit sequence in memory 400.

Next, still under the control of unit 300, the training sequence is transmitted by the transmitter 30 in at least some bursts transmitted by the latter, depending on the system frame structure.

The learning device 34 then obtains the training sequence transmitted and compares it with the training sequence generated, and accordingly performs actions such as adaptations or other predistortion coefficients of the linearization device 33, according to an algorithm determined to learn. This algorithm can be adaptive. workout talking to denote these operations.

It may be noted that for any modulation, it is possible to find a length N signal sequence determined whose characteristics meet the constraints in terms of spectral width, depth of amplitude modulation, and / or others. In one example, N is equal to 10. The AGC sequence comprises at least N symbols. It can therefore have a length greater than that of the training sequence, when it comprises more than N symbols. In this case, the symbols of the training sequence are preferably the symbols of the AGC sequence which are transmitted first. In this way, the convergence of the learning algorithm and thus the linearization of the RF amplifier are obtained faster.

Learning phases can be performed periodically or otherwise. Other constraints may need to be taken into account after the initial learning phase, when just be corrected drift of the transmitter. The training sequence can evolve both in content and length. The number N is not necessarily fixed in a transmission of the training sequence to another. If an increase in the size of the sequence poses problems (for example if the frame structure is very flexible), one can fix the N in the sequence and size just change its content according to the evolution constraints on system .

The diagram of Figure 2 illustrates a first example of burst, which does not include AGC sequence. In this example, the burst has a duration equal to 20 ms. It firstly comprises a mounting ramp 51 ( "ramping- up" in English) of 625 microseconds, including five padding symbols, to ensure the rise. By padding symbols, is meant that the binary data transmitted in this ramp-up are stuffing bits that is to say, for example, a sequence of 0. It then comprises a timing data sequence 52, the duration is equal to approximately 5 ms. Then, it includes a payload data sequence 53. The payload data can be encoding of voice and generally traffic data or signaling data according to the burst is transmitted on a logical traffic channel or channel logic signal, respectively. Finally, it includes a ramp down 54, again having five padding symbols for the descent into power. Optionally, a further call time is scheduled after the issuance of a burst, to ensure the return to the reception of the issuer. In addition, in any frame structure is provided to issue isolated bursts, particularly each change of logical channel (occurring in particular in each turn, that is to say the passage of a receiving phase to a the terminal emission), each RF frequency change (when frequency hopping functionality is implemented by the system), at each change in transmit power level, or in other special cases too numerous to detail here.

Figure 3 shows an example of such an isolated frame comprising, before the synchronization sequence 52, an AGC sequence referenced 55. This block 55 is issued to allow the dynamic control, by the fixed network, of the transmission power of the transmitter (see above). In this example, the sequence 52 and the sequence 55 only last between 1 and 3 ms each. Other parts of the burst are unchanged from the burst of Figure 2. The user data sequence 53 may, however, be shorter than in the case of a normal burst according to Figure 2.

Advantageously, a portion of these bursts isolated to allow the learning device 34 of the transmitter 32 of radio frequency run a learning algorithm the linearization device 33. In the example of Figure 3, the sequence of linearization is well understood in the aforementioned AGC sequence.

It is thus possible to use the time necessary for transmission of the training sequence for other purposes such as the adjustment of the AGC in reception, according to the method which has been mentioned above with reference schema Figure 1.

The AGC sequence, and therefore the training sequence, are preferably issued at the beginning of frame, then during a change of logical channel when an RF frequency change and / or during a step change power and / or in other cases too numerous to detail here. This is why it is particularly advantageous to combine these sequences (the training sequence being comprised in the sequence of CAG). According to another advantage, the AGC sequence is closest to the ramp rise of the signal, for example, just after the ramp. In this way, learning the linearization device can be achieved faster and thus disrupt the shortest possible transmission.

It is preferable that the length of the training sequence is such that it does not occupy too large a portion of the burst in order to keep a maximum of symbols for distributing useful information. This duration obviously depends on the desired accuracy for the learning algorithm but a compromise between accuracy and time is often necessary to maintain a maximum of useful information in the burst. A reasonable compromise is reached when about 5% of the total duration of the burst. In the case of a 20 ms burst transmitted at a bit rate of 8 ksymbols / s, the duration of a training sequence of N = 10 symbols is thus equal to 1, 25 ms or 6.25% of the Total duration of the frame.

The diagram of Figure 4 illustrates the implementation of an AGC method (known per se) between the first unit 5 and a second device 5 'of a radiocommunication system.

The equipment 5 is here a mobile terminal for example as described above with reference to Figure 1. It comprises the RE 30 and the RF transceiver 50, the latter comprising the variable gain amplifier 59. The equipment 5 'is here a base station with which the mobile terminal communicates 5, which includes an RF transmitter 30' and an RF receiver 50 'having a variable gain amplifier 59'. Functionally, the components 30 ', 50' and 59 'of the base station 5 are identical or comparable with respective components 30, 50 and 59 of the mobile terminal 5'. These components are not described again here.

An AGC sequence transmitted by the mobile terminal 5 allows the dynamic control of the gain of the amplifier 59 of the receiver 50 'of the base station 5'. Conversely, an AGC sequence transmitted by the base station 5 'allows the dynamic control of the gain of the amplifier

59 of the receiver 50 of the mobile terminal.