WO2012119400A1

WO2012119400A1 - Signal sequence processing method and base station

Info

Publication number: WO2012119400A1
Application number: PCT/CN2011/078621
Authority: WO
Inventors: 郭江; 唐海正; 李应希
Original assignee: 华为技术有限公司
Priority date: 2011-08-19
Filing date: 2011-08-19
Publication date: 2012-09-13
Also published as: CN102763389A; CN102763389B

Abstract

Disclosed are a signal sequence processing method and a base station. The method includes: obtaining a signal sequence; forming a training signal with a set signal attribute in the obtained signal sequence; pre-distorting the signal sequence formed with the training signal using a pre-distortion model; amplifying the pre-distorted output signals using a power amplification module and then outputting the same; collecting the amplified training signal in the signal sequence amplified and outputted by the power amplification module; and modifying the pre-distortion model according to the collected training signal. In the present invention, by forming a training signal meeting the set requirements, the signal attribute of the training signal can be controlled actively; and the above solution is adapted to the feedback requirements on the modification of the pre-distortion model, so that the nonlinear features of the training signal after power amplification can be collected timely and effectively to modify the pre-distortion model, optimizing the compatibility of the pre-distortion model and the power amplification module.

Description

Signal sequence processing method and base station

The embodiments of the present invention relate to communication technologies, and in particular, to a signal sequence processing method and a base station. Background technique

The power amplifier module used in wireless communication usually implements the required power amplifier model in the form of hardware combined with software. However, the power amplifier module used in practice usually has a nonlinear amplification problem for the signal sequence, and the prior art generally uses the power amplifier pre- The distortion algorithm is used to correct the nonlinearity of the power amplifier module. According to the implementation of the predistortion model, it can be divided into analog and digital. Currently, digital predistortion algorithm

(Dig i ta l Pre-Di s tor t ion, abbreviated as DPD) is widely used.

In practical applications, the nonlinear characteristics of the power amplifier module are not always constant, but vary with temperature, frequency, time, etc., so the power amplifier pre-distortion algorithm must constantly track the characteristics of the power amplifier module, so as to ensure The output is not distorted.

The prior art adopts a typical pre-feedback scheme to correct the predistortion model. During the transmission of the service signal, the output signal of the power amplifier module is continuously detected, and then the nonlinear characteristic between the output and the input of the power amplifier module is identified, according to the nonlinear characteristic. Correct the predistortion model.

However, the inventors found in the research process of the present invention that the prior art has the following drawbacks: Many of the current wireless communication services are bursty, which causes the power amplifier model implemented by the power amplifier module to also be abrupt. Due to the abrupt nature of the traffic signal and power amplifier model, it is difficult to effectively correct the predistortion model to match the power amplifier module. Summary of the invention

The invention provides a signal sequence processing method and a base station, which are used for effectively correcting a predistortion model and improving the matching between the predistortion model and the power amplifier module.

An embodiment of the present invention provides a signal sequence processing method, including: Acquire a signal sequence;

Forming a training signal having a set signal property in the acquired signal sequence;

The signal sequence formed with the training signal is pre-distorted by a predistortion model; the output signal subjected to the predistortion processing is amplified and processed by the power amplifier module; and the amplified signal is acquired in a signal sequence amplified by the power amplifier module. Training signal

The predistortion model is corrected based on the collected training signals.

The embodiment of the present invention further provides a base station, including a signal sequence processing device, where the signal sequence processing device includes:

a signal acquisition module, configured to acquire a signal sequence;

a signal forming module, configured to form a training signal having a set signal attribute in the acquired signal sequence;

a predistortion module, configured to perform predistortion processing on the signal sequence formed with the training signal by using a predistortion model;

a power amplifier module, configured to amplify and process the output signal after pre-distortion processing by the power amplifier module;

The model correction module is configured to collect the amplified training signal in a signal sequence amplified by the power amplifier module, and correct the predistortion model according to the collected training signal.

The signal sequence processing method and the base station provided by the embodiments of the present invention enable the signal properties of the training signal to be actively controlled by forming a training signal that satisfies the setting requirements. By actively setting the training signal instead of passively using the traffic signal of any signal property, the accurate characteristics of the power amplifier module can be obtained. The above scheme adapts to the feedback requirement for the correction of the predistortion model, so that the training signal can be acquired in time and effectively through the nonlinear characteristics of the power amplifier to correct the predistortion model, thereby optimizing the matching between the predistortion model and the power amplifier module.

DRAWINGS

1 is a flowchart of a signal sequence processing method according to Embodiment 1 of the present invention; 2A is a schematic structural diagram of a power amplifier predistortion system according to an embodiment of the present invention;

2B is a non-linear characteristic curve of a power amplifier according to an embodiment of the present invention;

2C is a predistortion characteristic curve applied to an embodiment of the present invention;

2D is a characteristic curve of an output signal to which the embodiment of the present invention is applied;

3 is a schematic structural diagram of a frame of a T2 frame used in Embodiment 2 of the present invention;

4 is a schematic structural diagram of a signal sequence processing apparatus in a base station according to Embodiment 4 of the present invention; and FIG. 5 is a schematic structural diagram of a signal sequence processing apparatus in a base station according to Embodiment 5 of the present invention. detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is a partial embodiment of the invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

It is to be noted that the power amplifier used in the wireless communication may be a power amplifier that supports multiple carriers, or a power amplifier that supports a single carrier. The technical solutions of the embodiments of the present invention are applicable.

In fact, the nonlinear problem of the power amplifier module is caused by the fact that after the carrier signal is amplified by the power amplifier module, its output power is generally not linearly amplified with the input power, but is rendered nonlinear, so the signal is directly passed through the power amplifier module. Then there will be distortion. The predistortion algorithm pre-identifies the nonlinear features of the power amplifier module, inverts the nonlinear features, and establishes a predistortion model. Specifically, the predistortion model can be implemented by an analog device or a digital system, so that the input signal is first inverted by the predistortion model, and then the predistorted signal can be linearly output after passing through the actual power amplifier module.

To solve the deficiencies of the prior art, the embodiments of the present invention provide various implementations, which are specifically described as follows.

Embodiment 1 FIG. 1 is a flowchart of a signal sequence processing method according to Embodiment 1 of the present invention. The method is applicable to the power amplifier predistortion system shown in FIG. 2A. As shown in FIG. 2A, the predistortion system generally includes a predistortion module 210, a power amplifier module 220, and a model correction module 230. The predistortion module 210 has a predistortion model, and the power amplifier module 220 implements the power amplifier model in the form of hardware and/or software. The nonlinear characteristics of the power amplifier module may be as shown in FIG. 2B, and the signal characteristics after pre-distortion are as shown in FIG. 2C. After the pre-distortion processing and the power amplifier, the theoretical undistorted signal shown in FIG. 2D is formed. The predistortion system shown in FIG. 2A is a logical structure relationship, and when implemented by hardware, it is not limited to the independent module structure as shown in FIG. 2A. For example, the predistortion module 210 and the power amplifier module 220 may both be carried in a power amplifier device of the base transceiver station, and the predistortion module 210 may be implemented by software carried in the power amplifier hardware device. The model modification module 230 can be implemented by software, carried in a digital signal processor of the base transceiver station, or can be carried in other hardware entities capable of interacting with the predistortion module 210 and the power amplifier module 220, as long as the implementation of the present invention can be implemented. The required function can be used.

After the signal sequence is inverted by the predistortion model, it is amplified by the power amplifier module to obtain a theoretically distortion-free amplified output signal. The model correction module 230 is configured to acquire an output signal from the output of the power amplifier module 220, and then re-correct the predistortion model in the predistortion module 210 by comparing the input signal sequences.

The processing method of this embodiment can be performed by a model correction module. The model correction module can be implemented by using software and/or hardware. The processing method of this embodiment includes the following steps:

Step 110: Acquire a signal sequence.

The step may be specifically performed by a digital signal processor of the base transceiver station to obtain a signal sequence to be amplified and transmitted from the pre-sequence device;

Step 120: Form a training signal with a set signal attribute in the acquired signal sequence; this step may be specifically performed by a digital signal processor or a model correction module carried in other hardware. Setting the signal attribute means presetting the signal or some signal attribute of the training signal. For example, setting signal properties includes, but is not limited to, setting transmit power, setting frequency points, and/or setting bandwidth (one or a combination of the above signal attributes). The so-called training signal has a set transmission power, which means that the training signal is Set the transmit power transmission. The training signal having a set frequency point means that the training signal is transmitted at a set frequency point of one or more carriers. For the existing multi-carrier technology, multi-frequency point variation of each carrier can be realized by frequency hopping technology. The training signal having the set bandwidth means that when the training signal is transmitted by multiple carriers, the difference between the two carrier frequencies satisfies the requirement of the set bandwidth.

Step 130: Perform a predistortion process on the signal sequence formed with the training signal by using a predistortion model;

This step can be performed by a predistortion module of the base transceiver station, which processes the signal sequence through a predistortion model.

Step 140: The pre-distorted output signal is amplified by the power amplifier module, and then output;

This step can be performed by the power amplifier module of the base transceiver station, and the output signal sequence can be transmitted through the antenna.

Step 150: Acquire an amplified training signal in a signal sequence amplified by the power amplifier module. This step may be performed by a model correction module in the digital signal processor, and may specifically be between the digital signal processor and the power amplifier module. The communication channel is simulated to read the training signal in the signal sequence output after the power amplifier.

Step 160: Correct the predistortion model according to the collected training signal.

This step can be performed by a model modification module in the digital signal processor, and the manner of correction is not limited. For example, the nonlinear uncharacteristic signal sequence and the post-amplifier signal sequence can be used to obtain the nonlinear characteristics of the power amplifier module, and then based on This nonlinear feature is used to modify the predistortion model.

In the technical solution of the embodiment, by forming a training signal that satisfies the setting requirement, signal attributes such as power, frequency, bandwidth, and the like and training timing can be actively controlled. By actively setting the signal properties of the training signal, rather than passively employing traffic signals of any power, frequency or bandwidth, it is possible to obtain accurate features of the power amplifier module. The above solution adapts to the feedback requirement of the predistortion model correction, so that the training signal can be timely and effectively collected through the non-linear characteristics of the power amplifier to correct the predistortion model, thereby optimizing the matching between the predistortion model and the power amplifier module. In the embodiment of the present invention, there are many alternative implementations for setting signal attributes, as described below: For example, the transmit power is preferably equal to the maximum transmit power of the transmitter using the power amplifier module. When the training signal is transmitted with the maximum transmission power, the power amplifier module to which the maximum transmission power is applied can be corrected, so that the power amplifier signal is not distorted, and when the transmitter adopts a smaller transmission power, the signal is not distorted.

For another example, the bandwidth is set to be the bandwidth between the frequency points of the at least two carriers transmitting the training signal, and the set bandwidth is equal to the maximum bandwidth between the carrier frequency points used by the power amplifier module of the transmitter, that is, each configured on the power amplifier module. The maximum bandwidth that can be achieved between all frequency points of the carrier. This setting also enables the modified predistortion model to meet the requirements of the maximum bandwidth, covering the effective frequency of each carrier of the entire transmitter. In addition, the set frequency can be set according to different situations, or can be set by setting the frequency. The bandwidth requirement. When the transmitter has multiple carriers, and each carrier can choose to use multiple frequency points, the frequency of each carrier can be set to meet the set bandwidth requirements. The setting of the maximum bandwidth does not limit the difference between the frequency points of the specific two carriers as the maximum bandwidth, and the frequency of setting the different carriers may be replaced, as long as the bandwidth between the frequencies of the at least two carriers is equal to the carrier frequency used by the power amplifier module. The maximum bandwidth between points can be.

The technical solution of the embodiment of the present invention is particularly applicable to the current situation of a business burst. For example, the situation of a business burst, a cell along the railway usually does not have a mobile call, but when the train passes, the cell will suddenly switch to the multi-call service. When there is usually no mobile call service, the base station in this cell cannot use the service signal as the training signal to correct the predistortion model, or even if there are a small number of service signals, the transmission frequency, power, time and other factors of the service signal are also determined by the service signal. It is determined by itself and is inconsistent with the situation when the train passes. Therefore, the predistortion model modified based on such training signals is not adapted to the power amplifier module of the service signal when the train passes. Such a predistortion algorithm can only cause tracking to be untimely if only the existing output is tracked to establish a predistortion model. The superposition of bursty traffic signals onto an inaccurate predistortion model can result in poor signal quality and even spurs. The technical solution of the embodiment of the present invention can effectively solve the problem that the correction caused by the business sudden phenomenon is not timely question.

Embodiment 2

Embodiment 2 of the present invention provides a signal sequence processing method. In practical applications, there are various ways to form a training signal having a set signal attribute. This embodiment is one of the preferred modes, that is, the operation of forming a training signal having a set signal attribute in the obtained signal sequence is specifically:

A training signal is added to the idle time slot of the acquired signal sequence, and the training signal is set to have a set transmit power, a set frequency point, and/or a set bandwidth.

The technical solution of the embodiment is easier to implement. According to the communication protocol, some idle time slots are often reserved in the logic signal, and the positions of these idle time slots are fixed and predictable according to the protocol. Adding training signals to these idle time slots does not affect the transmission of existing traffic signals, and the power, frequency, and bandwidth of the training signals can be set as needed. In which idle time slot the training signal is transmitted, the base station implementation can be initiated by command setting, or the base station can automatically and silently transmit the training signal in the idle time slot.

There are many options for idle time slots, which can be applied to global mobile communication systems in multi-carrier technology.

GSM T2 frame includes multiple logical channels, such as Traffic Channel (referred to as TCH) and Packet Data CHannel (Packet Data CHannel, referred to as GSM) service neighborhood. PDCH), the 25th time slot of the T2 frame is an idle time slot (idle), as shown in FIG. The time slot range of T2 (5 b i t s) is 0 to 25, which is equal to "FN mod 26", where FN is the number of frames of TDMA. This idle time slot can be utilized to actively transmit the training signal. Because it is a free time slot, the user can combine the most suitable training signals including "power, frequency" by active setting. For example, a burst signal that transmits the maximum power on the carrier at idle time can be selected as the training signal while selecting the maximum transmission bandwidth. Because it is a free time slot, sending a training signal different from the service signal at this time does not affect the service. The training signal can be sent periodically, so that the digital pre-distortion algorithm always obtains the real-time characteristics of the power amplifier, thereby improving the performance of the digital power amplifier pre-distortion algorithm and achieving the optimal effect of establishing the power amplifier module.

Those skilled in the art can understand that the idle time slot is not limited to the T2 frame in GSM, and can also be Suitable for Wideband Code Division Multiple Access (WCDMA), Code Division Multiple Access (CDMA), Time Division Synchronous Code (Multi-Carrier) Idle time slots specified in protocols such as Division Multiple Access (TDSCDMA) and Long Term Evolution (LTE).

Since the idle time slot is determined according to the protocol, according to the formation position of the training signal in the signal sequence, the operation of acquiring the amplified training signal in the signal sequence amplified by the power amplifier module may specifically be: The position in the signal sequence is used to acquire the amplified training signal at the idle time slot position in the signal sequence amplified by the power amplifier module. The identification of the idle time slot can be identified based on the free time slot position specified by the predetermined protocol being executed.

Embodiment 3

Embodiment 3 of the present invention provides a signal sequence processing method, which differs from Embodiment 2 in that a method of forming a training signal is different. That is, the operation of forming the training signal having the set signal attribute in the acquired signal sequence is specifically: detecting, in the acquired signal sequence, a service signal having a set transmit power, a set frequency point, and/or a set bandwidth, as a The training signal.

In this embodiment, by detecting a service signal, a service signal that meets the set requirement is actively selected as a training signal. Therefore, the power, bandwidth, frequency point and other factors of the training signal can satisfy the requirements of detection and correction, and can not affect existing service signals.

Preferably, the operation of forming the training signal with the set signal attribute in the acquired signal sequence is specifically: detecting a service signal having a set frequency point and a set bandwidth in the acquired signal sequence, and detecting the detected service The transmit power of the signal is adjusted to set the transmit power to form a training signal.

The above solution may not affect the frequency and bandwidth of the existing service signal, because the frequency and bandwidth are usually fixed parameter values of the service signal, and should not be adjusted. The transmit power of the traffic signal can be forcibly adjusted, and is preferably adjusted to the maximum transmit power, which does not affect the transmission of the traffic signal.

On the basis of the technical solutions of the foregoing embodiments, after the training signal having the set signal attribute is formed in the acquired signal sequence, the following operations may also be performed: And generating a location identifier of the training signal, used to identify a location of the training signal in the signal sequence, to indicate that the amplified training signal is collected according to the location identifier of the training signal.

The above manner of indicating the location of the training signal acquisition is applicable to the manner in which the training signal is transmitted in the idle time slot and the service signal is selected as the training signal. When the training signal is selected, the component that performs the acquisition operation is notified, and the training signal is collected at the corresponding position.

In the specific application, the two methods of transmitting the training signal by using the idle time slot and selecting the service signal as the training signal may be used separately or in combination. The training signal may be formed periodically in the signal sequence, or may be combined in accordance with the set rules. For example, if the selected service signal is preferentially executed as the training signal, and there is no suitable service signal within the set time length, the training signal is actively generated and generated.

Embodiment 4

4 is a schematic structural diagram of a signal sequence processing apparatus in a base station according to Embodiment 4 of the present invention. The signal sequence processing apparatus includes a signal acquisition module 410, a signal forming module 420, a predistortion module 430, a power amplifier module 440, and a model correction module 450.

The signal acquisition module 410 is configured to acquire a signal sequence, and the signal forming module is configured to form a training signal having a set signal attribute in the acquired signal sequence, where the predistortion module 420 is configured to form the training signal. The signal sequence is pre-distorted by the pre-distortion model; the power amplifier module 430 is configured to output the pre-distorted output signal through the power amplifier module for amplification processing; and the model correction module 440 is configured to be used in the signal sequence amplified by the power amplifier module 430. The amplified training signal is acquired, and the predistortion model is corrected according to the collected training signal.

The technical solution of this embodiment can perform the signal sequence processing method provided by the embodiment of the present invention, and has a corresponding functional module.

In the technical solution of the embodiment, by forming a training signal that satisfies the requirements of the set signal attribute, the signal properties such as the power, frequency, bandwidth, and the formation timing of the training signal can be actively controlled.

Preferably, the set transmit power is equal to the maximum transmit power of the transmitter using the power amplifier module. The set bandwidth is a bandwidth between frequency points of at least two carriers transmitting the training signal, the setting The fixed bandwidth is equal to the maximum bandwidth between the carrier frequencies used by the power amplifier module.

The above solution can adapt to the feedback requirement of the predistortion model correction, so that the training signal can be timely and effectively collected through the non-linear characteristics of the power amplifier to correct the predistortion model, thereby optimizing the matching between the predistortion model and the power amplifier module.

Embodiment 5

FIG. 5 is a schematic structural diagram of a signal sequence processing apparatus in a base station according to Embodiment 5 of the present invention. The present embodiment may be based on the foregoing embodiment. Preferably, the signal forming module 420 includes: a signal adding unit 421, a signal detecting unit 422, and/or Or power adjustment unit 423 (ie one of the above units or a combination thereof).

The signal adding unit 421 is configured to add a training signal to the idle time slot of the acquired signal sequence, and set the training signal to have a set transmit power, a set frequency point, and/or a set bandwidth. The signal detecting unit 422 is configured to: Detecting, as the training signal, a service signal having a set transmit power, a set frequency point, and/or a set bandwidth in the acquired signal sequence; the power adjustment unit 423 is configured to detect the set frequency in the acquired signal sequence Point and set the bandwidth of the service signal, and adjust the transmit power of the detected service signal to set the transmit power to form a training signal.

The manner in which the training signal is transmitted using the idle time slot and the appropriate service signal is selected as the training signal can be implemented independently or in combination.

For the case of transmitting the training signal using the idle time slot, the model correction module preferably includes: a position recognition unit, a signal acquisition unit, and a model correction unit. The location identifying unit is configured to identify a location of the idle time slot in the signal sequence; the signal acquisition unit is configured to collect the amplified training signal in a free time slot position in the signal sequence amplified by the power amplifier module; The predistortion model is corrected based on the collected training signals.

In addition, the signal sequence processing device may further include a location indication module 460, configured to generate a location identifier of the training signal when the signal forming module 420 forms the training signal, to identify the training signal in the signal sequence. A position is formed to instruct the model correction module 450 to acquire the amplified training signal according to the position identifier of the training signal. The signal sequence processing device in the base station provided by the embodiment of the present invention may be integrated in the transceiver of the base station, for example, may be carried by a power device such as a power amplifier device and a digital signal processor in the transceiver, and implemented by using software.

The technical solution of the embodiment of the present invention is particularly applicable to a service burst situation of wireless communication, and the pre-distortion model established by tracking the negative feedback according to the service may cause a mismatch with the power amplifier module used in the burst service, thereby causing a sudden The technical solution of the embodiment of the present invention can effectively solve the problem.

Embodiments of the present invention provide a scheme for actively transmitting a training signal, training a digital pre-distortion algorithm, and the training signal does not affect an existing service. The signal properties of the training signal may mainly include power information, frequency information, and the like. That is, on a transmitter using multiple carriers, the training signal is transmitted at an appropriate timing, and by sampling the excitation of the training signal, the optimal model and the predistortion model of the power amplifier module can be obtained. The choice of the appropriate timing makes it unnecessary to transmit the training signal to the existing service.

The description of the method embodiments and device embodiments provided by the present invention can be referred to each other by reference. A person skilled in the art can understand that all or part of the steps of implementing the above method embodiments may be completed by using hardware related to program instructions, and the foregoing program may be stored in a computer readable storage medium, and the program is executed when executed. The foregoing steps include the steps of the foregoing method embodiments; and the foregoing storage medium includes: a medium that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.

It should be noted that the above embodiments are only for explaining the technical solutions of the present invention, and are not intended to be limiting; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: The technical solutions described in the foregoing embodiments are modified, or some of the technical features are equivalently replaced. The modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

Claim

A signal sequence processing method, comprising:

Acquire a signal sequence;

The predistortion model is corrected based on the collected training signals.

The signal sequence processing method according to claim 1, wherein the forming the training signal having the set signal attribute in the acquired signal sequence comprises:

The signal sequence processing method according to claim 2, wherein the acquiring the amplified training signal in the signal sequence amplified by the power amplifier module comprises:

The position of the idle time slot in the signal sequence is identified, and the amplified training signal is acquired at the idle time slot position in the signal sequence amplified by the power amplifier module.

Detecting, as the training signal, a service signal having a set transmit power, a set frequency point, and/or a set bandwidth in the acquired signal sequence; or

A service signal having a set frequency point and a set bandwidth is detected in the acquired signal sequence, and a transmission power of the detected service signal is adjusted to a set transmission power to form a training signal.

The signal sequence processing method according to claim 1 or 2 or 4, wherein after the training signal having the set signal attribute is formed in the acquired signal sequence, the method further includes: generating the training signal And a location identifier, configured to identify a location of the training signal in the signal sequence, to indicate that the amplified training signal is acquired according to the location identifier of the training signal.

The signal sequence processing method according to any one of claims 2 to 4, wherein: the set transmission power is equal to a maximum transmission power of a transmitter using the power amplifier module; a bandwidth between frequency points of at least two carriers of the training signal, the value of the set bandwidth being equal to a maximum bandwidth between carrier frequency points used by the power amplifier module.

A base station, comprising: a signal sequence processing device, wherein the signal sequence processing device comprises:

a signal acquisition module, configured to acquire a signal sequence;

The base station according to claim 7, wherein the signal forming module comprises one or a combination of the following units:

a signal adding unit, configured to add a training signal in an idle time slot of the acquired signal sequence, and set the training signal to have a set transmit power, a set frequency point, and/or a set bandwidth;

a signal detecting unit, configured to detect, in the acquired signal sequence, a service signal having a set transmit power, a set frequency point, and/or a set bandwidth, as the training signal;

And a power adjustment unit, configured to detect a service signal having a set frequency point and a set bandwidth in the acquired signal sequence, and adjust a transmit power of the detected service signal to a set transmit power to form a training signal.

The base station according to claim 8, wherein the model correction module comprises: a location identifying unit, configured to identify a location of the idle time slot in the signal sequence; a signal acquisition unit, configured to acquire an amplified training signal at a free time slot position in a signal sequence amplified by the power amplifier module;

The model correction unit is configured to correct the predistortion model according to the collected training signal.

The base station according to claim 7 or 8, wherein the signal sequence processing device further comprises:

a location indication module, configured to: when the signal forming module forms the training signal, generate a location identifier of the training signal, to identify a location of the training signal in a signal sequence, to indicate that the model correction module is The position of the training signal identifies the amplified training signal.

11. A base station according to any of claims 7-9, characterized in that:

The set transmit power is equal to a maximum transmit power of a transmitter using the power amplifier module; the set bandwidth is a bandwidth between frequency points of at least two carriers transmitting the training signal, the set bandwidth The value is equal to the maximum bandwidth between the carrier frequencies used by the power amplifier module.