CN1210967C - Linear power amplification method and power amplifier unit based on feedforward technique - Google Patents

Abstract

The invention discloses a method and device for realizing a linear power amplifier based on feedforward technology. The method adopts digital filtering technology to directly eliminate the carrier frequency component in the output sampling signal of the main power amplifier, obtains an error signal representing the nonlinear distortion product of the main power amplifier, and uses signal cancellation technology to eliminate the nonlinearity in the output signal of the main power amplifier at the system output Distortion products, to achieve the purpose of linearization. The invention makes the design of the distortion elimination circuit-error loop not limited by the gain of the main power amplifier and can be completed by itself, the difficulty of circuit design is greatly reduced, the system structure is simplified, the degree of freedom of development is increased, and the risk and cost are reduced.

Description

A linear power amplification method and power amplifier device based on feedforward technology

Technical field:

The invention belongs to the field of mobile communication, in particular to a linear power amplification method and power amplifier device based on feedforward technology.

Background technique:

With the continuous development of modern mobile communication technology, the linearity requirements for RF power amplifiers have increased sharply. Independent power amplifiers based on power combining technology and high power amplifiers based on power fallback technology have completely failed to meet the performance of mobile communication base station RF subsystems. The comprehensive demand of power consumption, cost and other factors. Because the DC conversion efficiency of Class A and B power amplifiers is much higher than that of pure Class A power amplifier tubes, and the linearity index is still relatively good, so the industry mainly uses Class A and B power amplifiers based on LDMOS (Lateral Expansion Metal Oxide Semiconductor) technology to assist in linearization. To realize high-power linear power amplifiers, the widely used linearization methods include predistortion linear method and feedforward linear method.

The pre-distortion linearization method is mainly to insert a pre-distortion component before the power amplifier, which causes the input signal of the power amplifier to be pre-distorted and changed, and its distortion characteristics are just opposite to the nonlinear performance of the power amplifier device, so as to achieve an elimination at the output of the system. Except that the amplitude increases a lot, other characteristics are basically the same as the linear power amplification effect of the input signal. In order to ensure a certain linearization bandwidth, an external monitoring circuit is required, combined with pre-distortion components, to simulate the nonlinear characteristics of AM-AM (amplitude modulation-amplitude modulation) and AM-PM (amplitude modulation-phase modulation) of the power amplifier. However, it is very difficult to accurately simulate the AM-AM and AM-PM characteristics of power amplifier tubes under various conditions such as ambient temperature, feed voltage, operating frequency, and power level. Therefore, the realized linearization index and adaptive work Bandwidth is less than ideal. Therefore, this type of linear power amplifier is still in the laboratory and small batch production stage, or as an auxiliary means of realization of the feedforward linear power amplifier.

The feed-forward linearization method is to separate the distorted signal generated by the power amplifier, then adjust the gain, phase and delay matching, and add it to the output of the power amplifier to eliminate the distortion component in the output signal of the power amplifier to obtain the maximum Linearization index. This method is less affected by factors such as ambient temperature and operating voltage, has a higher linearization index, and has a wider operating bandwidth. And because the technology is mature, the application range is very wide.

Current feed-forward linear power amplifiers all include at least two signal cancellation loops: a main loop and an error loop. The main loop cancels and eliminates the carrier frequency component in the output sampling signal of the main power amplifier, and obtains an error signal containing only the nonlinear distortion products of the main power amplifier; the error loop amplifies and adjusts the error signal to be equal to the amplitude of the nonlinear distortion signal in the output signal of the main power amplifier level, and then reverse phase cancellation to eliminate the distortion component in the output signal of the main power amplifier. Although the main loop and the error loop work independently, because they intersect with each other, the design status of the main loop directly affects the design result of the error loop, especially the gain setting of the main power amplifier directly determines the realization of the error amplifier. A main power amplifier with general gain requirements will This results in a very high error amplifier gain requirement, making it difficult to implement the error amplifier in a single block. If multiple amplifying circuit modules are cascaded to realize the error amplifier, not only the structure is complex, the system design is more difficult, the hardware cost increases, but also the multi-level circuit cascading causes a large delay, and the delay requirement of the error loop delay line increases accordingly. , the length of the delay line increases, and under the same conditions, the insertion loss increases. Since the delay line is connected in series to the output channel of the power amplifier, the insertion loss of the delay line directly consumes the output power of the system. In order to reduce the insertion loss of the delay line, cables (or delay devices) with better performance (lower insertion loss per unit length) must be selected, which will further increase the difficulty and cost of system implementation. However, since the main power amplifier is the direct implementer of the gain of the linear power amplifier system, it is generally impossible to set its gain too low. At present, no relevant technology patents and literatures that fundamentally solve this problem have been retrieved.

Invention content:

One of the purposes of the present invention is to propose a linear power amplifier based on feedforward technology that eliminates the distortion information of the main power amplifier through open-loop elimination of the working carrier frequency signal in the output sampling signal of the main power amplifier, and then eliminates the distortion signal at the output of the power amplifier. Amplifier device

The second object of the present invention is to propose a method for realizing linear power amplification using the above-mentioned device.

The linear power amplifier device based on feedforward technology provided by the present invention is realized with the following technical solutions:

The device provided by the present invention is composed of four circuit modules: a channel detection module, a main power amplifier module, a carrier elimination module, and a distortion elimination module;

The channel detection module further includes a coupler A, a mixer A, an intermediate frequency filter, a detector, and a monitor;

The carrier elimination module further includes a coupler B, a mixer B, an analog/digital converter, a digital filter, a digital/analog converter, a mixer C, and a local oscillator;

The distortion elimination module further includes a delay line, a coupler C, an electronically adjustable attenuator, an electronically adjustable phase shifter, and an error amplifier;

After the working carrier frequency signal is input to the channel detection module, it is sent to the coupler A, and the through signal enters the main power amplifier module, and the power is amplified by the main power amplifier module; the coupled signal is sent to the monitoring through the down-conversion mixer, intermediate frequency filter and detector The monitor feeds the detected carrier frequency channel information to the digital filter of the carrier cancellation module for setting digital filter parameters; the amplified power signal output by the main power amplifier module enters the carrier cancellation module and passes through the coupler B Sampling, and then through the down-conversion mixer, ADC, digital filter, DAC, and up-conversion mixer to process and eliminate the carrier frequency working signal, and obtain the error signal that only reflects the nonlinear distortion information of the main power amplifier and send it to the distortion elimination module; in the distortion elimination module, the error signal is adjusted for gain and phase through the ESC attenuator and the ESC phase shifter, then passes through the error amplifier, is coupled into the main channel by the coupler C, and is connected with the coupler B in the carrier elimination module After the direct signal is delayed by the delay line of the distortion elimination module, the output signal of the main power amplifier is cancelled, and the nonlinear distortion component is eliminated, and a highly linear high-power working carrier frequency signal output is obtained.

The linear power amplifier implementation method based on feedforward technology provided by the present invention comprises the following steps:

Step 1: Determine the channel number of the system operating carrier frequency;

Step 2: Set the digital filter according to the discrimination result of the channel where the working carrier frequency is located;

Step 3: Establish a carrier cancellation circuit;

Step 4: Establish a distorted signal elimination circuit.

Judging the channel number where the system operating carrier frequency is located in the step 1 further includes the following process:

First, coupler A samples part of the input signal power and enters mixer A, through the down-conversion and mixing of mixer A, the radio frequency signal is reduced to intermediate frequency, and then filtered by an intermediate frequency bandpass filter whose bandwidth is exactly the width of a communication channel , and then get the DC voltage through the detector and send it to the monitor;

Secondly, the monitor controls the output frequency of the frequency synthesizer to scan from low to high or from high to low at the same frequency step as the standard channel interval of the communication system, that is, channel scanning, and correspondingly detects the signal detection voltage corresponding to each working channel ;

Then, with reference to a fixed threshold, the monitor judges whether the signal detection voltage of each channel exceeds the threshold. If the threshold exceeds the threshold, it is considered that the channel has a working carrier frequency signal, otherwise it does not, and all the working channel numbers of the system are judged in real time.

Described step 2 sets the digital filter according to the discriminated result of the channel where the working carrier frequency is located and further refers to: according to the working carrier frequency channel number transmitted by the monitor that the first step obtains, the DSP (Digital Signal Processing) program module entry is set parameters, and a digital filtering algorithm with multi-window notch filter characteristics is established by the corresponding program module, so that the notch "slot" position of the digital filter corresponds to the working carrier frequency channel one by one, so that the filter can be used in the third Step to eliminate the working carrier frequency signal.

Described step 3 sets up the carrier cancellation circuit and further includes the following process:

First, sample the output signal of the main power amplifier; use the coupler B connected in series in the output channel of the main power amplifier to sample part of the output power signal of the main power amplifier, which includes both the system carrier frequency information and the intermodulation introduced by the nonlinear characteristics of the main power amplifier Distortion information, send this signal to mixer B;

Secondly, down-conversion and frequency mixing; the main power amplifier output sampling signal is mixed with the local oscillator signal generated by the local oscillator, and the frequency is down-converted to a lower frequency range;

Again, analog/digital conversion; convert low-frequency analog signals into digital information and send them to digital filters;

Then, digital filtering; by the digital filtering algorithm completed in step 2, the digital information input into the digital filter is filtered out the information corresponding to the carrier signal;

Then, digital/analog conversion; restore the digital signal output by the digital filter to an analog signal;

Finally, up-conversion and frequency mixing; the restored analog signal and the local oscillator signal generated by the local oscillator are mixed and up-converted to restore the low-frequency signal to the radio frequency.

Described step 4 establishes the distorted signal elimination circuit and further includes the following process:

First, delay; use the fixed cable delay line to delay the high-power signal output by the main power amplifier and send it to the system output end;

Secondly, error amplification; the radio frequency signal obtained in the third step and eliminated the carrier frequency signal is also sent to the system output after amplitude, phase adjustment and linear amplification;

Finally, the distortion is eliminated; use the coupler C to realize the subtraction of the high-power signal output by the main power amplifier and the error radio frequency signal that has eliminated the carrier frequency signal, cancel the distortion component in the output of the main power amplifier, and complete the linearization function of distorted signal elimination .

Description of drawings:

The present invention will be described in further detail below in conjunction with the accompanying drawings.

Fig. 1 is a structural diagram of an existing feedforward linear power amplifier device;

Fig. 2 is a method flowchart of the present invention;

Fig. 3 is a structure diagram of the feedforward linear power amplifier device of the present invention.

Detailed ways:

Figure 1 shows the basic framework of an existing feedforward linear power amplifier. The input signal is equally divided and output by the power splitter 1, one signal is sent to the main power amplifier 2 for power amplification, and the other signal is passed through the delay line 2, power amplifier

The attenuator 1 and the electric phase shifter 1 are adjusted and then input into the summing circuit 9; after the signal output by the main power amplifier 2 is sampled by the sampling coupler 3, it enters the other input terminal of the summing circuit 9 and is connected with the delay branch The incoming input signal is cancelled, and the input signal component in the output sampling signal of the main power amplifier is eliminated, and the representative signal of the distortion information of the power amplifier—the error signal is obtained, and sent to into the error amplifier 12; the amplified error signal enters the power amplifier output channel through the directional coupler 5, and performs signal cancellation with the main power amplifier output signal delayed by the delay line 1 to eliminate the distortion product in the main power amplifier output signal, A high-power signal with high linearity is obtained at the output of the power amplifier.

It can be seen from Figure 1 that the feedforward linear power amplifier includes two basic signal cancellation loops—the main loop and the error loop. Each loop completes its own function independently: the main loop eliminates the working signal and extracts distortion information; the error loop Eliminate distortion components in the output of the power amplifier to achieve linearization goals. However, the design of the error loop is directly dependent on the state of the main loop: the gain design of the error amplifier should compensate for the sampling coupler 3, the combiner 9, the directional coupler 5, the ESC attenuator 2, and the ESC phase shifter 2. The loss caused by the error channel, and the coupling degree selection of the sampling coupler 3 should refer to the gain of the main power amplifier 2 and the insertion loss of the ESC attenuator 1 and the ESC phase shifter 1. The main power amplifier 2 is the main source of the system gain. For the implementer, if its gain is set to 50dB, and the sum of the insertion losses of 7 and 8 is 10dB, then only the coupling degree of the sampling coupler C is 40dB (the realization of the 40dB coupler is very complicated), so as to ensure the (9) The amplitudes of the two signals participating in the cancellation are equal (the basis for the cancellation of the two signals is: the amplitude is equal, and the phase difference is 180 degrees). In this way, if the coupling degree of the directional coupler 5 is taken as 20dB (because the sampling coupler 3, the directional coupler 5, etc. are connected in series in the output channel of the power amplifier, the insertion loss directly consumes the output power of the power amplifier, it must be small enough, and the corresponding coupling degree It should preferably not be less than 20dB, that is, the coupling extracts no more than 1% of the energy in the output signal of the main power amplifier), then the error amplifier gain should be greater than 70dB. Generally speaking, it is very difficult to realize an amplifier greater than 70dB in one circuit module. Even if multiple amplifying circuit modules are cascaded to realize, not only the structure is complicated, but also the difficulty and cost of system design are increased, and the multi-level circuit cascading causes delay The time is very large, correspondingly, the delay requirement of delay line 1 (4) increases, and the length of the delay line increases, which not only causes a large delay of the linear power amplifier (because the delay line 1 (4) is connected in series with the main channel of the linear power amplifier) , and when the length of the delay line 1(4) increases, its insertion loss increases accordingly, and its insertion loss will directly "eat" the output power of the main power amplifier, which is obviously unacceptable. In order to reduce the insertion loss of the delay line, a delay cable (or delay device) with better performance must be selected, which will increase the system cost, size and design difficulty. All the above derivation results show the high difficulty of system implementation, the cost of development and production, and the high risk of research and development.

Fig. 2 is a flow chart of the method of the present invention. With the development and maturity of digital signal processing technology, a large number of DSP commercial hardware chips have been introduced to the market, and various application development software has also been widely released to users, and its application scope and depth have expanded rapidly. The present invention is a signal elimination method realized based on the most basic and mature digital filter technology in DSP technology. As shown in the figure, the method described in the present invention can be realized in the following steps:

The first step is to determine the channel number of the system operating carrier frequency. We know that almost all communication systems have their own channel division principles, that is to say, as long as the channel number of the carrier frequency signal is known, its standard spectrum characteristic diagram can be obtained. Therefore, the present invention adopts the simplest and fastest method to determine the signal channel transmitted by the linear power amplifier. Through a series of coupling sampling, frequency mixing down conversion, intermediate frequency filtering, DC detection, frequency band scanning and beyond threshold judgment, the channel numbers of all carrier signals in the system working frequency band are detected and sent to the digital filter circuit in real time. Of course, if the linear power amplifier can communicate with the base transceiver station (BTS), the working channel data can be obtained directly, and all hardware involved in this step can be omitted except the monitor 316;

In the second step, a digital filter is set according to the discrimination result of the working channel. Since the linear power amplifier only provides a transparent transmission channel for the radio frequency signal of the communication system, and the carrier frequency and the number of carrier frequencies transmitted in the channel change with the user's needs, it is necessary to dynamically filter out all carrier frequency signals in the system's working frequency band , it is obviously impossible to rely on a fixed frequency filter, so here we choose a digital filter based on DSP technology to eliminate the working carrier frequency signal in the output sampling signal of the main power amplifier. For a kind of communication system, its channel frequency characteristics are known and determined. According to the channel detection result of the first step, the digital filter "slot" position is set, and these "slot" positions vary with the program module entry parameters ( (from the channel detection result) can be changed instantly, which can realize the function of dynamically tracking the working frequency of the system and establishing a carrier signal elimination filter;

The third step is to build a carrier cancellation circuit. Digital signal processing (DSP) is completely based on digitized binary digital information. Therefore, before digital filtering, the signal to be processed must be sampled, held, and encoded, that is, analog-to-digital conversion (ADC). The sampling rate is less than twice the signal frequency to digitize the analog signal so as not to lose its frequency information, so the signal to be processed should be reduced to the acceptable frequency range of the ADC and DSP hardware, and then restored to the original radio frequency after digital filtering frequency. The specific process of carrier cancellation is: first sample part of the output signal of the main power amplifier, then frequency mixing down conversion, DAC, digital filtering, ADC, frequency mixing up conversion. Since the same local oscillator source signal is used for the two frequency mixing, except for the filtered carrier frequency information, the recovered RF signal is the same as the output sampling signal of the main power amplifier, which completely represents the nonlinear distortion of the main power amplifier;

The fourth step is to realize the elimination of distorted signals. This part of the circuit is basically the same as the general feed-forward signal cancellation circuit, the difference is: the selection of the coupling degree of the main power amplifier output sampling circuit - coupler B332) no longer depends on the main power amplifier gain (because the main loop does not exist anymore) , the circuit design can select the coupling degree of the coupler according to the actual output power of the linear power amplifier. If the coupling degree of the coupler B332 and the coupler C343 is a coupler of 20dB, and the insertion loss of the electric attenuator 346 and the electric phase shifter 347 is 10dB, the gain of the error amplifier 348 only needs 50dB. Obviously, using 2 The ~3-level amplifier circuit can be realized, and it is a single module mode, and the circuit delay is also greatly reduced. Correspondingly, the length of the delay line 342 in the output channel of the main power amplifier is reduced, and the cable performance and cost requirements for the delay line will also be greatly reduced. reduce. The specific implementation process is: one way delays the output signal of the direct main power amplifier and sends it to the coupler C343; the other way sends the error signal 345 obtained by carrier cancellation to the coupler C343 after amplitude phase adjustment and linear amplification; the two-way signal In the coupler C343, the distortion components in the output signal of the main power amplifier are subtracted and eliminated to achieve the linearization goal.

Fig. 3 is a configuration diagram of a feedforward linear power amplifier device provided by the present invention. After the working carrier frequency signal 311 is input into the channel detection module 31 of the linear power amplifier device, it is directly sent to its coupler A312, and the direct signal 321 enters the main power amplifier module 32, and the power is amplified by the main power amplifier 322; the coupled signal passes through the frequency mixing 313, Filtering 314 and wave detection 315 are sent to the monitor 316, and the monitor 316 controls the frequency synthesizer 317 to perform channel scanning and detect and judge the user carrier frequency working channel, and then immediately feed the judgment result or the channel data 318 directly provided by the BTS to the carrier cancellation The digital filter 335 of the module 33 is used for setting filter parameters; the power signal 331 output by the main power amplifier enters the carrier cancellation module 33, is sampled by the coupler B332, and then passes through the down-conversion frequency mixing 333, ADC334, digital filter 335, and DAC336 , up-conversion and mixing 337 and other processes to eliminate the carrier frequency working signal, and obtain the error signal 345 that only reflects the nonlinear distortion information of the main power amplifier and send it to the distortion elimination module 34; in the distortion elimination module 34, the error signal 345 is passed through the electric adjustment The attenuator 346 and the electric phase shifter 347 perform gain and phase adjustment, and then through the error amplifier 348, the coupler C343 couples into the main channel, and the direct signal 341 of the coupler B332 in the carrier elimination module 33 passes through the distortion elimination module 34. The delay line 342 cancels the output signal of the main power amplifier after the delay, and eliminates the nonlinear distortion component therein, so as to obtain a highly linear high-power working carrier frequency signal 344 output.

The invention discloses a method for realizing carrier cancellation based on DSP digital filtering technology and a device for realizing a linear power amplifier using the method. The method is unique and novel, and has not been found in current patents and related documents. The described The structure of the device is simple, the realization cost is low, and the risk of development technology is small.

A linear power amplifier implementation method and device based on feedforward technology. Compared with the previous feedforward signal cancellation carrier cancellation method, the carrier cancellation method has the following characteristics:

First, the present invention completely removes the main ring hardware circuit. The present invention uses a digital filter to directly filter out the carrier frequency signal in the sampling signal output by the main power amplifier, and obtains the distortion error information caused by the nonlinearity of the main power amplifier, thereby eliminating a pair of feedforward signals that require closed-loop adaptive control to achieve a certain performance. The elimination of loops simplifies the system composition from a technical point of view and reduces the difficulty of implementation. In addition, by using the carrier frequency detection and the adaptive filtering function of the digital filter, the method provided by the present invention can adjust the carrier frequency elimination effect (reaching, exceeding or lower than the signal cancellation type carrier frequency elimination method) and the working bandwidth according to user needs;

Second, the present invention reduces the difficulty of implementing the error loop. Since the main loop no longer exists, the error loop is completely independent, and the circuit design can be completed by itself according to the needs of the system. The specific advantages have been analyzed before, so I won’t repeat them here;

Third, the production process is standardized. The present invention simplifies the system configuration, and carrier frequency detection and carrier elimination can be realized by standard signal processing circuits, without the need for large-length, space-coiled delay cables (the main ring delay line is omitted, and the error rate is shortened. Ring delay line), the internal RF connection of the system is simple. These are very meaningful to the standardization of product production process.

In a word, using the method and device provided by the present invention to realize a linear power amplifier will not only simplify the system structure, reduce technical difficulty, shorten the research and development cycle, but also greatly reduce the development and production costs, and the production process is standard and simple, which is conducive to large-scale Mass production.

Claims (6)

1, a kind of linear power amplifier realization device based on feed-forward technology is characterized in that comprising following module: device provided by the invention is made up of four circuit modules: channel detection module (31), main power amplifier module (32), carrier elimination Module (33), distortion elimination module (34); The channel detection module (31) further includes a coupler A (312), a down-conversion mixer (313), an intermediate frequency filter (314), a wave detector (315), and a monitor (316); Described carrier elimination module (33) further comprises coupler B (332), down-conversion mixer (333), ADC (334), digital filter (335), DAC (336), up-conversion mixer ( 337), local oscillator (338); Described distortion elimination module (34) further comprises by coupler C (343), electrical control attenuator (346), electrical control phase shifter (347), error amplifier (348); After the working carrier frequency signal (311) enters the channel detection module (31), it is sent to the coupler A (312), and the through signal (321) enters the main power amplifier module (32), and the power amplification is carried out by the main power amplifier module (32); The signal is sent to the monitor (316) through the down-conversion mixer (313), intermediate frequency filter (314), and detector (315), and the detected carrier frequency channel information (318) is fed back by the monitor (316). Give the digital filter (335) of the carrier elimination module (33) for it to set the digital filter parameter; After the power signal (331) after the main power amplifier module (32) output amplifies enters the carrier elimination module (33), through the coupler B (332) sampling, and then through down-conversion mixer (333), ADC (334), digital filter (335), DAC (336), up-conversion mixer (337) processing to eliminate the carrier frequency work signal, obtain the error signal (345) that only reflects the nonlinear distortion information of the main power amplifier and send it into the distortion elimination module (34); The electronically adjustable phase shifter (347) performs gain and phase adjustment, and then through the error amplifier (348), is coupled into the main channel by the coupler C (343), and is directly connected to the coupler B (332) in the carrier elimination module (33). After the signal (341) is delayed by the delay line (342) of the distortion elimination module (34), the output signal of the main power amplifier is cancelled, and the non-linear distortion components are eliminated to obtain a highly linear high-power operating carrier frequency Signal (344) output. 2, a kind of realization method based on the linear power amplification of feedforward technology, it is characterized in that comprising the following steps: Step 1: Determine the channel number of the system operating carrier frequency; Step 2: Set the digital filter according to the discrimination result of the channel where the working carrier frequency is located; Step 3: complete carrier cancellation; Step 4: Eliminate the distorted signal. 3. A method for realizing linear power amplification based on feedforward technology according to claim 2, wherein the step 1 of judging the channel number of the system operating carrier frequency further includes the following process: First, the power of part of the input signal 311 is sampled by the coupler A (312) and enters the down-conversion mixer A (313), and the radio frequency signal is reduced to an intermediate frequency through the down-conversion frequency mixing of the mixer A (313), and then by An intermediate frequency band-pass filter (314) whose bandwidth is exactly the width of a communication channel is filtered, and then the DC voltage obtained by the detector (315) is sent to the monitor (316); Secondly, the monitor (316) controls the output frequency of the frequency synthesizer (317) to perform channel scanning from low to high or from high to low in frequency steps identical to the standard channel spacing of the communication system; Then, with reference to a fixed threshold, the monitor (316) detects whether the detection voltage of the corresponding channel is higher than the threshold, and determines whether the channel has a working carrier frequency signal. The system working channel number. 4. A method for realizing linear power amplification based on feedforward technology according to claim 2, characterized in that: said step 2 sets the digital filter according to the channel discrimination result of the operating carrier frequency and further refers to setting the digital filter according to the first The work carrier frequency channel number that one step obtains by monitor (316) transmits, the digital signal processing DSP program module entrance parameter is set, and the digital filter algorithm with multi-window notch filter characteristic is set up by corresponding program module, makes digital filter The notch "grooving" position of the device (335) corresponds to the working carrier frequency channel one by one. 5. A method for realizing linear power amplification based on feedforward technology according to claim 2, characterized in that: the completion of carrier cancellation in step 3 further includes the following process: First, the output signal of the main power amplifier (322) is sampled; the coupler B (332) connected in series in the output channel of the main power amplifier is used to sample part of the output power signal of the main power amplifier. The intermodulation distortion information introduced by the linear characteristic sends this signal to mixer 2 (333): Secondly, down-conversion; the main power amplifier output sampling signal is mixed with the local oscillator signal generated by the local oscillator (338), and the frequency is down-converted to a lower frequency range; Again, the analog/digital conversion ADC; Utilize the analog/digital converter (334) to convert the low-frequency analog signal into digital information, and send it into the digital filter (335); Then, digital filtering; by the digital filtering algorithm that step 2 completes, the digital information of input digital filter (335) is filtered out the information corresponding to carrier signal; Then, the digital/analog conversion DAC; Utilize the digital/analog converter (336) to restore the output digital signal of the digital filter (335) into an analog signal; Finally, the frequency is up-converted; after the local oscillator signal generated by the mixer C (337) and the local oscillator (338) is mixed and up-converted, the low-frequency signal restored by the DAC is restored to the radio frequency. 6. A method for realizing linear power amplification based on feedforward technology according to claim 2, characterized in that: said step 4 to realize distorted signal elimination further includes the following process: First, delay; use the fixed cable delay line to delay the high-power signal output by the main power amplifier (322) and send it to the system output end after delay processing; Secondly, the error is amplified; the radio frequency signal obtained in the third step that has eliminated the carrier frequency signal is also sent to the system output after amplitude phase adjustment and linear amplification; Finally, the distortion is eliminated; use the coupler C (343) to realize the subtraction of the high-power signal output by the main power amplifier and the error radio frequency signal that has eliminated the carrier frequency signal, and cancel the distortion component in the output of the main power amplifier to complete the elimination of the distorted signal Linearization function.

Images