WO2024051520A1 - 数字预失真电路及方法、射频芯片以及电子设备 - Google Patents

数字预失真电路及方法、射频芯片以及电子设备 Download PDF

Info

Publication number
WO2024051520A1
WO2024051520A1 PCT/CN2023/115425 CN2023115425W WO2024051520A1 WO 2024051520 A1 WO2024051520 A1 WO 2024051520A1 CN 2023115425 W CN2023115425 W CN 2023115425W WO 2024051520 A1 WO2024051520 A1 WO 2024051520A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
frequency
circuit
frequency signal
predistortion
Prior art date
Application number
PCT/CN2023/115425
Other languages
English (en)
French (fr)
Inventor
邓森洋
宁东方
陈文华
张作锋
陈龙
李运华
Original Assignee
中兴通讯股份有限公司
清华大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司, 清华大学 filed Critical 中兴通讯股份有限公司
Publication of WO2024051520A1 publication Critical patent/WO2024051520A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
    • H03F1/32—Modifications of amplifiers to reduce non-linear distortion
    • H03F1/3241—Modifications of amplifiers to reduce non-linear distortion using predistortion circuits
    • H03F1/3258—Modifications of amplifiers to reduce non-linear distortion using predistortion circuits based on polynomial terms
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
    • H03F1/02—Modifications of amplifiers to raise the efficiency, e.g. gliding Class A stages, use of an auxiliary oscillation
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
    • H03F1/32—Modifications of amplifiers to reduce non-linear distortion
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
    • H03F3/189—High-frequency amplifiers, e.g. radio frequency amplifiers
    • H03F3/19—High-frequency amplifiers, e.g. radio frequency amplifiers with semiconductor devices only
    • H03F3/195—High-frequency amplifiers, e.g. radio frequency amplifiers with semiconductor devices only in integrated circuits
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
    • H03F3/20—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers
    • H03F3/21—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers with semiconductor devices only
    • H03F3/213—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers with semiconductor devices only in integrated circuits
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/38—Synchronous or start-stop systems, e.g. for Baudot code
    • H04L25/40—Transmitting circuits; Receiving circuits
    • H04L25/49—Transmitting circuits; Receiving circuits using code conversion at the transmitter; using predistortion; using insertion of idle bits for obtaining a desired frequency spectrum; using three or more amplitude levels ; Baseband coding techniques specific to data transmission systems
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03F—AMPLIFIERS
    • H03F2200/00—Indexing scheme relating to amplifiers
    • H03F2200/451—Indexing scheme relating to amplifiers the amplifier being a radio frequency amplifier

Definitions

  • the present disclosure relates to the field of communication technology, and in particular, to a digital predistortion circuit and method, a radio frequency chip and an electronic device.
  • the baseband signal transmitted to the radio frequency front-end circuit may be a single-frequency signal, a dual-frequency signal, or even a tri-frequency signal.
  • inventions of the present disclosure provide a digital predistortion circuit.
  • the digital predistortion circuit includes a first spectrum splicing circuit and a digital predistorter.
  • the first spectrum splicing circuit is used to splice the spectrum of the first multi-frequency signal to obtain a first broadband signal.
  • the first multi-frequency signal includes signals of two or more frequency bands, and the bandwidth of the first wide-band signal is smaller than the frequency interval between the maximum frequency band and the minimum frequency band in the first multi-frequency signal.
  • the digital predistorter is coupled to the first spectrum splicing circuit.
  • the digital predistortion circuit is used to receive the first wideband signal and perform predistortion on the first wideband signal according to a first predistortion coefficient to obtain a predistorted wideband signal.
  • the first predistortion coefficient is determined based on the first wideband signal and the nonlinear characteristics of the power amplifier in the radio frequency front-end circuit.
  • the above-mentioned digital predistortion circuit may also include a second spectrum splicing circuit and a model extraction circuit.
  • the second spectrum splicing circuit is used to couple with the radio frequency front-end circuit.
  • the second spectrum splicing circuit is used to splice the spectrum of the second multi-frequency signal output by the power amplifier to obtain a second broadband signal.
  • the bandwidth of the second wideband signal is the same as the bandwidth of the first wideband signal.
  • the model extraction circuit is coupled to the first spectrum splicing circuit and the second spectrum splicing circuit, and is used to perform modeling according to the first wideband signal and the second wideband signal to extract a first predistortion coefficient.
  • the digital predistorter is also coupled to a model extraction circuit for receiving the first predistortion coefficient.
  • the above-mentioned digital predistortion circuit may also include a spectrum separation circuit.
  • the spectrum separation circuit is coupled to the digital predistorter and is used to separate the spectrum of the predistorted broadband signal into the predistorted multi-frequency signal, so that the output of the pre-distorted multi-frequency signal through the power amplifier meets the linearity index.
  • the above-mentioned digital predistorter is selectively coupled to the first spectrum splicing circuit.
  • the above-mentioned digital predistorter is also used.
  • the first single frequency signal is predistorted according to the second predistortion coefficient to obtain a predistorted single frequency signal; the second predistortion coefficient is based on the first single frequency signal and the power. determined by the nonlinear characteristics of the amplifier.
  • the model extraction circuit is further configured to receive the first single frequency signal and the second single frequency signal output by the power amplifier, and perform modeling based on the first single frequency signal and the second single frequency signal. , to extract the second predistortion coefficient.
  • the digital predistorter is also coupled to the model extraction circuit and is used for receiving the second predistortion coefficient.
  • the first wideband signal is obtained by frequency-shifting the signals of each frequency band in the first multi-frequency signal.
  • the signals of each frequency band in the first multi-frequency signal are frequency-shifted.
  • the frequency interval satisfies the N-order nonlinearity of the power amplifier, and N is an odd number greater than or equal to 3.
  • an embodiment of the present disclosure provides a radio frequency processing circuit.
  • the radio frequency processing circuit includes a radio frequency front-end circuit and And the digital predistortion circuit in any embodiment of the first aspect above.
  • the digital predistortion circuit is coupled to the radio frequency front-end circuit.
  • an embodiment of the present disclosure provides a radio frequency chip.
  • the radio frequency chip includes a package structure, and the radio frequency processing circuit of the second aspect packaged in the package structure.
  • inventions of the present disclosure provide an electronic device.
  • the electronic device includes a baseband processor, and the radio frequency processing circuit in the above second aspect or the radio frequency chip in the above third aspect.
  • embodiments of the present disclosure provide a digital predistortion method.
  • the method is applied to a radio frequency processing circuit that includes a radio frequency front-end circuit that includes a power amplifier.
  • the method includes receiving a first multi-frequency signal, the multi-frequency signal including signals of two or more frequency bands.
  • the first multi-frequency signal is spectrum spliced to obtain a first wide-band signal, and the bandwidth of the first wide-band signal is smaller than the frequency interval between the maximum frequency band and the minimum frequency band in the first multi-frequency signal.
  • the first wideband signal is predistorted according to the first predistortion coefficient to obtain a predistorted wideband signal; the first predistortion coefficient is determined based on the first wideband signal and the nonlinear characteristics of the power amplifier in the radio frequency front-end circuit. .
  • the method may further include: performing spectrum separation on the pre-distorted broadband signal to obtain a pre-distorted multi-frequency signal, so that the pre-distorted multi-frequency signal is output through the power amplifier to meet the linearity index.
  • the first predistortion coefficient is modeled and extracted based on the first wideband signal and the second wideband signal.
  • the second wideband signal is spectrum spliced to the second multi-frequency signal output by the power amplifier. owned.
  • the method may further include: receiving a first single frequency signal.
  • the first single-frequency signal is predistorted according to the second predistortion coefficient to obtain a predistorted single-frequency signal; the second predistortion coefficient is determined based on the first single-frequency signal and the nonlinear characteristics of the power amplifier.
  • the second predistortion coefficient is extracted based on modeling of the first single frequency signal and the second single frequency signal output by the power amplifier.
  • the first wideband signal is obtained by frequency-shifting the signals of each frequency band in the first multi-frequency signal.
  • the signals of each frequency band in the first multi-frequency signal are frequency-shifted.
  • the frequency interval satisfies the N-order nonlinearity of the power amplifier, and N is an odd number greater than or equal to 3.
  • embodiments of the present disclosure provide a computer-readable storage medium.
  • the computer-readable storage medium includes computer instructions that, when executed on the electronic device, cause the electronic device to perform the method in any embodiment of the fifth aspect.
  • embodiments of the present disclosure provide a computer program product.
  • the computer program product When the computer program product is run on a computer, it causes the computer to execute the method described in any of the possible implementations in the fifth aspect. .
  • Figure 1 is a structural diagram of an electronic device according to some embodiments.
  • Figure 2 is a schematic diagram of nonlinear distortion of a power amplifier according to some embodiments.
  • Figure 3 is a structural diagram of a digital predistortion circuit according to some embodiments.
  • Figure 4 is a schematic diagram of the processing process of the first spectrum splicing circuit according to some embodiments.
  • Figure 5 is a structural diagram of another digital predistortion circuit according to some embodiments.
  • Figure 6 is a schematic diagram of a processing process of a spectrum separation circuit according to some embodiments.
  • Figure 7 is a structural diagram of yet another digital predistortion circuit according to some embodiments.
  • Figure 8 is a schematic diagram of a connection method of the digital predistortion circuit in Figure 7;
  • Figure 9 is a structural diagram of a radio frequency processing circuit according to some embodiments.
  • FIG. 10 is a structural diagram of another radio frequency processing circuit according to some embodiments.
  • Figure 11 is a flow chart of a digital predistortion method according to some embodiments.
  • Figure 12 is a flow chart of another digital predistortion method according to some embodiments.
  • Figure 13 is a flow chart of yet another digital predistortion method according to some embodiments.
  • first and second are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of indicated technical features. Thus, features defined as “first” and “second” may explicitly or implicitly include one or more of these features. In the description of the embodiments of the present disclosure, unless otherwise specified, "plurality” means two or more.
  • At least one of A, B and C has the same meaning as “at least one of A, B or C” and includes the following combinations of A, B and C: A only, B only, C only, A and B The combination of A and C, the combination of B and C, and the combination of A, B and C.
  • a and/or B includes the following three combinations: A only, B only, and a combination of A and B.
  • the term “if” is optionally interpreted to mean “when” or “in response to” or “in response to determining” or “in response to detecting,” depending on the context.
  • the phrase “if it is determined" or “if [stated condition or event] is detected” is optionally interpreted to mean “when it is determined" or “in response to the determination" or “on detection of [stated condition or event]” or “in response to detection of [stated condition or event]”.
  • FIG. 1 is a schematic structural diagram of an electronic device according to some embodiments.
  • the electronic device may be a base station or a terminal device.
  • the electronic device includes a baseband processor, radio frequency processing circuit and antenna.
  • the baseband processor processes the signals to be transmitted by the electronic device through the antenna.
  • electronic devices can extract useful information or data from signals to be transmitted, or convert these information and data into baseband signals to be transmitted.
  • This information or data can be user data such as voice, text, video, or data that controls information.
  • a baseband processor can implement signal processing operations such as modulation and demodulation, encoding and decoding.
  • a baseband processor may include multiple processing cores.
  • the baseband processor can be a central processing unit (CPU) or a digital signal processor (DSP).
  • the baseband processor can also be a microcontroller (micro control unit, MCU), a graphics processor (graphics processing unit, GPU), an image signal processor (image signal processor, ISP), an audio signal processor (audio signal processor, ASP), as well as processors specifically designed for artificial intelligence (AI) applications.
  • MCU microcontroller
  • MCU graphics processor
  • image signal processor image signal processor
  • ISP audio signal processor
  • ASP audio signal processor
  • the radio frequency processing circuit may include radio frequency front-end circuitry.
  • the radio frequency front-end circuit may include a radio frequency transmitting channel and a radio frequency receiving channel.
  • An electronic device may include one or more radio frequency transmitting channels and radio frequency receiving channels.
  • the baseband signal can be converted into an analog signal through a digital-to-analog converter DAC, and the analog signal is converted into a radio frequency signal through up-conversion processing by a local oscillator (referred to as local oscillator).
  • the radio frequency signal is processed by a power amplifier (PA), and finally radiated outward from a suitable antenna through the selection of an antenna switch.
  • PA power amplifier
  • the radio frequency signal received from the antenna is selected by the antenna switch and sent to the radio frequency receiving channel. Since the radio frequency signal received from the antenna is usually very weak, a low noise amplifier (LNA) is usually used to amplify it.
  • LNA low noise amplifier
  • the amplified signal first undergoes down-conversion processing of the local oscillator, then passes through the analog-to-digital converter ADC, and finally becomes a baseband signal and is transmitted to the baseband processor for processing.
  • the power amplifier operates in a nonlinear region, which causes nonlinear distortion of the radio frequency signal amplified by the power amplifier, resulting in an increase in the bit error rate during electronic device communication.
  • digital predistortion processing is usually performed on the signal to be transmitted to adjust the nonlinear distortion of the in-band signal.
  • the radio frequency processing circuit may also include a digital predistortion circuit.
  • the digital predistortion circuit is coupled to the baseband processor and the radio frequency front-end circuit, and is used to perform digital predistortion on signals from the baseband processor.
  • the digitally predistorted baseband signal is processed into a radio frequency signal through the radio frequency front-end circuit, and then the radio frequency signal is transmitted through the antenna.
  • the signal in each frequency band requires a digital predistortion module (or digital predistortion circuit) to process. That is, for a dual-frequency signal, two digital predistortion modules (which can be called dual-frequency DPD models) are required to perform predistortion processing of the signal. Similarly, for a tri-band signal with three frequency bands, three digital predistortion circuits (which can be called a tri-band DPD model) may be needed to process reason. In addition, for single-frequency signals, only one digital predistortion module (which can be called a single-frequency DPD model) is needed for processing.
  • a digital predistortion module which can be called a single-frequency DPD model
  • the nonlinear characteristics of the power amplifier of the RF front-end circuit can be described by a memory polynomial (MP) model. If the input signal and output signal of the single-frequency signal are recorded as xs(n) and ys(n) respectively, the nonlinear behavior of the power amplifier (ie, single-frequency DPD model or broadband model) can be expressed as the following equation (1).
  • K and M represent the nonlinear order and memory depth respectively
  • amk is the model coefficient (ie, predistortion coefficient).
  • x_1 ⁇ d(n) represents the input signal of the low frequency band
  • x_2 ⁇ d(n) represents the input signal of the high frequency band
  • ⁇ _1 ⁇ /2
  • ⁇ is the frequency interval of the two frequency bands.
  • y_1 ⁇ d(n) represents the output signal of the low-frequency band
  • y_2 ⁇ d(n) represents the output signal of the high-frequency band.
  • the dual-frequency output signal can use a two-dimensional digital predistortion (2-D DPD ) model (i.e. dual-frequency DPD model), as shown in the following equation (4).
  • b_mkl ⁇ ((1)) and b_mkl ⁇ ((2)) are model coefficients (ie, predistortion coefficients).
  • Equation (4) It can be seen from equation (4) that the frequency interval ⁇ does not appear in the model expression.
  • the dual-frequency DPD model in equation (4) can be equivalent to the broadband model in equation (1), so the dual-frequency model can be modeled by modeling the broadband model.
  • the dual-frequency signal can be equivalent to a broadband signal to model the nonlinear behavior of the power amplifier. Therefore, the single-frequency DPD model can be used for the dual-frequency signal to perform digital predistortion to suppress or eliminate the power amplifier. nonlinear distortion.
  • the signals in each frequency band of the dual-frequency signal can be spectrum moved to reduce the frequency interval between the two adjacent frequency band signals, thereby narrowing the dual-frequency signal, etc.
  • the effect is the bandwidth of the wideband signal to reduce the sampling rate of the single-frequency DPD model, thereby reducing system power consumption.
  • (a) of FIG. 2 shows the input signal x_1 ⁇ d(n) of the low frequency band and the input signal x_2 ⁇ d(n) of the high frequency band.
  • the frequency interval ⁇ between the input signals of the two frequency bands is 2 ⁇ _1.
  • (b) of Figure 2 shows the output signals of the two frequency bands after the input signals of the above two frequency bands are amplified by the power amplifier, that is, the output signal of the low frequency band y_1 ⁇ d(n) and the output signal of the high frequency band y_2 ⁇ d (n). Comparing (a) and (b) of Figure 2, the output signals of these two frequency bands have nonlinear distortion.
  • the input signal x_1 ⁇ d(n) of the low frequency band and the input signal x_2 ⁇ d(n) of the high frequency band are reduced
  • the frequency interval ⁇ between the signals of the two frequency bands is reduced to 2 ⁇ _2 ( ⁇ _2 ⁇ _1).
  • the output signals of the two frequency bands amplified by the power amplifier are shown in (d) in Figure 2. Comparing (b) and (d) in Figure 2, it can be seen that the frequency intervals of the signals in the two frequency bands are different, which does not produce a large difference in the nonlinear distortion of the signal output by the power amplifier.
  • embodiments of the present disclosure provide a digital predistortion circuit that can perform digital predistortion processing on multi-frequency signals using a single-frequency DPD model, thereby simplifying the system architecture, increasing the data rate, and reducing system power consumption.
  • Figure 3 shows a schematic structural diagram of a digital predistortion circuit according to some embodiments.
  • the digital predistortion circuit includes a first spectrum splicing circuit and a digital pre-distortion (DPD).
  • the first spectrum splicing circuit performs spectrum splicing on the first multi-frequency signal to obtain a first broadband signal.
  • the first multi-frequency signal includes signals of two or more frequency bands, and the bandwidth of the second wide-band signal is smaller than the frequency interval between the maximum frequency band signal and the minimum frequency band signal in the first multi-frequency signal.
  • the digital predistorter is coupled to the first spectrum splicing circuit.
  • the digital predistorter is used to receive a first broadband signal, and perform predistortion on the first broadband signal according to a first predistortion coefficient to obtain a predistorted broadband signal, so that the output signal of the power amplifier in the radio frequency front-end circuit meets the linearity index.
  • the first predistortion coefficient is determined based on the first wideband signal and the nonlinear characteristics of the power amplifier in the radio frequency front-end circuit.
  • the frequency interval between adjacent frequency bands in the first multi-frequency signal is reduced through the first spectrum splicing circuit, so that a broadband signal with a smaller bandwidth is formed through spectrum splicing, so that it is possible to Increase data processing rate and reduce system power consumption.
  • the single-frequency DPD model i.e., single-frequency digital predistorter
  • the single-frequency DPD model can also be used to pre-distort the multi-frequency signals, thereby achieving simultaneous processing of single-frequency signals and multi-frequency signals through the single-frequency DPD model. signal requirements, thereby simplifying the system architecture and reducing system power consumption.
  • the above-mentioned first multi-frequency signal may be a multi-frequency signal from the baseband processor.
  • the above-mentioned first multi-frequency signal may be a dual-frequency signal, that is, a signal including two frequency bands.
  • the above-mentioned first multi-frequency signal may also be a tri-frequency signal, that is, a signal including three frequency bands.
  • the above-mentioned first multi-frequency signal can also be a four-frequency signal, a five-frequency signal, or even a six-frequency signal.
  • the embodiment of the present disclosure does not specifically limit the number of frequency bands of the first multi-frequency signal.
  • first multi-frequency signal as a dual-frequency signal or a three-frequency signal as an example
  • processing process of the first spectrum splicing circuit will be exemplarily explained below.
  • Figure 4 is a schematic diagram of the process of the first spectrum splicing circuit processing the multi-frequency signal into the first broadband signal.
  • the dual-frequency signal includes a signal X1 and a signal X2, and the signal X1 is a signal in the high-frequency band, and the signal X2 is a signal in the low-frequency band.
  • the baseband signal rate is low.
  • the signal can be first X1 and signal X2 are upsampled (also called upsampling) to increase the sampling rate of the signal.
  • the signal X1 can be down-converted to reduce the center frequency of the signal X1
  • the signal X2 can be up-converted to increase the center frequency of the signal X2.
  • the frequency interval between the down-converted signal X1 and the up-converted signal X2 is smaller than the frequency interval between the un-down-converted signal X1 and the un-up-converted signal The frequency separation between the highest frequency band signal and the lowest frequency band signal).
  • the down-converted signal X1 and the up-converted signal X2 are spectrum spliced through an adder to form a broadband signal (ie, the first broadband signal).
  • the bandwidth of the broadband signal is approximately equal to the frequency interval between the down-converted signal X1 and the up-converted signal X2, so that the bandwidth of the broadband signal output by the first spectrum splicing circuit is smaller than that of the dual-frequency signal.
  • the frequency separation between the highest frequency band signal and the lowest frequency band signal In this way, compared to broadband signals with wider bandwidth, In terms of number, the processing rate of the digital predistorter can be reduced and the power consumption of the circuit can be reduced.
  • the three-frequency signal includes signal X1, signal X2 and signal X3, and signal X1 is a signal in the high frequency band, and signal X2 is a signal in the middle frequency band.
  • the signal on the frequency band, signal X3 is the signal on the low frequency band.
  • the signal X1, the signal X2 and the signal X3 can be up-sampled (also called up-sampling) first to increase the sampling rate of the signal.
  • the signal X1 can be down-converted to reduce the center frequency of the signal
  • Signal X2 is up-converted or down-converted to reduce the frequency interval between signal X2 and signal X1, or the frequency interval between signal X2 and signal X3
  • signal X3 can be up-converted to increase the center frequency of signal X3.
  • the frequency intervals between signals in adjacent frequency bands are reduced, that is, the frequency intervals between signals X1 and X2, and the frequency intervals between signals X2 and signals X3 are both reduced.
  • the frequency-converted signal X1, signal X2 and signal X3 are spectrum spliced through an adder to form a broadband signal (ie, the first broadband signal).
  • the bandwidth of the broadband signal is roughly equal to the frequency interval between the down-converted signal X1 and the up-converted signal The frequency separation between the band signal and the lowest band signal.
  • the processing rate of the digital predistorter can be reduced and the power consumption of the circuit can be reduced.
  • the N-order nonlinearity of the power amplifier can be considered when splicing the spectrum of multi-frequency signals.
  • N is an odd number greater than or equal to 3. That is, according to the N-order nonlinearity of the power amplifier, the up-conversion or down-conversion of each frequency band signal is adjusted to reduce the variation range of the frequency interval between two adjacent frequency band signals. Therefore, the first wideband signal is obtained by frequency-shifting the signals of each frequency band in the first multi-frequency signal, and the frequency intervals of the frequency-shifted signals of each frequency band in the first multi-frequency signal satisfy the requirements of the power amplifier. N-order nonlinearity, thus improving the predistortion effect of multi-frequency signals, making the nonlinear distortion of the radio frequency signal output by the power amplifier smaller, higher signal quality, and better communication performance.
  • the above-mentioned N order may be, for example, third order, fifth order or seventh order.
  • the frequency intervals of the signals in each frequency band in the first multi-frequency signal after frequency shifting can satisfy the fifth order nonlinearity of the power amplifier. Since the frequency interval between each frequency band is very small when seventh-order nonlinearity is satisfied, it can usually be ignored. Furthermore, when the frequency intervals of the frequency-shifted signals in each of the above frequency bands satisfy the fifth-order nonlinearity of the power amplifier, they usually also satisfy the third-order nonlinearity. Therefore, when fifth-order nonlinearity is generally satisfied, the linearity index of the output of the power amplifier will be better.
  • multi-frequency signals can be realized by reducing the frequency interval between adjacent frequency bands and forming a broadband signal with a smaller bandwidth through spectrum splicing to increase the data rate and reduce system power consumption.
  • the single-frequency DPD model can be used to pre-distort the above-mentioned multi-frequency signal. Therefore, the above-mentioned digital predistorter can be a single-frequency DPD model. For example, as shown in (a) in Figure 3 and Figure 4, in a dual-frequency signal including a high-frequency band signal X1 and a low-frequency band signal X2, a smaller bandwidth (for example, a bandwidth of 2 ⁇ _2) wideband signal, which is digitally predistorted by the digital predistorter DPD to form a digitally predistorted wideband signal.
  • a smaller bandwidth for example, a bandwidth of 2 ⁇ _2
  • the digital predistorter in the above-mentioned digital predistortion circuit it is possible to achieve predistortion of single-frequency signals and predistortion of multi-frequency signals, thereby simplifying the system structure and improving the processing rate of the digital predistorter. Reduce system power consumption of the entire circuit.
  • the signal input to the digital predistorter is a broadband signal
  • the output signal of the digital predistorter is also a broadband signal
  • the broadband signal output by the digital predistorter is a broadband signal after linear predistortion.
  • the signal input to the RF front-end circuit needs to be a multi-frequency signal with the same frequency band as the first multi-frequency signal. Therefore, the pre-distorted broadband signal output by the digital predistorter is input to the RF front-end circuit. Before the front-end circuit, spectrum separation needs to be performed to form a multi-frequency signal with the same frequency band as the first multi-frequency signal.
  • the above-mentioned digital predistortion circuit may also include a spectrum separation circuit.
  • the spectrum separation circuit is coupled to the digital predistorter and is used to separate the spectrum of the predistorted broadband signal into the predistorted multi-frequency signal, so that the pre-distorted multi-frequency signal is output through the power amplifier to meet the linearity index.
  • the pre-distorted broadband signal can be processed into a pre-distorted multi-frequency signal through the spectrum separation circuit, so that the RF front-end circuit can modulate it to form a multi-frequency RF signal, which can be amplified by the power amplifier and radiated outward via the antenna, thus meeting the linearity index.
  • the processing process of the spectrum separation circuit may be a reverse process to the processing process of the first spectrum splicing circuit.
  • Figure 6 is a schematic diagram of the process of the spectrum separation circuit processing the pre-distorted broadband signal into a pre-distorted multi-frequency signal.
  • the first broadband signal can be passed through a band-pass filter to obtain multi-frequency signals in two frequency bands, such as signal X1' and signal X2', and then up-convert the signal X1', so that the signal X1' returns to the center frequency point of the original signal X1 (usually, for the baseband signal, it returns to the zero frequency); perform down-conversion of the signal X2' , causing the signal X2' to return to the center frequency point of the original signal X2.
  • the signal X1' and the signal X2' can be down-sampled and returned to the original sampling rate to form a pre-distorted multi-frequency signal.
  • the pre-distorted multi-frequency signal includes the pre-distorted high-frequency band signal.
  • the first broadband signal can be passed through a band-pass filter to obtain a multi-frequency signal in three frequency bands, such as signal X1 ', signal X2' and signal X3'. Then the signal X1' is up-converted, so that the signal X1' returns to the center frequency point of the original signal X1 (usually, for the baseband signal, it returns to the zero frequency); the signal X2' is down-converted or up-converted.
  • the signal X1', the signal X2' and the signal X3' can be down-sampled and returned to the original sampling rate to form a pre-distorted multi-frequency signal, which includes the pre-distorted high-frequency band signal.
  • the pre-distorted broadband signal is spectrum separated to obtain the pre-distorted multi-frequency signal, thus realizing the digital pre-distortion processing of the multi-frequency signal through the single-frequency digital pre-distorter DPD , to increase the data processing rate and reduce system power consumption.
  • the predistortion coefficient is used to characterize the nonlinear characteristic relationship between the signal input to the power amplifier and the signal output from the power amplifier. Therefore, in embodiments of the present disclosure, the predistortion coefficient can be extracted by modeling based on the input signal of the power amplifier and the output signal of the power amplifier. In some embodiments of the present disclosure, the first wideband signal may be regarded as an input signal of the power amplifier. Therefore, the above-mentioned first predistortion coefficient is determined based on the first broadband signal output by the first spectrum splicing circuit and the nonlinear characteristics of the power amplifier in the radio frequency front-end circuit.
  • modeling in order to determine the predistortion coefficient (for example, the first predistortion coefficient), modeling can be performed based on the signal input to the power amplifier (ie, the first broadband signal) and the signal output from the power amplifier, thereby extracting the predistortion coefficient. Distortion coefficient.
  • the digital predistortion circuit may further include a second spectrum splicing circuit and a model extraction circuit.
  • the second spectrum splicing circuit is used to couple with the radio frequency front-end circuit, and is used to perform spectrum splicing on the second multi-frequency signal output by the power amplifier to obtain a second broadband signal.
  • the model extraction circuit may be coupled to the first spectrum splicing circuit and the second spectrum splicing circuit for modeling based on the first wideband signal output by the first spectrum splicing circuit and the second wideband signal output by the second spectrum splicing circuit, Thus, the first predistortion coefficient is extracted. In this way, through repeated iterations, the predistortion coefficient that satisfies the nonlinear characteristics of the power amplifier can be extracted.
  • the model extraction circuit may also be coupled to the digital predistorter DPD for transmitting the extracted first predistortion coefficient to the digital predistorter DPD, or in other words, the digital predistorter may receive the first predistortion coefficient.
  • the pre-distortion coefficient is determined based on the input signal and output signal of the power amplifier, and the first broadband signal output by the above-mentioned first spectrum splicing circuit can be regarded as the power
  • the input signal of the amplifier and the output signal of the power amplifier are related to the nonlinear characteristics of the power amplifier, so the above-mentioned first predistortion coefficient is determined based on the first broadband signal and the nonlinear characteristics of the power amplifier.
  • the first broadband signal output by the above-mentioned first spectrum splicing circuit can be the input signal of the power amplifier, and the second spectrum splicing circuit is obtained by processing the signal output by the power amplifier, so the first The second broadband signal output by the two spectrum splicing circuits can be used as the output signal of the power amplifier.
  • the input signal of the power amplifier can be modeled as a reference signal and the output signal of the power amplifier to extract the predistortion coefficient (i.e., the first predistortion coefficient), that is, the first wideband signal and the second wideband signal can be constructed. module, extract the predistortion coefficient (i.e., the first predistortion coefficient).
  • the following describes the process of modeling based on the first wideband signal and the second wideband signal and extracting the predistortion coefficient (ie, the first predistortion coefficient).
  • the normalized signals of the input signal and the output signal of the power amplifier can be recorded as x_1(n), x_2(n), y_1(n) and y_2 (n)
  • the first broadband signal after spectrum splicing can be recorded as x ⁇ d(n)
  • the second broadband signal after spectrum splicing is recorded as x ⁇ d(n).
  • the nonlinear behavior of the power amplifier under concurrent multi-frequency signals can be modeled.
  • a general memory polynomial model can be used when modeling.
  • a direct learning architecture can be used to iteratively solve the model coefficients.
  • b i is the model coefficient vector of the i-th iteration
  • ⁇ is the iteration control factor.
  • Matrix A is the basis function matrix
  • y i and xi are the output and input sampling point sequence matrices respectively. It is not difficult to find from equation (6) that when solving the model coefficients, The least squares algorithm is used to find the optimal solution in the least squares sense. In addition, in order to improve the stability of numerical calculations, Tikhonov regularization can be performed first when using the least squares algorithm.
  • I is the identity matrix
  • parameter ⁇ is the regularization factor.
  • a search algorithm can be used to find the optimal parameter ⁇ , so that the model coefficients can be solved with the highest accuracy.
  • the model used to model the first wideband signal and the second wideband signal is not limited to the general memory polynomial model.
  • Other simplified forms of the Volterra series model can also be used, which are not limited by the embodiments of the present disclosure.
  • the solution architecture of the predistortion coefficient the above-mentioned direct learning architecture can be used, or the indirect learning architecture can be used, which is not limited by the embodiments of the present disclosure.
  • the solution method of the predistortion coefficient the above-mentioned least squares algorithm can be used, or the recursive least squares algorithm can be used, which is not limited by the embodiments of the present disclosure.
  • the first spectrum splicing circuit and the spectrum separation circuit can be bypassed. That is to say, as shown in FIG. 8 , the digital predistorter can be selectively coupled to the first spectrum splicing circuit. Of course, the digital predistorter can also be selectively coupled to the spectrum separation circuit. In the case where the digital predistorter is not coupled to the first spectrum splicing circuit, the digital predistorter can be used to receive the first single-frequency signal, and predistort the first single-frequency signal according to the second predistortion coefficient to obtain the predetermined predistorter. Distorted single frequency signal. The second predistortion coefficient is determined based on the first single frequency signal and the nonlinear characteristics of the power amplifier. In this way, compatible processing of single-frequency signals can be achieved, the system architecture can be simplified, and the system power consumption can be reduced.
  • the digital predistorter can be used to be directly coupled to the baseband processor for receiving the first single frequency signal from the baseband processor ( As the signal X1 in Figure 8), the first single-frequency signal is predistorted according to the second predistortion coefficient to obtain a predistorted single-frequency signal (the signal The output signal of the power amplifier satisfies the linearity index.
  • the pre-distorted single-frequency signal does not need to pass through the spectrum separation circuit, so the digital pre-distorter can also be directly coupled with the RF front-end circuit to transmit the pre-distorted single-frequency signal to the RF front-end circuit for processing. Processed and modulated into radio frequency signals for transmission.
  • the above-mentioned second predistortion coefficient is similar to the above-mentioned first predistortion coefficient and is determined based on the input signal and output signal of the power amplifier, and the first single frequency signal can be regarded as the input signal of the power amplifier. Therefore, for the above model extraction circuit, the model extraction circuit can also be used to be directly coupled with the baseband processor and the radio frequency front-end circuit to receive the first single frequency signal from the baseband processor and the output of the power amplifier from the radio frequency front-end circuit The second single-frequency signal at the terminal is used to extract the second predistortion coefficient for the single-frequency signal through the model extraction module.
  • the digital predistorter may also be coupled to the model extraction circuit for receiving the second predistortion coefficient, so as to predistort the first single-frequency signal according to the second predistortion coefficient.
  • the digital predistorter may also be coupled to the model extraction circuit for receiving the second predistortion coefficient, so as to predistort the first single-frequency signal according to the second predistortion coefficient.
  • the digital predistortion circuit provided by the embodiment of the present disclosure can use a single-frequency digital predistorter (ie, single-frequency DPD model) to achieve compatible processing of multi-frequency signals and single-frequency signals, thereby making the system
  • a single-frequency digital predistorter ie, single-frequency DPD model
  • the architecture is more simplified, which not only increases data rates but also reduces system power consumption.
  • each circuit module in the above-mentioned digital predistortion circuit can be implemented as a hardware circuit or can be implemented using software, or each circuit module can be partially implemented using hardware circuits and partially implemented using software algorithms.
  • the first spectrum splicing circuit and the digital predistorter can be implemented using hardware circuits, while other circuit modules (such as model Extraction circuit, spectrum separation circuit) are implemented using software algorithms. Therefore, the embodiments of the present disclosure do not place special restrictions on the implementation of each circuit module in the digital predistortion circuit.
  • some embodiments of the present disclosure also provide a radio frequency processing circuit.
  • the radio frequency processing circuit includes any one of the digital predistortion circuits and radio frequency front-end circuits in the above embodiments.
  • the RF front-end circuit includes a transmit path.
  • the transmitting path is used for coupling with the spectrum separation circuit, and for modulating the pre-distorted multi-frequency signal output by the spectrum separation circuit to form a multi-frequency radio frequency signal.
  • the power amplifier PA is located in the transmission path and is used to amplify multi-frequency radio frequency signals and then transmit them through the antenna.
  • the above-mentioned transmission path includes a plurality of first signal processing channels, and the number of the first signal processing channels is equal to or greater than the number of frequency bands of the multi-frequency signal.
  • the number of frequency bands of the signal For example, as shown in Figure 9, if the multi-frequency signal is a dual-frequency signal, the transmission path includes two or more first signal processing channels. For another example, as shown in FIG. 10 , if the multi-frequency signal is a tri-frequency signal, the transmission path includes three or more first signal processing channels.
  • each first signal processing channel includes a digital-to-analog converter (DAC) ) and local oscillators (such as f1 and f2 in Figure 9, or f1, f2, and f3 in Figure 10).
  • the digital-to-analog converter is used to convert the signals in each frequency band of the pre-distorted multi-frequency signal from digital to analog into analog signals, and the local oscillator is used to up-convert the analog signals to form radio frequency signals.
  • the RF signals of each frequency band that have been up-converted by the local oscillator can be combined by an adder, input to the power amplifier PA for amplification, and then radiated outward through the antenna.
  • the radio frequency front-end circuit also includes a feedback path.
  • the feedback path is used to couple with the second spectrum splicing circuit, and the feedback circuit is used to process the output signal of the power amplifier and feed it back to the second spectrum splicing circuit.
  • the feedback path includes two or more second signal processing channels, and the number of the second signal processing channels is equal to or greater than the number of frequency bands of the third signal.
  • the feedback path includes two or more second signal processing channels.
  • the feedback path includes three or more second signal processing channels.
  • each second signal processing channel includes a local oscillator (such as f1 and f2 in Figure 9, or f1, f2, and f3 in Figure 10), a bandpass filter, and an analog-to-digital converter. , ADC).
  • the local oscillator in the second signal processing channel is used to down-convert the analog signal output by the power amplifier PA to form a baseband signal.
  • the down-converted baseband signal can pass through a filter (such as a bandpass filter) to form each frequency band signal.
  • Each frequency band signal is then converted from analog to digital by an analog-to-digital converter to obtain a digital signal, which is then transmitted to the second spectrum splicing module to pass
  • the above-mentioned second broadband signal is formed by spectrum splicing.
  • the transmitting path can be directly coupled to the digital predistorter in order to receive the pre-distorted single-frequency signal.
  • the feedback path can also be directly coupled to the model extraction circuit, so that after processing the signal at the output end of the power amplifier into a single frequency signal, it is fed back to the model extraction module to extract the predistortion coefficient.
  • the radio frequency chip includes a packaging structure and a packaging structure
  • the packaging structure may be a packaging substrate, a silicon-based interposer, etc.
  • Some embodiments of the present disclosure also provide a digital predistortion method.
  • the digital predistortion method can be applied to a radio frequency processing circuit, the radio frequency processing circuit includes a radio frequency front-end circuit, and the radio frequency front-end circuit includes a power amplifier PA.
  • the digital predistortion method includes the following steps.
  • S1102. Perform spectrum splicing on the first multi-frequency signal to obtain a first wide-band signal.
  • the bandwidth of the first wide-band signal is smaller than the frequency interval between the maximum frequency band and the minimum frequency band in the first multi-frequency signal.
  • the signals of each frequency band in the first multi-frequency signal can be up-sampled, and the signals of each frequency band in the up-sampled signal can be up-converted or down-converted to reduce the frequency between two adjacent frequency band signals.
  • frequency intervals between; the first broadband signal is obtained by spectrum splicing the frequency band signals after reducing the frequency interval.
  • the frequency interval between each frequency band signal satisfies the N-order nonlinearity of the power amplifier, and N is an odd number greater than or equal to 3.
  • the implementation process can refer to the relevant description in the above-mentioned digital predistortion circuit, and will not be described again here.
  • the first predistortion coefficient is determined based on the first wideband signal and the nonlinear characteristics of the power amplifier.
  • the implementation process can refer to the relevant description in the above-mentioned digital predistortion circuit, and will not be described again here.
  • the signal output by the digital predistorter is also a broadband signal
  • the broadband signal is a broadband signal after linear predistortion.
  • the signal input to the radio frequency front-end circuit needs to be a multi-frequency signal with the same frequency band as the first multi-frequency signal. Therefore, as shown in Figure 12, the above-mentioned digital predistortion method may also include S1104.
  • S1104 Perform spectrum separation on the pre-distorted broadband signal to obtain a pre-distorted multi-frequency signal, so that the pre-distorted multi-frequency signal is output through the power amplifier to meet the linearity index.
  • the pre-distorted broadband signal can be filtered to obtain a multi-frequency signal including two or more frequency bands; and each frequency band signal in the multi-frequency signal can be up-converted or down-converted to make the signal in each frequency band
  • the center frequency point is returned to the center frequency point of the corresponding frequency band signal in the first multi-frequency signal; and then each frequency band signal in the first multi-frequency signal is down-sampled to obtain a pre-distorted multi-frequency signal.
  • spectrum separation of the pre-distorted width signal please refer to the relevant description in the above-mentioned digital pre-distortion circuit, and will not be described again here.
  • the first predistortion coefficient is obtained by modeling based on the first wideband signal and the second wideband signal.
  • the second wideband signal is obtained by spectrum splicing the second multi-frequency signal output by the power amplifier.
  • the above-mentioned digital pre-distortion method may also include S1301 to S1302.
  • S1302 Predistort the first single-frequency signal according to the second predistortion coefficient to obtain a predistorted single-frequency signal.
  • the second predistortion coefficient is determined based on the first single-frequency signal and the nonlinear characteristics of the power amplifier.
  • the second predistortion coefficient is provided based on modeling of the first single frequency signal and the second single frequency signal output by the power amplifier. obtained.
  • the baseband signal is usually transmitted continuously for a period of time, so the above method process can be executed in a loop.
  • the feedback path can collect the first multi-frequency signal output by the power amplifier PA.
  • spectrum splicing is performed on the first multi-frequency signal and the second multi-frequency signal output by the power amplifier PA, respectively, to obtain the first wide-band signal and the second wide-band signal respectively.
  • modeling is performed based on the first wideband signal and the second wideband signal to extract the predistortion coefficient.
  • the extracted predistortion coefficients are input to the digital predistorter, which can predistort the first broadband signal to obtain a predistorted broadband signal. Then, spectrum separation is performed on the pre-distorted broadband signal, and each of the first signal processing paths in the transmission channel is input to the power amplifier PA for transmission. During this process, the linearity of the multi-frequency signal output by the power amplifier can be observed. If the linearity requirements are not met, the above process can be continued.
  • the computer storage medium includes computer instructions.
  • the computer instructions When the computer instructions are run on an electronic device, the electronic device causes the electronic device to perform each step in the above method embodiment.
  • the computer storage media is non-transitory computer storage media.
  • Another embodiment of the present disclosure provides a computer program product.
  • the computer program product When the computer program product is run on a computer, it causes the computer to execute each step in the above method embodiment.
  • Each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the above integrated units can be implemented in the form of hardware or software functional units.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium.
  • the technical solution of the embodiments of the present disclosure is essentially or contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage
  • the medium includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present disclosure.
  • the aforementioned storage media include: flash memory, mobile hard disk, read-only memory, random access memory, magnetic disk or optical disk and other media that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Physics & Mathematics (AREA)
  • Nonlinear Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Algebra (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Amplifiers (AREA)

Abstract

一种数字预失真电路及方法、射频芯片以及电子设备。数字预失真电路包括第一频谱拼接电路和数字预失真器；第一频谱拼接电路用于将第一多频信号进行频谱拼接得到第一宽带信号，第一多频信号包括两个或更多个频带的信号，且第一宽带信号的带宽小于第一多频信号中的最大频带与最小频带之间的频率间隔；数字预失真器与第一频谱拼接电路耦接，用于接收第一宽带信号，并根据第一预失真系数对第一宽带信号进行预失真，得到预失真的宽带信号；第一预失真系数是根据第一宽带信号和射频前端电路的功率放大器的非线性特性确定的。

Description

数字预失真电路及方法、射频芯片以及电子设备
本公开要求于2022年09月08日提交的、申请号为202211096716.7的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本公开涉及通信技术领域，尤其涉及一种数字预失真电路及方法、射频芯片以及电子设备。
背景技术
随着第五代移动通信技术的发展，通信频谱范围越来越广，对频谱资源的调度也越来越复杂。在不同的信道场景或者频谱资源调度下，传输至射频前端电路的基带信号可能是单频信号，也可能是双频信号，甚至还可能是三频信号。
发明内容
第一方面，本公开实施例提供一种数字预失真电路。该数字预失真电路包括第一频谱拼接电路和数字预失真器。该第一频谱拼接电路用于将第一多频信号进行频谱拼接得到第一宽带信号。该第一多频信号包括两个或更多个频带的信号，且该第一宽带信号的带宽小于该第一多频信号中的最大频带与最小频带之间的频率间隔。该数字预失真器与该第一频谱拼接电路耦接。该数字预失真电路用于接收该第一宽带信号，并根据第一预失真系数对该第一宽带信号进行预失真，得到预失真的宽带信号。该第一预失真系数是根据该第一宽带信号以及射频前端电路中的功率放大器的非线性特性确定的。
在一些实施例中，上述数字预失真电路还可以包括第二频谱拼接电路和模型提取电路。该第二频谱拼接电路用于与射频前端电路耦接。该第二频谱拼接电路用于将功率放大器输出的第二多频信号进行频谱拼接得到第二宽带信号。该第二宽带信号的带宽与该第一宽带信号的带宽相同。该模型提取电路与该第一频谱拼接电路和该第二频谱拼接电路耦接，用于根据该第一宽带信号和该第二宽带信号进行建模，以提取第一预失真系数。该数字预失真器还与模型提取电路耦接，用于接收该第一预失真系数。
在一些实施例中，上述数字预失真电路还可以包括频谱分离电路。该频谱分离电路与数字预失真器耦接，用于将预失真的宽带信号频谱分离为预失真的多频信号，使得该预失真的多频信号经由功率放大器的输出而满足线性指标。
在一些实施例中，上述数字预失真器选择性地耦接该第一频谱拼接电路，在该数字预失真器不与该第一频谱拼接电路耦接的情况下，上述数字预失真器还用于接收第一单频信号，并根据第二预失真系数对该第一单频信号进行预失真得到预失真的单频信号；该第二预失真系数是根据该第一单频信号与该功率放大器的非线性特性确定的。
在一些实施例中，模型提取电路还用于接收该第一单频信号和该功率放大器的输出的第二单频信号，并根据该第一单频信号以及该第二单频信号进行建模，以提取第二预失真系数。该数字预失真器还与该模型提取电路耦接，用于接收该第二预失真系数。
在一些实施例中，该第一宽带信号是通过将该第一多频信号中的各个频带的信号进行频移后得到的，该第一多频信号中的各个频带的信号经过频移后的频率间隔满足该功率放大器的N阶非线性，N为大于或等于3的奇数。
第二方面，本公开实施例提供一种射频处理电路。该射频处理电路包括射频前端电路、以 及如上第一方面中任一些实施例中的数字预失真电路。该数字预失真电路与该射频前端电路耦接。
第三方面，本公开实施例提供一种射频芯片。该射频芯片包括封装结构、以及封装于该封装结构内的第二方面中的该射频处理电路。
第四方面，本公开实施例提供一种电子设备。该电子设备包括基带处理器、以及如上第二方面中的该射频处理电路或者如上第三方面中的该射频芯片。
第五方面，本公开实施例提供一种数字预失真方法。该方法应用于射频处理电路，该射频处理电路包括射频前端电路，该射频前端电路包括功率放大器。该方法包括：接收第一多频信号，该多频信号包括两个或更多个频带的信号。将该第一多频信号进行频谱拼接得到第一宽带信号，该第一宽带信号的带宽小于该第一多频信号中的最大频带与最小频带之间的频率间隔。根据第一预失真系数对该第一宽带信号进行预失真，得到预失真的宽带信号；该第一预失真系数是根据该第一宽带信号以及射频前端电路中的功率放大器的非线性特性确定的。
在一些实施例中，该方法还可以包括：将预失真的宽带信号进行频谱分离，得到预失真的多频信号，使得该预失真的多频信号经由功率放大器输出而满足线性指标。
在一些实施例中，该第一预失真系数是根据该第一宽带信号和第二宽带信号进行建模提取到的，该第二宽带信号是对功率放大器输出的第二多频信号进行频谱拼接得到的。
在一些实施例中，该方法还可以包括：接收第一单频信号。根据第二预失真系数对该第一单频信号进行预失真得到预失真的单频信号；该第二预失真系数是根据第一单频信号以及功率放大器的非线性特性确定的。
在一些实施例中，该第二预失真系数是根据该第一单频信号以及该功率放大器输出的第二单频信号进行建模而提取到的。
在一些实施例中，该第一宽带信号是通过将该第一多频信号中的各个频带的信号进行频移后得到的，该第一多频信号中的各个频带的信号经过频移后的频率间隔满足该功率放大器的N阶非线性，N为大于或等于3的奇数。
第六方面，本公开实施例提供一种计算机可读存储介质。该计算机可读存储介质包括计算机指令，当该计算机指令在电子设备上运行时，使得电子设备执行如上第五方面中任一些实施例中的方法。
第七方面，本公开实施例提供一种计算机程序产品，当所述计算机程序产品在计算机上运行时，使得所述计算机执行如上第五方面中任一种可能的实现方式中的所述的方法。
附图说明
为了更清楚地说明本公开中的技术方案，下面将对本公开一些实施例中所需要使用的附图作简单地介绍，然而，下面描述中的附图仅仅是本公开的一些实施例的附图，对于本领域普通技术人员来讲，还可以根据这些附图获得其他的附图。此外，以下描述中的附图可以视作示意图，并非对本公开实施例所涉及的产品的实际尺寸、方法的实际流程、信号的实际时序等的限制。
图1为根据一些实施例的电子设备的结构图；
图2为根据一些实施例的功率放大器的非线性失真的示意图；
图3为根据一些实施例的一种数字预失真电路的结构图；
图4为根据一些实施例的第一频谱拼接电路的处理过程示意图；
图5为根据一些实施例的另一种数字预失真电路的结构图；
图6为根据一些实施例的频谱分离电路的处理过程示意图；
图7为根据一些实施例的又一种数字预失真电路的结构图；
图8为图7中的数字预失真电路的一种连接方式示意图；
图9为根据一些实施例的一种射频处理电路的结构图；
图10为根据一些实施例的另一种射频处理电路的结构图；
图11为根据一些实施例的一种数字预失真方法的流程图；
图12为根据一些实施例的另一种数字预失真方法的流程图；
图13为根据一些实施例的又一种数字预失真方法的流程图。
具体实施方式
下面将结合附图，对本公开一些实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本公开一部分实施例，而不是全部的实施例。基于本公开所提供的实施例，本领域普通技术人员所获得的所有其他实施例，都属于本公开保护的范围。
除非上下文另有要求，否则，在整个说明书和权利要求书中，术语“包括”被解释为开放、包含的意思，即为“包含，但不限于”。在说明书的描述中，术语“一个实施例”、“一些实施例”、“示例性实施例”、“示例”或“一些示例”等旨在表明与该实施例或示例相关的特定特征、结构、材料或特性包括在本公开的至少一个实施例或示例中。上述术语的示意性表示不一定是指同一实施例或示例。此外，所述的特定特征、结构、材料或特点可以以任何适当方式包括在任何一个或多个实施例或示例中。
以下，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括一个或更多个该特征。在本公开实施例的描述中，除非另有说明，“多个”的含义是两个或两个以上。
在描述一些实施例时，可能使用了“耦接”和“连接”及其衍伸的表达。例如，描述一些实施例时可能使用了术语“连接”以表明两个或两个以上部件彼此间有直接物理接触或电接触。又如，描述一些实施例时可能使用了术语“耦接”以表明两个或两个以上部件有直接物理接触或电接触。然而，术语“耦接”或“通信耦合(communicatively coupled)”也可能指两个或两个以上部件彼此间并无直接接触，但仍彼此协作或相互作用。这里所公开的实施例并不必然限制于本文内容。
“A、B和C中的至少一个”与“A、B或C中的至少一个”具有相同含义，均包括以下A、B和C的组合：仅A，仅B，仅C，A和B的组合，A和C的组合，B和C的组合，及A、B和C的组合。
“A和/或B”，包括以下三种组合：仅A，仅B，及A和B的组合。
如本文中所使用，根据上下文，术语“如果”任选地被解释为意思是“当……时”或“在……时”或“响应于确定”或“响应于检测到”。类似地，根据上下文，短语“如果确定……”或“如果检测到[所陈述的条件或事件]”任选地被解释为是指“在确定……时”或“响应于确定……”或“在检测到[所陈述的条件或事件]时”或“响应于检测到[所陈述的条件或事件]”。
本文中“适用于”或“被配置为”的使用意味着开放和包容性的语言，不排除适用于或被配置为执行额外任务或步骤的设备。
另外，“基于”的使用意味着开放和包容性，因为“基于”一个或多个所述条件或值的过程、步骤、计算或其他动作在实践中可以基于额外条件或超出所述的值。
如本文所使用的那样，“约”、“大致”或“近似”包括所阐述的值以及处于特定值的可接受偏差范围内的平均值，所述可接受偏差范围如由本领域普通技术人员考虑到正在讨论的测量以及与特定量的测量相关的误差(即，测量系统的局限性)所确定。
图1为根据一些实施例的电子设备的结构示意图。该电子设备可以是基站，也可以是终端设备。如图1所示，该电子设备包括基带处理器、射频处理电路和天线。基带处理器对电子设备待通过天线发射的信号进行处理。例如，电子设备可以从待发射的信号中提取有用的信息或数据，或者将这些信息和数据转换为待发射的基带信号。这些信息或数据可以是语音、文本、视频等用户数据或控制信息的数据。例如，基带处理器可以实现诸如调制和解调，编码和解码等信号处理操作。
基带处理器可以包括多个处理核心。例如，该基带处理器可以是中央处理单元(center processing unit，CPU)，也可以是数字信号处理器(digital signal processor，DSP)。该基带处理器也可以是微控制器(micro control unit，MCU)，图形处理器(graphics processing unit，GPU)、图像信号处理器(image signal processor，ISP)，音频信号处理器(audio signal processor，ASP)，以及为人工智能(artificial intelligence，AI)应用专门设计的处理器。
射频处理电路可以包括射频前端电路。该射频前端电路可以包括射频发送通道和射频接收通道。在一个电子设备中，可以包括一条或多条射频发送通道以及射频接收通道。
对于射频发送通道而言，基带信号可经过数模转换器DAC变为模拟信号，该模拟信号经过本地振荡器(简称本振)的上变频处理变为射频信号。该射频信号经过功率放大器(power amplifier，PA)的处理，最终经过天线开关的选择，从合适的天线向外辐射。
对于射频接收通道而言，从天线处接收的射频信号经过天线开关的选择，送入射频接收通道。由于从天线接收的射频信号通常很微弱，通常采用低噪声放大器LNA放大。放大后的信号先经过本振的下变频处理，再经过模数转换器ADC，最终变为基带信号，传输至基带处理器中进行处理。
然而，随着功率放大器的发射功率的不断提高，功率放大器工作在非线性区域，会使得经过功率放大器放大的射频信号产生非线性失真，从而导致电子设备通信时的误码率增加。为了抑制信号的非线性失真，提高功率放大器的线性度，通常会对待发射的信号进行数字预失真处理以调节带内信号的非线性失真。
因此，如图1所示，射频处理电路还可以包括数字预失真电路。该数字预失真电路与基带处理器和射频前端电路耦接，用于对来自基带处理器的信号进行数字预失真。经过数字预失真的基带信号再通过射频前端电路处理为射频信号，然后通过天线发射该射频信号。
通常情况下，针对来自基带处理器的具有两个频带的双频信号，在进行数字预失真处理时，每个频带的信号均需要一个数字预失真模块(或数字预失真电路)来处理。即，对于双频信号，需要两个数字预失真模块(可以称为双频DPD模型)来进行信号的预失真处理。类似地，对于具有三个频带的三频信号，可能需要三个数字预失真电路(可以称为三频DPD模型)来处 理。此外，对于单频信号，仅需要一个数字预失真模块(可以称为单频DPD模型)来处理。由此可见，对于不同的频带数量的信号，需要建立不同的DPD模型来实现信号的数字预失真处理，这会增加DPD模块的配置，增加资源的消耗以及硬件的功耗。为提高DPD的处理速率，降低系统功耗，需要一种利用单频DPD模型对多频信号进行预失真的DPD电路。
不失一般性，射频前端电路的功率放大器的非线性特性可以用一个记忆多项式(memory polynomial，MP)模型来描述。如果把单频信号的输入信号和输出信号分别记为xs(n)和ys(n)，则功率放大器的非线性行为(即单频DPD模型或宽带模型)可以如下式(1)表示。
在上述公式中，K和M分别表示非线性阶数和记忆深度，amk是模型系数(即预失真系数)。
对于一个并发的双频信号可以表示为高、低两个频带的信号相加，如下式(2)所示。
在上述公式中，x_1^d(n)表示低频带的输入信号，x_2^d(n)表示高频带的输入信号，ω_1＝Δω/2，Δω是两个频带的频率间隔。将式(2)带入式(1)中，只考虑输入信号中落在原信号频带附近的分量，则可以得到双频信号的表达式，如下式(3)所示。
在上述公式中，y_1^d(n)表示低频带的输出信号，y_2^d(n)表示高频带的输出信号，该双频输出信号可以利用一个二维数字预失真(2-D DPD)模型(即双频DPD模型)表示，如下式(4)所示。
在上述公式中，b_mkl^((1))和b_mkl^((2))是模型系数(即预失真系数)。
从式(4)中可以看出，模型表达式中没有出现频率间隔Δω。式(4)中的双频DPD模型与公式(1)中的宽带模型可以进行等效，因此可以通过建模宽带模型来实现对双频模型的建模。
根据上面的推导，可以将双频信号等效为一个宽带信号来对功率放大器的非线性行为进行建模，从而可以针对双频信号使用单频DPD模型来进行数字预失真来抑制或消除功率放大器的非线性失真。
为了提高数据速率、降低系统功耗，可以对双频信号(或多频信号)中各个频带的信号进行频谱搬移，缩小相邻的两个频带信号之间的频率间隔，从而缩小双频信号等效为宽带信号的带宽，以降低对单频DPD模型的采样率，进而降低系统功耗。
如图2所示，在图2的(a)中显示了低频带的输入信号x_1^d(n)和高频带的输入信号x_2^d(n)。该两个频带的输入信号之间的频率间隔Δω为2ω_1。图2的(b)显示了上述两个频带的输入信号经过功率放大器放大后的两个频带的输出信号，即低频带的输出信号y_1^d(n)和高频带的输出信号y_2^d(n)。对比图2的(a)和(b)，这两个频带的输出信号均产生了非线性失真。
如图2中的(c)所示，对低频带的输入信号x_1^d(n)和高频带的输入信号x_2^d(n)缩小 频率间隔后，两个频带的信号之间的频率间隔Δω缩小为2ω_2(ω_2＜ω_1)，经过功率放大器放大后的两个频带的输出信号如图2中的(d)所示。对比图2中的(b)和(d)可以看出，两个频带的信号的频率间隔不同，并不会对功率放大器的输出的信号的非线性失真产生较大的差异。
因此，本公开实施例提供了一种数字预失真电路，该数字预失真电路能够使用单频DPD模型对多频信号进行数字预失真处理，从而简化系统架构，提高数据速率，降低系统功耗。
图3示出了根据一些实施例的数字预失真电路的结构示意图。如图3所示，该数字预失真电路包括第一频谱拼接电路和数字预失真器(digital pre-distortion，DPD)。第一频谱拼接电路将第一多频信号进行频谱拼接得到第一宽带信号。该第一多频信号包括两个或更多个频带的信号，且第二宽带信号的带宽小于该第一多频信号中的最大频带信号与最小频带信号之间的频率间隔。数字预失真器与所述第一频谱拼接电路耦接。该数字预失真器用于接收第一宽带信号，并根据第一预失真系数对第一宽带信号进行预失真，得到预失真的宽带信号，以使射频前端电路中的功率放大器的输出信号满足线性指标。该第一预失真系数是根据第一宽带信号以及射频前端电路中的功率放大器的非线性特性确定的。
可以理解的是，基于上述数字预失真电路，通过第一频谱拼接电路缩小第一多频信号中相邻频带之间的频率间隔，使得通过频谱拼接的方式形成带宽较小的宽带信号，从而可以提高数据处理速率，降低系统功耗。此外，将多频信号处理为宽带信号还可以利用单频DPD模型(即单频数字预失真器)来对多频信号进行预失真，从而实现通过单频DPD模型同时处理单频信号和多频信号的需求，进而简化系统架构，降低系统功耗。
可以理解的是，上述第一多频信号可以是来自基带处理器的多频信号。上述第一多频信号可以双频信号，即包括两个频带的信号。上述第一多频信号也可以是三频信号，即包括三个频带的信号。当然，上述第一多频信号也可以是四频信号、五频信号，甚至是六频信号，本公开实施例对第一多频信号的频带数量不做特殊限定。
下面以上述第一多频信号为双频信号或者三频信号为例，对第一频谱拼接电路的处理过程进行示例性的说明。
图4为第一频谱拼接电路将多频信号处理为第一宽带信号的过程示意图。对于双频信号而言，如图4中的(a)所示，假设双频信号包括信号X1和信号X2，且信号X1为高频带上的信号，信号X2为低频带上的信号。在对该双频信号进行频谱拼接时，由于该双频信号是来自基带处理器的基带信号，该基带信号速率低，为防止基带信号频谱拼接后因速率不够导致信号混叠，可以先将信号X1和信号X2进行上采样(也称为升采样)，以提高信号的采样速率。针对完成上采样的信号X1和信号X2，可以对信号X1进行下变频，降低信号X1的中心频率，并且可以对信号X2进行上变频，提高信号X2的中心频率。这样，经过下变频的信号X1和经过上变频的信号X2之间的频率间隔，小于未经过下变频的信号X1和未经过上变频的信号X2之间的频率间隔(即该双频信号中的最高频带信号与最低频带信号之间的频率间隔)。然后将经过下变频的信号X1和经过上变频的信号X2通过加法器进行频谱拼接，形成一个宽带信号(即第一宽带信号)。此时，该宽带信号的带宽大致等于经过下变频的信号X1和经过上变频的信号X2之间的频率间隔，使得通过第一频谱拼接电路输出的宽带信号的带宽、小于该双频信号中的最高频带信号与最低频带信号之间的频率间隔。这样，相比较带宽更宽的宽带信 号而言，可以降低数字预失真器的处理速率，降低电路的功耗。
类似地，对于三频信号而言，如图4中的(b)所示，假设三频信号包括信号X1、信号X2和信号X3，且信号X1为高频带上的信号，信号X2为中频带上的信号，信号X3为低频带上的信号。类似地，在对该三频信号进行频谱拼接时，可以先将信号X1、信号X2和信号X3进行上采样(也称为升采样)，以提高信号的采样速率。针对完成上采样的信号X1、信号X2和信号X3，可以对信号X1进行下变频，降低信号X1的中心频率；可以根据信号X2与信号X1、以及信号X2与信号X3之间的频率间隔，对信号X2进行上变频或下变频，以缩小信号X2与信号X1之间的频率间隔、或者信号X2与信号X3之间的频率间隔；可以对信号X3进行上变频，提高信号X3的中心频率。这样，相邻的频带的信号之间的频率间隔都被缩小了，即信号X1与信号X2之间的频率间隔、以及信号X2与信号X3之间的频率间隔均被缩小。然后将变频后的信号X1、信号X2和信号X3通过加法器进行频谱拼接，形成一个宽带信号(即第一宽带信号)。此时，该宽带信号的带宽大致等于经过下变频的信号X1和经过上变频的信号X3之间的频率间隔，使得通过第一频谱拼接电路输出的宽带信号的带宽，小于三频信号中的最高频带信号与最低频带信号之间的频率间隔。类似地，相比较带宽更宽的宽带信号而言，可以降低数字预失真器的处理速率，降低电路的功耗。
对于频谱拼接后的宽带信号，为了保证非线性的频谱再生分量互相不发生混叠，可以在对多频信号进行频谱拼接时，考虑功率放大器的N阶非线性，N为大于等于3的奇数，即根据功率放大器的N阶非线性来调整各频带信号在进行上变频或下变频、以缩小相邻两个频带信号之间的频率间隔的变化幅度。因此，第一宽带信号是通过将第一多频信号中的各个频带的信号进行频移后得到的，且第一多频信号中的各个频带的信号经过频移后的频率间隔满足功率放大器的N阶非线性，从而提高对多频信号进行预失真的效果，使功率放大器输出的射频信号的非线性失真更小，信号质量更高，通信性能更好。
需要说明的是，上述N阶例如可以是三阶、五阶或七阶，例如可以使第一多频信号中的各个频带的信号经过频移后的频率间隔满足功率放大器的五阶非线性。由于在满足七阶非线性时各个频带之间的频率间隔很小，通常可以不考虑。并且，在上述各个频带的信号经过频移后的频率间隔满足功率放大器的五阶非线性的情况下，通常也满足三阶非线性。因此，通常满足五阶非线性时，功率放大器的输出的线性指标会更好。
通过如上的方式可以实现多频信号通过缩小相邻频带之间的频率间隔，并通过频谱拼接的方式形成带宽较小的宽带信号，以提高数据速率，降低系统功耗。
需要说明的是，在通过上述第一频谱拼接电路对多频信号进行处理后，可以使用单频DPD模型对上述多频信号进行预失真。因此，上述数字预失真器可以是单频DPD模型。示例性地，如图3和图4中的(a)所示，在包括高频带信号X1和低频带信号X2的双频信号，经过第一频谱拼接电路形成了带宽更小(例如带宽为2ω_2)的宽带信号，该宽带信号通过数字预失真器DPD进行数字预失真后，形成了数字预失真后的宽带信号。
因此，通过上述数字预失真电路中的数字预失真器，既能够实现单频信号的预失真，也能够实现多频信号的预失真，从而简化系统结构，提高数字预失真器的处理速率，进而降低整个电路的系统功耗。
此外，可以理解的是，由于数字预失真器输入的信号为宽带信号，因此数字预失真器输出 的信号也为宽带信号，该数字预失真器输出的宽带信号是经过线性预失真后的宽带信号。然而，为了保证信号的正确传输，输入至射频前端电路中的信号需要是与第一多频信号的频带相同的多频信号，因此数字预失真器输出的预失真后的宽带信号在输入至射频前端电路之前，需要进行频谱分离形成与第一多频信号的频带相同的多频信号。
因此，在一些实施例中，如图5所示，上述数字预失真电路还可以包括频谱分离电路。该频谱分离电路与数字预失真器耦接，用于将预失真的宽带信号频谱分离为预失真的多频信号，使得该预失真的多频信号经由功率放大器输出而满足线性指标。这样，通过频谱分离电路可以将预失真的宽带信号处理为预失真的多频信号，以便射频前端电路进行调制形成多频射频信号，以便通过功率放大器放大后经由天线向外辐射，从而满足线性指标。该频谱分离电路的处理过程可以是与第一频谱拼接电路的处理过程相反的过程。
下面以上述第一多频信号为双频信号或者三频信号为例，对频谱分离电路的处理过程进行示例性的说明。
图6为频谱分离电路将预失真后的宽带信号处理为预失真的多频信号的过程示意图。在对双频信号进行预失真的情况下，如图6中的(a)所示，可以将第一宽带信号经过带通滤波器，得到两个频带的多频信号，如信号X1’和信号X2’，然后对信号X1’进行上变频，使信号X1’返回至原信号X1的中心频点处(通常，对于基带信号而言，即返回至零频处)；对信号X2’进行下变频，使信号X2’返回至原信号X2的中心频点处。最后，可以对信号X1’和信号X2’进行下采样，返回至原来的采样速率上，从而形成预失真后的多频信号，该预失真后的多频信号包括预失真后的高频带信号X1’和预失真后的低频带信号X2’。
类似地，在对三频信号进行预失真的情况下，如图6中的(b)所示，可以将第一宽带信号经过带通滤波器，得到三个频带的多频信号，如信号X1’、信号X2’和信号X3’。然后对信号X1’进行上变频，使信号X1’返回至原信号X1的中心频点处(通常，对于基带信号而言，即返回至零频处)；对信号X2’进行下变频或上变频，使信号X2’返回至原信号X2的中心频点处；对信号X3’进行下变频，使信号X3’返回至原信号X3的中心频点处。最后，可以对信号X1’、信号X2’和信号X3’进行下采样，返回至原来的采样率上，从而形成预失真后的多频信号，该多频信号包括预失真后的高频带信号X1’、预失真后的中频带信号X2’和预失真后的低频带信号X3’。
这样一来，通过频谱分离电路，对预失真后的宽带信号进行频谱分离便得到了预失真后的多频信号，从而实现了通过单频数字预失真器DPD对多频信号的数字预失真处理，以提高数据处理速率，降低系统功耗。
此外，还需要说明的是，通常情况下，预失真系数用于表征输入至功率放大器中的信号与从功率放大器中输出的信号之间的非线性特征的关系。因此，在本公开的实施例中，预失真系数可以根据功率放大器的输入信号和功率放大器的输出信号进行建模来提取。在本公开一些实施例中，第一宽带信号可以看作为功率放大器的输入信号。因此，上述第一预失真系数是根据第一频谱拼接电路输出的第一宽带信号、以及射频前端电路中的功率放大器的非线性特征确定的。在公开一些实施例中，为了确定预失真系数(例如第一预失真系数)，可以根据输入至功率放大器的信号(即第一宽带信号)以及输出功率放大器的信号进行建模，从而提取出预失真系数。
由于在实际的信号传递和处理的过程中，在来自基带处理器的信号为多频信号的情况下，功率放大器的输出信号也为多频信号。为了便于根据第一宽带信号以及功率放大器的输出信号进行建模，功率放大器的输出信号也需要处理为宽带信号。在此情况下，在一些实施例中，如图7所示，数字预失真电路还可以包括第二频谱拼接电路和模型提取电路。第二频谱拼接电路用于与射频前端电路耦接，以用于对功率放大器输出的第二多频信号进行频谱拼接得到第二宽带信号。模型提取电路可以与第一频谱拼接电路和第二频谱拼接电路耦接，以便用于根据第一频谱拼接电路输出的第一宽带信号和第二频谱拼接电路输出的第二宽带信号进行建模，从而提取第一预失真系数。这样，通过反复迭代的方式，可以提取出满足功率放大器的非线性特性的预失真系数。模型提取电路还可以与数字预失真器DPD耦接，以用于将提取得到的第一预失真系数传输给数字预失真器DPD，或者说，数字预失真器可以接收第一预失真系数。
可以理解的是，针对功率放大器的输入信号进行预失真，通常预失真系数是根据功率放大器的输入信号和输出信号来确定的，而上述第一频谱拼接电路输出的第一宽带信号可以视为功率放大器的输入信号，且功率放大器的输出信号与功率放大器的非线性特性有关，因此上述第一预失真系数是根据第一宽带信号和功率放大器的非线性特性确定的。
在实际确定第一预失真系数时，上述第一频谱拼接电路输出的第一宽带信号可以为功率放大器的输入信号，而第二频谱拼接电路是对功率放大器输出的信号进行处理得到的，因此第二频谱拼接电路输出的第二宽带信号可以作为功率放大器的输出信号。此时，功率放大器的输入信号可以作为参考信号与功率放大器的输出信号进行建模，提取预失真系数(即第一预失真系数)，即可以通过对第一宽带信号和第二宽带信号进行建模，提取预失真系数(即第一预失真系数)。
下面描述根据第一宽带信号和第二宽带信号进行建模，提取预失真系数(即第一预失真系数)的过程。
在对第一宽带信号和第二宽带信号进行建模时，可以将功率放大器的输入信号和输出信号归一化后的信号记为x_1(n)、x_2(n)、y_1(n)和y_2(n)，经过频谱拼接后的第一宽带信号可以记为x^d(n)，经过频谱拼接后的第二宽带信号记为x^d(n)。在频谱拼接时，两个频带的频率间隔设置为Δω＝Δω_1+Δω_2，所以经过频谱拼接后的第一宽带信号x^d(n)和第二宽带信号y^d(n)可以为如下式(5)。
利用频谱拼接后的第一宽带信号(作为功率放大器的输入信号)和第二宽带信号(作为功率放大器的输出信号)就可以对该功率放大器在并发多频信号下的非线性行为进行建模，建模时可以采用一般记忆多项式模型。
为了提取模型系数，可以采用直接学习架构，迭代求解模型系数，按照直接学习，模型系数更新公式如下式(6)。
bi+1＝bi-μ(AHA)-1AH(yi-xi)      式(6)。
在上述公式中，bi是第i次迭代的模型系数向量，μ是迭代控制因子，μ越大迭代速度越快，迭代稳定性越差，实际系统中需要在迭代速度和迭代稳定性之间折中。矩阵A为基函数矩阵，yi和xi分别是输出和输入采样点序列矩阵。从式(6)中不难发现，在求解模型系数时， 采用的是最小二乘算法，求解最小二乘意义下的最优解。此外，为了提高数值计算的稳定性，在利用最小二乘算法时可以先做吉洪洛夫正则化。当矩阵的最小特征值接近0时，条件数会很大时，利用正则化可以增大矩阵的最小特征值，从而防止矩阵的病态性，改善方程求解精度。吉洪洛夫正则化的详细实现在公式(6)的基础上可以用如下式(7)表示。
bi+1＝bi-μ(AHA+λI)-1AH(yi-xi)       式(7)。
在上述公式中，I为单位矩阵，参数λ为正则化因子。在实际操作时，可以采用搜索算法找到最优的参数λ，使得模型系数的求解精度最高。
需要说明的是，对第一宽带信号和第二宽带信号进行建模而采用的模型并不限于一般记忆多项式模型，也可以使用Volterra级数模型的其他简化形式，本公开实施例不做限制。另外，对于预失真系数的求解架构，可以采用上述直接学习架构，也可以采用间接学习架构，本公开实施例不做限制。此外，对于预失真系数的求解方法，可以使用上述最小二乘法算法，也可以采用递归最小二乘法算法，本公开实施例不做限制。
针对上述数字预失真电路，在对单频信号进行预失真处理时，可以将第一频谱拼接电路和频谱分离电路旁路掉。也就是说，如图8所示，数字预失真器可以选择性地耦接第一频谱拼接电路，当然数字预失真器也可以选择性地耦接频谱分离电路。在数字预失真器不与第一频谱拼接电路耦接的情况下，数字预失真器可以用于接收第一单频信号，并根据第二预失真系数对第一单频信号进行预失真得到预失真的单频信号。该第二预失真系数是根据第一单频信号和功率放大器的非线性特性确定的。这样，可以满足对单频信号进行兼容处理，简化系统架构，降低系统功耗。
例如，在数字预失真器不与第一频谱拼接电路耦接的情况下，数字预失真器可以用于直接与基带处理器耦接，以便用于接收来自基带处理器的第一单频信号(如图8中的信号X1)，并根据第二预失真系数对该第一单频信号进行预失真，得到预失真的单频信号(如图8中的信号X1’)，以使射频前端电路中的功率放大器的输出信号满足线性指标。此时，预失真的单频信号也不需要再经过频谱分离电路，因此该数字预失真器也可以直接用于与射频前端电路耦接，以便将预失真的单频信号传输至射频前端电路进行处理，并调制为射频信号发射。
需要说明的是，上述第二预失真系数与上述第一预失真系数类似，是根据功率放大器的输入信号和输出信号来确定的，且第一单频信号可以视为功率放大器的输入信号。因此，对于上述模型提取电路，该模型提取电路也可以用于直接与基带处理器和射频前端电路耦接，以便接收来自基带处理器的第一单频信号和来自射频前端电路的功率放大器的输出端的第二单频信号，从而通过模型提取模块提取针对该单频信号的第二预失真系数。此时，数字预失真器还可以与模型提取电路耦接，用于接收第二预失真系数，以便根据第二预失真系数对第一单频信号进行预失真。这样，可以满足对单频信号进行兼容处理，简化系统架构，降低系统功耗。
这样一来，通过上述实施方式，本公开的实施例提供的数字预失真电路可以利用单频数字预失真器(即单频DPD模型)实现多频信号和单频信号的兼容处理，从而使得系统架构更加简化，不仅可以提高数据速率，还能够降低系统功耗。
需要说明的是，上述数字预失真电路中的各个电路模块可以实现为硬件电路，也可以使用软件来实现，或者各个电路模块可以部分采用硬件电路来实现，而部分采用软件算法实现。例如，可以将第一频谱拼接电路和数字预失真器采用硬件电路实现，而其他电路模块(例如模型 提取电路、频谱分离电路)采用软件算法实现。因此，本公开实施例针对数字预失真电路中各个电路模块的实现方式，不做特殊限制。
另外，本公开一些实施例还提供了一种射频处理电路。该射频处理电路包括上述实施例中的任意一种数字预失真电路以及射频前端电路。
基于图7所示的数字预失真电路，射频前端电路包括发射通路。发射通路，用于与频谱分离电路耦接，且用于将频谱分离电路输出的预失真的多频信号进行调制形成多频射频信号。功率放大器PA位于发射通路中，用于将多频射频信号进行放大后经由天线发射。
示例性地，为了对预失真后的多频信号进行处理，上述发射通路包括多个第一信号处理通道，且所述第一信号处理通道的数量与多频信号的频带数量相等或者大于多频信号的频带数量。例如，如图9所示，若多频信号为双频信号，则发射通路包括两个或更多个第一信号处理通道。又例如，如图10所示，若多频信号的为三频信号，则发射通路包括三个或更多个第一信号处理通道。
通常情况下，来自基带处理器的多频信号为数字信号，而经过天线发射的射频信号为模拟信号，因此每一个第一信号处理通道均包括数模转换器(digital-to-analog converter，DAC)和本振(例如图9中的f1和f2，或者图10中的f1、f2和f3)。数模转换器用于将预失真后的多频信号中各个频带的信号进行数模转换，变为模拟信号，本振用于对模拟信号进行上变频以形成射频信号。如图9和图10所示，经过本振进行上变频处理后的各个频带的射频信号可以通过加法器进行合并，输入至功率放大器PA进行放大后，经由天线向外辐射。
为了获取功率放大器的输出信号用于提取预失真系数，该射频前端电路还包括反馈通路。该反馈通路用于与所述第二频谱拼接电路耦接，所述反馈电路用于对所述功率放大器的输出信号进行处理，并反馈至所述第二频谱拼接电路。
示例性地，上述反馈通路包括两个或更多个第二信号处理通道，且所述第二信号处理通道的数量与所述第三信号的频带数量相等或者大于所述第三信号的频带数量。例如，如图9所示，若多频信号为双频信号，则反馈通路包括两个或更多个第二信号处理通道。又例如，如图10所示，若多频信号的为三频信号，则反馈通路包括三个或更多个第二信号处理通道。
通常情况下，多频射频信号为模拟信号，而数字预失真器是对数字信号进行预失真处理，因此需要将功率放大器输出的模拟信号转换为数字信号。因此每一个第二信号处理通道均包括本振(例如图9中的f1和f2，或者图10中的f1、f2和f3)、带通滤波器和模数转换器(analog-to-digital converter，ADC)。在反馈通道中，第二信号处理通道中的本振用于对功率放大器PA输出的模拟信号进行下变频，形成基带信号。经过下变频后的基带信号可以通过滤波器(如带通滤波器)形成各个频带信号，各个频带信号再通过模数转换器进行模数转换得到数字信号，传输给第二频谱拼接模块，以通过频谱拼接的方式形成上述第二宽带信号。
当然，在来自基带处理器的信号为单频信号时，发射通路可以直接与数字预失真器耦接，以便接收预失真后的单频信号。类似地，反馈通路也可以直接与模型提取电路耦接，以便将对功率放大器的输出端的信号处理为单频信号后，反馈给模型提取模块以提取预失真系数。
可以理解的是，针对该射频处理电路的技术效果可以参照上述数字预失真电路的技术效果，此处不再赘述。
本公开一些实施例还提供一种射频芯片。该射频芯片包括封装结构、以及封装于封装结构 内的如上述任意一个实施例所述的射频处理电路。该封装结构可以为封装基板、硅基转接板(interposer)等。
本公开一些实施例还提供一种数字预失真方法。该数字预失真方法可以应用于射频处理电路，该射频处理电路包括射频前端电路，射频前端电路包括功率放大器PA。如图11所示，该数字预失真方法包括以下步骤。
S1101，接收第一多频信号，该第一多频信号包括两个或更多个频带的信号。
关于多频信号的描述，可以参照上述数字预失真电路中的相关描述，此处不再赘述。
S1102，将第一多频信号进行频谱拼接得到第一宽带信号，该第一宽带信号的带宽小于第一多频信号中的最大频带与最小频带之间的频率间隔。
示例性地，可以对第一多频信号中各个频带的信号进行上采样，并通过对上采样得到的信号中的各个频带的信号进行上变频或下变频，以缩小相邻两个频带信号之间的频率间隔；通过对缩小频率间隔后的各个频带信号进行频谱拼接得到第一宽带信号。对上采样得到的信号中的各频带信号进行上变频或下变频后，各频带信号之间的频率间隔满足功率放大器的N阶非线性，N为大于或等于3的奇数。实现过程可以参照上述数字预失真电路中的相关描述，此处不再赘述。
S1103，根据第一预失真系数对第一宽带信号进行预失真，得到预失真的宽带信号，该第一预失真系数是根据第一宽带信号以及功率放大器的非线性特性确定的。
实现过程可以参照上述数字预失真电路中的相关描述，此处不再赘述。
与上述数字预失真电路相对应，由于数字预失真器输入的信号为宽带信号，因此数字预失真器输出的信号也为宽带信号，该宽带信号是经过线性预失真后的宽带信号。然而，为了保证信号的正确传输，输入至射频前端电路中的信号需要是与第一多频信号的频带相同的多频信号，因此如图12所示，上述数字预失真方法还可以包括S1104。
S1104，将预失真的宽带信号进行频谱分离，得到预失真的多频信号，使得该预失真的多频信号经由功率放大器输出而满足线性指标。
示例性地，可以对预失真的宽带信号进行滤波，得到包括两个或更多个频带的多频信号；并对多频信号中各频带信号进行上变频或者下变频，使各个频带的信号的中心频点返回至第一多频信号中相应频带信号的中心频点；然后对第一多频信号中各频带信号进行下采样，得到预失真的多频信号。对预失真的宽度信号进行频谱分离的详细过程，可以参照上述数字预失真电路中的相关描述，此处不再赘述。
针对第一预失真系数，该第一预失真系数是根据第一宽带信号和第二宽带信号进行建模得到的，该第二宽带信号是对功率放大器输出的第二多频信号进行频谱拼接得到的。详细的过程可以参考上述关于数字预失真电路实施例中的描述，此处不再赘述。
此外，对于单频信号的预失真处理，如图13所示，上述数字预失真方法还可以包括S1301至S1302。
S1301，接收第一单频信号。
S1302，根据第二预失真系数对第一单频信号进行预失真，得到预失真的单频信号，第二预失真系数是根据第一单频信号以及功率放大器的非线性特性确定的。
第二预失真系数是根据第一单频信号以及功率放大器输出的第二单频信号进行建模而提 取到的。
关于S1301和S1302的详细过程，可以参照上述数字预失真电路中的相关描述，此处不再赘述。
需要说明的是，针对上述数字预失真方法中的各个步骤可以按照实际情况调整各个步骤的执行顺序，本公开实施例对上述方法的各步骤的执行顺序不做特殊限定。
此外，通常对于基带信号的发射会在一段时间内持续发射，因此上述方法过程可以循环执行。例如，当第一多频信号经过预失真并通过发射通道输入至功率放大器PA之后，反馈通路可以采集功率放大器PA输出的第一多频信号。然后再分别对第一多频信号以及功率放大器PA输出的第二多频信号分别进行频谱拼接，分别得到第一宽带信号和第二宽带信号。再根据第一宽带信号和第二宽带信号进行建模，以提取预失真系数。提取到的预失真系数，输入至数字预失真器，可以对第一宽带信号进行预失真，得到预失真后的宽带信号。接着，再对预失真后的宽带信号进行频谱分离，分别通过发射通道中的各个第一信号处理通路输入至功率放大器PA进行发射。在这个过程中可以观测功率放大器输出的多频信号的线性度，在不满足线性度要求的情况下，可以持续执行上述过程。
类似地，对于上述数字预失真方法的技术效果，可以参照上述数字预失真电路的技术效果，此处不再赘述。
本公开另一实施例提供一种计算机存储介质，该计算机存储介质包括计算机指令，当所述计算机指令在电子设备上运行时，使得电子设备执行上述方法实施例中的各个步骤。在一些实施例中，该计算机存储介质为非暂态计算机存储介质。
本公开另一实施例提供一种计算机程序产品，当所述计算机程序产品在计算机上运行时，使得所述计算机执行上述方法实施例中各个步骤。
通过以上的实施方式的描述，所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，仅以上述各功能模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即将装置的内部结构划分成不同的功能模块，以完成以上描述的全部或者部分功能。上述描述的系统，装置和单元的工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本公开实施例各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本公开实施例的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或处理器执行本公开各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：快闪存储器、移动硬盘、只读存储器、随机存取存储器、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本公开的具体实施方式，但本公开的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本公开揭露的技术范围内，相关变化或替换，都应涵盖在本公开的保护 范围之内。因此，本公开的保护范围应以所述权利要求的保护范围为准。

Claims (16)

  1. 一种数字预失真电路，包括：
    第一频谱拼接电路，用于将第一多频信号进行频谱拼接得到第一宽带信号；其中，所述第一多频信号包括两个或更多个频带的信号，且所述第一宽带信号的带宽小于所述第一多频信号中的最大频带与最小频带之间的频率间隔；
    数字预失真器，与所述第一频谱拼接电路耦接，所述数字预失真器用于接收所述第一宽带信号，并根据第一预失真系数对所述第一宽带信号进行预失真得到预失真的宽带信号；其中，所述第一预失真系数是根据所述第一宽带信号以及射频前端电路中的功率放大器的非线性特性确定的。
  2. 根据权利要求1所述的数字预失真电路，还包括：
    第二频谱拼接电路，用于与所述射频前端电路耦接，用于将所述功率放大器输出的第二多频信号进行频谱拼接得到第二宽带信号；所述第二宽带信号的带宽与所述第一宽带信号的带宽相同；
    模型提取电路，与所述第一频谱拼接电路和所述第二频谱拼接电路耦接，用于根据所述第一宽带信号和所述第二宽带信号进行建模，以提取所述第一预失真系数；所述数字预失真器还与所述模型提取电路耦接，用于接收所述第一预失真系数。
  3. 根据权利要求1或2所述的数字预失真电路，还包括：
    频谱分离电路，与所述数字预失真器耦接，用于将所述预失真的宽带信号频谱分离为预失真的多频信号，使得所述预失真的多频信号经由所述功率放大器输出后满足线性指标。
  4. 根据权利要求1所述的数字预失真电路，其中，所述数字预失真器选择性地耦接所述第一频谱拼接电路，在所述数字预失真器不与所述第一频谱拼接电路耦接的情况下，所述数字预失真器还用于接收第一单频信号，并根据第二预失真系数对所述第一单频信号进行预失真得到预失真的单频信号；所述第二预失真系数是根据所述第一单频信号与所述功率放大器的非线性特性确定的。
  5. 根据权利要求4所述的数字预失真电路，还包括：
    模型提取电路，用于接收所述第一单频信号和所述功率放大器输出的第二单频信号，并根据所述第一单频信号以及所述第二单频信号进行建模，以提取所述第二预失真系数；所述数字预失真器还与所述模型提取电路耦接，用于接收所述第二预失真系数。
  6. 根据权利要求1或2所述的数字预失真电路，其中，所述第一宽带信号是通过将所述第一多频信号中的各个频带的信号进行频移后得到的，所述第一多频信号中的各个频带的信号经过频移后的频率间隔满足所述功率放大器的N阶非线性,其中，N为大于或等于3的奇数。
  7. 一种射频处理电路，包括射频前端电路、以及如权利要求1至6中任一项所述的数字预失真电路，所述数字预失真电路与所述射频前端电路耦接。
  8. 一种射频芯片，包括封装结构，以及封装于所述封装结构内的如权利要求7所述的射频处理电路。
  9. 一种电子设备，其中，所述电子设备包括基带处理器、以及如权利要求7所述的射频处理电路或者如权利要求8所述的射频芯片。
  10. 一种数字预失真方法，其中，所述方法应用于射频处理电路，所述射频处理电路包括射频前端电路，所述射频前端电路包括功率放大器；所述方法还包括：
    接收第一多频信号；所述第一多频信号包括两个或更多个频带的信号；
    将所述第一多频信号进行频谱拼接得到第一宽带信号；所述第一宽带信号的带宽小于所述第一多频信号中的最大频带与最小频带之间的频率间隔；
    根据第一预失真系数对所述第一宽带信号进行预失真得到预失真的宽带信号；所述第一预失真系数是根据所述第一宽带信号以及所述功率放大器的非线性特性确定的。
  11. 根据权利要求10所述的方法，还包括：
    将所述预失真的宽带信号进行频谱分离，得到预失真的多频信号，使得所述预失真的多频信号经由所述功率放大器输出而满足线性指标。
  12. 根据权利要求10或11所述的方法，其中，所述第一预失真系数是根据所述第一宽带信号和第二宽带信号进行建模提取到的，所述第二宽带信号是对所述功率放大器输出的第二多频信号进行频谱拼接得到的。
  13. 根据权利要求10所述的方法，还包括：
    接收第一单频信号；
    根据第二预失真系数对所述第一单频信号进行预失真得到预失真的单频信号；所述第二预失真系数是根据所述第一单频信号以及所述功率放大器的非线性特性确定的。
  14. 根据权利要求13所述的方法，其中，所述第二预失真系数是根据所述第一单频信号以及所述功率放大器输出的第二单频信号进行建模而提取到的。
  15. 根据权利要求10或11所述的方法，其中，所述第一宽带信号是通过将所述第一多频信号中的各个频带的信号进行频移后得到的，所述第一多频信号中的各个频带的信号经过频移后的频率间隔满足所述功率放大器的N阶非线性，其中，N为大于或等于3的奇数。
  16. 一种计算机可读存储介质，包括计算机指令，当所述计算机指令在电子设备上运行时，使得所述电子设备执行如权利要求10-15任一项所述的方法。
PCT/CN2023/115425 2022-09-08 2023-08-29 数字预失真电路及方法、射频芯片以及电子设备 WO2024051520A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211096716.7 2022-09-08
CN202211096716.7A CN117674743A (zh) 2022-09-08 2022-09-08 数字预失真电路及方法、射频芯片以及电子设备

Publications (1)

Publication Number Publication Date
WO2024051520A1 true WO2024051520A1 (zh) 2024-03-14

Family

ID=90068830

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/115425 WO2024051520A1 (zh) 2022-09-08 2023-08-29 数字预失真电路及方法、射频芯片以及电子设备

Country Status (2)

Country Link
CN (1) CN117674743A (zh)
WO (1) WO2024051520A1 (zh)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101056288A (zh) * 2007-06-14 2007-10-17 中兴通讯股份有限公司 预失真模型装置和信号的预失真处理装置、系统及方法
KR20100040497A (ko) * 2008-10-10 2010-04-20 세원텔레텍 주식회사 다중대역 중계기
US20130094610A1 (en) * 2011-10-14 2013-04-18 Fadhel Ghannouchi Digital Multi-band Predistortion Linearizer with Nonlinear Subsampling Algorithm in the Feedback Loop
CN104040883A (zh) * 2011-09-09 2014-09-10 奥普蒂斯蜂窝技术有限责任公司 多频带发射机中的单一功率放大器的线性化
CN104883140A (zh) * 2015-06-03 2015-09-02 中国科学院微电子研究所 基于宽带射频功率放大器的数字预失真装置
CN109889166A (zh) * 2019-03-12 2019-06-14 北京邮电大学 一种基于时间交织采样的单反馈回路并发双频带数字预失真方法
CN110574288A (zh) * 2017-02-25 2019-12-13 纳诺塞米有限公司 多频带数字预失真器

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101056288A (zh) * 2007-06-14 2007-10-17 中兴通讯股份有限公司 预失真模型装置和信号的预失真处理装置、系统及方法
KR20100040497A (ko) * 2008-10-10 2010-04-20 세원텔레텍 주식회사 다중대역 중계기
CN104040883A (zh) * 2011-09-09 2014-09-10 奥普蒂斯蜂窝技术有限责任公司 多频带发射机中的单一功率放大器的线性化
US20130094610A1 (en) * 2011-10-14 2013-04-18 Fadhel Ghannouchi Digital Multi-band Predistortion Linearizer with Nonlinear Subsampling Algorithm in the Feedback Loop
CN104883140A (zh) * 2015-06-03 2015-09-02 中国科学院微电子研究所 基于宽带射频功率放大器的数字预失真装置
CN110574288A (zh) * 2017-02-25 2019-12-13 纳诺塞米有限公司 多频带数字预失真器
CN109889166A (zh) * 2019-03-12 2019-06-14 北京邮电大学 一种基于时间交织采样的单反馈回路并发双频带数字预失真方法

Also Published As

Publication number Publication date
CN117674743A (zh) 2024-03-08

Similar Documents

Publication Publication Date Title
US12089232B2 (en) Wireless devices and systems including examples of configuration modes for baseband units and remote radio heads
US9774364B2 (en) Interference phase estimate system and method
EP3170265B1 (en) Cancelling crosstalk
US20180212633A1 (en) Cancelling Intermodulation Interference
CN115298978B (zh) 用于谐波干扰消除的基带芯片和无线通信方法
Campo et al. Cascaded spline-based models for complex nonlinear systems: Methods and applications
US12003225B2 (en) Method and apparatus for nonlinear signal processing
CN115865594A (zh) 一种基于fdd系统的无源交调校正方法及系统
CN112532271A (zh) 射频电路和电子设备
WO2022237395A1 (zh) 预失真处理方法及其装置、通信设备、存储介质
WO2024051520A1 (zh) 数字预失真电路及方法、射频芯片以及电子设备
JP2003174332A (ja) プリディストーション型増幅装置
EP3151500B1 (en) Frequency-selective quadrature baseband coupling cancellation
US20150055731A1 (en) Digital Transmitter With Sample Rate Digital Predistortion
WO2023116834A1 (zh) 一种非线性校正方法、装置及系统
CN116436539A (zh) 放大器非线性校准装置及方法
Amiri et al. Estimation of crossover DPD using orthogonal polynomials in fixed point arithmetic
CN112543157B (zh) 用于无线发射机中的数字预失真dpd的设备和方法
US20240291508A1 (en) Systems and methods for multiband linearization architecture using kernel regression
Shokair et al. Wide band digital predistortion using iterative feedback decomposition
Tarver et al. Low-complexity, multi sub-band digital predistortion: Novel algorithms and SDR verification
KR101818834B1 (ko) 광 무선 통신 시스템의 등화기 및 상기 등화기 제어 방법
JP2015231056A (ja) 電力増幅装置
CN117546414A (zh) 级间匹配网络衰减器
CN116192280A (zh) 量子计算多比特操控及读取系统和信号处理方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23862215

Country of ref document: EP

Kind code of ref document: A1