WO2016154933A1 - 数字预失真校正方法及装置 - Google Patents

数字预失真校正方法及装置 Download PDF

Info

Publication number
WO2016154933A1
WO2016154933A1 PCT/CN2015/075606 CN2015075606W WO2016154933A1 WO 2016154933 A1 WO2016154933 A1 WO 2016154933A1 CN 2015075606 W CN2015075606 W CN 2015075606W WO 2016154933 A1 WO2016154933 A1 WO 2016154933A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
input signal
dynamic
adjustment factor
amplifier
Prior art date
Application number
PCT/CN2015/075606
Other languages
English (en)
French (fr)
Inventor
肖宇翔
朱尔霓
尤览
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to CN201580078331.XA priority Critical patent/CN107431495B/zh
Priority to PCT/CN2015/075606 priority patent/WO2016154933A1/zh
Publication of WO2016154933A1 publication Critical patent/WO2016154933A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B1/00Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/02Transmitters
    • H04B1/04Circuits

Definitions

  • the present invention relates to the field of communications, and in particular, to a digital predistortion correction method and apparatus.
  • a power amplifier is an important component of a transmitting device in a communication system.
  • power amplifiers need to operate at high efficiency while maintaining high linearity, while efficiency and linearity often contradict each other in the design of power amplifiers.
  • Digital Pre-Distortion (DPD) technology is a key technology that specifically compensates for the nonlinear characteristics of power amplifiers. The technique establishes a DPD model according to the characteristics of the power amplifier, and performs predistortion processing on the current input signal in the digital domain, so that the processed predistortion signal can cancel the nonlinear characteristic of the power amplifier itself after entering the power amplifier.
  • DPD technology the power amplifier can be operated in a high efficiency state while meeting the specification requirements in terms of linearity.
  • the power of the output signal of the power amplifier is often dynamically adjusted in real time as the traffic volume changes.
  • the power variation of such an output signal is often accompanied by a change in the characteristics of the power amplifier.
  • the nonlinear correction for the power dynamic change scene of the output signal is often realized by establishing a lookup table model related to the power value.
  • the lookup table model saves the correspondence between the model parameters and the power values of the preset DPD model in a table.
  • the corresponding model may be firstly found according to the power value of the current input signal.
  • the parameter determines a DPD model corresponding to the current input signal according to the acquired model parameters, and performs predistortion correction on the current input signal according to the DPD model.
  • the lookup table model needs to calculate and store the model parameters of the DPD model covering a sufficient power range in order to meet the correction requirements of different input signals, the amount of data calculated and stored is very large. Therefore, the acquisition process of model parameters is more complicated and the correction efficiency is lower.
  • the present invention provides a digital predistortion correction method and device.
  • the technical solution is as follows:
  • a digital predistortion correction method comprising:
  • a dynamic linear model and a dynamic nonlinear model are included, wherein the dynamic linear model is used to indicate a linear characteristic in a variable of an input signal, the dynamic nonlinear model being used to indicate a nonlinear characteristic in a variable of an input signal
  • the preselected input signal is at least two signals in the preset input signal group, and the preset input signal group includes a plurality of input signals having different power values;
  • the adjusting factor of a dynamic linear model of the preselected input signal and an adjustment factor of a dynamic nonlinear model of the preselected input signal establish an amplifier model of a current input signal ,include:
  • the N represents the number of sampling points of each input signal in the preset input signal group
  • the L1 represents the number of model coefficients of the static model
  • the L2 represents the model coefficient of the dynamic model.
  • Model coefficients based on the static model of the current input signal And model coefficients of the dynamic model Substituting the amplifier model formula, establishing an amplifier model corresponding to the current input signal to obtain the current input signal.
  • model coefficients of the dynamic model include:
  • an adjustment factor of the dynamic linear model of the preset input signal group and an adjustment factor of the dynamic nonlinear model of the preset input signal group are determined by substituting into the amplifier model formula to obtain the preset input signal group.
  • model coefficient of the static model of the preset input signal group as the model coefficient of the static model of the current input signal
  • model coefficient of the dynamic model of the preset input signal group as the model coefficient of the dynamic model of the current input signal
  • the preselected input signal is a first input signal having the highest power value and a second input signal having the lowest power value in the preset input signal group,
  • the determining an adjustment factor of the dynamic linear model of the preselected input signal in the amplifier model based on the preselected input signal comprises:
  • the linear adjustment factor formula is:
  • ⁇ i represents a parameter value of a dynamic linear characteristic parameter of the first input signal or a parameter value of a dynamic linear characteristic parameter of the second input signal
  • the i represents a power amount of the first input signal a power level of the second input signal
  • the ⁇ 1 representing a parameter value of a dynamic linear characteristic parameter of the first input signal
  • An adjustment factor representing a dynamic linear model of the first input signal or an adjustment factor of a dynamic linear model of the second input signal.
  • the dynamic linear characteristic parameter is an average power of an output signal corresponding to the preset input signal group in the power amplifier or a gain of an output signal corresponding to the preset input signal group in the power amplifier.
  • Determining, by the preselected input signal, an adjustment factor of a dynamic nonlinear model of the preselected input signal in the amplifier model comprising:
  • an adjustment factor of the dynamic linear model of the preset input signal group including:
  • the r indicates that the power level of the input signal in the preset input signal group is smaller than the power level M of the first input signal, and is greater than the power level m of the second input signal,
  • An adjustment factor representing a dynamic linear model of the first input signal An adjustment factor representing a dynamic linear model of the second input signal
  • the w (r) representing a weighting factor
  • the weighting factor w (r) being determined according to a weight formula, the weighting formula is:
  • P r represents a power value of an input signal of a power level r
  • P M represents a power value of the first input signal of a power level of M
  • the method After the digital pre-distortion correction of the current input signal according to the DPD model, the method also includes:
  • the DPD model of the current input signal is updated to obtain an updated amplifier model.
  • the state change of the power amplifier is device aging, temperature fluctuation, or bias voltage variation.
  • the updating the amplifier model of the current input signal to obtain an updated amplifier model includes:
  • model coefficients of the dynamic model of the updated amplifier model including:
  • model coefficients of the static model of the updated amplifier model according to the amplifier model of the current input signal including:
  • a difference between a model coefficient of a static model of the input signal after changing the state of the power amplifier and a model coefficient of a static model of the input signal before the change is used as a second difference;
  • a sum of a model coefficient of the static model of the current input signal and the second difference is used as a model coefficient of a static model of the updated amplifier model.
  • a digital predistortion correction apparatus comprising:
  • a first determining unit configured to determine, according to the preselected input signal, an adjustment factor of a dynamic linear model of the preselected input signal in an amplifier model, the amplifier model being used to indicate a static model of the output signal and the input signal, and a dynamic model of the input signal
  • the dynamic model includes a dynamic linear model and a dynamic nonlinear model, wherein the dynamic linear model is used to indicate a linear characteristic in a variable of the input signal, the dynamic nonlinear model being used to indicate a variable of the input signal
  • the preselected input signal is at least two signals in the preset input signal group, and the preset input signal group includes a plurality of input signals having different power values;
  • a second determining unit configured to determine, according to the preselected input signal, an adjustment factor of a dynamic nonlinear model of the preselected input signal in the amplifier model
  • a establishing unit configured to establish an amplifier model of the current input signal according to an adjustment factor of the dynamic linear model of the preselected input signal and an adjustment factor of a dynamic nonlinear model of the preselected input signal;
  • a processing unit configured to obtain a digital pre-distortion DPD model of the current input signal according to an amplifier model of the current input signal
  • a correcting unit configured to perform digital predistortion correction on the current input signal according to the DPD model of the current input signal.
  • the establishing unit includes:
  • a first determining module configured to determine a model of the static model of the current input signal by using an amplifier model formula according to an adjustment factor of a dynamic linear model of the preselected input signal, an adjustment factor of a dynamic nonlinear model of the preselected input signal coefficient And model coefficients of the dynamic model
  • the amplifier model formula is:
  • the N represents the number of sampling points of each input signal in the preset input signal group
  • the L1 represents the number of model coefficients of the static model
  • the L2 represents the model coefficient of the dynamic model.
  • a first generation module for using a model coefficient of a static model of the current input signal And model coefficients of the dynamic model Substituting the amplifier model formula to obtain an amplifier model corresponding to the current input signal.
  • the first determining module includes:
  • a first determining submodule configured to determine an adjustment factor of a dynamic linear model of the preset input signal group according to an adjustment factor of a dynamic linear model of the preselected input signal
  • a second determining submodule configured to determine an adjustment factor of a dynamic nonlinear model of the preset input signal group according to an adjustment factor of a dynamic nonlinear model of the preselected input signal
  • Substituting a sub-module for substituting an adjustment factor of a dynamic linear model of the preset input signal group and an adjustment factor of a dynamic nonlinear model of the preset input signal group into the amplifier model formula to obtain the preset input The model coefficients of the static model of the signal group and the model coefficients of the dynamic model of the preset input signal group;
  • a first processing submodule configured to use a model coefficient of a static model of the preset input signal group as a model coefficient of a static model of the current input signal Using the model coefficient of the dynamic model of the preset input signal group as the model coefficient of the dynamic model of the current input signal
  • the preselected input signal is a first input signal having the highest power value and a second input signal having the lowest power value in the preset input signal group.
  • the first determining unit includes:
  • a second determining module configured to determine an adjustment factor of the dynamic linear model of the first input signal according to a dynamic linear characteristic parameter and a linear adjustment factor formula of the first input signal
  • a third determining module configured to determine an adjustment factor of the dynamic linear model of the second input signal according to a parameter value of the dynamic linear characteristic parameter of the second input signal and a linear adjustment factor formula
  • the linear adjustment factor formula is:
  • ⁇ i represents a parameter value of a dynamic linear characteristic parameter of the first input signal or a parameter value of a dynamic linear characteristic parameter of the second input signal
  • the i represents a power amount of the first input signal a power level of the second input signal
  • the ⁇ 1 representing a parameter value of a dynamic linear characteristic parameter of the first input signal
  • An adjustment factor representing a dynamic linear model of the first input signal or an adjustment factor of a dynamic linear model of the second input signal.
  • the dynamic linear characteristic parameter is an average power of an output signal corresponding to the preset input signal group in the power amplifier or a gain of an output signal corresponding to the preset input signal group in the power amplifier.
  • the second determining unit includes:
  • a fourth determining module configured to determine a parameter value of a dynamic nonlinear characteristic parameter of the preselected input signal according to a parameter value and a nonlinear characteristic formula of the dynamic linear characteristic parameter of the preselected input signal, where the nonlinear characteristic formula is:
  • a fifth determining module configured to determine, according to a parameter value of a dynamic nonlinear characteristic parameter of the preselected input signal and a nonlinear adjustment factor formula, an adjustment factor of a dynamic nonlinear model of the preselected input signal, the nonlinear adjustment factor formula for:
  • the first determining sub-module is specifically configured to: determine, according to an adjustment factor of the dynamic linear model of the preselected input signal, a first interpolation formula to determine an adjustment factor of a dynamic linear model of the preset input signal group
  • the first interpolation formula is:
  • the r indicates that the power level of the input signal in the preset input signal group is smaller than the power level M of the first input signal, and is greater than the power level m of the second input signal,
  • An adjustment factor representing a dynamic linear model of the first input signal An adjustment factor representing a dynamic linear model of the second input signal
  • the w (r) representing a weighting factor
  • the weighting factor w (r) being determined according to a weight formula, the weighting formula is:
  • P r represents a power value of an input signal of a power level r
  • P M represents a power value of the first input signal of a power level of M
  • the second determining sub-module is specifically configured to: determine, according to an adjustment factor of the dynamic nonlinear model of the preselected input signal, a second interpolation formula to determine an adjustment factor of the dynamic nonlinear model of the preset input signal group
  • the second interpolation formula is:
  • the digital predistortion correction apparatus further includes:
  • an updating unit configured to update an amplifier model of the current input signal to obtain an updated amplifier model when a state of the power amplifier changes.
  • the state change of the power amplifier is device aging, temperature fluctuation, or bias voltage variation.
  • the updating unit includes:
  • a first acquiring module configured to acquire, according to an amplifier model of the current input signal, model coefficients of a dynamic model of the updated amplifier model
  • a second acquiring module configured to acquire, according to an amplifier model of the current input signal, model coefficients of a static model of the updated amplifier model
  • a second generation module for substituting the model coefficients of the dynamic model of the updated amplifier model and the model coefficients of the static model of the updated amplifier model into the amplifier model to obtain the updated amplifier model.
  • the first acquiring module includes:
  • a first difference submodule configured to use a difference between a model coefficient of a dynamic model of the input signal after the state change of the power amplifier and a model coefficient of a dynamic model of the input signal before the change as a first difference
  • a second processing submodule configured to use a sum of a model coefficient of the dynamic model of the current input signal and the first difference as a model coefficient of a dynamic model of the updated amplifier model.
  • the second obtaining module includes:
  • a second difference submodule wherein a difference between a model coefficient of a static model of the input signal after changing the state of the power amplifier and a model coefficient of a static model of the input signal before the change is used as a second difference value;
  • a third processing submodule configured to use a sum of a model coefficient of the static model of the current input signal and the second difference as a model coefficient of a static model of the updated amplifier model.
  • a digital predistortion correction apparatus comprising:
  • a processor for determining an adjustment factor of a dynamic linear model of the preselected input signal in the amplifier model according to the preselected input signal, the amplifier model for indicating a relationship between the output signal and a static model of the input signal, and a dynamic model of the input signal
  • the dynamic model includes a dynamic linear model for indicating a linear characteristic in a variable of an input signal, and a dynamic nonlinear model for indicating a non-variation in a variable of the input signal a linear characteristic, the preselected input signal is at least two signals in a preset input signal group, and the preset input signal group includes a plurality of input signals having different power values;
  • the processor is further configured to determine an adjustment factor of a dynamic nonlinear model of the preselected input signal in the amplifier model according to the preselected input signal;
  • the processor is further configured to establish an amplifier model of the current input signal according to an adjustment factor of the dynamic linear model of the preselected input signal and an adjustment factor of a dynamic nonlinear model of the preselected input signal;
  • the processor is further configured to obtain a digital pre-distortion DPD model of the current input signal according to an amplifier model of the current input signal;
  • the processor is further configured to perform digital predistortion correction on the current input signal according to a DPD model of the current input signal.
  • the processor is specifically configured to:
  • the N represents the number of sampling points of each input signal in the preset input signal group
  • the L1 represents the number of model coefficients of the static model
  • the L2 represents the model coefficient of the dynamic model.
  • Model coefficients of the static model of the current input signal And model coefficients of the dynamic model Substituting the amplifier model formula to obtain an amplifier model corresponding to the current input signal.
  • the processor is specifically configured to:
  • model coefficient of the static model of the preset input signal group as the model coefficient of the static model of the current input signal
  • model coefficient of the dynamic model of the preset input signal group as the model coefficient of the dynamic model of the current input signal
  • the preselected input signal is a first input signal having the highest power value and a second input signal having the lowest power value in the preset input signal group.
  • the processor is specifically used to:
  • the linear adjustment factor formula is:
  • ⁇ i represents a parameter value of a dynamic linear characteristic parameter of the first input signal or a parameter value of a dynamic linear characteristic parameter of the second input signal
  • the i represents a power amount of the first input signal a power level of the second input signal
  • the ⁇ 1 representing a parameter value of a dynamic linear characteristic parameter of the first input signal
  • An adjustment factor representing a dynamic linear model of the first input signal or an adjustment factor of a dynamic linear model of the second input signal.
  • the dynamic linear characteristic parameter is an average power of an output signal corresponding to the preset input signal group in the power amplifier or a gain of an output signal corresponding to the preset input signal group in the power amplifier.
  • the processor is specifically configured to:
  • the processor is specifically configured to:
  • the r indicates that the power level of the input signal in the preset input signal group is smaller than the power level M of the first input signal, and is greater than the power level m of the second input signal,
  • An adjustment factor representing a dynamic linear model of the first input signal An adjustment factor representing a dynamic linear model of the second input signal
  • the w (r) representing a weighting factor
  • the weighting factor w (r) being determined according to a weight formula, the weighting formula is:
  • P r represents a power value of an input signal of a power level r
  • P M represents a power value of the first input signal of a power level of M
  • the processor is further configured to determine, according to an adjustment factor of the dynamic nonlinear model of the preselected input signal, a second interpolation formula to determine an adjustment factor of the dynamic nonlinear model of the preset input signal group
  • the second interpolation formula is:
  • the processor is further configured to:
  • the amplifier model of the current input signal is updated to obtain an updated amplifier model.
  • the state change of the power amplifier is device aging, temperature fluctuation, or bias voltage variation.
  • the processor is specifically configured to:
  • the processor is also specifically configured to:
  • the processor is also specifically configured to:
  • a difference between a model coefficient of a static model of the input signal after changing the state of the power amplifier and a model coefficient of a static model of the input signal before the change is used as a second difference;
  • a sum of a model coefficient of the static model of the current input signal and the second difference is used as a model coefficient of a static model of the updated amplifier model.
  • the invention provides a digital predistortion correction method and device, which can determine an adjustment factor of a dynamic linear model of the preselected input signal and an adjustment factor of the dynamic nonlinear model according to the preselected input signal, and then according to the preselected input
  • the adjustment factor of the dynamic linear model of the signal, the adjustment factor of the dynamic nonlinear model of the preselected input signal establishes the amplifier model of the current input signal, and then obtains the DPD model of the current input signal according to the amplifier model of the current input signal, and finally the current DPD model according to the current DPD model
  • the input signal is digitally predistort corrected.
  • FIG. 1 is a basic working principle diagram of an existing DPD technology according to an embodiment of the present invention.
  • FIG. 2 is a flowchart of a digital predistortion correction method according to an embodiment of the present invention
  • FIG. 3 is a flowchart of another digital predistortion correction method according to an embodiment of the present invention.
  • FIG. 4 is a schematic diagram of an amplifier model according to an embodiment of the present invention.
  • FIG. 5 is a flowchart of a method for determining an adjustment factor of a dynamic linear model of a preselected input signal in an amplifier model according to a preselected input signal according to an embodiment of the present invention
  • FIG. 6 is a flowchart of a method for determining an adjustment factor of a dynamic nonlinear model of a preselected input signal in an amplifier model according to a preselected input signal according to an embodiment of the present invention
  • FIG. 7 is a schematic diagram of determining model coefficients of a static model corresponding to a preselected input signal and model coefficients of a dynamic model according to an embodiment of the present invention
  • FIG. 8 is a flowchart of a method for establishing an amplifier model of a current input signal according to an adjustment factor of a dynamic linear model of a preselected input signal and an adjustment factor of a dynamic nonlinear model of a preselected input signal according to an embodiment of the present invention
  • FIG. 9 is an adjustment factor of a dynamic linear model according to a preselected input signal and an adjustment factor of a dynamic nonlinear model of a preselected input signal according to an embodiment of the present invention, and determining a model coefficient of a static model of the current input signal by using an amplifier model formula and Flow chart of the model coefficient method of the dynamic model;
  • FIG. 10 is a schematic diagram of determining an adjustment factor of a dynamic linear model and an adjustment factor of a dynamic nonlinear model corresponding to an input signal that is not trained in a preset input signal group according to an embodiment of the present disclosure
  • FIG. 11 is a flowchart of a method for updating an amplifier model obtained by updating an amplifier model of a current input signal according to an embodiment of the present invention
  • FIG. 12 is a flowchart of a method for obtaining a model coefficient of a dynamic model of an updated amplifier model according to an embodiment of the present invention
  • FIG. 13 is a flowchart of a method for obtaining a model coefficient of a static model of an updated amplifier model according to an embodiment of the present invention
  • FIG. 14 is a schematic diagram of model coefficient update of an amplifier model according to an embodiment of the present invention.
  • 15 is a schematic structural diagram of a digital predistortion correction apparatus according to an embodiment of the present invention.
  • 16 is a schematic structural diagram of another digital predistortion correction apparatus according to an embodiment of the present invention.
  • 17 is a schematic structural diagram of a establishing unit of a digital predistortion correction apparatus according to an embodiment of the present invention.
  • FIG. 18 is a schematic structural diagram of a first determining module of a digital predistortion correction apparatus according to an embodiment of the present disclosure
  • FIG. 19 is a schematic structural diagram of a first determining unit of a digital predistortion correction apparatus according to an embodiment of the present invention.
  • FIG. 20 is a second determining unit of a digital predistortion correction apparatus according to an embodiment of the present invention. Schematic;
  • 21 is a schematic structural diagram of an update unit of a digital predistortion correction apparatus according to an embodiment of the present invention.
  • FIG. 22 is a schematic structural diagram of a first acquiring module of a digital predistortion correction apparatus according to an embodiment of the present disclosure
  • FIG. 23 is a schematic structural diagram of a second acquiring module of a digital predistortion correction apparatus according to an embodiment of the present invention.
  • FIG. 24 is a schematic structural diagram of still another digital predistortion correction apparatus according to an embodiment of the present invention.
  • DPD technology is a key technology that compensates for the nonlinearity of the power amplifier in the digital domain, allowing the power amplifier to operate in a highly efficient saturation state without losing linearity.
  • the basic working principle of the DPD technology is to establish a DPD module 01 in the digital baseband, and pre-distort the input signal before entering the power amplifier (English: Power Amplifier; PA: 02).
  • the DPD module 01 The model is the DPD model, and the PA02 model is the amplifier model. If the DPD model is the inverse function of the amplifier model, the input signal will be linearly amplified after passing through the cascaded DPD modules 01 and PA02, thereby avoiding the input signal passing through. The output signal after PA02 is distorted.
  • the abscissa X represents the input power of the input signal
  • the ordinate Y represents the output power of the input signal
  • the first power graph on the left represents the power curve of the input signal
  • the second power graph represents the generation of the DPD module 01.
  • the third power graph represents the power curve of the output signal output from PA02 after the predistortion correction of the input signal.
  • the embodiment of the invention provides a digital predistortion correction method. As shown in FIG. 2, the method includes:
  • Step 101 Determine, according to the preselected input signal, an adjustment factor of a dynamic linear model of the preselected input signal in the amplifier model, the amplifier model is used to indicate a static model of the output signal and the input signal,
  • the relationship between the dynamic models of the input signals, the dynamic model includes a dynamic linear model and a dynamic nonlinear model, wherein the dynamic linear model is used to indicate linear characteristics in the variables of the input signal, and the dynamic nonlinear model is used to indicate the variables of the input signal.
  • the nonlinear characteristic, the preselected input signal is at least two signals in the preset input signal group, and the preset input signal group includes a plurality of input signals having different power values.
  • Step 102 Determine an adjustment factor of a dynamic nonlinear model of the preselected input signal in the amplifier model according to the preselected input signal.
  • Step 103 Establish an amplifier model of the current input signal according to an adjustment factor of the dynamic linear model of the preselected input signal and an adjustment factor of the dynamic nonlinear model of the preselected input signal.
  • Step 104 Obtain a DPD model of the current input signal according to an amplifier model of the current input signal.
  • Step 105 Perform digital predistortion correction on the current input signal according to the DPD model of the current input signal.
  • the digital predistortion correction method provided by the embodiment of the present invention can determine the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model according to the preselected input signal, and then The amplifier model of the current input signal is established according to the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model of the preselected input signal, and then the DPD model of the current input signal is obtained according to the amplifier model of the current input signal, and finally according to the DPD.
  • the model performs digital predistortion correction on the current input signal.
  • step 103 specifically includes:
  • the model coefficient of the static model of the current input signal is determined by using the amplifier model formula.
  • model coefficients of the dynamic model The amplifier model formula is:
  • N represents the number of sampling points of each input signal in the preset input signal group
  • L1 represents the number of model coefficients of the static model
  • L2 represents the number of model coefficients of the dynamic model
  • r represents the preset input signal group.
  • the power level of the input signal, r is an integer greater than or equal to 1,
  • An output signal group representing a preset input signal group, a matrix of N ⁇ 1, a static model representing a preset input signal group, a matrix of N ⁇ L1, a dynamic model representing a preset input signal group, a matrix of N ⁇ L2, a dynamic linear model representing a preset input signal group, a dynamic nonlinear model representing a preset input signal group,
  • model coefficient of the static model of the current input signal is determined by using the amplifier model formula
  • model coefficients of the dynamic model include:
  • the adjustment factor of the dynamic linear model of the preset input signal group is determined according to the adjustment factor of the dynamic linear model of the preselected input signal; the adjustment of the dynamic nonlinear model of the preset input signal group is determined according to the adjustment factor of the dynamic nonlinear model of the preselected input signal Factor; the adjustment factor of the dynamic linear model of the preset input signal group and the adjustment factor of the dynamic nonlinear model of the preset input signal group are substituted into the amplifier model formula, and the model coefficients and presets of the static model for determining the preset input signal group are obtained.
  • the model coefficient of the dynamic model of the preset input signal group is used as the model coefficient of the dynamic model of the current input signal
  • the preselected input signal is at least two signals in the preset input signal group. Taking the first input signal with the highest power value and the second input signal with the lowest power value in the preset input signal group as an example, the adjustment factor of the dynamic linear model of the preselected input signal in the amplifier model is determined according to the preselected input signal, including:
  • ⁇ i represents a parameter value of a dynamic linear characteristic parameter of the first input signal or a parameter value of a dynamic linear characteristic parameter of the second input signal
  • i represents a power magnitude of the first input signal or a power magnitude of the second input signal
  • ⁇ 1 represents the parameter value of the dynamic linear characteristic parameter of the first input signal
  • An adjustment factor representing the dynamic linear model of the first input signal or an adjustment factor of the dynamic linear model of the second input signal.
  • the dynamic linear characteristic parameter may be the average power of the output signal corresponding to the preset input signal group in the power amplifier or the output signal corresponding to the preset input signal group in the power amplifier.
  • the adjustment factor of the dynamic nonlinear model of the preselected input signal in the amplifier model is determined according to the preselected input signal, including:
  • Determining a parameter value of a dynamic nonlinear characteristic parameter of the preselected input signal according to a parameter value of the dynamic linear characteristic parameter of the preselected input signal and a nonlinear characteristic formula, and the nonlinear characteristic formula may be:
  • a parameter value indicating a dynamic nonlinear characteristic parameter of the preselected input signal x (i) (n) represents a signal value of the preselected input signal, y (i) (n) represents an output signal corresponding to the preselected input signal, and ⁇ i represents the first The parameter value of the dynamic linear characteristic parameter of the input signal or the parameter value of the dynamic linear characteristic parameter of the second input signal.
  • the adjustment factor of the dynamic nonlinear model of the preselected input signal is determined according to the parameter value of the dynamic nonlinear characteristic parameter of the preselected input signal and the nonlinear adjustment factor formula, and the nonlinear adjustment factor formula is:
  • a parameter value representing a dynamic nonlinear characteristic parameter of the first input signal a parameter value representing a dynamic nonlinear characteristic parameter of the preselected input signal, An adjustment factor representing a dynamic nonlinear model of the first input signal or an adjustment factor of a dynamic nonlinear model of the second input signal.
  • the adjustment factor of the dynamic linear model of the preset input signal group is determined according to an adjustment factor of the dynamic linear model of the preselected input signal, including:
  • the first interpolation formula is used to determine the adjustment factor of the dynamic linear model of the preset input signal group
  • the first interpolation formula can be:
  • r indicates that the power level of the input signal in the preset input signal group is smaller than the power level M of the first input signal, and greater than the power level m of the second input signal
  • An adjustment factor representing the dynamic linear model of the first input signal An adjustment factor representing a dynamic linear model of the second input signal
  • w (r) represents a weighting factor
  • a weighting factor w (r) is determined according to a weighting formula, which may be:
  • P r represents the power value of the input signal of the power level r
  • P M represents the power value of the first input signal of the power level M.
  • the adjustment factor of the dynamic nonlinear model of the preset input signal group is determined according to an adjustment factor of the dynamic nonlinear model of the preselected input signal, including:
  • the second interpolation formula is used to determine the adjustment factor of the dynamic nonlinear model of the preset input signal group.
  • the second interpolation formula can be:
  • An adjustment factor representing a dynamic nonlinear model of the first input signal An adjustment factor representing a dynamic nonlinear model of the second input signal, w (r) representing a weighting factor.
  • the method may further include: updating the amplifier model of the current input signal to obtain an updated amplifier model when the state of the power amplifier changes.
  • the state change of the power amplifier is device aging, temperature fluctuations, or bias voltage variations.
  • updating the amplifier model of the current input signal results in an updated amplifier model, including:
  • model coefficients of the dynamic model of the updated amplifier model according to an amplifier model of the current input signal; acquiring model coefficients of the static model of the updated amplifier model according to an amplifier model of the current input signal; and dynamics of the updated amplifier model
  • the model coefficients of the model and the model coefficients of the static model of the updated amplifier model are substituted into the amplifier model to obtain an updated amplifier model.
  • the model coefficients of the dynamic model of the updated amplifier model are obtained according to an amplifier model of the current input signal, including:
  • the difference between the model coefficient of the dynamic model of the input signal after changing the state of the power amplifier and the model coefficient of the dynamic model of the input signal before the change is taken as the first difference; the model coefficient of the dynamic model of the current input signal is the first difference
  • the sum of the values is used as the model coefficient of the dynamic model of the updated amplifier model.
  • model coefficients of the static model of the updated amplifier model based on the amplifier model of the current input signal including:
  • the model coefficient of the static model of the input signal after changing the state of the power amplifier and before the change The difference between the model coefficients of the static model of the input signal is taken as the second difference; the sum of the model coefficients of the static model of the current input signal and the second difference is used as the model coefficient of the static model of the updated amplifier model.
  • the digital predistortion correction method provided by the embodiment of the present invention can determine the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model according to the preselected input signal, and then The amplifier model of the current input signal is established according to the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model of the preselected input signal, and then the DPD model of the current input signal is obtained according to the amplifier model of the current input signal, and finally according to the DPD.
  • the model performs digital predistortion correction on the current input signal.
  • Another embodiment of the present invention provides a digital predistortion correction method. As shown in FIG. 3, the method includes:
  • Step 201 Establish an amplifier model.
  • the amplifier model formula is:
  • N represents the number of sampling points of each input signal in the preset input signal group
  • the preset input signal group includes a plurality of input signals having different power values
  • L1 represents the number of model coefficients of the static model
  • L2 represents the dynamic model.
  • r represents the power level of the input signal in the preset input signal group, r is an integer greater than or equal to 1
  • An output signal group representing a preset input signal group, a matrix of N ⁇ 1, a static model representing a preset input signal group, a matrix of N ⁇ L1, a dynamic model representing a preset input signal group, a matrix of N ⁇ L2, a dynamic linear model representing a preset input signal group, a dynamic nonlinear model representing a preset input signal group,
  • An adjustment factor representing a dynamic linear model of a preset input signal group An adjustment factor representing a dynamic nonlinear model of a preset input signal group, a model coefficient representing a static model, Is a matrix of L1
  • the input signals in the preset input signal group are arranged in descending order of power values, in order: S1, S2, S3, S4, S5, S6, S7, S8, S9 and S10, then S1, S2
  • the power levels r corresponding to S3, S4, S5, S6, S7, S8, S9, and S10 are 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10, respectively.
  • the amplifier model includes three parts, a static model X (r) of a preset input signal group, and a dynamic linear model.
  • Dynamic nonlinear model It should be noted that the adjustment factor of the dynamic linear model in the model formula of the amplifier Adjustment factor of dynamic nonlinear model Model coefficient of static model And model coefficients of the dynamic model It is unknown.
  • Step 202 Determine an adjustment factor of a dynamic linear model of the preselected input signal in the amplifier model according to the preselected input signal.
  • the amplifier model is used to indicate the relationship between the output signal and the static model of the input signal, and the dynamic model of the input signal.
  • the dynamic model includes a dynamic linear model and a dynamic nonlinear model, wherein the dynamic linear model is used to indicate linearity in the variables of the input signal.
  • the characteristic, the dynamic nonlinear model is used to indicate a nonlinear characteristic in the variable of the input signal
  • the preselected input signal is at least two signals in the preset input signal group
  • the preset input signal group includes a plurality of input signals having different power values.
  • the preselected input signal may be the input signal with the highest power value and the input signal with the lowest power value in the preset input signal group, or may be The input signal with the highest power value, the input signal with the lowest power value, and the input signal corresponding to other power values are preset in the input signal group.
  • the input signals in the preset input signal group are arranged in descending order of power values, in order: S1, S2, S3, S4, S5, S6, S7, S8, S9 and S10, then the preselected input signals are selected. It can be S1 and S10, or S1, S4, S7 and S10.
  • the method for selecting the preselected input signal and the number of the selected preselected input signals are not limited.
  • the selected method may perform an average division according to the power value corresponding to the input signal in the preset input signal group to determine the preselected input signal.
  • the number of pre-selected input signals can be two or more than two.
  • Step 202 may include:
  • Step 2021 Determine an adjustment factor of a dynamic linear model of the first input signal according to a dynamic linear characteristic parameter of the first input signal and a linear adjustment factor formula.
  • the linear adjustment factor formula is:
  • ⁇ i represents a parameter value of a dynamic linear characteristic parameter of the first input signal or a parameter value of a dynamic linear characteristic parameter of the second input signal
  • i represents a power magnitude of the first input signal or a power magnitude of the second input signal
  • ⁇ 1 represents the parameter value of the dynamic linear characteristic parameter of the first input signal
  • An adjustment factor representing a dynamic linear model of the first input signal or an adjustment factor of a dynamic linear model of the second input signal.
  • i represents the power level of the preselected input signal
  • the preselected input signal may be the first input signal with the highest power value and the second input signal with the lowest power value in the preset input signal group, or may be pre-selected.
  • the input signal with the highest power value, the input signal with the lowest power value, and the input signal corresponding to other power values are set in the input signal group.
  • the selection of the pre-selected input signal is not limited in the embodiment of the present invention.
  • the parameter that can be used to represent the dynamic linear characteristic of the input signal is a dynamic linear characteristic parameter.
  • the dynamic linear characteristic parameter may be the average power of the output signal corresponding to the preset input signal group in the power amplifier or in the power amplifier. The gain of the output signal corresponding to the preset input signal group.
  • Step 2022 Determine an adjustment factor of a dynamic linear model of the second input signal according to a parameter value of the dynamic linear characteristic parameter of the second input signal and a linear adjustment factor formula.
  • the specific process of determining the adjustment factor of the dynamic linear model of the second input signal can be referred to the determination process in step 2021.
  • Step 203 Determine an adjustment factor of a dynamic nonlinear model of the preselected input signal in the amplifier model according to the preselected input signal.
  • the embodiment of the present invention adopts a method of nonlinear measurement, that is, by calculating the difference between the optimal dynamic linear characteristic approximation of the characteristics of the power amplifier to represent the remaining dynamic non- Linear characteristics.
  • the step 203 may include:
  • Step 2031 Determine a parameter value of a dynamic nonlinear characteristic parameter of the preselected input signal according to a parameter value of the dynamic linear characteristic parameter of the preselected input signal and a nonlinear characteristic formula.
  • a parameter value indicating a dynamic nonlinear characteristic parameter of the preselected input signal x (i) (n) represents a signal value of the preselected input signal, y (i) (n) represents an output signal corresponding to the preselected input signal, and ⁇ i represents the first The parameter value of the dynamic linear characteristic parameter of the input signal or the parameter value of the dynamic linear characteristic parameter of the second input signal.
  • Step 2032 Determine an adjustment factor of a dynamic nonlinear model of the preselected input signal according to a parameter value of the dynamic nonlinear characteristic parameter of the preselected input signal and a nonlinear adjustment factor formula.
  • the nonlinear adjustment factor formula is:
  • a parameter value representing a dynamic nonlinear characteristic parameter of the first input signal a parameter value representing a dynamic nonlinear characteristic parameter of the preselected input signal, An adjustment factor representing a dynamic nonlinear model of the first input signal or an adjustment factor of a dynamic nonlinear model of the second input signal.
  • the adjustment factor of the dynamic linear model of the first input signal can be determined according to the parameter value of the dynamic linear characteristic parameter of the first input signal, and according to the parameter value of the dynamic linear characteristic parameter of the second input signal, Determining an adjustment factor of a dynamic linear model of the second input signal; determining a parameter value of the dynamic nonlinear characteristic parameter of the first input signal according to the parameter value of the dynamic linear characteristic parameter of the first input signal, thereby determining the first input signal
  • the adjustment factor of the dynamic nonlinear model can determine the parameter value of the dynamic nonlinear characteristic parameter of the second input signal according to the parameter value of the dynamic linear characteristic parameter of the second input signal, thereby determining the dynamic nonlinear model of the second input signal Adjustment factor.
  • Step 204 Establish an amplifier model of the current input signal according to an adjustment factor of the dynamic linear model of the preselected input signal and an adjustment factor of the dynamic nonlinear model of the preselected input signal.
  • the adjustment factor of the dynamic linear model of the first input signal and the adjustment factor of the dynamic nonlinear model of the first input signal, the adjustment factor of the dynamic linear model of the second input signal, and the adjustment factor of the dynamic nonlinear model of the second input signal , the adjustment factor of the dynamic linear model of the preset input signal group and the adjustment factor of the dynamic nonlinear model can be determined, and then the model coefficient and the pre-determination of the static model of the preset input signal group can be determined according to the least square method using the amplifier model formula.
  • model coefficients of the dynamic model Establish an amplifier model of the current input signal.
  • the amplifier model formula can refer to the amplifier model formula in step 201.
  • FIG. 7 is a process for determining a model coefficient of a static model corresponding to a preselected input signal and a model coefficient of a dynamic model according to an adjustment factor of a dynamic linear model of the preselected input signal and an adjustment factor of the dynamic nonlinear model, as shown in FIG. 7 .
  • a preselected input signal in the preset input signal group is selected, and the preselected input signal may be the input signal with the highest power value and the input signal with the lowest power value in the preset input signal group, or may be the preset input signal.
  • the adjustment factors of the corresponding dynamic linear model and the adjustment factors of the dynamic nonlinear model are calculated, and then the model coefficients and dynamics of the static model corresponding to the preselected input signals are determined according to the least squares method using the amplifier model formula.
  • the model factor of the model is determined according to the least squares method using the amplifier model formula.
  • x (1) (n) to x (R) (n) represents the input signal in the preset input signal group
  • y (1) (n) to y (R) (n) represents the output corresponding to the preset input signal group signal
  • An adjustment factor indicating a dynamic linear model corresponding to an input signal in a preset input signal group to An adjustment factor indicating a dynamic nonlinear model corresponding to an input signal in a preset input signal group
  • a model coefficient representing a static model of a preset input signal group
  • a model coefficient representing a dynamic model of a preset input signal group is
  • step 204 may specifically include:
  • Step 2041 determining an model coefficient of a static model of the current input signal by using an amplifier model formula according to an adjustment factor of a dynamic linear model of the preselected input signal and an adjustment factor of a dynamic nonlinear model of the preselected input signal. And model coefficients of the dynamic model
  • step 2041 may specifically include:
  • Step 2041a Determine an adjustment factor of a dynamic linear model of the preset input signal group according to an adjustment factor of the dynamic linear model of the preselected input signal.
  • step 2041a may include:
  • the first interpolation formula is used to determine the adjustment factor of the dynamic linear model of the preset input signal group
  • the first interpolation formula can be:
  • An adjustment factor representing a dynamic linear model of the first input signal An adjustment factor representing a dynamic linear model of the second input signal
  • w (r) represents a weighting factor
  • a weighting factor w (r) is determined according to a weighting formula, which may be:
  • P r represents the power value of the input signal of the power level r
  • Step 2041b Determine an adjustment factor of the dynamic nonlinear model of the preset input signal group according to an adjustment factor of the dynamic nonlinear model of the preselected input signal.
  • step 2041b may include:
  • the second interpolation formula is used to determine the adjustment factor of the dynamic nonlinear model of the preset input signal group.
  • the second interpolation formula can be:
  • An adjustment factor representing a dynamic nonlinear model of the first input signal An adjustment factor representing a dynamic nonlinear model of the second input signal, w (r) representing a weighting factor.
  • the preselected input signal in the preset input signal group is also referred to as a trained input signal, and the input signal other than the preselected input signal in the preset input signal group is referred to as an untrained input signal, as shown by step 2041a and step 2041b.
  • the adjustment factor and the dynamic nonlinear model of the dynamic linear model corresponding to the input signal can be obtained by the first interpolation formula and the second interpolation formula. Adjustment factor.
  • the step 2041a and the step 2041b may be performed at the same time, and the sequence of the embodiment of the present invention is not limited.
  • FIG. 10 is a method for determining an adjustment factor of a dynamic linear model corresponding to an input signal signal that is not trained in a preset input signal group and an adjustment factor of a dynamic nonlinear model by using a first interpolation formula and a second interpolation formula. For details, refer to step 2041a and step 2041b.
  • M represents the power level of the first input signal
  • m represents the power level of the second input signal
  • x (M) (n) represents the first input signal
  • C M represents the model coefficient of the first input signal
  • y ( M) (n) represents the output signal corresponding to the first input signal
  • x (r) (n) represents that the power value is smaller than the power level M of the first input signal, and the power level m greater than the second input signal is not trained.
  • the input signal, (C r ) represents the model coefficient of the input signal
  • y (r) (n) represents the output signal corresponding to the input signal
  • x (m) (n) represents the second input signal
  • (C m ) represents The model coefficient of the second input signal
  • y (m) (n) represents the output signal corresponding to the second input signal
  • w (r) represents the weighting factor
  • the first interpolation formula and the second interpolation formula are used to determine the adjustment factor of the dynamic linear model corresponding to the input signal and the dynamic nonlinear model. Adjustment factor.
  • the input signals in the preset input signal group are arranged in descending order of power values, in order: S1, S2, S3, S4, S5, S6, S7, S8, S9 and S10, if the preselected input signal is S1 And S10, according to S1 and S10, the adjustment factor of the dynamic linear model corresponding to any one of the input signals S2, S3, S4, S5, S6, S7, S8 and S9 can be determined by the first interpolation formula and the second interpolation formula.
  • the adjustment factor of the dynamic nonlinear model if the preselected input signals are S1, S2, S6 and S10, since the power value of S3 is between the power values of S2 and S6, the first interpolation formula and the first can be adopted according to S2 and S6.
  • the two interpolation formula determines the adjustment factor of the dynamic linear model corresponding to S3 and the adjustment factor of the dynamic nonlinear model.
  • the adjustment factor and dynamics of the dynamic linear model corresponding to any one of S4, S5, S7, S8 and S9 can be determined.
  • the adjustment factor of the nonlinear model if the preselected input signals are S1, S2, S6 and S10, since the power value of S3 is between the power values of S2 and S6, the first interpolation formula and the first can be adopted according to S2 and S6.
  • the two interpolation formula determines the adjustment factor of the dynamic linear model corresponding to S3 and the adjustment factor of the dynamic nonlinear model.
  • the adjustment factor and dynamics of the dynamic linear model corresponding to
  • Step 2041c Substituting the adjustment factor of the dynamic linear model of the preset input signal group and the adjustment factor of the dynamic nonlinear model of the preset input signal group into the amplifier model formula to obtain model coefficients and presets of the static model of the preset input signal group The model coefficients of the dynamic model of the input signal group.
  • the adjustment factor of the dynamic linear model of the preset input signal group and the adjustment factor of the dynamic nonlinear model of the preset input signal group are obtained according to the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model of the preselected input signal.
  • the static model model is used to determine the static of the preset input signal group according to the least square method.
  • the model coefficients of the model and the model coefficients of the dynamic model of the preset input signal group are obtained according to the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model of the preselected input signal.
  • Step 2041d using a model coefficient of a static model of the preset input signal group as a model coefficient of a static model of the current input signal
  • the model coefficient of the dynamic model of the preset input signal group is used as the model coefficient of the dynamic model of the current input signal
  • the model coefficients of the static model of the preset input signal group and the model coefficients of the dynamic model of the preset input signal group are the model coefficients of the static model of the current input signal.
  • Step 2042 Apply model coefficients of a static model of the current input signal And model coefficients of the dynamic model Substitute the amplifier model formula to get the amplifier model corresponding to the current input signal.
  • Model coefficients for static models of current input signals are known And model coefficients of the dynamic model
  • the amplifier model corresponding to the current input signal can be established.
  • the method for acquiring the model coefficients in the digital predistortion correction method provided by the embodiment of the present invention can reduce the number of power values corresponding to different input signals that need to participate in the calculation, thereby reducing the complexity of the amplifier model establishment.
  • the way of obtaining the model coefficients corresponding to the remaining power values by means of linear interpolation can be combined with the adjustment factor of the original dynamic linear model in the amplifier model formula and the adjustment factor of the dynamic nonlinear model, thereby being simple and Quickly determine the model coefficients of the static model of the current input signal And model coefficients of the dynamic model
  • Step 205 Obtain a DPD model of the current input signal according to an amplifier model of the current input signal.
  • the inverse function of the amplifier model of the current input signal can be used as the DPD model of the current input signal.
  • the current input signal is linearly amplified after the cascaded DPD modules 01 and PA02, thereby avoiding the output signal of the current input signal after passing through PA02. Produces distortion.
  • Step 206 Perform digital predistortion correction on the current input signal according to the DPD model of the current input signal.
  • the predistortion signal corresponding to the inverse function of the DPD model of the current input signal is superimposed on the current input signal, so that the superposed signal passes through the power amplifier module to achieve the purpose of predistortion correction of the current input signal.
  • Step 207 When the state of the power amplifier changes, the amplifier model of the current input signal is updated to obtain an updated amplifier model.
  • steps 201 to 206 can better fit the characteristics of the power amplifier when the output power of the power amplifier changes, the operating state of the power amplifier may vary with device aging, temperature fluctuation, or bias voltage variation. The reason changes. Therefore, when the state of the power amplifier changes, the amplifier model of the current input signal needs to be updated to obtain an updated amplifier model to accommodate these changes. Illustrative, the state change of the power amplifier can be device aging, temperature fluctuations, or bias voltage variations. It should be noted that the method for detecting whether the state of the power amplifier is changed may be various. For details, refer to the prior art, and details are not described herein again.
  • the amplifier model of the current input signal is updated according to the specific update mode in step 207 to obtain the updated amplifier model.
  • step 207 may specifically include:
  • Step 2071 Acquire a model coefficient of a dynamic model of the updated amplifier model according to an amplifier model of the current input signal.
  • step 2071 may specifically include:
  • Step 2071a the difference between the model coefficient of the dynamic model of the input signal after changing the state of the power amplifier and the model coefficient of the dynamic model of the input signal before the change is taken as the first difference.
  • Step 2071b using the sum of the model coefficients of the dynamic model of the current input signal and the first difference as the model coefficients of the dynamic model of the updated amplifier model.
  • Step 2072 Acquire a model coefficient of the static model of the updated amplifier model according to an amplifier model of the current input signal.
  • step 2072 may specifically include:
  • Step 2072a the difference between the model coefficient of the static model of the input signal after changing the state of the power amplifier and the model coefficient of the static model of the input signal before the change is used as the second difference.
  • Step 2072b using the sum of the model coefficients of the static model of the current input signal and the second difference as the model coefficients of the static model of the updated amplifier model.
  • updating the model coefficients of the dynamic model of the amplifier model and the model coefficients of the static model can include two aspects:
  • model coefficients after the state change can be determined. (model coefficient of static model of input signal after state change) Model coefficients with dynamic models ) to determine the model coefficients after the state changes Difference between the model coefficient C (r') before the state change (second difference) First difference ):
  • Step 2073 Substituting the model coefficients of the dynamic model of the updated amplifier model and the model coefficients of the static model of the updated amplifier model into the amplifier model to obtain an updated amplifier model.
  • FIG. 14 is a process of updating the model coefficients of the amplifier model.
  • the upper part of the dotted line indicates the model coefficients of the dynamic model of the input signal after the state change of the power amplifier and the model coefficient of the dynamic model of the input signal before the change.
  • the first difference, and the process of obtaining the second difference between the model coefficient of the static model of the input signal after the state change of the power amplifier and the model coefficient of the static model of the input signal before the change refer to step 2071a and the step. 2072a
  • the portion below the dotted line indicates the model coefficient of the dynamic model of the dynamic model of the current input signal as the model coefficient of the dynamic model of the updated amplifier model, and the model coefficient of the static model of the current input signal.
  • the process of using the sum of the second difference as the model coefficient of the static model of the updated amplifier model can be specifically referred to steps 2071b and 2072b, and x (1) (n) to x (R) (n) in FIG. Current input signal.
  • Step 208 The inverse function of the updated amplifier model is used as a DPD model of the current input signal.
  • the amplifier model of the current input signal is updated to obtain the updated amplifier model.
  • the inverse function of the updated amplifier model is used as the DPD model of the current input signal to avoid the current input signal.
  • the output signal after passing through PA02 produces distortion.
  • Step 209 Perform digital predistortion correction on the current input signal according to the DPD model corresponding to the inverse function of the updated amplifier model.
  • step 206 For details, refer to step 206 in step 209, and details are not described herein again.
  • an input signal of a higher power value in the preset input signal group may be selected as the input signal of the amplifier module in FIG. No., and then calculate the model coefficient after the state change and the model coefficient before the state change, the model coefficient of the updated amplifier model according to the recalculated model coefficients, and then apply it to other power values to obtain the power amplifier.
  • the new amplifier model factor after the state change In this way, the amplifier model can quickly adapt to changes in the state of the power amplifier through occasional one or two coefficient updates, and adjust the model coefficients in the entire power dynamic range in time to adapt to this change in the power amplifier.
  • the amplifier model is better able to adapt to the dynamic changes in power amplifier power in modern wireless communication systems.
  • the amplifier model established according to the embodiment of the present invention is an accurate but simple amplifier model, which satisfies the solution of the output signal of the power amplifier. Requirements for key issues of DPD technology in power dynamics scenarios.
  • the size of the sequence numbers of the foregoing processes does not mean the order of execution sequence, and the execution order of each process should be determined by its function and internal logic, and the present invention should not be The implementation of the embodiments constitutes any limitation.
  • the digital predistortion correction method provided by the embodiment of the present invention can determine the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model according to the preselected input signal, and then The amplifier model of the current input signal is established according to the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model of the preselected input signal, and then the DPD model of the current input signal is obtained according to the amplifier model of the current input signal, and finally according to the DPD.
  • the model performs digital predistortion correction on the current input signal.
  • the embodiment of the present invention provides a digital predistortion correction apparatus 50.
  • the apparatus provided by all embodiments of the present invention may be applied to a communication system, for example, the apparatus may be a radio frequency unit or a base station, or may be a part of a radio frequency unit or a base station. It can also be applied to other systems that require digital predistortion correction, such as part of a radar system or radar system.
  • the digital predistortion correction device 50 includes:
  • a first determining unit 501 configured to determine, according to the preselected input signal, an adjustment factor of a dynamic linear model of the preselected input signal in the amplifier model, the amplifier model being used to indicate a relationship between the output signal and a static model of the input signal, and a dynamic model of the input signal
  • the dynamic model includes a dynamic linear model for indicating a linear characteristic in a variable of the input signal, and a dynamic nonlinear model for indicating a nonlinear characteristic in the variable of the input signal, the preselected input signal To preset at least two signals in the input signal group, the preset input signal group includes a plurality of input signals having different power values.
  • the second determining unit 502 is configured to determine an adjustment factor of the dynamic nonlinear model of the preselected input signal in the amplifier model according to the preselected input signal.
  • the establishing unit 503 is configured to establish an amplifier model of the current input signal according to an adjustment factor of the dynamic linear model of the preselected input signal and an adjustment factor of the dynamic nonlinear model of the preselected input signal.
  • the processing unit 504 is configured to obtain a digital pre-distortion DPD model of the current input signal according to an amplifier model of the current input signal.
  • the correcting unit 505 is configured to perform digital predistortion correction on the current input signal according to the DPD model of the current input signal.
  • the digital predistortion correction apparatus can determine the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model according to the preselected input signal, and then The amplifier model of the current input signal is established according to the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model of the preselected input signal, and then the DPD model of the current input signal is obtained according to the amplifier model of the current input signal, and finally according to the DPD.
  • the model performs digital predistortion correction on the current input signal.
  • the digital predistortion correction apparatus 50 includes:
  • a first determining unit 501 configured to determine, according to the preselected input signal, an adjustment factor of a dynamic linear model of the preselected input signal in the amplifier model, the amplifier model being used to indicate the output signal and the input signal
  • the static model, the dynamic model of the input signal, the dynamic model includes a dynamic linear model and a dynamic nonlinear model, wherein the dynamic linear model is used to indicate the linear characteristic in the variable of the input signal, and the dynamic nonlinear model is used to indicate the input signal
  • the non-linear characteristic of the variable, the pre-selected input signal is at least two signals in the preset input signal group, and the preset input signal group includes a plurality of input signals having different power values.
  • the second determining unit 502 is configured to determine an adjustment factor of the dynamic nonlinear model of the preselected input signal in the amplifier model according to the preselected input signal.
  • the establishing unit 503 is configured to establish an amplifier model of the current input signal according to an adjustment factor of the dynamic linear model of the preselected input signal and an adjustment factor of the dynamic nonlinear model of the preselected input signal.
  • the processing unit 504 is configured to obtain a digital pre-distortion DPD model of the current input signal according to an amplifier model of the current input signal.
  • the correcting unit 505 is configured to perform digital predistortion correction on the current input signal according to the DPD model of the current input signal.
  • the updating unit 506 is configured to update the amplifier model of the current input signal to obtain an updated amplifier model when the state of the power amplifier changes.
  • the establishing unit 503 may include:
  • the first determining module 5031 and the first substituting module 5032 are identical to each other.
  • the first determining module 5031 is configured to determine, according to an adjustment factor of the dynamic linear model of the preselected input signal, an adjustment factor of the dynamic nonlinear model of the preselected input signal, and use an amplifier model formula to determine a model coefficient of the static model of the current input signal.
  • model coefficients of the dynamic model The amplifier model formula is:
  • N represents the number of sampling points of each input signal in the preset input signal group
  • L1 represents the number of model coefficients of the static model
  • L2 represents the number of model coefficients of the dynamic model
  • r represents the preset input signal group.
  • the power level of the input signal, r is an integer greater than or equal to 1,
  • An output signal group representing a preset input signal group, a static model representing a preset input signal group, a dynamic model representing a preset input signal group, a dynamic linear model representing a preset input signal group, a dynamic nonlinear model representing a preset input signal group,
  • An adjustment factor representing a dynamic linear model of a preset input signal group
  • the first generation module 5032 is used to model the coefficient of the static model of the current input signal And model coefficients of the dynamic model Substitute the amplifier model formula to get the amplifier model corresponding to the current input signal.
  • the first determining module 5031 may include:
  • the first determining submodule 50311, the second determining submodule 50312, is substituted into the submodule 50313 and the first processing submodule 50314.
  • the first determining submodule 50311 is configured to determine an adjustment factor of the dynamic linear model of the preset input signal group according to an adjustment factor of the dynamic linear model of the preselected input signal.
  • the second determining submodule 50312 is configured to determine an adjustment factor of the dynamic nonlinear model of the preset input signal group according to an adjustment factor of the dynamic nonlinear model of the preselected input signal.
  • a first processing sub-module 50314 configured to use a model coefficient of a static model of the preset input signal group as a model coefficient of a static model of the current input signal
  • the model coefficient of the dynamic model of the preset input signal group is used as the model coefficient of the dynamic model of the current input signal
  • the first determining unit 501 may include:
  • the second determining module 5011 is configured to determine an adjustment factor of the dynamic linear model of the first input signal according to the dynamic linear characteristic parameter and the linear adjustment factor formula of the first input signal.
  • the third determining module 5012 is configured to determine an adjustment factor of the dynamic linear model of the second input signal according to the parameter value of the dynamic linear characteristic parameter of the second input signal and the linear adjustment factor formula.
  • the linear adjustment factor formula is:
  • ⁇ i represents a parameter value of a dynamic linear characteristic parameter of the first input signal or a parameter value of a dynamic linear characteristic parameter of the second input signal
  • i represents a power magnitude of the first input signal or a power magnitude of the second input signal
  • ⁇ 1 represents the parameter value of the dynamic linear characteristic parameter of the first input signal
  • An adjustment factor representing a dynamic linear model of the first input signal or an adjustment factor of a dynamic linear model of the second input signal.
  • the dynamic linear characteristic parameter is the average power of the output signal corresponding to the preset input signal group in the power amplifier or the gain of the output signal corresponding to the preset input signal group in the power amplifier.
  • the second determining unit 502 can include:
  • the fourth determining module 5021 and the fifth determining module 5022 are identical to each other.
  • the fourth determining module 5021 is configured to determine a parameter value of a dynamic nonlinear characteristic parameter of the preselected input signal according to a parameter value of the dynamic linear characteristic parameter of the preselected input signal and a nonlinear characteristic formula, where the nonlinear characteristic formula is:
  • a parameter value indicating a dynamic nonlinear characteristic parameter of the preselected input signal x (i) (n) represents a signal value of the preselected input signal, and y (i) (n) represents an output signal corresponding to the preselected input signal.
  • the fifth determining module 5022 is configured to determine, according to the parameter value of the dynamic nonlinear characteristic parameter of the preselected input signal and the nonlinear adjustment factor formula, an adjustment factor of the dynamic nonlinear model of the preselected input signal, where the nonlinear adjustment factor formula is:
  • a parameter value representing a dynamic nonlinear characteristic parameter of the first input signal is a parameter value representing a dynamic nonlinear characteristic parameter of the first input signal.
  • the first determining submodule is specifically configured to: determine, according to an adjustment factor of the dynamic linear model of the preselected input signal, a first interpolation formula to determine an adjustment factor of the dynamic linear model of the preset input signal group
  • the first interpolation formula is:
  • r indicates that the power level of the input signal in the preset input signal group is smaller than the power level M of the first input signal, and greater than the power level m of the second input signal
  • An adjustment factor representing a dynamic linear model of the first input signal An adjustment factor representing a dynamic linear model of the second input signal
  • w (r) represents a weighting factor
  • w (r) is determined according to a weighting formula, the weighting formula is:
  • P r represents the power value of the input signal of the power level r
  • P M represents the power value of the first input signal of the power level M.
  • the second determining sub-module is specifically configured to: determine, according to an adjustment factor of the dynamic nonlinear model of the preselected input signal, a second interpolation formula to determine an adjustment factor of the dynamic nonlinear model of the preset input signal group
  • the second interpolation formula is:
  • An adjustment factor representing a dynamic nonlinear model of the first input signal An adjustment factor representing a dynamic nonlinear model of the second input signal, w (r) representing a weighting factor.
  • the state change of the power amplifier may be device aging, temperature fluctuation or bias voltage variation.
  • the update unit 506 can include:
  • the first obtaining module 5061, the second obtaining module 5062 and the second joining module 5063 are identical to each other.
  • the first obtaining module 5061 is configured to obtain model coefficients of the dynamic model of the updated amplifier model according to an amplifier model of the current input signal.
  • the second obtaining module 5062 is configured to obtain model coefficients of the static model of the updated amplifier model according to the amplifier model of the current input signal.
  • the second generation module 5063 is configured to substitute the model coefficients of the dynamic model of the updated amplifier model and the model coefficients of the static model of the updated amplifier model into the amplifier model to obtain an updated amplifier model.
  • the first obtaining module 5061 may include:
  • the first difference sub-module 50611 is configured to use a difference between a model coefficient of a dynamic model of the input signal after the state of the power amplifier is changed and a model coefficient of the dynamic model of the input signal before the change as the first difference.
  • the second processing sub-module 50612 is configured to use the sum of the model coefficients of the dynamic model of the current input signal and the first difference as the model coefficients of the dynamic model of the updated amplifier model.
  • the second obtaining module 5062 may include:
  • the second difference sub-module 50621 is configured to use a difference between a model coefficient of the static model of the input signal after the state change of the power amplifier and a model coefficient of the static model of the input signal before the change as the second difference.
  • a third processing sub-module 50622 configured to calculate a model coefficient of the static model of the current input signal The sum of the two differences is used as the model coefficient of the static model of the updated amplifier model.
  • the digital predistortion correction apparatus can determine the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model according to the preselected input signal, and then The amplifier model of the current input signal is established according to the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model of the preselected input signal, and then the DPD model of the current input signal is obtained according to the amplifier model of the current input signal, and finally according to the DPD.
  • the model performs digital predistortion correction on the current input signal.
  • the digital predistortion correction apparatus includes: at least one processor 701, at least one input/output (English: Input-Output; IO) interface. 702 or other communication interface, memory 703, and at least one communication bus 704 are used to effect connection communication between these devices.
  • the processor 701 is configured to execute a computer execution instruction 7031 in the memory 703, and control the IO interface 702 to perform transmission and reception.
  • the memory 703 may include a random access memory (English: Random Access Memory; RAM), and may also include a non-volatile memory, such as at least one disk storage.
  • the digital predistortion correction device is configured to receive and transmit data via at least one IO interface 702 (which may be wired or wireless), including a communication connection with at least one other network element.
  • IO interface 702 which may be wired or wireless
  • the processor executes the instructions in the execution of the computer, the steps in the foregoing method embodiments may be performed. For details, refer to the description in the foregoing method embodiments, and details are not described herein.
  • the embodiment of the present invention further provides a digital predistortion correction device.
  • the digital predistortion correction device includes:
  • the processor 701 is configured to determine, according to the preselected input signal, an adjustment factor of a dynamic linear model of the preselected input signal in the amplifier model, where the amplifier model is used to indicate a relationship between the output signal and a static model of the input signal, a dynamic model of the input signal, and dynamic
  • the model includes a dynamic linear model and a dynamic nonlinear model, wherein the dynamic linear model is used to indicate linear characteristics in the variables of the input signal, and the dynamic nonlinear
  • the qualitative model is used to indicate a nonlinear characteristic in the variable of the input signal, and the preselected input signal is at least two signals in the preset input signal group, and the preset input signal group includes a plurality of input signals having different power values.
  • the processor 701 is further configured to determine an adjustment factor of the dynamic nonlinear model of the preselected input signal in the amplifier model based on the preselected input signal.
  • the processor 701 is further configured to establish an amplifier model of the current input signal according to an adjustment factor of the dynamic linear model of the preselected input signal and an adjustment factor of the dynamic nonlinear model of the preselected input signal.
  • the processor 701 is further configured to obtain a digital pre-distortion DPD model of the current input signal according to an amplifier model of the current input signal.
  • the inverse function of the amplifier model of the current input signal can be used as a digital pre-distortion DPD model of the current input signal.
  • the processor 701 is further configured to perform digital predistortion correction on the current input signal according to the DPD model of the current input signal.
  • the digital predistortion correction apparatus can determine the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model according to the preselected input signal, and then The amplifier model of the current input signal is established according to the adjustment factor of the dynamic linear model of the preselected input signal and the adjustment factor of the dynamic nonlinear model of the preselected input signal, and then the DPD model of the current input signal is obtained according to the amplifier model of the current input signal, and finally according to the DPD.
  • the model performs digital predistortion correction on the current input signal.
  • processor 701 is specifically configured to:
  • the model coefficient of the static model of the current input signal is determined by using the amplifier model formula.
  • model coefficients of the dynamic model The amplifier model formula is:
  • N represents the number of sampling points of each input signal in the preset input signal group
  • L1 represents the number of model coefficients of the static model
  • L2 represents the number of model coefficients of the dynamic model
  • r represents the preset input signal group.
  • the power level of the input signal, r is an integer greater than or equal to 1,
  • An output signal group representing a preset input signal group, a static model representing a preset input signal group, a dynamic model representing a preset input signal group, a dynamic linear model representing a preset input signal group, a dynamic nonlinear model representing a preset input signal group, An adjustment factor representing a dynamic linear model of a preset input signal group, An adjustment factor representing a dynamic nonlinear model of a preset input signal group; a model coefficient of a static model of the current input signal And model coefficients of the dynamic model Substitute the amplifier model formula to get the amplifier model corresponding to the current input signal.
  • the processor 701 is also specifically configured to:
  • the adjustment factor of the dynamic linear model of the preset input signal group is determined according to the adjustment factor of the dynamic linear model of the preselected input signal; the adjustment of the dynamic nonlinear model of the preset input signal group is determined according to the adjustment factor of the dynamic nonlinear model of the preselected input signal Factor; the adjustment factor of the dynamic linear model of the preset input signal group and the adjustment factor of the dynamic nonlinear model of the preset input signal group are substituted into the amplifier model formula to obtain the model coefficient and the preset input of the static model of the preset input signal group.
  • the model coefficient of the dynamic model of the signal group; the model coefficient of the static model of the preset input signal group is used as the model coefficient of the static model of the current input signal
  • the model coefficient of the dynamic model of the preset input signal group is used as the model coefficient of the dynamic model of the current input signal
  • the pre-selected input signal may be a first input signal with the highest power value and a second input signal with the lowest power value in the preset input signal group, and the processor 701 is specifically configured to:
  • ⁇ i represents a parameter value of a dynamic linear characteristic parameter of the first input signal or a parameter value of a dynamic linear characteristic parameter of the second input signal
  • i represents a power magnitude of the first input signal or a power magnitude of the second input signal
  • ⁇ 1 represents the parameter value of the dynamic linear characteristic parameter of the first input signal
  • An adjustment factor representing a dynamic linear model of the first input signal or an adjustment factor of a dynamic linear model of the second input signal.
  • the dynamic linear characteristic parameter may be the average power of the output signal corresponding to the preset input signal group in the power amplifier or the output signal corresponding to the preset input signal group in the power amplifier.
  • the processor 701 is also specifically configured to:
  • the nonlinear characteristic formula is:
  • a parameter value indicating a dynamic nonlinear characteristic parameter of the preselected input signal x (i) (n) represents a signal value of the preselected input signal, and y (i) (n) represents an output signal corresponding to the preselected input signal.
  • the adjustment factor of the dynamic nonlinear model of the preselected input signal is determined according to the parameter value of the dynamic nonlinear characteristic parameter of the preselected input signal and the nonlinear adjustment factor formula, and the nonlinear adjustment factor formula is:
  • a parameter value representing a dynamic nonlinear characteristic parameter of the first input signal is a parameter value representing a dynamic nonlinear characteristic parameter of the first input signal.
  • the processor 701 is also specifically configured to:
  • the first interpolation formula is used to determine the adjustment factor of the dynamic linear model of the preset input signal group
  • the first interpolation formula is:
  • r indicates that the power level of the input signal in the preset input signal group is smaller than the power level M of the first input signal, and greater than the power level m of the second input signal
  • An adjustment factor representing a dynamic linear model of the first input signal An adjustment factor representing a dynamic linear model of the second input signal
  • w (r) represents a weighting factor
  • w (r) is determined according to a weighting formula, the weighting formula is:
  • P r represents the power value of the input signal of the power level r
  • P M represents the power value of the first input signal of the power level M.
  • the processor 701 is further configured to determine, according to an adjustment factor of the dynamic nonlinear model of the preselected input signal, a second interpolation formula to determine an adjustment factor of the dynamic nonlinear model of the preset input signal group.
  • the second interpolation formula is:
  • An adjustment factor representing a dynamic nonlinear model of the first input signal An adjustment factor representing a dynamic nonlinear model of the second input signal, w (r) representing a weighting factor.
  • processor 701 is further configured to:
  • the amplifier model of the current input signal is updated to obtain an updated amplifier model.
  • the state change of the power amplifier can be device aging, temperature fluctuations, or bias voltage variations.
  • the processor 701 is further configured to: obtain a model coefficient of the dynamic model of the updated amplifier model according to an amplifier model of the current input signal; and acquire a model coefficient of the static model of the updated amplifier model according to an amplifier model of the current input signal; The model coefficients of the dynamic model of the updated amplifier model and the model coefficients of the static model of the updated amplifier model are substituted into the amplifier model to obtain an updated amplifier model.
  • the processor 701 is further configured to: use a difference between a model coefficient of a dynamic model of the input signal after changing a state of the power amplifier and a model coefficient of a dynamic model of the input signal before the change as a first difference; and dynamically change a current input signal.
  • the sum of the model coefficients of the model and the first difference is used as the model coefficient of the dynamic model of the updated amplifier model.
  • the processor 701 is further configured to: use a difference between a model coefficient of a static model of the input signal after changing a state of the power amplifier and a model coefficient of a static model of the input signal before the change as a second difference; and statically use the current input signal The sum of the model coefficients of the model and the second difference is used as the model coefficient of the static model of the updated amplifier model.
  • the processor 701 may be a central processing unit (English: Central Processing Unit; CPU), and the processor may also be other general-purpose processors and digital signal processors.
  • Digital Sgnal Processing referred to as: DSP
  • Application Specific Integrated Circuit ASIC
  • Ready-to-Programmable Gate Array English: Field-Programmable Gate Array; FPGA
  • a general purpose processor can be A microprocessor, which can also be any conventional processor or the like.
  • the memory 703 can include read only memory and random access memory and provides computer executing instructions and data to the processor 701.
  • a portion of the memory may also include a non-volatile random access memory.
  • the memory can also store information of the device type.
  • each step in the foregoing method embodiments may be completed by an integrated logic circuit of hardware in the processor 701 or an instruction in a form of software.
  • the steps of the method disclosed in the embodiments of the present invention may be directly implemented as a hardware processor, or may be performed by a combination of hardware and software modules in the processor.
  • the software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.
  • the storage medium is located in the memory, and the processor reads the information in the memory and combines the hardware to complete the steps of the above method. To avoid repetition, it will not be described in detail here.
  • the disclosed apparatus and method may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the functions may be stored in a computer readable storage medium if implemented in the form of a software functional unit and sold or used as a standalone product.
  • the technical solution of the present invention which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including
  • the instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention.
  • the aforementioned storage medium Including: U disk, mobile hard disk, read-only memory (English: Read-Only Memory; for short: ROM), RAM, disk or optical disk and other media that can store program code.
  • the digital predistortion correction apparatus can determine, according to the preselected input signal, an adjustment factor of the dynamic linear model of the preselected input signal in the amplifier model and an adjustment factor of the dynamic nonlinear model. Then, according to the adjustment factor of the dynamic linear model of the preselected input signal, the adjustment factor of the dynamic nonlinear model of the preselected input signal, an amplifier model of the current input signal is established, and then the DPD model of the current input signal is obtained according to the amplifier model of the current input signal, and finally Digital predistortion correction of the current input signal according to the DPD model.
  • the lookup table model Compared to the lookup table model, it is not necessary to calculate and store the model parameters of the amplifier model covering a sufficient power range, thereby eliminating the need to calculate and store the DPD model covering a sufficient power range.
  • the model parameters are used to meet the calibration requirements of different input signals. Therefore, the acquisition process of the model parameters is simplified and the correction efficiency is improved.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Amplifiers (AREA)

Abstract

本发明提供了一种数字预失真校正方法及装置,涉及通信领域。所述方法包括:根据预选输入信号确定放大器模型中预选输入信号动态线性模型调整因子,预选输入信号为预设输入信号组中至少两个信号;根据预选输入信号确定放大器模型中预选输入信号的动态非线性模型调整因子;根据预选输入信号的动态线性模型调整因子、预选输入信号的动态非线性模型调整因子建立当前输入信号放大器模型;根据当前输入信号的放大器模型获得当前输入信号的DPD模型;根据DPD模型对当前输入信号进行数字预失真校正。本发明解决了模型参数获取过程较复杂,建模效率较低的问题,实现了简化模型参数的获取过程,提高校正效率的效果,用于预失真校正。

Description

数字预失真校正方法及装置 技术领域
本发明涉及通信领域,特别涉及一种数字预失真校正方法及装置。
背景技术
功率放大器是通信***中发送装置的重要组件。在通信***中,为满足发射要求,功率放大器需要在高效率运行的同时保持高线性度,而效率和线性度在功率放大器的设计中往往是相互抵触的。数字预失真(英文:Digital Pre-Distortion;简称:DPD)技术是一种专门针对功率放大器的非线性特性进行补偿的关键技术。该技术根据功率放大器的特性建立DPD模型,在数字域上对当前输入信号进行预失真处理,使得处理后的预失真信号在进入功率放大器之后,可以与功率放大器自身的非线性特性相抵消。借助DPD技术,可以使功率放大器既工作在高效率状态,同时在线性度上也满足指标要求。
在实际应用中,功率放大器的输出信号的功率经常会随着业务量的变化而实时进行动态调整,这种输出信号的功率的变化往往伴随着功率放大器特性的改变。在现有的DPD技术中,针对输出信号的功率动态变化场景下的非线性校正,常常是通过建立一个与功率值相关的查找表模型来实现的。查找表模型是将预设的DPD模型的模型参数和功率值的对应关系保存在表格中,当需要对当前输入信号进行预失真校正时,可以先根据该当前输入信号的功率值找到相应的模型参数,根据获取的模型参数确定当前输入信号对应的DPD模型,根据该DPD模型对当前输入信号进行预失真校正。
但是,由于查找表模型为了满足不同输入信号的校正要求,需要计算并存储覆盖足够的功率范围的DPD模型的模型参数,计算并存储的数据量是非常庞大的。因此,模型参数的获取过程较复杂,校正效率较低。
发明内容
为了解决模型参数的获取过程较复杂,校正效率较低的问题,本发明提供了一种数字预失真校正方法及装置。所述技术方案如下:
第一方面,提供了一种数字预失真校正方法,所述方法包括:
根据预选输入信号确定放大器模型中的所述预选输入信号的动态线性模型的调整因子,所述放大器模型用于指示输出信号与输入信号的静态模型、输入信号的动态模型的关系,所述动态模型包括动态线性模型和动态非线性模型,其中,所述动态线性模型用于指示输入信号的变量中的线性特性,所述动态非线性模型用于指示输入信号的变量中的非线性特性,所述预选输入信号为预设输入信号组中的至少两个信号,所述预设输入信号组包括功率值不同的多个输入信号;
根据所述预选输入信号确定所述放大器模型中的所述预选输入信号的动态非线性模型的调整因子;
根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型;
根据所述当前输入信号的放大器模型获得所述当前输入信号的数字预失真DPD模型;
根据所述当前输入信号的DPD模型对所述当前输入信号进行数字预失真校正。
结合第一方面,在第一种可实现方式中,所述根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型,包括:
根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定所述当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000001
和动态模型的模型系数
Figure PCTCN2015075606-appb-000002
所述放大器模型公式为:
Figure PCTCN2015075606-appb-000003
其中,所述N表示所述预设输入信号组中每个输入信号的采样点个数,所述L1表示所述静态模型的模型系数的个数,所述L2表示所述动态模型的模型系数的个数,所述r表示所述预设输入信号组中的输入信号的功率量级,所述r为大于或等于1的整数,所述
Figure PCTCN2015075606-appb-000004
表示所述预设输入信号组的输出信号组,所 述
Figure PCTCN2015075606-appb-000005
表示所述预设输入信号组的静态模型,所述
Figure PCTCN2015075606-appb-000006
表示所述预设输入信号组的动态模型,所述
Figure PCTCN2015075606-appb-000007
表示所述预设输入信号组的动态线性模型,所述
Figure PCTCN2015075606-appb-000008
表示所述预设输入信号组的动态非线性模型,所述
Figure PCTCN2015075606-appb-000009
表示所述预设输入信号组的动态线性模型的调整因子,所述
Figure PCTCN2015075606-appb-000010
表示所述预设输入信号组的动态非线性模型的调整因子;
将根据所述当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000011
和动态模型的模型系数
Figure PCTCN2015075606-appb-000012
代入所述放大器模型公式,建立所述当前输入信号对应的得到所述当前输入信号对应的放大器模型。
结合第一种可实现方式,在第二种可实现方式中,
所述根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定所述当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000013
和动态模型的模型系数
Figure PCTCN2015075606-appb-000014
包括:
根据所述预选输入信号的动态线性模型的调整因子确定所述预设输入信号组的动态线性模型的调整因子;
根据所述预选输入信号的动态非线性模型的调整因子确定所述预设输入信号组的动态非线性模型的调整因子;
将所述预设输入信号组的动态线性模型的调整因子、所述预设输入信号组的动态非线性模型的调整因子,采用代入所述放大器模型公式确定,得到所述预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数;
将所述预设输入信号组的静态模型的模型系数作为所述当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000015
将所述预设输入信号组的动态模型的模型系数作为所述当前输入信号的动态模型的模型系数
Figure PCTCN2015075606-appb-000016
结合第二种可实现方式,在第三种可实现方式中,所述预选输入信号为所述预设输入信号组中功率值最高的第一输入信号和功率值最低的第二输入信号,
所述根据预选输入信号确定放大器模型中的所述预选输入信号的动态线性模型的调整因子,包括:
根据所述第一输入信号的动态线性特性参数和线性调整因子公式,确定所 述所述第一输入信号的动态线性模型的调整因子;
根据所述第二输入信号的动态线性特性参数的参数值和线性调整因子公式,确定所述所述第二输入信号的动态线性模型的调整因子;
所述线性调整因子公式为:
Figure PCTCN2015075606-appb-000017
其中,所述θi表示所述第一输入信号的动态线性特性参数的参数值或所述第二输入信号的动态线性特性参数的参数值,所述i表示所述第一输入信号的功率量级或所述第二输入信号的功率量级,所述θ1表示第一输入信号的动态线性特性参数的参数值,所述
Figure PCTCN2015075606-appb-000018
表示第一输入信号的动态线性模型的调整因子或第二输入信号的动态线性模型的调整因子。
结合第三种可实现方式,在第四种可实现方式中,
所述动态线性特性参数为功率放大器中所述预设输入信号组对应的输出信号的平均功率或者功率放大器中所述预设输入信号组对应的输出信号的增益。
结合第三种可实现方式,在第五种可实现方式中,
所述根据所述预选输入信号确定所述放大器模型中的所述预选输入信号的动态非线性模型的调整因子,包括:
根据所述预选输入信号的动态线性特性参数的参数值和非线性特性公式确定所述预选输入信号的动态非线性特性参数的参数值,所述非线性特性公式为:
Figure PCTCN2015075606-appb-000019
其中,所述
Figure PCTCN2015075606-appb-000020
表示所述预选输入信号的动态非线性特性参数的参数值,所述x(i)(n)表示所述预选输入信号的信号值,所述y(i)(n)表示所述预选输入信号对应的输出信号;
根据所述预选输入信号的动态非线性特性参数的参数值和非线性调整因子公式确定所述预选输入信号的动态非线性模型的调整因子,所述非线性调整因子公式为:
Figure PCTCN2015075606-appb-000021
其中,所述
Figure PCTCN2015075606-appb-000022
表示所述第一输入信号的动态非线性特性参数的参数值。
结合第五种可实现方式,在第六种可实现方式中,
所述根据所述预选输入信号的动态线性模型的调整因子确定所述预设输入信号组的动态线性模型的调整因子,包括:
根据所述预选输入信号的动态线性模型的调整因子,采用第一插值公式确定所述预设输入信号组的动态线性模型的调整因子
Figure PCTCN2015075606-appb-000023
所述第一插值公式为:
Figure PCTCN2015075606-appb-000024
其中,所述r表示所述预设输入信号组中的输入信号的功率量级小于所述第一输入信号的功率量级M,大于所述第二输入信号的功率量级m,所述
Figure PCTCN2015075606-appb-000025
表示所述第一输入信号的动态线性模型的调整因子,所述
Figure PCTCN2015075606-appb-000026
表示所述第二输入信号的动态线性模型的调整因子,所述w(r)表示权重因子,所述权重因子w(r)根据权重公式确定得到,所述权重公式为:
Figure PCTCN2015075606-appb-000027
其中,所述Pr表示功率量级为r的输入信号的功率值,所述PM表示功率量级为M的所述第一输入信号的功率值;
所述根据所述预选输入信号的动态非线性模型的调整因子确定所述预设输入信号组的动态非线性模型的调整因子,包括:
根据所述预选输入信号的动态非线性模型的调整因子,采用第二插值公式确定所述预设输入信号组的动态非线性模型的调整因子
Figure PCTCN2015075606-appb-000028
所述第二插值公式为:
Figure PCTCN2015075606-appb-000029
其中,所述
Figure PCTCN2015075606-appb-000030
表示所述第一输入信号的动态非线性模型的调整因子,所述
Figure PCTCN2015075606-appb-000031
表示所述第二输入信号的动态非线性模型的调整因子,所述w(r)表示权重因子。
结合第一方面至第六种可实现方式中的任意一种,在第七种可实现方式中,在所述根据所述DPD模型对所述当前输入信号进行数字预失真校正之后,所述方法还包括:
当所述功率放大器的状态变化时,更新所述当前输入信号的DPD模型得到更新后的放大器模型。
结合第七种可实现方式,在第八种可实现方式中,
所述功率放大器的状态变化为器件老化、温度波动或者偏置电压变化。
结合第七或第八种可实现方式,在第九种可实现方式中,所述更新所述当前输入信号的放大器模型得到更新后的放大器模型,包括:
根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的动态模型的模型系数;
根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的静态模型的模型系数;
将所述更新后的放大器模型的动态模型的模型系数和所述更新后的放大器模型的静态模型的模型系数代入所述放大器模型得到所述更新后的放大器模型。
结合第九种可实现方式,在第十种可实现方式中,
所述根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的动态模型的模型系数,包括:
将所述功率放大器的状态变化后的输入信号的动态模型的模型系数与变化前的输入信号的动态模型的模型系数之差作为第一差值;
将所述当前输入信号的动态模型的模型系数与所述第一差值之和作为所述更新后的放大器模型的动态模型的模型系数。
结合第九种可实现方式,在第十一种可实现方式中,
所述根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的静态模型的模型系数,包括:
将所述功率放大器的状态变化后的输入信号的静态模型的模型系数与变化前的输入信号的静态模型的模型系数之差作为第二差值;
将所述当前输入信号的静态模型的模型系数与所述第二差值之和作为所述更新后的放大器模型的静态模型的模型系数。
第二方面,提供了一种数字预失真校正装置,所述数字预失真校正装置包括:
第一确定单元,用于根据预选输入信号确定放大器模型中的所述预选输入信号的动态线性模型的调整因子,所述放大器模型用于指示输出信号与输入信号的静态模型、输入信号的动态模型的关系,所述动态模型包括动态线性模型和动态非线性模型,其中,所述动态线性模型用于指示输入信号的变量中的线性特性,所述动态非线性模型用于指示输入信号的变量中的非线性特性,所述 预选输入信号为预设输入信号组中的至少两个信号,所述预设输入信号组包括功率值不同的多个输入信号;
第二确定单元,用于根据所述预选输入信号确定所述放大器模型中的所述预选输入信号的动态非线性模型的调整因子;
建立单元,用于根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型;
处理单元,用于根据所述当前输入信号的放大器模型获得所述当前输入信号的数字预失真DPD模型;
校正单元,用于根据所述当前输入信号的DPD模型对所述当前输入信号进行数字预失真校正。
结合第二方面,在第一种可实现方式中,所述建立单元包括:
第一确定模块,用于根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定所述当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000032
和动态模型的模型系数
Figure PCTCN2015075606-appb-000033
所述放大器模型公式为:
Figure PCTCN2015075606-appb-000034
其中,所述N表示所述预设输入信号组中每个输入信号的采样点个数,所述L1表示所述静态模型的模型系数的个数,所述L2表示所述动态模型的模型系数的个数,所述r表示所述预设输入信号组中的输入信号的功率量级,所述r为大于或等于1的整数,所述
Figure PCTCN2015075606-appb-000035
表示所述预设输入信号组的输出信号组,所述
Figure PCTCN2015075606-appb-000036
表示所述预设输入信号组的静态模型,所述
Figure PCTCN2015075606-appb-000037
表示所述预设输入信号组的动态模型,所述
Figure PCTCN2015075606-appb-000038
表示所述预设输入信号组的动态线性模型,所述
Figure PCTCN2015075606-appb-000039
表示所述预设输入信号组的动态非线性模型,所述
Figure PCTCN2015075606-appb-000040
表示所述预设输入信号组的动态线性模型的调整因子,所述
Figure PCTCN2015075606-appb-000041
表示所述预设输入信号组的动态非线性模型的调整因子;
第一代入模块,用于将所述当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000042
和动态模型的模型系数
Figure PCTCN2015075606-appb-000043
代入所述放大器模型公式,得到所述当前输入信号对应的放大器模型。
结合第一种可实现方式,在第二种可实现方式中,所述第一确定模块包括:
第一确定子模块,用于根据所述预选输入信号的动态线性模型的调整因子确定所述预设输入信号组的动态线性模型的调整因子;
第二确定子模块,用于根据所述预选输入信号的动态非线性模型的调整因子确定所述预设输入信号组的动态非线性模型的调整因子;
代入子模块,用于将所述预设输入信号组的动态线性模型的调整因子、所述预设输入信号组的动态非线性模型的调整因子代入所述放大器模型公式,得到所述预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数;
第一处理子模块,用于将所述预设输入信号组的静态模型的模型系数作为所述当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000044
将所述预设输入信号组的动态模型的模型系数作为所述当前输入信号的动态模型的模型系数
Figure PCTCN2015075606-appb-000045
结合第二种可实现方式,在第三种可实现方式中,所述预选输入信号为所述预设输入信号组中功率值最高的第一输入信号和功率值最低的第二输入信号,所述第一确定单元包括:
第二确定模块,用于根据所述第一输入信号的动态线性特性参数和线性调整因子公式,确定所述所述第一输入信号的动态线性模型的调整因子;
第三确定模块,用于根据所述第二输入信号的动态线性特性参数的参数值和线性调整因子公式,确定所述所述第二输入信号的动态线性模型的调整因子;
所述线性调整因子公式为:
Figure PCTCN2015075606-appb-000046
其中,所述θi表示所述第一输入信号的动态线性特性参数的参数值或所述第二输入信号的动态线性特性参数的参数值,所述i表示所述第一输入信号的功率量级或所述第二输入信号的功率量级,所述θ1表示第一输入信号的动态线性特性参数的参数值,所述
Figure PCTCN2015075606-appb-000047
表示第一输入信号的动态线性模型的调整因子或第二输入信号的动态线性模型的调整因子。
结合第三种可实现方式,在第四种可实现方式中,
所述动态线性特性参数为功率放大器中所述预设输入信号组对应的输出信号的平均功率或者功率放大器中所述预设输入信号组对应的输出信号的增益。
结合第三种可实现方式,在第五种可实现方式中,
所述第二确定单元包括:
第四确定模块,用于根据所述预选输入信号的动态线性特性参数的参数值和非线性特性公式确定所述预选输入信号的动态非线性特性参数的参数值,所述非线性特性公式为:
Figure PCTCN2015075606-appb-000048
其中,所述
Figure PCTCN2015075606-appb-000049
表示所述预选输入信号的动态非线性特性参数的参数值,所述x(i)(n)表示所述预选输入信号的信号值,所述y(i)(n)表示所述预选输入信号对应的输出信号;
第五确定模块,用于根据所述预选输入信号的动态非线性特性参数的参数值和非线性调整因子公式确定所述预选输入信号的动态非线性模型的调整因子,所述非线性调整因子公式为:
Figure PCTCN2015075606-appb-000050
其中,所述
Figure PCTCN2015075606-appb-000051
表示所述第一输入信号的动态非线性特性参数的参数值。
结合第五种可实现方式,在第六种可实现方式中,
所述第一确定子模块具体用于:根据所述预选输入信号的动态线性模型的调整因子,采用第一插值公式确定所述预设输入信号组的动态线性模型的调整因子
Figure PCTCN2015075606-appb-000052
所述第一插值公式为:
Figure PCTCN2015075606-appb-000053
其中,所述r表示所述预设输入信号组中的输入信号的功率量级小于所述第一输入信号的功率量级M,大于所述第二输入信号的功率量级m,所述
Figure PCTCN2015075606-appb-000054
表示所述第一输入信号的动态线性模型的调整因子,所述
Figure PCTCN2015075606-appb-000055
表示所述第二输入信号的动态线性模型的调整因子,所述w(r)表示权重因子,所述权重因子w(r)根据权重公式确定得到,所述权重公式为:
Figure PCTCN2015075606-appb-000056
其中,所述Pr表示功率量级为r的输入信号的功率值,所述PM表示功率量级为M的所述第一输入信号的功率值;
所述第二确定子模块具体用于:根据所述预选输入信号的动态非线性模型 的调整因子,采用第二插值公式确定所述预设输入信号组的动态非线性模型的调整因子
Figure PCTCN2015075606-appb-000057
所述第二插值公式为:
Figure PCTCN2015075606-appb-000058
其中,所述
Figure PCTCN2015075606-appb-000059
表示所述第一输入信号的动态非线性模型的调整因子,所述
Figure PCTCN2015075606-appb-000060
表示所述第二输入信号的动态非线性模型的调整因子,所述w(r)表示权重因子。
结合第二方面至第六种可实现方式中的任意一种,在第七种可实现方式中,所述数字预失真校正装置还包括:
更新单元,用于当所述功率放大器的状态变化时,更新所述当前输入信号的放大器模型得到更新后的放大器模型。
结合第七种可实现方式,在第八种可实现方式中,
所述功率放大器的状态变化为器件老化、温度波动或者偏置电压变化。
结合第七或第八种可实现方式,在第九种可实现方式中,所述更新单元包括:
第一获取模块,用于根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的动态模型的模型系数;
第二获取模块,用于根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的静态模型的模型系数;
第二代入模块,用于将所述更新后的放大器模型的动态模型的模型系数和所述更新后的放大器模型的静态模型的模型系数代入所述放大器模型得到所述更新后的放大器模型。
结合第九种可实现方式,在第十种可实现方式中,所述第一获取模块包括:
第一差值子模块,用于将所述功率放大器的状态变化后的输入信号的动态模型的模型系数与变化前的输入信号的动态模型的模型系数之差作为第一差值;
第二处理子模块,用于将所述当前输入信号的动态模型的模型系数与所述第一差值之和作为所述更新后的放大器模型的动态模型的模型系数。
结合第九种可实现方式,在第十一种可实现方式中,所述第二获取模块包括:
第二差值子模块,用于将所述功率放大器的状态变化后的输入信号的静态模型的模型系数与变化前的输入信号的静态模型的模型系数之差作为第二差 值;
第三处理子模块,用于将所述当前输入信号的静态模型的模型系数与所述第二差值之和作为所述更新后的放大器模型的静态模型的模型系数。
第三方面,提供了一种数字预失真校正装置,所述数字预失真校正装置包括:
处理器,用于根据预选输入信号确定放大器模型中的所述预选输入信号的动态线性模型的调整因子,所述放大器模型用于指示输出信号与输入信号的静态模型、输入信号的动态模型的关系,所述动态模型包括动态线性模型和动态非线性模型,其中,所述动态线性模型用于指示输入信号的变量中的线性特性,所述动态非线性模型用于指示输入信号的变量中的非线性特性,所述预选输入信号为预设输入信号组中的至少两个信号,所述预设输入信号组包括功率值不同的多个输入信号;
所述处理器还用于根据所述预选输入信号确定所述放大器模型中的所述预选输入信号的动态非线性模型的调整因子;
所述处理器还用于根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型;
所述处理器还用于根据所述当前输入信号的放大器模型获得所述当前输入信号的数字预失真DPD模型;
所述处理器还用于根据所述当前输入信号的DPD模型对所述当前输入信号进行数字预失真校正。
结合第三方面,在第一种可实现方式中,所述处理器具体用于:
根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定所述当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000061
和动态模型的模型系数
Figure PCTCN2015075606-appb-000062
所述放大器模型公式为:
Figure PCTCN2015075606-appb-000063
其中,所述N表示所述预设输入信号组中每个输入信号的采样点个数,所述L1表示所述静态模型的模型系数的个数,所述L2表示所述动态模型的模型系数的个数,所述r表示所述预设输入信号组中的输入信号的功率量级,所述r为大于或等于1的整数,所述
Figure PCTCN2015075606-appb-000064
表示所述预设输入信号组的输出信号组,所 述
Figure PCTCN2015075606-appb-000065
表示所述预设输入信号组的静态模型,所述
Figure PCTCN2015075606-appb-000066
表示所述预设输入信号组的动态模型,所述
Figure PCTCN2015075606-appb-000067
表示所述预设输入信号组的动态线性模型,所述
Figure PCTCN2015075606-appb-000068
表示所述预设输入信号组的动态非线性模型,所述
Figure PCTCN2015075606-appb-000069
表示所述预设输入信号组的动态线性模型的调整因子,所述
Figure PCTCN2015075606-appb-000070
表示所述预设输入信号组的动态非线性模型的调整因子;
将所述当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000071
和动态模型的模型系数
Figure PCTCN2015075606-appb-000072
代入所述放大器模型公式,得到所述当前输入信号对应的放大器模型。
结合第一种可实现方式,在第二种可实现方式中,
所述处理器具体用于:
根据所述预选输入信号的动态线性模型的调整因子确定所述预设输入信号组的动态线性模型的调整因子;
根据所述预选输入信号的动态非线性模型的调整因子确定所述预设输入信号组的动态非线性模型的调整因子;
将所述预设输入信号组的动态线性模型的调整因子、所述预设输入信号组的动态非线性模型的调整因子代入所述放大器模型公式,得到所述预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数;
将所述预设输入信号组的静态模型的模型系数作为所述当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000073
将所述预设输入信号组的动态模型的模型系数作为所述当前输入信号的动态模型的模型系数
Figure PCTCN2015075606-appb-000074
结合第二种可实现方式,在第三种可实现方式中,所述预选输入信号为所述预设输入信号组中功率值最高的第一输入信号和功率值最低的第二输入信号,所述处理器具体用于:
根据所述第一输入信号的动态线性特性参数和线性调整因子公式,确定所述所述第一输入信号的动态线性模型的调整因子;
根据所述第二输入信号的动态线性特性参数的参数值和线性调整因子公式,确定所述所述第二输入信号的动态线性模型的调整因子;
所述线性调整因子公式为:
Figure PCTCN2015075606-appb-000075
其中,所述θi表示所述第一输入信号的动态线性特性参数的参数值或所述 第二输入信号的动态线性特性参数的参数值,所述i表示所述第一输入信号的功率量级或所述第二输入信号的功率量级,所述θ1表示第一输入信号的动态线性特性参数的参数值,所述
Figure PCTCN2015075606-appb-000076
表示第一输入信号的动态线性模型的调整因子或第二输入信号的动态线性模型的调整因子。
结合第三种可实现方式,在第四种可实现方式中,
所述动态线性特性参数为功率放大器中所述预设输入信号组对应的输出信号的平均功率或者功率放大器中所述预设输入信号组对应的输出信号的增益。
结合第三种可实现方式,在第五种可实现方式中,
所述处理器具体用于:
根据所述预选输入信号的动态线性特性参数的参数值和非线性特性公式确定所述预选输入信号的动态非线性特性参数的参数值,所述非线性特性公式为:
Figure PCTCN2015075606-appb-000077
其中,所述
Figure PCTCN2015075606-appb-000078
表示所述预选输入信号的动态非线性特性参数的参数值,所述x(i)(n)表示所述预选输入信号的信号值,所述y(i)(n)表示所述预选输入信号对应的输出信号;
根据所述预选输入信号的动态非线性特性参数的参数值和非线性调整因子公式确定所述预选输入信号的动态非线性模型的调整因子,所述非线性调整因子公式为:
Figure PCTCN2015075606-appb-000079
其中,所述
Figure PCTCN2015075606-appb-000080
表示所述第一输入信号的动态非线性特性参数的参数值。
结合第五种可实现方式,在第六种可实现方式中,
所述处理器具体用于:
根据所述预选输入信号的动态线性模型的调整因子,采用第一插值公式确定所述预设输入信号组的动态线性模型的调整因子
Figure PCTCN2015075606-appb-000081
所述第一插值公式为:
Figure PCTCN2015075606-appb-000082
其中,所述r表示所述预设输入信号组中的输入信号的功率量级小于所述第一输入信号的功率量级M,大于所述第二输入信号的功率量级m,所述
Figure PCTCN2015075606-appb-000083
表示所述第一输入信号的动态线性模型的调整因子,所述
Figure PCTCN2015075606-appb-000084
表示所述第二输入信号的动态线性模型的调整因子,所述w(r)表示权重因子,所述权重因子w(r)根据权重公式确定得到,所述权重公式为:
Figure PCTCN2015075606-appb-000085
其中,所述Pr表示功率量级为r的输入信号的功率值,所述PM表示功率量级为M的所述第一输入信号的功率值;
所述处理器还具体用于根据所述预选输入信号的动态非线性模型的调整因子,采用第二插值公式确定所述预设输入信号组的动态非线性模型的调整因子
Figure PCTCN2015075606-appb-000086
所述第二插值公式为:
Figure PCTCN2015075606-appb-000087
其中,所述
Figure PCTCN2015075606-appb-000088
表示所述第一输入信号的动态非线性模型的调整因子,所述
Figure PCTCN2015075606-appb-000089
表示所述第二输入信号的动态非线性模型的调整因子,所述w(r)表示权重因子。
结合第三方面至第六种可实现方式中的任意一种,在第七种可实现方式中,所述处理器还用于:
当所述功率放大器的状态变化时,更新所述当前输入信号的放大器模型得到更新后的放大器模型。
结合第七种可实现方式,在第八种可实现方式中,
所述功率放大器的状态变化为器件老化、温度波动或者偏置电压变化。
结合第七或第八种可实现方式,在第九种可实现方式中,所述处理器具体用于:
根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的动态模型的模型系数;
根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的静态模型的模型系数;
将所述更新后的放大器模型的动态模型的模型系数和所述更新后的放大器模型的静态模型的模型系数代入所述放大器模型得到所述更新后的放大器模型。
结合第九种可实现方式,在第十种可实现方式中,
所述处理器还具体用于:
将所述功率放大器的状态变化后的输入信号的动态模型的模型系数与变化前的输入信号的动态模型的模型系数之差作为第一差值;
将所述当前输入信号的动态模型的模型系数与所述第一差值之和作为所述更新后的放大器模型的动态模型的模型系数。
结合第九种可实现方式,在第十一种可实现方式中,
所述处理器还具体用于:
将所述功率放大器的状态变化后的输入信号的静态模型的模型系数与变化前的输入信号的静态模型的模型系数之差作为第二差值;
将所述当前输入信号的静态模型的模型系数与所述第二差值之和作为所述更新后的放大器模型的静态模型的模型系数。
本发明提供的技术方案的有益效果是:
本发明提供了一种数字预失真校正方法及装置,由于可以根据预选输入信号确定放大器模型中的该预选输入信号的动态线性模型的调整因子和动态非线性模型的调整因子,然后再根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型,继而根据当前输入信号的放大器模型获得当前输入信号的DPD模型,最后根据DPD模型对当前输入信号进行数字预失真校正,相较于查找表模型,无需计算并存储覆盖足够的功率范围的放大器模型的模型参数,从而无需计算并存储覆盖足够的功率范围的DPD模型的模型参数来满足不同输入信号的校正要求,因此,简化了模型参数的获取过程,提高了校正效率。
附图说明
为了更清楚地说明本发明实施例中的技术方案,下面将对实施例描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本发明实施例提供的现有DPD技术的基本工作原理图;
图2是本发明实施例提供的一种数字预失真校正方法的流程图;
图3是本发明实施例提供的另一种数字预失真校正方法的流程图;
图4是本发明实施例提供的一种放大器模型的示意图;
图5是本发明实施例提供的一种根据预选输入信号确定放大器模型中的预选输入信号的动态线性模型的调整因子方法的流程图;
图6是本发明实施例提供的一种根据预选输入信号确定放大器模型中的预选输入信号的动态非线性模型的调整因子方法的流程图;
图7是本发明实施例提供的一种确定预选输入信号对应的静态模型的模型系数和动态模型的模型系数的示意图;
图8是本发明实施例提供的一种根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型方法的流程图;
图9是本发明实施例提供的一种根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定当前输入信号的静态模型的模型系数和动态模型的模型系数方法的流程图;
图10是本发明实施例提供的一种确定预设输入信号组中没有被训练的任意一个输入信号对应的动态线性模型的调整因子和动态非线性模型的调整因子的示意图;
图11是本发明实施例提供的一种更新当前输入信号的放大器模型得到更新后的放大器模型方法的流程图;
图12是本发明实施例提供的一种获取更新后的放大器模型的动态模型的模型系数方法的流程图;
图13是本发明实施例提供的一种获取更新后的放大器模型的静态模型的模型系数方法的流程图;
图14是本发明实施例提供的一种放大器模型的模型系数更新的示意图;
图15是本发明实施例提供的一种数字预失真校正装置的结构示意图;
图16是本发明实施例提供的另一种数字预失真校正装置的结构示意图;
图17是本发明实施例提供的一种数字预失真校正装置的建立单元的结构示意图;
图18是本发明实施例提供的一种数字预失真校正装置的第一确定模块的结构示意图;
图19是本发明实施例提供的一种数字预失真校正装置的第一确定单元的结构示意图;
图20是本发明实施例提供的一种数字预失真校正装置的第二确定单元的 结构示意图;
图21是本发明实施例提供的一种数字预失真校正装置的更新单元的结构示意图;
图22是本发明实施例提供的一种数字预失真校正装置的第一获取模块的结构示意图;
图23是本发明实施例提供的一种数字预失真校正装置的第二获取模块的结构示意图;
图24是本发明实施例提供的又一种数字预失真校正装置的结构示意图。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚,下面将结合附图对本发明实施方式作进一步地详细描述。
DPD技术是一种通过在数字域对功率放大器的非线性进行补偿的关键技术,可以使功率放大器工作在高效率的饱和状态而不丢失线性度。
如图1所示,DPD技术的基本工作原理是:在数字基带建立一个DPD模块01,对输入信号在进入功率放大器(英文:Power Amplifier;简称:PA)02之前进行预失真,该DPD模块01的模型为DPD模型,PA02的模型为放大器模型,如果DPD模型为放大器模型的逆函数,则当输入信号经过级联的DPD模块01和PA02后,会被线性放大,从而避免了输入信号在经过PA02后的输出信号产生失真。图1中,横坐标X表示输入信号的输入功率,纵坐标Y表示输入信号的输出功率,左数第一个功率曲线图表示输入信号的功率曲线,第二个功率曲线图表示DPD模块01产生的预失真信号的功率曲线,第三个功率曲线图表示输入信号经预失真校正后从PA02输出的输出信号的功率曲线。由此可知,DPD技术关键问题在于建立一个精确但又简单的放大器模型:首先,放大器模型应该精确,因为只有根据功率放大器的特征准确建立放大器模型,才能对输入信号进行预失真校正;其次,放大器模型应该足够简单,因为复杂的放大器模型会增加实际硬件实现的成本和***复杂度,在实际中无法得到广泛的应用。
本发明实施例提供一种数字预失真校正方法,如图2所示,该方法包括:
步骤101、根据预选输入信号确定放大器模型中的预选输入信号的动态线性模型的调整因子,该放大器模型用于指示输出信号与输入信号的静态模型、 输入信号的动态模型的关系,动态模型包括动态线性模型和动态非线性模型,其中,动态线性模型用于指示输入信号的变量中的线性特性,动态非线性模型用于指示输入信号的变量中的非线性特性,预选输入信号为预设输入信号组中的至少两个信号,预设输入信号组包括功率值不同的多个输入信号。
步骤102、根据预选输入信号确定放大器模型中的预选输入信号的动态非线性模型的调整因子。
步骤103、根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型。
步骤104、根据当前输入信号的放大器模型获得当前输入信号的DPD模型。
步骤105、根据当前输入信号的DPD模型对当前输入信号进行数字预失真校正。
综上所述,本发明实施例提供的数字预失真校正方法,由于可以根据预选输入信号确定放大器模型中的该预选输入信号的动态线性模型的调整因子和动态非线性模型的调整因子,然后再根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型,继而根据当前输入信号的放大器模型获得当前输入信号的DPD模型,最后根据DPD模型对当前输入信号进行数字预失真校正,相较于查找表模型,无需计算并存储覆盖足够的功率范围的放大器模型的模型参数,从而无需计算并存储覆盖足够的功率范围的DPD模型的模型参数来满足不同输入信号的校正要求,因此,简化了模型参数的获取过程,提高了校正效率。
进一步的,步骤103具体包括:
根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000090
和动态模型的模型系数
Figure PCTCN2015075606-appb-000091
该放大器模型公式为:
Figure PCTCN2015075606-appb-000092
其中,N表示预设输入信号组中每个输入信号的采样点个数,L1表示静态模型的模型系数的个数,L2表示动态模型的模型系数的个数,r表示预设输入信号组中的输入信号的功率量级,r为大于或等于1的整数,
Figure PCTCN2015075606-appb-000093
表示预设输入信号组的输出信号组,
Figure PCTCN2015075606-appb-000094
为N×1的矩阵,
Figure PCTCN2015075606-appb-000095
表示预设输入信号组的静 态模型,
Figure PCTCN2015075606-appb-000096
为N×L1的矩阵,
Figure PCTCN2015075606-appb-000097
表示预设输入信号组的动态模型,
Figure PCTCN2015075606-appb-000098
为N×L2的矩阵,
Figure PCTCN2015075606-appb-000099
表示预设输入信号组的动态线性模型,
Figure PCTCN2015075606-appb-000100
表示预设输入信号组的动态非线性模型,
Figure PCTCN2015075606-appb-000101
表示预设输入信号组的动态线性模型的调整因子,
Figure PCTCN2015075606-appb-000102
表示预设输入信号组的动态非线性模型的调整因子,
Figure PCTCN2015075606-appb-000103
为L1×1的矩阵,
Figure PCTCN2015075606-appb-000104
为L2×1的矩阵;将当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000105
和动态模型的模型系数
Figure PCTCN2015075606-appb-000106
代入放大器模型公式,得到当前输入信号对应的放大器模型。
其中,根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000107
和动态模型的模型系数
Figure PCTCN2015075606-appb-000108
包括:
根据预选输入信号的动态线性模型的调整因子确定预设输入信号组的动态线性模型的调整因子;根据预选输入信号的动态非线性模型的调整因子确定预设输入信号组的动态非线性模型的调整因子;将预设输入信号组的动态线性模型的调整因子、预设输入信号组的动态非线性模型的调整因子代入放大器模型公式,得到确定预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数;将预设输入信号组的静态模型的模型系数作为当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000109
将预设输入信号组的动态模型的模型系数作为当前输入信号的动态模型的模型系数
Figure PCTCN2015075606-appb-000110
需要说明的是,预选输入信号为预设输入信号组中的至少两个信号。以预设输入信号组中功率值最高的第一输入信号和功率值最低的第二输入信号为例,则根据预选输入信号确定放大器模型中的预选输入信号的动态线性模型的调整因子,包括:
根据第一输入信号的动态线性特性参数和线性调整因子公式,确定第一输入信号的动态线性模型的调整因子;根据第二输入信号的动态线性特性参数的参数值和线性调整因子公式,确定第二输入信号的动态线性模型的调整因子,该线性调整因子公式为:
Figure PCTCN2015075606-appb-000111
其中,θi表示第一输入信号的动态线性特性参数的参数值或第二输入信号的动态线性特性参数的参数值,i表示第一输入信号的功率量级或第二输入信号的功率量级,θ1表示第一输入信号的动态线性特性参数的参数值,
Figure PCTCN2015075606-appb-000112
表示第 一输入信号的动态线性模型的调整因子或第二输入信号的动态线性模型的调整因子。
需要说明的是,动态线性特性参数可以为功率放大器中预设输入信号组对应的输出信号的平均功率或者功率放大器中预设输入信号组对应的输出信号的增益。
相应的,根据预选输入信号确定放大器模型中的预选输入信号的动态非线性模型的调整因子,包括:
根据预选输入信号的动态线性特性参数的参数值和非线性特性公式确定预选输入信号的动态非线性特性参数的参数值,该非线性特性公式可以为:
Figure PCTCN2015075606-appb-000113
其中,
Figure PCTCN2015075606-appb-000114
表示预选输入信号的动态非线性特性参数的参数值,x(i)(n)表示预选输入信号的信号值,y(i)(n)表示预选输入信号对应的输出信号,θi表示第一输入信号的动态线性特性参数的参数值或第二输入信号的动态线性特性参数的参数值。
根据预选输入信号的动态非线性特性参数的参数值和非线性调整因子公式确定预选输入信号的动态非线性模型的调整因子,该非线性调整因子公式为:
Figure PCTCN2015075606-appb-000115
其中,
Figure PCTCN2015075606-appb-000116
表示第一输入信号的动态非线性特性参数的参数值,
Figure PCTCN2015075606-appb-000117
表示预选输入信号的动态非线性特性参数的参数值,
Figure PCTCN2015075606-appb-000118
表示第一输入信号的动态非线性模型的调整因子或第二输入信号的动态非线性模型的调整因子。
进一步的,根据预选输入信号的动态线性模型的调整因子确定预设输入信号组的动态线性模型的调整因子,包括:
根据预选输入信号的动态线性模型的调整因子,采用第一插值公式确定预设输入信号组的动态线性模型的调整因子
Figure PCTCN2015075606-appb-000119
该第一插值公式可以为:
Figure PCTCN2015075606-appb-000120
其中,r表示预设输入信号组中的输入信号的功率量级小于第一输入信号的功率量级M,大于第二输入信号的功率量级m,
Figure PCTCN2015075606-appb-000121
表示第一输入信号的动 态线性模型的调整因子,
Figure PCTCN2015075606-appb-000122
表示第二输入信号的动态线性模型的调整因子,w(r)表示权重因子,权重因子w(r)根据权重公式确定得到,该权重公式可以为:
Figure PCTCN2015075606-appb-000123
其中,Pr表示功率量级为r的输入信号的功率值,PM表示功率量级为M的第一输入信号的功率值。
根据预选输入信号的动态非线性模型的调整因子确定预设输入信号组的动态非线性模型的调整因子,包括:
根据预选输入信号的动态非线性模型的调整因子,采用第二插值公式确定预设输入信号组的动态非线性模型的调整因子
Figure PCTCN2015075606-appb-000124
该第二插值公式可以为:
Figure PCTCN2015075606-appb-000125
其中,
Figure PCTCN2015075606-appb-000126
表示第一输入信号的动态非线性模型的调整因子,
Figure PCTCN2015075606-appb-000127
表示第二输入信号的动态非线性模型的调整因子,w(r)表示权重因子。
在步骤105之后,该方法还可以包括:当功率放大器的状态变化时,更新当前输入信号的放大器模型得到更新后的放大器模型。功率放大器的状态变化为器件老化、温度波动或者偏置电压变化。
具体的,更新当前输入信号的放大器模型得到更新后的放大器模型,包括:
根据当前输入信号的放大器模型,获取更新后的放大器模型的动态模型的模型系数;根据当前输入信号的放大器模型,获取更新后的放大器模型的静态模型的模型系数;将更新后的放大器模型的动态模型的模型系数和更新后的放大器模型的静态模型的模型系数代入放大器模型得到更新后的放大器模型。
其中,根据当前输入信号的放大器模型,获取更新后的放大器模型的动态模型的模型系数,包括:
将功率放大器的状态变化后的输入信号的动态模型的模型系数与变化前的输入信号的动态模型的模型系数之差作为第一差值;将当前输入信号的动态模型的模型系数与第一差值之和作为更新后的放大器模型的动态模型的模型系数。
根据当前输入信号的放大器模型,获取更新后的放大器模型的静态模型的模型系数,包括:
将功率放大器的状态变化后的输入信号的静态模型的模型系数与变化前 的输入信号的静态模型的模型系数之差作为第二差值;将当前输入信号的静态模型的模型系数与第二差值之和作为更新后的放大器模型的静态模型的模型系数。
综上所述,本发明实施例提供的数字预失真校正方法,由于可以根据预选输入信号确定放大器模型中的该预选输入信号的动态线性模型的调整因子和动态非线性模型的调整因子,然后再根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型,继而根据当前输入信号的放大器模型获得当前输入信号的DPD模型,最后根据DPD模型对当前输入信号进行数字预失真校正,相较于查找表模型,无需计算并存储覆盖足够的功率范围的放大器模型的模型参数,从而无需计算并存储覆盖足够的功率范围的DPD模型的模型参数来满足不同输入信号的校正要求,因此,简化了模型参数的获取过程,提高了校正效率。
本发明实施例提供另一种数字预失真校正方法,如图3所示,该方法包括:
步骤201、建立放大器模型。
该放大器模型公式为:
Figure PCTCN2015075606-appb-000128
其中,N表示预设输入信号组中每个输入信号的采样点个数,预设输入信号组包括功率值不同的多个输入信号,L1表示静态模型的模型系数的个数,L2表示动态模型的模型系数的个数,r表示预设输入信号组中的输入信号的功率量级,r为大于或等于1的整数,
Figure PCTCN2015075606-appb-000129
表示预设输入信号组的输出信号组,
Figure PCTCN2015075606-appb-000130
为N×1的矩阵,
Figure PCTCN2015075606-appb-000131
表示预设输入信号组的静态模型,
Figure PCTCN2015075606-appb-000132
为N×L1的矩阵,
Figure PCTCN2015075606-appb-000133
表示预设输入信号组的动态模型,
Figure PCTCN2015075606-appb-000134
为N×L2的矩阵,
Figure PCTCN2015075606-appb-000135
表示预设输入信号组的动态线性模型,
Figure PCTCN2015075606-appb-000136
表示预设输入信号组的动态非线性模型,
Figure PCTCN2015075606-appb-000137
表示预设输入信号组的动态线性模型的调整因子,
Figure PCTCN2015075606-appb-000138
表示预设输入信号组的动态非线性模型的调整因子,
Figure PCTCN2015075606-appb-000139
表示静态模型的模型系数,
Figure PCTCN2015075606-appb-000140
为L1×1的矩阵,
Figure PCTCN2015075606-appb-000141
表示动态模型的模型系数,
Figure PCTCN2015075606-appb-000142
为L2×1的矩阵。
示例的,预设输入信号组中的输入信号按照功率值从大到小的顺序排列,依次为:S1,S2,S3,S4,S5,S6,S7,S8,S9和S10,那么S1,S2,S3,S4,S5,S6,S7,S8,S9,S10对应的功率量级r分别为1,2,3,4,5,6,7,8,9和10。
如图4所示,该放大器模型包括三个部分,分别为预设输入信号组的静态模型X(r)、动态线性模型
Figure PCTCN2015075606-appb-000143
和动态非线性模型
Figure PCTCN2015075606-appb-000144
需要说明的是,该放大器模型公式中的动态线性模型的调整因子
Figure PCTCN2015075606-appb-000145
动态非线性模型的调整因子
Figure PCTCN2015075606-appb-000146
静态模型的模型系数
Figure PCTCN2015075606-appb-000147
和动态模型的模型系数
Figure PCTCN2015075606-appb-000148
是未知的。
步骤202、根据预选输入信号确定放大器模型中的预选输入信号的动态线性模型的调整因子。
该放大器模型用于指示输出信号与输入信号的静态模型、输入信号的动态模型的关系,动态模型包括动态线性模型和动态非线性模型,其中,动态线性模型用于指示输入信号的变量中的线性特性,动态非线性模型用于指示输入信号的变量中的非线性特性,预选输入信号为预设输入信号组中的至少两个信号,预设输入信号组包括功率值不同的多个输入信号。
需要说明的是,为了使预选输入信号能够覆盖整个预设输入信号组的功率范围,预选输入信号可以为预设输入信号组中功率值最高的输入信号和功率值最低的输入信号,也可以为预设输入信号组中功率值最高的输入信号、功率值最低的输入信号和其它功率值对应的输入信号。示例的,预设输入信号组中的输入信号按照功率值从大到小的顺序排列,依次为:S1,S2,S3,S4,S5,S6,S7,S8,S9和S10,则预选输入信号可以为S1和S10,也可以为S1,S4,S7和S10。本发明对预选输入信号选取的方法和选取的预选输入信号的个数不做限定,如选取的方法可以根据预设输入信号组中的输入信号对应的功率值进行平均划分,来确定预选输入信号;而预选输入信号的个数可以为两个,也可以大于两个。
假设,预选输入信号为预设输入信号组中功率值最高的第一输入信号和功率值最低的第二输入信号。则步骤202,如图5所示,可以包括:
步骤2021、根据第一输入信号的动态线性特性参数和线性调整因子公式,确定第一输入信号的动态线性模型的调整因子。
该线性调整因子公式为:
Figure PCTCN2015075606-appb-000149
其中,θi表示第一输入信号的动态线性特性参数的参数值或第二输入信号的动态线性特性参数的参数值,i表示第一输入信号的功率量级或第二输入信号的功率量级,θ1表示第一输入信号的动态线性特性参数的参数值,
Figure PCTCN2015075606-appb-000150
表示第一输入信号的动态线性模型的调整因子或第二输入信号的动态线性模型的调整因子。需要说明的是,i表示的是预选输入信号的功率量级,预选输入信号可以为预设输入信号组中功率值最高的第一输入信号和功率值最低的第二输入信号,也可以为预设输入信号组中功率值最高的输入信号、功率值最低的输入信号和其它功率值对应的输入信号,本发明实施例对预选输入信号的选取不做限定。
进一步的,能够用来表示输入信号的动态线性特性的参数即为动态线性特性参数,示例的,动态线性特性参数可以为功率放大器中预设输入信号组对应的输出信号的平均功率或者功率放大器中预设输入信号组对应的输出信号的增益。
步骤2022、根据第二输入信号的动态线性特性参数的参数值和线性调整因子公式,确定第二输入信号的动态线性模型的调整因子。
确定第二输入信号的动态线性模型的调整因子的具体过程可以参考步骤2021中的确定过程。
步骤203、根据预选输入信号确定放大器模型中的预选输入信号的动态非线性模型的调整因子。
为了量化功率放大器的动态非线性特性的变化,本发明实施例采用一种非线性测量的方法,即通过计算与功率放大器的特性的最佳动态线性特性近似之间的差异来表示剩余的动态非线性特性。
如步骤202中预选输入信号为预设输入信号组中功率值最高的第一输入信号和功率值最低的第二输入信号,则相应的,步骤203如图6所示,可以包括:
步骤2031、根据预选输入信号的动态线性特性参数的参数值和非线性特性公式确定预选输入信号的动态非线性特性参数的参数值。
该非线性特性公式可以为:
Figure PCTCN2015075606-appb-000151
其中,
Figure PCTCN2015075606-appb-000152
表示预选输入信号的动态非线性特性参数的参数值,x(i)(n)表示 预选输入信号的信号值,y(i)(n)表示预选输入信号对应的输出信号,θi表示第一输入信号的动态线性特性参数的参数值或第二输入信号的动态线性特性参数的参数值。
步骤2032、根据预选输入信号的动态非线性特性参数的参数值和非线性调整因子公式确定预选输入信号的动态非线性模型的调整因子。
该非线性调整因子公式为:
Figure PCTCN2015075606-appb-000153
其中,
Figure PCTCN2015075606-appb-000154
表示第一输入信号的动态非线性特性参数的参数值,
Figure PCTCN2015075606-appb-000155
表示预选输入信号的动态非线性特性参数的参数值,
Figure PCTCN2015075606-appb-000156
表示第一输入信号的动态非线性模型的调整因子或第二输入信号的动态非线性模型的调整因子。
由步骤202和步骤203可知,根据第一输入信号的动态线性特性参数的参数值可以确定第一输入信号的动态线性模型的调整因子,根据第二输入信号的动态线性特性参数的参数值,可以确定第二输入信号的动态线性模型的调整因子;根据第一输入信号的动态线性特性参数的参数值可以确定第一输入信号的动态非线性特性参数的参数值,从而可以确定第一输入信号的动态非线性模型的调整因子,根据第二输入信号的动态线性特性参数的参数值可以确定第二输入信号的动态非线性特性参数的参数值,从而可以确定第二输入信号的动态非线性模型的调整因子。
步骤204、根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型。
根据第一输入信号的动态线性模型的调整因子和第一输入信号的动态非线性模型的调整因子、第二输入信号的动态线性模型的调整因子和第二输入信号的动态非线性模型的调整因子,即可以确定预设输入信号组的动态线性模型的调整因子和动态非线性模型的调整因子,进而可以根据最小二乘法,采用放大器模型公式确定预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数,最终确定当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000157
和动态模型的模型系数
Figure PCTCN2015075606-appb-000158
建立当前输入信号的放大器模型。其中,放大器模型公式可以参考步骤201中的放大器模型公式。
示例的,图7为根据预选输入信号的动态线性模型的调整因子和动态非线性模型的调整因子,确定预选输入信号对应的静态模型的模型系数和动态模型 的模型系数的求解过程,如图7所示,选取预设输入信号组中的一组预选输入信号,该预选输入信号可以为预设输入信号组中功率值最高的输入信号和功率值最低的输入信号,也可以为预设输入信号组中功率值最高的输入信号、功率值最低的输入信号和其它功率值对应的输入信号。根据采集到的预选输入信号,计算对应的动态线性模型的调整因子和动态非线性模型的调整因子,然后再根据最小二乘法,采用放大器模型公式确定预选输入信号对应的静态模型的模型系数和动态模型的模型系数。x(1)(n)至x(R)(n)表示预设输入信号组中的输入信号,y(1)(n)至y(R)(n)表示预设输入信号组对应的输出信号,
Figure PCTCN2015075606-appb-000159
Figure PCTCN2015075606-appb-000160
表示预设输入信号组中的输入信号对应的动态线性模型的调整因子,
Figure PCTCN2015075606-appb-000161
Figure PCTCN2015075606-appb-000162
表示预设输入信号组中的输入信号对应的动态非线性模型的调整因子,
Figure PCTCN2015075606-appb-000163
表示预设输入信号组的静态模型的模型系数,
Figure PCTCN2015075606-appb-000164
表示预设输入信号组的动态模型的模型系数。
如图8所示,步骤204具体可以包括:
步骤2041、根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000165
和动态模型的模型系数
Figure PCTCN2015075606-appb-000166
如图9所示,步骤2041具体可以包括:
步骤2041a、根据预选输入信号的动态线性模型的调整因子确定预设输入信号组的动态线性模型的调整因子。
具体的,步骤2041a可以包括:
根据预选输入信号的动态线性模型的调整因子,采用第一插值公式确定预设输入信号组的动态线性模型的调整因子
Figure PCTCN2015075606-appb-000167
该第一插值公式可以为:
Figure PCTCN2015075606-appb-000168
其中,r表示预设输入信号组中的输入信号的功率量级小于第一输入信号的功率量级M(M=1),大于第二输入信号的功率量级m,
Figure PCTCN2015075606-appb-000169
表示第一输入信号的动态线性模型的调整因子,
Figure PCTCN2015075606-appb-000170
表示第二输入信号的动态线性模型的调整因子,w(r)表示权重因子,权重因子w(r)根据权重公式确定得到,该权重公式可以为:
Figure PCTCN2015075606-appb-000171
其中,Pr表示功率量级为r的输入信号的功率值,PM表示功率量级为M(M=1)的第一输入信号的功率值。
当预设输入信号组中的输入信号的功率值小于第一输入信号的功率值,大于第二输入信号的功率值,则认为该输入信号的功率量级小于第一输入信号的功率量级M(M=1),大于第二输入信号的功率量级m。
步骤2041b、根据预选输入信号的动态非线性模型的调整因子确定预设输入信号组的动态非线性模型的调整因子。
具体的,步骤2041b可以包括:
根据预选输入信号的动态非线性模型的调整因子,采用第二插值公式确定预设输入信号组的动态非线性模型的调整因子
Figure PCTCN2015075606-appb-000172
该第二插值公式可以为:
Figure PCTCN2015075606-appb-000173
其中,
Figure PCTCN2015075606-appb-000174
表示第一输入信号的动态非线性模型的调整因子,
Figure PCTCN2015075606-appb-000175
表示第二输入信号的动态非线性模型的调整因子,w(r)表示权重因子。
预设输入信号组中的预选输入信号也称为被训练的输入信号,预设输入信号组中除预选输入信号以外的其他输入信号称为没有被训练的输入信号,由步骤2041a和步骤2041b可知,对于预设输入信号组中没有被训练的输入信号中的任意一个输入信号,通过第一插值公式和第二插值公式,可以得到该输入信号对应的动态线性模型的调整因子和动态非线性模型的调整因子。步骤2041a和步骤2041b可以是同时执行的,本发明实施例对其先后顺序不作限定。
示例的,图10为通过第一插值公式和第二插值公式,确定预设输入信号组中没有被训练的任意一个输入信号对应的动态线性模型的调整因子和动态非线性模型的调整因子的求解过程,具体说明可以参考步骤2041a和步骤2041b。M表示第一输入信号的功率量级,m表示第二输入信号的功率量级,x(M)(n)表示第一输入信号,(CM)表示第一输入信号的模型系数,y(M)(n)表示第一输入信号对应的输出信号,x(r)(n)表示功率值小于第一输入信号的功率量级M,大于第二输入信号的功率量级m的没有被训练的输入信号,(Cr)表示该输入信号的模型系数,y(r)(n)表示该输入信号对应的输出信号,x(m)(n)表示第二输入信号,(Cm)表示第二输入信号的模型系数,y(m)(n)表示第二输入信号对应的输出信号,w(r)表示权重因子。
对于预设输入信号组中没有被训练的任意一个输入信号,首先判断该输入 信号的功率值位于哪两个被训练的输入信号的功率值之间,然后再通过第一插值公式和第二插值公式,确定该输入信号对应的动态线性模型的调整因子和动态非线性模型的调整因子。假设预设输入信号组中的输入信号按照功率值从大到小的顺序排列,依次为:S1,S2,S3,S4,S5,S6,S7,S8,S9和S10,如果预选输入信号为S1和S10,则可以根据S1和S10,通过第一插值公式和第二插值公式确定S2,S3,S4,S5,S6,S7,S8和S9中任意一个输入信号对应的动态线性模型的调整因子和动态非线性模型的调整因子;如果预选输入信号为S1,S2、S6和S10,由于S3的功率值位于S2和S6的功率值之间,则可以根据S2和S6,通过第一插值公式和第二插值公式确定S3对应的动态线性模型的调整因子和动态非线性模型的调整因子,同样,可以确定S4,S5,S7,S8和S9中任意一个输入信号对应的动态线性模型的调整因子和动态非线性模型的调整因子。
步骤2041c、将预设输入信号组的动态线性模型的调整因子、预设输入信号组的动态非线性模型的调整因子代入放大器模型公式,得到预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数。
根据预选输入信号的动态线性模型的调整因子和预选输入信号的动态非线性模型的调整因子得到预设输入信号组的动态线性模型的调整因子和预设输入信号组的动态非线性模型的调整因子之后,便可以根据预设输入信号组的动态线性模型的调整因子、预设输入信号组的动态非线性模型的调整因子,根据最小二乘法,采用放大器模型公式来确定预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数。
步骤2041d、将预设输入信号组的静态模型的模型系数作为当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000176
将预设输入信号组的动态模型的模型系数作为当前输入信号的动态模型的模型系数
Figure PCTCN2015075606-appb-000177
预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数即为当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000178
和动态模型的模型系数
Figure PCTCN2015075606-appb-000179
步骤2042、将当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000180
和动态模型的模型系数
Figure PCTCN2015075606-appb-000181
代入放大器模型公式,得到当前输入信号对应的放大器模型。
已知当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000182
和动态模型的模型系数
Figure PCTCN2015075606-appb-000183
便可建立当前输入信号对应的放大器模型。
由此可知,本发明实施例提供的数字预失真校正方法中的模型系数的获取方式,一方面可以减少需要参与计算的不同输入信号对应的功率值的数量,因而降低了放大器模型建立的复杂度;另一方面,通过线性插值的方式来获取其余功率值对应的模型系数的方式,可以与放大器模型公式中原有的动态线性模型的调整因子和动态非线性模型的调整因子相结合,从而简单且快速地确定当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000184
和动态模型的模型系数
Figure PCTCN2015075606-appb-000185
步骤205、根据当前输入信号的放大器模型获得当前输入信号的DPD模型。
具体的,可以将当前输入信号的放大器模型的逆函数作为当前输入信号的DPD模型。如图1所示,当DPD模型为放大器模型的逆函数时,则当前输入信号经过级联的DPD模块01和PA02后,会被线性放大,从而避免了当前输入信号在经过PA02后的输出信号产生失真。
步骤206、根据当前输入信号的DPD模型对当前输入信号进行数字预失真校正。
将当前输入信号的DPD模型的逆函数对应产生的预失真信号,叠加到当前输入信号上,使叠加后的信号经过功率放大器模块,达到对当前输入信号进行预失真校正的目的。
步骤207、当功率放大器的状态变化时,更新当前输入信号的放大器模型得到更新后的放大器模型。
虽然步骤201至步骤206可以较好地在功率放大器的输出功率变化时,很好地拟合功率放大器的特性,但是由于功率放大器的工作状态会随着器件老化、温度波动或者偏置电压变化等原因而发生变化,因此,需要在功率放大器的状态变化时,更新当前输入信号的放大器模型得到更新后的放大器模型,以适应这些变化。示例的,功率放大器的状态变化可以为器件老化、温度波动或者偏置电压变化。需要说明的是,检测功率放大器的状态是否变化可以有多种方式,具体可以参考现有技术,本发明实施例在此不再赘述。当功率放大器的状态发生变化,按照步骤207中具体的更新方式更新当前输入信号的放大器模型得到更新后的放大器模型即可。
如图11所示,步骤207具体可以包括:
步骤2071、根据当前输入信号的放大器模型,获取更新后的放大器模型的动态模型的模型系数。
如图12所示,步骤2071具体可以包括:
步骤2071a、将功率放大器的状态变化后的输入信号的动态模型的模型系数与变化前的输入信号的动态模型的模型系数之差作为第一差值。
步骤2071b、将当前输入信号的动态模型的模型系数与第一差值之和作为更新后的放大器模型的动态模型的模型系数。
步骤2072、根据当前输入信号的放大器模型,获取更新后的放大器模型的静态模型的模型系数。
如图13所示,步骤2072具体可以包括:
步骤2072a、将功率放大器的状态变化后的输入信号的静态模型的模型系数与变化前的输入信号的静态模型的模型系数之差作为第二差值。
步骤2072b、将当前输入信号的静态模型的模型系数与第二差值之和作为更新后的放大器模型的静态模型的模型系数。
由于放大器模型公式中的静态模型和动态线性模型可以线性相关,也可以线性不相关,因此,更新放大器模型的动态模型的模型系数和静态模型的模型系数,可以包含两个方面:
一方面,当放大器模型公式中的静态模型和动态线性模型不相关时,可以确定状态变化后的模型系数
Figure PCTCN2015075606-appb-000186
(状态变化后的输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000187
与动态模型的模型系数
Figure PCTCN2015075606-appb-000188
),从而确定状态变化后的模型系数
Figure PCTCN2015075606-appb-000189
与状态变化前的模型系数C(r′)之间的差值(第二差值
Figure PCTCN2015075606-appb-000190
和第一差值
Figure PCTCN2015075606-appb-000191
):
Figure PCTCN2015075606-appb-000192
再将当前输入信号的动态模型的模型系数
Figure PCTCN2015075606-appb-000193
与第一差值
Figure PCTCN2015075606-appb-000194
之和作为更新后的放大器模型的动态模型的模型系数;将当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000195
与第二差值
Figure PCTCN2015075606-appb-000196
之和作为更新后的放大器模型的静态模型的模型系数:
Figure PCTCN2015075606-appb-000197
另一方面,当放大器模型公式中的静态模型和动态线性模型线性相关时,则该训练过程只得到一部分有效模型系数
Figure PCTCN2015075606-appb-000198
即存在:
Figure PCTCN2015075606-appb-000199
其中,
Figure PCTCN2015075606-appb-000200
再将当前输入信号的动态模型的模型系数
Figure PCTCN2015075606-appb-000201
与第一差值
Figure PCTCN2015075606-appb-000202
之和作为更新后的放大器模型的动态模型的模型系数;将当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000203
与第二差值
Figure PCTCN2015075606-appb-000204
之和作为更新后的放大器模型的静态模型的模型系数:
Figure PCTCN2015075606-appb-000205
步骤2073、将更新后的放大器模型的动态模型的模型系数和更新后的放大器模型的静态模型的模型系数代入放大器模型得到更新后的放大器模型。
示例的,图14为放大器模型的模型系数的更新过程,虚线以上的部分指示的是获取功率放大器的状态变化后的输入信号的动态模型的模型系数与变化前的输入信号的动态模型的模型系数的第一差值,及获取功率放大器的状态变化后的输入信号的静态模型的模型系数与变化前的输入信号的静态模型的模型系数的第二差值的过程,具体可以参考步骤2071a和步骤2072a,虚线以下的部分指示的是将当前输入信号的动态模型的模型系数与第一差值之和作为更新后的放大器模型的动态模型的模型系数,及将当前输入信号的静态模型的模型系数与第二差值之和作为更新后的放大器模型的静态模型的模型系数的过程,具体可以参考步骤2071b和2072b,图14中的x(1)(n)至x(R)(n)表示当前输入信号。
步骤208、将更新后的放大器模型的逆函数作为当前输入信号的DPD模型。
当功率放大器的状态变化时,更新当前输入信号的放大器模型得到更新后的放大器模型,如步骤205所述,再将更新后的放大器模型的逆函数作为当前输入信号的DPD模型,避免当前输入信号在经过PA02后的输出信号产生失真。
步骤209、根据更新后的放大器模型的逆函数对应的DPD模型对当前输入信号进行数字预失真校正。
步骤209具体可以参考步骤206,在此不再赘述。
需要说明的是,当需要更新当前输入信号的放大器模型时,可以选取预设输入信号组中一个较高功率值的输入信号作为图1中的放大器模块的输入信 号,然后根据重新求解出的模型系数计算出状态变化后的模型系数与状态变化前的模型系数、更新后的放大器模型的模型系数,再将其应用于其它的功率值中,获得功率放大器的状态改变后的新的放大器模型系数。这样放大器模型就可以通过偶尔的一两次系数更新,很快地适应功率放大器状态的变化,并对整个功率动态范围内的模型系数都进行及时地调整,以适应功率放大器的这种改变,使得放大器模型能够更好地适应现代无线通信***中的功率放大器功率的动态变化。
因此,根据本发明实施例提供的数字预失真校正方法中的建立放大器模型的过程可知,根据本发明实施例建立的放大器模型是一个精确但又简单的放大器模型,满足解决功率放大器的输出信号的功率动态变化场景下的DPD技术关键问题的要求。
需要说明的是,在本发明的各种实施例中,上述各过程的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本发明实施例的实施过程构成任何限定。
综上所述,本发明实施例提供的数字预失真校正方法,由于可以根据预选输入信号确定放大器模型中的该预选输入信号的动态线性模型的调整因子和动态非线性模型的调整因子,然后再根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型,继而根据当前输入信号的放大器模型获得当前输入信号的DPD模型,最后根据DPD模型对当前输入信号进行数字预失真校正,相较于查找表模型,无需计算并存储覆盖足够的功率范围的放大器模型的模型参数,从而无需计算并存储覆盖足够的功率范围的DPD模型的模型参数来满足不同输入信号的校正要求,因此,简化了模型参数的获取过程,提高了校正效率。
本发明实施例提供一种数字预失真校正装置50,本发明所有实施例提供的装置可以应用于通信***中,比如,该装置可以为射频单元或基站,还可以为射频单元或基站中的一部分,还可以应用于其他需要进行数字预失真校正的***,如雷达***或雷达***中的一部分。如图15所示,该数字预失真校正装置50包括:
第一确定单元501,第二确定单元502,建立单元503,处理单元504和校正单元505。
第一确定单元501,用于根据预选输入信号确定放大器模型中的预选输入信号的动态线性模型的调整因子,该放大器模型用于指示输出信号与输入信号的静态模型、输入信号的动态模型的关系,动态模型包括动态线性模型和动态非线性模型,其中,动态线性模型用于指示输入信号的变量中的线性特性,动态非线性模型用于指示输入信号的变量中的非线性特性,预选输入信号为预设输入信号组中的至少两个信号,预设输入信号组包括功率值不同的多个输入信号。
第二确定单元502,用于根据预选输入信号确定放大器模型中的预选输入信号的动态非线性模型的调整因子。
建立单元503,用于根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型。
处理单元504,用于根据当前输入信号的放大器模型获得当前输入信号的数字预失真DPD模型。
校正单元505,用于根据当前输入信号的DPD模型对当前输入信号进行数字预失真校正。
综上所述,本发明实施例提供的数字预失真校正装置,由于可以根据预选输入信号确定放大器模型中的该预选输入信号的动态线性模型的调整因子和动态非线性模型的调整因子,然后再根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型,继而根据当前输入信号的放大器模型获得当前输入信号的DPD模型,最后根据DPD模型对当前输入信号进行数字预失真校正,相较于查找表模型,无需计算并存储覆盖足够的功率范围的放大器模型的模型参数,从而无需计算并存储覆盖足够的功率范围的DPD模型的模型参数来满足不同输入信号的校正要求,因此,简化了模型参数的获取过程,提高了校正效率。
本发明实施例提供另一种数字预失真校正装置50,如图16所示,该数字预失真校正装置50包括:
第一确定单元501,第二确定单元502,建立单元503,处理单元504,校正单元505和更新单元506。
第一确定单元501,用于根据预选输入信号确定放大器模型中的预选输入信号的动态线性模型的调整因子,该放大器模型用于指示输出信号与输入信号 的静态模型、输入信号的动态模型的关系,动态模型包括动态线性模型和动态非线性模型,其中,动态线性模型用于指示输入信号的变量中的线性特性,动态非线性模型用于指示输入信号的变量中的非线性特性,预选输入信号为预设输入信号组中的至少两个信号,预设输入信号组包括功率值不同的多个输入信号。
第二确定单元502,用于根据预选输入信号确定放大器模型中的预选输入信号的动态非线性模型的调整因子。
建立单元503,用于根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型。
处理单元504,用于根据当前输入信号的放大器模型获得当前输入信号的数字预失真DPD模型。
校正单元505,用于根据当前输入信号的DPD模型对当前输入信号进行数字预失真校正。
更新单元506,用于当功率放大器的状态变化时,更新当前输入信号的放大器模型得到更新后的放大器模型。
进一步的,如图17所示,建立单元503可以包括:
第一确定模块5031和第一代入模块5032。
第一确定模块5031,用于根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000206
和动态模型的模型系数
Figure PCTCN2015075606-appb-000207
该放大器模型公式为:
Figure PCTCN2015075606-appb-000208
其中,N表示预设输入信号组中每个输入信号的采样点个数,L1表示静态模型的模型系数的个数,L2表示动态模型的模型系数的个数,r表示预设输入信号组中的输入信号的功率量级,r为大于或等于1的整数,
Figure PCTCN2015075606-appb-000209
表示预设输入信号组的输出信号组,
Figure PCTCN2015075606-appb-000210
表示预设输入信号组的静态模型,
Figure PCTCN2015075606-appb-000211
表示预设输入信号组的动态模型,
Figure PCTCN2015075606-appb-000212
表示预设输入信号组的动态线性模型,
Figure PCTCN2015075606-appb-000213
表示预设输入信号组的动态非线性模型,
Figure PCTCN2015075606-appb-000214
表示预设输入信号组的动态线性模型的调整因子,
Figure PCTCN2015075606-appb-000215
表示预设输入信号组的动态非线 性模型的调整因子。
第一代入模块5032,用于将当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000216
和动态模型的模型系数
Figure PCTCN2015075606-appb-000217
代入放大器模型公式,得到当前输入信号对应的放大器模型。
如图18所示,第一确定模块5031可以包括:
第一确定子模块50311,第二确定子模块50312,代入子模块50313和第一处理子模块50314。
第一确定子模块50311,用于根据预选输入信号的动态线性模型的调整因子确定预设输入信号组的动态线性模型的调整因子。
第二确定子模块50312,用于根据预选输入信号的动态非线性模型的调整因子确定预设输入信号组的动态非线性模型的调整因子。
代入子模块50313,用于将预设输入信号组的动态线性模型的调整因子、预设输入信号组的动态非线性模型的调整因子代入放大器模型公式,得到预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数。
第一处理子模块50314,用于将预设输入信号组的静态模型的模型系数作为当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000218
将预设输入信号组的动态模型的模型系数作为当前输入信号的动态模型的模型系数
Figure PCTCN2015075606-appb-000219
如图19所示,第一确定单元501可以包括:
第二确定模块5011和第三确定模块5012。
第二确定模块5011,用于根据第一输入信号的动态线性特性参数和线性调整因子公式,确定第一输入信号的动态线性模型的调整因子。
第三确定模块5012,用于根据第二输入信号的动态线性特性参数的参数值和线性调整因子公式,确定第二输入信号的动态线性模型的调整因子。
该线性调整因子公式为:
Figure PCTCN2015075606-appb-000220
其中,θi表示第一输入信号的动态线性特性参数的参数值或第二输入信号的动态线性特性参数的参数值,i表示第一输入信号的功率量级或第二输入信号的功率量级,θ1表示第一输入信号的动态线性特性参数的参数值,
Figure PCTCN2015075606-appb-000221
表示第一输入信号的动态线性模型的调整因子或第二输入信号的动态线性模型的调整因子。
需要说明的是,动态线性特性参数为功率放大器中预设输入信号组对应的输出信号的平均功率或者功率放大器中预设输入信号组对应的输出信号的增益。
如图20所示,第二确定单元502可以包括:
第四确定模块5021和第五确定模块5022。
第四确定模块5021,用于根据预选输入信号的动态线性特性参数的参数值和非线性特性公式确定预选输入信号的动态非线性特性参数的参数值,该非线性特性公式为:
Figure PCTCN2015075606-appb-000222
其中,
Figure PCTCN2015075606-appb-000223
表示预选输入信号的动态非线性特性参数的参数值,x(i)(n)表示预选输入信号的信号值,y(i)(n)表示预选输入信号对应的输出信号。
第五确定模块5022,用于根据预选输入信号的动态非线性特性参数的参数值和非线性调整因子公式确定预选输入信号的动态非线性模型的调整因子,该非线性调整因子公式为:
Figure PCTCN2015075606-appb-000224
其中,
Figure PCTCN2015075606-appb-000225
表示第一输入信号的动态非线性特性参数的参数值。
进一步的,第一确定子模块具体用于:根据预选输入信号的动态线性模型的调整因子,采用第一插值公式确定预设输入信号组的动态线性模型的调整因子
Figure PCTCN2015075606-appb-000226
该第一插值公式为:
Figure PCTCN2015075606-appb-000227
其中,r表示预设输入信号组中的输入信号的功率量级小于第一输入信号的功率量级M,大于第二输入信号的功率量级m,
Figure PCTCN2015075606-appb-000228
表示第一输入信号的动态线性模型的调整因子,
Figure PCTCN2015075606-appb-000229
表示第二输入信号的动态线性模型的调整因子,w(r)表示权重因子,权重因子w(r)根据权重公式确定得到,该权重公式为:
Figure PCTCN2015075606-appb-000230
其中,Pr表示功率量级为r的输入信号的功率值,PM表示功率量级为M的第一输入信号的功率值。
第二确定子模块具体用于:根据预选输入信号的动态非线性模型的调整因子,采用第二插值公式确定预设输入信号组的动态非线性模型的调整因子
Figure PCTCN2015075606-appb-000231
该第二插值公式为:
Figure PCTCN2015075606-appb-000232
其中,
Figure PCTCN2015075606-appb-000233
表示第一输入信号的动态非线性模型的调整因子,
Figure PCTCN2015075606-appb-000234
表示第二输入信号的动态非线性模型的调整因子,w(r)表示权重因子。
需要说明的是,功率放大器的状态变化可以为器件老化、温度波动或者偏置电压变化。
如图21所示,更新单元506可以包括:
第一获取模块5061,第二获取模块5062和第二代入模块5063。
第一获取模块5061,用于根据当前输入信号的放大器模型,获取更新后的放大器模型的动态模型的模型系数。
第二获取模块5062,用于根据当前输入信号的放大器模型,获取更新后的放大器模型的静态模型的模型系数。
第二代入模块5063,用于将更新后的放大器模型的动态模型的模型系数和更新后的放大器模型的静态模型的模型系数代入放大器模型得到更新后的放大器模型。
如图22所示,第一获取模块5061可以包括:
第一差值子模块50611和第二处理子模块50612。
第一差值子模块50611,用于将功率放大器的状态变化后的输入信号的动态模型的模型系数与变化前的输入信号的动态模型的模型系数之差作为第一差值。
第二处理子模块50612,用于将当前输入信号的动态模型的模型系数与第一差值之和作为更新后的放大器模型的动态模型的模型系数。
如图23所示,第二获取模块5062可以包括:
第二差值子模块50621和第三处理子模块50622。
第二差值子模块50621,用于将功率放大器的状态变化后的输入信号的静态模型的模型系数与变化前的输入信号的静态模型的模型系数之差作为第二差值。
第三处理子模块50622,用于将当前输入信号的静态模型的模型系数与第 二差值之和作为更新后的放大器模型的静态模型的模型系数。
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,上述描述的装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
综上所述,本发明实施例提供的数字预失真校正装置,由于可以根据预选输入信号确定放大器模型中的该预选输入信号的动态线性模型的调整因子和动态非线性模型的调整因子,然后再根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型,继而根据当前输入信号的放大器模型获得当前输入信号的DPD模型,最后根据DPD模型对当前输入信号进行数字预失真校正,相较于查找表模型,无需计算并存储覆盖足够的功率范围的放大器模型的模型参数,从而无需计算并存储覆盖足够的功率范围的DPD模型的模型参数来满足不同输入信号的校正要求,因此,简化了模型参数的获取过程,提高了校正效率。
本发明实施例提供又一种数字预失真校正装置,如图24所示,该数字预失真校正装置包括:至少一个处理器701,至少一个输入输出(英文:Input-Output;简称:IO)接口702或者其他通信接口,存储器703,和至少一个通信总线704,用于实现这些装置之间的连接通信。处理器701用于执行存储器703中的计算机执行指令7031,控制IO接口702进行收发。存储器703可能包含随机存取存储器(英文:Random Access Memory;简称:RAM),也可能包含非不稳定的存储器(non-volatile memory),例如至少一个磁盘存储器。通过至少一个IO接口702(可以是有线或者无线)实现该数字预失真校正装置进行数据的接收与发送,包括与至少一个其他网元之间的通信连接。当处理器执行存储器中的计算机执行指令时,可以执行上述方法实施例中的步骤,具体可以参考上述方法实施例中的描述,在此不再赘述。
本发明实施例还提供一种数字预失真校正装置,如图24所示,该数字预失真校正装置包括:
处理器701,用于根据预选输入信号确定放大器模型中的预选输入信号的动态线性模型的调整因子,该放大器模型用于指示输出信号与输入信号的静态模型、输入信号的动态模型的关系,动态模型包括动态线性模型和动态非线性模型,其中,动态线性模型用于指示输入信号的变量中的线性特性,动态非线 性模型用于指示输入信号的变量中的非线性特性,预选输入信号为预设输入信号组中的至少两个信号,预设输入信号组包括功率值不同的多个输入信号。
处理器701还用于根据预选输入信号确定放大器模型中的预选输入信号的动态非线性模型的调整因子。
处理器701还用于根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型。
处理器701还用于根据当前输入信号的放大器模型获得当前输入信号的数字预失真DPD模型。可选的,可以将当前输入信号的放大器模型的逆函数作为当前输入信号的数字预失真DPD模型。
处理器701还用于根据当前输入信号的DPD模型对当前输入信号进行数字预失真校正。
需要说明的是,本发明实施例中的处理器执行存储器中的计算机执行指令时,可以执行上述方法实施例中的步骤,具体可以参考上述方法实施例中的描述。
综上所述,本发明实施例提供的数字预失真校正装置,由于可以根据预选输入信号确定放大器模型中的该预选输入信号的动态线性模型的调整因子和动态非线性模型的调整因子,然后再根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型,继而根据当前输入信号的放大器模型获得当前输入信号的DPD模型,最后根据DPD模型对当前输入信号进行数字预失真校正,相较于查找表模型,无需计算并存储覆盖足够的功率范围的放大器模型的模型参数,从而无需计算并存储覆盖足够的功率范围的DPD模型的模型参数来满足不同输入信号的校正要求,因此,简化了模型参数的获取过程,提高了校正效率。
进一步的,处理器701具体用于:
根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000235
和动态模型的模型系数
Figure PCTCN2015075606-appb-000236
该放大器模型公式为:
Figure PCTCN2015075606-appb-000237
其中,N表示预设输入信号组中每个输入信号的采样点个数,L1表示静态模型的模型系数的个数,L2表示动态模型的模型系数的个数,r表示预设输 入信号组中的输入信号的功率量级,r为大于或等于1的整数,
Figure PCTCN2015075606-appb-000238
表示预设输入信号组的输出信号组,
Figure PCTCN2015075606-appb-000239
表示预设输入信号组的静态模型,
Figure PCTCN2015075606-appb-000240
表示预设输入信号组的动态模型,
Figure PCTCN2015075606-appb-000241
表示预设输入信号组的动态线性模型,
Figure PCTCN2015075606-appb-000242
表示预设输入信号组的动态非线性模型,
Figure PCTCN2015075606-appb-000243
表示预设输入信号组的动态线性模型的调整因子,
Figure PCTCN2015075606-appb-000244
表示预设输入信号组的动态非线性模型的调整因子;将当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000245
和动态模型的模型系数
Figure PCTCN2015075606-appb-000246
代入放大器模型公式,得到当前输入信号对应的放大器模型。
处理器701还具体用于:
根据预选输入信号的动态线性模型的调整因子确定预设输入信号组的动态线性模型的调整因子;根据预选输入信号的动态非线性模型的调整因子确定预设输入信号组的动态非线性模型的调整因子;将预设输入信号组的动态线性模型的调整因子、预设输入信号组的动态非线性模型的调整因子代入放大器模型公式,得到预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数;将预设输入信号组的静态模型的模型系数作为当前输入信号的静态模型的模型系数
Figure PCTCN2015075606-appb-000247
将预设输入信号组的动态模型的模型系数作为当前输入信号的动态模型的模型系数
Figure PCTCN2015075606-appb-000248
需要说明的是,预选输入信号可以为预设输入信号组中功率值最高的第一输入信号和功率值最低的第二输入信号,处理器701具体用于:
根据第一输入信号的动态线性特性参数和线性调整因子公式,确定第一输入信号的动态线性模型的调整因子;根据第二输入信号的动态线性特性参数的参数值和线性调整因子公式,确定第二输入信号的动态线性模型的调整因子,该线性调整因子公式为:
Figure PCTCN2015075606-appb-000249
其中,θi表示第一输入信号的动态线性特性参数的参数值或第二输入信号的动态线性特性参数的参数值,i表示第一输入信号的功率量级或第二输入信号的功率量级,θ1表示第一输入信号的动态线性特性参数的参数值,
Figure PCTCN2015075606-appb-000250
表示第一输入信号的动态线性模型的调整因子或第二输入信号的动态线性模型的调整因子。
需要说明的是,动态线性特性参数可以为功率放大器中预设输入信号组对应的输出信号的平均功率或者功率放大器中预设输入信号组对应的输出信号的增益。
处理器701还具体用于:
根据预选输入信号的动态线性特性参数的参数值和非线性特性公式确定预选输入信号的动态非线性特性参数的参数值,该非线性特性公式为:
Figure PCTCN2015075606-appb-000251
其中,
Figure PCTCN2015075606-appb-000252
表示预选输入信号的动态非线性特性参数的参数值,x(i)(n)表示预选输入信号的信号值,y(i)(n)表示预选输入信号对应的输出信号。
根据预选输入信号的动态非线性特性参数的参数值和非线性调整因子公式确定预选输入信号的动态非线性模型的调整因子,该非线性调整因子公式为:
Figure PCTCN2015075606-appb-000253
其中,
Figure PCTCN2015075606-appb-000254
表示第一输入信号的动态非线性特性参数的参数值。
处理器701还具体用于:
根据预选输入信号的动态线性模型的调整因子,采用第一插值公式确定预设输入信号组的动态线性模型的调整因子
Figure PCTCN2015075606-appb-000255
该第一插值公式为:
Figure PCTCN2015075606-appb-000256
其中,r表示预设输入信号组中的输入信号的功率量级小于第一输入信号的功率量级M,大于第二输入信号的功率量级m,
Figure PCTCN2015075606-appb-000257
表示第一输入信号的动态线性模型的调整因子,
Figure PCTCN2015075606-appb-000258
表示第二输入信号的动态线性模型的调整因子,w(r)表示权重因子,权重因子w(r)根据权重公式确定得到,该权重公式为:
Figure PCTCN2015075606-appb-000259
其中,Pr表示功率量级为r的输入信号的功率值,PM表示功率量级为M的第一输入信号的功率值。
处理器701还具体用于根据预选输入信号的动态非线性模型的调整因子,采用第二插值公式确定预设输入信号组的动态非线性模型的调整因子
Figure PCTCN2015075606-appb-000260
第 二插值公式为:
Figure PCTCN2015075606-appb-000261
其中,
Figure PCTCN2015075606-appb-000262
表示第一输入信号的动态非线性模型的调整因子,
Figure PCTCN2015075606-appb-000263
表示第二输入信号的动态非线性模型的调整因子,w(r)表示权重因子。
需要说明的是,处理器701还用于:
当功率放大器的状态变化时,更新当前输入信号的放大器模型得到更新后的放大器模型。
功率放大器的状态变化可以为器件老化、温度波动或者偏置电压变化。
处理器701还具体用于:根据当前输入信号的放大器模型,获取更新后的放大器模型的动态模型的模型系数;根据当前输入信号的放大器模型,获取更新后的放大器模型的静态模型的模型系数;将更新后的放大器模型的动态模型的模型系数和更新后的放大器模型的静态模型的模型系数代入放大器模型得到更新后的放大器模型。
处理器701还具体用于:将功率放大器的状态变化后的输入信号的动态模型的模型系数与变化前的输入信号的动态模型的模型系数之差作为第一差值;将当前输入信号的动态模型的模型系数与第一差值之和作为更新后的放大器模型的动态模型的模型系数。
处理器701还具体用于:将功率放大器的状态变化后的输入信号的静态模型的模型系数与变化前的输入信号的静态模型的模型系数之差作为第二差值;将当前输入信号的静态模型的模型系数与第二差值之和作为更新后的放大器模型的静态模型的模型系数。
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,上述描述的装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
应该理解的是,在本发明实施例中,处理器701可以是中央处理单元(英文:Central Processing Unit;简称:CPU),该处理器还可以是其他通用处理器、数字信号处理器(英文:Digital Sgnal Processing;简称:DSP)、专用集成电路(英文:Application Specific Integrated Circuit;简称:ASIC)、现成可编程门阵列(英文:Field-Programmable Gate Array;简称:FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是 微处理器,该处理器也可以是任何常规的处理器等。
该存储器703可以包括只读存储器和随机存取存储器,并向处理器701提供计算机执行指令和数据。存储器的一部分还可以包括非易失性随机存取存储器。例如,存储器还可以存储设备类型的信息。
在实现过程中,上述方法实施例中的各步骤可以通过处理器701中的硬件的集成逻辑电路或者软件形式的指令完成。结合本发明实施例所公开的方法的步骤可以直接体现为硬件处理器执行完成,或者用处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器,闪存、只读存储器,可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器,处理器读取存储器中的信息,结合其硬件完成上述方法的步骤。为避免重复,这里不再详细描述。
在本申请所提供的几个实施例中,应该理解到,所揭露的装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个***,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本发明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。
所述功能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,本发明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)执行本发明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:U盘、移动硬盘、只读存储器(英文:Read-Only Memory;简称:ROM)、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
综上所述,本发明实施例提供的数字预失真校正装置,由于处理器可以根据预选输入信号确定放大器模型中的该预选输入信号的动态线性模型的调整因子和动态非线性模型的调整因子,然后再根据预选输入信号的动态线性模型的调整因子、预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型,继而根据当前输入信号的放大器模型获得当前输入信号的DPD模型,最后根据DPD模型对当前输入信号进行数字预失真校正,相较于查找表模型,无需计算并存储覆盖足够的功率范围的放大器模型的模型参数,从而无需计算并存储覆盖足够的功率范围的DPD模型的模型参数来满足不同输入信号的校正要求,因此,简化了模型参数的获取过程,提高了校正效率。
本领域普通技术人员可以意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本发明的范围。
以上所述仅为本发明的较佳实施例,并不用以限制本发明,凡在本发明的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本发明的保护范围之内。

Claims (24)

  1. 一种数字预失真校正方法,其特征在于,所述方法包括:
    根据预选输入信号确定放大器模型中的所述预选输入信号的动态线性模型的调整因子,所述放大器模型用于指示输出信号与输入信号的静态模型、输入信号的动态模型的关系,所述动态模型包括动态线性模型和动态非线性模型,其中,所述动态线性模型用于指示输入信号的变量中的线性特性,所述动态非线性模型用于指示输入信号的变量中的非线性特性,所述预选输入信号为预设输入信号组中的至少两个信号,所述预设输入信号组包括功率值不同的多个输入信号;
    根据所述预选输入信号确定所述放大器模型中的所述预选输入信号的动态非线性模型的调整因子;
    根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型;
    根据所述当前输入信号的放大器模型获得所述当前输入信号的数字预失真DPD模型;
    根据所述当前输入信号的DPD模型对所述当前输入信号进行数字预失真校正。
  2. 根据权利要求1所述的数字预失真校正方法,其特征在于,所述根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型,包括:
    根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定所述当前输入信号的静态模型的模型系数
    Figure PCTCN2015075606-appb-100001
    和动态模型的模型系数
    Figure PCTCN2015075606-appb-100002
    所述放大器模型公式为:
    Figure PCTCN2015075606-appb-100003
    其中,所述N表示所述预设输入信号组中每个输入信号的采样点个数,所述L1表示所述静态模型的模型系数的个数,所述L2表示所述动态模型的模型系数的个数,所述r表示所述预设输入信号组中的输入信号的功率量级,所述r 为大于或等于1的整数,所述
    Figure PCTCN2015075606-appb-100004
    表示所述预设输入信号组的输出信号组,所述
    Figure PCTCN2015075606-appb-100005
    表示所述预设输入信号组的静态模型,所述
    Figure PCTCN2015075606-appb-100006
    表示所述预设输入信号组的动态模型,所述
    Figure PCTCN2015075606-appb-100007
    表示所述预设输入信号组的动态线性模型,所述
    Figure PCTCN2015075606-appb-100008
    表示所述预设输入信号组的动态非线性模型,所述
    Figure PCTCN2015075606-appb-100009
    表示所述预设输入信号组的动态线性模型的调整因子,所述
    Figure PCTCN2015075606-appb-100010
    表示所述预设输入信号组的动态非线性模型的调整因子;
    将所述当前输入信号的静态模型的模型系数
    Figure PCTCN2015075606-appb-100011
    和动态模型的模型系数
    Figure PCTCN2015075606-appb-100012
    代入所述放大器模型公式,得到所述当前输入信号对应的放大器模型。
  3. 根据权利要求2所述的数字预失真校正方法,其特征在于,
    所述根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定所述当前输入信号的静态模型的模型系数
    Figure PCTCN2015075606-appb-100013
    和动态模型的模型系数
    Figure PCTCN2015075606-appb-100014
    包括:
    根据所述预选输入信号的动态线性模型的调整因子确定所述预设输入信号组的动态线性模型的调整因子;
    根据所述预选输入信号的动态非线性模型的调整因子确定所述预设输入信号组的动态非线性模型的调整因子;
    将所述预设输入信号组的动态线性模型的调整因子、所述预设输入信号组的动态非线性模型的调整因子代入所述放大器模型公式,得到所述预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数;
    将所述预设输入信号组的静态模型的模型系数作为所述当前输入信号的静态模型的模型系数
    Figure PCTCN2015075606-appb-100015
    将所述预设输入信号组的动态模型的模型系数作为所述当前输入信号的动态模型的模型系数
    Figure PCTCN2015075606-appb-100016
  4. 根据权利要求3所述的数字预失真校正方法,其特征在于,所述预选输入信号为所述预设输入信号组中功率值最高的第一输入信号和功率值最低的第二输入信号,
    所述根据预选输入信号确定放大器模型中的所述预选输入信号的动态线性模型的调整因子,包括:
    根据所述第一输入信号的动态线性特性参数和线性调整因子公式,确定所述所述第一输入信号的动态线性模型的调整因子;
    根据所述第二输入信号的动态线性特性参数的参数值和线性调整因子公式,确定所述所述第二输入信号的动态线性模型的调整因子;
    所述线性调整因子公式为:
    Figure PCTCN2015075606-appb-100017
    其中,所述θi表示所述第一输入信号的动态线性特性参数的参数值或所述第二输入信号的动态线性特性参数的参数值,所述i表示所述第一输入信号的功率量级或所述第二输入信号的功率量级,所述θ1表示第一输入信号的动态线性特性参数的参数值,所述
    Figure PCTCN2015075606-appb-100018
    表示第一输入信号的动态线性模型的调整因子或第二输入信号的动态线性模型的调整因子。
  5. 根据权利要求4所述的数字预失真校正方法,其特征在于,
    所述动态线性特性参数为功率放大器中所述预设输入信号组对应的输出信号的平均功率或者功率放大器中所述预设输入信号组对应的输出信号的增益。
  6. 根据权利要求4所述的数字预失真校正方法,其特征在于,
    所述根据所述预选输入信号确定所述放大器模型中的所述预选输入信号的动态非线性模型的调整因子,包括:
    根据所述预选输入信号的动态线性特性参数的参数值和非线性特性公式确定所述预选输入信号的动态非线性特性参数的参数值,所述非线性特性公式为:
    Figure PCTCN2015075606-appb-100019
    其中,所述
    Figure PCTCN2015075606-appb-100020
    表示所述预选输入信号的动态非线性特性参数的参数值,所述x(i)(n)表示所述预选输入信号的信号值,所述y(i)(n)表示所述预选输入信号对应的输出信号;
    根据所述预选输入信号的动态非线性特性参数的参数值和非线性调整因子公式确定所述预选输入信号的动态非线性模型的调整因子,所述非线性调整因子公式为:
    Figure PCTCN2015075606-appb-100021
    其中,所述
    Figure PCTCN2015075606-appb-100022
    表示所述第一输入信号的动态非线性特性参数的参数值。
  7. 根据权利要求6所述的数字预失真校正方法,其特征在于,
    所述根据所述预选输入信号的动态线性模型的调整因子确定所述预设输入信号组的动态线性模型的调整因子,包括:
    根据所述预选输入信号的动态线性模型的调整因子,采用第一插值公式确定所述预设输入信号组的动态线性模型的调整因子
    Figure PCTCN2015075606-appb-100023
    所述第一插值公式为:
    Figure PCTCN2015075606-appb-100024
    其中,所述r表示所述预设输入信号组中的输入信号的功率量级小于所述第一输入信号的功率量级M,大于所述第二输入信号的功率量级m,所述
    Figure PCTCN2015075606-appb-100025
    表示所述第一输入信号的动态线性模型的调整因子,所述
    Figure PCTCN2015075606-appb-100026
    表示所述第二输入信号的动态线性模型的调整因子,所述w(r)表示权重因子,所述权重因子w(r)根据权重公式确定得到,所述权重公式为:
    Figure PCTCN2015075606-appb-100027
    其中,所述Pr表示功率量级为r的输入信号的功率值,所述PM表示功率量级为M的所述第一输入信号的功率值;
    所述根据所述预选输入信号的动态非线性模型的调整因子确定所述预设输入信号组的动态非线性模型的调整因子,包括:
    根据所述预选输入信号的动态非线性模型的调整因子,采用第二插值公式确定所述预设输入信号组的动态非线性模型的调整因子
    Figure PCTCN2015075606-appb-100028
    所述第二插值公式为:
    Figure PCTCN2015075606-appb-100029
    其中,所述
    Figure PCTCN2015075606-appb-100030
    表示所述第一输入信号的动态非线性模型的调整因子,所述
    Figure PCTCN2015075606-appb-100031
    表示所述第二输入信号的动态非线性模型的调整因子,所述w(r)表示权重因子。
  8. 根据权利要求1至7任意一项权利要求所述的数字预失真校正方法,其 特征在于,在所述根据所述DPD模型对所述当前输入信号进行数字预失真校正之后,所述方法还包括:
    当所述功率放大器的状态变化时,更新所述当前输入信号的放大器模型得到更新后的放大器模型。
  9. 根据权利要求8所述的数字预失真校正方法,其特征在于,
    所述功率放大器的状态变化为器件老化、温度波动或者偏置电压变化。
  10. 根据权利要求8或9所述的数字预失真校正方法,其特征在于,所述更新所述当前输入信号的放大器模型得到更新后的放大器模型,包括:
    根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的动态模型的模型系数;
    根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的静态模型的模型系数;
    将所述更新后的放大器模型的动态模型的模型系数和所述更新后的放大器模型的静态模型的模型系数代入所述放大器模型得到所述更新后的放大器模型。
  11. 根据权利要求10所述的数字预失真校正方法,其特征在于,
    所述根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的动态模型的模型系数,包括:
    将所述功率放大器的状态变化后的输入信号的动态模型的模型系数与变化前的输入信号的动态模型的模型系数之差作为第一差值;
    将所述当前输入信号的动态模型的模型系数与所述第一差值之和作为所述更新后的放大器模型的动态模型的模型系数。
  12. 根据权利要求10所述的数字预失真校正方法,其特征在于,
    所述根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的静态模型的模型系数,包括:
    将所述功率放大器的状态变化后的输入信号的静态模型的模型系数与变化前的输入信号的静态模型的模型系数之差作为第二差值;
    将所述当前输入信号的静态模型的模型系数与所述第二差值之和作为所述更新后的放大器模型的静态模型的模型系数。
  13. 一种数字预失真校正装置,其特征在于,所述数字预失真校正装置包括:
    处理器,用于根据预选输入信号确定放大器模型中的所述预选输入信号的动态线性模型的调整因子,所述放大器模型用于指示输出信号与输入信号的静态模型、输入信号的动态模型的关系,所述动态模型包括动态线性模型和动态非线性模型,其中,所述动态线性模型用于指示输入信号的变量中的线性特性,所述动态非线性模型用于指示输入信号的变量中的非线性特性,所述预选输入信号为预设输入信号组中的至少两个信号,所述预设输入信号组包括功率值不同的多个输入信号;
    所述处理器还用于根据所述预选输入信号确定所述放大器模型中的所述预选输入信号的动态非线性模型的调整因子;
    所述处理器还用于根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子建立当前输入信号的放大器模型;
    所述处理器还用于根据当前输入信号的放大器模型获得所述当前输入信号的数字预失真DPD模型;
    所述处理器还用于根据所述当前输入信号的DPD模型对所述当前输入信号进行数字预失真校正。
  14. 根据权利要求13所述的数字预失真校正装置,其特征在于,所述处理器具体用于:
    根据所述预选输入信号的动态线性模型的调整因子、所述预选输入信号的动态非线性模型的调整因子,采用放大器模型公式确定所述当前输入信号的静态模型的模型系数
    Figure PCTCN2015075606-appb-100032
    和动态模型的模型系数
    Figure PCTCN2015075606-appb-100033
    所述放大器模型公式为:
    Figure PCTCN2015075606-appb-100034
    其中,所述N表示所述预设输入信号组中每个输入信号的采样点个数,所述L1表示所述静态模型的模型系数的个数,所述L2表示所述动态模型的模型 系数的个数,所述r表示所述预设输入信号组中的输入信号的功率量级,所述r为大于或等于1的整数,所述
    Figure PCTCN2015075606-appb-100035
    表示所述预设输入信号组的输出信号组,所述
    Figure PCTCN2015075606-appb-100036
    表示所述预设输入信号组的静态模型,所述
    Figure PCTCN2015075606-appb-100037
    表示所述预设输入信号组的动态模型,所述
    Figure PCTCN2015075606-appb-100038
    表示所述预设输入信号组的动态线性模型,所述
    Figure PCTCN2015075606-appb-100039
    表示所述预设输入信号组的动态非线性模型,所述
    Figure PCTCN2015075606-appb-100040
    表示所述预设输入信号组的动态线性模型的调整因子,所述
    Figure PCTCN2015075606-appb-100041
    表示所述预设输入信号组的动态非线性模型的调整因子;
    将所述当前输入信号的静态模型的模型系数
    Figure PCTCN2015075606-appb-100042
    和动态模型的模型系数
    Figure PCTCN2015075606-appb-100043
    代入所述放大器模型公式,得到所述当前输入信号对应的放大器模型。
  15. 根据权利要求14所述的数字预失真校正装置,其特征在于,所述处理器具体用于:
    根据所述预选输入信号的动态线性模型的调整因子确定所述预设输入信号组的动态线性模型的调整因子;
    根据所述预选输入信号的动态非线性模型的调整因子确定所述预设输入信号组的动态非线性模型的调整因子;
    将所述预设输入信号组的动态线性模型的调整因子、所述预设输入信号组的动态非线性模型的调整因子代入所述放大器模型公式,得到所述预设输入信号组的静态模型的模型系数和预设输入信号组的动态模型的模型系数;
    将所述预设输入信号组的静态模型的模型系数作为所述当前输入信号的静态模型的模型系数
    Figure PCTCN2015075606-appb-100044
    将所述预设输入信号组的动态模型的模型系数作为所述当前输入信号的动态模型的模型系数
    Figure PCTCN2015075606-appb-100045
  16. 根据权利要求15所述的数字预失真校正装置,其特征在于,所述预选输入信号为所述预设输入信号组中功率值最高的第一输入信号和功率值最低的第二输入信号,所述处理器具体用于:
    根据所述第一输入信号的动态线性特性参数和线性调整因子公式,确定所述所述第一输入信号的动态线性模型的调整因子;
    根据所述第二输入信号的动态线性特性参数的参数值和线性调整因子公 式,确定所述所述第二输入信号的动态线性模型的调整因子;
    所述线性调整因子公式为:
    Figure PCTCN2015075606-appb-100046
    其中,所述θi表示所述第一输入信号的动态线性特性参数的参数值或所述第二输入信号的动态线性特性参数的参数值,所述i表示所述第一输入信号的功率量级或所述第二输入信号的功率量级,所述θ1表示第一输入信号的动态线性特性参数的参数值,所述
    Figure PCTCN2015075606-appb-100047
    表示第一输入信号的动态线性模型的调整因子或第二输入信号的动态线性模型的调整因子。
  17. 根据权利要求16所述的数字预失真校正装置,其特征在于,
    所述动态线性特性参数为功率放大器中所述预设输入信号组对应的输出信号的平均功率或者功率放大器中所述预设输入信号组对应的输出信号的增益。
  18. 根据权利要求16所述的数字预失真校正装置,其特征在于,所述处理器具体用于:
    根据所述预选输入信号的动态线性特性参数的参数值和非线性特性公式确定所述预选输入信号的动态非线性特性参数的参数值,所述非线性特性公式为:
    Figure PCTCN2015075606-appb-100048
    其中,所述
    Figure PCTCN2015075606-appb-100049
    表示所述预选输入信号的动态非线性特性参数的参数值,所述x(i)(n)表示所述预选输入信号的信号值,所述y(i)(n)表示所述预选输入信号对应的输出信号;
    根据所述预选输入信号的动态非线性特性参数的参数值和非线性调整因子公式确定所述预选输入信号的动态非线性模型的调整因子,所述非线性调整因子公式为:
    Figure PCTCN2015075606-appb-100050
    其中,所述
    Figure PCTCN2015075606-appb-100051
    表示所述第一输入信号的动态非线性特性参数的参数值。
  19. 根据权利要求18所述的数字预失真校正装置,其特征在于,所述处理 器具体用于:
    根据所述预选输入信号的动态线性模型的调整因子,采用第一插值公式确定所述预设输入信号组的动态线性模型的调整因子
    Figure PCTCN2015075606-appb-100052
    所述第一插值公式为:
    Figure PCTCN2015075606-appb-100053
    其中,所述r表示所述预设输入信号组中的输入信号的功率量级小于所述第一输入信号的功率量级M,大于所述第二输入信号的功率量级m,所述
    Figure PCTCN2015075606-appb-100054
    表示所述第一输入信号的动态线性模型的调整因子,所述
    Figure PCTCN2015075606-appb-100055
    表示所述第二输入信号的动态线性模型的调整因子,所述w(r)表示权重因子,所述权重因子w(r)根据权重公式确定得到,所述权重公式为:
    Figure PCTCN2015075606-appb-100056
    其中,所述Pr表示功率量级为r的输入信号的功率值,所述PM表示功率量级为M的所述第一输入信号的功率值;
    所述处理器还具体用于根据所述预选输入信号的动态非线性模型的调整因子,采用第二插值公式确定所述预设输入信号组的动态非线性模型的调整因子
    Figure PCTCN2015075606-appb-100057
    所述第二插值公式为:
    Figure PCTCN2015075606-appb-100058
    其中,所述
    Figure PCTCN2015075606-appb-100059
    表示所述第一输入信号的动态非线性模型的调整因子,所述
    Figure PCTCN2015075606-appb-100060
    表示所述第二输入信号的动态非线性模型的调整因子,所述w(r)表示权重因子。
  20. 根据权利要求13至19任意一项权利要求所述的数字预失真校正装置,其特征在于,所述处理器还用于:
    当所述功率放大器的状态变化时,更新所述当前输入信号的放大器模型得到更新后的放大器模型。
  21. 根据权利要求20所述的数字预失真校正装置,其特征在于,
    所述功率放大器的状态变化为器件老化、温度波动或者偏置电压变化。
  22. 根据权利要求20或21所述的数字预失真校正装置,其特征在于,所述处理器具体用于:
    根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的动态模型的模型系数;
    根据所述当前输入信号的放大器模型,获取所述更新后的放大器模型的静态模型的模型系数;
    将所述更新后的放大器模型的动态模型的模型系数和所述更新后的放大器模型的静态模型的模型系数代入所述放大器模型得到所述更新后的放大器模型。
  23. 根据权利要求22所述的数字预失真校正装置,其特征在于,所述处理器还具体用于:
    将所述功率放大器的状态变化后的输入信号的动态模型的模型系数与变化前的输入信号的动态模型的模型系数之差作为第一差值;
    将所述当前输入信号的动态模型的模型系数与所述第一差值之和作为所述更新后的放大器模型的动态模型的模型系数。
  24. 根据权利要求22所述的数字预失真校正装置,其特征在于,所述处理器还具体用于:
    将所述功率放大器的状态变化后的输入信号的静态模型的模型系数与变化前的输入信号的静态模型的模型系数之差作为第二差值;
    将所述当前输入信号的静态模型的模型系数与所述第二差值之和作为所述更新后的放大器模型的静态模型的模型系数。
PCT/CN2015/075606 2015-03-31 2015-03-31 数字预失真校正方法及装置 WO2016154933A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201580078331.XA CN107431495B (zh) 2015-03-31 2015-03-31 数字预失真校正方法及装置
PCT/CN2015/075606 WO2016154933A1 (zh) 2015-03-31 2015-03-31 数字预失真校正方法及装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2015/075606 WO2016154933A1 (zh) 2015-03-31 2015-03-31 数字预失真校正方法及装置

Publications (1)

Publication Number Publication Date
WO2016154933A1 true WO2016154933A1 (zh) 2016-10-06

Family

ID=57006540

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/075606 WO2016154933A1 (zh) 2015-03-31 2015-03-31 数字预失真校正方法及装置

Country Status (2)

Country Link
CN (1) CN107431495B (zh)
WO (1) WO2016154933A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023070438A1 (en) * 2021-10-28 2023-05-04 Telefonaktiebolaget Lm Ericsson (Publ) Digital predistortion method and digital predistortion apparatus

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112039445A (zh) * 2019-06-03 2020-12-04 大唐移动通信设备有限公司 一种数字预失真dpd的处理方法及装置
CN113014207B (zh) * 2021-03-03 2023-07-18 展讯通信(上海)有限公司 射频电路及其预失真补偿方法、装置及电子设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101040502A (zh) * 2004-01-21 2007-09-19 电力波技术公司 宽带增强型数字注射预失真***和方法
CN101807886A (zh) * 2010-03-26 2010-08-18 深圳市云海通讯股份有限公司 一种射频功率放大装置及其模拟预失真校正方法
US20120034886A1 (en) * 2010-08-03 2012-02-09 Crestcom, Inc. Transmitter Linearized Using Inversing and Non-Inversing Transform Processing Sections and Method Therefor
US20130200948A1 (en) * 2010-05-03 2013-08-08 Korea Advanced Institute Of Science And Technology Digital pre-distortion device and method for a broadband power amplifier

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7577211B2 (en) * 2004-03-01 2009-08-18 Powerwave Technologies, Inc. Digital predistortion system and method for linearizing an RF power amplifier with nonlinear gain characteristics and memory effects
US7336725B2 (en) * 2004-03-03 2008-02-26 Powerwave Technologies, Inc. Digital predistortion system and method for high efficiency transmitters

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101040502A (zh) * 2004-01-21 2007-09-19 电力波技术公司 宽带增强型数字注射预失真***和方法
CN101807886A (zh) * 2010-03-26 2010-08-18 深圳市云海通讯股份有限公司 一种射频功率放大装置及其模拟预失真校正方法
US20130200948A1 (en) * 2010-05-03 2013-08-08 Korea Advanced Institute Of Science And Technology Digital pre-distortion device and method for a broadband power amplifier
US20120034886A1 (en) * 2010-08-03 2012-02-09 Crestcom, Inc. Transmitter Linearized Using Inversing and Non-Inversing Transform Processing Sections and Method Therefor

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023070438A1 (en) * 2021-10-28 2023-05-04 Telefonaktiebolaget Lm Ericsson (Publ) Digital predistortion method and digital predistortion apparatus

Also Published As

Publication number Publication date
CN107431495B (zh) 2020-01-03
CN107431495A (zh) 2017-12-01

Similar Documents

Publication Publication Date Title
US11171614B2 (en) Multi-band digital compensator for a non-linear system
KR20210008073A (ko) 비선형 시스템용 디지털 보상기
CN103299542B (zh) 用于数字预失真器的正交基函数集
US20140314176A1 (en) Non-Linear Modeling of a Physical System Using Two-Dimensional Look-Up Table with Bilinear Interpolation
US8787494B2 (en) Modeling digital predistorter
JP6159890B2 (ja) デジタルプリディストーションパラメーターを求める方法及びプリディストーションシステム
CN103947106B (zh) 一种用于使非线性的***元件线性化的方法、***及设备
JP2019500782A (ja) デジタル補償器
JP5120178B2 (ja) 非線形システム逆特性同定装置及びその方法、電力増幅装置及び電力増幅器のプリディストータ
US20170302233A1 (en) Curve Fitting Circuit, Analog Predistorter, and Radio Frequency Signal Transmitter
CN108650034A (zh) 一种射频设备的增益校准方法及装置
WO2014002017A1 (en) Predistortion according to an artificial neural network (ann)-based model
WO2016154933A1 (zh) 数字预失真校正方法及装置
CN108111448B (zh) 预失真查找表的生成方法、装置与预失真校准设备
JP5299958B2 (ja) プリディストータ
CN108449294B (zh) 预失真处理方法、装置及预失真处理***
CN112243119A (zh) 一种白平衡处理方法、装置、电子设备及存储介质
CN106169985B (zh) 基于改进模型的包络跟踪放大器数字预失真方法
EP4356517A1 (en) Systems and methods for multiband linearization using kernel regression
CN111131104B (zh) 一种预失真处理方法、装置、存储介质和设备
US20060033569A1 (en) Method and apparatus for performing digital pre-distortion
WO2018054259A1 (zh) 一种基带信号的处理方法和装置
CN110545079B (zh) 一种dpd查找表生成方法及装置
JP6235899B2 (ja) 送信装置及び歪み補償方法
CN112688544B (zh) 逆变电路的谐波补偿方法、装置及终端设备

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: 15886918

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15886918

Country of ref document: EP

Kind code of ref document: A1