Radio frequency circuit, method of operating radio frequency circuit, and electronic device

Abstract

A radio frequency circuit, a method of operating a radio frequency circuit, and an electronic device are provided, the radio frequency circuit including: a multi-tone generation circuit configured to generate a multi-tone signal including a plurality of tones; a pre-distortion circuit configured to pre-distort the multi-tone signal or the modulated signal and output a pre-distorted signal; a power amplifier configured to amplify the pre-distorted signal and output an output signal; and a pre-distortion modeling circuit configured to model a pre-distortion model corresponding to the pre-distortion circuit based on a pre-distortion signal corresponding to the multi-tone signal and the output signal.

Description

[0001] This application claims the priority of Korean Patent Application No. 10-2023-0136167 filed on October 12, 2023 in the Korean Intellectual Property Office, the disclosure of which is incorporated herein by reference in its entirety. Technical Field

[0002] Example embodiments of the inventive concepts relate to a radio frequency (RF) circuit for predistortion and a method of operating the RF circuit for predistortion. Background Art

[0003] Wireless communication devices including smartphones, tablet personal computers (PCs), and Internet of Things (IoT) devices are equipped with a power amplifier (PA) for amplifying the power of wireless signals (e.g., boosting the strength of wireless signals). Since the power amplifier has nonlinear characteristics, a predistortion (PD) method (e.g., a digital PD method) is used to compensate for these nonlinearities in the digital domain. Summary of the invention

[0004] Example embodiments of the inventive concepts provide a radio frequency (RF) circuit for predistortion and a method of operating the RF circuit for predistortion.

[0005] According to an example embodiment, there is provided an RF circuit, the RF circuit comprising: a multi-tone generation circuit configured to generate a multi-tone signal including a plurality of tones; a predistortion circuit configured to predistort the multi-tone signal or a modulated signal and output the predistortion signal; a power amplifier configured to amplify the predistortion signal and output an output signal; and a predistortion modeling circuit configured to model a predistortion model corresponding to the predistortion circuit based on the predistortion signal corresponding to the multi-tone signal and the output signal.

[0006] According to an example embodiment, a method of operating an RF circuit is provided, the method comprising: generating a multi-tone signal including a plurality of tones; distorting the multi-tone signal and outputting a predistortion signal; amplifying the predistortion signal and outputting an output signal; and modeling a predistortion model based on the predistortion signal and the output signal corresponding to the multi-tone signal.

[0007] According to an example embodiment, an electronic device is provided, the electronic device comprising: a memory configured to store at least one instruction; and at least one processor configured to execute the at least one instruction, wherein the at least one processor executes the at least one instruction to: generate a multi-tone signal comprising a plurality of tones; distort the multi-tone signal or a modulated signal and output a predicted signal; amplify a predistortion signal and output an output signal; and model a predistortion model based on the predistortion signal and the output signal corresponding to the multi-tone signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The above and other features of the inventive concept will be more clearly understood from the following detailed description of example embodiments of the inventive concept taken in conjunction with the accompanying drawings.

[0009] Figure 1 is a diagram illustrating a radio frequency (RF) circuit according to example embodiments.

[0010] Figure 2 Included are plots showing the probability density function (PDF) and normalized mean square error (NMSE) of exemplary training signals.

[0011] Figure 3 is a diagram illustrating an exemplary training signal and a PDF of the training signal according to example embodiments.

[0012] Figure 4 is a diagram illustrating a pre-distortion modeling circuit according to an example embodiment.

[0013] Figure 5 is a graph showing a curve fit of a power amplifier for an exemplary training signal.

[0014] Figure 6 is a diagram illustrating curve fitting of a power amplifier for a training signal according to example embodiments.

[0015] Figure 7 is a graph showing error vector magnitude (EVM) for an exemplary training signal.

[0016] Figure 8 is a diagram illustrating an RF circuit according to example embodiments.

[0017] Fig. 9 It is shown Figure 8 Schematic diagram of the phase determination circuit shown in .

[0018] Fig.10 is a diagram showing the PDF of an exemplary training signal.

[0019] Fig.11 is a flow chart illustrating a method of operating an RF circuit according to an example embodiment.

[0020] Fig.12 is a flow chart illustrating a method of modeling a pre-distortion model of an RF circuit according to an example embodiment.

[0021] Fig.13 is a flow chart illustrating a method of determining a phase of an RF circuit according to an example embodiment.

[0022] Fig.14is a diagram illustrating a training information determination circuit according to example embodiments.

[0023] Fig.15 is a diagram illustrating a wireless communication device according to an example embodiment.

[0024] Fig.16 is a diagram illustrating an electronic device according to example embodiments. DETAILED DESCRIPTION

[0025] Hereinafter, example embodiments of the inventive concept will be described with reference to the accompanying drawings.

[0026] Figure 1 is a diagram illustrating a radio frequency (RF) circuit according to example embodiments.

[0027] Reference Figure 1 According to an example embodiment, an RF circuit 100 may include a multi-tone (or referred to as multi-tone, multi-audio) generation circuit 110 , a multiplexer (MUX) 120 , a pre-distortion (PD) circuit 130 , a power amplifier 140 , and a pre-distortion (PD) modeling circuit 150 .

[0028] The multi-tone generation circuit 110 may be configured to generate a multi-tone signal MT_S including a plurality of tones. The multi-tone signal MT_S may be used as a training signal for predistortion. The multi-tone signal MT_S may have a single sideband (SSB) or a double sideband (DSB). For example, the multi-tone generation circuit 110 may generate the multi-tone signal MT_S based on the following equation 1.

[0029] ...Equation 1.

[0030] Where x(t) is the multi-tone signal MT_S, N is the number of multiple tones and is a positive integer, and f k is the frequency of the kth tone (where k is a positive integer), and θ k is the phase of the kth tone.

[0031] Alternatively, for example, the multi-tone generation circuit 110 may generate the multi-tone signal MT_S based on Equation 2 below.

[0032] ...Equation 2.

[0033] The multi-tone signal MT_S generated based on Equation 2 may be referred to as a real signal of the signal in Equation 1. For example, the exemplary multi-tone signal MT_S may be a signal excluding an imaginary signal from the signal in Equation 1. For example, when the multi-tone generation circuit 110 is configured to generate the multi-tone signal MT_S referred to as a real signal, the multi-tone generation circuit 110 may include a single unit including a digital-to-analog converter (DAC). The DAC generates an analog signal corresponding to the multi-tone signal MT_S according to a digital signal. Therefore, hardware resources of the multi-tone generation circuit 110 according to the exemplary embodiment may be saved.

[0034] According to an example embodiment, the multi-tone generation circuit 110 may generate a multi-tone signal MT_S having a relatively uniform probability density function (PDF) compared to the modulation signal M_S. The modulation signal M_S may be a signal modulated based on, for example, 16 quadrature amplitude modulation (QAM), 64 QAM, 256 QAM, binary phase shift keying (BPSK), quadrature phase shift keying (QPSK), etc. In addition, the modulation signal M_S may include cyclic prefix-orthogonal frequency division multiplexing (CP-OFDM) and / or discrete Fourier transform (DFT)-spread-OFDM (DFT-s-OFDM) generated using the above-mentioned modulation scheme and multiplexing scheme. The modulation signal M_S may have a relatively non-uniform PDF. Using the modulation signal M_S as a training signal may not accurately capture the characteristics of the power amplifier 140, which may result in a reduction in the effectiveness of pre-distortion.

[0035] For example, the multi-tone generation circuit 110 may generate a multi-tone signal MT_S having a phase value determined for each of a plurality of tones. In this case, the multi-tone signal MT_S may have a uniform PDF.

[0036] The multiplexer 120 may be connected to the output terminal of the multi-tone generation circuit 110. The multiplexer 120 may be configured to select one of the multi-tone signal MT_S and the modulation signal M_S, and transmit the selected signal to the pre-distortion circuit 130. Alternatively, in addition to the multiplexer 120, various circuits may be provided between the multi-tone generation circuit 110 and the pre-distortion circuit 130 to select one of the input signals (e.g., the multi-tone signal MT_S and the modulation signal M_S) or switch between the input signals (e.g., the multi-tone signal MT_S and the modulation signal M_S).

[0037] According to some embodiments, when the predistortion modeling circuit 150 models the predistortion model PD_M, the multiplexer 120 may select and output the multi-tone signal MT_S for use as a training signal. Optionally, when the predistortion modeling is completed, the multiplexer 120 may select and output the modulation signal M_S. For example, when the predistortion modeling circuit 150 does not output the predistortion model PD_M, the multiplexer 120 may select and output the modulation signal M_S.

[0038] The predistortion circuit 130 may be configured to predistort any signal (e.g., a multi-tone signal MT_S or a modulation signal M_S) selected from the multiplexer 120, and output a predistortion signal PD_S. The predistortion circuit 130 may perform digital predistortion (DPD) to compensate for the nonlinearity of the power amplifier 140 in the digital domain. For example, because the power amplifier 140 has a nonlinear element, the power amplifier 140 exhibits a nonlinear characteristic. This results in nonlinear distortion in the output signal OUT. The predistortion circuit 130 may be connected to the input of the power amplifier 140 to compensate for such nonlinearity. The predistortion circuit 130 may be designed to receive a predistortion model PD_M modeled from a predistortion modeling circuit 150, and predistort an input signal (e.g., a multi-tone signal MT_S or a modulation signal M_S) based on the predistortion model PD_M. The model causes the input signal to have a distortion opposite to the nonlinear distortion of the power amplifier 140.

[0039] The input signal of the predistortion circuit 130 may be a digital signal in the baseband, and thus the predistortion operation of the predistortion circuit 130 may be performed in the digital domain. Therefore, even when the output signal OUT of the power amplifier 140 is amplified by the power amplifier 140 having nonlinear distortion, the output signal OUT of the power amplifier 140 may have linear characteristics because the output signal OUT of the power amplifier 140 has been converted into a signal having opposite distortion by the predistortion circuit 130. For example, an adjacent channel leakage ratio (ACLR) of the power amplifier 140 may be improved by compensating for the nonlinear characteristics of the power amplifier 140.

[0040] As an example, the predistortion model PD_M for predistortion may include a predistortion lookup table (LUT), a predistortion coefficient, etc. for distortion compensation. For example, the predistortion LUT may include at least one of an amplitude modulation to amplitude modulation (AM-AM) LUT and an AM to phase modulation (AM-PM) LUT. For example, the predistortion model PD_M may include a memory predistortion model targeting a power amplifier 140 that is subject to a memory effect, and a memoryless predistortion model targeting a power amplifier 140 that is not subject to a memory effect. The memory effect is characterized in that the response of the system is affected by previous inputs.

[0041] The power amplifier 140 may amplify the predistortion signal PD_S predistorted from the predistortion circuit 130 to output the output signal OUT. The power amplifier 140 has a nonlinear characteristic as described above, so that when a nonlinear compensation operation (such as a predistortion operation) is not performed, the output signal OUT may be distorted. The nonlinearity of the output signal OUT may be compensated by the predistortion circuit 130 described above.

[0042] The predistortion modeling circuit 150 may be configured to model a predistortion model PD_M corresponding to the predistortion circuit 130 based on the predistortion signal PD_S corresponding to the multi-tone signal MT_S and the output signal OUT. As described above, the predistortion model PD_M may include, for example, a predistortion LUT and predistortion coefficients.

[0043] In an example embodiment, the predistortion modeling circuit 150 may model a power amplifier model and then model the predistortion model PD_M based on the power amplifier model. For example, when the power amplifier model is affected by the memory effect, the predistortion modeling circuit 150 may model a predistortion model with memory, and when the power amplifier model is not affected by the memory effect, the predistortion modeling circuit 150 may model a predistortion model without memory.

[0044] In an example embodiment, the predistortion modeling circuit 150 may be configured to generate a training signal internally without directly capturing (or receiving) a training signal (e.g., a multi-tone signal MT_S). For example, even when the predistortion modeling circuit 150 captures only the output signal OUT of the power amplifier 140, the predistortion modeling circuit 150 may model the predistortion model PD_M through the internally generated training signal. In this case, the generated training signal may be the same as the signal generated by the multi-tone generation circuit 110.

[0045] The predistortion modeling circuit 150 may transmit the modeled predistortion model PD_M to the predistortion circuit 130. The transmission of the predistortion model PD_M may be referred to as updating a predistortion LUT, predistortion coefficients, etc. to be used in the predistortion circuit 130.

[0046] In the above embodiment, the RF circuit 100 can use a separately generated multi-tone signal MT_S as a training signal for pre-distortion modeling instead of using the modulation signal M_S as a training signal. In this case, the multi-tone signal MT_S has a uniform PDF, which helps prevent the degradation of pre-distortion performance. In addition, the RF circuit 100 generates the multi-tone signal MT_S through the multi-tone generation circuit 110, and does not require an additional memory to store the multi-tone signal MT_S. Without this memory constraint, the RF circuit 100 can utilize a larger number of training signals.

[0047] Figure 2 Included are graphs (a) and (b) showing a probability density function (PDF) and normalized mean square error (NMSE) of an exemplary training signal.

[0048] For example, as shown in FIG. (a), when the modulation signal M_S is used as a training signal, the modulation signal M_S may have a PDF that is non-uniform relative to the input power of the power amplifier 140. In FIG. (a), the PDF is relatively high in one range (e.g., a range of -20 dBm to -15 dBm) and relatively low in another range (e.g., a range of -40 dBm to -25 dBm). For example, in the first range, the number of samples of the training signal may be relatively large, and in the second range, the number of samples of the training signal may be relatively small. In this case, when modeling a power amplifier model defined as the relationship between a training signal and an output signal OUT of the training signal passing through the power amplifier 140, it may be challenging to accurately capture the characteristics of the power amplifier 140 in the power amplifier model.

[0049] Deterioration of predistortion performance may occur within a range where it is difficult to accurately reflect the characteristics of the power amplifier 140. As an example, it can be seen that in the case of diagram (a), PDF is low for lower input power and corresponding target power, and in diagram (b) corresponding to diagram (a), NMSE is high for lower input power. For example, degradation of predistortion performance may occur for lower input power.

[0050] As a result, when the modulation signal M_S is used as a training signal, additional time or memory for calibration may be required to prevent predistortion performance from being degraded due to the characteristics of the non-uniform PDF.

[0051] Figure 3 is a diagram illustrating an exemplary training signal and a PDF of the training signal according to example embodiments.

[0052] Figure 3 A PDF is shown in a case where a QPSK modulated CP-OFDM signal is used as an exemplary training signal and a multi-tone signal MT_S is used as a training signal according to an example embodiment.

[0053] like Figure 3As shown in , when a modulation signal M_S (e.g., a QPSK modulated CP-OFDM signal) is used as a training signal, the PDF may be relatively non-uniform with respect to the input power, but when a multi-tone signal MT_S is used as a training signal, the PDF may be relatively uniform. Therefore, by using a multi-tone signal MT_S having a uniform PDF as a training signal for predistortion, the RF circuit according to the example embodiment can prevent the degradation of predistortion performance.

[0054] Figure 4 is a diagram illustrating a pre-distortion modeling circuit according to an example embodiment.

[0055] like Figure 4 As shown in , the pre-distortion modeling circuit 150 according to example embodiments may include a processing circuit 151 and a memory 152 .

[0056] The processing circuit 151 may be configured to model the pre-distortion model PD_M according to the above-described example embodiments.

[0057] In an example embodiment, the processing circuit 151 may model a power amplifier model. The power amplifier model may be defined as a relationship between an input and an output of the power amplifier 140. For example, the processing circuit 151 may model the power amplifier model based on curve fitting. For example, the processing circuit 151 may perform curve fitting based on polynomial fitting or a method of dividing an input axis of the power amplifier model into a plurality of intervals and selecting a representative value of each of the plurality of intervals. For example, the processing circuit 151 may perform curve fitting based on a representative value selection method when the multi-tone signal MT_S is used as a training signal.

[0058] For example, the representative value selection method may be a moving average method for selecting an average value of outputs included in each of the plurality of intervals as a representative value of each of the plurality of intervals. For example, the moving average method may be performed based on Equation 3 below.

[0059] ...Equation 3.

[0060] in, is the representative value of the interval, L is the range of the interval, i is the index of the power amplifier model input included in the interval, and x i is the ith input value of the power amplifier model, y i is the power amplifier model and x i The corresponding output value, and M is the x included in the interval i The number of

[0061] The removal of Gaussian noise should be considered in all sample ranges to apply the representative value selection method including the moving average method. Sufficient samples should be ensured in all sample ranges to effectively remove Gaussian noise. For example, the more uniform the PDF of the training signal is, the easier it is to apply the representative value selection method.

[0062] Therefore, according to an example embodiment, when the multi-tone signal MT_S is used as a training signal, a representative value selection method (such as a moving average method) can be used to model the power amplifier model. The processing circuit 151 can model the power amplifier model based on the representative value selection method instead of polynomial fitting. Polynomial fitting that requires a high polynomial order may have problems in determining the highest polynomial order.

[0063] The processing circuit 151 may model a predistortion model PD_M for compensating for the nonlinearity of the power amplifier model based on the power amplifier model.

[0064] In an example embodiment, the processing circuit 151 may be connected to the memory 152. The memory 152 may store the predistortion model PD_M modeled by the multi-tone generation circuit 153 and the processing circuit 151. The multi-tone generation circuit 153 may be configured to generate the multi-tone signal MT_S in the same manner as the multi-tone generation circuit 110. Alternatively, the memory 152 may store the multi-tone signal MT_S.

[0065] The processing circuit 151 may model the predistortion model PD_M based on the output signal OUT and the multi-tone signal MT_S generated from the multi-tone generation circuit 153. For example, the processing circuit 151 may model the predistortion model PD_M by using the internally generated multi-tone signal MT_S (e.g., the multi-tone signal generated by the memory 152) as a training signal instead of capturing the training signal from the outside. In this case, since the processing circuit 151 does not need a memory (such as a non-volatile memory) for storing a large number of training signals, the processing circuit 151 can use more training signal samples.

[0066] The processing circuit 151 may be implemented as a hardware circuit or logic circuit including hardware, a hardware / software combination (such as a processor executing software and / or firmware), or a combination of hardware circuits and hardware / software combinations to perform operations according to the above-described example embodiments.

[0067] Figure 5 is a diagram showing a curve fit of a power amplifier for an exemplary training signal, and Figure 6 is a diagram illustrating curve fitting of a power amplifier for a training signal according to example embodiments.

[0068] Reference Figure 5, when a QPSK modulated CP-OFDM signal is used as an exemplary training signal, there may be a section where the samples of the input signal and the output signal OUT of the power amplifier 140 are not uniform. Figure 5 As shown in , the number of samples is small in the intervals where the power of the input signal is high and low. Therefore, it can be seen that the curve fitting of the sample correlation cannot accurately represent the characteristics of the power amplifier 140.

[0069] In this case, the power amplifier model based on curve fitting may also be inaccurate, which may lead to degradation of the predistortion performance.

[0070] Reference Figure 6 , when the multi-tone signal MT_S is used as a training signal, it can be seen that the samples are uniform in most intervals. Therefore, the sample-related curve fitting can also relatively accurately represent the characteristics of the power amplifier 140. According to example embodiments, predistortion can be performed based on a power amplifier model that more accurately reflects the characteristics of the power amplifier 140 through the multi-tone signal MT_S having a uniform PDF. As a result, degradation of predistortion performance can be prevented.

[0071] Figure 7 is a graph showing error vector magnitude (EVM) for an exemplary training signal.

[0072] Figure 7 The case where pre-distortion (e.g., DPD) and pre-distortion (e.g., DPD) are performed and the case where pre-distortion (e.g., DPD) is not performed using a QPSK-modulated CP-OFDM signal and a multi-tone signal MT_S as exemplary training signals is shown. Polynomial fitting with a fifth-order polynomial (poly5) (e.g., CP-QPSK_poly5), polynomial fitting with a ninth-order polynomial (poly9) (e.g., CP-QPSK_poly9 and Multi-tone_poly9), and moving average (e.g., Multi-tone_moving) are applied to the exemplary training signals.

[0073] As mentioned above, since moving average cannot be applied due to the non-uniform PDF, only polynomial fitting is used for the modulation signal M_S.

[0074] For example, in Figure 7 , it can be seen that when the multi-tone signal MT_S is used, the EVM is lower in the high power interval compared to the modulated signal M_S (e.g., QPSK modulated CP-OFDM signal). This is because the multi-tone signal MT_S has a uniform PDF, so sufficient samples can be ensured even in the high power interval.

[0075] Furthermore, when the multi-tone signal MT_S is used, a moving average can be applied. Figure 7 , it can be seen that when the moving average is applied, the NMSE is almost similar to the NMSE of the polynomial fit with a ninth-order polynomial (poly9). This indicates that the disadvantages caused by using a high-order polynomial (e.g., calibration time for polynomial determination, etc.) can be reduced by using the multi-tone signal MT_S.

[0076] Figure 8 is a diagram showing an RF circuit according to an example embodiment. Figure 1 Detailed description of repeated elements may be omitted.

[0077] Reference Figure 8 , the RF circuit 100A according to example embodiments may include a phase determination circuit 160 in addition to the multi-tone generation circuit 110 , the multiplexer 120 , the predistortion circuit 130 , the power amplifier 140 , and the predistortion modeling circuit 150 .

[0078] The phase determination circuit 160 may be configured to determine the phase of the multi-tone generation circuit 110 (or Figure 4 The phase value of each of the multiple tones in the multi-tone signal MT_S generated by the multi-tone generation circuit 153 of FIG. 154 is shown in FIG. 155 . For example, the phase value may be θ of Equation 1 or Equation 2. k The phase determination circuit 160 may adjust the phase value of each of the plurality of tones so that the multi-tone signal MT_S has a uniform PDF.

[0079] According to an example embodiment, the phase determination circuit 160 may calculate an initial phase value (or referred to as, a phase initial value) for each of the plurality of tones, and may adjust the initial phase value based on a peak-to-average power ratio (PAPR) of the multi-tone signal MT_S to determine a final phase value. For example, the phase determination circuit 160 may adjust a plurality of initial phase values ​​corresponding to the plurality of tones to determine a plurality of final phase values ​​corresponding to the plurality of tones. For example, the phase determination circuit 160 may determine a phase value when the PAPR of the multi-tone signal MT_S is significantly reduced as the final phase value. In other words, the phase determination circuit 160 may identify a phase value as the final phase value when the PAPR of the multi-tone signal MT_S is significantly reduced.

[0080] When the training signal has a relatively high PAPR, the PDF of the training signal is also likely to be non-uniform. Therefore, the phase determination circuit 160 according to the example embodiment may calculate the PAPR of the training signal and may adjust the phase value in a direction in which the PAPR is significantly reduced. This adjustment ensures that the multi-tone signal MT_S has a relatively uniform PDF.

[0081] According to example embodiments, the phase determination circuit 160 may determine a phase value at which the PAPR of the multi-tone signal MT_S is significantly reduced, and may transmit phase information PI including the determined phase value to the multi-tone generation circuit 110 and / or the predistortion modeling circuit 150 .

[0082] The multi-tone generation circuit 110 may receive the phase information PI from the phase determination circuit 160, and may generate the multi-tone signal MT_S based on the phase value included in the phase information PI. Therefore, the generated multi-tone signal MT_S may have a relatively uniform PDF. In addition, the multi-tone generation circuit 110 may generate the multi-tone signal MT_S to have a predetermined phase value without additionally determining the phase of each of the plurality of tones. Based on the phase value included in the phase information PI, the pre-distortion modeling circuit 150 may generate an updated pre-distortion model PD_M.

[0083] Then, the multiplexer 120 may select the multi-tone signal MT_S as a training signal and transmit the selected multi-tone signal MT_S to the predistortion circuit 130. The predistortion circuit 130 may generate a predistortion signal PD_S based on the updated predistortion model PD_M from the predistortion modeling circuit 150. The power amplifier 140 may amplify the predistortion signal PD_S and may output an output signal OUT.

[0084] According to the above exemplary embodiment, the RF circuit 100A can determine the phase value to significantly reduce the PAPR so that the multi-tone signal MT_S has a uniform PDF. Therefore, the RF circuit 100A can use the multi-tone signal MT_S with a uniform PDF as a training signal for predistortion to prevent the degradation of predistortion performance.

[0085] In the following, some examples of the phase determination circuit 160 will be described in more detail.

[0086] Fig. 9 It is shown Figure 8 Schematic diagram of the phase determination circuit shown in .

[0087] Reference Fig. 9 , the phase determination circuit 160 according to example embodiments may include a phase calculation circuit 161 , a PAPR calculation circuit 162 , and a controller 163 .

[0088] The phase calculation circuit 161 may be configured to calculate a plurality of initial phase values ​​of a plurality of tones included in the multi-tone signal MT_S. For example, the phase calculation circuit 161 may calculate the initial phase values ​​based on an algorithm for calculating initial phase values, such as a Newman algorithm, a Kitayoshi algorithm, a Schroeder algorithm, and a Narahashi algorithm. For example, the following equations 4 to 7 may be defined as initial phase values ​​based on the Newman algorithm, the Kitayoshi algorithm, the Schroeder algorithm, and the Narahashi algorithm, respectively.

[0089] ...Equation 4.

[0090] ...Equation 5.

[0091] ...Equation 6.

[0092] ...Equation 7.

[0093] In Equations 4 to 7, θ k is the initial phase value of the kth tone, k is the index of each tone, and N is the number of all tones. Here, k and N may be positive integers. In addition, the phase calculation circuit 161 may calculate the phases of the plurality of tones based on various methods for obtaining the phases of the plurality of tones, without being limited to the above-mentioned algorithm. The initial phase value will be calculated for each tone, so that a plurality of initial phase values ​​may be calculated for the multi-tone signal MT_S by the phase calculation circuit 161.

[0094] The PAPR calculation circuit 162 may calculate the PAPR of the multi-tone signal MT_S. Hereinafter, for ease of description, the PAPR of the multi-tone signal MT_S corresponding to a plurality of initial phase values ​​will be referred to as an initial PAPR (or referred to as, PAPR initial value). In addition, the PAPR corresponding to an adjusted phase set including at least one adjusted phase value will be referred to as a plurality of adjusted PAPR values ​​(or referred to as, a plurality of PAPR adjustment values), and the at least one adjusted phase value is determined after adjusting at least one of the plurality of initial phase values. The adjusted phase set (or referred to as, an adjusted phase set) may be at least one set of phase values ​​including the at least one adjusted phase value. For example, when θ of the first tone among the N tones in Equations 4 to 7 is 1 and the second tone θ 2 When the phases are adjusted sequentially, the adjusted phase sets may include a first set and a second set, the first set including the adjusted θ 1 and maintain theta 2 to θ N , the second set includes the adjusted θ2 and maintain theta 1 and θ 3 to θ N Similarly, when θ k When adjusted, the adjusted phase set may include the adjusted θ k and the remaining set of retained phases.

[0095] In response thereto, the plurality of PAPR adjustment values ​​may be a set of PAPRs calculated from each of different sets included in the adjustment phase set (eg, the set may include a PAPR calculated from the above-mentioned first set and a PAPR calculated from the above-mentioned second set).

[0096] In an example embodiment, the PAPR calculation circuit 162 may calculate the PAPR of the multi-tone signal MT_S corresponding to the plurality of phase initial values ​​as the PAPR initial value. As at least one of the plurality of phase initial values ​​is adjusted, the PAPR calculation circuit 162 may calculate the PAPR of the multi-tone signal MT_S corresponding to the adjusted phase set as the plurality of PAPR adjustment values. For example, the PAPR may be a ratio of a peak value of power to an average value of power.

[0097] The controller 163 may be connected to the phase calculation circuit 161 and the PAPR calculation circuit 162 , and may determine a phase value of the multi-tone signal MT_S based on the phase initial value and the PAPR.

[0098] In an example embodiment, the controller 163 may compare the PAPR initial value with a threshold. When the PAPR initial value is less than the threshold, the controller 163 may determine the phase initial value as the final phase value. The threshold may be preset to a value required for the PAPR, or may be adjusted by the controller 163. When the PAPR initial value is less than the threshold, the PAPR of the multi-tone signal MT_S is sufficiently low that the multi-tone signal MT_S is likely to have a uniform PDF. Therefore, the controller 163 may immediately determine the phase initial value corresponding to the PAPR initial value less than the threshold as the final phase value.

[0099] Optionally, when the PAPR initial value is greater than or equal to the threshold, the controller 163 may adjust at least one of the multiple phase initial values. For example, the controller 163 may adjust the phase value based on various optimization algorithms for gradually adjusting (e.g., increasing or decreasing) the phase value. For example, the controller 163 may use a hill climbing algorithm, a simulated annealing algorithm, or a genetic algorithm as the optimization algorithm.

[0100] When the controller 163 adjusts at least one of the plurality of phase initial values, the PAPR calculation circuit 162 may calculate a plurality of PAPR adjustment values ​​of the output signal OUT corresponding to the set of adjusted phases. For example, when the phase of the k-th tone (k in equations 4 to 7) among the plurality of phase initial values ​​is adjusted, the PAPR calculation circuit 162 may calculate the PAPR adjustment value for a set including the adjusted phase and the maintained phase.

[0101] The controller 163 may compare the initial PAPR value with all the multiple PAPR adjustment values. When the initial PAPR value is less than the multiple PAPR adjustment values ​​(e.g., when the initial PAPR value is less than all the PAPR adjustment values), the controller 163 may determine the initial phase value corresponding to the initial PAPR value as the final phase value. For example, when the PAPR initial value is less than the multiple PAPR adjustment values, the controller 163 may determine the initial phase value as the final phase value. This is because there is no more room for PAPR improvement through phase adjustment.

[0102] Optionally, when at least one of the plurality of PAPR adjustment values ​​is less than the PAPR initial value (for example, when there is at least one PAPR adjustment value lower than the PAPR initial value), the controller 163 may set the minimum value of the at least one PAPR adjustment value as the PAPR initial value. After setting the minimum value as the PAPR initial value, the controller 163 may determine the phase value based on the reset PAPR initial value.

[0103] The controller 163 may transmit the phase information PI including the finally determined phase value.

[0104] According to the above-described exemplary embodiment, the RF circuit can determine the phase of each tone through the phase determination circuit 160 so that the multi-tone signal MT_S has a uniform PDF. Therefore, the component for generating the multi-tone signal MT_S (e.g., the multi-tone generation circuit 110) does not require an additional phase determination operation. In addition, the degradation of the pre-distortion performance can be prevented by the multi-tone signal MT_S having a determined phase.

[0105] Fig.10 is a diagram showing the PDF of an exemplary training signal.

[0106] Reference Fig.10 , the exemplary training signals are all multi-tone signals (MT_S), and the following three cases can be provided: a case where the phases of the tones are all zero (in-phase), a case where the phases of the tones are randomly different from each other (out-of-phase), and a case where the phases of the tones are appropriately determined according to the exemplary embodiment (custom phase).

[0107] In the case of in-phase, it can be seen that the PDF non-uniformity is significantly high, and the difference between the PDF of the interval with low input power and the PDF of the interval with high input power is particularly large. In the case of out-of-phase, it can be seen that the PDF uniformity is relatively increased compared to the case of in-phase, but there are still some differences between the PDF of the interval with low input power and the PDF of the interval with high input power.

[0108] When the phase is determined (for example, when the phase is determined to be a custom phase that significantly reduces the PAPR), it can be seen that the PDF uniformity is the highest. As a result, when the corresponding multi-tone signal MT_S is used as a training signal, the characteristics of the power amplifier 140 can be more accurately reflected to prevent the degradation of the pre-distortion performance.

[0109] Fig.11 is a flow chart illustrating a method of operating an RF circuit according to an example embodiment.

[0110] Reference Fig.11 In operation S110, the RF circuit 100 or 100A may generate a multi-tone signal MT_S including a plurality of tones. The RF circuit 100 or 100A may generate the multi-tone signal MT_S having a phase determined according to an example embodiment. Each of the plurality of tones included in the multi-tone signal MT_S may have a phase determined to significantly reduce a PAPR or have a uniform PDF.

[0111] In operation S120, the RF circuit 100 or 100A may predistort the multi-tone signal MT_S to output a predistortion signal PD_S. The predistortion in operation S120 may be based on the predistortion model PD_M before or after being updated. The multi-tone signal MT_S may be used as a training signal for predistortion.

[0112] In operation S130, the RF circuit 100 or 100A may amplify the predistortion signal PD_S to output an output signal OUT. The output signal OUT output in operation S130 may be used to model the predistortion model PD_M in operation S140.

[0113] In operation S140, the RF circuit 100 or 100A may model the predistortion model PD_M based on the predistortion signal PD_S corresponding to the multi-tone signal MT_S (e.g., the predistortion signal PD_S output in operation S120) and the output signal OUT of the power amplifier 140. The process of modeling the predistortion model PD_M may be described as an action of establishing a predistortion LUT or predistortion coefficients. As described above, these elements are designed to compensate for the nonlinearity of the output signal OUT, and then update the predistortion LUT or predistortion coefficients in the predistortion circuit 130.

[0114] In the above method according to example embodiments, operations S120 to S140 may be repeated until the requirement of compensating for nonlinearity of the output signal OUT is satisfied to the greatest extent. The multi-tone signal MT_S used as a training signal has a uniform PDF so that degradation of predistortion performance may be prevented.

[0115] According to an example embodiment, the method may further include predistorting the modulation signal M_S using a predistortion model PD_M optimized to compensate for the nonlinearity of the power amplifier 140. This is followed by amplifying and outputting the predistorted modulation signal M_S.

[0116] According to an example embodiment, the method may further include determining a phase value for each of the plurality of tones in the multi-tone signal MT_S.

[0117] Fig.12 is a flow chart illustrating a method for modeling a pre-distortion model of an RF circuit according to an example embodiment. For example, Fig.12 Can be used with Fig.11 Corresponding to operation S140.

[0118] Reference Fig.12 In operation S210, the RF circuit 100 or 100A may model a power amplifier model based on a moving average method, the power amplifier model being the relationship between the predistortion signal PD_S corresponding to the multi-tone signal MT_S and the output signal OUT. As described above, when the multi-tone signal MT_S is used as a training signal, a moving average method requiring a uniform PDF in curve fitting for modeling the power amplifier model may be adopted. Therefore, the power amplifier model may be modeled without the burden of having to perform a specific operation (e.g., an operation for determining a high-order polynomial in polynomial fitting).

[0119] In operation S220 , the RF circuit 100 or 100A may model a predistortion model PD_M to compensate for nonlinearity of the power amplifier model modeled in operation S210 .

[0120] Fig.13 is a flow chart illustrating a method of determining a phase of an RF circuit according to an example embodiment.

[0121] Reference Fig.13 In operation S310, the RF circuit 100A may calculate a plurality of initial values ​​of phases of a plurality of tones. For example, calculation of the initial value of the phase may be performed based on Equations 4 to 7.

[0122] In operation S320, the RF circuit 100A may calculate a PAPR initial value (PAPR) of the multi-tone signal MT_S corresponding to the plurality of phase initial values.init ).

[0123] In operation S330 , the RF circuit 100A may compare the PAPR initial value with a threshold value.

[0124] In operation S330, when the PAPR initial value is less than the threshold, the process proceeds to operation S340, in which the RF circuit 100A may determine the phase initial value as the phase value (e.g., the final phase value). For example, when the PAPR initial value satisfies the required threshold, the RF circuit 100A may use the calculated phase initial value as is.

[0125] Optionally, in operation S330, when the PAPR initial value is greater than the threshold, the process proceeds to operation S350, in which the RF circuit 100A may adjust at least one of the plurality of phase initial values, and may calculate a plurality of PAPR adjustment values ​​of the output signal OUT corresponding to the adjustment phase set including the adjusted at least one phase initial value. Through operation S350, the PAPR adjustment value of each of the plurality of sets included in the adjustment phase set may be calculated. For example, the RF circuit 100A may adjust the phase initial value based on various optimization algorithms.

[0126] In operation S360 , the RF circuit 100A may compare the PAPR initial value with the obtained plurality of PAPR adjustment values ​​(PAPR{phase set}).

[0127] In operation S360 , when the PAPR initial value is less than the plurality of PAPR adjustment values ​​(eg, when the PAPR initial value is less than all the PAPR adjustment values), the RF circuit 100A may determine the phase initial value as the phase value.

[0128] Optionally, in operation S360, when at least one of the multiple PAPR adjustment values ​​is less than the PAPR initial value (for example, when even at least one PAPR adjustment value is lower than the PAPR initial value), the process proceeds to operation S370, in which the RF circuit 100A may set the minimum value (minPAPR{phase set}) of the at least one PAPR adjustment value as the PAPR initial value.

[0129] Then, the RF circuit 100A may repeat operations S330, S350, S360, and S370 for the reset PAPR initial value. For example, the RF circuit 100A may determine the phase value by repeating the following operations: comparing the reset PAPR initial value with a threshold, calculating a PAPR adjustment value of an adjusted phase set, and comparing the PAPR initial value with the PAPR adjustment value of the adjusted phase set.

[0130] Fig.14is a diagram illustrating a training information determination circuit according to example embodiments.

[0131] Reference Fig.14 The training information determination circuit 200 may include Figure 8 and Fig. 9 The operation of the phase determination circuit 160 corresponds to the operation described above.

[0132] The training information determination circuit 200 may be configured to generate training information TI including various types of information for generating a multi-tone signal MT_S to be used as a training signal, and provide the training information TI to the multi-tone generation circuit 110 and / or the predistortion modeling circuit 150. For example, the training information TI may include a phase value of each of the plurality of tones determined by the phase determination circuit 160, the number of tones, and the bandwidth of the multi-tone signal MT_S. The training information determination circuit 200 may generate a training signal having a uniform PDF by selecting, for example, a bandwidth that is closest to the bandwidth of a signal to which predistortion is ultimately applied (e.g., the modulation signal M_S). Alternatively, the training information determination circuit 200 may generate a training signal having a uniform PDF by selecting, for example, the number of tones that can represent all frequency characteristics of the power amplifier 140 within the bandwidth.

[0133] In an example embodiment, the training information determination circuit 200 may determine the training information TI based on at least one of the following items: 1) whether a pre-distortion model PD_M reflecting the memory effect has been used (e.g., a pre-distortion model with memory or a pre-distortion model without memory), 2) the number of samples of the training signal and / or the sampling rate of the samples, and 3) the bandwidth of the signal to which pre-distortion will ultimately be applied.

[0134] In an example embodiment, when the predistortion model PD_M is memoryless, the training information determination circuit 200 may determine that a bandwidth lower than the bandwidth of the modulation signal M_S is the bandwidth of the multi-tone signal MT_S. For example, the memoryless predistortion model may be a model that averages performance characteristics based on frequency. In this case, it may be determined which of averaging the frequency characteristics of a training signal with a lower bandwidth and averaging the frequency characteristics of a training signal with a wider bandwidth is better in terms of predistortion performance based on the frequency characteristics of the power amplifier 140. In addition, the wider the bandwidth of the training signal, the more memory effects can be reflected. However, the need to widen the bandwidth in a memoryless predistortion model that does not consider the memory effect can be reduced.

[0135] Therefore, when the predistortion model PD_M has no memory, the training information determination circuit 200 may determine the bandwidth of the multi-tone signal MT_S to be lower than the bandwidth of the modulation signal M_S. In this case, the multi-tone generation circuit 110 may also generate a multi-tone signal MT_S having a lower bandwidth (e.g., a multi-tone signal MT_S having a bandwidth lower than the bandwidth of the modulation signal M_S). The training information determination circuit 200 may also determine whether to reduce the bandwidth by considering the number of available samples.

[0136] Alternatively, in an example embodiment, when the pre-distortion model PD_M reflects the memory effect, the training information determination circuit 200 may increase the bandwidth to more reflect the memory effect.

[0137] As the number of tones increases, the number of frequency characteristics considered also increases. Therefore, the pre-distortion model PD_M may reflect the memory effect more. As a result, in an example embodiment, when the pre-distortion model reflects the memory effect, the training information determination circuit 200 may increase the number of tones.

[0138] According to an example embodiment, when the predistortion model PD_M has no memory, the training information determination circuit 200 may reduce the number of tones to save resources for generating a training signal. However, when the number of tones is significantly reduced, performance degradation may be more likely to occur when predistortion is applied to a modulation signal M_S having a higher bandwidth than the multi-tone signal MT_S due to a smaller number of samples. Accordingly, the training information determination circuit 200 may determine the number of tones of the multi-tone signal MT_S to be greater than the number of required tones.

[0139] Fig.15 is a diagram illustrating a wireless communication device according to an example embodiment.

[0140] Reference Fig.15 , the wireless communication device 300 may include a modem 310 , a radio frequency integrated circuit (RFIC) 320 , a power amplifier 330 , a duplexer 340 , and an antenna 350 .

[0141] The modem 310 may include a digital processing circuit 311, an analog-to-digital converter (ADC) 315, and a digital-to-analog converter (DAC) 314. The modem 310 may process a baseband signal BB_T including information to be transmitted (including, for example, an I signal and a Q signal) through the digital processing circuit 311 using various communication methods. The modem 310 may process a received baseband signal BB_R through the digital processing circuit 311 using various communication methods.

[0142] For example, the modem 310 may use a communication method such as orthogonal frequency division multiplexing (OFDM), orthogonal frequency division multiple access (OFDMA), wideband code division multiple access (WCDMA), or high-speed packet access + (HSPA +) to process a signal to be transmitted or a received signal. In addition, the modem 310 may use various communication methods to process the baseband signal BB_T or BB_R. For example, the modem 310 may use various communication methods that employ a technique for modulating or demodulating the amplitude and frequency of the baseband signal BB_T or BB_R.

[0143] The digital processing circuit 311 may perform various processing operations on the baseband signal in the digital domain. The digital processing circuit 311 may perform predistortion operations according to the various embodiments described above. The digital processing circuit 311 may include a multi-tone generation circuit 312 and a predistortion circuit 313, wherein the multi-tone generation circuit 312 generates a training signal (e.g., a multi-tone signal) for predistortion.

[0144] The multi-tone generation circuit 312 may generate a multi-tone signal to be used as a training signal for the predistortion circuit 313 according to the above-described example embodiments.

[0145] The predistortion circuit 313 may perform a predistortion operation according to the above-described embodiments. For example, the predistortion circuit 313 may be configured to predistort the baseband signal BB_T based on the multi-tone signal and the predistortion model. The predistortion circuit 313 may predistort the baseband signal BB_T to compensate for the nonlinearity of the power amplifier 330. For example, in a changed scenario, the nonlinearity of the power amplifier 330 may be compensated by adopting a gain providing circuit.

[0146] In an example embodiment, modem 310 may generate training information including phase values ​​of tones, the number of tones, and bandwidths of tones used to generate a multi-tone signal. For example, modem 310 may determine the phase values ​​based on the PAPR of the multi-tone signal. For example, modem 310 may determine the training information based on at least one of the following: 1) whether a predistortion model reflecting memory effects has been used, 2) the number of samples of the training signal and / or the sampling rate of the samples, and 3) the bandwidth of the signal to which predistortion will ultimately be applied.

[0147] Each of the ADC 315 and the DAC 314 may be provided in a singular or plural number. The modem 310 may convert the baseband signal BB_T into an analog signal using the DAC 314 to generate a transmission signal TX. In addition, the modem 310 may receive a reception signal RX as an analog signal from the RFIC 320. The modem 310 may perform digital conversion on the reception signal RX through the ADC 315 provided in the modem 310 to extract the baseband signal BB_R as a digital signal. For example, the reception signal RX may be a differential signal including a positive signal and a negative signal.

[0148] The RFIC 320 may perform up-conversion on a transmission signal TX to generate an RF input signal RF_IN, or may perform down-conversion on an RF reception signal RF_R to generate a reception signal RX. For example, the RFIC 320 may include a transmission circuit TXC, a reception circuit RXC, and a local oscillator LO.

[0149] The transmission circuit TXC may include a first analog baseband filter ABF1, a first mixer MX1, and a driver amplifier 321. For example, the first analog baseband filter ABF1 may include a low-pass filter.

[0150] The first analog baseband filter ABF1 may filter the transmission signal TX received from the modem 310 and provide it to the first mixer MX1. Then, the first mixer MX1 ​​may perform up-conversion to convert the frequency of the transmission signal TX from the baseband to the high frequency band by the frequency signal provided by the local oscillator LO. Through the up-conversion, the transmission signal TX may be provided to the driver amplifier 321 as the RF input signal RF_IN, and the driver amplifier 321 may amplify the RF input signal RF_IN for the first time and provide it to the power amplifier 330.

[0151] The power amplifier 330 may receive a DC voltage or a variable power supply voltage (eg, a dynamically variable output voltage), and may amplify the power of the RF input signal RF_IN for the second time based on the received power supply voltage to generate an RF output signal RF_OUT. In this case, the nonlinearity of the RF output signal RF_OUT may be compensated by the predistortion circuit 313.

[0152] The receiving circuit RXC may include a second analog baseband filter ABF2, a second mixer MX2, and a low noise amplifier (LNA) 322. For example, the second analog baseband filter ABF2 may include a low pass filter.

[0153] The LNA 322 may amplify the RF reception signal RF_R received from the duplexer 340 and provide it to the second mixer MX2. Then, the second mixer MX2 may perform down-conversion to convert the frequency of the RF reception signal RF_R from a high frequency band to a baseband by a frequency signal provided by the local oscillator LO. Through the down-conversion, the RF reception signal RF_R may be provided as a reception signal RX to the second analog baseband filter ABF2, and the second analog baseband filter ABF2 may filter the reception signal RX and provide it to the modem 310.

[0154] For reference, the wireless communication device 300 may transmit a transmission signal (eg, TX) through a plurality of frequency bands using carrier aggregation (CA). To achieve this, the wireless communication device 300 may include a plurality of power amplifiers 330 for amplifying a plurality of RF input signals RF_IN respectively corresponding to a plurality of carriers. In the present embodiment, for clarity of description and explanation, an example in which the power amplifier 330 is provided in a singular has been described.

[0155] The duplexer 340 may be connected to the antenna 350 to separate the transmission frequency and the reception frequency. For example, the duplexer 340 may separate the RF output signal RF_OUT provided by the power amplifier 330 for each frequency band and provide the separated signal to the corresponding antenna 350. In addition, the duplexer 340 may provide an external signal received from the antenna 350 to the LNA 322 of the receiving circuit RXC of the RFIC 320. For example, the duplexer 340 may include a front end module with an integrated duplexer (FEMiD).

[0156] For reference, the wireless communication device 300 may have a switch structure for separating the transmission frequency and the reception frequency instead of the duplexer 340. In addition, the wireless communication device 300 may have a structure including the duplexer 340 and the switch for separating the transmission frequency and the reception frequency. For ease of description, in the present embodiment, an example in which the duplexer 340 for separating the transmission frequency and the reception frequency is provided in the wireless communication device 300 has been described.

[0157] The antenna 350 may transmit an RF output signal RF_OUT frequency-divided by the duplexer 340 to an external entity or provide an RF reception signal RF_R received from the external entity to the duplexer 340. For example, the antenna 350 may include an array antenna, but example embodiments are not limited thereto.

[0158] For reference, each of the modem 310, the RFIC 320, the power amplifier 330, and the duplexer 340 may be implemented as an integrated circuit (IC), a chip, or a module. In addition, the modem 310, the RFIC 320, the power amplifier 330, and the duplexer 340 may be mounted together on a printed circuit board (PCB), but example embodiments are not limited thereto. In some embodiments, at least a portion of the modem 310, the RFIC 320, the power amplifier 330, and the duplexer 340 may be implemented as a single communication chip.

[0159] also, Fig.15 The wireless communication device 300 shown in FIG. 1 may be included in a wireless communication system using a cellular network such as 5G or LTE, or may be included in a wireless local area network (WLAN) system or any other arbitrary wireless communication system. For reference, Fig.15 The configuration of the wireless communication device 300 shown in FIG. 1 is only an example, and the exemplary embodiments are not limited thereto. The wireless communication device 300 may be configured in various ways according to a communication protocol or a communication method. Fig.15 The wireless communication device 300 is shown in FIG.

[0160] Fig.16 is a diagram illustrating an electronic device according to example embodiments.

[0161] Reference Fig.16 , the electronic device 400 may be, for example, a mobile terminal, and may include an application processor (AP) 410, a memory 420, a display 430, and an RF module 440. The electronic device 400 may further include various components such as a lens, a sensor, or an audio module.

[0162] The AP 410 may be implemented as a system on chip (SoC) and may include a central processing unit (CPU) 411, a RAM 412, a power management unit (PMU) 413, a memory interface (memory I / F) 414, a display controller (DCON) 415, a modem (MODEM) 416, and a bus (system bus) 417. The AP 410 may also include various intellectual property (IP) cores. Since the functions of the modem chip are integrated in the AP 410, the AP 410 may be referred to as a ModAP.

[0163] The CPU 411 may control the overall operation of the AP 410 and the electronic device 400. The CPU 411 may be configured to execute at least one instruction stored in the RAM 412, and may control the operation of each component of the AP 410 through the instruction. The CPU 411 may also be implemented as a multi-core. A multi-core may be a computing component having two or more independent cores.

[0164] In an example embodiment, the CPU 411 may control the modem 416 and / or the RF module 440 to perform the following operations: generate a multi-tone signal including a plurality of tones, predistort the multi-tone signal or a modulated signal to output a predistorted signal, amplify the predistorted signal to output an output signal OUT, and model a predistortion model based on the predistorted signal corresponding to the multi-tone signal and the output signal OUT.

[0165] For example, the CPU 411 may control the modem 416 and / or the RF module 440 to perform the following operations: modeling a power amplifier model based on a moving average method, and modeling a predistortion model to compensate for the nonlinearity of the power amplifier model, wherein the power amplifier model is the relationship between a predistortion signal corresponding to a multi-tone signal and the output signal OUT.

[0166] For example, the CPU 411 may control the modem 416 and / or the RF module 440 to perform the following operations: calculate multiple phase initial values ​​of multiple tones, calculate a PAPR initial value of a multi-tone signal corresponding to the multiple initial phase values, and determine the phase initial value as a phase value of each of the multiple tones based on the initial PAPR value being less than a threshold.

[0167] The RAM 412 may temporarily store a program, data, or at least one instruction. For example, the program and / or data stored in the memory 420 may be temporarily stored in the RAM 412 under the control of the CPU 411 or the boot code. The RAM 412 may be implemented as a DRAM or an SRAM.

[0168] The PMU 413 may manage power of each component of the AP 410. The PMU 413 may determine an operation state of each component of the AP 410 and control the operation of each component of the AP 410.

[0169] The memory interface 414 may control the overall operation of the memory 420, and may control data exchange between the memory 420 and each component of the AP 410. The memory interface 414 may write data in the memory 420 or read data from the memory 420 according to a request of the CPU 411. For example, according to the above-described embodiments, the memory 420 may store various types of information (e.g., multi-tone signals, training information, etc.). For example, the memory 420 may store a power amplifier model and a predistortion model as described in the above-described embodiments.

[0170] The display controller 415 may transmit video data to be displayed on the display 430 to the display 430. The display 430 may be implemented as a flat display such as a liquid crystal display (LCD), an organic light emitting diode (OLED), or a flexible display.

[0171] For wireless communication, the modem 416 may appropriately modulate data to be transmitted for a wireless environment and recover received data. The modem 416 may perform digital communication with the RF module 440.

[0172] For reference, the modem 416 may utilize the above reference Fig.15 The described modem 310 is implemented.

[0173] The RF module 440 may convert a high frequency signal received through the antenna into a low frequency signal, and transmit the converted low frequency signal to the modem 416. In addition, the RF module 440 may convert a low frequency signal received from the modem 416 into a high frequency signal, and transmit the converted high frequency signal through the antenna to the outside of the electronic device 400. In addition, the RF module 440 may amplify or filter the signal.

[0174] For reference, the RF module 440 can utilize the above reference Fig.15 The RFIC 320, power amplifier 330, duplexer 340 and antenna 350 described above are implemented.

[0175] Alternatively, the modem 416 and the RF module 440 may be implemented using at least one of the RF circuit 100 or 100A, the phase determination circuit 160 , and the training information determination circuit 200 according to example embodiments.

[0176] As previously described, according to example embodiments, an RF circuit for predistortion and a method of operating the RF circuit for predistortion may be provided.

[0177] While example embodiments have been shown and described above, it will be apparent to those skilled in the art that modifications and variations may be made thereto without departing from the scope of the inventive concept as set forth in the appended claims.

Claims

1. A radio frequency circuit, comprising: a multi-tone generation circuit configured to generate a multi-tone signal including a plurality of tones; A predistortion circuit configured to predistort a multi-tone signal or a modulated signal and output a predistorted signal; a power amplifier configured to amplify the predistortion signal and output an output signal; as well as The predistortion modeling circuit is configured to model a predistortion model corresponding to the predistortion circuit based on the predistortion signal corresponding to the multi-tone signal and the output signal.

2. The radio frequency circuit according to claim 1, wherein: A multi-tone signal is a real signal.

3. The radio frequency circuit according to claim 1, further comprising: A multiplexer is configured to select one of the multi-tone signal and the modulated signal and send the selected signal to the predistortion circuit.

4. The radio frequency circuit according to claim 1, wherein: The predistortion circuit is configured to receive a predistortion model from the predistortion modeling circuit and perform predistortion based on the predistortion model.

5. The radio frequency circuit according to claim 1, wherein: The predistortion modeling circuit is configured as: Modeling a power amplifier model based on a moving average method, the power amplifier model being a relationship between a predistortion signal corresponding to a multi-tone signal and an output signal; and A predistortion model is modeled to compensate for the nonlinearity of the power amplifier model.

6. The radio frequency circuit according to claim 5, wherein: The moving average method is a method of dividing an input axis of a power amplifier model into a plurality of intervals, and selecting an average value of outputs included in each of the plurality of intervals as a representative value of each of the plurality of intervals.

7. The radio frequency circuit according to any one of claims 1 to 6, further comprising: The phase determination circuit is configured to determine a plurality of phase values ​​for the plurality of tones in the multi-tone signal.

8. The radio frequency circuit according to claim 7, wherein: The phase determination circuit is configured to: calculate multiple phase initial values ​​of the multiple tones, calculate peak-to-average power ratio initial values ​​of the multi-tone signal corresponding to the multiple phase initial values, and determine the multiple phase initial values ​​as the multiple phase values ​​when the peak-to-average power ratio initial values ​​are less than a threshold.

9. The radio frequency circuit according to claim 8, wherein: The phase determination circuit is configured to: adjust at least one of the multiple phase initial values ​​when the peak-to-average power ratio initial value is greater than a threshold, calculate multiple peak-to-average power ratio adjustment values ​​of the output signal corresponding to the adjustment phase set including the adjusted at least one phase initial value, and determine the multiple phase initial values ​​as the multiple phase values ​​when the peak-to-average power ratio initial value is less than the multiple peak-to-average power ratio adjustment values.

10. The radio frequency circuit according to claim 9, wherein: The phase determination circuit is configured to: when at least one of the multiple peak-to-average power ratio adjustment values ​​is less than the peak-to-average power ratio initial value, set the minimum value of the at least one peak-to-average power ratio adjustment value as the peak-to-average power ratio initial value, and determine the multiple phase values ​​based on the peak-to-average power ratio initial value after setting the minimum value.

11. The radio frequency circuit according to any one of claims 1 to 6, wherein: The multi-tone generation circuit is configured to generate the multi-tone signal to have a bandwidth lower than a bandwidth of the modulation signal when the pre-distortion model has no memory.

12. A method of operating a radio frequency circuit, the method comprising: generating a multi-tone signal comprising a plurality of tones; Predistorting a multi-tone signal and outputting a predistorted signal; amplifying the predistortion signal and outputting an output signal; as well as A predistortion model is modeled based on a predistortion signal corresponding to the multi-tone signal and an output signal.

13. The method according to claim 12, further comprising: The modulated signal is predistorted based on the predistortion model.

14. The method according to claim 12, wherein: The steps to model the predistortion model include: Modeling a power amplifier model based on a moving average method, the power amplifier model being a relationship between a predistortion signal corresponding to a multi-tone signal and an output signal; and A predistortion model is modeled to compensate for the nonlinearity of the power amplifier model.

15. The method according to any one of claims 12 to 14, further comprising: A plurality of phase values ​​for the plurality of tones in the multi-tone signal are determined.

16. The method according to claim 15, wherein: The step of determining the plurality of phase values ​​comprises: calculating a plurality of phase initial values ​​of the plurality of tones; calculating initial values ​​of peak-to-average power ratios of the multi-tone signal corresponding to the plurality of initial phase values; and When the peak-to-average power ratio initial value is less than a threshold, the multiple phase initial values ​​are determined as the multiple phase values.

17. The method according to claim 16, wherein: The step of determining the plurality of phase values ​​further comprises: When the peak-to-average power ratio initial value is greater than a threshold, adjusting at least one of the multiple phase initial values; Calculating a plurality of peak-to-average power ratio adjustment values ​​of the output signal corresponding to the adjusted phase set including the at least one adjusted phase initial value; and When the peak-to-average power ratio initial value is less than the multiple peak-to-average power ratio adjustment values, the multiple phase initial values ​​are determined as the multiple phase values.

18. An electronic device comprising: a memory configured to store at least one instruction; as well as at least one processor configured to execute the at least one instruction, in, The at least one processor executes the at least one instruction to: generating a multi-tone signal comprising a plurality of tones; Predistorting a multi-tone signal or a modulated signal and outputting a predistorted signal; amplifying the predistortion signal and outputting an output signal; and A predistortion model is modeled based on a predistortion signal corresponding to the multi-tone signal and an output signal.

19. The electronic device according to claim 18, wherein: The at least one processor executes the at least one instruction to: Modeling a power amplifier model based on a moving average method, the power amplifier model being a relationship between a predistortion signal corresponding to a multi-tone signal and an output signal; and A predistortion model is modeled to compensate for the nonlinearity of the power amplifier model.

20. The electronic device according to claim 18, wherein: The at least one processor executes the at least one instruction to: calculating a plurality of phase initial values ​​of the plurality of tones; Calculating initial values ​​of peak-to-average power ratios of the multi-tone signal corresponding to the plurality of initial phase values; and When the peak-to-average power ratio initial value is less than a threshold, the multiple phase initial values ​​are determined as multiple phase values ​​of the multiple tones.

Images