Distortion compensation device

Abstract

To provide a highly practical distortion compensation device.SOLUTION: A distortion compensation device includes a signal output terminal to which an amplifier is connected, a first distortion compensation portion provided between a signal input terminal and the signal output terminal, which compensates for a first distortion occurring in an output of the amplifier for an input signal input from the signal input terminal, a time constant derivation portion which derives in real time a time constant for a second distortion occurring in the output of the amplifier and changing more quickly over time than the first distortion, a second distortion compensation portion connected to the signal input terminal in parallel with the first distortion compensation portion, which compensates for the second distortion in the input signal on the basis of the time constant for the second distortion derived by the time constant derivation portion, and an adder having a first input terminal connected to the output terminal of the first distortion compensation portion, a second input terminal connected to the output terminal of the second distortion compensation portion, and an output terminal connected to the signal output terminal, which adds the input signal for which the second distortion has been compensated for by the second distortion compensation portion to the input signal for which the first distortion has been compensated for by the first distortion compensation portion, and outputs the result to the signal output terminal.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a distortion compensation device. [Background technology]

[0002] A conventional digital predistortion (DPD) system includes an input section receiving a DPD input signal associated with a first sampling frequency, a first predistortion circuit having a first signal path connected to the input section, generating a first predistortion signal, and including a first infinite impulse response (IIR) filter, a second predistortion circuit having a second signal path connected to the input section in parallel with the first signal path, generating a second predistortion signal, and including a second IIR filter, and a combining circuit combining the first predistortion signal and the second predistortion signal to generate a DPD output signal. The DPD output signal is coupled to an input of a power amplifier having multiple memory effects, and a first parameter of the first IIR filter is determined based on the first sampling frequency and a first memory effect time constant associated with the first memory effect (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] U.S. Patent No. 9,866,269 Summary of the Invention [Problem to be solved by the invention]

[0004] In a DPD system, it is necessary to measure in advance the first memory effect time constant (time constant) associated with the first memory effect (memory effect). However, the time constant differs for each power amplifier and varies with the amplifier's temperature. For this reason, in a conventional DPD system, it is necessary to measure the time constant at various temperatures for each amplifier. Since it is not realistic to measure the time constant at various temperatures for each amplifier, conventional DPD systems are not very practical.

[0005] Therefore, an object of the present invention is to provide a highly practical distortion compensation device. [Means for solving the problem]

[0006] a first distortion compensation unit provided between the signal input terminal and the signal output terminal and configured to compensate for a first distortion occurring in an output of the amplifier for an input signal input from the signal input terminal; a time constant derivation unit configured to derive in real time a time constant of a second distortion occurring in the output of the amplifier, the second distortion varying more quickly over time than the first distortion; a second distortion compensation unit connected to the signal input terminal in parallel with the first distortion compensation unit and configured to compensate for the second distortion in the input signal input from the signal input terminal based on the time constant of the second distortion derived by the time constant derivation unit; and an adder having a first input terminal connected to an output terminal of the first distortion compensation unit, a second input terminal connected to an output terminal of the second distortion compensation unit, and an output terminal connected to the signal output terminal, the adder adding the input signal for which the first distortion has been compensated for by the first distortion compensation unit to the input signal for which the second distortion has been compensated for by the second distortion compensation unit and outputting the result to the signal output terminal. [Effects of the Invention]

[0007] A highly practical distortion compensation device can be provided. [Brief explanation of the drawings]

[0008] [Figure 1]1 is a diagram illustrating an example of the configuration of a wireless communication device 10 including a distortion compensation device 100 according to an embodiment. [Figure 2] 10 is a diagram illustrating an example of processing performed by the time constant derivation section 150 to determine the time constant τ of the second distortion. FIG. [Figure 3] 10 is a flowchart showing an example of processing executed by a time constant derivation section 150 of the distortion compensation device 100. [Figure 4] FIG. 10 is a diagram illustrating an example of the configuration of a wireless communication device 10 including a distortion compensation device 100M1 according to a first modified example of an embodiment. [Figure 5] FIG. 10 is a diagram illustrating an example of the configuration of a wireless communication device 10 including a distortion compensation device 100M1 according to a first modified example of an embodiment. [Figure 6] FIG. 10 is a diagram illustrating an example of the configuration of a wireless communication device 10 including a distortion compensation device 100M1 according to a first modified example of an embodiment. [Figure 7] 10 is a flowchart showing an example of processing executed by a time constant derivation unit 150 of a distortion compensation device 100M1 according to a first modified example of the embodiment. [Figure 8] 10 is a diagram showing an example of a calculation result of the degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic. FIG. [Figure 9] FIG. 10 is a diagram illustrating an example of the configuration of a wireless communication device 10 including a distortion compensation device 100M2 according to a second modified example of the embodiment. [Figure 10] FIG. 1 is a diagram illustrating an example of a transmission test signal of the E-TM2a standard defined in the LTE (Long Term Evolution) specifications. DETAILED DESCRIPTION OF THE INVENTION

[0009] The embodiments for carrying out the invention are described below.

[0010] [Description of the embodiments of the present disclosure]

[0011] [1] a first distortion compensation unit arranged between the signal input terminal and the signal output terminal, the first distortion compensation unit compensating for a first distortion occurring in an output of the amplifier for an input signal input from the signal input terminal; a time constant derivation unit deriving in real time a time constant of a second distortion occurring in the output of the amplifier, the second distortion changing more quickly over time than the first distortion; a second distortion compensation unit connected to the signal input terminal in parallel with the first distortion compensation unit, compensating for the second distortion in the input signal input from the signal input terminal based on the time constant of the second distortion derived by the time constant derivation unit; and an adder having a first input terminal connected to an output terminal of the first distortion compensation unit, a second input terminal connected to an output terminal of the second distortion compensation unit, and an output terminal connected to the signal output terminal, the adder adding the input signal for which the first distortion has been compensated for by the first distortion compensation unit to the input signal for which the second distortion has been compensated for by the second distortion compensation unit and outputting the result to the signal output terminal.

[0012] A distortion compensation device according to one embodiment of the present disclosure derives a time constant for a second distortion of an amplifier in real time, and compensates for the second distortion (distortion due to long-term memory effect) that changes faster over time than the first distortion in an input signal input from an input terminal based on the derived time constant. The input signal compensated for the first distortion is added to the input signal compensated for the second distortion, and the resulting signal is output to the amplifier. This allows compensation for the first distortion and the second distortion contained in the output of the amplifier based on the real-time time constant of the amplifier. In particular, the second distortion (distortion due to long-term memory effect) can be compensated for based on the real-time time constant of the amplifier. Therefore, a highly practical distortion compensation device can be provided.

[0013] [2] In [1], the time constant derivation unit may derive the time constant of the second distortion in real time based on the input and output of the amplifier. This allows the time constant to be accurately derived in real time based on the input signal and output signal of the amplifier, and the second distortion (distortion due to long-term memory effect) included in the output of the amplifier to be accurately compensated for in real time. Therefore, it is possible to provide a highly practical distortion compensation device that can accurately compensate for the long-term memory effect of the amplifier in real time.

[0014] [3] In [2], the time constant derivation unit may derive a time constant of the second distortion in real time based on the input and output of the amplifier using a model of the first distortion and the second distortion occurring in the output of the amplifier. This allows the time constant to be accurately derived in real time using the model of the first distortion and the second distortion occurring in the output of the amplifier, and the second distortion (distortion due to long-term memory effect) included in the output of the amplifier to be accurately compensated for in real time. Therefore, it is possible to provide a highly practical distortion compensation device that can accurately compensate for the long-term memory effect of the amplifier in real time using the model of the first distortion and the second distortion occurring in the output of the amplifier.

[0015] [4] In [2], the model may be expressed by a mathematical formula that adds a term representing the first distortion and a term representing the second distortion, and the time constant derivation unit may derive, as the time constant of the second distortion, a time constant at which the square of a predetermined coefficient included in the term representing the second distortion takes a maximum value, and the predetermined coefficient may be a coefficient included in the term representing the second distortion and representing the magnitude of the time constant of the second distortion. This makes it possible to accurately derive in real time the time constant at which the square of the predetermined coefficient included in the term representing the second distortion takes a maximum value, and to more accurately compensate in real time for the second distortion (distortion due to long-term memory effect) included in the output of the amplifier. Therefore, it is possible to provide a highly practical distortion compensation device that can more accurately compensate for the long-term memory effect of the amplifier in real time using models of the first distortion and second distortion occurring in the output of the amplifier.

[0016] [5] In the method of [2], the present invention further includes a bypass line connecting the signal input terminal and the signal output terminal without passing through the first distortion compensation unit, the second distortion compensation unit, and the adder, a first path passing through the first distortion compensation unit and the adder, or the bypass line, between the signal input terminal and the signal output terminal, and a second path connecting the second distortion compensation unit in parallel to the first path between the signal input terminal and the adder, and switching between connection and disconnection of the second path, between the signal input terminal and the adder, and the time constant derivation unit The present invention may further comprise the steps of: deriving a first time-frequency characteristic of an input to an input terminal of the amplifier when the first switching unit connects the bypass line between the signal input terminal and the amplifier and the second switching unit disconnects the second path from the first path; deriving a second time-frequency characteristic of an input to an input terminal of the amplifier when the first switching unit connects the first path between the signal input terminal and the amplifier and the second switching unit disconnects the second path from the first path; and deriving a time constant for the second distortion based on the degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic. Therefore, the second distortion (distortion due to long-term memory effect) included in the output of the amplifier can be accurately compensated for in real time based on the degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic. Therefore, a highly practical distortion compensation device can be provided that can accurately compensate for the long-term memory effect of an amplifier in real time.

[0017] [6] In [5], the time constant derivation unit may derive the first time-frequency characteristic and the second time-frequency characteristic by wavelet transform. Therefore, the second distortion (distortion due to long-term memory effect) contained in the output of the amplifier can be compensated for more accurately in real time based on the degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic derived by wavelet transform. Therefore, it is possible to provide a highly practical distortion compensation device that can compensate for the long-term memory effect of the amplifier more accurately in real time.

[0018] [7] In [5], the time constant derivation unit may derive the time constant of the second distortion based on the degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic by wavelet coherence analysis. Therefore, the second distortion (distortion due to long-term memory effect) included in the output of the amplifier can be compensated for more accurately in real time based on the degree of coincidence derived by wavelet coherence analysis. Therefore, it is possible to provide a highly practical distortion compensation device that can compensate for the long-term memory effect of the amplifier more accurately in real time.

[0019] [8] In any one of paragraphs [1] to [7], The time constant derivation unit may further include a band-limiting filter that passes components of a band equal to or greater than a first predetermined frequency and equal to or less than a second predetermined frequency higher than the first predetermined frequency, from the output signal of the amplifier that is input to the time constant derivation unit. By using the band-limiting filter, the range of frequencies for calculating the time constant can be narrowed, thereby reducing the amount of calculation by the time constant derivation unit. Therefore, it is possible to provide a highly practical distortion compensation device that can reduce the amount of calculation required to derive the time constant.

[0020] [Details of the embodiments of the present disclosure] Hereinafter, embodiments of the present disclosure will be described in detail, but the present disclosure is not limited to these. In this specification and drawings, components having substantially the same functional configurations may be designated by the same reference numerals to avoid redundant description.

[0021] <Embodiment> <Wireless Communication Device 10> 1 is a diagram illustrating an example of the configuration of a wireless communication device 10 including a distortion compensation device 100 according to an embodiment. The wireless communication device 10 includes the distortion compensation device 100, a transceiver integrated circuit (IC) 20, a power amplifier (PA) 30, an antenna 40, and a coupler 41. The PA 30 is an example of an amplifier. The wireless communication device 10 is, for example, a base station or a mobile station for mobile communication.

[0022] <Transceiver IC 20> The transceiver IC 20 has a DAC (Digital to Analog Converter) 50, an upconverter 60, a local signal source 70, a downconverter 80, and an ADC (Analog to Digital Converter) 90. The transceiver IC 20 is configured as an IC (integrated circuit) as an example.

[0023] The DAC 50 performs analog conversion on the signal input from the signal output terminal 102 of the distortion compensation device 100 and outputs it to the upconverter 60. The upconverter 60 upconverts the analog signal input from the DAC 50 using the local signal input from the local signal source 70 and outputs it to the PA 30. The local signal source 70 outputs local signals to the upconverter 60 and the downconverter 80.

[0024] The downconverter 80 downconverts the signal input from the coupler 41 using the local signal input from the local signal source 70 and outputs it to the ADC 90. The ADC 90 digitally converts the analog signal input from the downconverter 80 and outputs it to the input terminal 103 of the distortion compensation device 100. Hereinafter, the signal output by the ADC 90 is referred to as the signal y[n].

[0025] [[ID=**14**]]<PA 30, Antenna 40, and Coupler 41> The input terminal of the PA 30 is connected to the output terminal of the upconverter 60 of the transceiver IC 20, and the output terminal of the PA 30 is connected to the antenna 40. The PA 30 amplifies the signal (transmission signal) input from the upconverter 60 and outputs it to the antenna 40. Also, a coupler 41 is provided in the line between the output terminal of the PA 30 and the antenna 40. The coupler 41 detects the transmission signal output from the PA 30 to the antenna 40 and outputs it to the downconverter 80 of the transceiver IC 20.

[0026] The PA30 is, for example, a compound semiconductor amplifier made of gallium nitride (hereinafter referred to as a "GaN amplifier"). The PA30 may be any amplifier that generates a first distortion and a second distortion in its output. The first distortion includes nonlinear distortion and memory effect distortion, and the second distortion includes memory effect distortion. The first distortion hardly changes over time, but the second distortion is a distortion that changes over time faster than the first distortion and changes relatively more over time. The change over time of the second distortion is related to the time constant τ of the second distortion included in the output signal of the PA30. Hereinafter, the second distortion included in the output signal of the PA30 may be referred to as the second distortion of the PA30 or simply as the second distortion.

[0027] The second distortion is, for example, distortion caused by the trapping effect or buffer trapping effect of the GaN amplifier. The degree of distortion caused by the trapping effect or buffer trapping effect varies depending on temperature and the past input signal of the GaN amplifier, and it is known that variations occur even within a single wafer when manufacturing the PA30. The degree to which the second distortion depends on the past input signal of the GaN amplifier is related to the time constant τ of the second distortion of the PA30, and variations in the PA30 lead to variations in the time constant τ of the second distortion.

[0028] Although the embodiment in which the PA30 is a GaN amplifier will be described here, the PA30 does not have to be a GaN amplifier. For example, the PA30 may be a HEMT (High Electron Mobility Transistor) device made of compound semiconductors such as AlN (aluminum nitride), InN (indium nitride), and their crystal systems, such as AlGaN, InAIN, and InGaN. In other words, the PA30 may be an amplifier that is a HEMT device made of a III-V group compound semiconductor or a crystal system thereof.

[0029] <Distortion compensation device 100> The distortion compensation device 100 includes a signal input terminal 101, a signal output terminal 102, an input terminal 103, a first distortion compensation unit 110, a second distortion compensation unit 120, an adder 130, a DPD (Digital Predistortion) coefficient calculation unit 140, and a time constant derivation unit 150. The distortion compensation device 100 is configured by a wired logic circuit such as a field-programmable gate array (FPGA), for example. Note that the wired logic circuit may be a reconfigurable logic circuit such as an FPGA, or may be a non-reconfigurable logic circuit.

[0030] <Signal input terminal 101> The signal input terminal 101 is a terminal to which a baseband signal output from a baseband processing unit (not shown) is input. Inside the distortion compensation device 100, the signal input terminal 101 is connected to a first distortion compensation unit 110, a second distortion compensation unit 120, and a DPD coefficient calculation unit 140. The baseband signal is input from the signal input terminal 101 to the first distortion compensation unit 110, the second distortion compensation unit 120, and the DPD coefficient calculation unit 140.

[0031] <Signal output terminal 102> The signal output terminal 102 is connected to the output terminal 133 of the adder 130 inside the distortion compensation device 100, and is connected to the DAC50 of the transceiver IC 20 outside the distortion compensation device 100. The signal output terminal 102 outputs the signal that has been compensated for distortion by the first distortion compensation unit 110 and the second distortion compensation unit 120 and added by the adder 130 to the DAC50 of the transceiver IC 20.

[0032] <Input terminal 103> The input terminal 103 is connected to a DPD coefficient calculation unit 140 and a time constant derivation unit 150 inside the distortion compensation device 100, and is connected to an ADC 90 of the transceiver IC 20 outside the distortion compensation device 100.

[0033] <First distortion compensation unit 110> The first distortion compensation unit 110 is configured to perform digital predistortion (DPD). The first distortion compensation unit 110 is connected between the signal output terminal 102 and a first input terminal 131 of the adder 130. The first distortion compensation unit 110 is also connected to the DPD coefficient calculation unit 140, and receives a coefficient h from the DPD coefficient calculation unit 140.

[0034] The first distortion compensation unit 110 has a first distortion compensation characteristic that compensates for a first distortion occurring in the output of the PA 30, and performs distortion compensation using the first distortion compensation characteristic on the baseband signal input from the signal output terminal 102, using a coefficient h input from the DPD coefficient calculation unit 140. The first distortion compensation characteristic is an inverse characteristic of the characteristic of the first distortion included in the output of the PA 30. The coefficient h is a DPD coefficient, and is a parameter that characterizes the first distortion. For example, the strength of third-order distortion, fifth-order distortion, etc. is reflected in this coefficient. The first distortion compensation characteristic will be explained together with the explanation of the distortion characteristic of the time constant derivation unit 150.

[0035] <Second distortion compensation unit 120> The second distortion compensation unit 120 is configured to perform digital pre-distortion (DPD). The second distortion compensation unit 120 is connected between the signal output terminal 102 and the second input terminal 132 of the adder 130. That is, the second distortion compensation unit 120 is connected to the signal input terminal 101 in parallel with the first distortion compensation unit 110.

[0036] The first distortion compensation unit 110 is also connected to a DPD coefficient calculation unit 140 and a time constant derivation unit 150. The first distortion compensation unit 110 receives the coefficient g τ is input, and the time constant τ of the second distortion of the PA 30 is input in real time from the time constant derivation unit 150. τ is an example of a predetermined coefficient, and is a coefficient representing the magnitude of the time constant τ of the second distortion of the PA 30.

[0037] The second distortion compensation unit 120 has a second distortion compensation characteristic that compensates for the second distortion occurring in the output of PA30 for the baseband signal input from the signal input terminal 101, based on the time constant τ of the second distortion of PA30 derived in real time by the time constant derivation unit 150. Because the second distortion compensation characteristic includes a component of the time constant τ of the second distortion of PA30, the second distortion compensation unit 120 sets the time constant τ of the second distortion of PA30 derived in real time by the time constant derivation unit 150 as the second distortion compensation characteristic, and performs distortion compensation on the baseband signal input from the signal output terminal 102 using the second distortion compensation characteristic.

[0038] The second distortion changes faster over time than the first distortion due to the trapping effect, changes with temperature, and differs due to individual differences in PA 30. These changes and differences are caused by changes in the time constant τ of the second distortion of PA 30 and differences in the time constant τ of the second distortion due to individual differences in PA 30. For this reason, the time constant τ of the second distortion of PA 30, which is derived in real time by time constant derivation unit 150, is set as the second distortion compensation characteristic, and the second distortion compensation characteristic is adjusted in real time to the time constant τ of the second distortion of PA 30, thereby compensating for differences due to the trapping effect, temperature, and individual differences of PA 30.

[0039] The second distortion compensation characteristic is an inverse characteristic of the second distortion characteristic included in the output of the PA 30. The second distortion compensation characteristic will be explained together with the explanation of the distortion characteristic of the time constant derivation section 150.

[0040] <Adder 130> The adder 130 has a first input terminal 131, a second input terminal 132, and an output terminal 133. The first input terminal 131 is connected to the output terminal of the first distortion compensation unit 110. The second input terminal 132 is connected to the output terminal of the second distortion compensation unit 120. The output terminal 133 is connected to the DAC 50 of the transceiver IC 20 via the signal output terminal 102. Since the DAC 50 is connected to the input terminal of the PA 30 via the upconverter 60, the output terminal 133 is connected to the input terminal of the PA 30 via the signal output terminal 102, the DAC 50, and the upconverter 60.

[0041] The adder 130 adds the baseband signal with the first distortion compensated by the first distortion compensation unit 110 and the baseband signal with the second distortion compensated by the second distortion compensation unit 120, and outputs the result from the output terminal 133 to the DAC 50. Hereinafter, the signal output from the output terminal 133 of the adder 130 is referred to as signal x[n].

[0042] <DPD coefficient calculation unit 140> The DPD coefficient calculation unit 140 is connected to the signal input terminal 101, the input terminal 103, the first distortion compensation unit 110, and the second distortion compensation unit 120. The DPD coefficient calculation unit 140 calculates the coefficient h and the coefficient g τ based on the baseband signal input from the signal input terminal 101 and the input signal input from the ADC 90 via the input terminal 103. The DPD coefficient calculation unit 140 outputs the coefficient h to the first distortion compensation unit 110, and outputs the coefficient g τ to the second distortion compensation unit 120. The coefficient g τ is an example of a predetermined coefficient. The coefficient h and the coefficient g τ are calculated by the least squares method using the data of the input signal and the data of the output signal corresponding to the data of the input signal. The output signal corresponding to the data of the input signal is the output signal output from the signal output terminal 102 corresponding to the data of the input signal. The coefficient g τ is a parameter characterized by the second distortion for each time constant τ of the second distortion. The larger the absolute value of the coefficient g τ , the stronger the distortion caused by the time constant τ.

[0043] <Time constant derivation unit <150>> The time constant derivation unit 150 has input terminals 151 and 152 and an output terminal 153. The input terminal 151 is connected to the output terminal 133 of the adder 130. The input terminal 152 is connected to the output terminal of the ADC 90 via the input terminal 103. The output terminal 153 is connected to the second distortion compensation unit 120.

[0044] The time constant derivation unit 150 derives the time constant τ of the second distortion of the PA30 in real time based on the signal x[n] input from the output terminal 133 of the adder 130 and the signal y[n] input from the ADC 90, and outputs it to the second distortion compensation unit 120 from the output terminal 153.

[0045] The signal x[n] is a signal input to the PA30 via the DAC 50 and the upconverter 60. Although it undergoes analog conversion and upconversion before being input to the PA30, it may be essentially regarded as the input signal of the PA30. Furthermore, the signal y[n] is a signal input to the time constant derivation unit 150 via the downconverter 80 and the ADC 90, which is the transmission signal of the PA30 detected by the coupler 41. Although it undergoes downconversion and digital conversion before being input to the time constant derivation unit 150, it may be essentially regarded as the output signal of the PA30. Therefore, the time constant derivation unit 150 derives the time constant of the second distortion of the PA30 in real time based on the input signal and output signal of the PA30. The input signal and output signal of the PA30 are synonymous with the input and output of the PA30.

[0046] In order to calculate the time constant τ of the second distortion of PA 30, time constant derivation unit 150 stores data representing a mathematical formula that models the relationship between the input and output of PA 30 in an internal memory. Hereinafter, the mathematical formula that models the relationship between the input and output of PA 30 will be referred to as a mathematical model. The mathematical model is expressed by the following formula (1).

[0047]

number

[0048] The mathematical model has a first term representing a first distortion occurring in the output of PA30 and a second term representing a second distortion occurring in the output of PA30, and is expressed as a mathematical formula in which the first and second terms are added together.

[0049] In the mathematical model, x[n] is the input signal of the time constant derivation unit 150, y[n] is the output signal of the ADC 90, k is an index related to the compensation function of the first distortion, l is an index related to the compensation function of the second distortion, M2, K, and L are constants, n is the discrete time, and h mlk ,gτ mlk denotes a coefficient. Pτ[n] represents the square root of the exponential function type moving average power of the input signal to the time constant derivation unit 150 for the past τ time.

[0050] Furthermore, since the first term in equation (1) represents the first distortion occurring in the output of the PA 30, the first distortion compensation characteristic of the first distortion compensation section 110 may be an inverse function of the first term in equation (1). Note that the inverse function used as the first distortion compensation characteristic of the first distortion compensation section 110 is not limited to a complete inverse function of the first term in equation (1), and may be an inverse function that has been simplified by reducing the coefficients of the inverse function of the first term in equation (1), for example.

[0051] Similarly, since the second term of equation (1) represents the second distortion occurring in the output of the PA 30, the second distortion compensation characteristic of the second distortion compensation section 120 may be an inverse function of the second term of equation (1). Note that the inverse function used as the second distortion compensation characteristic of the second distortion compensation section 120 is not limited to a complete inverse function of the second term of equation (1), and may be an inverse function that has been simplified by reducing the coefficients of the inverse function of the second term of equation (1), for example.

[0052] Here, the second term of equation (1), P τ [n] can be calculated as follows: β τ is expressed by the following equation (2), where fs is the frequency of the signal input to the PA 30.

[0053]

number

[0054] p τ [n] is expressed by the following equation (3).

[0055]

number

[0056] P τ [n] is the p obtained from equation (3) τ Using [n], it can be calculated as shown in equation (4).

[0057]

number

[0058] As can be seen from equation (4), P included in the second term of equation (1) τ [n] is a function of the time constant τ of the second distortion of PA 30. That is, the second distortion included in the output of PA 30 is a function of the time constant τ of the second distortion of PA 30. Furthermore, since the second distortion is a function of the time constant τ of the second distortion of PA 30, the second distortion compensation characteristic of second distortion compensating section 120, which is formed by the inverse function of the second term in equation (1), is also a function of the time constant τ of the second distortion.

[0059] Therefore, by calculating the time constant τ of the second distortion of PA30 in real time and substituting it into the second distortion compensation characteristic of the second distortion compensation section 120, even if the second distortion occurring in the output of PA30 changes over time, due to temperature changes, etc., the second distortion can be compensated for with the second distortion compensation characteristic that matches the changed second distortion.

[0060] In addition, the coefficient g included in the second term of equation (1) τmlk is a coefficient that represents the magnitude of the time constant τ of the second distortion of the PA 30, and takes a larger value as the time constant τ of the second distortion increases. τmlk is the coefficient g τ is the coefficient equivalent to

[0061] For this reason, the time constant derivation section 150 obtains the time constant τ of the second distortion as follows. Fig. 2 is a diagram illustrating an example of the processing performed by the time constant derivation section 150 to obtain the time constant τ of the second distortion. In Fig. 2, the horizontal axis represents the time constant τ of the second distortion, and the vertical axis represents the coefficient S[τ].

[0062] Coefficient g τmlk Since the coefficient S[τ], which represents the magnitude of the time constant τ of the second distortion, increases as the time constant τ of the second distortion increases, is calculated using the following equation (5): The coefficient S[τ] is a function of the time constant τ of the second distortion, and represents the spectrum of the second distortion with respect to the time constant τ of the second distortion.

[0063]

number

[0064] The time constant derivation unit 150 determines the time constant τ at which the coefficient S[τ] determined using the formula (5) takes a maximum value as the time constant τ of the second distortion of the PA 30. -6 The region below 10 is the region where compensation is performed by the first distortion compensation unit 110. -6 The region where the time constant τ is less than 10 -3 The area larger than is the area compensated by the equalizer, so when finding the maximum value, -3 In other words, the time constant derivation unit 150 may exclude a region where the time constant τ is greater than 10 -6 That's it, 10 -3 The time constant τ at which the coefficient S[τ] takes a maximum value within the range of the following region A can be found as the time constant of the second distortion of the PA 30. In the spectrum shown in FIG. 2, the time constant τ at which the coefficient S[τ] takes a maximum value within region A is 9×10 -4 [sec]. There may be multiple time constants τ at which the coefficient S[τ] takes a maximum value.

[0065] <Flowchart> FIG. 3 is a flowchart showing an example of processing executed by the time constant derivation section 150 of the distortion compensation device 100.

[0066] The time constant derivation unit 150 acquires the input signal and output signal of the PA 30 (step S1). The input signal of the PA 30 is the signal x[n], and the output signal is the signal y[n].

[0067] The time constant derivation unit 150 calculates the coefficient S[τ] (step S2).

[0068] The time constant derivation unit 150 derives the time constant τ at which the coefficient S[τ] takes a maximum value within the region A of the spectrum of the coefficient S[τ] (step S3).

[0069] The time constant derivation section 150 outputs the derived time constant τ to the second distortion compensation section 120 (step S4). As a result, the time constant τ is substituted into the equation representing the second distortion compensation characteristic of the second distortion compensation section 120.

[0070] The time constant derivation section 150 repeatedly executes the processes of steps S1 to S4 at a predetermined cycle. As a result, the time constant τ of the second distortion of PA30 derived by the time constant derivation section 150 is substituted in real time into the equation expressing the second distortion compensation characteristic of the second distortion compensation section 120. As a result, in the second distortion compensation section 120, the second distortion compensation characteristic corresponding to the inverse characteristic of the second distortion occurring in the output of PA30 is adjusted in real time based on the time constant τ of the second distortion of PA30, and the second distortion occurring in the output of PA30 can be compensated for in real time.

[0071] As described above, the distortion compensation device 100 can compensate for the first distortion and the second distortion contained in the output of the PA 30 based on the time constant τ of the second distortion of the PA 30 derived in real time. In particular, the distortion compensation device 100 can compensate for the second distortion (distortion due to the long-term memory effect) based on the time constant τ of the second distortion of the PA 30 derived in real time. Therefore, it is possible to provide a highly practical distortion compensation device 100.

[0072] It would require a huge amount of effort to measure the time constant τ of the second distortion of PA 30 in advance, taking into consideration changes over time in the time constant τ of the second distortion of PA 30, changes due to temperature in the time constant τ of the second distortion, or differences in the time constant τ of the second distortion due to individual differences in PA 30. In contrast, the distortion compensation device 100 derives the time constant τ of the second distortion in real time using the time constant derivation section 150 and determines it as the second distortion compensation characteristic of the second distortion compensation section 120, thereby easily achieving distortion compensation that matches the real-time time constant τ.

[0073] <First Modification> 4 is a diagram showing an example of the configuration of a wireless communication device 10 including a distortion compensation device 100M1 according to a first modified example of an embodiment. The distortion compensation device 100M1 according to the first modified example of an embodiment has a configuration in which a bypass line 160 and switches 170A to 170D are added to the distortion compensation device 100 according to the embodiment (see FIG. 1). The switches 170A and 170B are examples of a first switching unit, and the switches 170C and 170D are examples of a second switching unit. The switches 170A to 170D are switched by a time constant derivation unit 150.

[0074] Furthermore, the distortion compensation apparatus 100M1 of the first modified example of the embodiment differs from the distortion compensation apparatus 100 of the embodiment in the way of determining the time constant τ of the second distortion of the PA 30. The differences from the distortion compensation apparatus 100 of the embodiment will be described below.

[0075] The detour line 160 is connected to the signal input terminal 101 via a switch 170A, and is also connected to the signal output terminal 102 via a switch 170B. The detour line 160 is a line that connects the signal input terminal 101 and the signal output terminal 102 without passing through the first distortion compensator 110, the second distortion compensator 120, and the adder 130.

[0076] The switch 170A has terminals 1, 2A, and 2B. Terminal 1 is connected to the signal input terminal 101, terminal 2A is connected to the bypass line 160, and terminal 2B is connected to the input terminal of the first distortion compensation unit 110. The switch 170A can switch the connection destination of terminal 1 between terminal 2A and terminal 2B.

[0077] The switch 170B has terminals 3, 4A, and 4B. Terminal 3 is connected to the signal output terminal 102, terminal 4A is connected to the bypass line 160, and terminal 4B is connected to the output terminal of the first distortion compensation unit 110. The switch 170B can switch the connection destination of terminal 3 between terminal 4A and terminal 4B.

[0078] The switch 170C has terminals 5, 6A, and 6B. Terminal 5 is connected to the input terminal of the second distortion compensation unit 120 and the input terminal of the DPD coefficient calculation unit 140. Terminal 6A is open, and terminal 6B is connected to a line between terminal 2B of the switch 170A and the input terminal of the first distortion compensation unit 110. The switch 170C can switch the connection destination of terminal 5 between terminal 6A and terminal 6B.

[0079] The switch 170D has terminals 7, 8A, and 8B. The terminal 7 is connected to the output terminal of the second distortion compensation unit 120. The terminal 8A is open, and the terminal 8B is connected to the second input terminal 132 of the adder 130. The switch 170D can switch the connection destination of the terminal 7 between the terminal 8A and the terminal 8B.

[0080] Next, three connection states of the distortion compensation device 100M1 will be described using Fig. 5 and Fig. 6 in addition to Fig. 4. Fig. 5 and Fig. 6 are diagrams showing an example of the configuration of a wireless communication device 10 including a distortion compensation device 100M1 of a first modified example of the embodiment, and show connection states different from the connection state of Fig. 4.

[0081] <Connection status in Figure 4> 4 shows a state in which terminal 1 of switch 170A is connected to terminal 2A, terminal 3 of switch 170B is connected to terminal 4A, terminal 5 of switch 170C is connected to terminal 6A, and terminal 7 of switch 170D is connected to terminal 8A. The signal output from output terminal 133 of adder 130 in the connection state of FIG. 4 is referred to as signal x1[n]. Signal x1[n] in the connection state of FIG. 4 is an example of a first input to an input terminal of PA30. In the connection state of FIG. 4, time constant derivation section 150 does not output the time constant τ of the second distortion to second distortion compensation section 120, and therefore in FIG. 4, an x ​​is indicated on the path that sends the time constant τ of the second distortion from time constant derivation section 150 to second distortion compensation section 120.

[0082] <Connection status in Figure 5> 5 shows a state in which terminal 1 of switch 170A is connected to terminal 2B, terminal 3 of switch 170B is connected to terminal 4B, terminal 5 of switch 170C is connected to terminal 6A, and terminal 7 of switch 170D is connected to terminal 8A. The signal output from output terminal 133 of adder 130 in the connection state of FIG. 5 is referred to as signal x2[n]. Signal x2[n] in the connection state of FIG. 5 is an example of a second input to the input terminal of PA30. In the connection state of FIG. 5, time constant derivation section 150 does not output the time constant τ of the second distortion to second distortion compensation section 120, and therefore in FIG. 5, an x ​​is indicated on the path that sends the time constant τ of the second distortion from time constant derivation section 150 to second distortion compensation section 120.

[0083] <Connection status in Figure 6> 6 shows a state in which terminal 1 of switch 170A is connected to terminal 2B, terminal 3 of switch 170B is connected to terminal 4B, terminal 5 of switch 170C is connected to terminal 6B, and terminal 7 of switch 170D is connected to terminal 8B. The circuit configuration in the connection state shown in FIG. 6 is the same as the circuit configuration of distortion compensation device 100 shown in FIG. 1. The signal output from output terminal 133 of adder 130 in the connection state of FIG. 6 is signal x[n]. In the connection state of FIG. 6, time constant derivation section 150 outputs the time constant τ of the second distortion to second distortion compensation section 120, and therefore does not show an x ​​mark.

[0084] <Flowchart> FIG. 7 is a flowchart showing an example of processing executed by the time constant derivation unit 150 of the distortion compensation device 100M1 of the first modified example of the embodiment.

[0085] 4 and acquires a first gain time characteristic |y[n]| / |x1[n]| of the absolute value of signal y[n] with respect to the absolute value of signal x1[n] (step S11). The first gain time characteristic |y[n]| / |x1[n]| is a characteristic that represents a change over time in the gain of the output with respect to the input of PA30.

[0086] The time constant derivation unit 150 calculates the first time-frequency characteristic by performing a wavelet transform on the first gain-time characteristic |y[n]| / |x1[n]| (step S12). The first time-frequency characteristic is the time-frequency characteristic of the first gain-time characteristic |y[n]| / |x1[n]|.

[0087] 5, acquires the signal x2[n], and acquires a second gain-time characteristic |x2[n]| / |x1[n]| of the absolute value of the signal x2[n] with respect to the absolute value of the signal x1[n] acquired in step S11 (step S13). The second gain-time characteristic |x2[n]| / |x1[n]| is the gain-time characteristic of the first distortion compensation unit 110.

[0088] The time constant derivation unit 150 calculates the second time-frequency characteristic by performing a wavelet transform on the second gain time characteristic |x2[n]| / |x1[n]| (step S14). The second time-frequency characteristic is the time-frequency characteristic of the second gain time characteristic |x2[n]| / |x1[n]|.

[0089] The time constant derivation unit 150 calculates the degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic by wavelet coherence analysis (step S15), which will be described in detail later with reference to FIG.

[0090] Based on the degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic, the time constant derivation unit 150 derives the time constant τ of the second distortion of PA30 (step S16). Details of this will be described later using FIG. 8.

[0091] The time constant derivation unit 150 sets the switches 170A to 170D to the connection states shown in FIG. 6, and outputs the time constant τ of the second distortion derived in step S16 to the second distortion compensation unit 120 (step S17). As a result, the time constant τ of the second distortion is substituted into the expression representing the second distortion compensation characteristic of the second distortion compensation unit 120.

[0092] <Derivation of the time constant τ of the second distortion of PA30> The time constant τ of the second distortion of PA30 can be derived by wavelet coherence analysis. Let the frequency obtained by wavelet coherence analysis be fw, then the time constant τ of the second distortion can be expressed by the following equation (6).

[0093]

Equation

[0094] FIG. 8 is a diagram showing an example of the calculation result of the degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic. In FIG. 8, the horizontal axis is time [msec], and the vertical axis is frequency [kHz]. The degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic is higher as the color is lighter and lower as the color is darker.

[0095] In FIG. 8, as an example, the frequency fw obtained by wavelet coherence analysis is derived as a value of 67 kHz, and when converted to the time constant τ of the second distortion, it is 14.9 μsec.

[0096] The time constant derivation unit 150 repeatedly executes the processes of steps S11 to S17 at a predetermined cycle. As a result, the degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic is calculated in step S15, and the time constant τ of the second distortion of PA30 is derived in step S16. <000040​The time constant τ of the second distortion of PA30 derived by the time constant derivation section 150 is substituted in real time into the equation expressing the second distortion compensation characteristic of the second distortion compensation section 120. As a result, in the second distortion compensation section 120, the second distortion compensation characteristic corresponding to the inverse characteristic of the second distortion occurring in the output of PA30 is adjusted in real time based on the time constant τ of the second distortion of PA30, and the second distortion occurring in the output of PA30 can be compensated for in real time.

[0098] As described above, the distortion compensation device 100M1 can compensate for the first distortion and the second distortion contained in the output of the PA30 based on the time constant τ of the second distortion of the PA30 derived in real time. In particular, the second distortion (distortion due to the long-term memory effect) can be compensated for based on the time constant τ of the second distortion of the PA30 derived in real time. Therefore, it is possible to provide a highly practical distortion compensation device 100M1.

[0099] <Second Modification> 9 is a diagram showing an example of the configuration of a wireless communication device 10 including a wireless communication device 10 including a distortion compensation device 100M2 according to a second modified example of the embodiment. The distortion compensation device 100M2 according to the second modified example of the embodiment has a configuration in which a band-limiting filter 180 is added to the distortion compensation device 100 according to the embodiment (see FIG. 1).

[0100] Band-limiting filter 180 is provided on a line between input terminal 103 and an input terminal to which signal y[n] is input from time constant derivation unit 150. Band-limiting filter 180 has a passband that passes components of signal y[n] in a band that is equal to or greater than a first predetermined frequency and equal to or less than a second predetermined frequency that is higher than the first predetermined frequency.

[0101] Therefore, when the time constant derivation section 150 of the distortion compensation device 100M2 calculates the spectrum of the coefficient S[τ], it can narrow the range of the time constant τ of the second distortion for calculating the spectrum of the coefficient S[τ], similar to the time constant derivation section 150 of the distortion compensation device 100 shown in Fig. 1. This allows the amount of calculation by the time constant derivation section 150 to be reduced.

[0102] The band-limiting filter 180 can also be applied to the distortion compensation device 100M1 of the first modified example of the embodiment shown in Figures 4 to 6. When the band-limiting filter 180 is applied to the distortion compensation device 100M1, it is possible to reduce the amount of calculation when calculating the degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic as shown in Figure 8. This will be explained using Figure 10.

[0103] FIG. 10 shows an example of a transmission test signal conforming to the E-TM2a standard defined in the LTE specifications. This signal is composed of 14 sections, each of which includes frequencies ranging from low frequencies below 180 kHz to high frequencies up to 20 MHz. In this signal, the area most affected by distortion is the interval between sections 2 and 4. The signal frequency in this section is 180 kHz, and the interval is approximately 210 μs. The memory time constant frequencies that affect this section are 20 kHz to 2000 kHz, and it is not necessary to calculate the degree of match for other frequencies (e.g., 2 MHz to 20 MHz). The time constant derivation unit 150 only needs to calculate the degree of match between the first time-frequency characteristic and the second time-frequency characteristic for the band between 20 kHz and 2000 kHz. This reduces the amount of calculation required by the time constant derivation unit 150.

[0104] The above describes a distortion compensation device according to an exemplary embodiment of the present invention, but the present invention is not limited to the specifically disclosed embodiment, and various modifications and variations are possible without departing from the scope of the claims. [Explanation of symbols]

[0105] 10. Radio Communication Device 20 Transceiver IC 30 PA 40 Antenna 41 Coupler 50 DAC 60 Upconverter 70 Local signal source 80 Down Converter 90 ADC 100, 100M1, 100M2 distortion compensator 101 Signal input terminal 102 Signal output terminal 103 Input terminal 110 First distortion compensation section 120 Second distortion compensation section 130 Adder 131 First input terminal 132 Second input terminal 133 Output terminal 140 DPD coefficient calculation unit 150 Time constant derivation section 151, 152 input terminals 153 Output terminal 160 Detour Line 170A~170D Switch 180 Bandwidth Limiting Filter

Claims

1. A signal input terminal; a signal output terminal to which an amplifier is connected; a first distortion compensator provided between the signal input terminal and the signal output terminal, which compensates for a first distortion occurring in an output of the amplifier with respect to an input signal input from the signal input terminal; a time constant derivation unit that derives in real time a time constant of a second distortion that occurs in the output of the amplifier, the second distortion changing more quickly over time than the first distortion; a second distortion compensator connected to the signal input terminal in parallel with the first distortion compensator, and configured to compensate for the second distortion in the input signal input from the signal input terminal based on the time constant of the second distortion derived by the time constant derivation unit; an adder having a first input terminal connected to the output terminal of the first distortion compensator, a second input terminal connected to the output terminal of the second distortion compensator, and an output terminal connected to the signal output terminal, which adds the input signal in which the first distortion has been compensated for by the first distortion compensator to the input signal in which the second distortion has been compensated for by the second distortion compensator, and outputs the result to the signal output terminal; A distortion compensation device comprising:

2. The distortion compensation device according to claim 1 , wherein the time constant deriving section derives the time constant of the second distortion in real time based on the input and the output of the amplifier.

3. 3. The distortion compensation device according to claim 2, wherein the time constant derivation unit derives the time constant of the second distortion in real time based on the input and the output of the amplifier using models of the first distortion and the second distortion occurring in the output of the amplifier.

4. the model is expressed by a mathematical formula that adds a term representing the first distortion and a term representing the second distortion, the time constant deriving unit derives, as the time constant of the second distortion, a time constant at which a square of a predetermined coefficient included in a term representing the second distortion gives a maximum value; 4. The distortion compensation device according to claim 3, wherein the predetermined coefficient is included in a term representing the second distortion and is a coefficient representing the magnitude of a time constant of the second distortion.

5. a bypass line connecting the signal input terminal and the signal output terminal without passing through the first distortion compensator, the second distortion compensator, and the adder; a first switching unit that connects either a first path passing through the first distortion compensation unit and the adder or the bypass line between the signal input terminal and the signal output terminal; a second switching unit that switches between connection and disconnection of a second path that connects the second distortion compensation unit in parallel to the first path between the signal input terminal and the adder; further comprising When deriving the time constant of the second distortion in real time, the time constant derivation unit deriving a first time-frequency characteristic of an input to an input terminal of the amplifier in a state in which the first switching unit connects the bypass line between the signal input terminal and the amplifier and the second switching unit disconnects the second path from the first path; deriving a second time-frequency characteristic of an input to an input terminal of the amplifier in a state in which the first switching unit connects the first path between the signal input terminal and the amplifier and the second switching unit disconnects the second path from the first path; The distortion compensation device according to claim 2 , wherein the time constant of the second distortion is derived based on a degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic.

6. The distortion compensation device according to claim 5 , wherein the time constant deriving unit derives the first time-frequency characteristic and the second time-frequency characteristic by wavelet transformation.

7. The distortion compensation device according to claim 5 , wherein the time constant deriving unit derives the time constant of the second distortion based on a degree of coincidence between the first time-frequency characteristic and the second time-frequency characteristic by wavelet coherence analysis.

8. 8. The distortion compensation device according to claim 1, further comprising a band-limiting filter that passes components of a band equal to or greater than a first predetermined frequency and equal to or less than a second predetermined frequency higher than the first predetermined frequency, from the output of the amplifier that is input to the time constant derivation unit.

Images