Information processing device, communication device, and information processing method

Abstract

The present invention provides an information processing device, a communication device, and an information processing method that suppress leakage power in adjacent channels. [Solution] The distortion compensation coefficient calculation unit 12 determines a first distortion compensation coefficient based on the power or amplitude of the first signal and the power or amplitude of the second signal. The multiplier 131 generates a third signal based on the first signal and the first distortion compensation coefficient determined by the distortion compensation coefficient calculation unit 12. The multiplier 132 generates a fourth signal based on the second signal and the first distortion compensation coefficient determined by the distortion compensation coefficient calculation unit 12. The adder 15 generates a fifth signal based on the third signal generated by the multiplier 131 and the fourth signal generated by the multiplier 132.

Description

[Technical Field]

[0001] The present invention relates to an information processing device, a communication device, and an information processing method. [Background technology]

[0002] Wireless communication equipment such as base stations are equipped with power amplifiers that amplify the power of the transmitted signal. Generally, wireless communication equipment operates the power amplifier near the saturation region to improve its power efficiency. However, operating the power amplifier near the saturation region increases the nonlinear distortion generated in the power amplifier. Therefore, in order to suppress this nonlinear distortion and meet standards such as ACLR (Adjacent Channel Leakage Ratio) and SEM (Spectrum Emission Mask), wireless transmission equipment is equipped with a distortion compensation device that compensates for the nonlinear distortion.

[0003] Conventionally, in the field of wireless mobile communications, techniques have been proposed to compensate for the nonlinear distortion of power amplifiers used in base stations and the like. One of the distortion compensation methods used in distortion compensation devices is the pre-distortion method. In a pre-distortion distortion compensation device, a distortion compensation coefficient having the inverse characteristics of the nonlinear distortion of the power amplifier is multiplied in advance by the transmission signal input to the power amplifier. This increases the linearity of the output of the power amplifier and suppresses the nonlinear distortion of the output of the power amplifier. Known pre-distortion methods include the LUT (Look-Up Table) type pre-distortion method and the series type pre-distortion method.

[0004] In recent years, communication traffic has been steadily increasing, and as a countermeasure, the adoption of wireless communication equipment that supports broadband and multiband transmission is progressing. Conventional wireless communication equipment that supports single-band transmission requires separate equipment to be installed for each band. In contrast, wireless communication equipment that supports multiband transmission can handle multiple bands with a single device, which allows for space saving and more efficient installation.

[0005] In wireless communication devices that support multiband transmission, there are two configurations for power amplifiers that amplify the power of multiband signals: an individual amplification configuration in which each band is amplified individually by separate power amplifiers, and a common amplification configuration in which the multiband signal is amplified commonly by a single power amplifier. In the individual amplification configuration, a distortion compensation section is used to generate a pre-distortion signal corresponding to each individual power amplifier for each band, according to the number of bands. In the common amplification configuration, the wireless communication device can use a distortion compensation section that generates a pre-distortion signal by batch processing for the multiband signal. In this case, it is common to ensure that the sampling rate of the distortion compensation section is such that the digital processing bandwidth is 3 to 5 times the IBW (Instantaneous Bandwidth). On the other hand, by commonly amplifying the multiband signal with a single power amplifier, the number of devices can be reduced compared to the individual amplification configuration, and the mounting area can be reduced, allowing for miniaturization of the device.

[0006] Furthermore, as a distortion compensation technique for multiband transmission, a technique has been proposed that compensates for distortion by considering the influence of the power of other bands on a specific band. Another proposed technique involves estimating the estimated envelope of the composite signal input to the amplifier from multiple input signals, and then determining the distortion compensation coefficient based on the estimated envelope. [Prior art documents] [Patent Documents]

[0007] [Patent Document 1] Special Publication No. 2017-503381 [Patent Document 2] International Publication No. 2015 / 045709 [Patent Document 3] Japanese Patent Publication No. 2017-208753 [Overview of the project] [Problems that the invention aims to solve]

[0008] In wireless communication devices with a common amplification configuration, if the IBW is generally 100 MHz or less, the nonlinear characteristics of the PA (Power Amplifier) ​​are relatively similar across each band. Therefore, even if distortion compensation processing is performed to generate a pre-distortion signal by batch processing for multiband signals, a decrease in distortion compensation performance is not a problem. However, if distortion compensation processing is performed to generate a pre-distortion signal by batch processing for wideband multiband signals, for example, where the IBW is generally several hundred MHz to several GHz, average compensation is performed even if the nonlinear characteristics of the PA differ across each band. Therefore, the difference from the average remains as an error that cannot be compensated for, which may lead to a decrease in distortion compensation performance. In other words, the Adjacent Channel Leakage Ratio (ACLR) may worsen.

[0009] Furthermore, even if the nonlinear characteristics of the PA differ in each band, a configuration can be considered in which a distortion compensation section that generates a pre-distortion signal for each band is used, equal to the number of bands, to ensure optimal compensation for the distortion of each band. However, while it is possible to compensate for the nonlinear distortion generated at the power of a specific band, it may be difficult to suppress leakage power in adjacent channels, for example. In other words, it may be difficult to compensate for the nonlinear distortion of a specific band that is affected by other bands.

[0010] Furthermore, in techniques that perform distortion compensation for a specific band by considering the influence of the power of other bands, it is difficult to suppress the influence of the combined power of the specific band and other bands, which may worsen the adjacent channel leakage power ratio. Similarly, in techniques that determine the distortion compensation coefficient from the estimated envelope of the combined signal, for example, when processing at a low sampling rate, it is difficult to suppress the influence of the combined power, which may worsen the adjacent channel leakage power ratio.

[0011] The disclosed technology was made in view of the above and aims to provide an information processing device, a communication device, and an information processing method that suppress leakage power of adjacent channels. [Means for solving the problem]

[0012] In one embodiment of the information processing apparatus, communication apparatus, and information processing method disclosed herein, a first determination unit determines a first distortion compensation coefficient based on the power or amplitude of a first signal and the power or amplitude of a second signal. A first signal generation unit generates a third signal based on the first signal and the first distortion compensation coefficient. A second signal generation unit generates a fourth signal based on the second signal and the first distortion compensation coefficient. A third signal generation unit generates a fifth signal based on the third signal and the fourth signal. [Effects of the Invention]

[0013] In one aspect, the present invention can suppress leakage power in adjacent channels. [Brief explanation of the drawing]

[0014] [Figure 1] Figure 1 is a schematic diagram of the wireless communication system. [Figure 2] Figure 2 is a block diagram of the wireless device according to Example 1. [Figure 3] Figure 3 is a flowchart of the wireless signal output processing by the wireless device according to Example 1. [Figure 4] Figure 4 is a hardware configuration diagram of the radio. [Figure 5] Figure 5 is a block diagram of the wireless device according to Example 2. [Figure 6] Figure 6 is a block diagram of the wireless device according to Example 3. [Figure 7] Figure 7 is a block diagram of the wireless device according to Example 4. [Figure 8] Figure 8 is a block diagram of the wireless device relating to Modification 4-2. [Figure 9] Figure 9 is a block diagram of the wireless device according to modified example 4-3. [Figure 10] Figure 10 is a block diagram of the wireless device according to Example 5. [Figure 11] Figure 11 is a block diagram of the wireless device according to Example 6. [Figure 12] Figure 12 is a block diagram of the wireless device according to modified example 6-1. [Figure 13] Figure 13 is a block diagram of a wireless device relating to Modification 6-2. [Figure 14] Figure 14 is a block diagram of the wireless device according to Example 7. [Figure 15] Figure 15 shows the relationship between bands and carriers. [Figure 16] Figure 16 shows the suppression of unwanted out-of-band signals in the transmitted signal. [Modes for carrying out the invention]

[0015] The following describes in detail, with reference to the drawings, embodiments of the information processing apparatus, communication apparatus, and information processing method disclosed in this application. However, the following embodiments do not limit the information processing apparatus, communication apparatus, and information processing method disclosed in this application. [Examples]

[0016] Figure 1 is a schematic diagram of a wireless communication system. The wireless communication system 100 includes a radio 1, a transmission device 2, and a UE (User Equipment) 3. The transmission device 2 generates a baseband signal based on the transmission data. For example, the transmission device 2 generates two baseband signals for two-band multiband transmission. The transmission device 2 then outputs the generated baseband signals to the radio 1. The transmission device 2 is also called a CU / DU (Central Unit / Distributed Unit).

[0017] Radio 1 receives baseband signals as input from transmission device 2. Radio 1 performs an IFFT (Inverse Fast Fourier Transform) on each received baseband signal to convert it from the frequency domain to the time domain. Radio 1 also performs oversampling to increase the sampling rate of each baseband signal. Radio 1 also performs peak suppression processing on each baseband signal.

[0018] Furthermore, radio 1 performs distortion compensation on each baseband signal. Next, radio 1 performs frequency shifting on each of the distortion-compensated baseband signals to match each band of the multiband transmission. Then, radio 1 combines the frequency-shifted transmission signals, converts the combined transmission signal from a digital signal to an analog signal, and further converts it into a carrier wave through modulation, which is then radiated into space as radio waves from the antenna. Radio 1 is also called RU (Radio Unit).

[0019] UE3 receives signals transmitted from radio 1 via multiband transmission. Then, UE3 demodulates the received signals, separates them by band using an LPF (Low Pass Filter), and processes each signal individually.

[0020] Figure 2 is a block diagram of the radio according to Embodiment 1. Next, the details of the distortion compensation processing by the radio 1 will be explained with reference to Figure 2. In Figure 2, for explanatory purposes, the functions used for distortion compensation processing and signal transmission in the radio 1 are shown, and other functions have been omitted. As shown in Figure 2, the radio 1 has an address generation unit 11, a distortion compensation coefficient calculation unit 12, multipliers 131 and 132, frequency shift units 141 and 142, and an adder 15. The radio 1 also has a DAC (Digital Analog Converter) 16, an upconverter 17, a local oscillator 18, and a PA (Power Amplifier) ​​19.

[0021] Furthermore, this embodiment will be described in the case where the radio 1 performs multiband transmission using two bands, a first band and a second band. This radio 1 is an example of an "information processing device" and a "communication device". The radio 1 has a signal path P1 that receives the input of the baseband signal of the first transmission signal transmitted on the first band, and a signal path P2 that receives the input of the baseband signal of the second transmission signal transmitted on the second band.

[0022] The address generation unit 11 generates addresses for calculating distortion compensation coefficients. The address generation unit 11 includes power calculation units 111 and 112, and an adder 121.

[0023] The power calculation unit 111 receives the input of the baseband first transmission signal input to the signal path P1. Then, for each sample, the power calculation unit 111 calculates the power of the first transmission signal by squaring the absolute value of the complex number of the first transmission signal. After that, the power calculation unit 111 outputs the calculated power of the first transmission signal as an address to the LUTs (Look Up Tables) 211 and 223 of the distortion compensation coefficient calculation unit 12. Hereafter, the address represented by the power of the first transmission signal will be referred to as the "address of the first transmission signal". The power calculation unit 111 also outputs the calculated power of the first transmission signal to the adder 121.

[0024] The power calculation unit 112 receives the input of the second baseband transmission signal input to the signal path P2. Then, for each sample, the power calculation unit 112 calculates the power of the second transmission signal by squaring the absolute value of the complex number of the second transmission signal. After that, the power calculation unit 112 outputs the calculated power of the second transmission signal as an address to LUTs 213 and 221 of the distortion compensation coefficient calculation unit 12. Hereafter, the address represented by the power of the second transmission signal will be referred to as the "address of the second transmission signal". The power calculation unit 112 also outputs the calculated power of the second transmission signal to the adder 121.

[0025] In this embodiment, an example of generating an address based on power has been described, but the address generation unit 11 may also generate an address using the amplitudes of the first and second transmission signals. Alternatively, the address generation unit 11 may generate an address using dB obtained by converting power to a logarithmic representation.

[0026] The adder 121 receives the power input of the first transmission signal from the power calculation unit 111. The adder 121 also receives the power input of the second transmission signal from the power calculation unit 112. The adder 121 then adds the power of the first transmission signal and the power of the second transmission signal to calculate the combined power of the first and second transmission signals. Afterward, the adder 121 outputs the added combined power as an address to the LUTs 212 and 222 of the distortion compensation coefficient calculation unit 12. Hereafter, the address represented by the combined power will be referred to as the "combined power address."

[0027] The distortion compensation coefficient calculation unit 12 has three LUTs 211 to 213 for calculating the distortion compensation coefficient for the first transmitted signal. The distortion compensation coefficient calculation unit 12 also has three LUTs 221 to 223 for calculating the distortion compensation coefficient for the second transmitted signal. Furthermore, the distortion compensation coefficient calculation unit 12 has adders 127 and 128.

[0028] LUT211 is a table in which distortion compensation coefficients for distortion compensation of the first transmitted signal for the power of the first transmitted signal itself are stored, associated with the address of the first transmitted signal. LUT212 is a table in which distortion compensation coefficients for distortion compensation of the first transmitted signal for the combined power are stored, associated with the address of the combined power. LUT213 is a table in which distortion compensation coefficients for distortion compensation of the first transmitted signal for the power of the second transmitted signal are stored, associated with the address of the second transmitted signal.

[0029] LUT221 is a table in which distortion compensation coefficients for distortion compensation of the second transmitted signal for the power of the second transmitted signal itself are stored, associated with the address of the second transmitted signal. LUT222 is a table in which distortion compensation coefficients for distortion compensation of the second transmitted signal for the combined power are stored, associated with the address of the combined power. LUT223 is a table in which distortion compensation coefficients for distortion compensation of the second transmitted signal for the power of the first transmitted signal are stored, associated with the address of the first transmitted signal.

[0030] LUT211 reads, for example, the distortion compensation coefficient corresponding to the address of the first transmission signal input from the power calculation unit 111 and outputs it to the adder 127 as the distortion compensation coefficient for distortion compensation of the first transmission signal for the power of the first transmission signal itself. LUT212 reads, for example, the distortion compensation coefficient corresponding to the address of the combined power input from the adder 121 and outputs it to the adder 127 as the distortion compensation coefficient for distortion compensation of the combined power for the first transmission signal. LUT213 reads, for example, the distortion compensation coefficient corresponding to the address of the second transmission signal input from the power calculation unit 112 and outputs it to the adder 127 as the distortion compensation coefficient for distortion compensation of the power of the second transmission signal for distortion compensation of the first transmission signal.

[0031] LUT221 reads the distortion compensation coefficient corresponding to the address of the second transmission signal input from the power calculation unit 112 and outputs it to the adder 128 as a distortion compensation coefficient for distortion compensation of the second transmission signal for the power of the second transmission signal itself. LUT222 reads the distortion compensation coefficient corresponding to the address of the combined power input from the adder 121 and outputs it to the adder 128 as a distortion compensation coefficient for distortion compensation of the combined power for the second transmission signal. LUT223 reads the distortion compensation coefficient corresponding to the address of the first transmission signal input from the power calculation unit 111 and outputs it to the adder 128 as a distortion compensation coefficient for distortion compensation of the power of the first transmission signal for distortion compensation of the second transmission signal.

[0032] Here, the first transmission signal is an example of the "first signal," and the second transmission signal is an example of the "second signal." LUTs 212 and 222 are examples of the "first determination unit." Furthermore, the distortion compensation coefficient for distortion compensation for the first transmission signal with respect to the combined power and the distortion compensation coefficient for distortion compensation for the second transmission signal with respect to the combined power are examples of the "first distortion compensation coefficient." That is, the first determination unit determines the first distortion compensation coefficient based on the power or amplitude of the first signal and the power or amplitude of the second signal. The first determination unit also determines the first distortion compensation coefficient using a lookup table.

[0033] The adder 127 receives the respective distortion compensation coefficients for the first transmitted signal from each of the LUTs 211 to 213. The adder 127 then adds the three input distortion compensation coefficients to calculate a distortion compensation coefficient for the first transmitted signal that collectively compensates for the nonlinear distortion caused by the power of the first transmitted signal itself, the power of the combined signal, and the power of the second transmitted signal. After that, the adder 127 outputs the calculated distortion compensation coefficient for the first transmitted signal to the multiplier 131.

[0034] The adder 128 receives the respective distortion compensation coefficients for the second transmission signal from each of the LUTs 221 to 223. The adder 128 then adds the three input distortion compensation coefficients to calculate a distortion compensation coefficient for the second transmission signal that collectively compensates for the nonlinear distortion caused by the power of the second transmission signal itself, the power of the combined signal, and the power of the first transmission signal. After that, the adder 128 outputs the calculated distortion compensation coefficient for the second transmission signal to the multiplier 132.

[0035] Here, the combination of LUT211 or 213 and adder 127, and the combination of LUT221 or 223 and adder 128 are examples of the "second determination unit". Also, the distortion compensation coefficient for distortion compensation for the first transmitted signal with respect to the power of the first transmitted signal itself, or the distortion compensation coefficient for distortion compensation for the first transmitted signal with respect to the power of the second transmitted signal, are examples of the "second distortion compensation coefficient". Also, the distortion compensation coefficient for distortion compensation for the second transmitted signal with respect to the power of the second transmitted signal itself, or the distortion compensation coefficient for distortion compensation for the second transmitted signal with respect to the power of the first transmitted signal, are examples of the "second distortion compensation coefficient". Furthermore, the "distortion compensation coefficient for the first transmitted signal" generated by adder 127 and the "distortion compensation coefficient for the second transmitted signal" generated by adder 128 are examples of the "third distortion compensation coefficient". In other words, the second determination unit determines a second distortion compensation coefficient based on the first signal or the second signal, and determines a third distortion compensation coefficient based on the first and second distortion compensation coefficients.

[0036] Furthermore, the combination of LUTs 211 and 213 and adder 127, and the combination of LUTs 221 and 223 and adder 128 are examples of the "third determination unit". Also, the distortion compensation coefficient for distortion compensation for the first transmission signal with respect to the power of the first transmission signal itself, and the distortion compensation coefficient for distortion compensation for the second transmission signal with respect to the power of the first transmission signal are examples of the "fourth distortion compensation coefficient". Furthermore, the distortion compensation coefficient for distortion compensation for the first transmission signal with respect to the power of the second transmission signal, and the distortion compensation coefficient for distortion compensation for the second transmission signal with respect to the power of the second transmission signal itself are examples of the "fifth distortion compensation coefficient". And the "distortion compensation coefficient for the first transmission signal" generated by adder 127 and the "distortion compensation coefficient for the second transmission signal" generated by adder 128 are examples of the "sixth distortion compensation coefficient". Specifically, the third determination unit determines a fourth distortion compensation coefficient based on the first signal, a fifth distortion compensation coefficient based on the second signal, and a sixth distortion compensation coefficient based on the first, fourth, and fifth distortion compensation coefficients.

[0037] The multiplier 131 receives the input of the first transmission signal input to the signal path P1. The multiplier 131 also receives the input of the distortion compensation coefficient of the first transmission signal from the adder 127. The multiplier 131 then multiplies the first transmission signal by the distortion compensation coefficient of the first transmission signal to perform distortion compensation on the first transmission signal. In this way, the multiplier 131 performs distortion compensation on the first transmission signal with respect to the power of the first transmission signal itself, the power of the combined signal, and the power of the second transmission signal. After that, the multiplier 131 outputs the distortion-compensated first transmission signal as a pre-distortion signal for the first band to the frequency shift unit 141.

[0038] This multiplier 131 is an example of a "first signal generation unit." The pre-distortion signal of the first band calculated by the multiplier 131 is an example of a "third signal." That is, the first signal generation unit generates a third signal based on the first signal and the first distortion compensation coefficient. The first signal generation unit also generates a third signal based on the first signal and the third distortion compensation coefficient. The first signal generation unit also generates a third signal based on the first signal and the sixth distortion compensation coefficient.

[0039] The multiplier 132 receives the input of the second transmission signal, which is input to the signal path P2. The multiplier 132 also receives the input of the distortion compensation coefficient of the second transmission signal from the adder 128. The multiplier 132 then multiplies the second transmission signal by the distortion compensation coefficient of the second transmission signal to perform distortion compensation on the second transmission signal. In this way, the multiplier 132 performs distortion compensation on the second transmission signal with respect to the power of the second transmission signal itself, the power of the combined signal, and the power of the first transmission signal. After that, the multiplier 132 outputs the distortion-compensated second transmission signal as a pre-distortion signal for the second band to the frequency shift unit 142.

[0040] This multiplier 132 is an example of a "second signal generation unit." The pre-distortion signal of the second band calculated by the multiplier 132 is an example of a "fourth signal." That is, the second signal generation unit generates the fourth signal based on the second signal and the first distortion compensation coefficient. The second signal generation unit also generates the fourth signal based on the second signal and the third distortion compensation coefficient. The second signal generation unit also generates the fourth signal based on the second signal and the sixth distortion compensation coefficient.

[0041] The frequency shift unit 141 receives the input of the first transmission signal, which is a pre-distortion signal for the first band. The frequency shift unit 141 then shifts the frequency of the first transmission signal to the frequency band allocated to the first band. After that, the frequency shift unit 141 outputs the frequency-shifted first transmission signal to the adder 15.

[0042] The frequency shift unit 142 receives the input of the second transmission signal, which is a pre-distortion signal for the second band. The frequency shift unit 142 then shifts the frequency of the second transmission signal to the frequency band allocated to the second band. After that, the frequency shift unit 142 outputs the frequency-shifted second transmission signal to the adder 15.

[0043] The adder 15 receives input from the frequency shifting unit 141, which is the first transmission signal, a pre-distortion signal whose frequency has been shifted to the frequency band of the first band. The adder 15 also receives input from the frequency shifting unit 142, which is the second transmission signal, a pre-distortion signal whose frequency has been shifted to the frequency band of the second band. The adder 15 then adds the first transmission signal and the second transmission signal to generate a combined signal of the first and second transmission signals, which are pre-distortion signals. After that, the adder 15 outputs the generated combined pre-distortion signal to the DAC 16.

[0044] This adder 15 is an example of a "third signal generation unit," and the combined signal of the first transmission signal and the second transmission signal, which are pre-distortion signals generated by the adder 15, is an example of a "fifth signal." In other words, the adder 15, which is the third signal generation unit, generates the fifth signal based on the third signal and the fourth signal.

[0045] DAC16 receives the combined signal, which is a pre-distortion signal, as input from adder15. Then, DAC16 converts the combined signal from a digital signal to an analog signal. After that, DAC16 outputs the combined signal, which is a pre-distortion signal converted to an analog signal, to upconverter17.

[0046] The upconverter 17 receives the combined signal, which is a pre-distorted signal converted to an analog signal, as input from the DAC 16. The upconverter 17 then multiplies the combined signal and the local signal input from the local oscillator 18 to convert it to a radio frequency. After that, the upconverter 17 outputs the combined signal, which is a pre-distorted signal converted to a radio frequency, to the PA 19.

[0047] PA19 receives the combined signal, which is a pre-distorted signal converted to a radio frequency, as input from the upconverter 17. PA19 then power-amplifies the combined signal, which is a pre-distorted signal. Subsequently, radio 1 radiates the power-amplified combined signal into space as radio waves from its antenna.

[0048] Here, we will explain the entire process using an example of the first and second transmitted signals. For example, if the frequency of the first band is f L The frequency of the second band is f H Let's assume that this is the case. Then, let the first transmitted signal be x1(n) and the second transmitted signal be x2(n), and we will explain the signal transitions.

[0049] The address generation unit 11 generates the address A of the first transmission signal for the first transmission signal. 11 =|x1(n)| 2 , combined power address A12 =|x1(n)| 2 +|x2(n)| 2 , address A of the second transmission signal 13 =|x2(n)| 2 is generated.

[0050] Further, the address generation unit 11 generates, for the second transmission signal, address A of the second transmission signal 21 =|x2(n)| 2 , address A of the combined power 22 =|x1(n)| 2 +|x2(n)| 2 , and address A of the first transmission signal 23 =|x1(n)| 2 is generated.

[0051] LUT 211 receives an input of address A of the first transmission signal 11 from the address generation unit 11, and outputs LUT 1,1 (A 11 ) as a distortion compensation coefficient. LUT 1,1 (A 11 ) is a distortion compensation coefficient for performing distortion compensation on the first transmission signal with respect to the power of the first transmission signal itself. LUT 212 receives an input of address A of the combined signal 12 from the address generation unit 11, and outputs LUT 1,2 (A 12 ) as a distortion compensation coefficient. LUT 1,2 (A 12 ) is a distortion compensation coefficient for performing distortion compensation on the first transmission signal with respect to the combined power. LUT 213 receives an input of address A of the second transmission signal 13 from the address generation unit 11, and outputs LUT 1,3 (A 13 ) as a distortion compensation coefficient. LUT 1,3 (A 13 ) is a distortion compensation coefficient for performing distortion compensation on the first transmission signal with respect to the power of the second transmission signal.

[0052] Further, LUT 221 receives an input of address A of the second transmission signal 21 from the address generation unit 11, and uses LUT as a distortion compensation coefficient 2,1(A 21 Outputs a LUT. 2,1 (A 21 ) is a distortion compensation coefficient for distortion compensation to the second transmitted signal regarding the power of the second transmitted signal itself. LUT222 is the address A of the combined signal. 22 The input is received from the address generation unit 11 and used as the distortion compensation coefficient LUT 2,2 (A 22 Outputs a LUT. 2,2 (A 22 ) is the distortion compensation coefficient for distortion compensation for the second transmitted signal with respect to the combined power. LUT223 is the address A of the first transmitted signal. 23 The input is received from the address generation unit 11 and used as the distortion compensation coefficient LUT 2,3 (A 23 Outputs a LUT. 2,3 (A 23 ) is the distortion compensation coefficient for distortion compensation for the power of the first transmitted signal relative to the second transmitted signal.

[0053] The adder 127 uses the LUT to calculate the distortion compensation coefficient of the first transmitted signal. SUM-1 =LUT 1,1 (A 11 )+LUT 1,2 (A 12 )+LUT 1,3 (A 13 The LUT calculates the distortion compensation coefficient of the second transmitted signal. SUM-2 =LUT 2,1 (A 21 )+LUT 2,2 (A 22 )+LUT 2,3 (A 23 It is calculated as follows:

[0054] The multiplier 131 takes x1(n), which is the first transmitted signal, and the LUT, which is the distortion compensation coefficient of the first transmitted signal. SUM-1 Multiplying by this, the pre-distortion signal u1(n) of the first band is calculated. That is, the multiplier 131 calculates u1(n) = x1(n) × LUT SUM-1This determines the pre-distortion signal u1(n) of the first band. The multiplier 132 then calculates the second transmission signal x2(n) and the LUT, which is the distortion compensation coefficient of the second transmission signal. SUM-2 By multiplying by the two, the pre-distortion signal u2(n) of the second band is calculated. That is, the multiplier 132 calculates u2(n) = x2(n) × LUT SUM-2 This allows us to determine u2(n), which is the pre-distortion signal for the second band.

[0055] Figure 3 is a flowchart of the wireless signal output processing by the wireless device according to Example 1. Next, the flow of the wireless signal output processing by the wireless device 1 according to Example 1 will be explained with reference to Figure 3.

[0056] Radio 1 receives the first and second baseband transmission signals (step S1).

[0057] The power calculation unit 111 receives the input of the first baseband transmission signal via the signal path P1. The power calculation unit 111 then calculates the power of the first transmission signal and generates the address of the first transmission signal (step S2).

[0058] The power calculation unit 112 receives the input of the second baseband transmission signal via the signal path P2. The power calculation unit 112 then calculates the power of the second transmission signal and generates the address of the second transmission signal (step S3).

[0059] The adder 121 generates an address for the combined power by adding the power of the first transmission signal input from the power calculation unit 111 and the power of the second transmission signal input from the power calculation unit 112 (step S4).

[0060] LUT211 receives the address of the first transmission signal from the power calculation unit 111. LUT212 receives the address of the combined power from the adder 121. LUT213 receives the address of the second transmission signal from the power calculation unit 112 (step S5).

[0061] LUT211 outputs a distortion compensation coefficient to adder 127 for distortion compensation of the first transmitted signal for the power of the first transmitted signal itself, corresponding to the address of the first transmitted signal. LUT212 outputs a distortion compensation coefficient to adder 127 for distortion compensation of the combined power for the first transmitted signal, corresponding to the address of the combined power. LUT213 outputs a distortion compensation coefficient to adder 127 for distortion compensation of the first transmitted signal for the power of the second transmitted signal, corresponding to the address of the second transmitted signal. Adder 127 adds the three input distortion compensation coefficients to calculate the distortion compensation coefficient for the first transmitted signal (step S6).

[0062] The multiplier 131 multiplies the first transmitted signal sent via the signal path P1 by the distortion compensation coefficient of the first transmitted signal to perform distortion compensation on the first transmitted signal and outputs it to the frequency shift unit 141 as a pre-distortion signal for the first band (step S7).

[0063] The frequency shifting unit 141 performs a frequency shift of the pre-distortion signal of the first band (step S8).

[0064] LUT221 receives the address of the second transmission signal from the power calculation unit 112. LUT222 receives the address of the combined power from the adder 121. LUT223 receives the address of the first transmission signal from the power calculation unit 111 (step S9).

[0065] LUT223 outputs a distortion compensation coefficient to adder 128 for distortion compensation of the second transmission signal for the power of the first transmission signal, corresponding to the address of the first transmission signal. LUT222 outputs a distortion compensation coefficient to adder 128 for distortion compensation of the second transmission signal for the combined power, corresponding to the address of the combined power. LUT221 outputs a distortion compensation coefficient to adder 128 for distortion compensation of the second transmission signal for the power of the second transmission signal itself, corresponding to the address of the second transmission signal. Adder 128 adds the three input distortion compensation coefficients to calculate the distortion compensation coefficient for the second transmission signal (step S10).

[0066] The multiplier 132 multiplies the second transmitted signal sent via the signal path P2 by the distortion compensation coefficient of the second transmitted signal to perform distortion compensation on the second transmitted signal, and outputs it to the frequency shift unit 142 as a pre-distortion signal for the second band (step S11).

[0067] The frequency shifting unit 142 performs a frequency shift of the pre-distortion signal of the second band (step S12).

[0068] The adder 15 adds the first transmission signal, which is a pre-distortion signal having a frequency in the first band, and the second transmission signal, which is a pre-distortion signal having a frequency in the second band, to generate a combined signal of the first and second transmission signals. The DAC 16 converts the combined signal, which is a pre-distortion signal, from a digital signal to an analog signal. The upconverter, including the upconverter 17 and local oscillator 18, converts the frequency of the combined signal, which is a pre-distortion signal, to a radio frequency (step S13).

[0069] PA19 power-amplifies the combined signal, which is a pre-distorted signal converted to a radio frequency. Then, radio 1 radiates the power-amplified combined signal into space as a radio wave from its antenna (step S14).

[0070] Figure 4 is a hardware configuration diagram of the radio. As shown in Figure 4, the radio 1 includes, for example, a communication interface 91, a processor 92, a memory 93, and a wireless communication circuit 94.

[0071] The communication interface 91 mediates communication between the processor 92 and external devices. For example, the communication interface 91 outputs the first transmission signal and the second transmission signal input from the transmission device 2 to the processor 92.

[0072] Memory 93 is a storage device, such as DRAM (Dynamic Random Access Memory), a hard disk, or ROM (Read Only Memory). Memory 93 can store three one-dimensional LUTs each for the first and second transmission signals, as illustrated in Figure 2, namely LUTs 211-213 and 221-223. Memory 93 also stores various programs, including programs for realizing the functions of the address generation unit 11, the distortion compensation coefficient calculation unit 12, the multipliers 131 and 132, and the frequency shift units 141 and 142. Data in Memory 93 is read and written by the processor 92.

[0073] The processor 92 is, for example, a CPU (Central Processing Unit) or an FPGA (Field Programmable Gate Array). The processor 92 operates using memory 93 to implement the functions of the address generation unit 11, distortion compensation coefficient calculation unit 12, multipliers 131 and 132, and frequency shift units 141 and 142 as illustrated in Figure 2. Specifically, the processor 92 performs a process to generate three types of address information for each of the first and second transmission signals based on the power or amplitude of the first transmission signal and the power or amplitude of the second transmission signal. The processor 92 also inputs the addresses into LUTs 211-213 and 221-223 respectively and reads out the distortion compensation coefficients from each. Then, for each of the first and second transmission signals, the processor 92 multiplies the sum of the three read distortion compensation coefficients by the transmission signal to generate a pre-distortion signal for each band. Furthermore, the processor 92 adds the pre-distortion signals for each band to generate a composite signal, which is a pre-distortion signal.

[0074] The wireless communication circuit 94 includes a DAC 16, an upconverter 17, an upconverter including a local oscillator 18, and a PA 19. The wireless communication circuit 94 converts the digital transmission signal output from the processor 92 into an analog wireless signal and outputs it to the antenna. The wireless communication circuit 94 may also include a downconverter and an ADC (Analog to Digital Converter), in which case a portion of the amplifier output signal can be fed back to the processor 92.

[0075] Here, we will explain a comparison with a configuration using two-dimensional LUTs for each band. In this embodiment, LUTs 211-213 and 221-223 use one-dimensional LUTs. In contrast, a configuration using two-dimensional LUTs for each band is also conceivable. A two-dimensional LUT outputs a single distortion compensation coefficient depending on the address of the first transmitted signal and the address of the second transmitted signal. That is, a two-dimensional LUT can consider the influence of power on both the address of the first transmitted signal and the address of the second transmitted signal for each of the first and second transmitted signals.

[0076] For example, if the first transmitted signal is x1(n) and the second transmitted signal is x2(n), then the pre-distortion signals for each band output from the band-by-band distortion compensation section using a two-dimensional LUT are denoted as u1'(n) and u2'(n). In this case, the pre-distortion signals u1'(n) and u2'(n) are given by u1'(n) = x1(n) × LUT 2D-1 {|x1(n)| 2 ,|x2(n)| 2} and u2'(n)=x2(n)×LUT 2D-2 {|x1(n)| 2 ,|x2(n)| 2 This is expressed as}. Here, LUT 2D-1 {|x1(n)| 2 ,|x2(n)| 2} is the distortion compensation coefficient output from the two-dimensional LUT for the first transmitted signal. 2D-2 {|x1(n)| 2 ,|x2(n)| 2} is the distortion compensation coefficient output from the two-dimensional LUT for the second transmitted signal.

[0077] In a configuration using a two-dimensional LUT for each band, for example, distortion compensation coefficients corresponding to both the address of the first transmitted signal and the address of the second transmitted signal are output, resulting in the calculation of highly accurate approximated distortion compensation coefficients. However, the circuit size of the digital section becomes large, making implementation difficult and impractical. The digital section refers to, for example, the part of the radio 1 that performs processing before the DAC 16, including the address generation unit 11 and the distortion compensation coefficient calculation unit 12.

[0078] Here, the pre-distortion signal u1'(n) of the first band can be approximately expanded as shown in the following equation (1) if it is approximated to, for example, fifth-order distortion or less. The approximation here is limited to fifth-order distortion or less because compensation terms of seventh-order distortion or higher have little effect on distortion compensation. By using the compensation coefficients approximated to fifth-order distortion or less, radio 1 can satisfy the ACLR (Adjacent Channel Power Leakage Ratio) > 45 dB standard of the 3GPP (registered trademark) (3rd Generation Partnership Project). Note that u2'(n) can be expanded in a similar manner.

[0079]

number

[0080] In this case, the expanded signal contains compensation terms from the 1st to the 6th term. Furthermore, the coefficients of each compensation term can be expressed independently. Therefore, the distortion compensation coefficients used in equation (1) are expressed from the distortion compensation coefficients obtained from LUTs 211-213 and 221-223, which are one-dimensional LUTs obtained according to the three types of addresses for each band. Here, for example, if there are three types of addresses for each band, and the value indicating how many steps the address is represented by is NA, then the number of distortion compensation coefficients for each band is NA × 3. In contrast, for example, when using a two-dimensional LUT, the number of distortion compensation coefficients for each band is NA × NA. Therefore, the memory capacity of LUTs 211-213 and 221-223 that store the distortion compensation coefficients can be kept small, and the circuit size and memory capacity of the digital section can be reduced compared to when using a two-dimensional LUT.

[0081] For example, if distortion of the 7th order or higher is to be considered, radio 1 can calculate the distortion compensation coefficient by increasing the number of LUTs 211-213 and 221-223 for each band.

[0082] As described above, the radio 1 according to this embodiment performs distortion compensation processing for each band, so even if the nonlinear characteristics of PA19 differ in each band, optimal compensation can be provided for the distortion of each band. Furthermore, the radio 1 can compensate for the three types of nonlinear distortion that occur in a common amplification configuration by using three one-dimensional LUTs each of LUTs 211~213 and 223~221. Therefore, the radio 1 can improve distortion compensation performance and improve ACLR.

[0083] Furthermore, by performing distortion compensation processing for each band, the radio 1 only needs to secure a digital processing bandwidth of 3 to 5 times the transmission signal bandwidth for each band, and the amount of computation can be reduced by processing at a low sampling rate, thus reducing the circuit size of the digital section. In addition, by commonly amplifying the multiband signal with a single power amplifier PA19, the radio 1 can reduce the number of devices compared to using an individual amplification configuration in which each band is individually amplified by a separate power amplifier. As a result, the radio 1 can keep the mounting area small and reduce costs by the amount of components that have been reduced. [Examples]

[0084] Figure 5 is a block diagram of the radio according to Embodiment 2. In this embodiment, the radio 1 compensates for the memory effect, in which the nonlinear distortion of PA19 is affected by past states, by changing the combination of delay amounts for the power delay of the transmission signals of the first and second bands, and the transmission delay of the first and second transmission signals. In the following description, the operation of each part, which is the same as in Embodiment 1, may be omitted.

[0085] The address generation unit 11 includes delay-adding units 101 and 102. The delay-adding unit 101 adds a delay to the power of the first transmission signal output from the power calculation unit 111. The delay-adding unit 102 adds a delay to the power of the second transmission signal output from the power calculation unit 112. For example, the delay-adding unit 101 adds a delay of qpw1, and the delay-adding unit 102 adds a delay of qpw2.

[0086] The following explanation will describe the case where the first transmitted signal is x1(n) and the second transmitted signal is x2(n). In this case, the delay unit 101 is the power of the first transmitted signal |x1(n)| 2 By giving a delay amount qpw1, |x1(n-qpw1)| 2 It outputs the power of the second transmission signal |x2(n)|. 2 By giving a delay amount qpw2, |x2(n-qpw2)| 2and outputs the same. Hereinafter, the delay amount qpw1 may be referred to as the power delay amount of the first transmission signal, and the delay amount qpw2 may be referred to as the power delay amount of the second transmission signal.

[0087] These delay adding units 101 and 102 correspond to an example of a "second delay adding unit". Further, the delay imparted by the delay adding unit 101 to the power of the first transmission signal and the delay imparted by the delay adding unit 102 to the power of the second transmission signal correspond to an example of a "second delay".

[0088] The address generation unit 11 generates, for the first transmission signal, an address A of the first transmission signal 11,qpw1 =|x1(n-qpw1)| 2 Furthermore, the address generation unit 11 generates, for the first transmission signal, a combined power address A 12,qpw1,qpw2 =|x1(n-qpw1)| 2 +|x2(n-qpw2)| 2 Moreover, the address generation unit 11 generates, for the first transmission signal, an address A of the second transmission signal 13,qpw2 =|x2(n-qpw2)| 2 .

[0089] Furthermore, the address generation unit 11 generates, for the second transmission signal, an address A of the second transmission signal 21,qpw2 =|x2(n-qpw2)| 2 Furthermore, the address generation unit 11 generates, for the second transmission signal, a combined power address A 22,qpw1,qpw2 =|x1(n-qpw1)| 2 +|x2(n-qpw2)| 2 Moreover, the address generation unit 11 generates, for the second transmission signal, an address A of the first transmission signal 23,qpw1 =|x1(n-qpw1)| 2 .

[0090] The LUT 211 receives an input of address A 11,qpw1 and outputs LUT 1,1,qpw1,qtx1 (A 11,qpw1 ). The LUT 212 receives an input of address A 12,qpw1,qpw2 and outputs LUT 1,2,qpw1,qtx1,qpw2 (A 12,qpw1,qpw2) outputs. LUT213 outputs address A 13,qpw2 For the input, LUT 1,3,qtx1,qpw2 (A 13,qpw2 Outputs ).

[0091] LUT221 is at address A 21,qpw2 For the input, LUT 2,1,qpw2,qtx2 (A 21,qpw2 ) outputs. LUT222 outputs address A 22,qpw1,qpw2 For the input, LUT 2,2,qpw2,qtx2,qpw1 (A 22,qpw1,qpw2 ) outputs. LUT223 outputs address A 23,qpw1 For the input, LUT 2,3,qtx2,qpw1 (A 23,qpw1 Outputs ).

[0092] The adder 127 controls the distortion compensation coefficient LUT of the first transmitted signal. SUM-1,qtx1 LUT SUM-1,qtx1 =LUT 1,1,qpw1,qtx1 (A 11,qpw1 )+LUT 1,2,qpw1,qtx1,qpw2 (A 12,qpw1,qpw2 )+LUT 1,3,qtx1,qpw2 (A 13,qpw2 ) is calculated as follows. In addition, the adder 128 is the distortion compensation coefficient LUT of the second transmitted signal. SUM-2,qtx2 LUT SUM-2,qtx2 =LUT 2,1,qpw2,qtx2 (A 21,qpw2 )+LUT 2,2,qpw2,qtx2,qpw1 (A 22,qpw1,qpw2 )+LUT 2,3,qtx2,qpw1 (A 23,qpw1 It is calculated as follows:

[0093] The delay unit 134 adds a delay to the first transmission signal. For example, if the delay amount is qtx1, the delay unit 134 adds a delay amount qtx1 to the first transmission signal x1(n) and outputs x1(n-qtx1).

[0094] The delay unit 135 adds a delay to the second transmission signal. For example, if the delay amount is qtx2, the delay unit 135 adds a delay amount of qtx2 to the second transmission signal x2(n) and outputs x2(n-qtx2).

[0095] These delay-adding units 134 and 135 are examples of "first delay-adding units." Furthermore, the delay that delay-adding unit 134 imparts to the first transmission signal and the delay that delay-adding unit 135 imparts to the second transmission signal are examples of "first delays."

[0096] The multiplier 131 combines x1(n-qtx1) output from the delay addition unit 134 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-1,qtx1 Multiplying these together, we get the pre-distortion signal u1(n) = x1(n-qtx1) × LUT for the first band. SUM-1,qtx1 Outputs.

[0097] The multiplier 132 combines x2(n-qtx2) output from the delay addition unit 135 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-2,qtx2 Multiplying by this, the second band pre-distortion signal u2(n) = x2(n-qtx2) × LUT SUM-2,qtx2 Outputs.

[0098] The above explanation formulates one pattern of delay amount combinations. Radio 1 can compensate for the memory effect and further improve distortion compensation performance by varying each of the delay amounts qtx1, qtx2, qpw1, and qpw2, and, for example, by using an adder (not shown in the figure) to add the signals and make them pre-distortion signals. Several examples of combinations of each delay amount qtx1, qtx2, qpw1, and qpw2 are described below.

[0099] As a first example, in the first and second transmission signals, the power of the first transmission signal |x1(n)| 2 The delay amount pqw1 and the power of the second transmitted signal |x2(n)| 2 Let's explain the case where the delay amount pqw1 and pqw2 are the same. That is, for both the first and second transmitted signals, pqw1 = pqw2 = pqw. However, the delay amount pqw in the first transmitted signal and the delay amount pqw in the second transmitted signal may be different. Here, the power of the first transmitted signal |x1(n)| 2The delay amount and the power of the second transmitted signal |x2(n)| 2 The delay amount pqw is set to a changeable delay amount q11. Also, the power |x1(n)| of the first transmitted signal in the second transmitted signal. 2 The delay amount and the power of the second transmitted signal |x2(n)| 2 The delay amount pqw is set to a changeable delay amount q21. Furthermore, the delay amount qtx1 of the first transmitted signal is set to a changeable delay amount q12, and the delay amount qtx2 of the second transmitted signal is set to a changeable delay amount q22.

[0100] The delay addition unit 134 adds a delay amount q12 to the first transmission signal x1(n) and outputs x1(n-q12). The delay addition unit 135 adds a delay amount q22 to the second transmission signal x2(n) and outputs x2(n-q22).

[0101] The address generation unit 11 generates the address A of the first transmission signal for the first transmission signal. 11,q11 =|x1(n-q11)| 2 The address generation unit 11 generates the combined power address A for the first transmission signal. 12,q11 =|x1(n-q11)| 2 +|x2(n-q11)| 2 The address generation unit 11 generates the address A of the second transmission signal for the first transmission signal. 13,q11 =|x2(n-q11)| 2 Generates.

[0102] Furthermore, the address generation unit 11 generates the address A of the second transmission signal for the second transmission signal. 21,q21 =|x2(n-q21)| 2 The address generation unit 11 generates the combined power address A for the second transmission signal. 22,q21 =|x1(n-q21)| 2 +|x2(n-q21)| 2 The address generation unit 11 generates the address A of the first transmission signal for the second transmission signal. 23,q21 =|x1(n-q21)| 2 Generates.

[0103] LUT211 is address A 11,q11 For the input, LUT 1,1,q11,q12 (A 11,q11 ) outputs. LUT212 outputs address A 12,q11 For the input, LUT 1,2,q11,q12 (A 12,q11 ) outputs. LUT213 outputs address A 13,q11 For the input, LUT 1,3,q11,q12 (A 13,q11 Outputs ).

[0104] LUT221 is at address A 21,q21 For the input, LUT 2,1,q21,q22 (A 21,q21 ) outputs. LUT222 outputs address A 22,q21 For the input, LUT 2,2,q21,q22 (A 22,q21 ) outputs. LUT223 outputs address A 23,q21 For the input, LUT 2,3,q21,q22 (A 23,q21 Outputs ).

[0105] The adder 127 controls the distortion compensation coefficient LUT of the first transmitted signal. SUM-1,q12 This is calculated using the following formula (2). Here, radio 1 sets q11 to -Q 1,1 From +Q 1,1 Change it to that extent.

[0106]

number

[0107] The adder 128 controls the distortion compensation coefficient LUT of the second transmitted signal. SUM-2,q22 This is calculated using the following formula (3). Here, radio 1 sets q21 to -Q 2,1 From +Q 2,1 Change it to that extent.

[0108]

number

[0109] The multiplier 131 combines x1(n-q12) output from the delay addition unit 134 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-1,q12 Multiplying by the following equation (4), a pre-distortion signal is output. Here, radio 1 sets q12 to -Q 1,2 From +Q 1,2 The value is changed to this extent. Although only one multiplier 131 is shown in Figure 5, multiple multipliers 131 are arranged according to the number of multiplications in equation (4), and the results output from each multiplier 131 are added together, for example by an adder (not shown in the figure), to perform the calculation of equation (4).

[0110]

number

[0111] The multiplier 132 combines x2(n-q22) output from the delay addition unit 135 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-2,q22 Multiplying by the following equation (5), a pre-distortion signal is output. Here, radio 1 sets q22 to -Q 2,2 From +Q 2,2 The value is changed to this extent. Although only one multiplier 132 is shown in Figure 5, multiple multipliers 132 are arranged depending on the number of multiplications in equation (5), and the results output from each multiplier 132 are added together, for example by an adder (not shown in the figure), to perform the calculation of equation (5).

[0112]

number

[0113] Next, as a second example, in the first and second transmission signals, the power of the first transmission signal |x1(n)| 2 The delay amount pqw1 and the power of the second transmitted signal |x2(n)| 2We will now explain the case where the delay amount pqw2 is different from that of the first transmitted signal. That is, for both the first and second transmitted signals, pqw1 ≠ pqw2. Here, the power of the first transmitted signal |x1(n)| 2 Let q11 be the amount of change in the delay, and the power of the second transmitted signal |x2(n)| 2 Let q13 be the amount of delay that changes. Also, the power of the second transmitted signal |x2(n)| 2 Let q21 be the delay amount to be converted, and the power of the first transmitted signal |x1(n)| 2 Let q23 be the amount of delay that changes. Also, let q12 be the amount of delay qtx1 of the first transmitted signal, and let q22 be the amount of delay qtx2 of the second transmitted signal.

[0114] The delay addition unit 134 adds a delay amount q12 to the first transmission signal x1(n) and outputs x1(n-q12). The delay addition unit 135 adds a delay amount q22 to the second transmission signal x2(n) and outputs x2(n-q22).

[0115] The address generation unit 11 generates the address A of the first transmission signal for the first transmission signal. 11,q11 =|x1(n-q11)| 2 The address generation unit 11 generates the combined power address A for the first transmission signal. 12,q11,q13 =|x1(n-q11)| 2 +|x2(n-q13)| 2 The address generation unit 11 generates the address A of the second transmission signal for the first transmission signal. 13,q13 =|x²(n-q13)| 2 Generates.

[0116] Furthermore, the address generation unit 11 generates the address A of the second transmission signal for the second transmission signal. 21,q21 =|x2(n-q21)| 2 The address generation unit 11 generates the combined power address A for the second transmission signal. 22,q21,q23 =|x2(n-q21)| 2 +|x1(n-q23)| 2The address generation unit 11 generates the address A of the first transmission signal for the second transmission signal. 23,q23 =|x1(n-q23)| 2 Generates.

[0117] LUT211 is address A 11,q11 For the input, LUT 1,1,q11,q12,q13 (A 11,q11 ) outputs. LUT212 outputs address A 12,q11,q13 For the input, LUT 1,2,q11,q12,q13 (A 12,q11,q13 ) outputs. LUT213 outputs address A 13,q13 For the input, LUT 1,3,q11,q12,q13 (A 13,q13 Outputs ).

[0118] LUT221 is at address A 21,q21 For the input, LUT 2,1,q21,q22,q23 (A 21,q21 ) outputs. LUT222 outputs address A 22,q21,q23 For the input, LUT 2,2,q21,q22,q23 (A 22,q21,q23 ) outputs. LUT223 outputs address A 23,q23 For the input, LUT 2,3,q21,q22,q23 (A 23,q23 Outputs ).

[0119] The adder 127 controls the distortion compensation coefficient LUT of the first transmitted signal. SUM-1,q12 This is calculated using the following formula (6). Here, radio 1 sets q11 to -Q 1,1 From +Q 1,1 Change it to -Q 1,3 From +Q 1,3 Change it to that extent.

[0120]

number

[0121] The adder 128 controls the distortion compensation coefficient LUT of the second transmitted signal. SUM-2,q22 This is calculated using the following formula (7). Here, radio 1 sets q21 to -Q 2,1From +Q 2,1 Change it to -Q 2,3 From +Q 2,3 Change it to that extent.

[0122]

number

[0123] The multiplier 131 combines x1(n-q12) output from the delay addition unit 134 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-1,q12 Multiplying by the above formula (4), a pre-distortion signal is output. Here, radio 1 sets q12 to -Q 1,2 From +Q 1,2 Change it to that extent.

[0124] The multiplier 132 combines x2(n-q22) output from the delay addition unit 135 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-2,q22 Multiplying by the above equation (5), a pre-distortion signal is output. Here, radio 1 sets q22 to -Q 2,2 From +Q 2,2 Change it to that extent.

[0125] Next, as a third example, we will describe a different method for calculating the sum of three distortion compensation coefficients based on the second example.

[0126] LUT 1,1,q11,q12,q13 (A 11,q11 ) does not need to include q13 as a variable, so adder 127 is LUT 1,1,q11,q12,q13 (A 11,q11 ) Regarding - Q 1,3 From +Q 1,3 It is not necessary to perform the calculation to find the sum up to that point. In other words, adder 127 is LUT 1,1,q11,q12,q13 (A 11,q11 The LUT obtained by removing the variable q13 from ) 1,1,q11,q12 (A 11,q11 Using -Q 1,1 From +Q 1,1 We just need to find the sum up to [the specified number].

[0127] Also, LUT 1,3,q11,q12,q13 (A 13,q13 ) does not need to include q11 as a variable, so adder 127 is LUT 1,3,q11,q12,q13 (A 13,q13 ) Regarding - Q 1,1 From +Q 1,1 It is not necessary to perform the calculation to find the sum up to that point. In other words, adder 127 is LUT 1,3,q11,q12,q13 (A 13,q13 The LUT obtained by removing the variable q11 from ) 1,3,q12,q13 (A 13,q13 Using -Q 1,3 From +Q 1,3 We just need to find the sum up to [the specified number].

[0128] LUT 2,1,q21,q22,q23 (A 21,q21 ) does not need to include q23 as a variable, so adder 128 is LUT 2,1,q21,q22,q23 (A 21,q21 ) Regarding - Q 2,3 From +Q 2,3 It is not necessary to perform the calculation to find the sum up to that point. In other words, adder 128 is LUT 2,1,q21,q22,q23 (A 21,q21 ) LUT with the variable q23 removed 2,1,q21,q22 (A 21,q21 Using -Q 2,1 From +Q 2,1 We just need to find the sum up to [the specified number].

[0129] Also, LUT 2,3,q21,q22,q23 (A 23,q23 ) does not need to include q21 as a variable, so adder 128 is LUT 2,3,q21,q22,q23 (A 23,q23 ) Regarding - Q 2,1 From +Q 2,1 It is not necessary to perform the calculation to find the sum up to that point. In other words, adder 128 is LUT 2,3,q21,q22,q23 (A 23,q23 The LUT obtained by removing the variable q21 from ) 2,3,q22,q23 (A 23,q23 Using -Q 2,3 From +Q 2,3 We just need to find the sum up to [the specified number].

[0130] By making the above changes, the memory capacity of LUTs 211-213 and 221-223, which store the distortion compensation coefficients, can be reduced. The calculation process for the distortion compensation coefficients is explained below.

[0131] The delay addition unit 134 adds a delay amount q12 to the first transmission signal x1(n) and outputs x1(n-q12). The delay addition unit 135 adds a delay amount q22 to the second transmission signal x2(n) and outputs x2(n-q22).

[0132] The address generation unit 11 generates the address A of the first transmission signal for the first transmission signal. 11,q11 =|x1(n-q11)| 2 The address generation unit 11 generates the combined power address A for the first transmission signal. 12,q11,q13 =|x1(n-q11)| 2 +|x2(n-q13)| 2 The address generation unit 11 generates the address A of the second transmission signal for the first transmission signal. 13,q13 =|x²(n-q13)| 2 Generates.

[0133] Furthermore, the address generation unit 11 generates the address A of the second transmission signal for the second transmission signal. 21,q21 =|x2(n-q21)| 2 The address generation unit 11 generates the combined power address A for the second transmission signal. 22,q21,q23 =|x2(n-q21)| 2 +|x1(n-q23)| 2 The address generation unit 11 generates the address A of the first transmission signal for the second transmission signal. 23,q23 =|x1(n-q23)| 2 Generates.

[0134] LUT211 is address A 11,q11 For the input, LUT 1,1,q11,q12 (A 11,q11 ) outputs. LUT212 outputs address A 12,q11,q13 For the input, LUT 1,2,q11,q12,q13 (A 12,q11,q13) outputs. LUT213 outputs address A 13,q13 For the input, LUT 1,3,q12,q13 (A 13,q13 Outputs ).

[0135] LUT221 is at address A 21,q21 For the input, LUT 2,1,q21,q22 (A 21,q21 ) outputs. LUT222 outputs address A 22,q21,q23 For the input, LUT 2,2,q21,q22,q23 (A 22,q21,q23 ) outputs. LUT223 outputs address A 23,q23 For the input, LUT 2,3,q22,q23 (A 23,q23 Outputs ).

[0136] The adder 127 controls the distortion compensation coefficient LUT of the first transmitted signal. SUM-1,q12 This is calculated using the following formula (8). Here, radio 1 sets q11 to -Q 1,1 From +Q 1,1 Change it to -Q 1,3 From +Q 1,3 Change it to that extent.

[0137]

number

[0138] The adder 128 controls the distortion compensation coefficient LUT of the second transmitted signal. SUM-2,q22 This is calculated using the following formula (9). Here, radio 1 sets q21 to -Q 2,1 From +Q 2,1 Change it to -Q 2,3 From +Q 2,3 Change it to that extent.

[0139]

number

[0140] The multiplier 131 combines x1(n-q12) output from the delay addition unit 134 and the LUT output from the distortion compensation coefficient calculation unit 12.SUM-1,q12 Multiplying by the above formula (4), a pre-distortion signal is output. Here, radio 1 sets q12 to -Q 1,2 From +Q 1,2 Change it to that extent.

[0141] The multiplier 132 combines x2(n-q22) output from the delay addition unit 135 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-2,q22 Multiplying by the above equation (5), a pre-distortion signal is output. Here, radio 1 sets q22 to -Q 2,2 From +Q 2,2 Change it to that extent.

[0142] In the first to third examples of Embodiment 1 described above, the radio 1 does not need to use distortion compensation coefficients or compensation terms corresponding to all combinations of delay amounts for q11, q12, q13 and q21, q22, q23. That is, the radio 1 can select some combinations of delay amounts that have a relatively large impact on distortion compensation performance and use the distortion compensation coefficients or compensation terms corresponding to them. As a result, the radio 1 can reduce the memory capacity of LUTs 211-213 and 221-223 that store the distortion compensation coefficients, and reduce the number of multipliers 131 and 132 that multiply the distortion compensation coefficients by the transmitted signal, while minimizing the degradation of distortion compensation performance. [Examples]

[0143] Next, we will describe Example 3. In Example 1, three tables for calculating the distortion compensation coefficient were used for each of the first and second bands, but this number may be reduced to two or one. Figure 6 is a block diagram of the radio according to Example 3. In this example, the radio 1 uses two tables for calculating the distortion compensation coefficient for each of the first and second bands. In the following description, the operation of each part, which is the same as in Example 1, may be omitted.

[0144] The radio 1 according to this embodiment has two LUTs, LUT211 and LUT212, for the first band. LUT211 is a table in which distortion compensation coefficients for distortion compensation for the power of the first transmitted signal itself are stored, associated with the address of the first transmitted signal. LUT212 is a table in which distortion compensation coefficients for distortion compensation for the combined power of the first transmitted signal are stored, associated with the address of the combined power.

[0145] Furthermore, the radio 1 has two LUTs, LUT221 and LUT222, for the second band. LUT221 is a table in which distortion compensation coefficients for distortion compensation of the second transmitted signal for the power of the second transmitted signal itself are stored, associated with the address of the second transmitted signal. LUT222 is a table in which distortion compensation coefficients for distortion compensation of the second transmitted signal for the combined power are stored, associated with the address of the combined power.

[0146] Here, we will explain the case where the first transmitted signal is x1(n) and the second transmitted signal is x2(n).

[0147] The address generation unit 11 generates the address A of the first transmission signal for the first transmission signal. 11 =|x1(n)| 2 , combined power address A 12 =|x1(n)| 2 +|x2(n)| 2 Generates.

[0148] Furthermore, the address generation unit 11 generates the address A of the second transmission signal for the second transmission signal. 21 =|x2(n)| 2 , combined power address A 22 =|x1(n)| 2 +|x2(n)| 2 Generates.

[0149] Then, LUT211 is used as the distortion compensation coefficient corresponding to address A11 of the first transmitted signal. 1,1 (A 11 Outputs a LUT. 1,1 (A11 ) is a distortion compensation coefficient for distortion compensation for the first transmitted signal with respect to the power of the first transmitted signal itself. LUT212 is the distortion compensation coefficient corresponding to address A12 of the combined power. 1,2 (A 12 Outputs a LUT. 1,2 (A 12 ) is the distortion compensation coefficient for distortion compensation with respect to the first transmitted signal for the combined power.

[0150] Furthermore, LUT221 is the address A of the second transmission signal. 21 LUT as the corresponding distortion compensation coefficient 2,1 (A 21 Outputs a LUT. 2,1 (A 21 ) is the distortion compensation coefficient for distortion compensation to the second transmit signal regarding the power of the second transmit signal itself. LUT222 is the combined power at address A 22 LUT as the corresponding distortion compensation coefficient 2,2 (A 22 Outputs a LUT. 2,2 (A 22 ) is the distortion compensation coefficient for distortion compensation to the second transmitted signal for the combined power.

[0151] The adder 127 uses the LUT to calculate the distortion compensation coefficient of the first transmitted signal. SUM-1 =LUT 1,1 (A 11 )+LUT 1,2 (A 12 The distortion compensation coefficient is calculated as follows. The radio 1 can use this distortion compensation coefficient to perform distortion compensation on the first transmitted signal for the power and combined power of the first transmitted signal. The adder 128 also calculates the distortion compensation coefficient of the second transmitted signal as LUT SUM-2 =LUT 2,1 (A 21 )+LUT 2,2 (A 22 The distortion compensation coefficient is calculated as follows. By using this distortion compensation coefficient, the radio 1 can perform distortion compensation on the second transmitted signal for the power and combined power of the second transmitted signal.

[0152] As described above, the radio 1 according to this embodiment performs distortion compensation of the transmitted signal using a distortion compensation coefficient for the power of the transmitted signal in a specific band in multiband transmission and a distortion compensation coefficient for the combined power with the transmitted signals of other bands. In this way, even if there are two tables for calculating the distortion compensation coefficient for each transmitted signal, optimal compensation can be achieved for the distortion of each band. Therefore, the radio 1 can improve distortion compensation performance and improve ACLR.

[0153] (Variation 3-1) Furthermore, the radio 1 may have two LUTs, LUT211 and LUT213, for the first band, and two LUTs, LUT221 and LUT223, for the second band. LUT211 is a table in which distortion compensation coefficients for distortion compensation for the first transmitted signal with respect to the power of the first transmitted signal itself are stored, associated with the address of the first transmitted signal. LUT213 is a table in which distortion compensation coefficients for distortion compensation for the second transmitted signal with respect to the power of the second transmitted signal are stored, associated with the address of the second transmitted signal. LUT221 is a table in which distortion compensation coefficients for distortion compensation for the second transmitted signal with respect to the power of the second transmitted signal are stored, associated with the address of the second transmitted signal. LUT223 is a table in which distortion compensation coefficients for distortion compensation for the second transmitted signal with respect to the power of the first transmitted signal are stored, associated with the address of the first transmitted signal.

[0154] The address generation unit 11 generates the address A of the first transmission signal for the first transmission signal. 11 =|x1(n)| 2 , address A of the second transmitted signal 13 =|x2(n)| 2 Generates.

[0155] Furthermore, the address generation unit 11 generates the address A of the second transmission signal for the second transmission signal. 21 =|x2(n)| 2 , address A of the first transmitted signal 23 =|x1(n)| 2 Generates.

[0156] Then, LUT211 is used as the distortion compensation coefficient corresponding to address A11 of the first transmitted signal. 1,1 (A 11 Outputs a LUT. 1,1 (A 11 ) is a distortion compensation coefficient for distortion compensation to the first transmitted signal with respect to the power of the first transmitted signal itself. LUT213 is the address A of the second transmitted signal. 13 LUT as the corresponding distortion compensation coefficient 1,3 (A 13 Outputs a LUT. 1,3 (A 13 ) is the distortion compensation coefficient for distortion compensation for the power of the second transmitted signal relative to the first transmitted signal.

[0157] Furthermore, LUT221 is the address A of the second transmission signal. 21 LUT as the corresponding distortion compensation coefficient 2,1 (A 21 Outputs a LUT. 2,1 (A 21 ) is a distortion compensation coefficient for distortion compensation to the second transmit signal regarding the power of the second transmit signal itself. LUT223 is the address A of the first transmit signal. 23 LUT as the corresponding distortion compensation coefficient 2,3 (A 23 Outputs a LUT. 2,3 (A 23 ) is the distortion compensation coefficient for distortion compensation for the power of the first transmitted signal relative to the second transmitted signal.

[0158] The adder 127 uses the LUT to calculate the distortion compensation coefficient of the first transmitted signal. SUM-1 =LUT 1,1 (A 11 )+LUT 1,3 (A 13 The distortion compensation coefficient is calculated as follows. The radio 1 can use this distortion compensation coefficient to perform distortion compensation on the power of the first transmitted signal and the power of the second transmitted signal for the first transmitted signal. The adder 128 also calculates the distortion compensation coefficient of the second transmitted signal as LUT SUM-2 =LUT 2,1 (A 21 )+LUT 2,3 (A 23The distortion compensation coefficient is calculated as follows. By using this distortion compensation coefficient, the radio 1 can perform distortion compensation on the second transmission signal with respect to the power of the second transmission signal and the power of the first transmission signal.

[0159] As described above, the radio 1 according to this embodiment performs distortion compensation of the transmitted signal using a distortion compensation coefficient for the power of the transmitted signal in a specific band and a distortion compensation coefficient for the power of the transmitted signals in other bands in multiband transmission. When two transmitted signals exist, the power of each transmitted signal may have a relatively large impact on the distortion compensation performance. Therefore, the radio 1 according to this embodiment can provide more optimal compensation for the distortion of each band compared to the case where a distortion compensation coefficient for combined power is used as one of the two distortion compensation coefficients. Consequently, the radio 1 can improve distortion compensation performance and improve ACLR.

[0160] (Variation 3-2) Furthermore, the radio 1 may have two LUTs, LUT212 and LUT213, for the first band, and two LUTs, LUT222 and LUT223, for the second band. LUT212 is a table in which distortion compensation coefficients for distortion compensation for the first transmitted signal with respect to the combined power are stored, associated with the address of the combined power. LUT213 is a table in which distortion compensation coefficients for distortion compensation for the first transmitted signal with respect to the power of the second transmitted signal are stored, associated with the address of the second transmitted signal. LUT222 is a table in which distortion compensation coefficients for distortion compensation for the second transmitted signal with respect to the combined power are stored, associated with the address of the combined power. LUT223 is a table in which distortion compensation coefficients for distortion compensation for the second transmitted signal with respect to the power of the first transmitted signal are stored, associated with the address of the first transmitted signal.

[0161] The address generation unit 11 generates the combined power address A for the first transmission signal. 12 =|x1(n)| 2 +|x2(n)| 2 , address A of the second transmitted signal 13 =|x2(n)| 2 Generates.

[0162] Furthermore, the address generation unit 11 generates the combined power address A for the second transmission signal. 22 =|x1(n)| 2 +|x2(n)| 2 , address A of the first transmitted signal 23 =|x1(n)| 2 Generates.

[0163] Then, LUT212 is the address A of the synthesized signal. 12 LUT as the corresponding distortion compensation coefficient 1,2 (A 12 Outputs a LUT. 1,2 (A 12 ) is the distortion compensation coefficient for distortion compensation for the first transmitted signal with respect to the combined power. LUT213 is the address A of the second transmitted signal. 13 LUT as the corresponding distortion compensation coefficient 1,3 (A 13 Outputs a LUT. 1,3 (A 13 ) is the distortion compensation coefficient for distortion compensation for the power of the second transmitted signal relative to the first transmitted signal.

[0164] Furthermore, LUT222 has a combined power address A 22 LUT as the corresponding distortion compensation coefficient 2,2 (A 22 Outputs a LUT. 2,2 (A 22 ) is the distortion compensation coefficient for distortion compensation for the second transmitted signal with respect to the combined power. LUT223 is the address A of the first transmitted signal. 23 LUT as the corresponding distortion compensation coefficient 2,3 (A 23 Outputs a LUT. 2,3 (A 23 ) is the distortion compensation coefficient for distortion compensation for the power of the first transmitted signal relative to the second transmitted signal.

[0165] The adder 127 uses the LUT to calculate the distortion compensation coefficient of the first transmitted signal. SUM-1 =LUT 1,2 (A 12 )+LUT 1,3 (A13 The distortion compensation coefficient is calculated as follows. The radio 1 can use this distortion compensation coefficient to perform distortion compensation for the combined power and the power of the second transmitted signal relative to the first transmitted signal. The adder 128 also calculates the distortion compensation coefficient of the second transmitted signal as LUT SUM-2 =LUT 2,2 (A 22 )+LUT 2,3 (A 23 The distortion compensation coefficient is calculated as follows. The radio 1 can use this distortion compensation coefficient to perform distortion compensation for the combined power and the power of the first transmitted signal with respect to the second transmitted signal.

[0166] As described above, the radio 1 according to this embodiment performs distortion compensation for a specific band's transmission signal in multiband transmission using a distortion compensation coefficient for the combined power of the transmission signal and the transmission signals of other bands, and a distortion compensation coefficient for the power of the transmission signals of other bands. In this way, distortion compensation for each band is possible without using a distortion compensation coefficient for the power of the transmission signal itself in the distortion compensation of a specific band's transmission signal.

[0167] Furthermore, the configurations of Example 3, and the modified examples 3-1 and 3-2, may be used in combination with the configuration of Example 2. For example, for the combinations of delay amounts q11, q12, and q13 and delay amounts q21, q22, and q23 in Example 2, the tables to be used can be determined from LUTs 211-213 and 221-223 as follows.

[0168] For example, radio 1 uses LUTs 211-213 and 221-223 for the first combination. Radio 1 also uses two LUTs each from 211-213 and 221-223 for the second combination. Radio 1 also uses one LUT each from 211-213 and 221-223 for the third combination.

[0169] Alternatively, the tables can be determined as follows: For example, since the Memory Polynomial terms q11=q12=q13 and q21=q22=q23 have a relatively large impact on distortion compensation performance, radio 1 uses LUTs 211-213 and 221-223 for these delay combinations. Then, for other delay combinations, radio 1 uses two or one of LUTs 211-213 and 221-223.

[0170] In addition, combinations of delay amounts such as q11=q12, q12=q13, q13=q11, q21=q22, q22=q23, or q23=q21 have a relatively large impact on distortion compensation performance, similar to the Memory Polynomial term. Therefore, the radio 1 may use LUTs 211-213 and 221-223 for these delay amount combinations. The radio 1 may then use two or one of LUTs 211-213 and 221-223 for the remaining delay amount combinations. [Examples]

[0171] Next, we will describe Embodiment 4. Figure 7 is a block diagram of the radio according to Embodiment 4. The radio 1 according to this embodiment adaptively updates the distortion compensation coefficient using a feedback signal obtained by feeding back a portion of the amplifier output signal. In the following description, the operation of each part, which is the same as in Embodiment 1, may be omitted. The radio 1 according to this embodiment has a feedback unit 20 and a coefficient update unit 30.

[0172] The feedback unit 20 generates a feedback signal by feeding back a portion of the output signal of PA19. The feedback unit 20 includes a coupler 21, a multiplier 22, and an ADC (Analog Digital Converter) 23.

[0173] Coupler 21 takes a portion of the amplifier output signal from PA19 and outputs it to multiplier 22.

[0174] The multiplier 22 and local oscillator 18 are downconverters. The multiplier 22 receives the amplifier output signal as input from the coupler 21. The multiplier 22 then downconverts the amplifier output signal by multiplying it with the local signal input from the local oscillator 18, converting the frequencies of the first and second bands to the baseband or intermediate frequency, respectively. After that, the multiplier 22 outputs the downconverted amplifier output signal to the ADC 23.

[0175] The ADC23 receives the down-converted amplifier output signal as input from the multiplier 22. The ADC23 then converts the amplifier output signal from an analog signal to a digital signal. After that, the ADC23 outputs the converted digital amplifier output signal as a feedback signal to the coefficient update unit 30.

[0176] The coefficient update unit 30 updates the distortion compensation coefficient using a feedback signal. The coefficient update unit 30 includes frequency shift units 311 and 321, LPFs (Low Pass Filters) 312 and 322, and update units 313 and 323.

[0177] Frequency shifting units 311 and 321 each receive the input of the distributed feedback signal. Frequency shifting unit 311 then shifts the frequency of the feedback signal so that the signal of the first band in the feedback signal becomes the baseband signal. Frequency shifting unit 321 also shifts the frequency of the feedback signal so that the signal of the second band in the feedback signal becomes the baseband signal.

[0178] LPF312 and 322 are digital filters, such as FIR (Finite Impulse Response) filters. Both LPF312 and 322 are filters that allow signals with frequencies near the baseband to pass through. LPF312 passes a feedback signal that has been frequency-shifted by the frequency shifting unit 311 through it and outputs a feedback signal corresponding to the first transmitted signal of the first band. Similarly, LPF322 passes a feedback signal that has been frequency-shifted by the frequency shifting unit 321 through it and outputs a feedback signal corresponding to the second transmitted signal of the second band.

[0179] The update unit 313 receives the input of a feedback signal corresponding to the first transmission signal of the first band from the LPF 312. The update unit 313 also receives the input of the first transmission signal before distortion compensation, which is branched from the signal path P1. Furthermore, the update unit 313 receives the input of the distortion compensation coefficients of LUTs 211 to 213 before the update. Then, the update unit 313 calculates the new distortion compensation coefficients for each of LUTs 211 to 213 using the feedback signal corresponding to the first transmission signal, the first transmission signal, and the distortion compensation coefficients before the update. Subsequently, the update unit 313 updates the distortion compensation coefficients of each of LUTs 211 to 213 with the newly calculated distortion compensation coefficients.

[0180] The update unit 323 receives the input of a feedback signal corresponding to the second transmit signal of the second band from the LPF 322. The update unit 323 also receives the input of the second transmit signal before distortion compensation, which is branched from the signal path P2. Furthermore, the update unit 323 receives the input of the distortion compensation coefficients of LUTs 221 to 223 before the update. Then, the update unit 323 calculates the new distortion compensation coefficients for each of LUTs 221 to 223 using the feedback signal corresponding to the second transmit signal, the second transmit signal, and the distortion compensation coefficients before the update. Subsequently, the update unit 323 updates the distortion compensation coefficients of each of LUTs 221 to 223 with the calculated new distortion compensation coefficients.

[0181] The update units 313 and 323 can calculate new distortion compensation coefficients using algorithms such as the LMS (Least Mean Square) algorithm. For example, an example of distortion compensation coefficient calculation by the update units 313 and 323 is described below under the following conditions: The first transmitted signal is x1(n), and the second transmitted signal is x2(n). Also, the feedback signal for the first transmitted signal is y1(n), and the feedback signal for the second transmitted signal is y2(n). Also, the addresses input to LUTs 211~213 are A 11 ~A 13 And each of the addresses entered into LUT221~223 is A 21 ~A 23 This will be explained as follows.

[0182] For example, the update unit 313 updates the LUT 211 to a new strain compensation coefficient LUT 1,1 (A 11 ) to LUT 1,1 (A 11 )=LUT 1,1 (A 11 )+μ·e1(n)·(y1(n)) * It is calculated as follows: Here, μ is the step size parameter of the LMS algorithm, * The symbol '' represents the complex conjugate, and e1(n) is the error signal for the first transmitted signal, expressed as e1(n) = x1(n) - y1(n). Furthermore, the update unit 313 provides a new distortion compensation coefficient LUT for LUT212. 1,2 (A 12 ) to LUT 1,2 (A 12 )=LUT 1,2 (A 12 )+μ·e1(n)·(y1(n)) * It is calculated as follows. In addition, the update unit 313 is a new strain compensation coefficient LUT for LUT213. 1,3 (A 13 ) to LUT 1,3 (A 13 )=LUT 1,3 (A 13 )+μ·e1(n)·(y1(n)) * It is calculated as follows.

[0183] Furthermore, for example, the update unit 323 provides a new strain compensation coefficient LUT for LUT221. 2,1 (A 21 ) to LUT 2,1 (A 21 )=LUT 2,1 (A 21 ) + μ·e2(n)·(y2(n)) * It is calculated as follows. Here, e2(n) is the error signal for the second transmitted signal, and is expressed as e2(n) = x2(n) - y2(n). Also, the update unit 323 is a new distortion compensation coefficient LUT for LUT222. 2,2 (A 22 ) to LUT 2,2 (A 22 )=LUT 2,2 (A 22 ) + μ·e2(n)·(y2(n)) * It is calculated as follows. In addition, the update unit 323 is a new distortion compensation coefficient LUT for LUT223. 2,3 (A 23 ) to LUT 2,3 (A 23 )=LUT 2,3 (A 23 ) + μ·e2(n)·(y2(n)) * It is calculated as follows.

[0184] As described above, the radio 1 according to this embodiment adaptively updates the distortion compensation coefficient using a feedback signal obtained by feeding back a portion of the amplifier output signal from PA19. As a result, the radio 1 can improve the distortion compensation performance even when the nonlinear characteristics of PA19 change over time.

[0185] (Variation 4-1) Next, a modified example 4-1 of Example 4 will be described. The configuration of Example 4 can also be combined with the configuration of Example 2 or 3. For example, an example of using it in combination with the first to third examples of Example 2 will be described below.

[0186] In this case, the radio 1 has a delay addition unit 101 positioned between the power calculation unit 111 and the distortion compensation coefficient calculation unit 12. Furthermore, the radio 1 has a delay addition unit 102 positioned between the power calculation unit 112 and the distortion compensation coefficient calculation unit 12. Additionally, the radio 1 has a delay addition unit 134 positioned between the branching point to the power calculation unit 111 in the signal path P1 and the multiplier 131. Finally, the radio 1 has a delay addition unit 135 positioned between the branching point to the power calculation unit 112 in the signal path P2 and the multiplier 132.

[0187] When combined with the first example of Example 2, the power of the first transmitted signal in the first band |x1(n)| 2 The delay amount qpw1 and the power |x2(n)| of the second transmitted signal in the second band. 2 The delay amount qpw2 is the same as qpw1, that is, qpw1 = qpw2 = qpw. Here, similar to Example 2, the power of the first transmitted signal |x1(n)| in the first transmitted signal 2 The delay amount and the power of the second transmitted signal |x2(n)| 2 The delay amount qpw is set to a changeable delay amount q11. Also, the power |x1(n)| of the first transmitted signal in the second transmitted signal. 2 The delay amount and the power of the second transmitted signal |x2(n)| 2 The delay amount qpw is set to a changeable delay amount q21. Additionally, the delay amount qtx1 of the first transmitted signal is set to a changeable delay amount q12, and the delay amount qtx2 of the second transmitted signal is set to a changeable delay amount q22.

[0188] The update section 313 is a new strain compensation coefficient LUT for LUT211. 1,1,q11,q12 (A 11,q11 ) to LUT 1,1,q11,q12 (A 11,q11 )=LUT 1,1,q11,q12 (A 11,q11 )+μ·e1(n)·(y1(n-q12)) * It is calculated as follows. In addition, the update unit 313 is a new strain compensation coefficient LUT for LUT212. 1,2,q11,q12 (A 12,q11 ) to LUT 1,2,q11,q12 (A 12,q11 )=LUT 1,2,q11,q12 (A 12,q11)+μ·e1(n)·(y1(n-q12)) * It is calculated as follows. In addition, the update unit 313 is a new strain compensation coefficient LUT for LUT213. 1,3,q11,q12 (A 13,q11 ) to LUT 1,3,q11,q12 (A 13,q11 )=LUT 1,3,q11,q12 (A 13,q11 )+μ·e1(n)·(y1(n-q12)) * It is calculated as follows.

[0189] Furthermore, the update unit 323 provides a new distortion compensation coefficient LUT for LUT221. 2,1,q21,q22 (A 21,q21 ) to LUT 2,1,q21,q22 (A 21,q21 )=LUT 2,1,q21,q22 (A 21,q21 )+μ·e2(n)·(y2(n-q22)) * It is calculated as follows. In addition, the update unit 323 is a new distortion compensation coefficient LUT for LUT222. 2,2,q21,q22 (A 22,q21 ) to LUT 2,2,q21,q22 (A 22,q21 )=LUT 2,2,q21,q22 (A 22,q21 )+μ·e2(n)·(y2(n-q22)) * It is calculated as follows. In addition, the update unit 323 is a new distortion compensation coefficient LUT for LUT223. 2,3,q21,q22 (A 23,q21 ) to LUT 2,3,q21,q22 (A 23,q21 )=LUT 2,3,q21,q22 (A 23,q21 )+μ·e2(n)·(y2(n-q22)) * It is calculated as follows.

[0190] When combined with the second example of Example 2, the power of the first transmission signal |x1(n)| is determined for each of the first and second transmission signals. 2 The delay amount pqw1 and the power of the second transmitted signal |x2(n)| 2 The delay amount pqw2 is different. That is, for both the first and second transmitted signals, pqw1 ≠ pqw2. Here, the power of the first transmitted signal |x1(n)| in the first transmitted signal. 2Let q11 be the amount of delay in which the power of the second transmitted signal |x2(n)| changes, and let q13 be the amount of delay in which the power of the second transmitted signal |x2(n)| changes. Also, in the second transmitted signal, the power of the second transmitted signal |x2(n)| 2 Let q21 be the delay amount to be converted, and the power of the first transmitted signal |x1(n)| 2 Let q23 be the amount of delay that changes. Also, let q12 be the amount of delay qtx1 of the first transmitted signal, and let q22 be the amount of delay qtx2 of the second transmitted signal.

[0191] The update section 313 is a new strain compensation coefficient LUT for LUT211. 1,1,q11,q12,q13 (A 11,q11 ) to LUT 1,1,q11,q12,q13 (A 11,q11 )=LUT 1,1,q11,q12,q13 (A 11,q11 )+μ·e1(n)·(y1(n-q12)) * It is calculated as follows. The update unit 313 is a new distortion compensation coefficient LUT for LUT212. 1,2,q11,q12,q13 (A 12,q11,q13 ) to LUT 1,2,q11,q12,q13 (A 12,q11,q13 )=LUT 1,2,q11,q12,q13 (A 12,q11,q13 )+μ·e1(n)·(y1(n-q12)) * It is calculated as follows. The update unit 313 is a new distortion compensation coefficient LUT for LUT213. 1,3,q11,q12,q13 (A 13,q13 ) to LUT 1,3,q11,q12,q13 (A 13,q13 )=LUT 1,3,q11,q12,q13 (A 13,q13 )+μ·e1(n)·(y1(n-q12)) * It is calculated as follows.

[0192] Furthermore, the update unit 323 provides a new distortion compensation coefficient LUT for LUT221. 2,1,q21,q22,q23 (A 21,q21 ) to LUT 2,1,q21,q22,q23 (A 21,q21 )=LUT 2,1,q21,q22,q23 (A 21,q21 )+μ·e2(n)·(y2(n-q22)) * It is calculated as follows. The update unit 323 is a new distortion compensation coefficient LUT for LUT222. 2,2,q21,q22,q23 (A 22,q21,q23 ) to LUT2,2,q21,q22,q23 (A 22,q21,q23 )=LUT 2,2,q21,q22,q23 (A 22,q21,q23 )+μ·e2(n)·(y2(n-q22)) * It is calculated as follows. The update unit 323 is a new distortion compensation coefficient LUT for LUT223. 2,3,q21,q22,q23 (A 23,q23 ) to LUT 2,3,q21,q22,q23 (A 23,q23 )=LUT 2,3,q21,q22,q23 (A 23,q23 )+μ·e2(n)·(y2(n-q22)) * It is calculated as follows.

[0193] The case where it is combined with the third example of Example 2 will be described. In this case, the LUT when combined with the second example of Example 2 1,1,q11,q12,q13 (A 11,q11 ) does not include q13 as a variable, LUT 1,1,q11,q12 (A 11,q11 ) will be the result. Also, LUT 1,3,q11,q12,q13 (A 13,q13 ) does not include q11 as a variable, LUT 1,3,q12,q13 (A 13,q13 ) will be the result. Also, LUT 2,1,q21,q22,q23 (A 21,q21 ) does not include q23 as a variable, LUT 2,1,q21,q22 (A 21,q21 ) will be the result. Also, LUT 2,3,q21,q22,q23 (A 23,q23 ) does not include q21 as a variable, LUT 2,3,q22,q23 (A 23,q23 )

[0194] The update section 313 is a new strain compensation coefficient LUT for LUT211. 1,1,q11,q12 (A 11,q11 ) to LUT 1,1,q11,q12 (A 11,q11 )=LUT 1,1,q11,q12 (A 11,q11 )+μ·e1(n)·(y1(n-q12)) * It is calculated as follows. In addition, the update unit 313 is a new strain compensation coefficient LUT for LUT212. 1,2,q11,q12,q13 (A 12,q11,q13 ) to LUT 1,2,q11,q12,q13 (A 12,q11,q13 )=LUT1,2,q11,q12,q13 (A 12,q11,q13 )+μ·e1(n)·(y1(n-q12)) * It is calculated as follows. In addition, the update unit 313 is a new strain compensation coefficient LUT for LUT213. 1,3,q12,q13 (A 13,q13 ) to LUT 1,3,q12,q13 (A 13,q13 )=LUT 1,3,q12,q13 (A 13,q13 )+μ·e1(n)·(y1(n-q12)) * It is calculated as follows.

[0195] Furthermore, the update unit 323 provides a new distortion compensation coefficient LUT for LUT221. 2,1,q21,q22 (A 21,q21 ) to LUT 2,1,q21,q22 (A 21,q21 )=LUT 2,1,q21,q22 (A 21,q21 )+μ·e2(n)·(y2(n-q22)) * It is calculated as follows. In addition, the update unit 323 is a new distortion compensation coefficient LUT for LUT222. 2,2,q21,q22,q23 (A 22,q21,q23 ) to LUT 2,2,q21,q22,q23 (A 22,q21,q23 )=LUT 2,2,q21,q22,q23 (A 22,q21,q23 )+μ·e2(n)·(y2(n-q22)) * It is calculated as follows. In addition, the update unit 323 is a new distortion compensation coefficient LUT for LUT223. 2,3,q22,q23 (A 23,q23 ) to LUT 2,3,q22,q23 (A 23,q23 )=LUT 2,3,q22,q23 (A 23,q23 )+μ·e2(n)·(y2(n-q22)) * It is calculated as follows.

[0196] (Modification 4-2) Next, a modified example 4-2 of Embodiment 4 will be described. Figure 8 is a block diagram of the radio according to modified example 4-2. The radio 1 according to this modified example further includes frequency shift units 314 and 324, and adders 315 and 325. In this case, the radio 1 does not need to have LPFs 312 and 322.

[0197] The frequency shift unit 314 receives the input of the second transmission signal, which is branched from the signal path P2 and has not yet been compensated for distortion. The frequency shift unit 314 then applies the same frequency shift to the second transmission signal as it did during signal transmission. After that, the frequency shift unit 314 outputs the inverted signal of the second transmission signal with the frequency shift applied to it to the adder 315.

[0198] The frequency shift unit 324 receives the input of the first transmission signal before distortion compensation, which is branched from the signal path P1. The frequency shift unit 324 then applies the same frequency shift to the first transmission signal as when the signal was transmitted. After that, the frequency shift unit 324 outputs the inverted signal of the first transmission signal with the frequency shift applied to it to the adder 325.

[0199] The adder 315 receives the input of the distributed feedback signal. The adder 315 also receives the input of the inverted signal of the frequency-shifted second transmission signal from the frequency shift unit 314. The adder 315 then adds the feedback signal and the inverted signal of the frequency-shifted second transmission signal to remove the component of the second transmission signal from the feedback signal and generate a feedback signal corresponding to the first transmission signal. After that, the adder 315 outputs the feedback signal corresponding to the first transmission signal to the frequency shift unit 311.

[0200] The adder 325 receives the input of the distributed feedback signal. The adder 325 also receives the input of the inverted signal of the frequency-shifted first transmission signal from the frequency shift unit 324. The adder 325 then adds the feedback signal and the inverted signal of the frequency-shifted first transmission signal to remove the component of the first transmission signal from the feedback signal and generate a feedback signal corresponding to the second transmission signal. After that, the adder 325 outputs the feedback signal corresponding to the second transmission signal to the frequency shift unit 321.

[0201] As described above, the radio 1 according to this embodiment extracts band-specific feedback signals by subtracting the transmission signals of other bands that have been frequency-shifted after the feedback signal has been distributed, instead of using frequency shifting and filtering. In this way, the radio 1 can generate band-specific feedback signals even by subtracting the transmission signals of other bands that have been frequency-shifted, and can improve distortion compensation performance even when the nonlinear characteristics of PA19 change over time.

[0202] In the feedback signals for each band, a particular band may contain nonlinear distortion components from other bands. However, linear distortion components from other bands are generally uncorrelated with the transmitted signal or nonlinear distortion of the particular band and can therefore be considered noise. Thus, by repeatedly updating the distortion compensation coefficient, the distortion compensation coefficient is averaged, allowing the radio 1 in this embodiment to neutralize the influence of nonlinear distortion components from other bands on a particular band.

[0203] (Modification 4-3) Next, a modification 4-3 of Example 4 will be described. Figure 9 is a block diagram of the radio according to modification 4-3. The radio 1 according to this modification has selectors 301 and 302. In this case, the radio 1 does not need to have a frequency shift unit 321, an LPF 322, and an update unit 323. The radio 1 according to this modification updates the distortion compensation coefficient for each band exclusively in a time-division manner.

[0204] The frequency shift unit 311 receives the feedback signal without distribution. The frequency shift unit 311 then changes the frequency to be shifted at predetermined timings, thereby outputting either the feedback signal of the first transmission signal in the first band or the feedback signal of the second transmission signal in the second band in a time-division multiplexer.

[0205] Selector 302 receives inputs of the first transmission signal before distortion compensation, which is branched from signal path P1, and the second transmission signal before distortion compensation, which is branched from signal path P2. Selector 302 then selects and outputs the first transmission signal in accordance with the timing of the output of the feedback signal of the first transmission signal from frequency shift unit 311, and selects and outputs the second transmission signal in accordance with the timing of the output of the feedback signal of the second transmission signal.

[0206] When the frequency shift unit 311 outputs a feedback signal for the first transmission signal, the selector 301 connects the update unit 313 to LUTs 211-213. Also, when the frequency shift unit 311 outputs a feedback signal for the second transmission signal, the selector 301 connects the update unit 313 to LUTs 221-223.

[0207] When the frequency shift unit 311 outputs a feedback signal of the first transmission signal, the update unit 313 calculates new distortion compensation coefficients for LUTs 211 to 213. Then, the update unit 313 updates the distortion compensation coefficients for LUTs 211 to 213 with the calculated distortion compensation coefficients.

[0208] Furthermore, when a feedback signal of the second transmission signal is output from the frequency shift unit 311, the update unit 313 calculates new distortion compensation coefficients for LUTs 221 to 223. Then, the update unit 313 updates the distortion compensation coefficients for LUTs 221 to 223 with the calculated distortion compensation coefficients.

[0209] As described above, the coefficient update unit 30 in this embodiment switches the band to be updated for the distortion compensation coefficient at a predetermined timing, and in response to this switching, switches the frequency to be shifted by the frequency shift unit 311 and the settings of selectors 301 and 302. As a result, the radio 1 does not need to be equipped with a frequency shift unit 321, an LPF 322, and an update unit 323, and the circuit size of the digital section can be reduced. [Examples]

[0210] Figure 10 is a block diagram of the radio according to Embodiment 5. The radio 1 according to this embodiment performs delay adjustment to the transmitted signal and various adjustments to the feedback signal during the distortion compensation coefficient update process. As shown in Figure 10, the radio 1 according to this embodiment has an adjustment unit 24 in the feedback unit 20, and adjustment units 316 and 326, as well as delay adjustment units 317 and 327 in the coefficient update unit 30.

[0211] The adjustment unit 24 compares the combined transmission signal, which is obtained by frequency shifting the first and second transmission signals respectively and then combining them, with the feedback signal output from the ADC 23. Here, in Figure 10, the input path of the combined transmission signal to the adjustment unit 24 is omitted for clarity, but the adjustment unit 24 can, for example, obtain the combined transmission signal from the adder 15. Then, the adjustment unit 24 performs various adjustments on the feedback signal according to the comparison result. For example, the adjustment unit 24 can perform all or a combination of gain adjustment, phase adjustment, delay adjustment, and frequency characteristic correction.

[0212] Furthermore, when comparing the first transmission signal and the feedback signal of the first transmission signal in the time domain, the feedback signal is delayed compared to the transmission signal due to the time it takes to pass through the path to PA19 in the analog section and the feedback path. The same applies to the second transmission signal. Therefore, the adjustment units 316 and 326, and the delay adjustment units 317 and 327 perform delay adjustments as follows to synchronize the timing of the first transmission signal and the feedback signal of the first transmission signal, and the timing of the second transmission signal and the feedback signal of the second transmission signal.

[0213] The adjustment unit 316 compares the first transmission signal of the first band with the feedback signal of the first transmission signal and performs delay adjustment on the feedback signal of the first transmission signal to synchronize the timing. Similarly, the adjustment unit 326 compares the second transmission signal of the second band with the feedback signal of the second transmission signal and performs delay adjustment on the feedback signal of the second transmission signal to synchronize the timing. In Figure 10, for the sake of clarity, the input path of the first transmission signal to the adjustment unit 316 is omitted, but the adjustment unit 316 can receive the first transmission signal from the update unit 313 and the delay adjustment unit 317. The same applies to the adjustment unit 326. In addition, the adjustment units 316 and 326 may perform all or part of gain adjustment, phase adjustment, and frequency characteristic adjustment.

[0214] Furthermore, the delay adjustment unit 317 compares the first transmission signal of the first band with the feedback signal of the first transmission signal and performs delay adjustment on the first transmission signal to synchronize the timing. Similarly, the delay adjustment unit 327 compares the second transmission signal of the second band with the feedback signal of the second transmission signal and performs delay adjustment on the second transmission signal to synchronize the timing. Here, in Figure 10, for the sake of clarity, the input path of the feedback signal of the first transmission signal to the delay adjustment unit 317 has been omitted, but the delay adjustment unit 317 can receive the feedback signal of the first transmission signal from the update unit 313 and the adjustment unit 316. The same applies to the delay adjustment unit 327.

[0215] Furthermore, the coefficient update unit 30 may adjust the delay amount for the first transmission signal and the feedback signal of the first transmission signal with different resolutions. Similarly, the coefficient update unit 30 may adjust the delay amount for the second transmission signal and the feedback signal of the second transmission signal with different resolutions.

[0216] The delay adjustment unit 317 performs delay adjustment for the first transmission signal, for example, in units of sample intervals corresponding to the sampling rate of the digital section. Similarly, the delay adjustment unit 327 performs delay adjustment for the second transmission signal, for example, in units of sample intervals corresponding to the sampling rate of the digital section. The delay adjustment units 317 and 327, which perform delay adjustment for the first and second transmission signals, perform delay adjustment with high resolution.

[0217] The adjustment unit 316 performs delay adjustment in units smaller than the sample interval corresponding to the sampling rate of the digital section, for example, as delay adjustment for the feedback signal of the first transmission signal. The adjustment unit 316 may perform delay adjustment using, for example, an FIR filter. The adjustment unit 316 may also perform frequency characteristic correction using an FIR filter or the like.

[0218] The adjustment unit 326 performs delay adjustment in units smaller than the sample interval corresponding to the sampling rate of the digital section, for example, as delay adjustment for the feedback signal of the second transmission signal. The adjustment unit 326 may perform delay adjustment using, for example, an FIR filter. The adjustment unit 326 may also perform frequency characteristic correction using an FIR filter or the like.

[0219] Here, an FIR filter can also be used as LPF312, in which case the FIR filter of LPF312 and the FIR filter for adjusting the delay of the feedback signal of the first transmitted signal in the adjustment unit 316 to less than a clock unit can be integrated. In other words, one FIR filter can perform both the function of adjusting the delay of the feedback signal of the first transmitted signal to less than a clock unit and the function of frequency bandwidth limiting. The same applies to LPF322 and adjustment unit 326.

[0220] In this example, the delay adjustment units 317 and 327 perform delay adjustment with a large resolution, while the adjustment units 316 and 326 perform delay adjustment in fine units; however, the reverse may also be true. Furthermore, the radio 1 may have all of the adjustment units 316 and 326, as well as the delay adjustment units 317 and 327, or it may have only some of them. For example, the delay adjustment units 317 and 327 may perform both clock-unit delay adjustment and subclock-unit delay adjustment for the first and second transmission signals, while omitting delay adjustment for the feedback signal.

[0221] Here, the delay added by the adjustment unit 316 to the feedback signal of the first transmission signal and the delay added by the adjustment unit 326 to the feedback signal of the second transmission signal are examples of the "third delay". Also, the delay added by the delay adjustment unit 317 to the first transmission signal and the delay added by the delay adjustment unit 327 to the second transmission signal are examples of the "fourth delay". In other words, the coefficient update unit 30 adds the third delay to the feedback signal and / or the fourth delay to the first and second signals, adjusts the timing of the feedback signal and the first and second signals, compares them, and updates the first distortion compensation coefficient.

[0222] Furthermore, the configuration of Embodiment 5 described above can also be combined with the configuration of Embodiment 2. For example, the adjustment unit 316 or delay adjustment unit 317 compares the first transmission signal of the first band with the feedback signal of the first transmission signal to calculate the delay amount of the first transmission signal. The adjustment unit 326 or delay adjustment unit 327 compares the second transmission signal of the second band with the feedback signal of the second transmission signal to calculate the delay amount of the second transmission signal obtained from that comparison. Then, the delay addition units 101 and 134 are given a delay equal to the difference between the delay amounts of the first transmission signal and the delay amount of the second transmission signal, which corresponds to the difference in delay amounts between the bands, thereby adjusting the delay amounts between the bands. Conversely, the delay addition units 102 and 135 are given a delay equal to the difference between the delay amounts of the first transmission signal and the delay amount of the second transmission signal, which corresponds to the difference in delay amounts between the bands, thereby adjusting the delay amounts between the bands. As a result, the distortion compensation performance of the radio 1 is improved even when there is a difference in delay amounts between the bands.

[0223] As described above, the wireless device 1 according to this embodiment performs delay adjustment for the transmitted signal and various adjustments for the feedback signal. As a result, the wireless device 1 can calculate a more optimal distortion compensation coefficient, and its distortion compensation performance is improved. [Examples]

[0224] In each of the above embodiments, the radio 1 frequency-shifts the pre-distortion signals of the first band and the second band, combines them, and converts the combined pre-distortion signal from digital to analog. Then, the radio 1 converts the combined pre-distortion signal, which has been converted to analog, to a radio frequency using an upconverter and performs power amplification. However, the radio 1 is not limited to this configuration.

[0225] Figure 11 is a block diagram of the radio according to Embodiment 6. In this embodiment, instead of combining the pre-distortion signals of the first and second bands in the digital section, the radio 1 combines the pre-distortion signals of the first and second bands in the analog section. As shown in Figure 11, the radio 1 according to this embodiment has a DAC 161 for the first transmission signal, an upconverter 171 and a local oscillator 181 for the first transmission signal, and a DAC 162 for the second transmission signal, an upconverter 172 and a local oscillator 182 for the second transmission signal. In the following description, the operation of each part, which is the same as in Embodiment 1, may be omitted.

[0226] DAC161 receives the pre-distortion signal of the first transmission signal as input from the multiplier 131. Then, DAC161 converts the pre-distortion signal of the first transmission signal from a digital signal to an analog signal and outputs it to the upconverter 171.

[0227] The local oscillator 181 controls the frequency f, which is the frequency for the first band. L A local signal is generated. The local oscillator 181 then has a frequency f LThe local signal is output to the upconverter 171.

[0228] The upconverter 171 receives the pre-distortion signal of the first transmission signal as input from the DAC 161. Then, the upconverter 171 applies frequency f to the pre-distortion signal of the first transmission signal. L By multiplying and upconverting the local signal, the pre-distortion signal of the first transmitted signal is converted to a radio frequency.

[0229] DAC162 receives the pre-distortion signal of the second transmission signal as input from multiplier 132. Then, DAC162 converts the pre-distortion signal of the second transmission signal from a digital signal to an analog signal and outputs it to upconverter 172.

[0230] The local oscillator 182 controls the frequency f, which is the frequency for the second band. H A local signal is generated. The local oscillator 182 then has a frequency f H The local signal is output to the upconverter 172.

[0231] The upconverter 172 receives the pre-distortion signal of the second transmission signal as input from the DAC 162. Then, the upconverter 172 applies frequency f to the pre-distortion signal of the second transmission signal. H By multiplying and upconverting the local signal, the pre-distortion signal of the second transmission signal is converted to a radio frequency.

[0232] The adder 15 has a frequency f L The pre-distortion signal input of the first transmitted signal, which has been converted to f, is received from the upconverter 171. The adder 15 also receives the frequency f H The pre-distortion signal of the second transmission signal, which has been converted to f, is received as input from the upconverter 172. Then, the adder 15 processes the frequency f L The pre-distortion signal and frequency f of the first transmitted signal converted to HThe pre-distortion signals of the second transmitted signal, which have been converted, are added and combined. The adder 15 then outputs the combined pre-distortion signal to PA19.

[0233] PA19 amplifies the combined pre-distortion signal and radiates it into space as radio waves from the antenna.

[0234] As described above, the radio 1 according to this embodiment synthesizes pre-distortion signals for each band after converting to an analog signal. Even with this configuration, the radio 1 can improve distortion compensation performance and improve ACLR in multiband transmission.

[0235] (Extreme Variation 6-1) Figure 12 is a block diagram of a radio according to modified example 6-1. The radio 1 according to this embodiment synthesizes pre-distortion signals for each band after conversion to an analog signal, and updates the distortion compensation coefficient using the feedback signals of the first and second transmission signals. As shown in Figure 12, the radio 1 according to this embodiment has a feedback unit 20 and a coefficient update unit 30.

[0236] The feedback unit 20 includes a coupler 21, BPFs 215 and 225, multipliers 216 and 226, and ADCs 217 and 227. The coefficient update unit 30 also includes update units 313 and 323.

[0237] A portion of the amplifier output signal from PA19 is extracted and distributed by coupler 21. The distributed amplifier output signals are then input to BPF (Band Pass Filter) 251 and 225, respectively.

[0238] The BPF215 and BPF225 are analog filters. The BPF215 receives the amplifier output signal as input and passes signals in the first frequency band. The BPF225 receives the amplifier output signal as input and passes signals in the second frequency band.

[0239] Multiplier 216 and local oscillator 181 are downconverters. Multiplier 226 and local oscillator 182 are also downconverters. Multiplier 216 downconverts the signal that has passed through BPF215 and outputs it. Multiplier 226 downconverts the signal that has passed through BPF225 and outputs it. Through downconversion, each signal is converted to either the baseband or intermediate frequency.

[0240] ADC217 converts the signal down-converted by multiplier 216 from an analog signal to a digital signal and outputs it as a feedback signal for the first transmission signal. ADC227 converts the signal down-converted by multiplier 226 from an analog signal to a digital signal and outputs it as a feedback signal for the second transmission signal.

[0241] The update unit 313 calculates new distortion compensation coefficients for LUTs 211 to 213 using the first transmission signal, the feedback signal of the first transmission signal, and the distortion compensation coefficients of LUTs 211 to 213 before updating. Then, the update unit 313 updates the distortion compensation coefficients of LUTs 211 to 213 to the calculated distortion compensation coefficients. The update unit 323 calculates new distortion compensation coefficients for LUTs 221 to 223 using the second transmission signal, the feedback signal of the second transmission signal, and the distortion compensation coefficients of LUTs 221 to 223 before updating. Then, the update unit 323 updates the distortion compensation coefficients of LUTs 221 to 223 to the calculated distortion compensation coefficients.

[0242] As described above, the radio 1 according to this embodiment synthesizes pre-distortion signals for each band after conversion to an analog signal and updates the distortion compensation coefficient using a feedback signal. As a result, the radio 1 can perform more appropriate distortion compensation in a configuration in which pre-distortion signals for each band are synthesized after conversion to an analog signal.

[0243] (Variation 6-2) Figure 13 is a block diagram of a radio according to modified example 6-2. The radio 1 according to this embodiment synthesizes pre-distortion signals for each band after conversion to an analog signal, and updates the distortion compensation coefficient using the feedback signals of the first and second transmission signals while switching between the first and second bands. As shown in Figure 13, the radio 1 according to this embodiment has a feedback unit 20 and a coefficient update unit 30.

[0244] The BPF215 receives the amplifier output signal as input and, at a predetermined timing, switches between the frequency bands of the first and second bands to allow the signal to pass through.

[0245] The selector 27, at a predetermined timing, selects the frequency f of the first band from the local oscillator 181. L The local signal or the frequency f of the second band from local oscillator 182 H Select the local signal and send it to the multiplier 216. The multiplier 216 receives the first band frequency f at a predetermined timing. L The local signal or the frequency f of the second band H The signal is down-converted using the local signal to convert it to baseband or intermediate frequency.

[0246] The ADC217 converts the down-converted signal from an analog signal to a digital signal and outputs it as a feedback signal.

[0247] The selector 302 switches between the first transmission signal obtained from signal path P1 and the second transmission signal obtained from signal path P2 at a predetermined timing and outputs them to the update unit 313.

[0248] The selector 301 switches the connection destination of the update unit 313 to either LUT211~213 or LUT221~223 at a predetermined timing.

[0249] The update unit 313 updates the distortion compensation coefficients of LUTs 211-213 or LUTs 221-223 using the first or second transmission signal from the selector 302 and the feedback signal of the first or second transmission signal from the ADC 217.

[0250] As described above, the radio 1 according to this embodiment synthesizes pre-distortion signals for each band after conversion to an analog signal, and updates the distortion compensation coefficient using a feedback signal while switching the target band. As a result, the radio 1, in a configuration that synthesizes pre-distortion signals for each band after conversion to an analog signal and updates the distortion compensation coefficient, can reduce the circuit size of the digital section. [Examples]

[0251] In the above embodiments, the case of two bands as multiband was described as an example, but the radio 1 can also perform multiband transmission of three or more bands. In this embodiment, the radio 1 performs multiband transmission using three bands. Figure 14 is a block diagram of the radio according to Embodiment 7.

[0252] The address generation unit 11 includes power calculation units 111-113 and adders 121-123. The distortion compensation coefficient calculation unit 12 includes LUTs 411-416, 421-426, 431-436, adders 127-129, and multipliers 131-133. The radio 1 also includes frequency shift units 141-143.

[0253] The first transmission signal x1(n) of the first band is input to signal path P1. The second transmission signal x2(n) of the second band is input to signal path P2. The third transmission signal x3(n) of the third band is input to signal path P3.

[0254] The address generation unit 11 uses the power calculation units 111 to 113 to generate an address A corresponding to the power of the first transmission signal itself for the first transmission signal. 11The address generation unit 11 also uses the power calculation units 111 to 113 to generate an address A corresponding to the power of the second transmission signal for the first transmission signal. 14 The address generation unit 11 also uses the power calculation units 111 to 113 to generate an address A corresponding to the power of the third transmission signal for the first transmission signal. 15 The address generation unit 11 also uses the power calculation units 111 to 113 and the adders 121 to 123 to generate an address A corresponding to the combined power of the first and second transmission signals for the first transmission signal. 12 The address generation unit 11 also uses the power calculation units 111 to 113 and the adders 121 to 123 to generate an address A corresponding to the combined power of the first and third transmission signals for the first transmission signal. 13 The address generation unit 11 also uses the power calculation units 111-113 and adders 121-123 to generate an address A corresponding to the combined power of the second and third transmission signals for the first transmission signal. 16 Generates.

[0255] Here, A 11 =|x1(n)| 2 That is. Also, A 12 =|x1(n)| 2 +|x2(n)| 2 A 13 =|x1(n)| 2 +|x3(n)| 2 That is. Also, A 14 =|x2(n)| 2 That is. Also, A 15 =|x3(n)| 2 That is. Also, A 16 =|x2(n)| 2 +|x3(n)| 2 That is the case.

[0256] Furthermore, the address generation unit 11 uses the power calculation units 111 to 113 to generate an address A corresponding to the power of the second transmission signal itself for the second transmission signal. 21 The address generation unit 11 also uses the power calculation units 111 to 113 to generate an address A corresponding to the power of the first transmission signal for the second transmission signal.24 The address generation unit 11 also uses the power calculation units 111 to 113 to generate an address A corresponding to the power of the third transmission signal for the second transmission signal. 25 The address generation unit 11 also uses the power calculation units 111 to 113 and the adders 121 to 123 to generate an address A corresponding to the combined power of the first and second transmission signals for the second transmission signal. 22 The address generation unit 11 also uses the power calculation units 111 to 113 and the adders 121 to 123 to generate an address A corresponding to the combined power of the second and third transmission signals for the second transmission signal. 23 The address generation unit 11 also uses the power calculation units 111 to 113 and the adders 121 to 123 to generate an address A corresponding to the combined power of the first and third transmission signals for the second transmission signal. 26 Generates.

[0257] Here, A 21 =|x2(n)| 2 That is. Also, A 22 =|x1(n)| 2 +|x2(n)| 2 A 23 =|x2(n)| 2 +|x3(n)| 2 That is. Also, A 24 =|x1(n)| 2 That is. Also, A 25 =|x3(n)| 2 That is. Also, A 26 =|x1(n)| 2 +|x3(n)| 2 That is the case.

[0258] Furthermore, the address generation unit 11 uses the power calculation units 111 to 113 to generate an address A corresponding to the power of the third transmission signal itself for the third transmission signal. 31 The address generation unit 11 also uses the power calculation units 111 to 113 to generate an address A corresponding to the power of the first transmission signal for the third transmission signal. 34The address generation unit 11 also uses the power calculation units 111 to 113 to generate an address A corresponding to the power of the second transmission signal for the third transmission signal. 35 The address generation unit 11 also uses the power calculation units 111 to 113 and the adders 121 to 123 to generate an address A corresponding to the combined power of the first and third transmission signals for the third transmission signal. 32 The address generation unit 11 also uses the power calculation units 111 to 113 and the adders 121 to 123 to generate an address A corresponding to the combined power of the second and third transmission signals for the third transmission signal. 33 The address generation unit 11 also uses the power calculation units 111 to 113 and the adders 121 to 123 to generate an address A corresponding to the combined power of the first and second transmission signals for the third transmission signal. 36 Generates.

[0259] Here, A 31 =|x3(n)| 2 That is. Also, A 32 =|x1(n)| 2 +|x3(n)| 2 A 33 =|x2(n)| 2 +|x3(n)| 2 That is. Also, A 34 =|x1(n)| 2 That is. Also, A 35 =|x2(n)| 2 That is. Also, A 36 =|x1(n)| 2 +|x2(n)| 2 That is the case.

[0260] LUT411 is a table in which distortion compensation coefficients for distortion compensation to the first transmitted signal regarding the power of the first transmitted signal itself are stored, associated with the address of the first transmitted signal. 11 For the input, the distortion compensation coefficient LUT 1,1 (A 11) outputs. LUT414 is a table in which distortion compensation coefficients for distortion compensation for the power of the second transmitted signal relative to the first transmitted signal are stored and associated with the address of the second transmitted signal. LUT414 outputs ). 14 For the input, the distortion compensation coefficient LUT 1,4 (A 14 ) outputs. LUT415 is a table in which distortion compensation coefficients for distortion compensation for the power of the third transmitted signal relative to the first transmitted signal are stored and associated with the address of the third transmitted signal. LUT415 outputs ). 15 For the input, the distortion compensation coefficient LUT 1,5 (A 15 Outputs ).

[0261] LUT412 is a table in which distortion compensation coefficients for distortion compensation of the first transmitted signal for the combined power of the first and second transmitted signals are stored, associated with the address of the combined power. 12 For the input, the distortion compensation coefficient LUT 1,2 (A 12 ) outputs. LUT413 is a table in which distortion compensation coefficients for distortion compensation for the first transmit signal for the combined power of the first and third transmit signals are stored, associated with the address of the combined power. LUT413 outputs ). 13 For the input, the distortion compensation coefficient LUT 1,3 (A 13 ) outputs. LUT416 is a table in which distortion compensation coefficients for distortion compensation for the first transmit signal for the combined power of the second and third transmit signals are stored, associated with the address of the combined power. LUT416 outputs ). 16 For the input, the distortion compensation coefficient LUT 1,6 (A 16 Outputs ).

[0262] LUT421 is a table in which distortion compensation coefficients for distortion compensation to the second transmit signal regarding the power of the second transmit signal itself are stored, associated with the address of the second transmit signal. 21For the input, the distortion compensation coefficient LUT 2,1 (A 21 ) outputs. LUT424 is a table in which distortion compensation coefficients for distortion compensation for the second transmit signal with respect to the power of the first transmit signal are stored and associated with the address of the first transmit signal. LUT424 outputs ). 24 For the input, the distortion compensation coefficient LUT 2,4 (A 24 ) outputs. LUT425 is a table in which distortion compensation coefficients for distortion compensation for the power of the third transmitted signal relative to the second transmitted signal are stored and associated with the address of the third transmitted signal. LUT425 outputs ). 25 For the input, the distortion compensation coefficient LUT 2,5 (A 25 Outputs ).

[0263] LUT422 is a table in which distortion compensation coefficients for distortion compensation of the second transmit signal for the combined power of the first and second transmit signals are stored, associated with the address of the combined power. 22 For the input, the distortion compensation coefficient LUT 2,2 (A 22 ) outputs. LUT423 is a table in which distortion compensation coefficients for distortion compensation to the second transmit signal for the combined power of the second and third transmit signals are stored, associated with the address of the combined power. LUT423 outputs ). 23 For the input, the distortion compensation coefficient LUT 2,3 (A 23 ) outputs. LUT426 is a table in which distortion compensation coefficients for distortion compensation for the second transmit signal for the combined power of the first and third transmit signals are stored, associated with the address of the combined power. LUT426 outputs ). 26 For the input, the distortion compensation coefficient LUT 2,6 (A 26 Outputs ).

[0264] LUT431 is a table in which distortion compensation coefficients for distortion compensation to the third transmit signal regarding the power of the third transmit signal itself are stored, associated with the address of the third transmit signal. 31 For the input, the distortion compensation coefficient LUT 3,1 (A 31 ) outputs. LUT434 is a table in which distortion compensation coefficients for distortion compensation for the third transmit signal with respect to the power of the first transmit signal are stored and associated with the address of the first transmit signal. LUT434 outputs ). 34 For the input, the distortion compensation coefficient LUT 3,4 (A 34 ) outputs. LUT435 is a table in which distortion compensation coefficients for distortion compensation for the third transmit signal with respect to the power of the second transmit signal are stored and associated with the address of the second transmit signal. LUT435 outputs ). 35 For the input, the distortion compensation coefficient LUT 3,5 (A 35 Outputs ).

[0265] LUT432 is a table in which distortion compensation coefficients for distortion compensation of the third transmit signal for the combined power of the first and third transmit signals are stored, associated with the address of the combined power. 32 For the input, the distortion compensation coefficient LUT 3,2 (A 32 ) outputs. LUT433 is a table in which distortion compensation coefficients for distortion compensation for the third transmit signal for the combined power of the second and third transmit signals are stored, associated with the address of the combined power. LUT433 outputs ). 33 For the input, the distortion compensation coefficient LUT 3,3 (A 33 ) outputs. LUT436 is a table in which distortion compensation coefficients for distortion compensation for the third transmit signal for the combined power of the first and second transmit signals are stored, associated with the address of the combined power. LUT436 outputs ). 36 For the input, the distortion compensation coefficient LUT 3,6 (A 36 Outputs ).

[0266] The adder 127 uses the LUT to calculate the distortion compensation coefficient of the first transmitted signal. SUM-1 =LUT 1,1 (A 11 )+LUT 1,2 (A 12 )+LUT 1,3 (A 13 )+LUT 1,4 (A 14 )+LUT 1,5 (A 15 )+LUT 1,6 (A 16 The LUT calculates the distortion compensation coefficient of the second transmitted signal. SUM-2 =LUT 2,1 (A 21 )+LUT 2,2 (A 22 )+LUT 2,3 (A 23 )+LUT 2,4 (A 24 )+LUT 2,5 (A 25 )+LUT 2,6 (A 26 The LUT calculates the distortion compensation coefficient of the third transmitted signal. SUM-3 =LUT 3,1 (A 31 )+LUT 3,2 (A 32 )+LUT 3,3 (A 33 )+LUT 3,4 (A 34 )+LUT 3,5 (A 35 )+LUT 3,6 (A 36 It is calculated as follows:

[0267] The multiplier 131 receives the distortion compensation coefficient LUT input from the adder 127. SUM-1 Multiply this by the first transmitted signal x1(n) to calculate the pre-distortion signal u1(n) of the first transmitted signal in the first band. That is, u1(n) = x1(n)·LUT SUM-1 That is the case.

[0268] The multiplier 132 receives the distortion compensation coefficient LUT input from the adder 128. SUM-2 Multiply this by the second transmitted signal x2(n) to calculate the pre-distortion signal u2(n) of the second transmitted signal in the second band. That is, u2(n) = x2(n)·LUT SUM-2 That is the case.

[0269] The multiplier 133 receives the distortion compensation coefficient LUT input from the adder 129. SUM-3 Multiply this by the third transmission signal x3(n) to calculate the pre-distortion signal u3(n) of the third transmission signal in the third band. That is, u3(n) = x3(n)·LUT SUM-3 That is the case.

[0270] Although this explanation describes the case using three bands, radio 1 can perform distortion compensation in the same way when using four or more bands.

[0271] (Variation 7-1) Next, a modified example 7-1 of Example 7 will be described. The configuration of Example 7 can also be combined with the configuration of Example 2. For example, an example of using it in combination with the first to third examples of Example 2 will be described below. In this case, the radio 1 has a delay addition unit 101 between the power calculation unit 111 and the distortion compensation coefficient calculation unit 12. The radio 1 also has a delay addition unit 102 between the power calculation unit 112 and the distortion compensation coefficient calculation unit 12. The radio 1 also has a delay addition unit 103 between the power calculation unit 113 and the distortion compensation coefficient calculation unit 12. The radio 1 also has a delay addition unit 134 between the branching point to the power calculation unit 111 in the signal path P1 and the multiplier 131. The radio 1 also has a delay addition unit 135 between the branching point to the power calculation unit 112 in the signal path P2 and the multiplier 132. Furthermore, in the radio 1, a delay addition unit 136 is positioned between the branching point to the power calculation unit 113 in the signal path P3 and the multiplier 133.

[0272] The case where it is combined with the first example of Example 2 will be explained. For each of the first to third transmission signals, the power of the first transmission signal in the first band |x1(n)| 2The delay amount qpw1 and the power |x2(n)| of the second transmitted signal in the second band. 2 The delay amount qpw2 and the power |x3(n)| of the third transmitted signal in the third band. 2 The delay amount qpw3 is the same. That is, qpw1 = qpw2 = qpw3 = qpw. Here, the power of the first transmitted signal |x1(n)| in the first transmitted signal. 2 The delay amount, the power of the second transmitted signal |x2(n)| 2 The delay amount, the power of the third transmitted signal |x3(n)| 2 Let the delay amount be a changeable delay amount q11. Also, the power |x1(n)| of the first transmitted signal in the second transmitted signal. 2 The delay amount, the power of the second transmitted signal |x2(n)| 2 The delay amount, the power of the third transmitted signal |x3(n)| 2 Let the delay amount be a changeable delay amount q21. Also, the power |x1(n)| of the first transmitted signal in the third transmitted signal. 2 The delay amount, the power of the second transmitted signal |x2(n)| 2 The delay amount, the power of the third transmitted signal |x3(n)| 2 The delay amount of the first transmission signal is set to a changeable delay amount q31. Furthermore, the delay amount qtx1 of the first transmission signal is set to a changeable delay amount q12, the delay amount qtx2 of the second transmission signal is set to a changeable delay amount q22, and the delay amount qtx3 of the third transmission signal is set to a changeable delay amount q32.

[0273] The delay addition unit 134 adds a delay amount q12 to the first transmission signal x1(n) and outputs x1(n-q12). The delay addition unit 135 adds a delay amount q22 to the second transmission signal x2(n) and outputs x2(n-q22). The delay addition unit 136 adds a delay amount q32 to the third transmission signal x3(n) and outputs x3(n-q32).

[0274] The address generation unit 11 generates the following address A for the first transmission signal. 11 ~A 16 The address generation unit 11 generates the address A of the first transmission signal. 11,q11 =|x1(n-q11)| 2 , address A of the second transmitted signal14,q11 =|x2(n-q11)| 2 and the address A of the third transmission signal 15,q11 =|x3(n-q11)| 2 The address generation unit 11 generates the address A of the combined power of the first transmission signal and the second transmission signal. 12,q11 =|x1(n-q11)| 2 +|x2(n-q11)| 2 The address generation unit 11 generates the address A of the combined power of the first transmission signal and the third transmission signal. 13,q11 =|x1(n-q11)| 2 +|x3(n-q11)| 2 The address generation unit 11 generates the address A of the combined power of the second transmission signal and the third transmission signal. 16,q11 =|x2(n-q11)| 2 +|x3(n-q11)| 2 Generates.

[0275] Furthermore, the address generation unit 11 generates the following address A for the second transmission signal. 21 ~A 26 The address generation unit 11 generates the address A of the second transmission signal. 21,q21 =|x2(n-q21)| 2 , address A of the first transmitted signal 24,q21 =|x1(n-q21)| 2 and the address A of the third transmission signal 25,q21 =|x3(n-q21)| 2 The address generation unit 11 generates the address A of the combined power of the first transmission signal and the second transmission signal. 22,q21 =|x1(n-q21)| 2 +|x2(n-q21)| 2 The address generation unit 11 generates the address A of the combined power of the second transmission signal and the third transmission signal. 23,q21 =|x2(n-q21)| 2 +|x3(n-q21)| 2 The address generation unit 11 generates the address A of the combined power of the first transmission signal and the third transmission signal. 26,q21 =|x1(n-q21)| 2+|x3(n-q21)| 2 Generates.

[0276] Furthermore, the address generation unit 11 generates the following address A for the third transmission signal. 31 ~A 36 The address generation unit 11 generates the address A of the third transmission signal. 31,q31 =|x3(n-q31)| 2 , address A of the first transmitted signal 34,q31 =|x1(n-q31)| 2 and address A of the second transmission signal 35,q31 =|x2(n-q31)| 2 The address generation unit 11 generates the address A of the combined power of the first transmission signal and the third transmission signal. 32,q31 =|x1(n-q31)| 2 +|x3(n-q31)| 2 The address generation unit 11 generates the address A of the combined power of the second transmission signal and the third transmission signal. 33,q31 =|x2(n-q31)| 2 +|x3(n-q31)| 2 The address generation unit 11 generates the address A of the combined power of the first transmission signal and the second transmission signal. 36,q31 =|x1(n-q31)| 2 +|x2(n-q31)| 2 Generates.

[0277] LUT411 is address A 11,q11 For the input, LUT 1,1,q11,q12 (A 11,q11 ) outputs. LUT412 outputs address A 12,q11 For the input, LUT 1,2,q11,q12 (A 12,q11 ) outputs. LUT413 outputs address A 13,q11 For the input, LUT 1,3,q11,q12 (A 13,q11 ) outputs. LUT414 outputs address A 14,q11 For the input, LUT 1,4,q11,q12 (A 14,q11 ) outputs. LUT415 outputs address A 15,q11 For the input, LUT1,5,q11,q12 (A 15,q11 ) is output. LUT416 responds to an input of address A 16,q11 by outputting LUT 1,6,q11,q12 (A 16,q11 ).

[0278] LUT421 responds to an input of address A 21,q21 by outputting LUT 2,1,q21,q22 (A 21,q21 ). LUT422 responds to an input of address A 22,q21 by outputting LUT 2,2,q21,q22 (A 22,q21 ). LUT423 responds to an input of address A 23,q21 by outputting LUT 2,3,q21,q22 (A 23,q21 ). LUT424 responds to an input of address A 24,q21 by outputting LUT 2,4,q21,q22 (A 24,q21 ). LUT425 responds to an input of address A 25,q21 by outputting LUT 2,5,q21,q22 (A 25,q21 ). LUT426 responds to an input of address A 26,q21 by outputting LUT 2,6,q21,q22 (A 26,q21 ).

[0279] LUT431 responds to an input of address A 31,q31 by outputting LUT 3,1,q31,q32 (A 31,q31 ). LUT432 responds to an input of address A 32,q31 by outputting LUT 3,2,q31,q32 (A 32,q31 ). LUT433 responds to an input of address A 33,q31 by outputting LUT 3,3,q31,q32 (A 33,q31 ). LUT434 responds to an input of address A 34,q31 by outputting LUT 3,4,q31,q32 (A 34,q31 ). LUT435 responds to an input of address A 35,q31 by outputting LUT 3,5,q31,q32 (A 35,q31 ). LUT436 responds to an input of address A 36,q31For the input, LUT 3,6,q31,q32 (A 36,q31 Outputs ).

[0280] The adder 127 controls the distortion compensation coefficient LUT of the first transmitted signal. SUM-1,q12 This is calculated using the following formula (10). Here, radio 1 sets q11 to -Q 1,1 From +Q 1,1 Change it to that extent.

[0281]

number

[0282] The adder 128 controls the distortion compensation coefficient LUT of the second transmitted signal. SUM-2,q22 This is calculated using the following formula (11). Here, radio 1 sets q21 to -Q 2,1 From +Q 2,1 Change it to that extent.

[0283]

number

[0284] The adder 129 controls the distortion compensation coefficient LUT of the first transmitted signal. SUM-3,q32 This is calculated using the following formula (12). Here, radio 1 sets q31 to -Q 3,1 From +Q 3,1 Change it to that extent.

[0285]

number

[0286] The multiplier 131 combines x1(n-q12) output from the delay addition unit 134 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-1,q12 Multiplying by the following equation (13), a pre-distortion signal is output. Here, radio 1 sets q12 to -Q 1,2 From +Q 1,2 Change it to that extent.

[0287]

number

[0288] The multiplier 132 combines x2(n-q22) output from the delay addition unit 135 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-2,q22 Multiplying by the following equation (14), a pre-distortion signal is output. Here, radio 1 sets q22 to -Q 2,2 From +Q 2,2 Change it to that extent.

[0289]

number

[0290] The multiplier 133 combines x3(n-q32) output from the delay addition unit 136 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-3,q32 Multiplying by the following equation (15), a pre-distortion signal is output. Here, radio 1 sets q32 to -Q 3,2 From +Q 3,2 Change it to that extent.

[0291]

number

[0292] The case where it is combined with the second example of Example 2 will be explained. For each of the first to third transmission signals, the power of the first transmission signal |x1(n)| 2 The delay amount pqw1 and the power of the second transmitted signal |x2(n)| 2 The delay amount pqw2 and the power of the third transmitted signal in the third band |x3(n)| 2 The delay amount qpw3 is different. Here, the power of the first transmitted signal |x1(n)| in the first transmitted signal. 2 Let q11 be the amount of change in the delay, and the power of the second transmitted signal |x2(n)| 2 Let q13 be the amount of change in the delay, and the power of the third transmitted signal |x3(n)| 2Let the varying delay amount be q14. Also, in the second transmission signal, the power of the second transmission signal |x2(n)| 2 Let the varying delay amount be q21, and the power of the first transmission signal |x1(n)| 2 Let the varying delay amount be q23, and the power of the third transmission signal |x3(n)| 2 Let the varying delay amount be q24. Also, in the third transmission signal, the power of the third transmission signal |x3(n)| 2 Let the varying delay amount be q31, and the power of the first transmission signal |x1(n)| 2 Let the varying delay amount be q33, and the power of the second transmission signal |x2(n)| 2 Let the varying delay amount be q34. Also, let the varying delay amount qtx1 of the first transmission signal be q12, let the varying delay amount qtx2 of the second transmission signal be q22, and let the varying delay amount qtx3 of the third transmission signal be q32.

[0293] The address generation unit 11 generates the following addresses A for the first transmission signal 11 ~A 16 That is, the address generation unit 11 generates the address A of the first transmission signal 11,q11 =|x1(n-q11)| 2 , the address A of the second transmission signal 14,q13 =|x2(n-q13)| 2 and the address A of the third transmission signal 15,q14 =|x3(n-q14)| 2 In addition, the address generation unit 11 generates the combined power address A of the first transmission signal and the second transmission signal 12,q11,q13 =|x1(n-q11)| 2 +|x2(n-q13)| 2 In addition, the address generation unit 11 generates the combined power address A of the first transmission signal and the third transmission signal 13,q11,q14 =|x1(n-q11)| 2 +|x3(n-q14)| 2 In addition, the address generation unit 11 generates the combined power address A of the second transmission signal and the third transmission signal 16,q13,q14 =|x2(n-q13)| 2 +|x3(n-q14)| 2 is generated.

[0294] Furthermore, the address generation unit 11 generates the following address A for the second transmission signal. 21 ~A 26 The address generation unit 11 generates the address A of the second transmission signal. 21,q21 =|x2(n-q21)| 2 , address A of the first transmitted signal 24,q23 =|x1(n-q23)| 2 and the address A of the third transmission signal 25,q24 =|x3(n-q24)| 2 The address generation unit 11 generates the address A of the combined power of the first transmission signal and the second transmission signal. 22,q21,q23 =|x1(n-q23)| 2 +|x2(n-q21)| 2 The address generation unit 11 generates the address A of the combined power of the second transmission signal and the third transmission signal. 23,q21,q24 =|x2(n-q21)| 2 +|x3(n-q24)| 2 The address generation unit 11 generates the address A of the combined power of the first transmission signal and the third transmission signal. 26,q23,q24 =|x1(n-q23)| 2 +|x3(n-q24)| 2 Generates.

[0295] Furthermore, the address generation unit 11 generates the following address A for the third transmission signal. 31 ~A 36 The address generation unit 11 generates the address A of the third transmission signal. 31,q31 =|x3(n-q31)| 2 , address A of the first transmitted signal 34,q33 =|x1(n-q33)| 2 and address A of the second transmission signal 35,q34 =|x2(n-q34)| 2 The address generation unit 11 generates the address A of the combined power of the first transmission signal and the third transmission signal. 32,q31,q33 =|x1(n-q33)| 2 +|x3(n-q31)| 2The address generation unit 11 generates the address A of the combined power of the second transmission signal and the third transmission signal. 33,q31,q34 =|x2(n-q34)| 2 +|x3(n-q31)| 2 The address generation unit 11 generates the address A of the combined power of the first transmission signal and the second transmission signal. 36,q33,q34 =|x1(n-q33)| 2 +|x2(n-q34)| 2 Generates.

[0296] LUT411 is address A 11,q11 For the input, LUT 1,1,q11,q12,q13,q14 (A 11,q11 ) outputs. LUT412 outputs address A 12,q11,q13 For the input, LUT 1,2,q11,q12,q13,q14 (A 12,q11,q13 ) outputs. LUT413 outputs address A 13,q11,q14 For the input, LUT 1,3,q11,q12,q13,q14 (A 13,q11,q14 ) outputs. LUT414 outputs address A 14,q13 For the input, LUT 1,4,q11,q12,q13,q14 (A 14,q13 ) outputs. LUT415 outputs address A 15,q14 For the input, LUT 1,5,q11,q12,q13,q14 (A 15,q14 ) outputs. LUT416 outputs address A 16,q13,q14 For the input, LUT 1,6,q11,q12,q13,q14 (A 16,q13,q14 Outputs ).

[0297] LUT421 is address A 21,q21 For the input, LUT 2,1,q21,q22,q23,q24 (A 21,q21 ) outputs. LUT422 outputs address A 22,q21,q23 For the input, LUT 2,2,q21,q22,q23,q24 (A 22,q21,q23 ) outputs. LUT423 outputs address A 23,q21,q24 For the input, LUT 2,3,q21,q22,q23,q24 (A 23,q21,q24 ) outputs. LUT424 is at address A 24,q23 For the input, LUT 2,4,q21,q22,q23,q24 (A24,q23 ) outputs. LUT425 outputs address A 25,q24 For the input, LUT 2,5,q21,q22,q23,q24 (A 25,q24 ) outputs. LUT426 outputs address A 26,q23,q24 For the input, LUT 2,6,q21,q22,q23,q24 (A 26,q23,q24 Outputs ).

[0298] LUT431 is address A 31,q31 For the input, LUT 3,1,q31,q32,q33,q34 (A 31,q31 ) outputs. LUT432 is at address A 32,q31,q33 For the input, LUT 3,2,q31,q32,q33,q34 (A 32,q31,q33 ) outputs. LUT433 outputs address A 33,q31,q34 For the input, LUT 3,3,q31,q32,q33,q34 (A 33,q31,q34 ) outputs. LUT434 outputs address A 34,q33 For the input, LUT 3,4,q31,q32,q33,q34 (A 34,q33 ) outputs. LUT435 outputs address A 35,q34 For the input, LUT 3,5,q31,q32,q33,q34 (A 35,q34 ) outputs. LUT436 outputs address A 36,q33,q34 For the input, LUT 3,6,q31,q32,q33,q34 (A 36,q33,q34 Outputs ).

[0299] The adder 127 controls the distortion compensation coefficient LUT of the first transmitted signal. SUM-1,q12 This is calculated using the following formula (16). Here, radio 1 sets q11 to -Q 1,1 From +Q 1,1 Change it to -Q 1,3 From +Q 1,3 Change it to -Q 1,4 From +Q 1,4 Change it to that extent.

[0300]

number

[0301] The adder 128 controls the distortion compensation coefficient LUT of the second transmitted signal. SUM-2,q22 This is calculated using the following formula (17). Here, radio 1 sets q21 to -Q 2,1 From +Q 2,1 Change it to -Q 2,3 From +Q 2,3 Change it to -Q 2,4 From +Q 2,4 Change it to that extent.

[0302]

number

[0303] The adder 129 controls the distortion compensation coefficient LUT of the third transmitted signal. SUM-3,q32 This is calculated using the following formula (18). Here, radio 1 sets q31 to -Q 3,1 From +Q 3,1 Change it to -Q 3,3 From +Q 3,3 Change it to -Q 3,4 From +Q 3,4 Change it to that extent.

[0304]

number

[0305] The multiplier 131 combines x1(n-q12) output from the delay addition unit 134 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-1,q12 Multiplying by the above formula (13), a pre-distortion signal is output. Here, radio 1 sets q12 to -Q 1,2 From +Q 1,2 Change it to that extent.

[0306] The multiplier 132 combines x2(n-q22) output from the delay addition unit 135 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-2,q22 By multiplying by the above formula (14), a pre-distortion signal is output. Here, radio 1 sets q22 to -Q 2,2 From +Q2,2 Change it to that extent.

[0307] The multiplier 133 combines x3(n-q32) output from the delay addition unit 136 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-3,q32 Multiplying by the above formula (15), a pre-distortion signal is output. Here, radio 1 sets q32 to -Q 3,2 From +Q 3,2 Change it to that extent.

[0308] Next, we will describe the case when combined with the third example of Example 2. In this case, the LUT when combined with the second example of Example 2. 1,1,q11,q12,q13,q14 (A 11,q11 ) does not include q13 and q14 as variables, LUT 1,1,q11,q12 (A 11,q11 ) will be the result. Also, LUT 1,2,q11,q12,q13,q14 (A 12,q11,q13 ) does not include q14 as a variable, LUT 1,2,q11,q12,q13 (A 12,q11,q13 ) will be the result. Also, LUT 1,3,q11,q12,q13,q14 (A 13,q11,q14 ) does not include q13 as a variable, LUT 1,3,q11,q12,q14 (A 13,q11,q14 ) will be the result. Also, LUT 1,4,q11,q12,q13,q14 (A 14,q13 ) does not include q11 and q14 as variables, LUT 1,4,q12,q13 (A 14,q13 ) will be the result. Also, LUT 1,5,q11,q12,q13,q14 (A 15,q14 ) does not include q11 and q13 as variables, LUT 1,4,q12,q14 (A 15,q14 ) will be the result. Also, LUT 1,6,q11,q12,q13,q14 (A 16,q13,q14 ) does not include q11 as a variable, LUT 1,6,q12,q13,q14 (A 16,q13,q14 )

[0309] Also, LUT 2,1,q21,q22,q23,q24 (A 21,q21 ) does not include q23 and q24 as variables, LUT 2,1,q21,q22 (A 21,q21 ) will be the result. Also, LUT 2,2,q21,q22,q23,q24 (A 22,q21,q23) does not include q24 as a variable, LUT 2,2,q21,q22,q23 (A 22,q21,q23 ) will be the result. Also, LUT 2,3,q21,q22,q23,q24 (A 23,q21,q24 ) does not include q23 as a variable, LUT 2,3,q21,q22,q24 (A 23,q21,q24 ) will be the result. Also, LUT 2,4,q21,q22,q23,q24 (A 24,q23 ) does not include q21 and q24 as variables, LUT 2,4,q22,q23 (A 24,q23 ) will be the result. Also, LUT 2,5,q21,q22,q23,q24 (A 25,q24 ) does not include q21 and q23 as variables, LUT 2,5,q22,q24 (A 25,q24 ) will be the result. Also, LUT 2,6,q21,q22,q23,q24 (A 26,q23,q24 ) does not include q21 as a variable, LUT 2,6,q22,q23,q24 (A 26,q23,q24 )

[0310] Also, LUT 3,1,q31,q32,q33,q34 (A 31,q31 ) does not include q33 and q34 as variables, LUT 3,1,q31,q32 (A 31,q31 ) will be the result. Also, LUT 3,2,q31,q32,q33,q34 (A 32,q31,q33 ) does not include q34 as a variable, LUT 3,2,q31,q32,q33 (A 32,q31,q33 ) will be the result. Also, LUT 3,3,q31,q32,q33,q34 (A 33,q31,q34 ) does not include q33 as a variable, LUT 3,3,q31,q32,q34 (A 33,q31,q34 ) will be the result. Also, LUT 3,4,q31,q32,q33,q34 (A 34,q33 ) does not include q31 and q34 as variables, LUT 3,4,q32,q33 (A 34,q33 ) will be the result. Also, LUT 3,5,q31,q32,q33,q34 (A 35,q34 ) does not include q31 and q33 as variables, LUT 3,5,q32,q34 (A 35,q34 ) will be the result. Also, LUT 3,6,q31,q32,q33,q34 (A 36,q33,q34 ) does not include q31 as a variable, LUT 3,6,q32,q33,q34 (A 36,q33,q34 )

[0311] The adder 127 controls the distortion compensation coefficient LUT of the first transmitted signal. SUM-1,q12 This is calculated using the following formula (19). Here, radio 1 sets q11 to -Q 1,1 From +Q 1,1 Change it to -Q 1,3 From +Q 1,3 Change it to -Q 1,4 From +Q 1,4 Change it to that extent.

[0312]

number

[0313] The adder 128 controls the distortion compensation coefficient LUT of the second transmitted signal. SUM-2,q22 This is calculated using the following formula (20). Here, radio 1 sets q21 to -Q 2,1 From +Q 2,1 Change it to -Q 2,3 From +Q 2,3 Change it to -Q 2,4 From +Q 2,4 Change it to that extent.

[0314]

number

[0315] The adder 129 controls the distortion compensation coefficient LUT of the third transmitted signal. SUM-3,q32 This is calculated using the following formula (21). Here, radio 1 sets q31 to -Q 3,1 From +Q 3,1 Change it to -Q 3,3 From +Q 3,3 Change it to -Q 3,4 From +Q 3,4 Change it to that extent.

[0316]

number

[0317] The multiplier 131 combines x1(n-q12) output from the delay addition unit 134 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-1,q12 Multiplying by the above formula (13), a pre-distortion signal is output. Here, radio 1 sets q12 to -Q 1,2 From +Q 1,2 Change it to that extent.

[0318] The multiplier 132 combines x2(n-q22) output from the delay addition unit 135 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-2,q22 By multiplying by the above formula (14), a pre-distortion signal is output. Here, radio 1 sets q22 to -Q 2,2 From +Q 2,2 Change it to that extent.

[0319] The multiplier 133 combines x3(n-q32) output from the delay addition unit 136 and the LUT output from the distortion compensation coefficient calculation unit 12. SUM-3,q32 Multiplying by the above formula (15), a pre-distortion signal is output. Here, radio 1 sets q32 to -Q 3,2 From +Q 3,2 Change it to that extent.

[0320] (Variation 7-2) Next, a modified example 7-2 of Example 7 will be described. The configuration of Example 7 can also be combined with the configuration of Example 4. That is, the radio 1 adaptively updates the distortion compensation coefficient using a feedback signal obtained by feeding back a portion of the amplifier output signal. For example, an example of using it in combination with the first to third examples of Example 2 will be described below. In this case, the radio 1 has a feedback unit 20 and a coefficient update unit 30.

[0321] For example, let y1(n) be the feedback signal corresponding to the first transmission signal x1(n). Also, let y2(n) be the feedback signal corresponding to the second transmission signal x2(n). Also, let y3(n) be the feedback signal corresponding to the third transmission signal x3(n). The calculation of the distortion compensation coefficient by the coefficient update unit 30 in this case will be explained. Each of the addresses input to LUT411~416 is A 11 ~A 16 And each of the addresses entered into LUT421~426 is A 21 ~A 26 And each of the addresses entered into LUT431~436 is A 31 ~A 36 This will be explained as follows.

[0322] For example, the coefficient update unit 30 updates the new strain compensation coefficient LUT of LUT 411. 1,1 (A 11 ) to LUT 1,1 (A 11 )=LUT 1,1 (A 11 )+μ·e1(n)·(y1(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new strain compensation coefficient LUT of LUT412. 1,2 (A 12 ) to LUT 1,2 (A 12 )=LUT 1,2 (A 12 )+μ·e1(n)·(y1(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new strain compensation coefficient LUT of LUT413. 1,3 (A 13 ) to LUT 1,3 (A 13 )=LUT 1,3 (A 13 )+μ·e1(n)·(y1(n)) * The coefficient update unit 30 calculates the new strain compensation coefficient LUT for LUT414. 1,4 (A 14 ) to LUT 1,4 (A 14 )=LUT 1,4 (A 14 )+μ·e1(n)·(y1(n))* It is calculated as follows. In addition, the coefficient update unit 30 updates the new strain compensation coefficient LUT of LUT415. 1,5 (A 15 ) to LUT 1,5 (A 15 )=LUT 1,5 (A 15 )+μ·e1(n)·(y1(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new strain compensation coefficient LUT of LUT416. 1,6 (A 16 ) to LUT 1,6 (A 16 )=LUT 1,6 (A 16 )+μ·e1(n)·(y1(n)) * It is calculated as follows.

[0323] Furthermore, for example, the coefficient update unit 30 updates the new strain compensation coefficient LUT of LUT421. 2,1 (A 21 ) to LUT 2,1 (A 21 )=LUT 2,1 (A 21 ) + μ·e2(n)·(y2(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new distortion compensation coefficient LUT of LUT422. 2,2 (A 22 ) to LUT 2,2 (A 22 )=LUT 2,2 (A 22 ) + μ·e2(n)·(y2(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new distortion compensation coefficient LUT of LUT423. 2,3 (A 23 ) to LUT 2,3 (A 23 )=LUT 2,3 (A 23 ) + μ·e2(n)·(y2(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new distortion compensation coefficient LUT of LUT424. 2,4 (A 24 ) to LUT 2,4 (A 24 )=LUT 2,4 (A 24) + μ·e2(n)·(y2(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new distortion compensation coefficient LUT of LUT425. 2,5 (A 25 ) to LUT 2,5 (A 25 )=LUT 2,5 (A 25 ) + μ·e2(n)·(y2(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new distortion compensation coefficient LUT of LUT426. 2,6 (A 26 ) to LUT 2,6 (A 26 )=LUT 2,6 (A 26 ) + μ·e2(n)·(y2(n)) * It is calculated as follows.

[0324] Furthermore, for example, the coefficient update unit 30 updates the new strain compensation coefficient LUT of LUT 431. 3,1 (A 31 ) to LUT 3,1 (A 31 )=LUT 3,1 (A 31 )+μ·e3(n)·(y3(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new distortion compensation coefficient LUT of LUT432. 3,2 (A 32 ) to LUT 3,2 (A 32 )=LUT 3,2 (A 32 )+μ·e3(n)·(y3(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new strain compensation coefficient LUT of LUT433. 3,3 (A 33 ) to LUT 3,3 (A 33 )=LUT 3,3 (A 33 )+μ·e3(n)·(y3(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new strain compensation coefficient LUT of LUT434. 3,4 (A 34 ) to LUT 3,4 (A 34 )=LUT3,4 (A 34 )+μ·e3(n)·(y3(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new distortion compensation coefficient LUT of LUT435. 3,5 (A 35 ) to LUT 3,5 (A 35 )=LUT 3,5 (A 35 )+μ·e3(n)·(y3(n)) * It is calculated as follows. In addition, the coefficient update unit 30 updates the new strain compensation coefficient LUT of LUT436. 3,6 (A 36 ) to LUT 3,6 (A 36 )=LUT 3,6 (A 36 )+μ·e3(n)·(y3(n)) * It is calculated as follows.

[0325] Furthermore, in addition to compensating for the memory effect, the radio 1 may adaptively update the distortion compensation coefficient using a feedback signal obtained by feeding back a portion of the amplifier output signal.

[0326] For example, let's describe the combination with the first example of Embodiment 2 in the modified example 7-1 described above. In this case, the coefficient update unit 30 updates each strain compensation coefficient as follows.

[0327] LUT 1,1,q11,q12 (A 11,q11 )=LUT 1,1,q11,q12 (A 11,q11 )+μ·e1(n)·(y1(n-q12)) * LUT 1,2,q11,q12 (A 12,q11 )=LUT 1,2,q11,q12 (A 12,q11 )+μ·e1(n)·(y1(n-q12)) * LUT 1,3,q11,q12 (A 13,q11 )=LUT 1,3,q11,q12 (A 13,q11 )+μ·e1(n)·(y1(n-q12)) * LUT 1,4,q11,q12 (A 14,q11 )=PAGE 1,4,q11,q12 (A 14,q11 )+µ·e1(n)·(y1(n-q12)) * LUT 1,5,q11,q12 (A 15,q11 )=PAGE 1,5,q11,q12 (A 15,q11 )+µ·e1(n)·(y1(n-q12)) * LUT 1,6,q11,q12 (A 16,q11 )=PAGE 1,6,q11,q12 (A 16,q11 )+µ·e1(n)·(y1(n-q12)) *

[0328] LUT 2,1,q21,q22 (A 21,q21 )=PAGE 2,1,q21,q22 (A 21,q21 )+µ·e2(n)·(y2(n-q22)) * LUT 2,2,q21,q22 (A 22,q21 )=PAGE 2,2,q21,q22 (A 22,q21 )+µ·e2(n)·(y2(n-q22)) * LUT 2,3,q21,q22 (A 23,q21 )=PAGE 2,3,q21,q22 (A 23,q21 )+µ·e2(n)·(y2(n-q22)) * LUT 2,4,q21,q22 (A 24,q21 )=PAGE 2,4,q21,q22 (A 24,q21 )+µ·e2(n)·(y2(n-q22)) * LUT 2,5,q21,q22 (A 25,q21 )=PAGE 2,5,q21,q22 (A 25,q21 )+µ·e2(n)·(y2(n-q22)) * LUT 2,6,q21,q22 (A 26,q21 )=PAGE 2,6,q21,q22 (A 26,q21 )+µ·e2(n)·(y2(n-q22))*

[0329] LUT 3,1,q31,q32 (A 31,q31 )=LUT 3,1,q31,q32 (A 31,q31 )+μ·e3(n)·(y3(n-q32)) * LUT 3,2,q31,q32 (A 32,q31 )=LUT 3,2,q31,q32 (A 32,q31 )+μ·e3(n)·(y3(n-q32)) * LUT 3,3,q31,q32 (A 33,q31 )=LUT 3,3,q31,q32 (A 33,q31 )+μ·e3(n)·(y3(n-q32)) * LUT 3,4,q31,q32 (A 34,q31 )=LUT 3,4,q31,q32 (A 34,q31 )+μ·e3(n)·(y3(n-q32)) * LUT 3,5,q31,q32 (A 35,q31 )=LUT 3,5,q31,q32 (A 35,q31 )+μ·e3(n)·(y3(n-q32)) * LUT 3,6,q31,q32 (A 36,q31 )=LUT 3,6,q31,q32 (A 36,q31 )+μ·e3(n)·(y3(n-q32)) *

[0330] Furthermore, we will describe the combination with the second example of Example 2 in the modified example 7-1 described above. In this case, the coefficient update unit 30 updates each strain compensation coefficient as follows.

[0331] LUT 1,1,q11,q12,q13,q14 (A 11,q11 )=LUT 1,1,q11,q12,q13,q14 (A 11,q11 )+μ·e1(n)·(y1(n-q12)) * LUT 1,2,q11,q12,q13,q14 (A12,q11,q13 )=PAGE 1,2,q11,q12,q13,q14 (A 12,q11,q13 )+µ·e1(n)·(y1(n-q12)) * LUT 1,3,q11,q12,q13,q14 (A 13,q11,q14 )=PAGE 1,3,q11,q12,q13,q14 (A 13,q11,q14 )+µ·e1(n)·(y1(n-q12)) * LUT 1,4,q11,q12,q13,q14 (A 14,q13 )=PAGE 1,4,q11,q12,q13,q14 (A 14,q13 )+µ·e1(n)·(y1(n-q12)) * LUT 1,5,q11,q12,q13,q14 (A 15,q14 )=PAGE 1,5,q11,q12,q13,q14 (A 15,q14 )+µ·e1(n)·(y1(n-q12)) * LUT 1,6,q11,q12,q13,q14 (A 16,q13,q14 )=PAGE 1,6,q11,q12,q13,q14 (A 16,q13,q14 )+µ·e1(n)·(y1(n-q12)) *

[0332] LUT 2,1,q21,q22,q23,q24 (A 21,q21 )=PAGE 2,1,q21,q22,q23,q24 (A 21,q21 )+µ·e2(n)·(y2(n-q22)) * LUT 2,2,q21,q22,q23,q24 (A 22,q21,q23 )=PAGE 2,2,q21,q22,q23,q24 (A 22,q21,q23 )+µ·e2(n)·(y2(n-q22)) * LUT 2,3,q21,q22,q23,q24 (A 23,q21,q24 )=PAGE 2,3,q21,q22,q23,q24 (A 23,q21,q24 )+µ·e2(n)·(y2(n-q22)) * LUT 2,4,q21,q22,q23,q24 (A 24,q23 )=PAGE 2,4,q21,q22,q23,q24 (A 24,q23 )+µ·e2(n)·(y2(n-q22)) * LUT 2,5,q21,q22,q23,q24 (A 25,q24 )=PAGE 2,5,q21,q22,q23,q24 (A 25,q24 )+µ·e2(n)·(y2(n-q22)) * LUT 2,6,q21,q22,q23,q24 (A 26,q23,q24 )=PAGE 2,6,q21,q22,q23,q24 (A 26,q23,q24 )+µ·e2(n)·(y2(n-q22)) *

[0333] LUT 3,1,q31,q32,q33,q34 (A 31,q31 )=PAGE 3,1,q31,q32,q33,q34 (A 31,q31 )+µ·e3(n)·(y3(n-q32)) * LUT 3,2,q31,q32,q33,q34 (A 32,q31,q33 )=PAGE 3,2,q31,q32,q33,q34 (A 32,q31,q33 )+µ·e3(n)·(y3(n-q32)) * LUT 3,3,q31,q32,q33,q34 (A 33,q31,q34 )=PAGE 3,3,q31,q32,q33,q34 (A 33,q31,q34 )+µ·e3(n)·(y3(n-q32)) * LUT 3,4,q31,q32,q33,q34 (A 34,q33 )=PAGE 3,4,q31,q32,q33,q34 (A 34,q33 )+µ·e3(n)·(y3(n-q32)) * LUT 3,5,q31,q32,q33,q34 (A 35,q34 )=PAGE 3,5,q31,q32,q33,q34 (A 35,q34 )+µ·e3(n)·(y3(n-q32)) * LUT 3,6,q31,q32,q33,q34 (A 36,q33,q34 )=PAGE 3,6,q31,q32,q33,q34 (A 36,q33,q34 )+µ·e3(n)·(y3(n-q32)) *

[0334] Furthermore, we will describe the combination with the third example of Example 2 in the modified example 7-1 described above. In this case, the coefficient update unit 30 updates each strain compensation coefficient as follows.

[0335] LUT 1,1,q11,q12 (A 11,q11 )=LUT 1,1,q11,q12 (A 11,q11 )+μ·e1(n)·(y1(n-q12)) * LUT 1,2,q11,q12,q13 (A 12,q11,q13 )=LUT 1,2,q11,q12,q13 (A 12,q11,q13 )+μ·e1(n)·(y1(n-q12)) * LUT 1,3,q11,q12,q14 (A 13,q11,q14 )=LUT 1,3,q11,q12,q14 (A 13,q11,q14 )+μ·e1(n)·(y1(n-q12)) * LUT 1,4,q12,q13 (A 14,q13 )=LUT 1,4,q12,q13 (A 14,q13 )+μ·e1(n)·(y1(n-q12)) * LUT 1,5,q12,q14 (A 15,q14 )=LUT 1,5,q12,q14 (A 15,q14 )+μ·e1(n)·(y1(n-q12)) * LUT 1,6,q12,q13,q14 (A 16,q13,q14 )=LUT 1,6,q12,q13,q14 (A 16,q13,q14 )+μ·e1(n)·(y1(n-q12)) *

[0336] LUT 2,1,q21,q22 (A 21,q21 )=LUT 2,1,q21,q22 (A 21,q21 )+μ·e2(n)·(y2(n-q22)) * LUT 2,2,q21,q22,q23 (A 22,q21,q23 )=LUT 2,2,q21,q22,q23 (A 22,q21,q23)+µ·e2(n)·(y2(n-q22)) * LUT 2,3,q21,q22,q24 (A 23,q21,q24 )=PAGE 2,3,q21,q22,q24 (A 23,q21,q24 )+µ·e2(n)·(y2(n-q22)) * LUT 2,4,q22,q23 (A 24,q23 )=PAGE 2,4,q22,q23 (A 24,q23 )+µ·e2(n)·(y2(n-q22)) * LUT 2,5,q22,q24 (A 25,q24 )=PAGE 2,5,q22,q24 (A 25,q24 )+µ·e2(n)·(y2(n-q22)) * LUT 2,6,q22,q23,q24 (A 26,q23,q24 )=PAGE 2,6,q22,q23,q24 (A 26,q23,q24 )+µ·e2(n)·(y2(n-q22)) *

[0337] LUT 3,1,q31,q32 (A 31,q31 )=PAGE 3,1,q31,q32 (A 31,q31 )+µ·e3(n)·(y3(n-q32)) * LUT 3,2,q31,q32,q33 (A 32,q31,q33 )=PAGE 3,2,q31,q32,q33 (A 32,q31,q33 )+µ·e3(n)·(y3(n-q32)) * LUT 3,3,q31,q32,q34 (A 33,q31,q34 )=PAGE 3,3,q31,q32,q34 (A 33,q31,q34 )+µ·e3(n)·(y3(n-q32)) * LUT 3,4,q32,q33 (A 34,q33 )=PAGE 3,4,q32,q33 (A 34,q33 )+µ·e3(n)·(y3(n-q32)) * LUT 3,5,q32,q34 (A 35,q34 )=PAGE3,5,q32,q34 (A 35,q34 )+μ·e3(n)·(y3(n-q32)) * LUT 3,6,q32,q33,q34 (A 36,q33,q34 )=LUT 3,6,q32,q33,q34 (A 36,q33,q34 )+μ·e3(n)·(y3(n-q32)) * [Examples]

[0338] In the embodiments described above, the case where distortion compensation is performed for each band was explained as an example, but the radio 1 can also perform distortion compensation for each carrier. A band has a bandwidth allocated to a mobile operator, and the mobile operator can freely determine the placement of carriers within each band.

[0339] Figure 15 shows the relationship between bands and carriers. Radio 1 can handle transmission signals for, for example, two carriers in each of the first and second bands.

[0340] The address generation unit 11, distortion compensation coefficient calculation unit 12, and multiplier 131 of the radio 1 can perform distortion compensation processing on the transmitted signals of carriers 551 and 552 in the first band together. Furthermore, the address generation unit 11, distortion compensation coefficient calculation unit 12, and multiplier 132 can perform distortion compensation processing on the transmitted signals of carriers 561 and 562 in the second band together.

[0341] Furthermore, when performing distortion compensation processing on the first carrier, the address generation unit 11, the distortion compensation coefficient calculation unit 12, and the multiplier 131 may perform distortion compensation processing on carrier 551 as a transmission signal, and independently perform distortion compensation processing on carrier 552 as a transmission signal. Similarly, when performing distortion compensation processing on the second carrier, the address generation unit 11, the distortion compensation coefficient calculation unit 12, and the multiplier 132 may perform distortion compensation processing on carrier 561 as a transmission signal, and independently perform distortion compensation processing on carrier 562 as a transmission signal.

[0342] Furthermore, in each of the embodiments and modifications described above, the wireless device 1 is capable of performing distortion compensation for each carrier. [Examples]

[0343] Next, Example 9 will be described. In the radio 1 according to each of the above-described embodiments, nonlinear distortion is suppressed in the vicinity of the first band and the second band. For example, if the first transmission signal is transmitted with bandwidth BW1 and the second transmission signal is transmitted with bandwidth BW2, then linear distortion within the bandwidths DPD-BW1 and DPD-BW2, which have 3 to 5 times the bandwidth, is suppressed. However, third-order distortion is 2f L -f H ya 2f H -f L It occurs in the frequency band. Also, fifth-order distortion occurs in 3f L -2f H ya 3f H -2f L This occurs in the bandwidth. Furthermore, nonlinear distortions that occur in bandwidths far from bandwidths BW1 and BW2, such as third-order and fifth-order distortions, may not be suppressed. Therefore, the radio 1 according to this embodiment suppresses nonlinear distortions that occur in bandwidths far from bandwidths BW1 and BW2.

[0344] Figure 16 shows the suppression of unwanted out-of-band signals from the transmitted signal. In this embodiment, the radio 1 has a BPF 50 placed between the PA 19 and the antenna.

[0345] The BPF50 has pass characteristics that attenuate nonlinear distortion in bandwidths far from bandwidths BW1 and BW2. For example, the BPF50 has pass characteristics 51 as shown in Figure 16.

[0346] The BPF 50 attenuates signals outside the range of the pass-through characteristic 51. This allows the BPF 50 to attenuate third-order distortions 512 and 513 and fifth-order distortions 511 and 514 in the signals being passed through, in bands far from the bandwidths BW1 and BW2 of the first and second transmitted signals. The signal with the nonlinear distortion in the bands far from the bandwidths BW1 and BW2 of the first and second transmitted signals attenuated is then radiated into space as radio waves from the antenna.

[0347] Furthermore, BPF50 can be applied to any of the above-described embodiments and modifications.

[0348] As described above, the wireless device 1 according to this embodiment can attenuate nonlinear distortion in a bandwidth far from the bandwidth of the transmitted signal, making it possible to transmit a signal with less distortion. [Explanation of symbols]

[0349] 1. Radio 2. Transmission device 3 UE 11 Address generation unit 12. Distortion Compensation Coefficient Calculation Unit 15, 121~123, 127~129, 315, 325 Adder 16,161,162 DAC 17,171,172 Upconverter 18,181,182 Local oscillators 19 PA 20 Feedback Department 21 Cup Noodles 22,131,132,133,216,226 multiplier 23,217,227 ADC 24,316,326 Adjustment section 27,301,302 Selector 30 Coefficient update section 50,215,225 BPF 101-103, 134-136 Delay Addition Section 111-113 Power Calculation Unit 141, 142, 143, 311, 321, 314, 324 Frequency shift section 211~213, 221~223, 411~416, 421~426, 431~436 LUT 312,322 LPF 313,323 Update Department 317,327 Delay adjustment section P1, P2, P3 signal path 100 Wireless Communication Systems

Claims

1. A first determination unit that determines a first distortion compensation coefficient based on the power or amplitude of a first signal and the power or amplitude of a second signal, A first signal generation unit generates a third signal based on the first signal and the first distortion compensation coefficient, A second signal generation unit generates a fourth signal based on the second signal and the first distortion compensation coefficient, A third signal generation unit generates a fifth signal based on the third signal and the fourth signal. An information processing device characterized by comprising:

2. The information processing apparatus according to claim 1, characterized in that the first determination unit determines the first distortion compensation coefficient using a lookup table.

3. The system further comprises a second determination unit that determines a second distortion compensation coefficient based on the first signal or the second signal, and determines a third distortion compensation coefficient based on the first distortion compensation coefficient and the second distortion compensation coefficient. The first signal generation unit generates the third signal based on the first signal and the third distortion compensation coefficient. The second signal generation unit generates the fourth signal based on the second signal and the third distortion compensation coefficient. The information processing apparatus according to feature 1.

4. The system further comprises a third determination unit that determines a fourth distortion compensation coefficient based on the first signal, a fifth distortion compensation coefficient based on the second signal, and a sixth distortion compensation coefficient based on the first, fourth, and fifth distortion compensation coefficients. The first signal generation unit generates the third signal based on the first signal and the sixth distortion compensation coefficient. The second signal generation unit generates the fourth signal based on the second signal and the sixth distortion compensation coefficient. The information processing apparatus according to feature 1.

5. A first delay addition unit that adds the same or different first delay to each of the first signal and the second signal, A second delay addition unit that adds the same or different second delay to the power or amplitude of the first signal and the power or amplitude of the second signal, respectively. Furthermore, The first determination unit determines the first distortion compensation coefficient based on the power or amplitude of the first signal to which the second delay has been applied and the power or amplitude of the second signal. The first signal generation unit generates the third signal based on the first signal to which the first delay has been applied and the first distortion compensation coefficient. The second signal generation unit generates the fourth signal based on the second signal to which the first delay has been applied and the first distortion compensation coefficient. The information processing apparatus according to feature 1.

6. A feedback unit that generates a feedback signal from the amplified fifth signal, A coefficient update unit updates the first distortion compensation coefficient based on the feedback signal, the first signal, and the second signal. The information processing apparatus according to claim 1, further comprising:

7. The information processing apparatus according to claim 6, characterized in that the coefficient update unit suppresses out-of-band signals of the first signal in the feedback signal by filtering to extract a first feedback signal corresponding to the first signal, suppresses out-of-band signals of the second signal in the feedback signal by filtering to extract a second feedback signal corresponding to the second signal, and updates the first distortion compensation coefficient based on the first signal, the first feedback signal, the second signal, and the second feedback signal.

8. The information processing apparatus according to claim 6, characterized in that the coefficient update unit extracts a first feedback signal corresponding to the first signal by removing the second signal from the feedback signal, extracts a second feedback signal corresponding to the second signal by removing the first signal from the feedback signal, and updates the first distortion compensation coefficient based on the first signal, the first feedback signal, the second signal, and the second feedback signal.

9. The information processing apparatus according to claim 6, characterized in that the coefficient update unit adds a third delay to the feedback signal and / or adds a fourth delay to the first signal and the second signal, adjusts the timing of the feedback signal and the first signal and the second signal, compares them, and updates the first distortion compensation coefficient.

10. An amplifier for amplifying the fifth signal described above, A suppression unit that suppresses nonlinear distortion in a frequency band separated by a predetermined frequency from the bandwidth of the first signal and the bandwidth of the second signal of the signal output from the amplifier. The information processing apparatus according to claim 1, further comprising:

11. A first determination unit that determines a first distortion compensation coefficient based on the power or amplitude of a first signal and the power or amplitude of a second signal, A first signal generation unit generates a third signal based on the first signal and the first distortion compensation coefficient, A second signal generation unit generates a fourth signal based on the second signal and the first distortion compensation coefficient, A third signal generation unit generates a fifth signal based on the third signal and the fourth signal. The amplifier that amplifies the third signal described above and outputs it from the antenna A communication device characterized by being equipped with the following features.

12. Information processing device, Based on the power or amplitude of the first signal and the power or amplitude of the second signal, the first distortion compensation coefficient is determined. A third signal is generated based on the first signal and the first distortion compensation coefficient. A fourth signal is generated based on the second signal and the first distortion compensation coefficient. A fifth signal is generated based on the third signal and the fourth signal. An information processing method characterized by performing a process.

Images