TECHNICAL FIELD The present invention relates to a quadrature digital power amplifier for converting an IQ data signal to an RF signal. Furthermore, the present invention also relates to a transmitter comprising such a quadrature digital power amplifier. The present invention also relates to a method for converting an IQ data signal to an RF signal. BACKGROUND

Embodiments of the present invention are related to quadrature digital power amplifiers, which convert digital l/Q signals directly to l/Q modulated analog RF signals. Such devices have various names and abbreviations in literature, such as all-digital quadrature transmitter, Direct Digital to RF Modulator DDRM, Digital to RF Converter DRFC, Digital to RF Amplitude Converter DRAC. However, in this document the device will be called a quadrature Digital Power Amplifier, Q-DPA.

A conventional quadrature Digital Power Amplifiers (Q-DPA) may comprise a unary power cell array in combination with a binary power cell array. If the whole power cell array was unary-coded (thermo-coded), 12 bit (plus sign bit) Q-DPA would require two arrays of 2 ^Λ 12=4096 individually controlled power cells. Routing LO and controls would be challenging and would introduce large amount of parasitic and varying delays to the control, which degrades the Q-DPA performance. In the other extreme if all units were binary coded, only 12 baseband control signals would be needed to control the 12 bit power cell array, but that would mean intolerable glitches and nonlinearity in the middle code region when a large number of separate power cells, corresponding to least significant bits, are turned off and the separate power cells corresponding to the next higher bit is turned on. Furthermore, due to manufacturing tolerances different power cells in the Q-DPA are not identical. This is one reason for the glitches. A power cell may be implemented in various ways. In the example on conventional technology, the power cell is a switch. Another well-known Q-DPA topology is based on switched-capacitors power cells.

A typical solution, as described above, is to have the least significant bit (LSB) units binary- coded and most significant bit (MSB) units unary coded (thermo-coded) but that still requires a large number of unary coded control signals to avoid binary glitches. The physical layout for a Q-DPA with large number of power units and mixed binary and unary coding is very challenging.

SUMMARY

An objective of the present invention is to provide a solution which mitigates or solves the drawbacks and problems of conventional solutions.

Another objective of the present invention is to provide a quadrature digital power amplifier system for converting an IQ data signal to an RF signal and a transmitter device comprising such a system, wherein the quadrature digital power amplifier system is an alternative to conventional quadrature digital power amplifier systems.

Another objective of the present invention is to provide a quadrature digital power amplifier system for converting an IQ data signal to an RF signal and a transmitter device comprising such a system, wherein the quadrature digital power amplifier system enables a simplified layout of the digital-to-RF converter devices in the quadrature digital power amplifier system compared to conventional solutions. An "or" in this description and the corresponding claims is to be understood as a mathematical OR which covers "and" and "or", and is not to be understood as an XOR (exclusive OR).

The above objectives are fulfilled by the subject matter of the independent claims. Further advantageous implementation forms of the present invention can be found in the dependent claims.

In the following, an IQ data signal is to be understood as a data signal comprising an in- phase data signal or an I data signal and a quadrature data signal or a Q signal.

According to a first aspect of the present invention, a quadrature digital power amplifier system is provided. The quadrature digital power amplifier system comprises a decomposition circuit comprising an input configured to receive an IQ data signal comprising an I data signal and a Q data signal, and a first suboutput and a second suboutput. The decomposition circuit is configured to decompose the IQ data signal into at least a first IQ data subsignal and a second IQ data subsignal if the signal level of the IQ data signal in a given sample is above a first threshold level, and to output the first IQ data subsignal on the first suboutput and the second IQ data subsignal on the second suboutput. The quadrature digital power amplifier system also comprises a first digital-to-RF converter device comprising a first converter input connected to the first suboutput of the decomposition circuit, a first digital power amplifier connected to the first converter input, and a first output. The first digital-to-RF converter device is configured to convert the first IQ data subsignal to a first RF signal and to output the first RF signal on the first output. The quadrature digital power amplifier system also comprises at least a second digital-to-RF converter device comprising a second converter input connected to the second suboutput of the decomposition circuit, a second digital power amplifier connected to the second converter input, and a second output. The second digital-to-RF converter device is configured to convert the second IQ data subsignal to a second RF signal and to output the second RF signal on the second output.

In this application, radio frequency may mean the frequency range from 1 MHz to 300 GHz. With a quadrature digital power amplifier system according to the first aspect of the present invention the layout of the first and second digital-to-RF converter devices may be tremendously simplified compared to conventional solutions. Furthermore the binary-coding- glitches may be made significantly smaller since each segment (e.g. each digital-to-RF converter device) can have its own binary coded units instead of reusing the same binary units over the whole array. Also the subarray (e.g. the second digital-to-RF converter device) that is used only for the signal peaks can have a lower resolution and lower efficiency than the subarray(s) (e.g. the first digital-to-RF converter device) used for the main part of the signal, because the peaks are rare and thus do not affect the average efficiency or the spectral content as much as the rest of the signal.

In a first possible implementation form of a quadrature digital power amplifier system according to the first aspect, the first output and the second output are connected to each other. In order to provide an RF signal corresponding to the IQ data signal, the first RF signal and the second RF signal are combined.

In a second possible implementation form of a quadrature digital power amplifier system according to the first implementation form of the first aspect or to the quadrature digital power amplifier system as such, the IQ data signal is decomposed so that the sum of the first IQ data subsignal and the second IQ data subsignal in the given sample is equal to the IQ data signal in the given sample. Such a decomposition of the IQ data signal may be relatively easily implemented in some cases. It is, however, possible to decompose the IQ data signal in other ways so that the sum of the first IQ data subsignal and the second IQ data subsignal in the given sample is not equal to the IQ data signal in the given sample. It is for example possible to have some scale factor between the IQ data signal and the sum of the first IQ data subsignal and the second IQ data subsignal or to omit the highest peaks of the IQ data signal as long as the resulting RF signal resembles the ideal RF signal.

In a third possible implementation form of a quadrature digital power amplifier system according to any of the preceding implementation forms of the first aspect or to the quadrature digital power amplifier system as such, the signal level of the IQ data signal is defined as l ² +Q ² , ² +Q ² , or abs(l) +abs(Q), wherein I is the value of the I data signal and

Q is the value of the Q data signal. In principle this corresponds to the amplitude of the signal. This is a favourable definition of the signal level.

In a fourth possible implementation form of a quadrature digital power amplifier system according to any of the preceding implementation forms of the first aspect or to the quadrature digital power amplifier system as such, the decomposition circuit is configured to decompose the I data signal into at least a first I data subsignal and a second I data subsignal, if the signal level of the I data signal in a given sample is above a second threshold level for the I data signal, and decompose the Q data signal into at least a first Q data subsignal and a second Q data subsignal, if the signal level of the Q data signal in a given sample is above a third threshold level for the Q data signal. The first I data subsignal and the first Q data subsignal constitute the first IQ data subsignal, wherein the second I data subsignal and the second Q data subsignal constitute the second IQ data subsignal, and wherein the signal level for the I data signal is defined as abs(l) and the signal level for the Q data signal is defined as abs(Q). Such a decomposition of the IQ data signal may be more easily implemented in some cases. In a fifth possible implementation form of a quadrature digital power amplifier system according to the fourth possible implementation form of the first aspect the IQ data signal is decomposed so that the sum of the first I data subsignal and the second I data subsignal in the given sample is equal to I data signal in the given sample, and so that the sum of the first Q data subsignal and the second Q data subsignal in a given sample is equal to the Q data signal in the given sample. Similarly to the second possible implementation form it is possible to decompose the I data signal and the Q data signal in other ways. It is for example possible to have some scale factor between the l/Q data signal and the sum of the first l/Q data subsignal and the second l/Q data subsignal as long as the resulting RF signal fulfils the specific system requirements. Such a decomposition of the IQ data signal may be relatively easily implemented in some cases.

In a sixth possible implementation form of a quadrature digital power amplifier system according to any of the preceding implementation forms of the first aspect or to the quadrature digital power amplifier system as such, at least one of the first and second digital- to-RF converter devices comprises a corresponding decoder connected between the converter input and the digital power amplifier. By having a decoder between the converter input and the digital power amplifier the layout of the digital power amplifier may be simplified. Possibly, also the overall layout of the digital-to-RF converters may be simplified.

In a seventh possible implementation form of a quadrature digital power amplifier system according to the sixth possible implementation form of the first aspect, at least a part of one of the digital power amplifiers is unary coded and the corresponding decoder is configured to provide a unary coded signal to said at least a part of one of the digital power amplifiers. By using unary coding binary glitches are avoided. Due to the simplified layout of each digital power amplifier, which is due to the use of at least two digital power amplifier, the digital power amplifiers are as a whole still less complicated in their layout in comparison with conventional digital power amplifiers.

In an eighth possible implementation form of a quadrature digital power amplifier system according to the sixth or seventh possible implementation form of the first aspect at least a part of one of the first digital power amplifier and the second digital power amplifier is binary coded, wherein the corresponding decoder is configured to provide a binary coded signal to said at least part of one of the first digital power amplifier and the second digital power amplifier. Binary coding simplifies the layout of the digital power amplifier even further. The glitches due to the binary coding are still smaller in the eighth possible implementation form compared to the binary glitches in conventional systems.

A ninth possible implementation form of a quadrature digital power amplifier system according to any of the preceding implementation forms of the first aspect or to the quadrature digital power amplifier system as such, further comprises a first digital quadrature modulator being connected between the first converter input and the first digital power amplifier, and being configured to perform at least one of upconverting and combining the I data subsignal of the first IQ data subsignal and the Q data subsignal of the first IQ data subsignal in the digital domain , and a second digital quadrature modulator being connected between the second converter input and the second digital power amplifier, and being configured to perform at least one of upconverting and combining the I data subsignal of the first IQ data subsignal and the Q data subsignal of the first IQ data subsignal in the digital domain. Combining the I data subsignals and Q data subsignals in the digital domain, instead of the analog domain in the digital power amplifier output, may have advantage in some implementations. For instance, IQ power combination in the analog domain may result in power loss due to quadrature signal cancellation. A tenth possible implementation form of a quadrature digital power amplifier system according to any of the preceding implementation forms of the first aspect or to the quadrature digital power amplifier system as such, further comprises up-sampling blocks, wherein one up-sampling block is arranged between each one of the suboutputs and the corresponding converter inputs, wherein each up-sampling block is configured to up-sample and filter each IQ data subsignal. The up-sampling blocks contribute to providing a more accurate digital signal. This improves the output RF signals from the outputs.

In an eleventh possible implementation form of a quadrature digital power amplifier system according to any of the preceding implementation forms of the first aspect or to the quadrature digital power amplifier system as such, the decomposition circuit is configured to decompose the IQ data signal into the first IQ data subsignal and the second IQ data subsignal such that signal components (e.g. signal peaks) above the first threshold level in the IQ data signal are comprised in the second IQ data subsignal and are not comprised in the first IQ data subsignal. By such a decomposition it can be achieved that the digital-to-RF converter devices can be efficiently chosen to with regards to needed performance verses implementation cost.

In a twelfth possible implementation form of a quadrature digital power amplifier system according to any of the preceding implementation forms of the first aspect or to the quadrature digital power amplifier system as such, the second digital-to-RF converter device has any of a lower resolution and a lower efficiency than the first digital-to-RF converter device. As the second digital-to-RF converter device is configured to support only the rare signal envelope peaks and due to the rare occurrence of the signal envelope peaks, the output RF signal may still resemble the correct RF signal within the specification limits even with a reduced resolution and/or efficiency. A digital-to-RF converter device with lower resolution and/or lower efficiency can be realized with transistors and/or topologies that support higher voltages, such as thick-oxide transistors and cascade switch topologies. According to a second aspect of the present invention a method for converting an IQ data signal to an RF signal is provided. The method comprises the steps of receiving an IQ data signal comprising an I data signal and a Q data signal, decomposing the IQ data signal into at least a first IQ data subsignal and a second IQ data subsignal if the signal level of the IQ data signal in a given sample is above a first threshold level, converting the first IQ data subsignal to a first RF signal, and converting the second IQ data subsignal to a second RF signal.

With a method according to the second aspect of the present invention the implementation of the conversion is simplified in comparison with conventional methods. Furthermore the binary-coding-glitches may be made significantly smaller since each segment can be implemented with its own binary coded units instead of reusing the same binary units over the whole array. In a first possible implementation form of a method according to the second aspect the IQ data signal is decomposed so that the sum of the first IQ data subsignal and the second IQ data subsignal in the given sample is equal to the IQ data signal in the given sample. Such a decomposition of the IQ data signal may be relatively easily implemented in most cases. It is, however, possible to decompose the IQ data signal in other ways so that the sum of the first IQ data subsignal and the second IQ data subsignal in the given sample is not equal to the IQ data signal in the given sample. It is for example possible to have some scale factor between the IQ data signal and the sum of the first IQ data subsignal and the second IQ data subsignal as long as the resulting RF signal resembles the ideal RF signal. In a second possible implementation form of a method according to the first possible implementation form of the second aspect or to the method for converting an IQ data signal to an RF signal as such, the signal level of the IQ data signal is defined as l ² +Q ² , ² + Q ² , or abs(l) +abs(Q). This corresponds in principal to the amplitude of the IQ data signal. In a third possible implementation form of a method according to any of the preceding implementation forms of the second aspect or to the method for converting an IQ data signal to an RF signal as such, the step of decomposing comprises the step of decomposing the I data signal into at least a first I data subsignal and a second I data subsignal, if the signal level of the I data signal in a given sample is above a second threshold level for the I data signal, and decomposing the Q data signal into at least a first Q data subsignal and a second Q data subsignal, if the signal level of the Q data signal in a given sample is above a third threshold level for the Q data signal, wherein the first I data subsignal and the first Q data subsignal constitutes the first IQ data subsignal IQ1 , wherein the second I data subsignal and the second Q data subsignal constitutes the second IQ data subsignal, and wherein the signal level for the I data signal is defined as abs(l) and the signal level for the Q data signal is defined as abs(Q). Such a decomposition of the IQ data signal may be more easily implemented in some cases.

In a fourth possible implementation form of a method according to any of the preceding implementation forms of the second aspect or to the method for converting an IQ data signal to an RF signal as such, the IQ data signal is decomposed so that the sum of the first I data subsignal and the second I data subsignal in the given sample is equal to I data signal in the given sample, and so that the sum of the first Q data subsignal and the second Q data subsignal in a given sample is equal to the Q data signal in the given sample. Such a decomposition of the IQ data signal may be relatively easily implemented in some cases. Similarly to the first possible implementation form it is possible to decompose the I data signal and the Q data signal in other ways. It is for example possible to have some scale factor between the l/Q data signal and the sum of the first l/Q data subsignal and the second l/Q data subsignal as long as the resulting RF signal resembles the ideal RF signal. In a fifth possible implementation form of a method according to the fourth possible implementation form of the second aspect, the method comprises the step of unary coding at least a part of one of the first IQ data subsignal and the second IQ data subsignal before the conversion step. This step reduces the binary coding glitches. A part of the first IQ data subsignal and the second IQ data subsignal may be kept binary coded to reduce the amount of control signals.

In a sixth possible implementation form of a method according to any of the preceding implementation forms of the second aspect or to the method for converting an IQ data signal to an RF signal as such, the method comprises the steps of upconverting and combining the I data subsignal of the first IQ data subsignal and the Q data subsignal of the first IQ data subsignal in the digital domain, and upconverting and combining the I data subsignal of the first IQ data subsignal and the Q data subsignal of the first IQ data subsignal in the digital domain. Combining the the I data signals and Q data signals in the digital domain, instead of the analog domain in the digital power amplifier output, may be advantageous in some implementations. For instance, IQ power combination in the analog domain may result in power loss due to quadrature signal cancellation. In a seventh possible implementation form of a method according to any of the preceding implementation forms of the second aspect or to the method for converting an IQ data signal to an RF signal as such, the method comprises the steps up-sampling and filtering each IQ data subsignal. The up-sampling blocks contribute to providing a more accurate digital signal and thus better resulting RF signals.

In an eighth possible implementation form of a method according to any of the preceding implementation forms of the second aspect or to the method for converting an IQ data signal to an RF signal as such, the IQ data signal is decomposed into the first IQ data subsignal and the second IQ data subsignal such that signal components (e.g. signal peaks) above the first threshold level in the IQ data signal are comprised in the second IQ data subsignal and are not comprised in the first IQ data subsignal.

According to a third aspect of the present invention a transmitter device for a wireless communication system is provided which comprises a quadrature digital power amplifier system according to the first aspect of the present invention. The third aspect provides similar advantages as was described for the first aspect.

According to a fourth aspect of the present invention a computer program is provided which computer program comprises a program code for performing a method according to the second aspect of the invention, when the computer program runs on a computer. The third aspect provides similar advantages as was described for the first aspect.

SHORT DESCRIPTION OF THE DRAWINGS

Fig. 1 shows schematically a quadrature digital power amplifier system according to a first embodiment of the present invention, which quadrature digital power amplifier system comprises digital-to-RF converter devices. Fig 2 shows a digital-to-RF converter device according to an embodiment of the present invention.

Fig 3 shows a digital-to-RF converter device according to another embodiment of the present invention. Fig 4 shows a digital-to-RF converter device according to a further embodiment of the present invention.

Fig. 5 shows schematically a quadrature digital power amplifier system according to a second embodiment of the present invention.

Fig. 6 shows schematically a quadrature digital power amplifier system according to a third embodiment of the present invention. Fig. 7 is a flow diagram over a method according to an embodiment of the present invention.

Fig. 8 shows schematically a transmitter device for a wireless communication system comprising a quadrature digital power amplifier system according to any one of the embodiments described above.

DETAILED DESCRIPTION OF THE EMBODIMENTS

In the following detailed description of the embodiments of the invention, the same reference numeral will be used for the corresponding feature in the different drawings.

Fig. 1 shows schematically a quadrature digital power amplifier system 100 according to an embodiment of the present invention. The quadrature digital power amplifier system 100 comprises a decomposition circuit 102 as shown in Fig. 1 . The decomposition circuit 102 comprises an input 104 configured to receive an IQ data signal IQ comprising an I data signal and a Q data signal, and a first suboutput 106 and a second suboutput 108. The IQ data signal is in Fig. 1 shown as a number of arrows, the arrows representing the different bits of the digital signal. The decomposition circuit 102 receives the IQ data signal and is configured to decompose the IQ data signal IQ into at least a first IQ data subsignal IQ1 and a second IQ data subsignal IQ2 if the signal level of the IQ data signal IQ in a given sample is above a first threshold level. The decomposition circuit 102 is further configured to output the first IQ data subsignal IQ1 on the first suboutput 106 and the second IQ data subsignal IQ2 on the second suboutput 108. The subsignals IQ1 and IQ2 are also shown with as a number of arrows, the arrows representing the different bits of the digital signal. As an example the decomposition of the IQ data signal into the two IQ data subsignals IQ1 and IQ2 can be performed such that all signal components of the IQ data signal IQ above the first threshold level are comprised in the second IQ data subsignal IQ2 but not in the first IQ data subsignal IQ1 and all signal components of the IQ data signal IQ below or equal to the first threshold level are comprised in the first IQ data subsignal IQ1 but not in the second IQ data subsignal IQ2. The signal level of the IQ data signal IQ is according to an embodiment defined as l ² +Q ² , l ² + Q ² , or abs(l) +abs(Q).

I is the value of the I data signal and Q is the value of the Q data signal. The quadrature digital power amplifier system 100 also comprises a first digital-to-RF converter device 1 10 comprising a first converter input 1 12 connected to the first suboutput 106 of the decomposition circuit 102, a first digital power amplifier DPA1 connected to the first converter input 1 12, and a first output 1 14. The first digital-to-RF converter device 1 10 is configured to convert the first IQ data subsignal IQ1 to a first RF signal and to output the first RF signal on the first output 1 14. Said conversion to a first RF signal is performed by the first digital power amplifier DPA1 in the embodiment shown in Fig. 1 . The quadrature digital power amplifier system 100 also comprises at least a second digital-to-RF converter device 1 16 comprising a second converter input 1 18 connected to the second suboutput 108 of the decomposition circuit 102, a second digital power amplifier DPA2 connected to the second converter input 1 18, and a second output 120. The second digital-to-RF converter device 1 16 is configured to convert the second IQ data subsignal IQ2 to a second RF signal and to output the second RF signal on the second output 120. Said conversion to a second RF signal is performed by the second digital power amplifier DPA2 in the embodiment shown in Fig. 1 . The first output 1 14 comprises a first positive output 1 14a and a first negative output 1 14b while the second output 120 comprises a second positive output 120a and a second negative output 120b. As can be seen in the embodiment shown in Fig 1 the first output 1 14 and the second output 120 are connected to each other. The common output 140 comprises a common positive output 140a which is connected to the first positive output 1 14a and the second positive output 120a. The common output 140 also comprises a common negative output 140b which is connected to the first negative output 1 14b and the second negative output 120b. The RF signal is output on the common output 140. The common output may be connected to, e.g., at least one of an impedance matching network, power combiner, filter, switch, and an antenna on a transmitter. Although the outputs shown in Figure 1 and the following figures are differential outputs according to further embodiments, the outputs can be single ended outputs. Fig. 2 shows a digital-to-RF converter device 1 10, 1 16, according to an embodiment of the present invention. The digital-to-RF converter device 1 10, 1 16, in Fig. 2 comprises a corresponding decoder 122, 124, connected between the converter input 1 12, 1 18, and the digital power amplifier DPA1 , DPA2. The decoder 122, 124, is separate from the digital power amplifier DPA1 , DPA2, and decodes part of the IQ data subsignal IQ1 , IQ2 from binary coded digital signals into a combination of unary coded and binary coded signals which are transmitted to the digital power amplifier DPA1 , DPA2. Unary coding increases the complexity but reduces the amount of glitches caused by binary coding. The increased complexity is represented by an increased number of arrows from the decoder 122, 124, in comparison with the number of arrows on the input of the decoder 122, 124.

Figs. 3 and 4 show a digital-to-RF converter device 1 10, 1 16, according to further embodiments of the present invention. The digital-to-RF converter device 1 10, 1 16, of Fig. 3 and Fig. 4 additionally comprises a digital quadrature modulator 130, 132, being connected between the first converter input 1 12, 1 18 and the digital power amplifier DPA1 , and being configured to upconvert the IQ data subsignal IQ1 , IQ2, to an RF signal in the digital domain.

In the embodiment shown in Fig 3, the digital quadrature modulator 130, 132, is connected between the converter input 1 12, 1 18, and the decoder 122, 124. In Fig. 4 the digital quadrature modulator 130, 132, is arranged between the decoder 122, 124 and the digital power amplifier DPA1 , DPA2. The choice between the embodiment in Fig. 3 and Fig. 4 depends among other things on how the decoder 122, 124, is arranged. In case the decoder is configured to decode the IQ data subsignal IQ1 , IQ2, into unary coded signals the decoder 122, 124, has to output a large number of control signals. In this case it might be favourable to have the decoder arranged after the digital quadrature modulator 130, 132, as shown in the embodiment in Fig. 3.

Fig. 5 shows schematically a quadrature digital power amplifier system 100 according to a further embodiment of the present invention. In Fig. 5 the shape with respect to time of the IQ data signal and the IQ data subsignals IQ1 , IQ2, are shown schematically. The difference between the embodiment shown in Fig. 1 and the embodiment shown in Fig. 5 is that the quadrature digital power amplifier system 100 shown in Fig. 5 comprises up-sampling blocks UP1 , UP2, wherein one up-sampling block UP1 , UP2, is arranged between each one of the suboutputs 106, 108, and the corresponding converter inputs 1 12, 1 18, wherein each up- sampling block UP1 , UP2, is configured to up-sample and filter each IQ data subsignal. As is shown in Fig. 5, the IQ data signal is decomposed by the decomposition circuit 102 into a first IQ data subsignal IQ1 and a second IQ data subsignal IQ2. The up-sampling blocks UP1 , UP2, then up-samples the first IQ data subsignal IQ1 and the second IQ data subsignal IQ2 into a first up-sampled IQ data subsignal IQ1 ' and a second up-sampled IQ data subsignal IQ2 ^' . The first up-sampled IQ data subsignal IQ1 ' and a second up-sampled IQ data subsignal IQ2 ^' are then converted by the first digital-to-RF converter device 1 10 and the second digital-to-RF converter device 1 16 to a first RF signal and a second RF signal which are output on the first output 1 14 and the second output 120, respectively. Fig. 6 shows schematically a quadrature digital power amplifier system 100 according to a third embodiment of the present invention. The quadrature digital power amplifier system 100 in Fig. 6 comprises a decomposition circuit 102 which is configured to decompose the IQ data signal IQ into four IQ data subsignals IQ1 , IQ2, IQ3, IQ4. The quadrature digital power amplifier system 100 in Fig. 6 comprises four up-sampling blocks UP1 , UP2, UP3, UP4, which are configured to up-sample and filter each IQ data subsignal IQ1 , IQ2, IQ3, IQ4. The quadrature digital power amplifier system 100 also comprises a first digital-to-RF converter device 1 10 comprising a first converter input 1 12 and a first output 1 14, a second digital-to- RF converter device 1 16 comprising a second converter input 1 18 and a second output 120, a third digital-to-RF converter device 142 comprising a third converter input 144 and a third output 146, and a fourth digital-to-RF converter device 148 comprising a fourth converter input 150 and a fourth output 152. The first output 1 14, the second output 120, the third output 144 and the fourth output 148 are connected into a common output 140 on which an RF-signal RF, is output. Thus, in operation the IQ data signal IQ is decomposed into four IQ data subsignals IQ1 , IQ2, IQ3, IQ4. The IQ data subsignals are up-sampled by the up-sampling blocks UP1 , UP2, UP3, UP4 and input into the first digital-to-RF converter device 1 10, the second digital-to-RF converter device 1 16, the third digital-to-RF converter device 142, and the fourth digital-to-RF converter device 146. By dividing the IQ data signal IQ into four IQ data subsignals IQ1 , IQ2, IQ3, IQ4, instead of only two the complexity of the converter devices 1 10, 1 16, 142, 146, may be further decreased.

In the embodiment shown in Fig. 5, the decomposition circuit 102 is configured to decompose the IQ data signal IQ into at least a first IQ data subsignal IQ1 and a second IQ data subsignal IQ2 if the signal level of the IQ data signal IQ in a given sample is above a first threshold level, and to output the first IQ data subsignal IQ1 on the first suboutput 106 and the second IQ data subsignal IQ2 on the second suboutput 108. The signal level of the IQ data signal IQ is according to an embodiment defined as l ² +Q ² , ² + ζ) ² , or abs(l) +abs(Q).

In the embodiment shown in Fig. 6, three different threshold levels are used, wherein the IQ data signal is decomposed into a first IQ data subsignal IQ1 , a second IQ data subsignal IQ2, a third IQ data subsignal IQ3, and a fourth IQ data subsignal IQ4. As an example the decomposition of the IQ data signal into the four IQ data subsignals IQ1 , IQ2, IQ3 and IQ4, can be performed in the following way. All signal components of the IQ data signal IQ below or equal to the first threshold level are comprised in the first IQ data subsignal IQ1 but not in the second IQ data subsignal IQ2, the third IQ data subsignal IQ3, or the fourth IQ data subsignal IQ4. All signal components of the IQ data signal IQ above the first threshold level and below or equal to the second threshold level are comprised in the second IQ data subsignal IQ2, but not in the first IQ data subsignal IQ1 , the third IQ data subsignal IQ3, or the fourth IQ data subsignal IQ4. All signal components of the IQ data signal IQ above the second threshold level and below or equal to the third threshold level are comprised in the third IQ data subsignal IQ3, but not in the first IQ data subsignal IQ1 , the second IQ data subsignal IQ2, or the fourth IQ data subsignal IQ4. Finally, all signal components of the IQ data signal IQ above the third threshold level are comprised in the fourth IQ data subsignal IQ4, but not in the first IQ data subsignal IQ1 , the second IQ data subsignal IQ2, or the third IQ data subsignal IQ3. The IQ data subsignals IQ1 , IQ2, IQ3, IQ4, are shown both before and after the up-sampling blocks UP1 , UP2, UP3, UP4, in Fig. 6.

The embodiment shown in Fig. 6 may also be used to describe a method in which the I data signal is decomposed into at least a first I data subsignal and a second I data subsignal, if the signal level of the I data signal in a given sample is above a second threshold level for the I data signal, and the Q data signal is decomposed into at least a first Q data subsignal and a second Q data subsignal, if the signal level of the Q data signal in a given sample is above a third threshold level for the Q data signal. The first I data subsignal should then correspond to IQ1 and the first Q data subsignal should correspond to IQ2 and together constitute the first IQ data subsignal. The second I data subsignal should correspond to IQ3 and the second Q data subsignal should correspond to IQ4 and together constitute the second IQ data subsignal. The signal level for the I data signal is defined as abs(l) and the signal level for the Q data signal is defined as abs(Q). According to an embodiment, the sum of the first I data subsignal and the second I data subsignal in the given sample is equal to I data signal in the given sample. According to an embodiment, the sum of the first Q data subsignal and the second Q data subsignal in a given sample is equal to the Q data signal in the given sample.

Fig. 7 is a flow diagram over a method according to the present invention. The method 200 for converting an IQ data signal IQ to an RF signal comprises the steps of receiving 202 an IQ data signal IQ comprising an I data signal and a Q data signal, decomposing 204 the IQ data signal IQ into at least a first IQ data subsignal IQ1 and a second IQ data subsignal IQ2 if the signal level of the IQ data signal IQ in a given sample is above a first threshold level, converting 206 the first IQ data subsignal IQ1 to a first RF signal, and converting 208 the second IQ data subsignal IQ2 to a second RF signal.

Fig. 8 shows schematically a transmitter device 300 for a wireless communication system 400 comprising a quadrature digital power amplifier system 100 according to any one of the embodiments described above. The wireless communication system 400 also comprises a base station 500 which may also comprise a quadrature digital power amplifier system 100 according to any one of the embodiments described above. The dotted arrow A1 represents transmissions from the transmitter device 300 to the base station 500. The full arrow A2 represents transmissions from the base station 500 to the transmitter device 300. The present transmitter device 300 may be any of a User Equipment (UE) in Long Term Evolution (LTE), mobile station (MS), wireless terminal or mobile terminal which is enabled to communicate wirelessly in a wireless communication system, sometimes also referred to as a cellular radio system. The UE may further be referred to as mobile telephones, cellular telephones, computer tablets or laptops with wireless capability. The UEs in the present context may be, for example, portable, pocket-storable, hand-held, computer-comprised, or vehicle-mounted mobile devices, enabled to communicate voice or data, via the radio access network, with another entity, such as another receiver or a server. The UE can be a Station (STA), which is any device that contains an IEEE 802.1 1 -conformant Media Access Control (MAC) and Physical Layer (PHY) interface to the Wireless Medium (WM).

The present base station 500 may be a (radio) network node or an access node or an access point or a base station, e.g., a Radio Base Station (RBS), which in some networks may be referred to as transmitter, "eNB", "eNodeB", "NodeB" or "B node", depending on the technology and terminology used. The radio network nodes may be of different classes such as, e.g., macro eNodeB, home eNodeB or pico base station, based on transmission power and thereby also cell size. The radio network node can be a Station (STA), which is any device that contains an IEEE 802.1 1 -conformant Media Access Control (MAC) and Physical Layer (PHY) interface to the Wireless Medium (WM).