TECHNICAL FIELD

The present invention relates to a device for processing digital signals and to a corresponding method. In particular, the device of the present invention includes a digital signal source for providing digital signals, and a converter circuit for converting the digital signals. The present invention addresses specifically an issue of supply current variations of the device occurring when the converter circuit processes the digital signals.

BACKGROUND

A conventional device for processing digital signals suffers from current supply variations while processing the digital signals. The problem is in particular that the current requirements of the device, specifically of a converter circuit in the device, depend on codewords of the digital signal. That means a supply voltage (and supply current) of the converter circuit changes as the codewords change. These supply changes severely impact the performance of the conventional device.

As an example of such a converter circuit, a Digital-to-Analog Converter (DAC) is one of the most important building blocks of many conventional devices for digital signal processing. DACs are used in communication systems but also in any other place where an interface with the analog physical world is required.

A DAC consists typically of individual cells that contribute to its output signal. In particular, the number of cells that are enabled, together with their size(s), determines the output signal. Switching the cells ON and/or OFF draws current from the supply. Therefore, the current drawn by the DAC from the supply is a function of the analog output signal (and thus of the codeword of the digital input signal). In fact, the current consumption tracks the output signal. In combination with the finite impedance of the supply network, the supply voltage will also vary with the output signal. The variations in the supply voltage may specifically cause timing variations when the DAC cells switch, thereby causing harmonic distortion and a performance degradation. Several solutions are possible for solving this issue. For example, the impedance of the supply could be lowered.

One solution would thereby be to reduce the output impedance of the voltage supply (typically a low-dropout regulator (LDO)). However, this has the drawback that the size of the LDO and/or the current consumption would increase significantly. This solution would accordingly increase the power consumption and also the production costs of the device.

Another solution would increase the speed of the voltage regulator, which would also come with a significant increase in power consumption.

Another solution would increase an on-chip decoupling capacitance. However, the required area for this capacitance would be very large, thus increasing production costs.

Another solution would add an off-chip decoupling capacitance. However, this would require an external component and a low impedance connection from the device to the external component, thus again increasing production costs.

A further solution would keep the consumption constant. For example, dummy DAC cells could be used, which do not contribute to the output signal, but have the same current consumption as the normal cells. Using such dummy cells, the total number of cells that are switched in the DAC may be kept constant. The major drawback, however, is the overall increased current consumption. That is, the DAC would always consume the current associated with a maximal output swing.

SUMMARY

The above-described solutions also suffer from the drawback that none of them allows for over-compensating or pre-distorting the DAC. For a DAC, digital pre-distortion may be applied, however, this has the drawback that a lot of current is required because of required digital calculations. In view of the above-mentioned problems and disadvantages, the object of the present invention is to improve the conventional devices and methods for digital signal processing, and to avoid the drawbacks of the above-mentioned solutions.

More precisely, the present invention aims at providing a device and method for digital signal processing with reduced supply variations. In particular, a converter circuit of the device does not show supply variations in dependence of a codeword of a digital signal to be converted. The device and method of the present invention also allow over compensating and/or pre-distorting the converter circuit. Additionally, the device and method of the present invention function independent of process, voltage and/or temperature changes (so called PVT variations).

The object of the present invention is achieved by the solution provided in the enclosed independent claims. Advantageous implementations of the present invention are further defined in the dependent claims.

The idea described in the present disclosure is to re-use digital information for a feed forward correction on the supply of the converter circuit. In other words, the present disclosure proposes to make use of the fact that the required current draw of the converter circuit is known to the device - or to a digital signal source controlling the converter circuit. This allows the device - or digital signal source - to instruct a feed forward circuit (also referred to as a Current Injection Bank, CIB) to provide the current required to the converter circuit.

A first aspect provides a device for processing digital signals, comprising a digital signal source configured to output codewords, a converter circuit configured to generate an output signal based on a first codeword received from the digital signal source, and a feed forward circuit configured to generate an output current based on a second codeword received from the digital signal source, wherein the output current of the feed forward circuit is connected to a current supply of the converter circuit, and wherein the digital signal source is configured to generate the second codeword based on the first codeword to compensate variations of a supply current of the converter circuit. The output current of the feed forward circuit is connected to the current supply of the converter circuit, in order to provide the required current for converting each first codeword provided by the digital signal source to the converter circuit. Thus, the device can correct - by means of generating a second codeword for each first codeword - for a time varying and codeword-dependent current consumption of the converter circuit. Thereby, the device can completely eliminate harmonic distortion due to supply ripple.

In addition, the device allows varying the supply voltage for the converter circuit based on the second codeword supplied to the feed forward circuit. This allows performing an analog pre-distortion by increasing or decreasing the supply voltage level, in order to further reduce spurs present in the device. The pre-distortion is thereby not limited to reducing spurs or artefacts arising from supply variations, but can reduce also other spurs present in the device, e.g. spurs caused by Local Oscillator (LO) leakage.

The device of the first aspect has several further advantages. For instance, the overall power consumption of the device does not increase much over a conventional device, since the converter circuit current is now supplied by the feed forward circuit instead of a voltage supply (like an LDO). Further, the requirement on the feed forward circuit is relatively relaxed, which supports a reduction of the device area and of a power required by the device.

In an implementation form of the first aspect, the digital signal source is configured to generate the second codeword so as to reduce supply voltage variations to which the performance of the converter circuit is sensitive.

In a further implementation form of the first aspect, the digital signal source is configured to generate the second codeword based on a current consumption of the converter circuit when generating the output signal based on the first codeword.

In a further implementation form of the first aspect, the digital signal source is configured to generate the second codeword such that the output current of the feed forward circuit at least matches a current required by the converter circuit for generating the output signal based on the first codeword. The above implementation forms use the knowledge of the current consumption required by the converter circuit to convert a first codeword for generating the second codeword. The second codeword is generated in a way that it leads to an output current of the feed forward circuit that compensates, when being fed to the current supply of the converter circuit, the supply variations that would arise while converting the first codeword. This effectively reduces supply variations of the converter circuit, and leads to a significant improvement of the performance of the device.

In a further implementation form of the first aspect, the digital signal source is configured to generate the second codeword based further on a configuration of the converter circuit.

In a further implementation form of the first aspect, the configuration of the converter circuit includes a number of a plurality of cells that contribute to the generation of the output signal based on the first codeword.

Converter circuits, like DACs, typically employ a number of current switching cells to convert digital input signals to analog output signals. A current circuit in each cell is operated to provide a defined current to a current switch in the cell, and a driver circuit in each cell causes it, in response to an input digital data signal, to produce an analog output from the defined current from the current circuit. Accordingly, the above implementation forms take further into account the configuration and design of the converter circuit. This leads to an even more effective suppression of supply variations, and thus to a further improvement of the performance of the device.

In a further implementation form of the first aspect, the digital signal source is configured to generate the second codeword based further on an operating frequency of the converter circuit.

This allows an even more effective suppression of supply variations, and is thus a further measure for improving the performance of the device.

In a further implementation form of the first aspect, the converter circuit includes a first DAC and the output current of the feed forward circuit is provided to a current supply of the first DAC. In a further implementation form of the first aspect, the feed forward circuit includes a second DAC.

The above implementation forms of the device accordingly use up to two DACs, and are particularly simple to realize. The supply variations of the first DAC can be completely eliminated by using the second DAC.

In a further implementation form of the first aspect, the feed forward circuit further includes a bias generation section connected to the second DAC, and the bias generation section comprises a mirroring device configured to mirror the operation of the second DAC, wherein the mirroring device includes a feedback circuit.

Thereby, the device performance can be designed to operate independently of process, voltage or temperature changes (PVT variations).

In a further implementation form of the first aspect, the device further comprises a supply unit, including a voltage regulator, configured to provide a DC output voltage to a current supply of the converter circuit.

In a further implementation form of the first aspect, the digital signal source is configured to generate the second codeword such that the output current of the feed forward circuit together with the output current of the supply unit exceeds a current required by the converter circuit for generating the output signal based on the first codeword.

In a further implementation form of the first aspect, the digital signal source is configured to generate the second codeword to perform a pre-distortion of the output signal generated by the converter circuit.

According to the above implementation forms, over-compensation and/or pre-distortion of the converter circuit are enabled. The performance of the device can thereby be further improved. For instance, by further reducing spurs present in the device. Notably, the pre distortion is not be limited to spurs or artefacts arising from supply variations, but can also address other spurs present in the device. A second aspect comprises providing, by a digital signal source,

a first codeword to a converter circuit, in order to generate an output signal, providing, by the digital signal source, a second codeword to a feed forward circuit, in order to generate an output current, wherein the output current of the feed forward circuit is supplied to the converter circuit, and wherein the second codeword is generated by the digital signal source based on the first codeword to compensate variations of a supply current of the converter circuit.

In an implementation form of the second aspect, the method comprises generating the second codeword so as to reduce supply voltage variations to which the performance of the converter circuit is sensitive.

In a further implementation form of the second aspect, the method comprises generating the second codeword based on a current consumption of the converter circuit when generating the output signal based on the first codeword.

In a further implementation form of the second aspect, the method comprises generating the second codeword such that the output current of the feed forward circuit at least matches a current required by the converter circuit for generating the output signal based on the first codeword.

In a further implementation form of the second aspect, the method comprises generating the second codeword based further on a configuration of the converter circuit.

In a further implementation form of the second aspect, the configuration of the converter circuit includes a number of a plurality of cells that contribute to the generation of the output signal based on the first codeword.

In a further implementation form of the second aspect, the method comprises generating the second codeword based further on an operating frequency of the converter circuit. In a further implementation form of the second aspect, the converter circuit includes a first digital-to-analog converter (DAC) and the output current of the feed forward circuit is provided to a current supply of the first DAC. In a further implementation form of the first aspect, the first device includes a first DAC connected in series at least with a power amplifier, and the output current of the second device is provided to a current supply of the power amplifier.

In a further implementation form of the second aspect, the feed forward circuit includes a second DAC.

In a further implementation form of the second aspect, the feed forward circuit further includes a bias generation section connected to the second DAC, and the bias generation section comprises a mirroring device mirroring the operation of the second DAC, wherein the mirroring device includes a feedback circuit.

The mirroring device particularly mirrors the operation of the second DAC, in order to keep constant the output current under changing operating conditions of the second DAC, such as supply voltage temperature and/or operating frequency.

In a further implementation form of the second aspect, the method further comprises providing a DC output voltage to a current supply of the converter circuit.

In a further implementation form of the second aspect, the digital signal source is configured to generate the second codeword such that the output current of the feed forward circuit together with the output current of the supply unit exceeds a current required by the converter circuit for generating the output signal based on the first codeword.

In a further implementation form of the second aspect, the digital signal source is configured to generate the second codeword to perform a pre-distortion of the output signal generated by the converter circuit. The method of the second aspect and its implementation forms achieve the same advantages and effects as described above with respect to the device of the first aspect and its implementation forms. It has to be noted that all devices, elements, units and means described in the present application could be implemented in the software or hardware elements or any kind of combination thereof. All steps which are performed by the various entities described in the present application as well as the functionalities described to be performed by the various entities are intended to mean that the respective entity is adapted to or configured to perform the respective steps and functionalities. Even if, in the following description of specific embodiments, a specific functionality or step to be performed by external entities is not reflected in the description of a specific detailed element of that entity which performs that specific step or functionality, it should be clear for a skilled person that these methods and functionalities can be implemented in respective software or hardware elements, or any kind of combination thereof.

BRIEF DESCRIPTION OF DRAWINGS

The above described aspects and implementation forms of the present invention will be explained in the following description of specific embodiments in relation to the enclosed drawings, in which

FIG. 1 shows a device according to an embodiment of the present invention. FIG. 2 shows a device according to an embodiment of the present invention using two DACs.

FIG. 3 shows a device according to an embodiment of the present invention using two DACs.

FIG. 4 shows a feed forward circuit of a device according to an embodiment of the present invention including a DAC and a bias generation section for biasing the DAC. FIG. 5 shows a feed forward circuit of a device according to an embodiment of the present invention including a DAC and a bias generation section for biasing the DAC. FIG. 6 shows a method according to an embodiment of the present invention.

FIG. 7 shows results of simulating the harmonic distortion of the converter circuit of a device according to an embodiment of the present invention. DETAILED DESCRIPTION OF EMBODIMENTS

FIG. 1 shows a device 100 according to an embodiment of the present invention. The device 100 is configured to process digital signals. The device 100 comprises a digital signal source 101, a converter circuit 102, and a feed forward circuit 105.

The digital signal source 101 is configured to output codewords of a digital signal, for instance, at least one first codeword 104 and at least one second codeword 107. The at least one first codeword 104 is supplied to the converter circuit 102, while the at least one second codeword 107 is supplied to the feed forward circuit 105. The digital signal source 101 may specifically be configured to supply multiple (different) first codewords 104 to the converter circuit 102 in a consecutive manner, e.g. changing codewords of a digital signal. At the same times it may supply multiple (different) second codewords 107 - one for each first codeword 104 - to the feed forward circuit 105. The converter circuit 102 is configured to generate an output signal 103 based on the at least one first codeword 104 received from the digital signal source 101. Thus, it can convert a digital signal into an analog signal. The feed forward circuit 105 is configured to generate an output current 106 based on the at least one second codeword 107 received from the digital signal source 101.

This output current 106 of the feed forward circuit 105 is advantageously connected (directly or indirectly) to a current supply 108 of the converter circuit 102. Moreover, the digital signal source 101 is configured to generate the at least one second codeword 107 based on the at least one first codeword 104. Thereby, the digital signal source 101 is able to compensate variations of a supply current of the converter circuit 105.

In an implementation, the digital signal source 101 is configured to generate the at least one second codeword 107, so as to reduce supply voltage variations to which the performance of the converter circuit 105 is sensitive. Thereby, the digital signal source 101 is configured to generate the second codeword 107 based on a current consumption of the converter circuit 105 when generating the output signal 103 based on the first codeword 104. The digital signal source 101 has knowledge of this codeword-dependent current consumption of the converter circuit 102, and can select the second codeword 107 accordingly. The digital signal source 101 may generate the second codeword 107 such that the output current 103 of the feed forward circuit 105 at least matches a current required by the converter circuit 102 for generating the output signal 103 based on the first codeword 104. Notably, over-compensating of the current required by the converter circuit 102 is also possible.

In the following, the above-described concept of the present invention is further illustrated for the case of a DAC, where the impact is on both the digital and analog subsystem. However, the invention is notably not limited to a DAC.

FIG. 2 shows a device 100 according to an embodiment of the present invention, which builds on the device 100 shown in FIG. 1. Same elements are labelled with same reference signs and function likewise. Accordingly, also the device 100 of FIG. 2 has the digital signal source 101 for providing the first and second codewords 104 and 107, respectively, the converter circuit 102 for outputting the output signal 103 based on the first codeword 104, and the feed forward circuit 105 for supplying the output current 106 produced based on the second codeword 107 towards the converter circuit 102.

FIG. 2 illustrates specifically a device 100, in which the converter circuit 102 is (or includes) a first DAC 202, in which the feed forward circuit 105 is (or includes) a second DAC 205, and in which the output current 106 of the feed forward circuit 105 is provided to a current supply 108 of the first DAC 202. Because the supply current uppiy required by the first DAC 202 is (indirectly) known to the digital signal source 101, an open loop correction can be performed with the second DAC 205. FIG. 2 shows in detail that the digital signal source 101 is a digital system including at least one data source 200. The digital signal source 101 is configured to output the first codeword 104, in order to set the output level of the first DAC 202. This causes the first DAC 202 to consume a certain amount of current related to the first codeword 104. At the same time, however, the digital signal source 101 is configured to program the second DAC 205 with a second (correction) codeword 107, thereby setting its output current 106 (Icon·). When this resulting output current 106 matches the current required by the first DAC 202, only a very small current (ILDO) would be drawn from a supply unit 201 (e.g. an LDO). The supply unit 201 may be part of the device 100 or not. As a result, voltage variations of the supply of the first DAC 202 are very strongly suppressed - if not eliminated. The second codeword 107 can be calculated from the first codeword 104, for instance taking into account the structure of the first DAC 202 (e.g. the construction of the first DAC 202 from current switching cells with unit weight or binary scaled weights).

The above-described current injection concept can also be applied to blocks of a device 100 other than the first DAC 202, as long as the current consumption profile of the block can be predicted.

FIG. 3 shows in this respect an exemplary device 100 according to the present invention, which builds on the device 100 shown in FIG. 1 and has several components in common with the device 100 shown in FIG. 2. Same elements are labelled with same reference signs and function likewise. Accordingly, also the device 100 of FIG. 3 has the digital signal source 101 for providing the first and second codewords 104 and 107, the converter circuit 102 for outputting the output signal 103 based on the first codeword 104, and the feed forward circuit 105 for supplying the output current 106 based on the second codeword 107 towards the converter circuit 102.

FIG. 3 illustrates specifically a device 100, in which the converter circuit 102 includes a first DAC 202 connected in series with a power amplifier 300 (as example of another block to which the current injection concept can be applied). Further, the feed forward circuit 105 is (or includes) a second DAC 202. The output current 106 of the feed forward circuit 105, which bases on the second codeword 107, is in this device 100 provided to a current supply of the power amplifier 300 (not of the first DAC 202 as in FIG. 2). Accordingly, the current injection concept is here used to reduce a supply ripple and distortion of the power amplifier 300, because this is correlated with the first codeword 104, and the second codeword 107 is generated on the second codeword 107 to compensate variations of a supply current of the power amplifier 300 of the converter circuit 102.

The converter circuit 105 may further include a bandpass filter 300. In particular, the first DAC 202 may be followed by the bandpass filter 301 and then the power amplifier 300, i.e. the bandpass filter 301 may be connected in series between the first DAC 202 and the power amplifier 300.

FIG. 3 shows in detail that the first codeword 104 from the digital signal source 101 is converted to an analog signal by the first DAC 102 (which is e.g. directly supplied by a supply unit, which may be an LDO, without any correction). The first DAC 202 drives the power amplifier 300. The current consumption of the power amplifier 300 depends on its input signal (i.e. on the codeword-dependent output signal of the first DAC 202) and will cause distortion. If, however, the second DAC 205 is connected, as in FIG. 3, in parallel with a supply unit 201 of the power amplifier 300, a supply ripple can be reduced. To this end, the second codeword 107 is specifically generated so as to compensate variations of a supply current of the power amplifier 300, which variations are caused indirectly by the first codeword 104. That is, knowledge of a current consumption required by the power amplifier 300 specifically for amplifying the output of the DAC 102 based on the first codeword 104 is used to generate the second codeword for the second DAC 205, in order to counter a supply ripple at the power amplifier 300. Thereby, the second codeword 107 generated for the second DAC 205 can, for example, be obtained from a one-time characterization of the device 100 or can be obtained from a calibration loop. Depending on the kind of non-linearities in the power amplifier 300, it is possible that this type of correction is less computationally intensive than a conventional digital pre-distortion solution.

In the devices 100 of FIG. 2 and FIG. 3, respectively, it is also possible to overcompensate by injecting more current than required by the first DAC 202. This may further reduce harmonic distortion. In effect, by overcompensating, a pre-distorting of the first DAC 202 in the analog domain is carried out. In an embodiment, only information regarding the current consumption of the first DAC 202 may be used when generating the second codeword 107 based on the first codeword 104, in order to compensate variations of a supply current of the second DAC 202. However, it is also possible that other information can be derived from the first codeword 104. Thus, it is also possible to reduce - in the same manner - other spurs present in the device 100 (e.g. LO leakage), using the second DAC 205.

The scale of the second DAC 205 may be programmable. This enables fine-tuning the required current of the first DAC 202 over process, voltage, temperature or other variations (PVT variations). The power consumption may also depend on the operation frequency of the first DAC 202, which can also be taken into account when changing the scale of the second DAC 205.

FIG. 4 shows an exemplary implementation of a feed forward circuit 105 for a device 100 according to an embodiment of the invention, e.g. a device 100 as shown in the FIGs. 1-3. This feed forward circuit 105 is specifically implemented for the case of a lO-bit RFDAC as the first DAC 202. The RFDAC consisted of 64 cells, which are used in pairs of two. Thus, a 5-bit second DAC 205 may be used in the feed forward circuit 105. The current switching cells 400 of this second DAC 205 may include PMOS devices, and are for switching a main supply current to an output (correction) current 106 provided towards the first DAC 202. FIG. 4 shows specifically that the second DAC 205 is implemented as a current DAC and is biased from a bias generation section 401 using a diode connected PMOS device 402.

While the feed forward circuit 105 of FIG. 4 allows compensating codeword-dependent supply variations of the first DAC 202, and thus improving the performance of the device 100, it is not yet optimized for PVT variations. In particular, in this exemplary implementation the drain voltage of the second DAC 205 (“vdd_dac”) can vary from the drain voltage of the diode device 402 in the bias generation section 401. Because of this, the amount of injected current could potentially vary over PVT variations. This would entail that during operation of the device 100, one would need to adjust the bias point. And such adjustments would require inputs such as operating voltages, temperature, etc. The settings of the second DAC 205 can be made independent of PVT variations as presented in the following. The presented solution notably could presents also a strong improvement on other bias strategies. In particular, the feed forward circuit 105 may include a mirroring device 501 its bias generation section 401 connected to the second DAC 205 - as shown in FIG. 5.

FIG. 5 shows again a feed forward circuit 105 of a device 100 according to an embodiment of the invention, e.g. a device 100 as shown in the FIGs. 1-3. The exemplary feed forward circuit 105 particularly builds on the one shown in FIG. 4.

FIG. 5 shows specifically that the mirroring device 501 includes the diode device 402 and a feedback circuit. When the reference voltage“vref” tracks the DAC supply voltage “vdd_dac”, a bias feedback loop can make the current per cell 400 of the second DAC 205 independent of PVT variations. This is because the reference PMOS device is then operating under identical conditions as the second DAC cells 400. The output current 106 is then also independent of the operating region of the feed forward circuit 105. Key is that the mirroring device 501 operates under identical conditions as the cells 400 of the secondary DAC 205. The mirroring device 501 is exemplarily implemented using a feedback circuit, but there are other ways of accomplishing it. For example, by making use of a cascode transistor.

An alternative to the bias feedback loop could be a voltage DAC to set in FIG. 4 the “vref_pmos” voltage. However, this would suffer from the drawback that this reference voltage needs to be adjusted over VT corners, i.e. during operation of the device 100.

FIG. 6 shows a method according to an embodiment of the present invention. The method 600 can be carried out for processing digital signals. For instance, the method 100 may be performed by a device 100 according to an embodiment of the present invention as shown FIGs. 1-5.

The method 600 includes a step 601 of providing, by a digital signal source 101, a first codeword 104 to a converter circuit 102, in order to generate an output signal 103. Further, the method 600 comprises a step 602 of providing, by the digital signal source 101, a second codeword 107 to a feed forward circuit 105, in order to generate an output current 106. The output current 106 of the feed forward circuit 105 is supplied to the converter circuit 102. Further, the second codeword 107 is generated by the digital signal source 101 based on the first codeword 104, particularly such that variations of a supply current of the converter circuit 102 are compensated.

FIG. 7 shows simulations performed on the dynamic performance of the first DAC 202 of a device 100 according to an embodiment of the present invention, while changing the scale (x-axis) of the second DAC 205. In particular, in the top graph results 702 and 703 on the harmonic distortion (in dB) of the first DAC 202 are shown. The simulation result 702 is obtained at a different operating frequency than the simulation result 703. The bottom graph shows results 700 and 701 of the corresponding supply variation of the first DAC 202 (in dB). Again, the result 700 is obtained at a different operating frequency than the result 701. The arrows between the top and bottom graph show that the result 702 corresponds to the result 701, and the result 703 corresponds to the result 700.

In particular, the largest (frequency-dependent) non-linearity/harmonic distortion (top graph) and the largest supply ripple/variation (bottom graph) are illustrated in FIG. 7 as a function of the scale of the second DAC 205. It can be seen that the scale at which the supply ripple is minimal does not correspond to the scale of lowest harmonic distortion. This entails that the present invention allows doing even better than simply reducing a supply ripple, but also performing an analog pre-distortion. At the same time, the input codeword 107 applied to the second DAC 205 is directly proportional to the input codeword 104 of the first DAC 202, leading to very little overhead in the digital data source 101 (as opposed to digital pre-distortion).

The present invention has been described in conjunction with various embodiments as examples as well as implementations. However, other variations can be understood and effected by those persons skilled in the art and practicing the claimed invention, from the studies of the drawings, this disclosure and the independent claims. In the claims as well as in the description the word“comprising” does not exclude other elements or steps and the indefinite article“a” or“an” does not exclude a plurality. A single element or other unit may fulfill the functions of several entities or items recited in the claims. The mere fact that certain measures are recited in the mutual different dependent claims does not indicate that a combination of these measures cannot be used in an advantageous implementation.