The invention relates to a modulator for a high efficiency switch mode RF power amplifier (SMPA) in wireless or wired communications. It describes a fully digital modulation technique which deals with a conversion of the baseband amplitude and phase modulating signals into an RF modulated carrier with constant envelope resulting in a two-level pulse train that drives the input of an SMPA.
The RF modulator is usually realized as a pulse-width modulator (PWM) or employs bandpass delta-sigma modulator (BDSM) techniques. In case of PWM, output pulses can be very short (maximum pulse width is 1/fM, where fM is the modulator sampling frequency) which requires high fT of a power device to process the signal without additional power losses. In case of DSM the minimum pulse length will be equal to 1/fM. Common BDSM architecture uses the sampling frequency four times higher than the input fM=4fC. Thus, if the system in
Another implementation of the modulator in the analogue domain including an upconversion function which decreases the average output frequency can be found in [M. Nielsen, T. Larsen, “A Transmitter Architecture Based on Delta-Sigma Modulation and Switch-Mode Power Amplification,” IEEE Trans. Circuits and Syst. II, Exp. Briefs, vol. 54, no. 8, pp. 735-739, August 2007.]. This modulator processes the envelope amplitude and phase modulated carrier such a way that at the input of the SMPA only pulses with a width of 1/(2fC) are occurred.
A fully digital modulator with decreased average output frequency and, possibly, with only 1/(2fC) pulse widths in the output pulse train is highly demanded for the SMPA.
According to the present invention, a digital polar modulator for transforming a baseband signal into a modulated digital modulator output signal, the digital polar modulator comprises
The fully digital polar modulator of the current invention improves the performance of the SMPA by reducing the average output frequency.
In the following, embodiments of the digital polar modulator are described. The additional features of the embodiments can be combined with each other to form further embodiments, unless explicitly described as forming alternative embodiments.
In the digital polar modulator of one embodiment, the first delta-sigma modulator has a drive input for receiving a drive signal determining a sampling frequency, and wherein the drive input is connected to the output of the multiplexer.
This embodiment may advantageously further comprise a divider unit, which is connected between the multiplexer and the first delta-sigma modulator and which is configured to provide to the drive input a drive signal that determines a sampling frequency forming a fraction of a frequency of the signal provided at the output of the multiplexer.
Another embodiment further comprises an encoder, which is connected downstream from the second delta-sigma modulator and upstream from the multiplexer, and which is configured to transform the multilevel quantized second delta-sigma output into the select signal with a width of N-bit, wherein N is selected such that 2N represents a number of carrier signals provided by the multiphase generator.
In the digital polar modulator of this embodiment the second delta-sigma modulator is preferably configured to provide the multilevel quantized output as an output with the number of levels higher than 2N, wherein at least one level exceeding 2N represents an extension of the second delta-sigma modulator input range.
In a further embodiment of the digital polar modulator, the combiner unit comprises at least one AND-gate connected on its input side with the multiplexer and the first delta-sigma modulator.
In an embodiment forming an alternative thereto, however, the multiplexer of the digital polar modulator additionally has an inverted output that provides an inverted multiplexer output signal forming an inverse of the multiplexer output signal, the combiner unit is configured to provide a differential three-level output signal. In this embodiment, the combiner unit preferably comprises a first AND-gate connected on its input side with the non-inverted multiplexer output providing the multiplexer output signal and with the first delta-sigma modulator, and a second AND-gate connected on its input side with the inverted output of the multiplexer and with the first delta-sigma modulator output.
A second aspect of the invention is formed by a transmitter, comprising a polar modulator according to the first aspect, or according to one of its embodiments.
The transmitter preferably further comprises power amplifier connected downstream from the digital polar modulator, and a bandpass filter connected downstream from the power amplifier.
The transmitter may further comprise a CORDIC unit, which is configured to receive a first signal component (I) and a quadrature second signal component (Q) and to provide to the first and second input parts of the digital polar modulator the amplitude and phase modulated signals, respectively. The CORDIC unit may be implemented as a digital signal processor (DSP).
Further embodiments are described in the claims and will, in the following, be described with reference to the enclosed Figures.
The proposed DPM architecture is depicted in
The target transmitted signal can be written as
sTX(t)=A(n)·sin(ωCt+φ(n)) (1)
where A(n) and φ(n) are digital representations of the amplitude and phase of the carrier, respectively; ωC is the carrier frequency.
The proposed modulator constructs a signal in the digital domain of a kind, which after amplification and bandpass filtering will result in the sTX(t) signal of equation (1). For this reason the signal according to (1) was split in two components, the amplitude envelope A(n) and the square-wave signal with the frequency of fC modulated by the phase signal φ(n). Amplitude values A(n) are converted in 1-bit pulse train by means of a low pass delta-sigma modulator DSMA. The sampling frequency fDSM-A is taken from the MUX output, optionally divided by some number, in order to align pulse edges in the amplitude and phase paths. This relaxes the speed requirements for the power device. If the speed requirements are not critical, the sampling frequency fDSM-A can be applied independently and has to satisfy the following expression fBB<<fDSM-A≦fC.
To construct a phase-modulated square-wave signal, the following approach is used. An MPG has to provide a set of square-wave carrier signals shifted by a discrete phase value, applied to the input of a multiplexer.
The phase values φ(n) come to the low pass multi-level delta-sigma modulator DSMP. The sampling frequency fDSM-P has to satisfy following expression fBB<<fDSM-P≦fC. The select signal of the MUX is generated depending on the input phase values φ(n). This allows to switch between the square-wave carrier signals shifted by a discrete phase values and thus to approach to the carrier frequency with the original phase. For example, consider a DPM with 4 discrete phases in the MPG. If an input phase value φ(n) equals to 340 degrees, then the select signal of the MUX switches its output between the MPG signals corresponding to 270 and 0 degrees (
The number of quantizer levels is equal to the number of carrier phases 2N plus 1 to cover the phase range. A sufficient number of levels can be added in the quantizer of DSMP to extend the input range in order to avoid errors if overloading occurs. For example, one additional level extends the input range from the top and another one from the bottom. Thus, the modulator DSMP will have 2N+3 outputs. Then, an encoding has to be applied to provide an N-bit select signal from 2N+3 DSM outputs.
Consider a DPM with four discrete phases in the MPG (N=2). An example of the encoder in this case is shown in
Finally, the constructed phase-modulated square wave signal and the delta-sigma modulated amplitude values are combined by means of the AND-gate resulting in a two-level or three level signal. After amplification and bandpass filtering at carrier frequency, the targeted signal sTX(t) is obtained.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/EP2011/073118 | 12/16/2011 | WO | 00 | 6/16/2014 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO2013/087124 | 6/20/2013 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
7551904 | Lee et al. | Jun 2009 | B2 |
20030206056 | Hietala | Nov 2003 | A1 |
20050190854 | Shakeshaft et al. | Sep 2005 | A1 |
20060038710 | Staszewski et al. | Feb 2006 | A1 |
20060052124 | Pottenger et al. | Mar 2006 | A1 |
20060170505 | Humphreys et al. | Aug 2006 | A1 |
20060203922 | Rezeq | Sep 2006 | A1 |
Number | Date | Country |
---|---|---|
1020060038134 | May 2006 | KR |
20110068605 | Jun 2011 | KR |
Entry |
---|
M. Helaoui, et. al, “A Novel Architecture of Delta-Sigma Modulator Enabling All-Digital Multiband Multistandard RF Transmitters Design,” IEEE Trans. Circuits and Syst. II: Exp. Briefs, vol. 55, No. 11, pp. 1129-1133, Nov. 2008. |
M. Nielsen, T. Larsen, “A Transmitter Architecture Based on Delta-Sigma Modulation and Switch-Mode Power Amplification,” IEEE Trans. Circuits and Syst. II, Exp. Briefs, vol. 54, No. 8, pp. 735-739, Aug. 2007. |
Bassoo V et al: “Analysis of distortion in pulse modulation converters for switching radio frequency power amplifiers”, Dec. 9, 2010, pp. 2088-2096. |
Bassoo V et al: “Distortion arising from polar to PWM/PPM conversion in an all digital upconverter for switching rf power amplifier”, Jun. 7, 2009, pp. 1533-1536. |
Assoo V et al: “Potential Architecture for Future Generation ‘Green’ Wireless Base Station”, Wireless Pervasive Computing, 2009. ISWPC 2009. Feb. 11, 2009, pp. 1-5. |
Keyzer J et al: “Generation of RF Pulsewidth Modulated˜”, 2002 IEEE MTT-S International Microwave Symposium Digest, Jun. 2, 2002, pp. 397-400. |
Parikh V K et al: “All Digital-Quadrature-Modulator Based Wideband Wireless Transmitters”, IEEE Transactions on Circuits and Systems, Nov. 1, 2009, pp. 2487-2497. |
Bjorn Thorsten Thiel et al: “Digital polyphase baseband for direct digital generation of high sample rate radio frequency signals”, Jul. 18, 2010, pp. 1-4. |
Bjorn Thorsten Thiel et al: “Digitally controlled pulse-width-pulse-position modulator in an 1.2V 65nm CMOS technology”, Aug. 29, 2011, pp. 81-84. |
Jack E. Volder, “The CORDIC Trigonometric Computing Technique”, IRE Trans. Electron. Comput. EC-8:330-334 (1959). |
Number | Date | Country | |
---|---|---|---|
20140307825 A1 | Oct 2014 | US |