Patent Grant
6888485
Digital-analog converters of this type are used in particular in mobile radiotelephone base stations in order to convert digital data having a high bit rate into high-frequency transmission signals. In this situation, the high-frequency transmission signals are to be generated with a high level of spectral purity, in other words they should exhibit as few unwanted spurious frequencies as possible. Moreover, this should be achieved cost-effectively with the smallest possible demand on technical resources. In the typical case, the high-frequency transmission signals should have one or more modulated frequency carriers and be capable of covering a large bandwidth.
To date, architectures have been employed for this purpose which use known digital-analog converters to generate high-bandwidth analog signals in a frequency range which is too low for transmission. Subsequently, the analog signals are then converted by one or two frequency converter stages, for example mixer stages, into the high-frequency transmission signals in order to be sent by radio from the mobile radiotelephone base station.
The mixer stages, however, require a high level of resources in terms of circuitry, in particular a high demand for electronic components. In general, the mixer stages cannot be implemented as integrated circuits since filters having a high quality level must be used and, as is known, these require a large number of passive components such as capacitors and resistors. Passive components cannot however be conveniently implemented in integrated circuits, particularly when precise and in part high capacitance and/or inductance values are required for high-quality filters. Furthermore, all intermediate frequencies must be screened off separately. In order to provide a cost-effective and thus economical solution, a digital-analog converter would therefore be required which comprises a small number of electronic components, passive components in particular, and is preferably suitable for implementation as an integrated circuit.
Various aspects and advantages of the invention will be set forth in part in the description that follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
One possible object of the invention is therefore to create a digital-analog converter which has a simple structure, in particular avoids high-quality filters, and can therefore preferably be implemented as an integrated circuit. Furthermore, the digital-analog converter should generate an analog output signal which can be used as a high-frequency transmission signal.
The digital-analog converter includes two basic structures, namely a D flip-flop and a modulator which is connected downstream of the D flip-flop. The D flip-flop is clocked by a clock signal and a digital input signal is fed to the input of the D flip-flop, which digital input signal is to be converted into an analog output signal. One possible of the purpose of the device is to generate an analog output signal in a frequency range which is suitable for transmission. The analog output signal must therefore lie in the frequency range for high-frequency transmission signals. To this end, the output signal from the D flip-flop is multiplied with the clock signal by the modulator. In the modulator, the clock signal is distorted in such a way that the zero transitions of the clock signal are extended in time. Accordingly, jitter contained in the output signal from the D flip-flop is effectively suppressed in the modulator during the zero transitions of the distorted dock signal.
The spectral purity of the analog output signal from the modulator is essentially determined by the spectral purity of the clock signal. In this situation, no clock signals having different phases for the D flip-flop and the modulator are required as in the case of previous approaches to a solution on the basis of the related art. For this reason, a 90 degree phase shifter is also dispensed with. In addition, the frequency is lower than in the case of known digital-analog converters whose output signals lie directly in the high-frequency range. This results from the fact that the modulator additionally converts the frequency of the output signal from the D flip-flop in such a way that all the harmonics are superimposed constructively and no unnecessary quantization noise is added.
By preference, the clock signal is a sinusoidal signal of high spectral purity. As a result of this, an analog output signal is generated at the modulator with a particularly high level of spectral purity, in other words essentially having no unwanted spurious frequencies.
The following values have proved to be particularly advantageous for the frequency of the digital input signal and the frequency of the clock signal: The digital input signal should occupy a frequency range at one quarter of the clock frequency. In the analog output spectrum, on the other hand, the frequency range proves to be particularly favorable at five quarters of the clock frequency.
It generally proves to be favorable if the frequency range of the digital input signal lies at one quarter of the clock frequency and the analog output signal lies at (2n+1) quarters of the clock frequency. In this situation, it is particularly favorable if the condition n=4k+1 or n=4k+2 where k=0, 1, 2, . . . is satisfied.
By preference, n=2 is chosen since in this instance the advantageous relationship mentioned previously results, namely that the frequency of the digital input signal lies at one quarter of the clock frequency and the analog output signal lies at five quarters of the clock frequency.
An embodiment of the D flip-flop and of the modulator employing bipolar technology is, by preference, suitable for the extremely high frequency which normally has to be produced for the analog output signal and also for the high frequency of the digital input signal. An embodiment employing a fast MOS technology is also conceivable. As is known, both aforementioned technologies are particularly suited for integration of the circuit employing silicon technology or also GaAs technology.
The digital-analog converter can be operated in a particularly advantageous manner if the digital input signal is noise-shaped, whereby the quantization noise of the digital signal is shifted into frequency ranges which contain no signal constituents. This can be realized for example by a sigma-delta modulator which is connected upstream.
The digital-analog converter is similarly advantageously suited for a multi-stage construction; to this end, simply a plurality of digital-analog converters is connected a parallel configuration whereby different digital input signals are fed to the inputs of the D flip-flops in each case and the corresponding analog output signals are simply added at the outputs of the modulators. This embodiment can additionally be advantageously implemented as an integrated circuit.
These and/or other aspects and advantages of the present invention will become more apparent and more readily appreciated from the following description of the preferred embodiments, taken in conjunction with the accompanying drawings of which:
Hereinafter, aspects of the present invention will be described in detail with reference to the attached drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the aspects set forth herein; rather, these aspects are provided so that the present disclosure will be thorough and complete, and will fully convey the concept of the invention to those skilled in the art.
Reference character 10 in
Finally, the D flip-flop 10 has a further differential input “CLK” for a clock signal 12. The clock signal 12 corresponds—as illustrated in FIG. 1—to a sinusoidal signal of high spectral purity with a frequency of four fifths of a default frequency, at which the frequency fOut of the analog output signal to be generated lies. The greater the stability of frequency of the clock signal 12, in other words the less phase noise it exhibits, the higher is also the level of purity of the frequency spectrum of the analog output signal to be generated. In the ideal situation, the clock signal is a sinusoidal signal of high spectral purity, in other words it is essentially free from unwanted spurious frequencies.
A modulator 11 which is similarly implemented in bipolar technology is connected downstream of the D flip-flop 10. Like the D flip-flop 10, it is possible to set the power consumption of the modulator 11 by way of the input denoted by “Bias1”. Moreover, the setting of the threshold value for the pulse distortion is effected by way of an input denoted by “Bias2”. In principle, the modulator 11 corresponds to a multiplier circuit which multiples the output signal 14 from the D flip-flop 10, whose idealized eye pattern is represented schematically, with the clock signal 12. However, before the two signals are multiplied the clock signal 12 is distorted in such a way in the modulator 11 that the zero transitions of the clock signal, or more precisely the zero transitions of the sinusoidal clock signal, are extended in time, as is shown schematically using the two small signal waveform diagrams 17 and 18. As a result of this distortion of the clock signal, any jitter which may be contained in the output signal 14 from the D flip-flop 10 is essentially suppressed in an effective manner.
After the clock signal 12 has been distorted in the modulator 11, the multiplication with the output signal 14 from the D flip-flop 10 is performed by two differential changeover stages 19 and 20 connected in parallel. At the collectors of the transistors of the two differential changeover stages 19 and 20 an analog digital signal 16 is present—as illustrated schematically—at an output 15 of the modulator 11. In this situation, the analog output signal 16 lies in a frequency range which corresponds to the transmission frequency range, in other words the frequency range for the transmission of radio signals in a mobile radiotelephone base station. A further conversion of the analog output signal to another frequency range is therefore no longer required. Corresponding mixer stages can be dispensed with.
The highly symmetrical structure both of the D flip-flop 10 and also of the modulator 11 can be recognized from
The basic frequency spectrum of the output signal at the output 15 of the modulator 11 is illustrated in FIG. 3. Depending on the shape of the individual pulses 17, 18 in the modulator, the analog output signal appears with differing strengths in the individual frequency ranges which lie at (2n+1) quarters of the clock frequency. Between the individual peaks in the frequency ranges where the analog output signal is present there are areas which contain no signal. These areas contain the shaped quantization noise which is shown hatched. This can be removed along with the undesirable signal bands by appropriate filtering, for example in a narrow-band transmitter stage.
It has proved to be particularly advantageous when the frequency of the digital input signal lies at one quarter of the clock frequency and the analog output signal occupies the frequency band at five quarters of the clock frequency.
It generally holds true that frequency values at one quarter of the clock frequency are favorable for the digital input signal and that output signals thus result at (2n+1) quarters of the clock frequency (advantageous where n=4k+1 or n=4k+2).
The many features and advantages of the invention are apparent from the detailed specification and, thus, it is intended by the appended claims to cover all such features and advantages of the invention that fall within the true spirit and scope of the invention. Further, since numerous modifications and changes will readily occur to those skilled in the art, it is not desired to limit the invention to the exact construction and operation illustrated and described, and accordingly all suitable modifications and equivalents may be resorted to, falling within the scope of the invention.
This application is based on and hereby claims priority to PCT Application No. PCT/DE01/02211 filed on 13 Jun. 2001 and German Application No. 100 31 538.0 filed 28 Jun. 2000, the contents of which are hereby incorporated by reference.
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCTDE01/02211 | 6/13/2001 | WO | 00 | 3/25/2003 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO0201724 | 1/3/2002 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 4663610 | Metz et al. | May 1987 | A |
| 5231395 | Irwin et al. | Jul 1993 | A |
| 5321401 | White | Jun 1994 | A |
| 5544081 | Yukawa | Aug 1996 | A |
| 5625360 | Garrity et al. | Apr 1997 | A |
| 6433720 | Libove et al. | Aug 2002 | B1 |
| 6486814 | Koyanagi | Nov 2002 | B2 |
| 6628220 | Cosand | Sep 2003 | B2 |
| 6664906 | Volk | Dec 2003 | B2 |
| Number | Date | Country |
|---|---|---|
| 3637650 | May 1987 | DE |
| 0577887 | Jan 1994 | EP |
| WO9501006 | Jan 1995 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 20030181172 A1 | Sep 2003 | US |
