SYSTEM AND METHOD FOR GENERATING A DATA-Dependent CLOCK FOR A DAC IN A MODULATOR

By determining time instances for phase crossings to generate a data-dependent clock, the polar modulator addresses inefficiencies in phase transition processing, achieving enhanced power efficiency and bandwidth support in wireless transceivers.

DE102015107198B4Inactive Publication Date: 2024-11-14APPLE INC
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
DE102015107198
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2014-06-13
Filing Date
2015-05-08
Publication Date
2024-11-14
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing polar modulators for wireless transceivers face challenges in efficiently generating output signals with precise phase transitions due to the use of fixed digital clocks, leading to inefficiencies and limitations in supporting larger signal bandwidths.

Method used

A method and modulator that determines time instances associated with predetermined phase crossings, such as integer multiples of 180°, to generate a data-dependent clock for clocking a digital-to-analog converter, allowing for synchronized amplitude and phase processing, thereby producing a square wave output signal with transit times aligned with these phase transitions.

Benefits of technology

This approach enhances the efficiency and flexibility of polar modulators by aligning clock edges with phase transitions, enabling higher output power and supporting larger signal bandwidths without back-intermodulation and remodulation issues.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Procedure, comprehensive: Receiving a plurality of digital samples of a modulated baseband signal, wherein the modulated baseband signal comprises a plurality of digital amplitude samples and a plurality of digital phase samples; and where the majority of digital phase scans comprise digital scans of a combined baseband and carrier phase; Determining time instances that are assigned to predefined phase transitions of the modulated baseband signal; Generating a data-dependent clock signal with rising and falling edges that are assigned to specific time instances, and Interpolating the plurality of digital amplitude samples based on the specified time instances to generate a plurality of interpolated digital amplitude samples representing amplitude values ​​corresponding to the predetermined phase transitions; and clocking a digital-to-analog converter using the generated data-dependent clock signal, wherein a square-wave output signal from the digital-to-analog converter includes transit times associated with the generated data-dependent clock signal; Providing the majority of interpolated digital amplitude samples to the digital-to-analog converter; and Modulating the size of the square wave output signal based on the plurality of interpolated digital amplitude samples.
Need to check novelty before this filing date? Find Prior Art

Description

BACKGROUND OF THE INVENTION

[0001] Polar modulators for wireless transceivers offer several implementation advantages compared to conventional I / Q transceivers, such as lower power consumption due to a lower ratio of maximum to average amplitude and a half-clock local oscillator (LO) distribution. Polar modulators also exhibit no mutual intermodulation, and higher output power is possible because a polar modulator is less susceptible to remodulation. Furthermore, a polar modulator has no limitation regarding the support of the wider signal bandwidth required for digital pre-distortion.

[0002] According to the polar modulator concept, the modulation signal is divided into an amplitude-modulated (AM) signal and a phase-modulated (PM) signal. The symbols or points used in polar modulation correspond to, or are derived from, Cartesian coordinates used in vector modulation concepts. The polar modulation concept offers, among other things, power efficiency advantages.

[0003] US 2010 / 0 111 222 A1 relates to a device and associated method that facilitates the communication of data in a radio communication system using a time-slot communication scheme. Multiple consecutive time slots can be assigned in which data can be transmitted to a receiving communication station. At selected intervals, a controller causes the operating mode of the receiving communication station to change, pausing its operation in receive mode and, during the pause, causing the receiving communication station to transmit data, such as control confirmations or other responses. BRIEF DESCRIPTION OF THE DRAWINGS Fig. is a block diagram illustrating a polar modulator according to an example in the present disclosure. Fig. are graphs that illustrate a continuous baseband phase, a continuous support phase, and a continuous combined complex baseband and support phase. Fig. is a graph that represents amplitude jumps at predetermined phase transitions that are integer multiples of 180°, according to an example in the present disclosure. Fig. is a graph that illustrates digital phase samples of a combined, complex baseband and carrier phase with adjacent samples spanning a predetermined phase transition, for example at integer multiples of 180°. Fig. is a graph that shows amplitude jumps at phase transitions other than those predetermined in Fig. illustrated by another example from the present revelation. Fig. is a graph that shows an exploded view of a portion of Fig. shows, in order to illustrate more clearly a calculation of a time instance that is assigned to a predetermined phase passage, according to an example from the present disclosure. Fig. is a graph that illustrates a plurality of amplitude samples that are assigned to the edges of a fixed clock signal. Fig. is a graph that shows an exploded view of a portion of the graph in Fig. represents, where amplitude values ​​that are associated with the edges of a fixed clock and that span a specific time instance are used to interpolate a digital amplitude value that is associated with the specific time instance. Fig. is a schematic representation illustrating a first processing circuit according to an example in the present disclosure. Fig. is a schematic representation illustrating a second processing circuit according to another example in the present disclosure. Fig. Figure 1 is a block diagram illustrating a digital-to-time converter (DTC) which receives a control signal and a high-frequency clock and outputs rising and falling edges that form a data-dependent clock, according to an example in the present disclosure. Fig. is a graph representing a rectangular wave output waveform, which has transitions that are assigned to the specific time instances, which are pre-determined phase transitions, and an amplitude that is assigned to the amplitude data synchronized with the specific time instances, according to an example in the present disclosure. Fig. is a block diagram illustrating a polar modulator according to another example in the present disclosure. Fig. is a schematic representation showing a second processing circuit according to the exemplary one, in Fig. The modulator shown illustrates this. Fig. is a flowchart illustrating a method for generating an output signal of a polar modulator using a data-independent clock according to an example of the present disclosure. Fig. is a block diagram illustrating a polar modulator according to another example in the present disclosure. Fig. is a block diagram illustrating a polar modulator according to another example in the present disclosure. Fig. is a block diagram illustrating a modulator that identifies predefined phase transitions in an I / Q range, according to an example in the present disclosure. Fig. is a block diagram illustrating a modulator that identifies pre-determined phase transitions in an I / Q range, according to another example in the present disclosure. DETAILED DESCRIPTION

[0004] The present disclosure comprises a method according to claim 1 and a modulator according to claim 7, which determine timing instances that are assigned to predetermined phase transitions of a modulated baseband signal and use the predetermined timing instances to generate a data-dependent clock. The data-dependent clock is then used to clock a digital-to-analog converter such that it generates an output of a polar modulator with timing instances corresponding to the predetermined phase transitions.

[0005] In a polar modulator, received Cartesian signals (i.e., an in-phase (I) and a quadrature (Q) signal) are converted into polar equivalents representing an amplitude part (R) and a phase part (9). The polar signals are then processed, with the amplitude signal part being processed by a digital-to-analog converter (DAC) and the phase signal part by some form of time-based DAC, such as a digitally controlled oscillator (DCO) or a digital-time converter (DTC). Digital clock signals used in such phase processing, for example, do not provide continuous phase information, as information is only present on the rising and falling edges of the digital clock signal.

[0006] In the present disclosure, a fixed digital clock signal is used to sample an RF phase (where the RF phase comprises the sum of the phase of the complex baseband signal and the linear phase ramp of the RF carrier frequency signal). The resulting phase samples are then used to determine time instances associated with a predetermined phase transition, for example, at integer multiples of 180°. The polar modulator of the present disclosure therefore determines the time instances when the RF phase has advanced by, for example, integer multiples of 180°.

[0007] The output of an analog I / Q modulator can be written as follows: y(t) = real {e 2nfcentert × (I(t) + jQ(t))}, where f centerThe RF channel frequency is and I(t) and Q(t) are the in-phase and quadrature components, respectively, of the complex baseband signal. The complex baseband signal can also be written in polar form, where I(t)+jQ(t)=A(t)ej9t.

[0008] This results in an RF signal of y(t)=real{eJ2nfcentertA(t)ej9t}=ieai:−\me2f entet+':=A(t)cos(2nfcentert+(t).

[0009] The RF signal will therefore exhibit a zero crossing on the rising edge when 2nf centert + ^(t) = 3n 2 + 2nn, and a zero crossing on the falling edge occurs at 2nf centert+ ^(t) = n / 2 + 2nn. The polar modulator of this disclosure can use the above information to calculate the time instances associated with the phase data occurring, for example, at integer multiples of 180°. Alternatively, as discussed in more detail below, other predetermined phase transitions can be used in conjunction with this disclosure. Such time instances (associated with the predetermined phase transitions) are used to generate a data-dependent clock, which is ultimately used to clock a digital-to-analog converter that generates the output signal of the polar modulator. Furthermore, although the examples given here address predetermined phase transitions at integer multiples of 180°, the disclosure is not limited to these examples.

[0010] Reference is now made to Fig. A block diagram is shown illustrating a polar modulator 100 according to an example from the present disclosure. The modulator 100 receives input data I and Q in Cartesian coordinates, which undergo various types of optional signal processing in the digital domain, for example with a processor 102. The processed I / Q values ​​104 are converted into a polar form comprising amplitude samples R[k] and phase samples Q[k], for which an I / Q polar converter 105 is used, such as a coordinate rotation digital computer (CORDIC). Optional processing blocks 106 and 108 in the amplitude and phase paths, respectively, can provide processing such as pre-distortion and can also provide upsampling, such that the resulting digital amplitude samples 110 and phase samples 112 are time instances of a fixed digital clock (CLKf). ixed) correspond.

[0028] Further reference is made to Fig. A computation block 114 receives the digital phase samples 112 and determines time instances corresponding to a predetermined phase crossing or demarcation and, in this example, integer multiples of 180° with respect to it. In a specific example, the time instances could be assigned to zero crossings of the phase. More generally, however, the computation block 114 can determine time instances with any crossing(s) in the phase data.

[0011] For example, reference is made to the Fig. ; Fig. illustrates a slowly varying baseband phase signal 150, while Fig. The linear ramp slope of an RF carrier phase signal 152 illustrates, for example, a sine signal with a linearly varying phase for each period from 0 to 2n, which then repeats periodically. Fig. Illustrates the combined baseband and carrier phase signal 154 (e.g. 2nf). center + ^(t)), where, for a cosine function, the zero crossings occur at integer multiples of n / 2 and 3n / 2. It should be noted that, since Fig. a change in the carrier phase of Fig. due to the addition of the baseband phase data from Fig. This causes the time instances of the zero crossings to be non-periodic and therefore do not correspond to the time instances of the digital phase data ^[k], which correspond to the fixed sampling clock CLK. fixed are equivalent to.

[0012] Reference will now be made again to Fig. combined with Fig. ; the calculation block 114 uses the samples ^[k] 112, which are connected to the fixed data clock CLK fixedcoincide to calculate the time instances of the zero crossings in the rising and falling edges (n / 2 and 3n / 2) of the combined baseband and carrier phase signal 154 for each 360° phase rotation.

[0013] For example, reference is made to Fig. , in which an example of a continuous combined baseband and carrier phase signal 160 is illustrated, where the amplitude 162 undergoes a polar phase shift at -n / 2 and +n / 2, which occurs every 180° (and which is a different phase shift than described above). In an actual implementation, this is illustrated Fig. The digital samples ^[k] 112, which represent the combined baseband and carrier phase signal. In this example, the time instances (T1, T2, ,,., Tn), which are assigned integer multiples of 180°, such as -n / 2 and +n / 2, are provided. An enlarged or exploded view of part of the digital phase data ^[k] is shown in Fig. represented, where each phase sample 170 corresponds to an edge 172 of the fixed clock 174. It should be noted that, due to the baseband phase data over the carrier phase, a time instance (T3) 180, which is assigned to the -n / 2 crossing, corresponds neither to a rising nor a falling edge of the fixed clock 174, but instead falls somewhere between a falling edge 182 and a rising edge 184 thereof, where the falling edge 182 and the rising edge 184 are assigned to known time instances.

[0014] In one example, calculation block 114 calculates the zero-crossing time instance (T1, T2, "., Tn) 180 (e.g., t cross) by solving a linear equation. If the time instance associated with a falling edge 182 is t0 = 22.5625 nS and the phase quantity Arg0 = -1.835409 (which is less than -n / 2) is known, and if the time instance associated with a rising edge 184 is t1 = 22.625 nS and the phase quantity Arg1 = -1.429997 (which is greater than -n / 2) is known, we can calculate the time instance 180 as follows: (−n / 2−Arg0) / (Arg1−Arg0)X(t1−t0)+t0=22.6033 nS.

[0015] Using these specific time instances 191, the time instance differences can then be easily calculated, where, for example, AT2 = T2 - T1. The same applies to the predetermined phase transition of n / 2 (time instance 181), t0 = 23.0625 nS and Arg0 = 1.4145112 (smaller than n / 2) and t1 = 23.125 nS and Arg1 = 1.821834 (larger than n / 2). Using the formula, one obtains: (−n / 2−Arg0) / (Arg1−Arg0)X(t1−t0)+t0=23.0865 nS.

[0016] Other integer multiples of 180°, for example, of such phase transitions, can be calculated by computation block 114 in a similar manner. Therefore, in one example, computation block 114 identifies the two phase instances spanning the predetermined phase transition or demarcation (as well as integer multiples of 180° thereof). In one example, the phase multiples of 180° correspond to the zero crossings n / 2 and 3n / 2; however, as in Fig. The phase transitions shown correspond to -n / 2 and +n / 2. Any other predetermined phase ratio may be used and is considered to fall within the scope of this disclosure. For example, as shown in 3C, the phase transition may be n / 4 or 3n / 4. Any such predetermined phase transition and any integer multiple thereof of 180° is considered within the scope of this disclosure. It should also be noted that although integer multiples of 180° are used in the examples herein, this disclosure is not limited to them.

[0017] As in Fig. As can be seen, calculation block 114 thus gives a calculated difference of 190 in the time instances T1, T2,_, T nfrom, where the difference AT1 = T2 - T1, AT2 = T3 - T2 etc., which is assigned to the combined baseband and carrier phase signal 112, passes through the predetermined phase transition or demarcation plus integer multiples of 180° therefrom. Alternatively, the calculation block 114 can solve a quadratic equation, for example: [a−c±[(c−a)2−8b(a−2b+c)]1 / 2] / 4(a−2b+c), where “b” is the sampling time instance closest to the phase transition, “a” is the preceding sampling time instance, and “c” is the subsequent sampling time instance. Other computation schemes may also be used by computation block 114 to compute the time instances, and all such alternatives are considered to fall within the scope of protection of this disclosure.

[0018] Reference is made again to Fig. While computation block 114 outputs the determined differences in the time instances 190, it provides the actual time instances 191 (T1, T2, ... Tn) to a variable-delay block 192, which acts as an interpolator. The variable-delay block 192 receives the digital amplitude values ​​R[k] 110 and uses the pre-determined time instances 191 to generate interpolated digital amplitude values ​​Ri, R2, ... R n to obtain 194, which are synchronized with the specific time instances 191. For example, as in Fig. As shown, a plurality of amplitude samples R[k] 110 are input into the block with variable delay 192, with each of the samples being triggered by an edge of the fixed clock (CLK). fixed ) coincides. As in Fig. As shown, the time instance T3 falls between flanks 182 and 184 of the fixed time signature, and Fig. represents two amplitude values ​​R[3a] and R[3b] on the same edges. To determine the amplitude value assigned to the time instance T3, which lies between edges 182 and 184 of the clock cycle, an interpolation, for example a linear interpolation, can be performed, as described in more detail in Fig. shown. In this case, the time instances 191 are known and the amplitude values ​​are known, and because the time instance T3 has been calculated, the interpolated amplitude R[3] can thus be easily determined.

[0019] The interpolated amplitude values ​​R1, R2, R n 194 are then forwarded to a first processing unit 196, which subsequently uses the interpolated values ​​to generate a single amplitude value for each 180° phase period. It should be noted that in Fig. For example, the amplitude for each 180° phase period is a single value, but the interpolated values ​​194 are a value of the phase across multiple time instances 191, where a first time instance is assigned to the beginning of a 180° phase period (see reference 198 in Fig. , while the next interpolated value 194 represents the value of the phase at the next time instance assigned to the end of this 180° phase segment (see reference 200 in Fig. Since the amplitude must be a single value during this 180° phase, the first processing circuit 196 calculates a single amplitude value using the two interpolated values ​​(e.g., R1 and R2 or R2 and R3, etc.). In a Fig. In the example shown, the first processing circuit 196 includes a ring buffer 197a, where the interpolated amplitude values ​​are input according to a first range (e.g., the fixed clock) and retrieved from it according to a second clock range (e.g., the data-dependent clock range). The amplitude values ​​leave the ring buffer 197a and are fed to an arithmetic circuit 197b, where two consecutive amplitude values ​​are added, followed by a shift of 1, which is equivalent to division by two. Thus, the output value of the first processing circuit 196 is equivalent to: (R1 + R2) / 2, (R2 + R3) / 2, "... Thus, the first processing circuit 196 generates a single amplitude value for the given duration of the 180° phase by means of interpolation or some other type of processing.The first processing circuit 196 repeats this operation for each of the pairs of interpolated amplitude values ​​to generate a series of amplitude values ​​202 that provide the digital input data for the digital-to-analog converter (DAC) 204. Fig. represent.

[0020] Reference is made again to Fig. The calculated time instance differences 190 (e.g., AT2 = T2 - T1) output by the computation block 114 are input into a second processing circuit 206, which receives the time instance differences and generates a control signal 208 that, in one example, has an integer part 208a and a fractional part 208b. The control signal 208 is then provided to a digital time converter (DTC) 210, which generates a clock signal 212. This clock signal is data-dependent because the clock 212 has edge transitions corresponding to the specific time instance differences 190 output by computation block 114.

[0037] Reference is made to Fig. In one example, the second processing circuit 206 includes a ring buffer 207, followed by an integrator circuit 209. The specified time instance differences are input to and output from the ring buffer 207 according to the first clock range, and the integrator 209 returns the least significant bit (LSB) so that the fractional samples accumulate with LSB overflow. Thus, in one example, the DTC might select 5 integer samples and one 1 / 3 fractional sample, then 5 integer samples and one 2 / 3 fractional sample, followed by 6 integer samples with zero fractional samples. This is an example of a generated clock without complex phase data. The complex phase data causes the number of integer samples, fractional samples, or both to change.

[0021] In one example, the DTC 210 receives from Fig. A high-frequency clock 214 is generated by an oscillator circuit 216, for example, a digitally controlled oscillator (DCO). In one example, the high-frequency clock has a sufficiently high frequency so that selected edges of it can be chosen by the DTC 210 and passed on as a constructed data-dependent clock 212 for further use by the DAC 204. If, for example, the channel frequency is 1 GHz, the period of a fixed clock would be 1 ns.If the first specific time instance T1 fell at 1.1 ns, for example due to the complex baseband phase information, the integer part 208a of the control signal 208 can instruct the DTC 210 to use one of each of 10 time periods of the DCO clock 214 (if the DCO clock is 10 times the channel frequency), while the fractional part 208b of the control signal 208 instructs the DTC to subsequently advance by one clock period (0.1 ns) of the DCO clock 214 and select this edge to be passed on as the edge of the data-dependent clock 212. Similarly, if the next time instance T2 fell at 0.9 ns later, for example due to the complex baseband phase information, the integer part 208a of the control signal can remain the same, and the fractional part 208b can indicate a negative 0.1 ns.These are merely examples to aid in understanding that the time instance differences 190 of the computation block 114 are used to generate a control signal 208, which is used by the DTC 210 to select specific edges of the high-frequency DCO clock 214 in order to construct a clock 213 that is data-dependent and not a fixed clock. This means that the data-dependent clock 212 has edges that correspond in time to the time instances output by the computation block 114 and that correspond to 180° phase periods. Any alternative control and data-dependent clock circuits can likewise be used and are considered to fall within the scope of protection of this disclosure.

[0022] The data-dependent clock 212 is then used to clock the digital-to-analog converter (DAC) 204. The DAC 204 subsequently outputs a square-wave output signal 220, which has transit times corresponding to the generated data-dependent clock 212 and an amplitude corresponding to the processed amplitude data 202 supplied to the DAC 204. The DAC 204 synchronized to the transit times via the processing in the variable-delay circuit 192 and the first processing circuit 196. An example of the output signal 220 is shown in Fig. represented, where the time of the passages of the rectangular wave corresponds to the specific time instances Ti, T 2,_ T n corresponds to and the amplitudes 2021-202 n corresponding to the processed amplitude data, which are synchronized with respect to the time instances.

[0023] Fig. is a block diagram illustrating a polar modulator 300 according to another example in the present disclosure. Insofar as the polar modulator 300 is from Fig. similar components and functionality to components in the polar modulator 100 Fig. Since similar reference numbers are used, a discussion of the components in question is omitted for the sake of brevity. In the polar modulator 300 from Fig. The time instance differences ATi, AT2, ... AT nThe 190 time instance differences determined by the calculation block 114 are forwarded to a frequency calculation circuit 302, which determines a period corresponding to each individual time instance difference (or time between successive time instances), calculates an associated frequency, and outputs a frequency control signal 304 based on this. The frequency control signal 304 serves as a control input to the DCO 306, which then modifies the frequency of its output signal such that the edges of the output signal 212 correspond to the determined time instance differences 190. Thus, the DCO 306 outputs a data-dependent clock signal 212, which is used to clock the DAC 204, in a manner similar to that discussed above.

[0024] Fig. is a schematic representation showing the second processing circuit 302 from Fig. This is illustrated by an example. The second processing circuit 302 includes a ring buffer 303, which receives the determined time instance differences 190 according to the first clock range and outputs the time instance differences 190 according to the second clock range. The output time instance differences 190 are supplied to a processing block 305, which divides a normalization constant by the time instance to calculate an associated frequency, which is then passed to the DCO 306.

[0025] The present disclosure also relates to a process 400 as described in Fig. As illustrated, in one example, the method 400 comprises receiving a plurality of digital samples at 402, wherein the digital samples comprise a plurality of digital amplitude samples and a plurality of digital phase samples, the plurality of digital phase samples comprising digital samples of a combined baseband and carrier phase. The method 400 further comprises determining time instances corresponding to predetermined phase transitions (e.g., at integer multiples of 180°, but not limited thereto) of the modulated baseband signal at 404. In one example, such time instances correspond to the instances 190, which are determined by computation block 114 from Fig. were determined, however, Procedure 400 is not limited to such an example.

[0026] Further reference is made to Fig. The procedure 300 is continued in 306 with the generation of a data-dependent clock with rising and falling edges, which are assigned to the specific time instances. Non-restrictive examples of such processes can be found in the converter circuits 206, 210 and 216 from Fig. as well as the converter circuits 302 and 306 from Fig. Furthermore, the method 400 includes clocking a digital-to-analog converter (DAC) using the generated data-dependent clock signal at 408. The output of the DAC comprises a square wave output signal with transit times that are assigned to the generated data-dependent clock.

[0027] Method 400 can further include, in an example, providing the amplitude samples to the DAC and then, at 412, modulating the square wave output signal using the amplitude samples.

[0028] Reference is now made to Fig. , in which another modulator 500 is shown, the digital phase samples 112 continue to be processed by the computation block 114 to generate the calculated time instance differences 190 and the time instances 191. However, instead of using the time instances 191 to interpolate amplitude samples and thus generate amplitude samples synchronized with respect to the time instances, a variable-delay circuit 502 receives digital in-phase (I) and quadrature (Q) samples 504 and uses the time instances 191 to interpolate the I / Q data 504 and generate interpolated I / Q data 506. The interpolated I / Q data 506 are then passed to a converter 508, such as a CORDIC, to generate amplitude values ​​510, which are interpolated and synchronized with the respective time instances 191. The other modulator components operate in the same way as described above, with reference to... Fig. described to generate the square wave output signal 220 with transitions that are assigned to the data-dependent clock 212.

[0029] Fig. represents a modulator 600 according to another example. In this example, the I / Q samples are interpolated using the time instances 191 in the same way as above, with reference to Fig. described. The difference is that the converter circuits 302 and 306 use the time instance differences 190 to calculate a frequency and a frequency control signal 304 to generate the data-dependent clock 212 as described above with reference to Fig. discussed.

[0030] Fig. is a block diagram representing another modulator 700, where the calculation of the time instances and time instance differences assigned to predefined phase transitions is performed using I / Q data sampling instead of phase sampling, as described above in connection with Fig. As discussed, in the I / Q domain, we can predetermine certain phase transitions and integer 180° multiples thereof, for example, in various ways, and all these alternatives are considered in the present disclosure. For example, we can predetermine certain phase transitions for each time point at which the in-phase (I) data are zero (0), corresponding to Q-axis crossings, which correspond to phases of 90° and 270°, respectively. Alternatively, one can select a phase transition for each time point at which the quadrature (Q) data are zero (0), corresponding to I-axis crossings or phases of 0° and 180°, respectively. Furthermore, one can select phase transitions for each time point at which the I and Q data are equal (in absolute values), corresponding to phases of 45° and 225° and 135° and 315°, respectively.Using these relationships, for example, if these predetermined phase transitions are identified via the I / Q data samples, the I / Q data samples spanning such transitions can be identified and used to calculate associated time instance differences 190.

[0031] In a manner analogous to the discussion in the Fig. For example, calculation block 714 can calculate the time instances in the I / Q domain for the same predetermined phase transitions (-n / 2 and +n / 2) as follows: At t0 = 22.5625 nS, we calculate Re0 = -0.326506 and Im0 = -1.20497, and at t1 = 22.625 nS, we calculate Re1 = 0.173612 and Im1 = -1.22488. From these values, we calculate the time instance as follows: Re0 / (Re0−Re1)X(t1−t0)+t0=22.6033 nS.

[0032] At the next predetermined phase transition, we calculate Re0 = 0.181111 and Im0 = 1.1494 at t0 = 23.0625 nS, and Re1 = -0.28665 and Im1 = 1.11776 at t1 = 23.125 nS. From these data, we calculate the time instance as follows: Re0 / (Re0 - Re1) x (t1 - t0) + t0 = 23.0867 nS. With these time instances 191, one can calculate time instance differences 190.

[0033] Further reference is made to Fig. The computation block 714 sends the calculated time instances 191 to the variable-delay block 792, which acts as an interpolator to synchronize the incoming I / Q data to the calculated time instances. The interpolated or synchronized I / Q data is then converted into amplitude data by a CORDIC converter 705. The rest of modulator 700 operates similarly to the one in Fig. and other corresponding illustrations described modulator 100.

[0034] Fig. This illustrates yet another example of a modulator 800, which uses a calculation block 714 as above with reference to Fig. used to calculate the time instances 191 and the time instance differences 190 in the I / Q domain and subsequently to synchronize the I / Q data samples to the time instances via the variable delay block 792. The CORDIC converter 705 then converts the synchronized I / Q data into synchronized amplitude samples, and the remaining part of the modulator 800 operates similarly to the above with reference to Fig. described.

[0035] In one example of the disclosure, a method comprises receiving a plurality of digital samples of a modulated baseband signal and determining time instances that correspond to predetermined phase transitions of the modulated baseband signal. The method further comprises generating a data-dependent clock signal with rising and falling edges that correspond to the determined time instances and clocking a digital-to-analog converter using the generated data-dependent clock signal, wherein a square-wave output signal from the digital-to-analog converter includes transit times that correspond to the generated data-dependent clock signal.

[0036] In one example of the method, the modulated baseband signal comprises a plurality of digital in-phase (I) samples and a plurality of digital quadrature (Q) samples. In another example of the method, the modulated baseband signal comprises a plurality of digital amplitude samples and a plurality of digital phase samples, wherein the plurality of digital phase samples includes digital samples of a combined baseband and carrier phase.

[0037] In one example of the disclosure, a method comprises receiving a plurality of digital amplitude samples and a plurality of digital phase samples, wherein the plurality of digital phase samples comprise digital samples of a combined baseband and carrier phase, and determining time instances that are assigned to predetermined phase transitions of the combined baseband and carrier phase. The method further comprises generating a data-dependent clock signal with rising and falling edges assigned to the determined time instances, and clocking a digital-to-analog converter using the generated data-dependent clock signal, wherein a square-wave output signal from the digital-to-analog converter comprises transit times assigned to the generated data-dependent clock signal.

[0038] In another example, the method further includes providing the plurality of digital amplitude values ​​to the digital-to-analog converter and modulating a size of the square wave output signal based on the plurality of digital amplitude values.

[0039] In another example of the disclosure, the method further comprises interpolating the plurality of digital amplitude samples based on the determined time instances to generate a plurality of interpolated digital amplitude samples representing amplitude values ​​corresponding to the predetermined phase transitions of the combined baseband and carrier phase, providing the plurality of interpolated digital amplitude samples to the digital-to-analog converter, and modulating a size of the square wave based on the plurality of interpolated digital amplitude samples.

[0040] In another example of the procedure, determining the time instances associated with the predetermined phase transitions involves identifying adjacent digital samples of the combined baseband and carrier phase spanning a predetermined phase transition, determining the time instances of a fixed digital clock signal associated with the identified adjacent samples, and using the determined time instances of the digital clock to determine the time instances associated with the predetermined phase transitions.

[0041] In another example of the procedure, using the determined time instances of the fixed digital clock signal involves solving a linear equation based on the determined time instances.

[0042] In another example of the procedure, generating the data-dependent clock signal involves calculating time instance differences based on the specified time instances assigned to the predetermined phase transitions, and generating a control signal based on the received specified time instance differences assigned to the predetermined phase transitions.

[0043] In another example of the method, generating the data-dependent clock further comprises receiving a digital signal with a frequency significantly higher than a frequency of the data-dependent clock signal, and selectively choosing rising and falling edges of the digital signal based on the control signal, wherein the selected rising and falling edges form the data-dependent clock or are used to construct the data-dependent clock.

[0044] In another example of the procedure, the control signal includes an integer part and a fractional part.

[0045] In one example, a modulator is disclosed and comprises a computation block designed to receive a plurality of digital samples of a modulated baseband signal and to determine time instances that correspond to predetermined phase transitions of the modulated baseband signal. The modulator further comprises a converter circuit designed to generate a data-dependent clock signal with rising and falling edges corresponding to the determined time instances, and a digital-to-analog converter designed to receive the data-dependent clock signal and generate a square-wave output signal with transit times corresponding to the generated data-dependent clock signal.

[0046] In one example of the modulator, the sampling of the modulated baseband signal comprises a plurality of digital in-phase (I) samples and a plurality of digital quadrature (Q) samples. In another example of the modulator, the sampling of the modulated baseband signal comprises a plurality of digital amplitude samples and a plurality of digital phase samples, where the plurality of digital phase samples includes digital samples of a combined baseband and carrier phase.

[0047] In another example, a polar modulator is disclosed and comprises a computation block designed to receive a plurality of digital phase samples, wherein the plurality of digital phase samples comprise digital samples of a combined baseband and carrier phase, and to determine time instances associated with predetermined phase transitions of the combined baseband and carrier phase. The modulator further comprises a converter circuit designed to generate a data-dependent clock signal with rising and falling edges associated with the determined time instances, and a digital-to-analog converter designed to receive the data-dependent clock signal and generate a square-wave output signal with transit times associated with the generated data-dependent clock signal.

[0048] In another example of the polar modulator, the digital-to-analog converter is further designed to receive a plurality of digital amplitude values ​​and to modulate a size of the square wave output signal based on the plurality of digital amplitude values.

[0049] In an example of the polar modulator, the polar modulator further comprises an interpolator circuit designed to receive a plurality of digital amplitude values ​​and the specified time instances from the computation block and to generate a plurality of interpolated digital amplitude values ​​based on the plurality of digital amplitude values ​​and the specified time instances.

[0050] In an example of the polar modulator, the digital-to-analog converter is further designed to receive the majority of interpolated digital amplitude values ​​and to modulate a size of the square wave output signal based on the majority of interpolated digital amplitude values.

[0051] In an example of the polar modulator, the computation block is further designed to identify adjacent digital phase signals spanning a predetermined phase crossing and to determine the time instances associated with the predetermined phase crossings based on the identified adjacent digital phase samples.

[0052] In an example of the polar modulator, the computation block is designed to determine the time instances associated with the predetermined phase transitions by identifying time instances of a fixed digital clock signal that are associated with the identified adjacent digital phase samples, and using the identified time instances of the fixed digital clock signal to determine the time instances associated with the predetermined phase transitions.

[0053] In one example, the polar modulator further includes a phase processing circuit designed to receive the time instances associated with the predetermined phase transitions, calculate time instance differences based on the time instances, and generate a control signal based on the calculated time instance differences.

[0054] In an example of the polar modulator, the converter circuit is designed to receive a digital signal with a frequency significantly higher than the frequency of the data-dependent clock signal and the control signal, and to selectively choose rising and falling edges of the digital signal based on the control signal, with the selected rising and falling edges forming the data-dependent clock or being used to construct the data-dependent clock.

[0055] In an example of the polar modulator, the control signal comprises an integer part and a fractional part.

[0056] In an example of the polar modulator, the converter further includes a processing circuit designed to receive the specified time instances, calculate time instance differences based on the specified time instances, calculate frequencies associated with the calculated time instance differences, and generate a control signal based on the calculated frequencies. The converter also includes a digitally controlled oscillator designed to receive the control signal and generate the data-dependent clock signal based on it.

[0057] In one example, the polar modulator further includes a processing circuit designed to receive the interpolated digital amplitude values ​​and process adjacent values ​​thereof to form a single processed interpolated digital amplitude value for two specified adjacent time instances for the duration of a single 180° phase.

[0058] In one example, a modulator is disclosed and comprises a computation block designed to receive a plurality of digital phase samples, wherein the plurality of digital phase samples comprise digital samples of a combined baseband and carrier phase, and to determine time instances that correspond to predetermined phase transitions of the combined baseband and carrier phase. The modulator further comprises an interpolator circuit designed to receive a plurality of digital amplitude values ​​and the determined time instances and to generate a plurality of interpolated digital amplitude values ​​based on the plurality of digital amplitude values ​​of the determined time instances, wherein the interpolated digital amplitude values ​​are synchronized with respect to the determined time instances.Furthermore, the modulator comprises a converter circuit designed to generate a data-dependent clock signal with rising and falling edges assigned to specific time instances, and a digital-to-analog converter designed to receive the data-dependent clock signal and the interpolated digital amplitude values ​​or related values, and to generate a square wave output signal with transit times assigned to the generated data-dependent clock signal and an amplitude that is a function of the interpolated digital amplitude values ​​or related values.

[0059] In one example of the modulator, the converter further comprises a processing circuit designed to receive the specified time instances, calculate time instance differences based on the specified time instances, and generate a control signal based on the calculated time instance differences, and a digital time converter designed to receive the control signal and a high-frequency clock signal and select specific rising and falling edges of the high-frequency clock signal based on the control signal to generate the data-dependent clock signal.

[0060] In one example of the modulator, the converter further includes a processing circuit designed to receive the specified time instances, calculate time instance differences based on the specified time instances, calculate frequencies associated with the calculated time instance differences, and generate a control signal based on the calculated frequencies, and a digitally controlled oscillator designed to receive the control signal and generate the data-dependent clock signal based on it.

[0061] In one example, the modulator further includes a processing circuit designed to receive the interpolated digital amplitude values ​​and process adjacent values ​​thereof to form a single processed interpolated digital amplitude value for two given adjacent time instances.

[0062] In one example, the modulator further includes a processing circuit that is placed upstream of the computation block and is designed to receive the majority of digital phase samples, pre-distort the majority of digital phase samples, up-convert the majority of digital phase samples, or both.

[0063] In one example, the modulator further includes a processing circuit placed upstream of the interpolator circuit and designed to receive the majority of digital amplitude samples, pre-distort the majority of digital amplitude samples, up-convert the majority of digital amplitude samples, or both.

[0064] Particularly with regard to the various functions performed by the components or structures described above (assemblies, devices, circuits, systems, etc.), the terms used to describe these components (including references to "means") shall, unless otherwise specified, correspond to any component or structure that performs the specified function of the described component (which is, for example, functionally equivalent), even if it is not structurally equivalent to the disclosed structure that performs the function in the exemplary implementations of the invention illustrated herein.Furthermore, if a particular feature of the invention has been disclosed in respect of only one of several implementations, this feature may be combined with one or more other features of the other implementations as desired and as may be advantageous for any or a particular application. It is also furthermore that, to the extent that the terms "including," "comprising," "has," "with," or their variants are used either in the detailed description or in the claims, these terms are to be understood as inclusive in a similar way to the term "comprehensive."

Claims

[1] Method comprising: Receiving a plurality of digital samples of a modulated baseband signal, wherein the modulated baseband signal comprises a plurality of digital amplitude samples and a plurality of digital phase samples; and wherein the plurality of digital phase samples comprises digital samples of a combined baseband and carrier phase; Determining time instances associated with predetermined phase transitions of the modulated baseband signal; Generating a data-dependent clock signal with rising and falling edges associated with the specific time instances, and Interpolating the plurality of digital amplitude samples based on the determined time instances to generate a plurality of interpolated digital amplitude samples representing amplitude values ​​corresponding to the predetermined phase crossings; and clocking a digital-to-analog converter using the generated data-dependent clock signal, wherein a square wave output signal from the digital-to-analog converter includes crossing times associated with the generated data-dependent clock signal; Providing the plurality of interpolated digital amplitude samples to the digital-to-analog converter; and Modulating a magnitude of the square wave output signal based on the plurality of interpolated digital amplitude samples. [2] The method of claim 1, wherein the modulated baseband signal comprises a plurality of digital in-phase (I) samples and a plurality of digital quadrature (Q) samples. [3] The method of claim 1, wherein determining the time instances associated with the predetermined phase transitions comprises: Identifying adjacent digital samples of the combined baseband and carrier phase that span a predetermined phase crossing; Determining the time instances of a fixed digital clock signal associated with the identified adjacent digital samples; and Using the determined time instances of the fixed digital clock signal to determine the time instances associated with the predetermined phase crossings. [4] The method of claim 3, wherein using the determined time instances of the fixed digital clock signal comprises solving a linear equation based on the determined time instances. [5] The method of claim 1, wherein generating the data-dependent clock signal comprises: Calculating time instance differences based on the determined time instances associated with the predetermined phase transitions; and Generating a control signal based on the calculated time instance differences associated with the predetermined phase transitions. [6] The method of claim 5, wherein generating the data-dependent clock signal further comprises: Receiving a digital signal having a frequency that is substantially higher than a frequency of the data-dependent clock signal; and selectively selecting rising and falling edges of the digital signal based on the control signal, wherein the selected rising and falling edges form the data-dependent clock or are used to construct the data-dependent clock signal. [7] Modulator comprising: a calculation block configured to receive a plurality of digital phase samples, wherein the plurality of digital phase samples comprises digital samples of a combined baseband and carrier phase, and to determine time instances associated with predetermined phase crossings of the combined baseband and carrier phase, an interpolator circuit configured to receive a plurality of digital amplitude values ​​and the determined time instances and to generate a plurality of interpolated digital amplitude values ​​based on the plurality of digital amplitude values ​​of the determined time instances, wherein the interpolated digital amplitude values ​​are synchronized with respect to the determined time instances; a converter circuit configured to generate a data-dependent clock signal having rising and falling edges associated with the particular time instances; and a digital-to-analog converter configured to receive the data-dependent clock signal and the interpolated digital amplitude values ​​and to generate a square wave output signal having transit times associated with the generated data-dependent clock signal and having an amplitude that is a function of the interpolated digital amplitude values. [8] The modulator of claim 7, wherein the samples of the modulated baseband signal comprise a plurality of digital in-phase (I) samples and a plurality of digital quadrature (Q) samples. [9] The modulator of claim 7, wherein the calculation block is further configured to identify adjacent digital phase samples spanning a predetermined phase crossing and to determine the time instances associated with the predetermined phase crossings based on the identified adjacent digital phase samples. [10] The modulator of claim 9, wherein the calculation block is configured to determine the time instances associated with the predetermined phase crossings by determining time instances of a fixed digital clock signal associated with the identified adjacent digital phase samples, and to use the determined time instances of the fixed digital clock signal to determine the time instances associated with the predetermined phase crossings. [11] The modulator of claim 7 or 10, further comprising a phase processing circuit configured to receive the time instances associated with the predetermined phase crossings, calculate time instance differences based on the time instances, and generate a control signal based on the calculated time instance differences. [12] The modulator of claim 11, wherein the converter circuit is configured to receive a digital signal having a frequency substantially higher than a frequency of the data-dependent clock signal and the control signal, and to selectively select rising and falling edges of the digital signal based on the control signal, the selected rising and falling edges forming or being used to construct the data-dependent clock. [13] A modulator according to claim 7 or 10, wherein the converter circuit further comprises: a processing circuit configured to receive the determined time instances, calculate time instance differences based on the determined time instances, calculate frequencies associated with the calculated time instance differences, and generate a control signal based on the calculated frequencies; and a digitally controlled oscillator configured to receive the control signal and generate the data-dependent clock signal based thereon. [14] Modulator according to claim 7 or 10, further comprising: a processing circuit configured to receive the interpolated digital amplitude values ​​and process adjacent values ​​thereof to form a single processed interpolated digital amplitude value for two given adjacent time instances. [15] The modulator of claim 7, wherein the converter circuit further comprises: a processing circuit configured to receive the determined time instances, calculate time instance differences based on the determined time instances, and generate a control signal based on the calculated time instance differences; and a digital-to-time converter configured to receive the control signal and a high-frequency clock signal and to select particular rising and falling edges of the high-frequency clock signal based on the control signal to generate the data-dependent clock signal.

Citation Information

Patent Citations

  • Digital Modulation Jitter Compensation for Polar Transmitter

    US20100111222A1