The present invention relates to a time-to-digital converters for converting signal phase differences to digital values.
In recent years, as digital phase locked loop circuits develop, time-to-digital converters, which digitize analog temporal information, have been actively developed. A typical time-to-digital converter receives a first signal into an inverter chain having a plurality of inverter circuits coupled in series, latches the output of each inverter circuit in synchronism with a second signal, and detects the point at which the status of the inverter chain changes, thereby digitizing the phase difference or time difference between the first and the second signals.
However, the numbers of latch circuits and comparators which the time-to-digital converter has to have in order to generate an n-bit digital code are each nth power of 2, which is disadvantageous in terms of circuit scale and power consumption. Thus, successive approximation time-to-digital converters for performing time-to-digital conversion by a binary search have been proposed.
In general, a successive approximation time-to-digital converter includes a plurality of time-to-digital conversion circuits coupled in cascade. In each time-to-digital conversion circuit, a phase comparator compares phases of two input signals with each other to generate a 1-bit digital value, and based on a result of the comparison of the phases, a route switch switches between routes of the two input signals such that one of the input signals which has a leading phase is input via a delay unit to a time-to-digital conversion circuit in a subsequent stage, and the other of the input signals which has a lagging phase is input directly to the time-to-digital conversion circuit in the subsequent stage. The delay units in the time-to-digital conversion circuits in the stages are set such that the delay time in each stage is reduced to a half of the delay time in a directly preceding one of the stages sequentially from the first stage, so that a time difference between the two signals input to the time-to-digital conversion circuit in the first stage is sequentially converted to a digital value by a binary search (see, e.g., Jinn-Shyan Wang, et. al, “An Ultra-Low-Power Fast-Lock-in Small-Jitter All-Digital DLL,” ISSCC 2005/SESSION 22/PLL, DLL, AND VCOs/22.7, Feb. 9, 2005, pp. 422-423,607).
In such a conventional time-to-digital conversion circuit, in order to prevent the two input signals from passing through the time-to-digital conversion circuit before determination of the result of the comparison of the phases, a delay unit has to be inserted in each of two input signal paths so as to delay input of the signals to the route switch. This increases not only latency, but also the circuit scale as well as power consumption of the time-to-digital conversion circuits and the time-to-digital converter including the time-to-digital conversion circuits coupled in cascade.
Thus, there are demands for a time-to-digital conversion circuit in which the latency, the circuit scale, and the power consumption are reduced.
A time-to-digital conversion circuit according to an example of the present invention is a time-to-digital conversion circuit for converting a time difference between two input signals to a 1-bit digital value, and adjusting the time difference between the two input signals to generate two output signals, and includes: a phase comparator configured to compare phases of the two input signals with each other to generate the digital value; a phase selector configured to output one of the two input signals which has a leading phase as a first signal, and the other of the two input signals which has a lagging phase as a second signal; and a delay unit configured to output the first signal with a delay, wherein the time-to-digital conversion circuit outputs the signal output from the delay unit and the second signal as the two output signals.
With this configuration, the phase selector distinguishes between one of the two input signals which has a leading phase and the other of the two input signals which has a lagging phase without waiting for a result of the comparison of the phases by the phase comparator, and the delay unit outputs the signal having the leading phase with a delay. Thus, it is not necessary to insert a delay unit for preventing the two input signals from passing through the circuit before determination of the result of the comparison of the phases.
The phase selector may include a first logic circuit configured to perform a logical AND operation of the two input signals, and a second logic circuit configured to perform a logical OR operation of the two input signals, and may output any one of output signals from the first and second logic circuits as the first signal and the other of the output signals from the first and second logic circuits as the second signal.
Moreover, the first logic circuit may include a NAND circuit, the second logic circuit may include a NOR circuit, and the NAND circuit and the NOR circuit may form a symmetric circuit configuration.
With this configuration, it is possible to eliminate or reduce an offset delay caused in an output of the phase selector due to a difference of gate configurations between the NOR circuit and the NAND circuit, which can improve the input/output characteristics of the phase selector.
Moreover, the phase selector may include a time difference amplifier configured to amplify a time difference between the output signals of the first and second logic circuits to output two signals, and may output the two signals output from the time difference amplifier as the first and second signals. Alternatively, the phase selector may include a time difference amplifier configured to amplify the time difference between the two input signals to output two signals, and the first and second logic circuits may perform a logical operation of the two signals output from the time difference amplifier.
With this configuration, when the time difference between the two input signals is significantly small, the input/output characteristics of the phase selector can be improved and approximated to the ideal characteristics.
Moreover, a time-to-digital converter according to an example of the present invention is configured to convert a time difference between two input signals to a digital code, and includes the above-described time-to-digital conversion circuit including a plurality of time-to-digital conversion circuits coupled in cascade, wherein delay times of the delay units in the time-to-digital conversion circuits are set such that the delay time of each delay unit is reduced to a half of the delay time of a directly preceding one of the delay units sequentially from the time-to-digital conversion circuit in a first stage.
With this configuration, it is possible to obtain a time-to-digital converter having a high speed, a small circuit area, and low power consumption.
(First Embodiment)
Specifically, the time-to-digital conversion circuit 10 includes a phase comparator 11 configured to compare phases of the signals A and B with each other to generate the digital value D, a phase selector 12 configured to output one of the signals A or B which has a leading phase as a signal LEAD and the other of the signals A or B which has a lagging phase as a signal LAG, and a delay unit 13 configured to output the signal LEAD with a delay. The signal output from the delay unit 13 corresponds to the signal A′, and the signal LAG corresponds to the signal B′.
The delay unit 13 can be a buffer circuit including a plurality of inverter circuits coupled in cascade. The delay amount can be adjusted by varying the number of inverter circuits coupled in cascade.
The phase comparator 11 performs comparison of timing of a rising edge or a falling edge between the signals A and B, thereby performing phase comparison between the signals A and B. The phase comparator 11 can be a RS flip-flop. For example, in performing comparison of timing of the rising edge, when the signal B transitions from a L level to a H level after the signal A has transitioned from the L level to the H level, the phase comparator 11 detects that the phase of the signal A leads the phase of the signal B, and outputs “1” as the digital value D. On the other hand, when the signal A transitions from the L level to the H level after the signal B has transitioned from the L level to the H level, the phase comparator 11 detects that the phase of the signal A lags the phase of the signal B, and outputs “0” as the digital value D.
The phase selector 12 performs a logical operation of the signals A and B to distinguish between one of the signals A or B which has a leading phase and the other of the signals A or B which has a lagging phase.
More specifically, the AND circuit 121 can include a NAND circuit 123 and an inverter circuit 124 configured to invert an output of the NAND circuit 123. The OR circuit 122 can include a NOR circuit 125, and an inverter circuit 126 configured to invert an output of the NOR circuit 125.
In general, a configuration of the gate is different between a NAND circuit and a NOR circuit, and thus gate delays are different. Thus, when the NAND circuit 123 and the NOR circuit 125 have general circuit configurations, an offset delay is added to an output of the phase selector 12 due to the difference between the gate delays. As described later, when multiple ones of the time-to-digital conversion circuit 10 are coupled in cascade to form a time-to-digital converter, an offset delay included in an output of at least one of the phase selectors 12 may reduce conversion accuracy. Thus, as illustrated in, for example,
Moreover, when the time difference between the signals A and B is significantly small, the NAND circuit 123 and the NOR circuit 125 behave in a manner comparable to the case where identical signals are input as the signals A and B, and operate as inverter circuits having substantially doubled drivability. Thus, as indicated by the broken line in
As descried above, according to the present embodiment, the time difference between the two input signals can be adjusted without waiting for the result of the comparison of the phases, and the adjusted time difference can be output. Thus, it is no longer necessary to insert a delay unit in the two input signal paths to wait for the result of the comparison of the phases, so that the latency of the time-to-digital conversion circuit can be reduced, and the circuit scale and the power consumption can also be reduced.
(Second Embodiment)
When the conversion input range of the time-to-digital converter 100 is ±T, the delay times of the delay units 13 of the time-to-digital conversion circuits 10 are set, for example, such that the delay time in the first stage is T, the delay time in the second stage is T/2, and the delay time in the last stage ((n−1)th stage) is T/2n−2, that is, the delay time of each delay unit is reduced to a half of the delay time of a directly preceding one of the delay units sequentially from the time-to-digital conversion circuit 10 in the first stage. The digital output D(1) of the time-to-digital conversion circuit 10 in the first stage corresponds to the MSB of the digital code, and the digital output D(n) of the phase comparator 11 corresponds to the LSB.
As described above, according to the present embodiment, time-to-digital conversion circuits 10 having low latency , a small circuit area , and low power consumption are used, so that it is possible to obtain a time-to-digital converter with a high speed, a small circuit area, and low power consumption.
A gray-coded n-bit digital code is output from the time-to-digital converter 100, but the gray code is easily converted to a normal binary code.
Moreover, in the last stage, the phase comparator 11 is provided, because signals do not have to be transmitted from the last stage to a subsequent stage, and thus the phase selector 12 and the delay unit 13 are omitted from the time-to-digital conversion circuit 10. However, the time-to-digital conversion circuit 10 may also be provided in the last stage.
Number | Date | Country | Kind |
---|---|---|---|
2011-049200 | Mar 2011 | JP | national |
This is a continuation of International Application No. PCT/JP2011/004285 filed on Jul. 8, 2011, which claims priority to Japanese Patent Application No. 2011-049200 filed on Mar. 7, 2011. The entire disclosures of these applications are incorporated by reference herein.
Number | Date | Country | |
---|---|---|---|
Parent | PCT/JP2011/004285 | Jul 2011 | US |
Child | 13942478 | US |