Jitter attenuation with a fractional-N clock synthesizer
View Patent ↗A circuit, such as, but not limited to, a digital phase-locked loop (PLL) or a transport timing loop, uses a fractional-N modulator and a fractional-N clock synthesizer to generate a clock signal, such as a transmit clock signal, from a reference clock signal. One embodiment uses a recovered clock signal derived from serial received data as a positive input to a feedback loop, and uses the transmit clock signal as a negative input to the feedback loop. After digital phase detection and digital filtering, a filtered error signal s is generated and used to control a modified fraction for control of the fractional-N synthesizer. Disclosed techniques advantageously exhibit jitter attenuation and have relatively little jitter accumulation, which are useful characteristics in telecommunication and data communication network clocking applications. Embodiments can be applied to loop timing, clock regeneration, and transport timing applications, and can be used when clock holdover is desirable.
1. A method comprising:
receiving a numerator integer and a denominator integer;
generating an error signal based on detection of a difference in phase between a first clock signal and a second clock signal;
filtering the error signal to generate a filtered error signal;
combining the filtered error signal with at least one of the numerator integer or the denominator integer to generate a modified integer;
modulating a fractional-N input of a fractional-N clock synthesizer based at least partly on the modified integer; and
using the fractional-N clock synthesizer to generate the second clock signal based on a reference clock signal and the fractional-N input;
wherein the method is performed in its entirety by one or more electronic circuits.
2. The method of claim 1 , wherein combining comprises adding the filtered error signal and the numerator integer to generate the modified integer, further comprising modulating the fractional-N input of the fractional-N clock synthesizer based on the modified integer and the denominator integer.
3. The method of claim 1 , wherein combining comprises subtracting the filtered error signal from the denominator integer to generate the modified integer, further comprising modulating the fractional-N input of the fractional-N clock synthesizer based on the modified integer and the numerator integer.
4. The method of claim 1 , further comprising:
receiving a serial data signal;
recovering the first clock signal from the serial data signal such that the first clock signal comprises a recovered clock signal;
wherein the second clock signal comprises a transmit clock signal.
5. The method of claim 4 , wherein a center frequency for the first clock signal is determined by the fractional-N clock synthesizer.
6. The method of claim 1 , wherein detecting, filtering, and combining are entirely performed by digital circuits.
7. An apparatus comprising:
a phase detector configured to detect a difference in phase between a first clock signal and a second clock signal, and to generate a error signal based on the detected difference;
a loop filter configured to filter the error signal to generate a filtered error signal;
an adder configured to combine the filtered error signal with at least one of a numerator integer or a denominator integer to generate a modified integer;
a fractional-N modulator configured to generate a modulated signal based at least partly on the modified integer; and
a fractional-N clock synthesizer configured to generate the second clock signal based on a reference clock signal and the modulated signal.
8. The apparatus of claim 7 , wherein the adder is configured to adder add the filtered error signal and the numerator integer to generate the modified integer, wherein the fractional-N modulator is configured to generate the modulated signal based on the modified integer and the denominator integer.
9. The apparatus of claim 7 , wherein the adder is configured to subtract the filtered error signal from the denominator integer to generate the modified integer, the fractional-N modulator is configured to generate the modulated signal based on the modified integer and the numerator integer.
10. The apparatus of claim 7 , further comprising a clock recovery circuit configured to receive a serial data signal and to generate the first clock signal from the serial data signal such that the first clock signal comprises a recovered clock signal, wherein the second clock signal comprises a transmit clock signal.
11. The apparatus of claim 10 , wherein a center frequency for the first clock signal is determined by the fractional-N clock synthesizer.
12. The apparatus of claim 7 , wherein the phase detector, the loop filter, and the adder consist of digital circuits.
13. A method comprising:
receiving a numerator integer and a denominator integer;
generating an error signal based on decoding of a clock measurement packets, wherein the clock measurement packets encode information associated with timing of a first clock signal;
filtering the error signal to generate a filtered error signal;
combining the filtered error signal with at least one of the numerator integer or the denominator integer to generate a modified integer;
modulating a fractional-N input of a fractional-N clock synthesizer based at least partly on the modified integer; and
using the fractional-N clock synthesizer to generate a second clock signal based on a reference clock signal and the fractional-N input;
wherein the method is performed in its entirety by one or more electronic circuits.
14. The method of claim 13 , wherein combining comprises adding the filtered error signal and the numerator integer to generate the modified integer, further comprising modulating the fractional-N input of the fractional-N clock synthesizer based on the modified integer and the denominator integer.
15. The method of claim 13 , wherein combining comprises subtracting the filtered error signal from the denominator integer to generate the modified integer, further comprising modulating the fractional-N input of the fractional-N clock synthesizer based on the modified integer and the numerator integer.
16. The method of claim 13 , further comprising:
receiving a serial data signal;
recovering the first clock signal from the serial data signal such that the first clock signal comprises a recovered clock signal;
wherein the second clock signal comprises a transmit clock signal.
17. The method of claim 16 , wherein a center frequency for the first clock signal is determined by the fractional-N clock synthesizer.
18. The method of claim 16 , further comprising encoding clock measurement packets with information associated with timing of the first clock signal.
19. The method of claim 13 , wherein generating, filtering, and combining are entirely performed by digital circuits.
20. An apparatus comprising:
a clock measurement packet decoder configured to decode clock measurement packets, wherein the clock measurement packets encode information associated with timing of a first clock signal, wherein the clock measurement packet decoder is configured to generate an error signal based on the decoded information;
a filter configured to filter the error signal to generate a filtered error signal;
an adder configured to combine the filtered error signal with at least one of a numerator integer or a denominator integer to generate a modified integer;
a fractional-N modulator configured to generate a modulated signal based at least partly on the modified integer; and
a fractional-N clock synthesizer configured to generate a second clock signal based on a reference clock signal and the modulated signal.
21. The apparatus of claim 20 , wherein the adder is configured to adder add the filtered error signal and the numerator integer to generate the modified integer, wherein the fractional-N modulator is configured to generate the modulated signal based on the modified integer and the denominator integer.
22. The apparatus of claim 20 , wherein the adder is configured to subtract the filtered error signal from the denominator integer to generate the modified integer, the fractional-N modulator is configured to generate the modulated signal based on the modified integer and the numerator integer.
23. The apparatus of claim 20 , further comprising a clock recovery circuit configured to receive a serial data signal and to generate the first clock signal from the serial data signal such that the first clock signal comprises a recovered clock signal, wherein the second clock signal comprises a transmit clock signal.
24. The apparatus of claim 23 , wherein a center frequency for the first clock signal is determined by the fractional-N clock synthesizer.
25. The apparatus of claim 23 , further comprising a clock measurement packet encoder configured to encode clock measurement packets with information associated with timing of the first clock signal.
26. The apparatus of claim 20 , wherein the clock measurement packet decoder, the filter, and the adder consist of digital circuits.