Time synchronization using skew estimation

Patent metadata

Field	Value
Publication number	US-11520372-B1
Application number	US-202017127565-A
Country	US
Kind code	B1
Filing date	Dec 18, 2020
Priority date	Feb 12, 2020
Publication date	Dec 6, 2022
Grant date	Dec 6, 2022

Abstract

Techniques are disclosed for performing time synchronization for a plurality of computing devices without relying upon a minimum measured delay. In one example, processing circuitry obtains time stamp data in accordance with an iteration of a synchronization operation for a timing protocol, wherein the time stamp data describes one or more measured delays for a path between a first computing device and a second computing device, computes a skewness estimate from the time stamp data using a regression analysis, the skewness estimate comprising a frequency difference between a first clock at the first computing device and a second clock at the second computing device, computes an offset estimate between the first clock and the second clock by applying a prediction model to the skewness estimate; and corrects at least one of the first clock or the second clock based at least on the offset estimate.

First claim

What is claimed is:

1. A method comprising:
obtaining, by processing circuitry of a computing device, time stamp data in accordance with an iteration of a synchronization operation for a timing protocol, wherein the time stamp data describes one or more measured delays for a path between a first computing device and a second computing device;
computing, by the processing circuitry, a skewness estimate from the time stamp data using a regression analysis, the skewness estimate comprising a frequency difference between a first clock at the first computing device and a second clock at the second computing device;
computing, by the processing circuitry, an offset estimate between the first clock and the second clock by applying a prediction model to the skewness estimate; and
correcting at least one of the first clock or the second clock based at least on the offset estimate.

2. The method of claim 1 , further comprising:
determining, by the processing circuitry, a level of network traffic between the first computing device and the second computing device of the network.

3. The method of claim 2 , wherein computing the skewness estimate and computing the offset estimate are in response to determining the level of network traffic exceeds a threshold.

4. The method of claim 1 , further comprising:
correcting, by the processing circuitry, a third clock at a third computing device based at least on the offset estimate.

5. The method of claim 1 , wherein computing the offset estimate further comprises:
determining an expected measured delay for a next iteration of the synchronization operation; and
determining the offset estimate based on the expected measured delay and a one-way trip time for the path.

6. The method of claim 1 , wherein the prediction model comprises a linear model for estimating at least one delay in the path according to:
{circumflex over (T)} 1 ( n+ 1)= {circumflex over (T)} 1 ( n )+ s ( n )·Δ t +Δθ( n ),

wherein:
{circumflex over (T)} 1 comprises an expected one-way delay,
n comprises an index of a time step,
s comprises a skewness estimate,
t comprises a time value, and
θ comprises a clock offset.

7. The method of claim 1 , wherein the prediction model comprises a non-linear model for estimating at least one delay in the path according to:

T
ˆ

1

(

n
+
1

)

=

T
ˆ

1

(
n
)

+

s
⁡
(
n
)

·
Δ

⁢
t

+

1
2

⁢

a
⁡
(
n
)

⁢

(

Δ
⁢
t

)

2

+

Δθ
⁡
(
n
)

,

wherein:
{circumflex over (T)} 1 comprises an expected one-way delay,
n comprises an index of a time step,
s comprises a skewness estimate,
t comprises a time value,
a comprises an acceleration term, and
θ comprises a clock offset.

8. The method of claim 1 , wherein the prediction model comprises a function for computing the offset estimate based on a minimum delay and an initial one-way trip time, the function according to:
θ( n+ 1)= {circumflex over (T)} 1 ( n+ 1)− d 1 (0),

wherein:
θ comprises a clock offset,
n comprises an index of a time step,
{circumflex over (T)} 1 comprises an expected one-way delay, and
d 1 comprises a measured one-way delay.

9. The method of claim 1 , wherein the prediction model comprises a function for computing the offset estimate from an upper band for measured delays in the path, the function according to:

θ
⁡
(

n
+
1

)

=

1
2

[

T
1

(

n
+
1

)

-

d
1

(

n
+
1

)

+

T
2

(

n
+
1

)

+

d
2

(

n
+
1

)

]

,

wherein:
θ comprises a clock offset,
n comprises an index of a time step,
T 1 comprises an upper band of a measured one-way delay,
d 1 comprises a first measured one-way delay from a first device to a second device,
T 2 comprises a lower band of a measured one-way delay, and
d 2 comprises a second measured one-way delay from the second device to the first device.

10. The method of claim 1 , wherein the prediction model com

Assignees

• Equinix Inc

Inventors

• Wang Lanfa
• Li Danjue

Classifications

• G06F1/14 Primary

  Time supervision arrangements, e.g. real time clock · CPC title

• G06F1/10

  Distribution of clock signals {, e.g. skew} · CPC title

• G06F1/12 Primary

  Synchronisation of different clock signals {provided by a plurality of clock generators} · CPC title