# US 7,411,867 B2 · Stable composite clock

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 7,411,867 B2
- Document ID: US7411867B2
- Status: Granted
- Kind: B2
- Application number: 10/967,405
- Filed: 2004-10-18
- Published: 2005-04-28
- Granted: 2008-08-12
- Earliest claimed: 2003-10-27
- Assignees: Lockheed Martin Corporation
- Inventors: Stephen R. McReynolds
- Web page: https://patentgrep.com/patent/US7411867B2

## CPC

- G01S 19/02: Details of the space or ground control segments
- G06F 1/04: Generating or distributing clock signals or signals derived directly therefrom
- G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Apparatus and method for producing a composite clock signal with optimized stability characteristics from individual clocks which have different stabilities or variances. The composite clock signal includes weighted individual clock signals. In a first (PPN) case, an optimum composite clock is a scale-factor-weighted linear combination of clock signals. In another (non-PPN) case, the optimum stable composite clock is the output of a linear filter of the input clocks signals. Both the phases and frequencies of the individual clocks over time are estimated.

## Claims

### Claim 1

A composite clock, comprising:
a plurality of individual clocks, each producing a clock signal, and the ith clock having known Allan variance curve give by a standard curve times a constant λi;
reading means coupled to said plurality of individual clocks, for simultaneously reading said clock signals; and
means coupled to said reading means and to a source of said variance information, for generating a composite clock signal according to
x
->
c
⁡
(
t
)
=
∑
i
=
1
n
⁢
K
i
*
⁢
x
->
i
⁡
(
t
)
(
57
)
where
{right arrow over (x)}c=Composite clock state (estimate of phase and/or frequency produced by the composite clock mechanism);
* represents multiplication;
{right arrow over (x)}i=Individual clock states read out by the hardware
Ki=the gain chosen to optimize the stability of the composite clock, such that
K
i
=
(
1
/
λ
i
)
/
(
∑
j
=
1
n
⁢
⁢
1
/
λ
j
)
.
(
58
)

### Claim 2 (depends on 1)

A clock according to claim 1, wherein the gains add to unity (1) in accordance with
1
=
∑
i
=
1
n
⁢
⁢
K
i
⁡
(
t
)
.
(
59
)

### Claim 3

A composite clock comprising:
a plurality of individual clocks, each producing a clock signal, and the ith clock having known Allan variance curve given by a standard curve times a constant vector
λ
→
i
=
[
λ
x
i
λ
y
i
]
;
λ
x
i
=
scale factor for short term (phase random walk) stability errors,
λiy=scale factor for long term (frequency random walk) stability errors,
reading means coupled to said plurality of individual clocks, for simultaneously reading said clock signals; and
means coupled to said reading means and to a source of said variance information, for generating a composite clock signal, the vector
[
x
c
y
c
]
according to
[
x
.
c
y
.
c
]
=
[
y
c
+
u
v
]
(
60
)
where
[
x
c
y
c
]
is the phase and frequency;
xc is phase; and
yc is the frequency,
u is the estimate of the composite clock phase noise,
v is the estimate of the composite clock frequency noise, and u and v are given by
[
u
v
]
=
∑
j
=
1
n
⁢
⁢
K
→
j
*
[
x
.
j
-
y
j
y
.
j
]
(
61
)
where
[
x
i
y
i
]
are phase and frequency estimates of the ith clock,
the optimized gains are given by
K
→
i
=
[
K
x
i
K
y
i
]
(
62
)
K
x
i
=
(
1
/
λ
x
j
)
/
(
∑
j
=
1
n
⁢
⁢
1
/
λ
x
j
)
(
63
)
K
y
i
=
(
1
/
λ
y
j
)
/
(
∑
j
=
1
n
⁢
⁢
1
/
λ
y
j
)
.
(
64
)