Symbol synchronizer for impulse noise channels
View Patent ↗A symbol synchronizer comprises a distance metric computation module for computing a metric from received samples using a distance metric function within a symbol sampling period and based on a timing signal for each of the possible symbols in a constellation, a control signal computation module for selecting a smallest and a second smallest metrics, producing a difference by subtracting the smallest metric from the second smallest metric, and deriving a control signal from the difference, and a voltage-control clock for producing the timing signal based on the control signal.
1. A symbol synchronizer comprising:
(a) means for deriving a control signal from received samples using a distance metric function; and
(b) timing adjustment means for producing a timing signal based on the control signal where the distance metric function is denoted as ρ(x), and is defined as follows:
ρ( x )= x 2 ,
where x is related to a distance between a received sample and each possible symbol in a constellation.
2. The symbol synchronizer of claim 1 wherein the distance metric function is non-Euclidean.
3. The symbol synchronizer of claim 2 wherein the distance metric function, denoted as ρ(x), is further defined as follows:
ρ
(
x
)
=
{
x
2
fo
r
-
k
<
x
<
k
k
2
o
t
h
e
r
w
i
s
e
}
where k may be integer.
4. The synchronizer of claim 1 wherein the deriving means further comprises first distance metric computation means for computing a first set of metrics from first N consecutive received samples using the distance metric function with respect to all possible symbols.
5. The symbol synchronizer of claim 4 wherein the control signal is derived from a first difference obtained by subtracting a smallest metric from a larger metric among the first set of metrics.
6. The symbol synchronizer of claim 5 wherein the larger metric is a second smallest metric in the first set of metrics.
7. The symbol synchronizer of claim 5 wherein denoting current and past first differences as dλ 1 and dλ 2 , the control signal is derived from the value of dλ 1 −dλ 2 .
8. The symbol synchronizer of claim 4 wherein the deriving means further comprises second distance metric computation means for computing a second set of metrics from second N consecutive received samples using the distance metric function, wherein there are P samples apart from the latest sample in the first N consecutive received samples to the earliest sample in the second N consecutive received samples and P<N.
9. The symbol synchronizer of claim 8 wherein denoting the smallest metrics among the first and second sets of metrics as λ min l and λ min θ . respectively, the control signal is derived from the value of λ min θ−λ min l .
10. The symbol synchronizer of claim 8 wherein denoting the difference between the smallest and the second smallest metrics among the first set of metrics as Δλ l and that among the second set of metrics as Δλ θ , the control signal is derived from the value of Δλ θ −Δλ l .
11. A method for synchronizing a communication channel, comprising the steps of:
(a) receiving samples;
(b) deriving a control signal from the received samples using a distance metric function; and
(c) producing a timing signal based on the control signal where the distance metric function is denoted as ρ(x), and is defined as follows:
ρ( x )= x 2
where x is related to a distance between a received sample and each possible symbol in a constellation.
12. The method of claim 11 wherein the distance metric function is non-Euclidean.
13. The method of claim 12 wherein the distance metric function, denoted as ρ(x), is defined as follows:
ρ
(
x
)
=
{
x
2
fo
r
-
k
<
x
<
k
k
2
o
t
h
e
r
w
i
s
e
}
where k may be an integer.
14. The method of claim 11 wherein step (b) further comprises the step of computing a first set of metrics from first N consecutive received samples using the distance metric function with respect to all possible signal constellations.
15. The method of claim 14 wherein the control signal is derived from a first difference obtained by subtracting a smallest metric from a larger metric among the first set of metrics.
16. The method of claim 15 wherein the larger metric is a second smallest metric in the first set of metrics.
17. The method of claim 15 wherein, denoting current and the past first differences as dλ 1 and dλ 2 , the control signal is derived from the value of dλ 1 −dλ 2 .
18. The method of claim 14 wherein step (b) further comprises the step of computing a second set of metrics from second N consecutive received samples using the distance metric function, wherein there are P samples apart from the latest sample in the first N consecutive received samples to the earliest sample in the second N consecutive received samples and P<N.
19. The method of claim 18 wherein denoting the smallest metrics among the first and second sets of metrics as λ min l and λ min θ , respectively, the control signal is derived from the value of λ min θ −λ min l .
20. The method of claim 18 wherein denoting the difference between the smallest and the second smallest metrics among the first set of metrics as Δλ l and that among the second set of metrics as Δλ θ , the control signal is derived from the value of Δλ θ −Δλ l .