Method for controlling the pump power of an optical amplifier
View Patent ↗The present disclosure relates to a method for regulating a pump power of an optical amplifier, where a multiplexed broadband optical signal having several channels is amplified at a certain gain value while a change in power is detected at the input or output of the amplifier. A new pump power is calculated and adjusted based on a previously provided stable gain state of the optical amplifier after detecting the change in power such that deviations of the gain value remain minimal as planned temporary interface transients during a specific interval. The new pump power is thus calculated very accurately and quickly according to a model that takes into account the actual gain value, the wavelength dependence of active channels, aging effects, and non-linear amplification effects.
1. A method for adapting a pump power of an optical amplifier, comprising:
receiving an optical wavelength multiplex signal having a number of channels of different wavelengths;
amplifying the optical wavelength multiplex signal;
calculating a state of gain curve for the amplification of new input channel powers;
detecting a change in input or output power, wherein when the change of input power occurs within a time interval that is smaller than a reaction time of the amplifier, the accumulated input and output power is measured, and, using the measured state of gain, a new pump power value is determined in a wavelength dependent manner so that a wavelength dependent gain curve of the amplifier becomes substantially constant.
2. The method in accordance with claim 1 , wherein the state of gain is measured from a stable state.
3. The method in accordance with claim 1 , wherein the new pump power (P pump after ) is defined in accordance with a switching process changing the input power by the following characteristics:
P
pump
after
=
P
0
·
[
exp
{
P
eff
after
P
0
}
-
1
]
with
P
eff
after
=
P
eff
before
+
λ
_
signal
λ
pump
·
1
G
norm
·
{
P
sig
,
out
after
-
P
sig
,
in
after
-
P
sig
,
out
before
+
P
sig
,
in
before
}
and
P
eff
before
=
P
0
·
ln
{
1
+
P
pump
before
P
0
}
with the variables (P sig,out/in after )being measurement variables which are recorded within a period of a few 10 μs after the switching process in which the gain of the amplifier changes, and wherein
(P sig,out after ) designates the accumulated output power after the switching process,
(P sig,in after ) is the accumulated input power after the switching process,
(P sig,out before ) is the accumulated output power before the switching process,
(P sig,in before ) is the accumulated input power before the switching process,
( μ signal and μ pump ) are the average signal wavelength after the switching process or the pump wavelength, (G norm , P 0 ) are two calibration parameters of the optical amplifier, (P pump before )is the measured pump power before the switching process and P eff before/after are effective powers which do not take account of any loss mechanisms.
4. The method in accordance with claim 2 , wherein given an average gain of an EDFA amplifier without smoothing filter, the new effective pump power P eff after is calculated in accordance with the characteristic:
P
eff
after
=
P
eff
before
+
μ
_
signal
μ
pump
·
G
sig
-
1
G
norm
·
{
P
sig
,
in
after
-
P
sig
,
in
before
}
with
(
G
sig
=
P
sig
,
out
before
P
sig
,
in
before
)
designating the ratio of the average gain over the entire wavelength range and being assumed to be approximately constant before and after the switching process.
5. The method in accordance with claim 1 , wherein for a slow change of the input power, the calculation and the setting are executed successively.
6. The method in accordance with claim 2 , wherein, during stable states, new values of the pump power are calculated and read into a table, said values serving as support points for an interpolation for setting new pump powers for switching processes.
7. The method in accordance to claim 3 , wherein for N pump sources the effective pump powers before the switching process P eff,1 before of each pump source are weighted and accumulated with the quotients from the average signal wavelength μ signal and the relevant pump wavelength μ pump according to the formula
μ
eff
before
=
∑
i
=
1
N
μ
pump
μ
signal
·
P
eff
,
i
before
,
and the effective overall pump power after the switching process X eff after is calculated from the measured accumulated input and output powers before and after the switching process and the calibration parameter G norm :
μ
eff
after
=
μ
eff
before
+
1
G
norm
·
{
P
sig
,
out
after
-
P
sig
,
in
after
-
P
sig
,
out
before
+
P
sig
,
in
before
}
.