616
14 Basic Measurement and Monitoring Techniques
is the power spectral density of the ASE noise (see Sect. 11.4), B e is the receiver
electrical bandwidth, and B o is the optical bandwidth, which usually is taken to be
0.1 nm.
For OSNR values greater than about 15 dB, the signal-ASE noise is the dominating
noise factor, so that the contributions from ASE-ASE beat noise and shot noise can
be neglected. In this case, the Q factor for a distorted eye pattern can be expressed
by the simplified relationship [20]
Q =
√
A −
√
B
√
P ave
√
S AS E B e
=
√
A −
√
B
√
B e
√
O SN R
(14.23)
where the relationship between P ave and OSNR is given by Eqs. (11.24) and (11.36).
Example 14.9 Consider an amplified transmission link for which the receiver Q
factor expression for a distorted eye pattern is given by Eq. (14.23) when OSNR >
15 dB. For an OSNR value of 16, compare the values of the Q factor when (a) there
is no eye closure penalty, that is, A = 1 and B = 0 and (b) with an eye closure penalty
that has A = 0.81 and B = 0.25.
Solution From Eq. (14.23) it follows that
(a) Q = 1 ×
4
√
B e
(b) Q = (0.9 − 0.5) ×
4
√
B e
= 0.4 ×
4
√
B e
.
14.8.4 OMA Measurement Method
Compared to long-haul lightwave networks, a different approach must be taken when
testing optical Ethernet links that are based on the 10-Gb/s IEEE 802.3ae standard.
One reason is that the lasers used for long-haul links usually are high-quality devices
that operate with high extinction ratios. Therefore, in the long-haul case, to characterize the sensitivity of an optical receiver to input signals, engineers simply can
use a slow responding power meter to measure the average optical signal power and
hence determine the BER.
In contrast, the need to reduce cost in optical Ethernet links means that often
one needs to use less expensive lasers that have lower extinction ratios but provide
adequate performance at 10-Gb/s in metro, access, and campus networks. In this
case, the low extinction ratio will result in the stressed (partially closed) eye shown
in Fig. 14.22, and an average optical power measurement does not give a good indication of receiver performance. Consequently, the traditional tests used to characterize
long-haul receivers need to be modified for 10-Gb/s optical Ethernet links. This
14 Basic Measurement and Monitoring Techniques
is the power spectral density of the ASE noise (see Sect. 11.4), B e is the receiver
electrical bandwidth, and B o is the optical bandwidth, which usually is taken to be
0.1 nm.
For OSNR values greater than about 15 dB, the signal-ASE noise is the dominating
noise factor, so that the contributions from ASE-ASE beat noise and shot noise can
be neglected. In this case, the Q factor for a distorted eye pattern can be expressed
by the simplified relationship [20]
Q =
√
A −
√
B
√
P ave
√
S AS E B e
=
√
A −
√
B
√
B e
√
O SN R
(14.23)
where the relationship between P ave and OSNR is given by Eqs. (11.24) and (11.36).
Example 14.9 Consider an amplified transmission link for which the receiver Q
factor expression for a distorted eye pattern is given by Eq. (14.23) when OSNR >
15 dB. For an OSNR value of 16, compare the values of the Q factor when (a) there
is no eye closure penalty, that is, A = 1 and B = 0 and (b) with an eye closure penalty
that has A = 0.81 and B = 0.25.
Solution From Eq. (14.23) it follows that
(a) Q = 1 ×
4
√
B e
(b) Q = (0.9 − 0.5) ×
4
√
B e
= 0.4 ×
4
√
B e
.
14.8.4 OMA Measurement Method
Compared to long-haul lightwave networks, a different approach must be taken when
testing optical Ethernet links that are based on the 10-Gb/s IEEE 802.3ae standard.
One reason is that the lasers used for long-haul links usually are high-quality devices
that operate with high extinction ratios. Therefore, in the long-haul case, to characterize the sensitivity of an optical receiver to input signals, engineers simply can
use a slow responding power meter to measure the average optical signal power and
hence determine the BER.
In contrast, the need to reduce cost in optical Ethernet links means that often
one needs to use less expensive lasers that have lower extinction ratios but provide
adequate performance at 10-Gb/s in metro, access, and campus networks. In this
case, the low extinction ratio will result in the stressed (partially closed) eye shown
in Fig. 14.22, and an average optical power measurement does not give a good indication of receiver performance. Consequently, the traditional tests used to characterize
long-haul receivers need to be modified for 10-Gb/s optical Ethernet links. This