 |
Description  |
|
|
BACKGROUND OF THE INVENTION
1. Technical Field
The present invention generally relates to the linearization of nonlinear
optical transmission systems by the generation of a compensation signal
and, more particularly, to a method and apparatus for compensating the
nonlinearities of an optical transmission system including at least one of
a semiconductor laser, an optical amplifier, and an optical fiber
communications link by the generation of a postdistortion signal.
2. Description of the Prior Art
Optical communications system which include a semiconductor laser, an
optical fiber communication link, and an optical receiver are well known
in the art. These communications systems are adapted to carry a wide range
of information including voice, video and data.
The typical optical communications system includes a laser transmitter
which transduces an electrical information signal into an optical signal.
The optical signal is then carried over an optical fiber communications
link where it is converted back to an electrical signal by a photodetector
of an optical receiver. The transmission scheme may be analog or digital
and the modulation scheme amplitude, phase, or frequency, or any
combination of the above.
One of the most advantageous optical communication systems from the
viewpoint of simplicity and bandwidth considerations is an analog scheme
where the optical intensity of the semiconductor laser is amplitude
modulated. The optical transmission system, including the semiconductor
laser, optionally an optical amplifier, and optical fiber communications
link is required to convert the electrical information signal linearly
into an optical signal and to transmit the optical signal linearly over
the communications link. In general, distortions caused by the
semiconductor laser, the optical amplifier, and the fiber optic
communications link cause the system to operate in less than an optimum
manner. Increasingly, this type of optical communication system is playing
an important role in the delivery of high quality signals in all types of
CATV architectures.
Distortion in optical transmission systems can originate from several
different sources. One of the primary sources is the electrical to optical
transducer, a laser diode in most systems. Another contributor is the
optical communications link and, more recently, any optical amplifier in
the optical link. Some of these sources produce similar distortion signals
which may even cancel others, but usually each distortion has its own
unique characteristics and should be compensated for independently.
A laser diode generally exhibits distortion from several identifiable
causes. The first is generally that caused by the non-linearities of its
LI (light intensity as a function of electrical current) transfer
function. The nonlinearity may be superlinear (L increases at increasing
rates for increases in I), or it may be sublinear (L increases at
decreasing rates for increases in i). This type of laser diode distortion
causes mainly second order and higher intermodulation products that are
independent of frequency. This type of distortion can be corrected by
generating a distortion signal which is generally similar (sum and
difference beats of the carrier frequencies) with a similar amplitude but
opposite phase. Another characteristic distortion of the laser diode is
where the amplitude and phase of the modulating signal is distorted as a
function of frequency changes. To correct for such a distortion, a
compensation signal which varies in amplitude and phase as a function of
frequency is advantageous.
The distortion generated by an optical link is generally caused by phase
and amplitude dispersion. Generally, phase dispersion causes the different
modulating frequencies of an optical signal to be phase shifted different
amounts. Phase dispersion is proportional to the length of the optical
link and causes second order distortion with amplitude proportional to
distortion frequency. Present optical communications systems use a 1330
nm. optical wavelength signal to minimize phase dispersion. However, newer
systems that operate at a 1550 nm. optical wavelength are capable of
minimizing amplitude signal losses but with effect of increasing phase
dispersion. It would be advantageous to compensate for the increased phase
dispersion of the 1550 nm. system while maintaining its lower optical
loss. Optical amplifiers, particularly Erbium doped fiber amplifiers
(EDFA), allow greater lengths of the optical fibers to be used but produce
their own distortion in the form of an amplitude versus frequency
characteristic.
One technique of compensating for distortion from the above-described
sources utilizes a predistortion signal to compensate the RF signal
modulating the laser transmitter for the distortion in the optical
communications system. Such predistortion is discussed in commonly
assigned, copending application Ser. No. 805,251 entitled "Method and
Apparatus for Predistortion" and application Ser. No. 805,259 entitled
"Method and Apparatus for Predistortion", which are incorporated herein by
reference. However, predistortion networks cannot generally compensate for
distortion which may be unique to a particular receiver or receiver
location. For example, distortion caused by fiber dispersion is dependent
upon fiber length and thus may be different at different receiver
locations serviced by the same transmitter. Additionally, the receiver
itself may introduce distortion which may vary from receiver to receiver.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide compensation for
distortion at a receiver location in an optical communications system.
It is another object of the present invention to provide compensation at a
receiver location for distortion caused by modulation of a laser diode.
It is another object of the present invention to provide compensation at a
receiver location for distortion generated by an optical amplifier.
It is another object of the present invention to provide compensation at a
receiver location for distortion due to chromatic dispersion in an optical
fiber.
It is another object of the present invention to provide an improved
optical communication system for compensating distortion.
The present invention provides a postdistortion method and apparatus for
the compensation of a nonlinear optical transmission system. In one
preferred embodiment, the transmission system can be used for the carriage
of a broadband television signal for a CATV system. The optical
transmission system includes a semiconductor laser which acts as an
electrical signal to optical signal transducer, optionally a fiber
amplifier which increases the optical signal strength, and a fiber optic
communications link which carries the optical signal to an optical
receiver including a photodetector. The postdistortion method and
apparatus includes a distortion generator which accepts the photodetector
output and generates a distortion signal of the same general type as the
distortion inherent in the optical transmission system.
In one preferred embodiment, the invention includes a direct path and a
distortion path. An RF output of the photodetector, which may be a
broadband multichannel CATV signal, is split between the direct path and
the distortion path. The distortion path has a distortion generator fed by
a portion of the output of the photodetector to produce a distortion
signal essentially equivalent to that which is produced by the optical
communications system. The distortion generated in the distortion path is
then recombined with the signal in the direct path to produce a
cancellation or substantial suppression of the distortion.
The present invention is adapted to compensate for the distortion caused by
the modulation of a laser diode with an RF input signal with a
multiplicity of carriers. In addition, it will compensate for distortion
generated in an optical amplifier due to its variation in gain with
respect to optical wavelength. Furthermore, the postdistortion network of
the invention is particularly well-suited to compensate for distortion due
to chromatic dispersion in the optical fiber. Since this distortion is
dependent on the length of the fiber path from the transmitter to a
receiver, its characteristics can vary from one receiver location to
another. Since postdistortion is implemented at receiver locations, the
postdistortion compensation signal may be adjusted to compensate for
distortion conditions unique to that location.
According to one aspect of this embodiment, a square-law device is used to
generate the distortion signal in the distortion generator. A square-law
device generates composite second order (CSO) distortion without
generating other tinwanted distortion products. CSO distortion in a
particular channel is the total power of the separate sum and difference
beats of the other channels falling within that channel. In one
implementation of the invention, the square law device selected is a field
effect transistor, such as a GaAsFET (Gallium Arsenide Field Effect
Transistor), operated in the non-linear region near pinch off. A GaAsFET
is chosen because of its good high frequency characteristics across the
bandwidth of interest. In addition, because it is a voltage controlled
device, the operating point of the device can be varied easily and
precisely to generate the distortion characteristic desired. A device to
invert of the phase of the distortion signal is further provided to
produce the proper phase for cancellation of the system distortion.
In another embodiment of the invention, a distortion generator is utilized
in a configuration which permits the cancellation of the fundamental
frequency and the composite triple beat (CTB) component of the distortion.
A triple beat is a third-order intermodulation product of two or three
fundamental carriers which combine such as (f1+f2+f3), (f1+f2-f3), etc.
and CTB is the composite of all the beat frequencies falling in a
particular channel for a particular range of frequencies.
Another embodiment includes a plurality of distortion generators, at least
one of which has provisions for generating and cancelling CTB distortion,
and at least one of which has provisions for generating and cancelling CSO
distortion.
The present invention also provides an optical communications system
including a transmitter having a predistortion network and a receiver
having a postdistortion network, thereby affording increased ability to
compensate for distortion in the system.
BRIEF DESCRIPTION OF THE DRAWINGS
These and other objects, features and advantages of the present invention
will be better understood from a reading of the following detailed
description in conjunction with the accompanying drawings in which:
FIG. 1 is a system block diagram of an optical communications system
including a postdistortion circuit in accordance with the present
invention;
FIG. 2 is a block diagram of a first embodiment of the postdistortion
circuit illustrated in FIG. 1;
FIG. 3A is a block diagram of a first embodiment of a distortion generator
for the postdistortion circuit of FIG. 2;
FIG. 3B is a schematic diagram of the distortion generator of FIG. 3A;
FIG. 4A is a block diagram of another embodiment of a distortion generator
for the postdistortion circuit of FIG. 2;
FIG. 4B is a schematic diagram of the distortion generator of FIG. 4A;
FIG. 5A is a block diagram of another embodiment of a distortion generator
for the postdistortion circuit of FIG. 2;
FIG. 5B is a schematic diagram of the distortion generator of FIG. 5A;
FIG. 6 is a block diagram of an embodiment of the I-channel variable-gain
network of FIG. 2;
FIG. 7 is a schematic diagram of the I-channel variable-gain network of
FIG. 6;
FIG. 8 is a schematic diagram of one embodiment of the Q-channel
variable-gain network of FIG. 2;
FIG. 9 is a block diagram of another embodiment of the postdistortion
circuit illustrated in FIG. 1 used to cancel multiple orders of
distortion;
FIG. 10 is a block diagram of an embodiment of the distortion generator of
FIG. 9;
FIG. 11 is a schematic diagram of the distortion generator of FIG. 10; and
FIG. 12 is a system block diagram of an optical communications system in
accordance with another embodiment of the present invention.
BRIEF DESCRIPTION OF THE PREFERRED EMBODIMENTS
FIG. 1 is a system block diagram of an optical communication system in
accordance with the present invention. The system includes an optical
transmitter 10, an optical communications link 11, and one or more optical
nodes 12. For clarity in the Figures, the details of one only optical node
12 are shown. The other nodes are similarly configured. The RF input to
optical transmitter 10 may, for example, be a broadband CATV signal
consisting of a plurality of frequency-division-multiplexed video
carriers, although the invention is not limited in this respect. The RF
input signal is amplified by RF amplifier 14, the output of which
modulates the bias current in laser diode 15. The bias current is provided
by a current source 16. Modulation of the laser current causes the
intensity of the optical output to vary in a nearly linear fashion with
respect to the input signal, The current source and amplitude of the
modulation current are adjusted for optimum link carrier-to-noise and
distortion performance.
The optical output from optical transmitter 10 is coupled to optical
communications link 11 for transmission to one or more optical nodes.
Optical communication link 11 includes optical fibers 20 and 21 and may
optionally include an optical amplifier 22 and an optical coupler 23 for
feeding a plurality of optical nodes.
Optical node 12 includes a photodetector 30, for example, a PIN photodiode,
which transduces the intensity modulated signal into an amplitude
modulated electric current signal. Photodetector 30 is reverse biased from
a voltage source 26 through resistors 28 and 32. The photodetector output
is amplified by photodetector amplifier 34, which may be a push-pull
transimpedance amplifier or other suitable type designed for CATV
applications. The RF output of photodetector amplifier 34 is processed by
postdistortion compensation network 36 to reduce distortions generated
from causes described above. The output of the postdistortion compensation
network is amplified by buffer amplifier 38 for distribution to the CATV
network.
FIG. 2 is a block diagram of one embodiment of the postdistortion
compensation network 36 illustrated in FIG. 1. The RF input to
postdistortion network 36 is the output from photodetector amplifier 34 of
FIG. 1. The RF signal with distortion is split into two paths by
directional coupler 110 with most of the output connected to a plug-in pad
112 in a first main path, and a smaller amount connected to a plug-in pad
115 in a second postdistortion path. The main path includes plug-in pad
112 and a delay network 114. Plug-in pad 112 is used to set the RF output
level from the postdistortion network. Delay network 114, which may be a
coaxial transmission line, distributed delay line, or other suitable delay
network, provides a main path signal delay which is substantially equal to
the time for a signal to pass through the postdistortion path.
The distortion path includes a plug-in pad 115, a distortion generator 120,
directional couplers 122, 132, and 134, and a distortion signal adjustment
network 123. Plug-in pad 115 is used to set the signal level to distortion
generator 120 to optimize the output of the distortion generator with
respect to the amplitude of the desired distortion, suppression of
high-order distortion, and suppression of the broadband RF signal. The
broadband RF signal with distortion passes from plug-in pad 115 to
distortion generator 120. The purpose of distortion generator 120 is to
generate intermodulation distortion, primarily second order (CSO), third
order (CTB), or both, while substantially suppressing the undistorted
broadband signal. The output of distortion generator 120 is provided to
adjustment network 123 for adjusting the phase and amplitude of the
distortion signal. Specifically, the output of distortion generator 120 is
split by directional coupler 122 into two paths: an in-phase path which
will be referred to as the I channel, and a 90-degree-shifted quadrature
path which will be referred to as the Q channel. The I channel includes an
I channel variable gain network 124 and the Q channel includes
differentiator 126, Q channel variable gain network 128, and delay 130.
The gain and phase of the signals in the I and Q channels are adjusted as
described in greater detail below. The outputs of the I and Q channels are
combined into a postdistortion compensation signal by directional coupler
132. Directional coupler 134 provides an output from the postdistortion
path for test purposes. The postdistortion compensation signal from
directional coupler 132 is combined with the RF input signal with
distortion in the main path by directional coupler 116. With appropriate
adjustments to the gain and phase of the signals in the I and Q channels,
the combination of the RF input signal with distortion and the
postdistortion compensation signal cancels or substantially suppresses the
distortion from the RF input signal.
FIG. 3A is a block diagram of one embodiment of distortion generator 120
illustrated in FIG. 2. In particular, the distortion generator illustrated
in FIG. 3A generates second-order (CS0) distortion and includes push-pull
transformer 200 and first and second non-linear devices 202 and 204. First
and second ports C and D of transformer 200 respectively provide out of
phase signals to non-linear devices 202 and 204, which are preferably FET
amplifiers. The outputs of non-linear devices 202 and 204 are combined
such that the even ordered distortion components add and the odd ordered
distortion components cancel.
FIG. 3B is a schematic diagram of distortion generator 120 illustrated in
FIG. 3A and the non-linear devices include respective field effect
transistors 220 and 220'. The push-pull transformer 200 illustrated in
FIG. 3A includes a balun 210 and a transformer 212 which provide a
balanced output from the unbalanced RF input. A first output of
transformer 212 is coupled to non-linear device 202 and a second output of
transformer 212 is coupled to non-linear device 204. Non-linear device 202
includes a capacitor 214 coupled between the first output of transformer
212 and the control gate of FET 220. Resistor 233 and capacitor 236 are
connected in series between a reference potential and a node between
capacitor 214 and the control gate of FET 220. A gate bias is connected to
a node between resistor 233 and capacitor 236. Capacitor 224 and resistor
240 are connected in series between output node 225 and a voltage source
of a voltage +V. Capacitor 244 is connected between ground and a node
between resistor 240 and the voltage source. The current terminals of
transistor 220 are connected between ground and a point between capacitor
224 and resistor 240. Non-linear device 204 is similarly arranged and will
not be discussed in detail.
The balanced output from transformer 212 is coupled through capacitors 214
and 214' to the gates of field effect transistors 220 and 220'. Resistors
233 and 233' are terminations for transformer 212. Field effect
transistors 220 and 220' are biased from gate bias sources 230 and 230' to
near pinch-off. In this region, the transistor drain cur:rent is a
nonlinear function of gate voltage and can be described by the Curtice
Cubic Model for field effect transistors. Accordingly, the drain current
in transistor 220 can be expressed as
I.sub.1.sbsb.--.sub.ds =a.sub.0 +a.sub.1 V.sub.gs +a.sub.2 V.sub.gs.sup.2
+a.sub.3 V.sub.gs.sup.3
where I.sub.1.sbsb.--.sub.ds is the drain current, V.sub.gs is the
gate-source voltage, and a.sub.0, a.sub.1, a.sub.2, and a.sub.3 are
constants in the nonlinear FET model. The gate-source voltage for FET 220'
is the negative of the gate-source voltage of FET 220 since the field
effect transistors are fed from a balanced source. Thus, the drain current
in 220' can be expressed as
I.sub.2.sbsb.--.sub.ds =a.sub.0 +a.sub.1 (-V.sub.gs)+a.sub.2
(-V.sub.GS).sup.2 +a.sub.3 (-V.sub.gs.sup.3).sup.3
The transistor outputs are connected in parallel to output terminal 225
through capacitors 224 and 224'. Summing the output currents gives
I.sub.ds.sbsb.--.sub.total =2a.sub.2 (V.sub.gs).sup.2
Thus, the linear broadband signal and odd-order CTB distortion is cancelled
at the output of the transistors. Drain voltage for the transistors is
supplied through resistors 240 and 240'. Capacitors 236, 236', 244, and
244' are RF bypass capacitors. The gate bias sources 230 and 230' may be
adjusted for best balance to minimize the fundamental and
odd-order-distortion components at the output of the distortion generator.
Transistors 220 and 220' are preferably GaAsFET devices because of their
highly non-linear pinch-off region, common in many FETs, and their
excellent high frequency characteristics across the CATV spectrum. The
device is also precisely controllable to tailor the distortion generated
by the gate voltage. Most FETs have a nonlinear region of their operating
drive which is of use.
FIG. 4A is a block diagram of a distortion generator 120' which may also be
utilized in the present invention. In particular, the distortion generator
illustrated in FIG. 4A generates second-order (CSO) distortion and
includes hybrid transformer 305 and diodes 307 and 309, which diodes are
respectively coupled to ports B and D of hybrid transformer 305.
Distortion generated in the diodes is reflected to port C of hybrid
transformer 305, which port constitutes the output of the distortion
generator.
FIG. 4B is a schematic diagram of the distortion generator 120' illustrated
in the block diagram of FIG. 4A. Transformers 302 and 304 form a 4-port
hybrid with ports labeled A, B, C, and D. Port A receives the RF input
from plug-in pad 115. Port C constitutes the output of distortion
generator 120'. Port D is coupled to ground through capacitor 3 12 and
diode 309. Port B is coupled to ground through capacitor 3 10 and diode
307. A first terminal of potentiometer 313 is connected to the anode of
diode 309 through resistor 314. A second terminal of potentiometer 313 is
connected to the anode of diode 307 through resistor 316. The first
terminal of potentiometer 3 13 is coupled to ground through capacitor 318.
The second terminal of potentiometer 313 is coupled to ground through
capacitor 320.
The RF input is fed into port A of hybrid 305. The output of hybrid 305 is
split equally between ports B and D with a 180-degree phase difference
between the signals at ports B and D. Port C is isolated from port A,
i.e., with ports B and D terminated, no power is coupled from port A to
port C. Signals reflected from ports B and D add in phase at port C and
out of phase at port A. Thus, the broadband input signal appears across
diodes 307 and 309 but with a 180-degree phase difference. The current
through the diodes is an exponential function of the voltage across the
diodes. This diode nonlinearity causes harmonic and intermodulation
currents in the diodes which are coupled to ports A and C of the hybrid.
Because the diodes are driven anti-phase, the even-ordered distortion
components reinforce at port C of the hybrid and cancel at port A.
Odd-order distortion components cancel at port C and add at port A.
Potentiometer 313 may be adjusted to minimize the amount of odd-order
distortion, principally composite triple beat (CTB). Accordingly, the
output of distortion generator 120' is second order (CSO) distortion.
FIG. 5A is a block diagram of a distortion generator 120'' which may also
be utilized in the present invention. In particular, the distortion
generator illustrated in FIG. 5A generates third-order (CTB) distortion
and includes directional coupler 406, diodes 408 and 410 coupled to port D
of directional coupler 406, and resistor 415. Diodes 408 and 410 are
connected in an opposed, parallel relationship between port D and a
reference potential such as ground. Distortion is generated in the diodes
408 and 410 and reflected by directional coupler 406 to port C thereof,
which port constitutes the output of the postdistortion generator.
Resistor 415 provides a termination for port B of directional coupler 406.
FIG. 5B is a schematic diagram of the distortion generator 120''
illustrated in the block diagram of FIG. 5A. Directional coupler 406
includes ports A, B, C, and D. Port A receives the RF input from plug-in
pad 115. Port B is coupled to ground through resistor 4 15. Port D of
directional coupler 406 is coupled to the cathode of diode 408 and to the
anode of diode 410. Resistor 412, capacitors 416 and 418 and resistor 4 14
are connected in series between voltages +V and -V. A point between
capacitors 416 and 418 is coupled to ground. The anode of diode 408 is
connected to a node between resistor 412 and capacitor 416 and the cathode
of diode 410 is connected to a node between capacitor 418 and resistor
414.
Directional coupler 406, which may be a 12-dB coupler, couples the
broadband RF input at port A to port B with a loss of approximately 0.5 dB
and to the diodes 408 and 410 at port D with a loss of approximately 12
dB. Port C is isolated from input port A. The current through the diodes
408 and 4 10 is an exponential function of the voltage across the diodes.
This diode nonlinearity causes harmonic and intermodulation currents in
the diodes which are coupled to port A of the hybrid with a loss of 12 db
and to port C with a loss of approximately 0.5 dB. Because the diodes are
connected back-to-back, the diode current is an odd function of the diode
voltage. Thus, even-order distortion components are cancelled and only
odd-order distortion (CTB) components are reflected to the output of the
distortion generator at port C of directional coupler 406. Diodes 408 and
410 are biased from voltage sources through resistors 412 and 414. The
voltage sources may be adjusted to balance the diode characteristics and
minimize the refection of the fundamental components and
second-order-distortion to the output of the distortion generator.
Capacitors 416 and 418 are RF bypass capacitors.
FIG. 6 is a block diagram of I-channel variable-gain network 124
illustrated in FIG. 2. I-channel variable-gain network 124 includes phase
selector 300, directional coupler 302, amplifiers 306 and 308 with
corresponding gain adjustments 314 and 316, diplex filter 318 including
high pass filter 310 and low pass filter 312, and an amplitude and delay
equalizer circuit 320. Distortion from the distortion generator is coupled
through directional coupler 122 of FIG. 2 to the 0-180 degree phase
selector network 300. Phase selector network 300 is used to select a phase
appropriate for compensating the distortion in the RF input signal. The
output of phase selector network 300 is supplied to a 3-dB coupler 302.
The output of coupler 302 feeds in antiphase the high-pass and low-pass
sections of the I-channel network. Because of the complex nature of the
distortion mechanisms described previously, the composite distortion will
in general vary in magnitude and phase as a function of the channel
frequency. Accordingly, independent gain adjustments 314 and 316 for
amplifiers 306 and 308, respectively, allow for independent adjustment of
distortion at high channels and low channels.
The output of the high-pass and low-pass sections are combined at the
output of the diplex filter 318 which is comprised of high-pass filter 310
and low-pass filter 312. Diplex filter 318 is of a Butterworth design. The
Butterworth filter has the characteristic that the phase versus frequency
characteristic of the high-pass and low pass sections is identical except
for an offset of 90 degrees times the filter order. The diplex filter as
shown in FIG. 6 is of order 2. Thus, the antiphase inputs from directional
coupler 302 add in phase at the output of the diplex filter. As a result
of the Butterworth characteristic, the gain of the high- and low-channels
can be changed without changing the phase at the output. In this
implementation, a linear phase characteristic is obtained for the
I-channel with independent high- and low-channel gain adjustments.
Amplitude and delay equalizer circuit 320 enables a flat amplitude and
linear phase response to be obtained. The delay equalizer equalizes
primarily for the characteristic delay of the Butterworth filter. The
amplitude equalizer equalizes for the 3-dB increase in gain at the
cross-over frequency of the diplex filter.
FIG. 7 is a detailed schematic of the I-channel variable gain network
illustrated in FIG. 6. The I-channel input from the distortion generator
via directional coupler 122 is connected to a jumper or switch assembly
which allows the I-channel output to be switched between 0 and 180 degrees
as necessary for cancellation of the optical-link distortions. When
terminal 1 is jumpered to terminal 3 and terminal 2 is jumpered to
terminal 4, the phase of the I channel input is shifted 180 degrees. When
terminal 1 is jumpered to terminal 2 and terminal 3 is jumpered to
terminal 4, the phase of the I channel input is not shifted. Transformer
302 includes a balun 702 and a 4:1 impedance transformer 704 which provide
a balanced input to the FET amplifiers. A first output of transformer 302
is coupled to the control gate of field effect transistor 720 through
capacitor 706. One current terminal of transistor 720 is coupled to ground
and the second terminal is coupled to a power supply +Vd through resistors
724 and 728. Capacitor 780 is coupled between ground and a point between
resistors 724 and 728. The output of amplifier 798 is coupled to the
control gate of transistor 720 through resistors 778 and 710. Capacitor
711 is coupled between ground and a point between resistors 778 and 710.
The output of amplifier 798 is coupled to the inve | | |
