Optical: systems and elements – Deflection using a moving element – Using a periodically moving element
Reexamination Certificate
1994-09-06
2001-02-06
Negash, Kinfe-Michael (Department: 2733)
Optical: systems and elements
Deflection using a moving element
Using a periodically moving element
C359S199200, C359S199200
Reexamination Certificate
active
06185025
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method of transmitting information where information is represented a series of binary digits (bits) each having one of two values The present invention also relates to a transmitter for generating such information for transmission, a transmission system for conveying such information, and a communications system for communicating such information between parties.
The present invention in its different aspects finds particular, but not exclusive, application to situations in which information is to be conveyed by means of an optical fibre waveguide in which spaced-apart, discrete, optical fibre amplifiers are used to compensate for losses caused during propagation along the optical fibre.
2. Related Art
For a receiver of a particular bandwidth at the end of such a communication system to be able to detect transmitted signals within a given error rate it must receive signals having a signal-to-noise (S/N) greater than some minimum value.
In an optical fibre transmission line with in-line optical fibre amplifiers, noise is generated by amplified spontaneous emission (ASE) in the amplifiers. The total noise generated by the optical transmission line therefore depends on the number of amplifiers in the line and the ASE noise generated by each amplifier.
The ASE noise is a function of the gain of amplifier which is given by
g=0.23L
S
&ggr;
dB (1)
where L
5
is the system length in km;
n is the total number of amplifiers, all assumed the same; and
&ggr; is the system loss in dB/km.
The solution pulses propagating down the optical fibre transmission line will lose energy and be subject to intermittent amplification. In order to have propagation in which the distance average power of the pulse is equal to a single soliton power it can be shown that an Nth order soliton has to be launched into the optical transmission line where N is given by
N
2
=log(g)/(1−1/g) (2)
The minimum average power Pmin necessary to achieve an S/N ratio sufficient to give a 10
−14
bit error rate is
P
min
=10
−4
Bn[exp(45L
5
)−1]mw (3)
where B is the bandwidth in G/bits.
It can be shown that the pulse width required to generated the required Nth-order soliton pulse of the desired minimum average power P and bandwidth B is given by
t
ASE
=0.658 N
2
BD/P (4)
where D is the dispersion.
Equation (4) puts a constraint on the maximum soliton pulse width, t, to achieve the desired S/N ratio.
Another source of noise which becomes increasingly important at higher bit rates is the Gordon-Haus effect: see J. P. Gordon and H. A. Haus, Random Walk of Coherently Amplified Solitons in Optical Fibre Transmission, Optics Lett 11 665-7 (1986).
This effect induces an error due to fluctuations in arrival times, t
n
, which occurs from the combined action of an ASE induced frequency fluctuation and dispersion. The mean square jitter can be expressed as
⟨
γ
t
a
2
⟩
=
(
D
2
L
s
9
t
2
)
(
n
noise
n
pulse
)
(
5
)
where n's are photon numbers. The error rate due to this effect can be calculated.
The operating nonlinear dynamics imposes a second requirement that the soliton period be rather longer than the amplifier spacing L
A
, namely
t
spacing
>(0.3DL
1
&agr;)
½
(6)
where &agr; is a safety factor of about 10. Once these two conditions of equations (4) and (7) are satisfied one observes essentially distortionless propagation of single pulses over arbitrarily large distances.
In
FIG. 1
is shown a plot of amplifier spacing against pulse duration showing the three limiting processes (equations 4,5 and 6) for the example of a 6000 km system length. The G-H effect is the only one which depends on bit rate and this is plotted for three bit rates (10, 8 and 5 Gbit/s).
To operate a prior art soliton transmission system, t, the soliton pulse width, must be less than the G-H limit, less than t
ASE
and greater than t
spacing
.
SUMMARY OF THE INVENTION
According to a first aspect of the present invention there is provided a method of transmitting information as a series of binary digits, each having one of two values, in which a series of n optical soliton pulses, where n is an integer>1, is coupled into an optical fibre at each occurrence of one of the two values.
According to a second aspect of the present invention there is provided an optical transmitter for transmitting information as a series of binary digits each having one of two values comprising a source of optical soliton pulses arranged to provide a series of n optical soliton pulses, where n is an integer>1, at each occurrence of one of the binary digit values, and optical coupling means for coupling the series of pulses into an optical fibre.
The present invention allows one to increase the average power within a bit without having to operate with shorter pulse that would otherwise be necessary with one soliton pulse per bit of prior art systems in order to avoid the limits imposed by equations 5 and 6 on single soliton pulse per bit prior art systems. The result is that the ASE noise limit is moved to longer pulses so allowing higher bit rates notwithstanding that there are a greater number of solitons in each bit of data than with single soliton per bit schemes of modulation.
The method also reduces G-H jitter and the thus opens the window of operation by pushing the G-H limit to longer pulses. The reason for this is that the pulse will be independently subject to a jitter so that an error will occur only if both pulses are sufficiently shifted. A simple delta function model for the pulses given a n
½
reduction in the jitter.
A potential problem with the multiple pulse per bit modulation scheme of the present invention is pulse-pulse (i.e. soliton-soliton) interaction. This is a well known phenomenon which leads to a collapse of the pulses if they are initially in phase. The total energy per bit is unaltered by the collapse and so this will not, in principle, affect the system performance. It is desirable in such cases that there is no interaction with the adjacent bit pulses. It is then preferable that each series of n optical soliton pulses representative of a binary digit is separated from the next series by a time interval greater than the pulses within each series.
When pulse-pulse interaction is significant it is also desirable that each pulse in each series of n optical soliton pulses is O or &pgr;/2 radians out of phase relative to the adjacent pulses in the series.
A convenient way to achieve variations in pulse spacing and phase control for two solitons per bit is a transmitter comprising means for generating a stream of optical soliton pulses, a beam splitter located to split the stream of optical pulses into two subsidiary streams, a pair of reflective means to reflect the two subsidiary streams back to the beam splitter, the reflective means being positioned such that positions of the two subsidiary streams are combined by the beam splitter to form as an output stream of optical soliton pulses alternate ones of the subsidiary streams. Series of pulses greater than two can be readily produced by multiplexing the appropriate number of such transmitters.
The distance of one of the reflective means, for example a mirror, can be made adjustable to provide equidistantly spaced or paired optical pulses of the desired phase relationship. Once this phase is fixed the collapse effect on propagation will not reduce the energy in the bit interval. The proposed scheme can also operate in the regime of complete overlap of the pulses in the centre of the bit. This is equivalent to operating with solitons of the initial form Nsech(t) with N an integer equivalent to the number of pulses per bit. This latter multiple soliton modulation scheme will have the same benefit as for the increase in average power but will not have the benefit of the reduced jitter.
Accordin
Blown Keith James
Doran Nicholas John
British Telecommunications public limited company
Negash Kinfe-Michael
Nixon & Vanderhye P.C.
LandOfFree
Method and apparatus for transmitting information does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for transmitting information, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for transmitting information will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2608208