Optical: systems and elements – Optical amplifier – Correction of deleterious effects
Reexamination Certificate
1999-12-20
2001-09-18
Hellner, Mark (Department: 3662)
Optical: systems and elements
Optical amplifier
Correction of deleterious effects
C359S199200
Reexamination Certificate
active
06292290
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to signal power adjustment and more particularly to methods and apparatus for adjusting power in an optical signal, for providing a seamless optical ring and for providing a bidirectional equalized amplifier for optical signals.
BACKGROUND OF INVENTION
In fiber optic communications, information is transmitted as optical signals, i.e. pulses of light at a particular wavelength. Using a system called Wavelength Division Multiplexing (WDM), a plurality of such optical signals, or channels, having respective unique optical wavelengths, are multiplexed to form a WDM signal for transmission on a single optical fiber. However, an optical fiber is not a loss free medium and causes attenuation of the WDM signal. Furthermore, the attenuation is not uniform at all wavelengths of light. Signals having relatively short wavelengths, less than 800 nm for example, are heavily attenuated due to Rayleigh scattering, while signals having relatively long wavelengths, greater than 1600 nm for example, are heavily attenuated due to infrared absorption. Between these extremes, there are local attenuation maxima caused by impurities in the fiber, for example. In addition, optical devices such as optical multiplexers and demultiplexers can cause further non-uniforn attenuation over the range of wavelengths used.
To compensate for the non-uniform attenuation caused by the fiber and other optical devices, respective powers of the individual channel signals which form the WDM signal are adjusted relative to one another such that the WDM signal will have an optimal spectrum when it is initially generated. Signals having wavelengths which are subject to relatively high attenuation will be adjusted to have relatively high power, and signals having wavelengths which are subject to relatively low attenuation will be adjusted to have relatively low power As the WDM signal propagates along the fiber and passes through optical devices, however, the non-uniform fiber attenuation will cause the spectrum of the WDM signal to depart from the optimal spectrum.
To counteract fiber and optical device attenuation losses, it is necessary to include optical amplifiers, such as erbium doped fiber amplifiers (EDFAs) for example, at intervals along the fiber to amplify an attenuated WDM signal so that none of the individual channels which form the WDM signal is attenuated below a minimum detectable power. However, the optical amplifiers may not have a uniform gain for all wavelengths of the WDM signal, and may be susceptible to saturation at some wavelengths thereby causing the overall spectrum of the WDM signal to further depart from the optimal spectrum. Failure to restore the optimal spectrum of the WDM signal can reduce the distance between which optical amplifiers can be spaced as the minimum detectable power of a most highly attenuated signal will be reached more quickly than would be the case for a WDM signal having an optimal spectrum.
In some existing optical communications systems, signal or channel power adjustment is provided by placing discrete narrow-band attenuators at intervals in the fibers according to prescribed rules, calculations and trial and error. The attenuators increase the attenuation of less attenuated channels, resulting in a more uniform attenuation over all WDM signals. Effectively, these discrete attenuators serve to make the loss characteristic of the fiber less wavelength dependent. However, this method may require a number of iterations and may require additional adjustment each time the network is expanded or modified.
Another method of providing channel power adjustment involves demultiplexing the WDM signal to separate the plurality of individual channels, individually attenuating each of the channels and remultiplexing the channels to reproduce the WDM signal prior to amplification. However, this method may result in unacceptably high signal losses since an optical demultiplexer typically uses a plurality of optical filters connected in series to separate the channels from the WDM signal. Each filter has an inherent loss, such that channels which must pass through several filters will experience a cumulative loss. If the WDM signal contains a large number of channels, 32 for example, the cumulative filter loss of 32 filters may be significant, reducing the spacing between amplifiers.
An additional need for channel power adjustment arises in the context of optical ring networks or meshed networks. Such networks have redundant paths between nodes, providing greater network reliability but also creating closed loop paths such that with the addition of optical amplifiers positive feedback can occur. Optical noise may accumulate and increase in magnitude with each circuit around the ring, leading to unacceptable signal-to-noise ratios, high bit error rates and, in the extreme case, to lasing and system overload. To prevent such noise accumulation, it is necessary to provide sufficient attenuation to prevent positive feedback from occurring. It has been found that a ring loss of at least 10 dB is necessary to prevent noise accumulation. However, while this amount of loss is desirable for system stability, it may have undesirable effects on signal transmission.
In most prior art optical rings, noise accumulation is prevented by providing a seam which interrupts the flow of optical signals around the ring. Conventionally, this seam is provided by an apparatus at a network node which performs electrical regeneration, by which a received WDM optical signal is demultiplexed, converted to electrical signals, filtered to remove noise and then converted back into a WDM optical signal for retransmission. However, as optical transmission rates increase, it has become difficult to perform the optical-to-electrical-to-optical conversion at sufficiently high speeds.
Therefore, there is a need for an apparatus which Provides variable attenuation of respective optical signals in a WDM signal to achieve an optimal WDM spectrum without causing unacceptable signal losses. There i s a particular need for such an apparatus for use in a seamless optical ring so as to provide selective attenuation representing a low loss path to an in-band WDM signal and representing a high loss path to an out-band noise signal.
SUMMARY OF THE INVENTION
The above problems with the prior art are addressed by the present invention which provides a method and apparatus for adjusting power in an optical signal. Effectively the method and apparatus involve splitting an input composite optical signal into respective drop signals in separate wavelength bands, attenuating the drop signals to produce attenuated drop signals and combining the attenuated drop signals to produce a first output composite optical signal.
In one embodiment, the input composite optical signal is split by producing respective drop signals and respective pass signals. The respective pass signals are cascaded through successive drop signal splitters to accumulate the pass signals into a remainder signal. At the same time, drop signals are produced by respective drop signal splitters and such drop signals are attenuated by separate respective attenuators to produce respective attenuated drop signals.
In one embodiment, splitting Is achieved by filtering respective predecessor pass signals and producing respective pass signals therefrom. filtering may be provided by dielectric interference filters, for example.
Preferably, the attenuation of the respective drop signals is variable and may be set by a control device or manually for example.
The combining of the attenuated drop signals may be achieved by passing the attenuated drop signals through respective cascaded drop signal combiners to successively add the drop signals together to produce the first output composite optical signal. A combined band signal is produced at at least some of the drop signal combiners in response to respective attenuated drop signals and a respective combined band signal from a predecessor drop signal combiner.
Signals which are not in
Adolph Stephen Kieran
Liu Kexing
Remedios Derrick Anthony John
Wan Ping Wai
Hellner Mark
Nortel Networks Limited
LandOfFree
Methods and apparatus for adjusting power in an optical... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods and apparatus for adjusting power in an optical..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods and apparatus for adjusting power in an optical... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2504768