JPH045296B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a device for transmitting electromagnetic pulse signals by means of a single mode fiber.

Prior Art Significant advances have been made in the field of optical telecommunications in recent years. Systems are currently being installed that transmit data at speeds of several megabits per second between repeaters over distances of several kilometers. However, since the economics of systems such as intercontinental submarine cable systems are strongly influenced by data rates and repeater spacing, efforts continue to improve these parameters of these systems.

Although currently available fibers are capable of transmitting signals with relatively low loss and low dispersion, further improvements are desired. Therefore, in fiber telecommunications links, it is necessary to perform signal regeneration at a so-called "repeater" located at an intermediate point between the sending end, or input location, and the receiving end, or output location, of the fiber transmission channel. Although the terms "input location" and "output location" here refer to unidirectional transmission, it is also possible to reverse the "input location" and "output location" for subsequent transmission.

A repeater typically performs two functions. i.e. increasing the power level of the signal pulse;
This is to reshape the pulse. In addition to this, repeaters often perform pulse retiming. The increase in power level is necessary due to the attenuation that the signal undergoes in the actual fiber. Reshaping is necessary because the pulse spreads due to dispersion effects in the fiber. Retiming is often required to maintain proper pulse spacing.

A repeater in a fiber telecommunications system typically includes a means for detecting a signal, e.g. a photodiode, a means for acting on the output of the photodiode, e.g. a means for amplifying and reshaping the electrical output signal of the detector, a detector. an optical radiation source, typically modulated by an amplified and reshaped output signal of the optical radiation source, and means for recombining the output of the optical source into the fiber. It is envisaged that repeaters of the type described above are not only currently in use, but will also be used in future fiber telecommunications systems. For example, P. E. Radley (PE
Radley) and AWHorsley's Proceedings of
Proceedings of the International Conference on Submarine Telecommunication Systems
Conference on Submarine
Telecommunication Systems), London;
See February 1980, pp. 173-176.

Conventional repeaters are typically complex devices containing multiple elements. For example, a typical optical repeater has approximately 50
contains transistors.

(See supra.) This ``electronic'' complexity makes it difficult to use repeaters for fiber telecommunications systems currently being considered, especially from the point of view of laser source reliability in high bit rate systems. will account for the major part of the cost.

Traditional countermeasures to these facts have been, inter alia, to improve the quality of the fibers, so that repeater spacings of about 50 km now seem possible. Nevertheless, the difficulties encountered with the use of repeaters require other solutions, and the present invention relates to other solutions. Some characteristics of fibers relevant to the present invention are discussed below.

Pulses of electromagnetic energy transmitted through optical fibers experience attenuation and dispersion. In particular, dispersion causes the pulse to spread out in the time domain.
If the pulse spread is large, adjacent pulses will overlap, making detection of the signal impossible. For single mode fibers (ie, fibers in which only the fundamental mode of the signal propagates at the operating wavelength of the system), the two primary dispersion mechanisms are material dispersion and waveguide dispersion.
A material with a refractive index n exhibits material dispersion when d ² n /dλ ² is not 0 at a wavelength λ. Physically,
This means that the phase velocity of a plane wave propagating in this medium varies non-linearly with wavelength, so that the light pulse spreads as it propagates through the medium.
Waveguide dispersion is also typically wavelength dependent. Here, the combination of material dispersion and waveguide dispersion will be referred to as "chromatic dispersion." For example, due to chromatic dispersion effects in a typical single mode fiber, a 10 picosecond pulse with a carrier wavelength of 1.5 μm doubles in width after traveling approximately 650 meters.

If d ² n/dλ ² > 0 over the entire wavelength range in a medium, the medium is said to be normally dispersive in that wavelength range. On the other hand, a wavelength range in which d ² n/dλ ² <0 forms a so-called anomalous dispersion range. For example, for silicon, the normal dispersion range is from short wavelengths to about 1.27 μm, and the anomalous dispersion range is from about 1.27 μm to longer wavelengths. The wavelength that separates the two ranges is the wavelength where d ² n/dλ ² =0, and the material dispersion becomes first-order 0 at this wavelength. This wavelength depends on the composition of the medium.
The wavelength at which the chromatic dispersion is first-order zero also depends on the composition, but in addition depends on fiber parameters such as fiber diameter and doping profile. This wavelength is, for example, approximately 1.5μ for a suitably designed single-mode silicon-based fiber.
It can be as long as m.

The most natural choice of carrier wavelength in a high data rate fiber telecommunications system is to choose a wavelength for which the chromatic dispersion in the fiber is first order zero. However, even at this wavelength there is pulse broadening due to higher order terms of dispersion. Regarding this, for example, F.P. Capron (FP
Electronics Letters, Vol. 13, pp. 96-97 (1977)
Please refer to 2013).

Recently, single-mode
Nonlinear variation of the dielectric constant of the fiber (Kerr)
It has been proposed to use ``solitons'', i.e. ``solitons''.

Soliton pulses occur when the broadening effects due to chromatic dispersion are balanced by contraction due to the nonlinear dependence of the refractive index on the electric field. The existence of solitons in single-mode fibers and the possibility of their stable transmission are discussed in a paper by A. Hasegawa and F. Tappert, Applied Physics Letters.
Physics Letters), Volume 23, No. 3, pp. 142-144
(1973). This paper deals with lossless single-mode fibers and reveals the existence of a minimum pulse peak power at which a soliton can exist, which depends on the fiber parameters, pulse width, and carrier wavelength. These predictions of Hazegawa and Tapert were demonstrated by dispersion-free transmission of a 7 picosecond pulse with a peak power of about 1 watt at 1.45 μm over a distance of about 700 meters through a single mode fiber. Regarding this, there are papers by L. F. Mollenauer et al., Physical Review Letters.
Letters, Vol. 45, No. 13, pp. 1095-1098 (1980
Please refer to 2013). Molenauer et al. also substantiated Hasegawa and Tapert's prediction that soliton pulses with peak powers above the so-called "equilibrium" peak power P ₀ will become narrower.

Recently, A. Hasegawa and Y. Kodama proposed the use of soliton pulses in high data rate single mode fiber telecommunications systems. Regarding this, see Proceedings of the IEEE, Volume 69, No. 9,
See September 1981, pp. 1145-1150. This paper describes in detail the properties of solitons in ideal optical fibers, the effects of higher-order dispersion and loss on solitons, and design examples and evaluation methods.

The proposed telecommunications system uses self-confining effects to achieve high data transmission rates. However, they do not address the problem of pulse regeneration or the problems inherent in the conventional regeneration methods mentioned above.

SUMMARY OF THE INVENTION A fiber optic telecommunications device in accordance with the present invention includes a single mode optical fiber capable of transmitting soliton pulses, each positioned at an intermediate location along the fiber to amplify the soliton pulses propagating through the fiber. one or more non-electronic amplifiers. A non-electronic amplifier is defined herein as an amplifier constructed to amplify a light pulse signal while the signal preserves the form of the light pulse throughout the amplification process.

An example of a suitable amplifier is a glass laser doped with suitable ions (i.e. ions with energy levels separated by hc/λ ₀ , where h is Planck's constant and c is the speed of light in vacuum). There is a glass medium (typically a fiber) pumped by electromagnetic radiation that causes a collective reversal of energy levels. Examples of other amplification means include Raman amplifiers;
For example, the "Stokes" of λ ₀ pumping radiation
There is a glass medium (typically a fiber) within the (Stokes) wavelength range. See, for example, the book "Optical Fiber Telecommunications" edited by SEMiller and AGChynoweth, Academic Press, 1979.
, pp. 127-132. An example of yet another amplification means is to inject a continuous wave (CW) with a wavelength equal to λ ₀ , in phase with the soliton, and with an amplitude less than the pulse amplitude, thereby
There are amplification means to increase the pulse amplitude by non-linear interaction between the CWs. Another example of amplification means is a semiconductor laser that acts as an amplification medium.

The amplifier examples described above are examples of non-electronic amplifiers. A property shared by these amplifiers is that they allow preservation of the phase of the pulse.

The soliton pulse changes its final (i.e., asymptotic) shape and pulse height at the point of "amplification", that is, when the energy is transferred to the pulse (see Hasegawa and Kodama, cited above). Rather than achieving a constant value, the pulse undergoes changes in pulse width and amplitude as it propagates through the fiber after being amplified, and after propagating a distance L _NL of the order of magnitude specified below, its final It will be appreciated that the shape and amplitude are achieved.

The initial pulse power and pulse width as well as the amplifier spacing and amplification factor are advantageously chosen such that the aforementioned changes result in a narrower pulse and an increased amplitude.

Embodiments of the present invention will be described below with reference to the accompanying drawings.

DESCRIPTION OF THE PREFERRED EMBODIMENTS As shown by Hasegawa and Kodama (ibid., p. 1147), the equilibrium peak electric field φ ₀ of a soliton pulse, i.e., a signal that maintains its pulse shape indefinitely in a lossless ideal fiber, is The peak electric field is given by the following equation.

φ ₀ =√2(−λ ² ∂ ² n/∂λ ² )/ω ₀ t ₀ √n ₂ ^1/2 (1) In this equation, λ is the wavelength of the carrier wave in free space,
n is the refractive index of the fiber, ω ₀ is the angular frequency of the carrier wave,
t ₀ is the half-maximum pulse width and n ₂ is the nonlinear refractive index of the fiber. Equation (1) can be used to define the equilibrium peak power P ₀ .

P ₀ = (1/2) v _g φ ₀ ² ε ₀ sn ² (2) In this equation, v _g is the group velocity c/n (c is the speed of light),
ε ₀ is the permittivity of vacuum, S is the cross-sectional area of the fiber, and n is the refractive index of the fiber.

As shown by Hasegawa and Kodama, one soliton pulse can exist for peak powers between 1/4P ₀ and 9/4P ₀ . P ₀
Solitons with a peak power between and 9/4P ₀ have a narrow pulse during the transmission period, and solitons with a peak power between 1/4P ₀ and P ₀ have a wide pulse during the transmission period. The aforementioned authors have also identified the requirements necessary for the generation of soliton pulses and the design conditions for a single mode fiber soliton transmission system, and these requirements and conditions will not be repeated here. Multiple solitons
The pulse has a peak power greater than 9/4P ₀ .
Although it is possible to use multiple soliton pulses in the present invention, it is not desirable.

Here it is more convenient to distinguish between pulse regeneration and pulse amplification. "Regeneration" shall mean at least a process in which the amplitude of the pulse is increased and the pulse width is decreased. This regeneration is typically performed in a device commonly called a repeater and involves changing the entity carrying the signal from a photon to, for example, an electron and then back again from an electron to a photon.

On the other hand, "amplification" refers to a process in which the amplitude of a pulse is substantially changed, and refers to the case where no means for changing the width or shape of the pulse is provided. A change in amplitude here refers to an increase. At least a portion of the amplification process is typically performed by a device referred to herein as an amplifier.

In implementing the invention, it is also conceivable to use, in addition to the amplifier, additional non-electronic pulse shaping means, for example a section of fiber made of a material with a large Kerr coefficient.

Since the only factor that degrades a soliton pulse due to pulse broadening is fiber loss, losses and chromatic dispersion (including higher order dispersion) do not substantially change the basic shape of the soliton pulse. It is possible to obtain a pulse width preserving channel by providing non-electronic amplification means for the soliton pulse without providing extraneous pulse shaping means. In other words, non-electronic amplification of soliton pulses will serve as the complex pulse regeneration that was previously required.

FIG. 1 is a schematic diagram of a generalized fiber telecommunications system embodying the present invention. The electromagnetic radiation pulses emitted by the pulse generating means (pulse source of electromagnetic radiation) 10 are coupled into the single mode fiber 12 by the coupling means 11 . The generation of pulses is controlled by input signal 15. Since any real fiber attenuates the pulses transmitted therethrough, the amplitude of the pulses arriving at non-electronic amplifier 13 is also reduced compared to when coupled to the input location of the fiber. amplifier 1
After being amplified at 3, the pulses are transmitted further through the fiber and periodically re-amplified by further amplifiers 13, and finally the pulses reach the output location of the transmission channel and are detected by detection means 14. Pulse reshaping typically occurs during transmission. A signal 16 is extracted from the detection means.
This signal 16 contains the information carried by signal 15.

The requirement for the existence of a soliton pulse is that the wavelength of the carrier wave lies within the anomalous dispersion region of the fiber. Here, "carrier wavelength" means the center wavelength of the pulse spectrum. For silicon-based fibers, this condition means that the wavelength of the carrier wave must be approximately 1.27 μm or greater. An advantageous wavelength for operation of the system according to the invention is in the vicinity of 1.5 μm. This is because silicon-based fibers typically have minimal loss in this wavelength range (fiber losses can be as low as 0.2 db/Km).

Any coherent electromagnetic radiation source of suitable wavelength and intensity can be used. For example, suitable semiconductor lasers or gas lasers can be used as radiation sources. Means of coupling pulsed radiation into fibers are also well known to those skilled in the art and will not be described here. Similarly, means for detecting signal pulses are well known to those skilled in the art and will not be discussed here.

Possible amplifier means include doped glass lasers, Raman lasers, semiconductor lasers, and amplifiers using continuous wave (CW) injection as detailed below.

Figure 2 shows that two pulses, each originally having a pulse width of about 14 picoseconds and separated by about 57 picoseconds, have a cross-sectional area
This figure shows the temporal order of the calculated pulse shape when propagating through a 20 μm ² silicon-based single mode fiber.
This pulse has a carrier wavelength of 1.5 μm and the fiber is assumed to have a loss of 0.2 db/Km at this wavelength frequency. Furthermore, the pulse is 1.26
Assume that the input voltage is ×10 ⁶ V/m and the equilibrium peak power P ₀ under assumed conditions is 105 mW. A curve 20 is a calculated pulse waveform for a soliton. That is, here n ₂ = 1.2
A nonlinear refractive index of ×10 ⁻²² (m/V) ² was used in the calculations. As can be seen in Figure 2, without amplification, the soliton pulse broadens significantly;
After 22.5 km, the pulses essentially merge.
Curve 21 is the calculated pulse waveform for two linear pulses having the same initial amplitude and width as the pulses of curve 20. Here, "linear" means that the nonlinear coefficient of refractive index is 0. As can be seen, the linear pulse undergoes extreme deformation after about 7.5 Km.

FIG. 3 shows the calculated pulse waveforms for two periodically amplified solitons. The fiber properties are assumed to be the same as in FIG. 2, and the same initial pulse waveforms and amplitudes are used. 9.4Km, 18.8Km and 28.2Km
It is assumed that 1.9 db amplification is performed afterwards. As can be seen, under these conditions the soliton pulse substantially retains its shape and other attributes.

FIG. 4 shows that two solitons with the same initial conditions as assumed in FIGS. 2 and 3 propagate through a fiber with the same properties as previously assumed, and
FIG. 3 is a calculated diagram of a pulse waveform when CW injection is performed. The CW is assumed to have the same wavelength as the pulse carrier wavelength, to be in phase with the soliton, and to have an amplitude of 11% of the peak amplitude of the initial soliton. Assume that injections occur at 9.4 Km, 18.8 Km and 28.2 Km. As can be seen in FIG. 4, under the assumed conditions, the soliton pulse also substantially retains its shape and other attributes.

Figures 5 and 6 show the calculated soliton pulse pairs at 1080 km and 5940 km when amplification of about 1.3 db is performed every 6.75 km. The assumed fiber characteristics are the same as in Figure 2, and the input peak power is 11.2m.
W, the pulse width at 1/2 of the maximum value is about 42 picoseconds, the pulse interval is about 170 picoseconds, and the carrier wavelength is
It is 1.5 μm. Curve 50 in Figures 5 and 6 represents the input pulse, and curve 51 in Figure 5 represents the fiber.
Curve 60 in FIG. 6 represents the same soliton pair after passing about 6000 km through the fiber. As can be seen, the shape of the pulse is very well preserved under the hypothesized conditions. The change in pulse interval seen in Figure 6 is 2
This is due to mutual interference between two solitons.

As stated by Hasegawa and Kodama (ibid., p. 1147), a pulse having an initial peak power other than the equilibrium peak power will undergo a change in pulse width and amplitude as it propagates. For example, the envelope
The input pulse given by aq ₀ sech (q ₀ τ) (where 1/2a3/2) is a∽q ₀ sech (a∽q ₀ τ) (where a∽=
Asymptotically give one soliton pulse (in the undamped case) with an envelope given by (1+2α), a=1+α, 1α1<1/2). Therefore, a=3/
2, the maximum possible amplitude of one soliton pulse is a∽=2, the peak power of the asymptotic soliton is about 4 times the peak power of the input pulse, and the asymptotic width is reduced to about 1/2 of the initial pulse width. do. Similarly,
When the soliton pulse is multiplied by a (a > 1), in the absence of fiber amplitude, the resulting asymptotic soliton has an amplitude approximately (2a - 1) times the original soliton pulse, and a width of the original soliton. Approximately (2a) of pulse width
-1) Decrease by ^-1 times.

Note that before the pulse waveform settles into its asymptotic single soliton, its shape oscillates and the pulse loses some energy. For silicon-based fibers, the vibration period L _NL (meters)
is approximately given by the following formula.

L _NL (m) = 5.3λ (μm) S (μm ² ) / P (W) where P (W) is the peak power of the pulse (watt),
λ (μm) is the wavelength of the carrier wave in μm and S is the core cross-sectional area of the fiber core in (μm) ² .

Injecting a CW into the fiber having the same wavelength and substantially the same phase as the soliton carrier can be used as a means to amplify the soliton pulses. Using this injection, soliton pulses with narrower width and larger amplitude are obtained. The portion of the CW that is not used for pulse amplification is removed by further injection of CW at a subsequent amplification point located at an appropriate location, resulting in destructive interaction of the CW.

According to the analysis results, the amplitude of a soliton whose carrier wave and CW have the same wavelength and phase is approximately π of the CW amplitude.
It is shown that the width increases by a factor of 1 and the width decreases by the same amount. In this way, when the amplitude of the soliton decreases to about 1/πE ₀ due to damping, the amplitude E ₀ becomes
If CW is injected into the fiber, the original soliton structure will be restored.

The undesired accumulation of unused CWs can be avoided by spacing the injection points so that they always have a constructive interaction with the soliton carrier while at the same time having a destructive interaction with successively injected CWs. I can do it. It is always possible to choose such an interval. This is because the phase of a soliton shifts continuously during propagation, whereas the phase of a CW remains constant. Hasegawa and Kodama (ibid.) gave a formula for determining the soliton phase as a function of propagation distance, but with a suitable CW
The injection point of can be determined by using this formula. If the initial phase of the soliton is τ ₀ , the CW injection may be performed at a propagation distance T such that the phase is τ(T)=2π/m+τ ₀ , where m is a positive integer.

As an example of amplification by repeated injections of CW using the same parameters described in connection with Figure 2, the original soliton has a width of about 14 ps and a power of about 105 mW.
When giving a balanced peak power of , the amplitude every 9.4Km
It is possible to keep the soliton pulse virtually unchanged by injecting a CW of 1.8×10 ⁵ V/m.

Single mode optical fiber 12 Non-electronic amplifier 13

[Brief explanation of drawings]

Fig. 1 is a diagram showing an example of a remote communication system embodying the present invention, and Fig. 2 is a diagram showing an example of a remote communication system in which no amplification is performed.
Figure 3 shows the pulse waveform calculated for a pair of soliton pulses and a pair of linear pulses. Figure 3 shows the pulse waveform calculated for a pair of periodically amplified soliton pulses. Figure 4 shows the calculated pulse waveforms for a pair of soliton pulses with periodic continuous wave (CW) injection and a pair of soliton pulses without continuous wave (CW) injection. Figures 5 and 6 show distances of approximately 1000 km and approximately 1,000 km, respectively.
FIG. 3 shows the calculated pulse waveform for a pair of periodically amplified soliton pulses propagated through a 6000 Km fiber. [Explanation of codes of main parts] Claim codes

Claims (1)

[Claims] 1 (a) Pulse source 1 of electromagnetic radiation with carrier wavelength λ ₀
(b) a fiber transmission channel having an input location and an output location spaced from the input location, the fiber being a single mode fiber for radiation of wavelength λ ₀ and having anomalous dispersion in a given wavelength range; a fiber transmission channel 12 consisting of an optical fiber, (c) means 11 for coupling one or more said pulses to said channel at an input location and detecting at an output location the pulses transmitted through the channel from the input location to the output location; 14, wherein the pulse shape at a given position in the channel is defined by a peak power and a pulse width, and the single mode fiber has a loss at wavelength λ ₀ and the loss means such that the peak power of the pulse decreases with increasing distance from the input location; and (d) at least one non-electronic amplifier located intermediate the input and output locations for increasing the peak power of the pulse. 13, in a fiber optic telecommunication system including a non-electronic amplifier designed to amplify a light pulse signal while the signal retains the form of a light pulse throughout the amplification process, (e) λ ₀ is the length of the optical fiber. (f) the pulse is of such a nature that its pulse width is reduced as a result of amplification in a non-electronic amplifier, thereby substantially preventing merging of adjacent pulses; A fiber optic telecommunications device characterized in that: 2. A fiber optic telecommunications device according to claim 1, characterized in that the pulses are soliton pulses in at least a portion of the fiber transmission channel. 3. The apparatus of claim 1, wherein the non-electronic amplifier includes means for injecting a substantially continuous waveform of electromagnetic radiation, referred to as "pump" radiation, into the single mode optical fiber. A fiber optic remote communication device characterized by: 4. The apparatus of claim 3, wherein each of the non-electronic amplifiers comprises a glass fiber amplifier. 5. The device according to claim 4, wherein the glass fiber amplifier has h as Planck's constant,
Optical fiber telecommunications characterized in that it consists of fibers doped with ions having energy levels separated by energies substantially equal to h·c/λ ₀ , where c is the speed of light in vacuum. Device. 6. A fiber optic telecommunications device according to claim 3, 4 or 5, characterized in that the wavelength of the pump radiation is shorter than λ ₀ . 7. A fiber optic telecommunications device according to claim 1, characterized in that λ ₀ is at least 1.5 μm.