KR20000039037A - Optical amplifier for wavelength division optical transmission apparatus - Google Patents

Abstract

PURPOSE: An optical amplifier for a wavelength division optical transmission apparatus is provided to increase the ratio of signal to noise of optical link since an attenuator for attenuating input in a section having a low loss of optical fiber, is needless and input of optical amplifier is capable of being increased, thereby improving the efficiency of a system. CONSTITUTION: An input monitor part(100) having a first tap coupler(12), an optical splitter(13), a variable optical filter(14) and a function generator(15), is configured. A controller for a voltage controlled variable attenuator(200) having a plurality of photoelectric converters(PD1,PD2), a first comparing part(17), a counter(17), a first averager(19) and a second comparing part(20), is configured. A voltage controlled variable attenuating part having a first amplifier, a VCVA(Voltage Controlled Variable Attenuator), and a second amplifier, is configured. An output monitoring part(300) having a second tap coupler, and a photoelectric converter(PD3), is configured. An output controlling part(400) having a second averager(25) and a third comparing part(26), is configured.

Description

Optical amplifier for wavelength division optical transmission device

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical amplifier for use in a wavelength division optical transmission system, and more particularly, to an optical amplifier having a flattened channel-to-channel gain and a fixed output.

Conventional methods for flattening the gain of a wavelength division optical amplifier include a gain fixing method, a wavelength feedback method, an input compensation method, and the like. The gain-locking method maintains a constant gain by monitoring inputs and outputs, and has the advantage that the gain for the entire signal and the gain per channel are constant even if the input power or the number of channels changes.

The wavelength feedback method is a method in which a part of the output (amplified natural emission light) is fed back to the input to flatten the gain. This method is similar to the gain-fixing method in that the density inversion of the erbium ions in the amplifier is kept constant, but has the disadvantage that the gain of the output signal is relatively small as the feedback signal. Therefore, there is an economic disadvantage of using many pump lights to obtain the same amount of gain. In addition, since the area of the feedback signal cannot be used as a service channel, the channel capacity is reduced. The advantage of this wavelength feedback method is that the control speed of the amplifier is faster than other methods because the control of the amplifier is made of light and not through electronic circuits. The input compensation method detects the change in the input and adds or subtracts the amount of the signal to the amplifier through a separate channel. The main disadvantage is that the variable input range is small and it is uneconomical to have a separate channel source. In addition, there is a method of using a gain flattening filter, but the main purpose of this method is to increase the usable bandwidth, not gain fixing, and thus will not be mentioned in the present specification.

To describe the present invention, the prior art will be described in more detail.

First, the gain lock method consists of three main parts. The input monitor unit divides a part of the input signal and converts it into an electrical signal. Tap couplers function to split part of the input, while photoelectric converters convert the split input into an electrical signal. Here, the photoelectric conversion linearity is important because nonlinearities generated in the input monitor may appear as a gain error of the output since all subsequent control is performed by the linear control method.

The output monitor divides a part of the signal amplified by the optical amplifier 1 and converts the signal into an electrical signal. For this purpose, tap couplers and photoelectric converters are used as in the input monitor unit.

The gain calculation unit 2 functions to compare the difference between the gain obtained by comparing the input monitor unit and the output monitor unit with a predetermined gain (initial gain) and adjust the pump light intensity by the difference. The gain flatness of the optical amplifier depends on the properties of the Erbium Doped Fiber (EDF) itself, the length of the EDF used, and the accuracy of the initial gain setting. The initial gain for maximum flatness is determined by the length of the EDF. Therefore, at the design stage of the optical amplifier, there must be precise consideration of the desired gain, EDF length, and internal losses of the optical amplifier. If the intensity of the input signal is changed or the number of channels is changed, the output of the input monitor unit and the output monitor unit is changed. If the optical amplifier is within the fixed gain guarantee range, the ratio of the input / output change amount is constant. Therefore, the appropriate pump light intensity must be used to ensure a fixed gain.

The wavelength feedback optical amplifier includes a detector that detects some wavelengths in the amplified output, a feedback gain adjuster that adjusts the magnitude of the detected signal, and a wavelength multiplexer that adds the feedback signal back to the input. When the optical signal is amplified by an Erbium Doped Fiber Amplifier (EDFA), the amplifying unit 3 gains a gain and generates noise light at the same time. The noise light is evenly distributed in the signal region and the no signal region. In the wavelength feedback method, gain flattening is performed using the noise light generated in the no signal region. The noise light separated by the detector is adjusted to an optimum state by the feedback gain adjuster. The variable optical attenuator 4 is mainly used to adjust the feedback gain. The returned signal is added to the input by the directional wavelength combiner of the multiplexer. When the input of the amplifier changes, the signal magnitude of the feedback channel changes in opposition to the input change. Thus, increasing or decreasing the feedback gain in the same direction as the change in input can flatten the gain for the overall signal. Therefore, the size of the pump light does not need to be adjusted and always outputs a fixed intensity.

Finally, the input compensation method includes only the compensation control unit 6 which fixes the intensity of the pump light and compensates the amount of change in the input at the input of the amplifier. Initially, the amplifier is fixed to be optimized for maximum flat gain for the maximum input that the amplifier can receive, and when the input light intensity is reduced or the number of channels is reduced, the amplifier amplifies the changed amount of optical power through the compensation channel. Enter in (5). Therefore, the total input of the amplifier 5 including the amplifier compensation controller is always constant, and the amplifier 5 may have a constant density inversion state. Therefore, the gain of the signal can be fixed. Since the compensation controller should be within the amplifiable bandwidth of the amplifier 5, the signal bandwidth of the amplifier is reduced. The adjustment of the compensation signal must be very precise because it directly determines the characteristics of the amplifier and affects the performance of the entire optical transmission system.

However, the input compensation method is not generally used due to the economical disadvantage of having a separate source, and the noise figure of the amplifier is degraded by the loss of the wavelength multiplexer entering the amplifier input to couple the compensation signal with the input.

The wavelength feedback method requires that the internal gain of the amplifier be much larger than the gain actually needed because the signal being fed reduces the gain of the amplifier. Therefore, a lot of pump light intensity is required, and it is difficult to obtain a signal gain required for a wavelength division optical transmission system using a commercially available pump laser diode. In all three methods mentioned above, the average gain of the amplifier with maximum flatness is fixed regardless of the change in the input signal. Therefore, the amplitude of the amplifier output changes as the input signal changes. In a practically installed system, the distance between each amplifier is not constant, so the losses are not. Therefore, the input of each amplifier is not constant. Amplifiers for wavelength division transmission devices have a limited range of variation in the input signal to ensure gain flatness. Therefore, inputs beyond the limit must be forcibly reduced using a device such as an optical attenuator. Optical noise generated by amplifiers is a major reason for reducing the performance of optical transmission systems. The greater the input of the optical amplifier, the smaller the amount of optical noise.

Therefore, forcibly reducing the input of the optical amplifier is undesirable in terms of systems. Therefore, it is necessary to keep the output of the optical amplifier constant at a constant value. In the conventional method, it is impossible to fix the output power per channel while maintaining the flat gain.

Accordingly, the present invention has been made to solve the above-described problems, and to this end, the present invention uses an optical amplifier having a fixed channel output so that an abnormal loss or amplification occurring in the middle of a link is transmitted to the receiving end without being filtered. The stability of the system can be improved by providing a blocking function against abnormal changes in the stability of the device. In addition, an optical amplifier with a fixed output per channel, which has a much wider input range, does not require the use of attenuators to forcibly reduce the input in the low loss period of the optical fiber, and can increase the input of the optical amplifier. This can increase the overall system performance.

These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter.

In one embodiment of the present invention, the first tap coupler 12 for branching a part of the optical signal at a specific ratio, the optical splitter 13 for dividing the branched signal at a specific ratio, and the wavelength multiplexed optical signal are periodically The input monitor unit 100 including a variable optical filter 14 for scanning a specific wavelength and outputting a specific wavelength, a function generator 15 for generating a periodic voltage, and measuring the magnitude of a partly branched optical signal A plurality of first photoelectric converters PD1 and PD2, a first comparator 17 which emits a signal when the measured value is greater than or equal to a specific value, a counter 18 that counts the number of input signals, A variable attenuator controller including a first averager 19 capable of calculating an average value by dividing the total power by a number, and a second comparator 20 for outputting an adjustment signal by comparing whether the average value is different from a predetermined value 200 and the first tap coupler 1 A variable for maintaining a constant size by receiving the output of the first amplifier 16 and the output of the first amplifier 16 and the output of the second comparator 20 to amplify another optical signal branched from 2) A variable attenuator including an attenuator 21 and a second amplifier 22 for amplifying the output of the variable attenuator 21, a second tap coupler 23 for branching a portion of the wavelength multiplexed optical amplifier, and wavelength multiplexed An output monitor unit 300 including a photoelectric converter PD3 used to measure the size of the entire optical signal, and a wavelength multiplexed optical signal measured and divided by the number of wavelengths to calculate an average optical output per channel. And an output control unit 400 for maintaining a constant optical output per channel including a second averager 25 and a third comparison unit 26 for comparing the calculated average value with a specific value and measuring a difference. Wavelength It is characterized by an optical amplifier for an optical transmission device.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a view for schematically explaining three forms of a conventional gain flattening optical amplifier.

2 shows the structure of an optical amplifier according to the present invention;

3 is a view for explaining the principle of operation of the optical amplifier according to the present invention.

4 is a view for explaining an optical amplifier using a wavelength branch device according to an embodiment of the present invention.

<Description of the code | symbol about the principal part of drawings>

12, 23, 42, 49: Tap coupler 14: Wavelength variable filter

15: function generator 17, 20, 26: comparison

16, 22, 44, 48: Amplifier 21: VCVA (variable attenuator)

100, 101: Input monitor 200, 201: VCVA control

300, 301: output monitor 400, 401: output control unit

The present invention basically consists of two stages of optical amplifiers 16 and 22, and the first stage sufficiently increases the density inversion of ions in order to improve the noise figure. The inserted voltage controlled variable attenuator (VCVA) 21 in order to keep the magnitude of the output constant has a loss proportional to the input signal in function. Therefore, in order to efficiently compensate for the loss, the VCVA 21 calculates the number of channels and the light intensity per channel of the input optical signal, and the VCVA controller 200 measures and controls the light intensity for each wavelength. The output monitor 300 divides a part of the amplifier output to monitor the total light intensity. The output control unit 400 controls the light intensity per channel of the amplifier by controlling the pump light intensity of the two stage amplifier using the value provided by the output monitor unit 300.

The tap coupler 12 of the input monitor unit 100 separates only part of the input light to control. When multiple channels (wavelengths) are input, the tap coupler does not distinguish between them and separates the entire signal by the same ratio for each wavelength. The separated signal is sent through a 5: 5 splitter 13, one of which is a variable optical filter and the other of which is sent to the first averager 19. The variable optical filter has a characteristic that a wavelength to pass through changes according to an applied voltage. Therefore, the function generator 15 enters the input monitor unit 100 to operate the VCVA 21. The function generator 15 may use a suitable one among commercially available elements for generating a sine wave or a triangle wave according to the configuration of the peripheral circuit.

If the VCVA supply voltage changes periodically, the output also alternately outputs the input channel. This signal Pi_i is inputted to the first comparator 17 of the VCVA control unit 200 and compared with a value Pi_o which is previously present. Each time Pi_i becomes larger than Pi_o, the first comparator 17 emits a signal once and the counter 18 counts this number. This process depends on the cycle of the function generator 15. Each time the cycle is finished, the counter is initialized. The average block of the VCVA controller 200 calculates the average optical power per channel of the input signal by dividing the magnitude of the signal input through the 5: 5 splitter 13 by the value of the counter. This calculation process is also possible with analog electronic circuits, but it is easier to adopt digital control circuits. The calculated average value is used to control the VCVA 21. The VCVA 21 has a characteristic that the amount of attenuation changes according to the magnitude of the applied voltage. Therefore, when the average power per channel is larger than the reference input, the amount of attenuation can be given to make the output light intensity of the first stage constant. At this time, when the pump light of the first stage is not large enough, the gain (gain spectrum) according to the wavelength also changes when the intensity of the input light changes. If the gain spectrum of the first stage changes, even if the output power of the first stage can be kept constant, the gain of the entire amplifier cannot be made flat. Therefore, the first stage first pump laser diode 27 should be capable of generating sufficient intensity in consideration of the input signal. When properly controlled, the output of the 1st stage will have a fixed per channel regardless of the size of the input. Since the gain leveling is not performed in the first stage, the output of each channel is different from each other. In addition, the density inversion of the first stage is a very large value close to 1, so the intensity difference for each channel of the output is constant when the input size is the same.

A band pass filter should be installed at the output of the first stage amplifier to block the noise light generated during the amplification process. This is to prevent the noise light from saturating the second stage. Through this process, the first stage output amplified and adjusted by the VCVA 21 is input to the second stage of the amplifier. The output monitor 300 at the rear of the second stage divides a part of the output signal to monitor the optical power of the divided whole signals. The output control unit 400 calculates the light intensity per channel of the output signal by using the number of channels sent from the VCVA control unit 200 and the value from the output monitor unit 300, and compares it with an initial value having the same value as that difference. The control signal is sent to the second pump laser diode 28 to maintain the optical power per channel of the output at a constant value. At this time, since the spectrum of the signal coming into the second stage is constant and has a fixed value for each channel, the gain flattening of the entire signal is automatically performed when the optical power per channel of the output is fixed at a specific value.

Fig. 3 shows the operation of the amplifier for the case where a small signal is input and a large signal is input. First, when a relatively small signal is input, since the output of the first stage 16 passes through the high density inversion region, the short wavelength signal gains more than the long wavelength signal. Thus, the shape of the output is the same as b in FIG. Since the input is small, the VCVA 21 is adjusted to give a small value of attenuation, and the signal shape after passing the VCVA 21 becomes as shown in FIG. In the second stage 22, since the light intensity per channel is made constant and gain flattening is performed at the same time, the overall amplifier output becomes as shown in FIG. Next, when a large value signal is input (e of FIG. 3), the input becomes large, but the density of the first stage 16 does not change much because the pump light of the first stage 16 is very strong. Therefore, the gain spectrum of the output is almost the same as when a small signal is input. Therefore, the same channel-specific gain as when a small signal is input is obtained. The output of the first stage 16 is shaped as shown in FIG. Since the change in density inversion of the first stage 16 is small, the output of the first stage 16 becomes larger as the input becomes larger. When the input monitor unit 100 and the VCVA control unit 200 compensate for this difference, the signal shape after passing the VCVA 21 is the same when the input is small. Therefore, the same process can be used for gain flattening and light intensity adjustment per channel.

4 is a configuration diagram implemented using a wavelength branching device according to the present invention. In this case, the output monitor 301 and the output control unit 401 do not change. In the input monitor 101, a wavelength diverter 43 enters in place of the wavelength variable filter. The wavelength branching device 43 functions to separate the wavelength multiplexed signal for each wavelength. Each separate channel is converted into an electrical signal through a plurality of photoelectric converters. This signal is also used to count the number of channels in the VCVA control unit 201 and calculates the average light intensity of the channel using this. The advantage of this method is that wavelength division multiplexer 43, which is used to monitor the input per channel of each amplifier in the wavelength division multiplexed optical transmission device, can be implemented without adding a separate device.

The first and second stage optical amplifiers 44 and 48 used in the present invention may be composed of components that are currently commercially available. Importantly, the output of the pump laser diode 53 of the first stage must be large enough. If the number of channels is large or the input is large, the pump light must be as large. Currently, 150mW or more laser diodes are commercially available. The tunable filter uses a Fabry parot, which is implemented by using a signal range in which a passable bandwidth is used and a search speed as fast as possible. The most important part of the VCVA 47 is a conventional motor, but the field-applied Mach-Zehnder type is suitable for the stability of the communication equipment. In addition, it is advantageous to use a flattening gain by using a uniform wavelength loss. Other electronic control circuits use well-known PID control circuits and photoelectric converter bias circuits.

The greatest advantage of the present invention is that the gain flattening of the optical amplifier for the wavelength division optical transmission device and the fixing of the channel output can be performed simultaneously. In a wavelength division transmission scheme, channel gain flattening is a basic technique, and conventional optical amplifiers use a method of fixing gain for this purpose. Therefore, if the input of the amplifier changes, the output also has a disadvantage. Optical links made with fixed-gain optical amplifiers are less reliable because they do not filter out abnormal losses or amplifications occurring in the middle of the link. Optical amplifiers with fixed channel outputs provide protection against these variations, thus improving system stability. Fixed-gain optical amplifiers also have a constant range to ensure gain flatness, while optical amplifiers with fixed outputs per channel have a much wider input range.

Therefore, it is not necessary to use an attenuator to forcibly reduce the input in the area where the loss of optical fiber is small, and the input of the optical amplifier can be increased, so that the signal-to-noise ratio of the entire optical link can be increased, thereby improving the overall system performance.

Claims (5)

A first tap coupler for dividing a part of the optical signal received at a specific ratio, an optical splitter for dividing the branched optical signal at a specific ratio, and a variable optical filter for periodically scanning the divided optical signal to output a specific wavelength An input monitor unit including a function generator for generating a periodic voltage; A plurality of first photoelectric converters used to measure the magnitude of the optical signal partially branched from the optical splitter and the variable optical filter, and a first comparator for outputting a specific signal when the measured value is greater than or equal to a specific value And a counter for counting the number of specific input signals, a first average for calculating an average value by dividing the total power by the number, and outputting a control signal by comparing whether the average value is different from a predetermined value. A variable attenuator controller including two comparison units, A variable amplifier and an output of the variable attenuator for maintaining a constant magnitude by receiving an output of a first amplifier, an output of the first amplifier, and an output of the second comparator that amplify another optical signal branched from the first tap coupler A variable attenuator comprising a second amplifier for amplifying a signal; An output monitor unit including a second tap coupler for dividing a part of the optical signals output from the second amplifier and a second photoelectric converter used to measure the magnitude of the entire wavelength-multiplexed optical signal; A second average unit for measuring the branched optical signal and dividing it by the number of wavelengths to calculate an average optical output per channel, and a third comparison unit for comparing the calculated average value with a specific value and measuring a difference. And an output control unit for maintaining the optical amplifier for wavelength division optical transmission device. The optical amplifier of claim 1, wherein the variable attenuator further includes a bandpass filter for blocking noise light generated during an amplification process at an output of the first amplifier. According to claim 1, The output using the number of channels sent from the variable attenuator control unit and the value output from the output monitor unit to calculate the light intensity per channel of the output signal and to compare the calculated value and the initial value and output the output And a pump laser diode connected to the control unit to maintain the optical power per channel of the output at a constant value using the output signal. The optical amplifier of claim 1, wherein the input monitor unit comprises an optical wavelength branching device for splitting optical power for each channel, and a plurality of photoelectric converters for measuring the optical power for each branched wavelength. . 5. The optical amplifier of claim 4, wherein the variable attenuator control unit includes a counter for counting the number of channels of signals passing through the plurality of photoelectric converters and an averager for calculating average power per channel.