JPH09153884A - Wavelength control method and optical communication system using the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate a chi-interval wavelength dependency without a new additional circuit by calculating the increased portion of a control variable corresponding to a ch-interval from a control variable value whenever the transmissive wavelength of a light transmitting means is swept. SOLUTION: A wavelength control circuit properly changes the increased portion ΔV of the control signal V corresponding to the ch-interval by the value of the control signal V so that the wavelength reliability of the ch-interval is eliminated. That is, the ΔV value is changed to be ΔV=ΔV1 when the control variable V is V1, ΔV=ΔV2 when it's V=v2 and ΔV=ΔV3 when it's V=V3 so that the corresponding ch-intervals Δλ1, Δλ2 and Δλ3 are kept to the equal value Δλch. In this case, the wavelength control circuit calculates the increased portion ΔV of the control signal corresponding to the ch-interval from the control signal by the use of a previously stored function. The wavelength control circuit actually grasps the position relation of respective kinds of transmission wavelength through the use of the control signal V to be transmitted to a filter drive circuit and controls the transmission wavelength of a self-station and the transmitted wavelength through a wavelength variable filter.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention uses a control node of a wavelength division multiplexing communication system, an optical transmitter incorporated in the communication node, a wavelength control method for controlling a wavelength in a transmission line of the wavelength division multiplexing communication system, and the method. The present invention relates to an optical communication system.

[0002]

2. Description of the Related Art A wavelength division multiplexing communication system is a kind of optical communication system and has a large number of independent channels (wavelengths) in one transmission line, and can transmit independent information to each channel. Since it does not require multiplexing on the time axis, it is not necessary to match the transmission rate of each channel, and it is also suitable for multimedia communication that requires network flexibility. As an example of the wavelength division multiplexing communication system, there is a passive star type network system in which each terminal station has a set of variable wavelength optical transmitters and optical receivers.

FIG. 5 shows the configuration of the wavelength division multiplexing communication system. The terminal stations 151 to 15n are communication nodes 161 to 16n.
n and the optical fibers 171 to 17n and 181 to 18n to be connected to the star coupler 14 to form a network. The optical signals sent from the communication nodes 161 to 16n are distributed by the star coupler 14 to all communication nodes including the own node. The distributed optical signal is received by the terminal station of each terminal device, it is determined whether the received optical signal is addressed to the own station, and if the optical signal is addressed to the own station, the optical signal is demodulated. Get that information.

Here, in order to increase the number of channels in the wavelength division multiplex communication system, when the wavelength interval between channels (hereinafter referred to as ch interval) is set to the order of 10 GHz (about 0.08 nm in wavelength conversion), interference is prevented. Needs to control the transmission wavelength of each communication node. Several methods have been proposed as methods for keeping the ch intervals of transmission wavelengths from distributedly arranged communication nodes constant. Among them, the method according to Japanese Patent Application No. 6-296660, which is being applied for by us, is easy to perform initial setting and resetting and easy to control.

According to this method, in the wavelength division multiplexing communication network as shown in FIG. 5, each communication node detects the ch interval between the wavelength of its own station and the wavelength of the channel adjacent to the long wavelength side on the wavelength axis, and ch The wavelength of its own station is controlled so that the interval becomes constant.

FIG. 6 shows an example of the operation of this method. First, at the start of light emission from the local station to the other station,
The presence of wavelengths on the transmission line is detected (for example, ch1 to ch
4) emits light to the short wavelength side of the wavelength range that the local station can emit (ch5 in FIG. 6a) and gradually shifts to the long wavelength side, and the shortest wavelength of the existing optical signal and the wavelength of a predetermined ch interval. Then, the wavelength interval is maintained and transmission is performed (FIG. 6b).
Also, when there is a communication node whose transmission has ended and light emission has stopped during the transmission of its own station (ch3 in FIG. 6c), the communication node emitting light on the short wavelength side is the wavelength (ch4, ch5).
Is gradually shifted to the long wavelength side (FIG. 6c). As a result, in the steady state, the wavelengths of the respective communication nodes are arranged at equal ch intervals from the long wavelength end of the wavelength range of the wavelength division multiplexing communication system (FIGS. 6b and 6d).

Further, FIG. 7 shows a configuration diagram of the communication nodes 161 to 16n of the wavelength division multiplexing communication network as shown in FIG. The communication node 16 includes an optical splitter 19 and an optical receiver 2.
0, an optical transmitter 21.

The optical splitter 19 splits an optical signal from the transmission line into two, one of which is an optical receiver 20 and the other of which is an optical transmitter 21.
Send to. The optical receiver 20 searches for a channel transmitting a signal addressed to itself from a large number of channels of the wavelength division multiplexing communication system. After finding the wavelength, the wavelength of the built-in tunable filter is matched with the wavelength, and the reception is continued while following the wavelength. The optical transmitter 21 detects the transmission wavelength of another station on the transmission path by the optical signal from the optical branching device 19, detects the transmittable wavelength of its own station, and converts the transmission signal from the terminal to electric / optical. It is sent to the optical fiber of the transmission line.

FIG. 8 shows a block diagram of the above-mentioned optical transmitter 21. The optical transmitter 21 includes a wavelength control circuit 1, a light source drive circuit 2, a wavelength variable light source 3, an optical modulator 4, an optical branching device 5, an optical switch 6, an optical multiplexer 7, a wavelength variable filter 8, a light receiving element 9, and an amplifier. 10, a discriminator 11, a filter drive circuit 12,
Consists of

Here, as an example of the variable wavelength light source 3,
IEICE Technical Report OQE (Optical and Quan)
tum Electronics) '89 -116 "Three-electrode long resonator λ / 4 shift MQW-DFB laser". As an example of the wavelength tunable filter 8, there is a Fabry-Perot resonator type filter.
C (Europian Conference on Optical Communicatio
n) '90 -605, "A field-worthy, high-performance, tun
The one described in “able fiber Fabry-Perot filter” ”can be used.

In such an optical transmitter, the wavelength control system 1 outputs a control signal for changing the optical filter control voltage of the wavelength tunable filter 8 and sweeps the transmission wavelength of the wavelength tunable filter 8 via the filter drive circuit 12. . Further, the wavelength interval between the transmission wavelength of the own station and the adjacent channel on the long wavelength side of the own station is detected from the output of the discriminator 11 based on the output of the wavelength tunable filter 8 during the sweep. Further, the wavelength control system 1 outputs a control signal via the light source drive circuit 2 based on the result of the wavelength interval, controls the wavelength of the wavelength tunable light source 3, and sets the transmission wavelength of the own station to the long wavelength side. The wavelength is controlled until it becomes a predetermined wavelength interval with the adjacent station, and then the wavelength is continuously maintained and controlled.

[0012]

However, the conventional wavelength control method has the following problems.

In the conventional wavelength control method, the control for maintaining the ch interval is performed by determining the difference between the control signal values corresponding to the transmission wavelength of the own station and the transmission wavelength of another station adjacent to the own station on the wavelength axis. I went by keeping it at a value. However, the value of the control signal to the wavelength tunable filter in the local station and the value of the transmission wavelength of the wavelength tunable filter are not necessarily in a linear relationship. For this reason, the actual channel spacing fluctuates depending on the wavelength, and the multiplexing is performed in consideration of the margin of the fluctuation, resulting in a decrease in the multiplicity.

The above state will be described with reference to FIGS. 3 and 4. FIG. 3 shows the filter drive circuit 12 of the wavelength control circuit 1.
Of the control signal V and the transmission wavelength λ of the wavelength tunable filter 8. The horizontal axis represents the control signal V and the vertical axis represents the transmission wavelength λ. In FIG. 3, in order to explain the problem of the relationship between the control voltage and the transmission wavelength, a part of this characteristic is represented by remarkable nonlinearity.

FIG. 4 also shows the relationship between the control signal V and the transmission wavelength λ. The horizontal axis of FIG. 4A is the control signal potential V, and the horizontal axis of FIG. 4B is the transmission wavelength λ of the wavelength tunable filter 8. Each of the vertical bars represents the transmission signal on the V axis of the control signal or on the λ axis of the transmission wavelength of the wavelength tunable filter 8.

The control signal V to the filter drive circuit 12 is
It is converted into a voltage signal by the D / A converter built in the filter drive circuit 12. This voltage signal is applied to the wavelength tunable filter 8, and as a result, the transmission wavelength λ changes. The relationship λ = g (V) between the control signal V and the transmission wavelength λ is generally deviated from the proportional relationship as shown in FIG. This is due to the nonlinearity of the piezo element used inside the wavelength tunable filter 8 and the structure of the wavelength tunable filter 8.

In such a case, the control signal V is set to the same value ΔV.
The increment Δλ of the transmission wavelength λ when it is changed depending on the value of the control signal V varies. As shown in FIG. 3, the transmission wavelength increments Δλ1 and Δλ2 when the control signals V1 and V2 are equally changed by ΔV have different values.

Therefore, when the transmission wavelengths of the respective stations are controlled so that the transmission signals of the respective stations have equal intervals ΔV on the axis of the control signal V, as shown in FIG.
1, Δλ2 ...) Are different from each other. That is, since the wavelength width of wavelength detection differs depending on the wavelength,
The h-interval changes depending on the wavelength. On the other hand, in order to prevent interference, it is necessary that the ch interval is a predetermined wavelength width Δλch or more, and the control variable interval ΔV is set so that the ch interval is never less than the wavelength width Δλch. For this reason,
There is a wavelength range in which the ch interval is larger than the wavelength width Δλch,
As a result, there is a problem that the number of usable channels is reduced.

A first object of the present invention is to numerically correct the non-linearity between the control variable to the filter driving circuit and the transmission wavelength of the filter, thereby eliminating the need for a new additional circuit. The purpose is to eliminate the wavelength dependence of the interval and increase the number of channels that can be used by the system. Further, a second object of the present invention is to provide a simple method for numerically correcting the nonlinearity of the control variable and the transmission wavelength.

Furthermore, a third object of the present invention is to provide another simple method for numerically correcting the nonlinearity of the control variable and the transmission wavelength.

[0021]

In order to achieve the above object, a first aspect of the present invention is to provide a light transmitting means which is connected to a transmission line and which can change a transmission wavelength, and the optical signal based on an input signal. A light having a drive circuit for generating a drive voltage for the transmission means, and a wavelength control system for controlling the transmission wavelength of the light transmission means by outputting the value of the control variable to the drive circuit and changing the value of the control variable. In the communication device, the wavelength control system is configured to calculate an increment of the control variable (hereinafter, an increment of the control variable) for changing the transmission wavelength of the light transmitting unit by a predetermined wavelength difference from the control variable. .

In the above structure, the wavelength control system sets the channel interval of the wavelength division multiplexing communication to a predetermined wavelength difference. The increment of the control variable corresponding to the ch interval depends on the value of the control variable. The wavelength control system calculates the increment of the control variable corresponding to the ch interval from the value of the control variable each time the transmission wavelength of the light transmitting means is swept. As a result, it is possible to eliminate the wavelength dependence of the ch interval without adding a new circuit.

In the second aspect of the present invention, the wavelength control system stores in advance a relational expression between the control variable and the increment of the control variable, and the relational expression is used to select the control variable from any of the control variables. It is characterized by calculating the increment of the control variable.

In the above structure, the wavelength control system obtains the increment of the control variable corresponding to the ch interval by substituting the value of the control variable for any control variable into the relational expression.

According to a third aspect of the invention, the wavelength control system stores in advance a plurality of sets of numerical values each including the control variable and an increment of the control variable in the control variable, and refers to the plurality of numerical values. It is characterized in that the increment of the control variable is calculated from the control variable.

In the above structure, the wavelength control system selects the increment of the control variable corresponding to the ch interval from the set of a plurality of prestored numerical values, and the value of the control variable is the closest to the current value of the control variable. We obtain one by choosing one and referencing the increment of the control variable in the chosen set of numbers.

Further, a fourth invention comprises a light transmitting means for transmitting a predetermined wavelength variable wavelength in an optical signal, and a wavelength controlling means capable of sweeping a transmission wavelength of the light transmitting means by a control signal. The wavelength control means is a wavelength control method for calculating the value of the increment of the control signal (hereinafter referred to as the increment of the control signal) from the control signal, the transmission wavelength and the transmission wavelength of the light transmission means. At least two or more sets of control signal values corresponding to the above are measured, wavelengths (hereinafter, referred to as target wavelengths) that are apart from the corresponding transmission wavelengths by a predetermined wavelength interval are obtained for the values of the control signals, and each of the above-mentioned respective The value of the control signal corresponding to the target wavelength (hereinafter referred to as the target control signal value) is obtained by approximation from the at least two sets of the transmission wavelength and the value of the control signal, and the value of the control signal and its target control signal Obtains an incremental value for each control signal mentioned above determined by taking the difference of values, and controlling the transmission wavelength of the light transmitting means based on the value of the calculated increment.

In such a method, the wavelength interval between the measured transmission wavelength and its target wavelength is equal to the channel interval at the time of communication. On the other hand, it is difficult to directly obtain the increment of the value of the control signal corresponding to the channel interval from a certain transmission wavelength. Therefore, for the measured transmission wavelength, the target wavelength and then the target control signal value are obtained, and finally the difference between the target control signal value and the value of the control signal is taken to obtain the incremental value stepwise. At that time, even if there is a nonlinear characteristic between the wavelength tunable filter control voltage and the transmission wavelength by using a predetermined function approximation or a mathematical table obtained in advance, the adjacent wavelength interval is over the entire transmission wavelength band on the transmission line. Therefore, it can be maintained at a constant interval.

[0029]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below in detail with reference to the drawings together with each embodiment.

[First Embodiment] A first embodiment of the present invention will be described in detail below with reference to the drawings. The configurations of the wavelength division multiplexing communication system, the optical node, and the optical transmitter in the present embodiment are the same as those described with reference to FIG. 5, FIG. 7, and FIG.

FIG. 1 is a diagram for explaining the wavelength control method of the present invention, and shows the relationship between the control signal V by the drive circuit for the optical filter having a variable transmission wavelength and the transmission wavelength λf of the optical filter. is there. The horizontal axis of FIG. 1a is the control signal V, and FIG.
The horizontal axis of b represents the transmission wavelength λf of the wavelength tunable filter 8, and each of the vertical bars represents the transmission signal on the control signal V axis or the transmission wavelength λf axis of the wavelength tunable filter.

In the present embodiment, the wavelength control circuit 1 appropriately changes the increment ΔV of the control signal V corresponding to the ch interval according to the value of the control signal V to eliminate the wavelength dependence of the ch interval. As shown in FIG. 1, when the control variable V = V1, the value of ΔV is ΔV = ΔV1, and when V = V2, ΔV = ΔV2 and V = V3.
Then, by changing ΔV = ΔV3 ...
The h intervals Δλ1, Δλ2, Δλ3 ... Are kept at equal values Δλch.

In this embodiment, the wavelength control circuit 1 calculates the increment ΔV of the control signal corresponding to the ch interval from the control signal V using the function f stored in advance.

This function f is obtained as follows. First of all, the control variable V for the wavelength tunable filter 8 is measured.
And the transmission wavelength λf, λf = g (V) is obtained. Change in transmission wavelength Δλ when the control variable changes from V to V + ΔV
f is given by Δλf = g (V + ΔV) −g (V). Therefore, for a certain control variable V, the difference ΔV in control signal such that the change Δλf in transmission wavelength is equal to the channel interval Δλch is obtained as ΔV = h (V) = g ⁻¹ (g (V) + Δλ) · V To be Here, g ⁻¹ is an inverse function of the function g. The function f is obtained by converting this function h into f (V) = a ₀ + a ₁ V ¹ + A ₂ V ² + A ₃ V ³ It is an approximation such as +.

The wavelength control circuit 1 controls the wavelength as follows using the above function f. As an example, in a system that starts transmission from the short wavelength side and changes the transmission wavelength to the long wavelength side, the local station transmits at wavelength λ1, and the other station adjacent to the local station at the long wavelength side transmits at wavelength λ2, A case where the own station controls the transmission wavelength λ1 of the own station so that the wavelength difference (λ2−λ1) from the other station matches the channel interval Δλch will be described.

FIG. 2 is a diagram for explaining the sweep operation of this embodiment. FIG. 2a shows the relationship between the transmission wavelengths of the own station and another station adjacent to the own station on the long wavelength side on the wavelength axis. The wavelength control circuit 1 knows the transmission wavelength λ1 of its own station from the result of the previous sweep, and first sets the transmission wavelength λf of the wavelength tunable filter 8 to λf = λ1−Δλmar. Δλ
mar is a sweep margin for surely detecting the transmission wavelength. Then, the transmission wavelength λf is set to λf = Δλch + Δλma
While sweeping to r, the sweep wavelength of another station is detected. Δ
λch is a channel interval. Here, the transmission wavelength of another station is detected at λf = λ2. The wavelength control circuit 1 compares the transmission wavelength λ2 of the other station with the magnitude of λ1 + Δλch, and changes the transmission wavelength of its own station in the direction in which the difference [λ2− (λ1 + Δλch)] between the two decreases.

The wavelength control circuit 1 actually grasps the positional relationship of each transmission wavelength by using the control signal V sent to the filter drive circuit 12, and controls the transmission wavelength of its own station and the transmission wavelength of the wavelength tunable filter. .

FIG. 2B shows the relationship between the transmission wavelengths of the own station and other stations adjacent to the own station on the long wavelength side on the V axis of the control variable of the wavelength control circuit 1. The wavelength control circuit 1 detects the transmission wavelength λ1 of its own station (the discriminator 11 outputs H).
The control variable V1 at that time is grasped. The wavelength control circuit 1 is
The control variable V1−ΔVmar is sent to the filter drive circuit 12 to set the transmission wavelength λf of the wavelength tunable filter to λf−λmar.
Set to. ΔVmar is a sweep margin on the V axis of the control variable, and corresponds to Δλmar in wavelength. Then, ΔV1 = f (V1) is calculated from the value of V1 and stored. ΔV1 is the increment of the control variable corresponding to the ch interval Δλch on the wavelength axis in the control variable V1 (corresponding to λ1 on the wavelength axis). Then, the wavelength control circuit 1 detects the transmission signal of another station while sweeping the control variable V to V = V1 + f (V1) + ΔVmar. Here, V = V2
The transmission wavelength of another station is detected at (corresponding to λ2 on the wavelength axis). The wavelength control circuit 1 has V2 and V1 + f
The magnitude of (V1) is compared, and the value of the control variable V is changed so that the difference between the two decreases.

The optical transmitter of each communication node stores the function f matched to the characteristic of each optical filter and is provided with the wavelength control system for performing the above operation, so that the transmission wavelengths on the transmission line are arranged at a constant channel. It is kept at the interval Δλch.

[Second Embodiment] The configurations of the wavelength division multiplex communication system, the communication node and the optical transmitter of the present embodiment are the same as those of the first embodiment. The difference from the first embodiment is that the wavelength control circuit 1 calculates the increment ΔV of the control signal corresponding to the ch interval for an arbitrary control signal V using a previously stored numerical table.

The wavelength control circuit 1 stores a plurality of sets (V, ΔV) of numerical values each consisting of the control variable V and the increment ΔV of the control variable corresponding to the ch interval. These sets of numbers are constructed so that from each stored value of the control variable V, 1) the corresponding value of ΔV, 2) the next largest and smallest value of V, respectively, can be obtained. .

This is, for example, the following two-dimensional array A
ij, (i = 1, 2, 3 ..., j = 0, 1). Among the plurality of sets of numerical values described above, the control variables V arranged in ascending order are Vi (i = 1, 2, 3 ...)
Let ΔV corresponding to i be ΔVi. Each element of the array Aij is set as Ai0 = Vi, Ai1 = ΔVi (i = 1, 2, 3 ...).

The values of Vi and ΔVi to be stored are determined as follows. First, the relation λf = g (V) between the control variable V and the transmission wavelength λf is obtained by measurement, and then the relation between the control variable V and the increment ΔV of the control signal such that the change Δλf of the transmission wavelength becomes equal to the ch interval Δλch. ΔV = h (V)
Ask for.

The values of the control variable V that the wavelength control circuit 1 may use when sweeping the filter transmission wavelength are set to V1, V2, V3, ... In ascending order. For example, the wavelength range used by the system is λmin to λmax, and the wavelength range λm is
The range of the control variable V corresponding to in to λmax is Vmin
.About.Vmax, and assuming that the wavelength control circuit 1 changes the control variable V in increments of .DELTA.Vs when sweeping the filter transmission wavelength, Vi becomes Vi = Vmin + (i-1) .DELTA.Vs. From these values of Vi, ΔVi is set to ΔVi = h (Vi).

Using the above array Aij, the wavelength control circuit 1 obtains the increment ΔV of the control signal corresponding to the ch interval from the arbitrary control signal V by the following procedure.

First, while changing the subscript i, Vi (= Ai
0) and the value of V are compared, and the value of i that satisfies Vi ≦ V <Vi0 is obtained. The value of ΔV can be obtained from this value of i and the array Aij as ΔV = Ai1.

[Third Embodiment] This embodiment relates to a mode different from the above embodiments. In the second embodiment, the increment Δ of the control variable corresponding to the control variable V and the ch interval.
A set of numerical values (V, ΔV) composed of V and an ascending index (index) i, and control variables V and ΔV 2
It is stored as a dimension array Aij. However, a configuration is used in which two addresses (hereinafter referred to as pointers) to a set of numerical values and a set of numerical values (hereinafter referred to as a structure) having four values of control variables V and ΔV as elements are used. Is also possible.

This structure has the i-th smallest control variable V
And a control variable increment ΔVi corresponding to Vi,
A pointer to a structure including the control variable Vi1 (hereinafter, referred to as a forward pointer) and a pointer to a structure including the control variable Vi + 1 (hereinafter, referred to as a backward pointer).
Consists of However, the pointer to the front of the structure including the control variable V1 has the address of its own structure as a value. Similarly, the pointer to the back of the structure containing the largest control variable Vn also has the address to its own structure as a value.

The wavelength control circuit 1 obtains the control variable increment ΔVi corresponding to the control variable Vi in ascending or descending order with respect to Vi, by sequentially referring to pointers to the front and back of each structure.

Further, in the first and second embodiments, the sweep margin Δλmar for surely detecting the transmission wavelength.
The increment ΔVmar (hereinafter, referred to as the sweep margin of the control variable V) of the control variable V corresponding to the above is constant regardless of the transmission wavelength λf (or the control voltage V). This control variable V
It is also possible to make the sweep margin ΔVmar variable.

Actually, 1) A function e (V) which gives an appropriate ΔVmar at an arbitrary control variable V is prepared by the same method as in the first embodiment. 2) A numerical table giving an appropriate ΔVmar for an arbitrary control variable V is prepared in the same manner as in the second embodiment. 3) In the first and second embodiments, an appropriate coefficient m for the value of ΔV obtained from the control variable V
ΔVmar is calculated by a method such as multiplying by.

[Fourth Embodiment] The fourth embodiment will be described with reference to FIG. 1 is different from that described in the first embodiment, it is operated as follows. The specific configuration of this embodiment is the same as that described with reference to FIGS. 5, 7, and 8, such as a wavelength division multiplexing communication system, an optical node, an optical transmitter, and an optical receiver, and a description thereof will be omitted. To do.

The wavelength control circuit 1 stores a relational expression ΔV = f (V) between the control signal increment ΔV corresponding to the channel interval and the control signal V, and increments from the control signal V by ΔV as necessary.
Is calculated. This function f is obtained in advance by measurement and calculation.

The wavelength control circuit 1 may store a numerical table of the control signal V and the increment ΔV instead of the above function f. The numerical table is a set of values (Vi, ΔVi) i of the control signal V and the increment ΔV obtained by the previous measurement and calculation.
= 1, 2, 3, ..., Each numerical value is arranged in ascending or descending order with respect to the control signal V.

When arranged in ascending order, the increment ΔV of the control signal V corresponding to the channel interval in the arbitrary control signal V
Is compared with the values of Vi and V while changing the subscript 1, and is obtained as an increment ΔVi from the value of i that satisfies Vi ≦ V <Vi + 1. In the case of arranging in descending order, the directions of the signs may be reversed.

The wavelength control circuit 1 controls the wavelength as follows using the above-mentioned function f to the mathematical table. As an example, in a system that starts transmission from the short wavelength side and changes the transmission wavelength to the long wavelength side, the local station transmits at wavelength λ1 and the other station adjacent to the local station at the long wavelength side transmits at wavelength λ2. A case will be described in which the own station controls its own transmission wavelength λ1 so that the wavelength difference λ2−λ1 from another station matches the channel interval Δλch.

FIG. 2 is a drawing explained in the first embodiment, and the reference numerals and the horizontal axis are the same as above. The wavelength control circuit 1 determines the transmission wavelength λ of its own station where the discriminator 11 outputs H.
The control variable V1 when 1 is detected is known. The wavelength control circuit 1 sends the control variable V1-Δ to the filter drive circuit 12.
Vmar is sent to set the transmission wavelength λf of the wavelength tunable filter 8 to λf−Δλmar. ΔVmar is a sweep margin on the V axis of the control variable in FIG. 2B, and a sweep margin Δλmar for surely detecting the transmission wavelength on the wavelength axis.
It corresponds to. In addition, from the value of the control variable V1 to the function f
.DELTA.V1 is calculated using the numerical table and stored. ΔV1 is the increment of the control variable in the control variable V1, and corresponds to the channel interval Δλch on the wavelength axis. Then, the wavelength control circuit 1 sets the control variable V to V = V1 + ΔV1 + ΔVma
While sweeping to r, the transmission signal of another station is detected. Here, the transmission wavelength of another station is detected at V = V2. The wavelength control circuit 1 compares the magnitudes of V2 and V1 + ΔV1 and changes the value of the control variable V so that the difference between them is reduced.

The case of using a polynomial for the function f will be described. In this polynomial, the order of the wavelength λ to be used is determined by the shape of the function and the required accuracy. As an example, consider the case where g (λ) is approximated by the third order of λ as follows.

G (λ) = a ₀ + a ₁ λ + a ₂ λ ² + a ₃ λ ³ measurement point (Vi, λi) i = 1, 2, 3, ...
To obtain (λ), a so-called normal equation, specifically, a quaternary simultaneous linear equation for coefficients a _{0 to} a ₃ ,

[0060]

(Equation 1) You just need to solve Here, it is possible to obtain the function g (λ) by substituting the numerical values of the measurement results for Vi and λi to obtain the respective coefficients. Note that at least m + 1 measurement points are necessary when approximating with a polynomial of degree m.

The optical transmitters of the respective optical nodes 161 to 16n store the function f to the numerical table corresponding to the characteristics of the respective optical filters and are provided with the wavelength control circuit 1 for performing the above operation, so that the transmission on the transmission line is performed. The wavelength arrangement is kept at a constant channel spacing Δλch.

[Fifth Embodiment] In this embodiment, an example of a method for obtaining the function f shown in the fourth embodiment will be described. FIG.
4 is a graph showing the relationship between the control signal V to the filter drive circuit 12 and the transmission wavelength λ of the wavelength tunable filter 8. The horizontal axis represents the control signal V, the vertical axis represents the transmission wavelength λ, the solid line represents the relationship between the two, and the black dots represent the measurement points (Vi, λi) i = 1, 2, 3,
It represents ... The following procedure is performed in order to obtain the function f which is an approximate function. First, the control signal V and the transmission wavelength λ
To measure the relationship. At this time, the wavelength λ is measured at least over the wavelength range used by the system. Hereinafter, the measurement result is expressed as (Vi, λi) i = 1, 2, 3, .... Note that Vi is arranged in ascending order. However, the order may be descending, but it will be simplified for simplification of the description.

Next, for each measured transmission wavelength λi, the value of the control signal V at λi + Δλch (hereinafter referred to as the target control signal value) V′i is obtained. In many cases, the number of measurement points cannot be taken so that all the values of V'i can be directly obtained. Therefore, each measured (Vi, λi)
The target control signal value V'i is calculated from the value of V by some method.

For example, (Vi, λi) i = 1, 2, 3,
There is a method of connecting each point of ... With a straight line and approximating the relationship between V and λ with a broken line. To obtain V'i from Vi, first find j such that λj ≤ λi + Δλch <λj + 1, and then calculate V'i as V'i = Vj + [(Vj + 1-Vj) / (λj + 1- λj)] ×
(Λj + 1-λi-Δλch).

At the measurement points, the transmission wavelengths λ are evenly spaced (λi + 1-λ
i = const. ), Various constant difference interpolation formulas can be used in the calculation of V'i. For the official form and how to use it, see [[Numerical calculation] Takashi Akasaka, Corona Publishing Co., Ltd.], for example, and refer to it to calculate an appropriate value.

When the intervals of the transmission wavelengths λ are not equal, or when the points are scattered due to measurement error and the approximation by the broken line or the interpolation is not appropriate, the approximation formula V = g (λ) is used by using the least squares method. Ask for.

From this, V'i is determined as V'i = g (λi + Δλch). Further, for each control signal Vi, the increment of the control signal (hereinafter, referred to as the increment of the control signal) ΔVi corresponding to the channel interval Δλch is obtained as ΔVi = V′i−Vi.

The function f is obtained by approximating each measured control signal and its relation of increment (Vi, ΔVi) i = 1, 2, 3, ... In accordance with the required accuracy by an appropriate method.

[Sixth Embodiment] In this embodiment, a method for obtaining the numerical table shown in the fourth embodiment will be described. FIG. 10 is a graph showing the relationship between the control signal V to the filter drive circuit 12 and the increment ΔV of the control signal V corresponding to the channel interval Δλch. The horizontal axis represents the control signal V, and the vertical axis represents the increment ΔV of the control signal V. The black dots are (V
The value of Vi calculated from i, λi), the solid line represents the value of ΔV provided by the numerical table, and dV represents the step when changing the control signal V, that is, the unit of quantization.

In order to obtain the above numerical table, the following procedure is taken. First, as in the first embodiment, measurement (Vi, λ
i) i = 1, 2, 3, ... (Vi, ΔVi) i = 1,
Get 2, 3, ... A plot of this is the black dots in FIG.

Since the control signal V is for numerical control,
Discrete values Vn = V0 + n · dV (n = 1, 2, 3, ...) Are taken. Here, V0 is an initial value, and dV is a unit (quantization unit) of the control signal. The value of ΔVi obtained by the calculation is quantized in units of dV. That is, the value of ΔV is replaced with the closest value of the discrete values Vn.
For example, in FIG. 10, ΔVi is Vk−dV / 2 ≦ ΔVi <Vk + dV / 2, so that value is set to Vk.

Quantization is performed on all the increments ΔV of the control signal V obtained by the calculation to obtain the relation (Vi, ΔVi) i = 1, 2, 3, ... Between the control signal V and the increment ΔV. These values can be used as the values in the above mathematical table.

As a result of quantizing the value of the control signal V and its increment ΔV, the increment ΔV of the control signal may continuously take the same value for i. For example, in FIG.
The values of Vi−2 to Vi + 2 are all equal to Vk. In this way, when the values of the increments ΔVi of the control signals are equal for a part (Vi, ΔVi) i = j, j + 1, j + 2, ... J + k of the numerical table (Vi, ΔVi) i = 1, 2, 3, ... Replace part of the above table with only one set of numbers. However, if Vi is arranged in ascending order or descending order, a part of the above-mentioned numerical table is replaced with a set of numerical values having the minimum or maximum control signal value.

Even if such replacement is performed, the value of the control signal referenced from the mathematical table does not change. On the other hand, the replacement can reduce the size of the numerical table, save the storage capacity of the wavelength control circuit 1 and speed up the search.

[Seventh Embodiment] In this embodiment, a method suitable for correcting the non-linearity of the wavelength tunable filter 8 will be described by obtaining the mathematical table shown in the fourth embodiment.

FIG. 11 is a graph for explaining the method of correcting the numerical table in this embodiment. FIG. 11A shows the relationship between the control signal V and the wavelength λ of the detector. Black points are measurement points, white points are points newly calculated (V
a, λa) and (Vb, λb). The solid line represents the increment value ΔV referenced from the value of the control signal V by the numerical table, and the thick dotted line represents the increment value referenced by the corrected numerical table.

When the numerical tables are prepared by the methods described in the fourth and sixth embodiments, the wavelength tunable filter 8 has a large non-linearity and the wavelength dependence of the channel spacing may become large when the number of measurement points is small. There is.

For example, from section Vi to section shown in FIG.
For Vi + 2, it is assumed that the increment ΔV of the control signal V is obtained, the rounding operation is performed, and then a mathematical table showing the relationship between the control signal V and the increment ΔV thereof is obtained. This is shown in FIG. 11 (b).
It is. If the nonlinearity of the wavelength tunable filter 8 is large and the number of measurement points is small, ΔVa (= ΔVi) and ΔVb (= Δ
As shown in (Vi + 1), the value of ΔV may change more than the step dV of the control signal. In this case, Vi ~ Vi + 2
Of the ideal increment ΔVi of the control signal V in FIG.
(Solid line in (a)) and the value of the increment referenced from the table (Fig. 1).
The difference of 1 (a) is large, which causes the channel interval variation.

Therefore, points in the middle of the measurement points, (Va, λa) and (Vb, λb) in FIG. 11B in this example, are predicted by interpolation or the least square method. Then, with respect to the predicted point, an increment corresponding to the channel interval is calculated, and (Va, ΔVa) and (Vb, ΔVb) are calculated and added to the mathematical table. As a result, the change in the increment value of the control signal as shown by the thick dotted line in FIG.

Specifically, the following procedure is performed to predict a point in the middle of the measurement points and correct the numerical table. (1) Numerical tables (Vi, Δ) obtained by the methods of the fourth and sixth embodiments
Vi) For i = 1, 2, 3, ...
When V is set, all values i such that | ΔVi + 1−ΔVi | ≧ 2dV are obtained. (2) With respect to the above i, the values of Vai and Vbi of the control signal V at λai = (λi + λi + 1) / 2, λbi = (λi + 1 + λi + 2) / 2, are the same as those in the fourth embodiment. Ask for. (3) Regarding Vai and Vbi, the increment Δ of the control signal
The values ΔVai and ΔVbi of V are obtained in the same manner as in the sixth embodiment. The obtained set of numerical values (Vai, ΔVai), and (V
bi, ΔVbi) is added to the table. (4) The ΔV of the newly obtained mathematical table is examined, and if the same ΔVai value continues continuously, replacement of the numerical value set is performed as in the sixth embodiment.

In this example, twice the step of the control signal is used as the reference value for whether or not the correction is performed. In practice, this reference value may be larger than the step dV of the control signal, and the value is set according to the required accuracy of the channel interval.

If one correction is insufficient, the above procedure is repeated on the numerical table to obtain a corrected numerical table.

[0083]

As described above, according to the first aspect of the present invention, a new addition is made by numerically correcting the nonlinearity between the control variable to the filter drive circuit and the transmission wavelength of the filter. It is possible to increase the number of channels that can be used by the system by eliminating the wavelength dependence of the channel spacing without requiring a circuit.

Further, according to the second invention, it is possible to provide a simple method for numerically correcting the nonlinearity of the control variable and the transmission wavelength.

Further, according to the third invention, it is possible to provide another simple method for numerically correcting the nonlinearity of the control variable and the transmission wavelength.

In addition, according to the fourth invention, a new additional circuit is obtained by numerically correcting the non-linearity between the control signal to the driving circuit of the wavelength tunable filter and the transmission wavelength of the wavelength tunable filter. It is possible to increase the number of channels that can be used by the system by eliminating the wavelength dependence of the channel spacing without the need for the. More specifically, from the measurement result of the control signal to the filter drive circuit and the transmission wavelength of the filter, the increment corresponding to the channel interval for each measured value of the control signal can be obtained. further,
Calculate the value of the increment of the control signal corresponding to the channel interval from the value of the control signal by an approximate expression, or obtain a numerical table that refers to the value of the increment, quantize the value of the numerical table and perform interpolation, etc., By obtaining an accurate value of the control signal, it is possible to eliminate the increase in the number of usable channels of the system and the wavelength dependence of the channel spacing.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining a wavelength control method according to first and second embodiments of the present invention.

FIG. 2 is a diagram for explaining a sweep operation according to the first and second embodiments of the present invention.

FIG. 3 is a diagram for explaining a problem in a conventional example.

FIG. 4 is a diagram for explaining a wavelength control method in a conventional example.

FIG. 5 is a configuration diagram showing a network of a wavelength division multiplexing communication system.

FIG. 6 is a diagram for explaining a wavelength control operation of the wavelength division multiplexing communication system.

FIG. 7 is a configuration diagram of a communication node used in the wavelength division multiplexing communication system.

FIG. 8 is a configuration diagram of an optical transmitter in a communication node used in a wavelength division multiplexing communication system.

FIG. 9 is a graph showing the relationship between the control signal of the filter drive circuit and the transmission wavelength of the wavelength tunable filter.

FIG. 10 is an explanatory diagram showing a relationship between measured values of control signals of the filter drive circuit and a numerical table.

FIG. 11 is an explanatory diagram for explaining a method of correcting a numerical table.

[Explanation of symbols]

 1 wavelength control circuit 2 light source drive circuit 3 wavelength variable light source 4 optical modulator 5 optical branching device 6 optical switch 7 optical multiplexer 8 wavelength tunable filter 9 light receiving element 10 amplifier 11 discriminator 12 filter drive circuit 14 star coupler 151 to 15n terminal 161 to 16n Communication node 171 to 17n, 181 to 18n Optical fiber 19 Optical brancher 20 Optical receiver 21 Optical transmitter

Claims (14)

[Claims] 1. A light transmission means connected to a transmission line and capable of changing a transmission wavelength, a drive means for generating a drive voltage for the light transmission means based on an input signal, and a control for the drive means. In a wavelength control method comprising a wavelength control system that outputs a value of a variable and changes the value of the control variable to control the transmission wavelength of the light transmission means, the wavelength control system is a transmission wavelength of the light transmission means. The wavelength control method is characterized in that an increment of the control variable (hereinafter referred to as an increment of the control variable) required to change the above-mentioned variable by a predetermined wavelength difference is calculated from the control variable. 2. An optical communication system using the wavelength control method according to claim 1. 3. The wavelength control system stores in advance a relational expression between the control variable and the increment of the control variable, and calculates the increment of the control variable from any of the control variables using the relational expression. The wavelength control method according to claim 1, wherein 4. An optical communication system using the wavelength control method according to claim 3. 5. The wavelength control system stores, in advance, a plurality of sets of numerical values each including the control variable and an increment of the control variable in the control variable, and refers to the plurality of numerical values,
The wavelength control method according to claim 1, wherein an increment of the control variable is calculated from the control variable. 6. An optical communication system using the wavelength control method according to claim 5. 7. A wavelength control means comprising: a light transmission means for transmitting a predetermined wavelength variable wavelength in an optical signal; and a wavelength control means capable of sweeping a transmission wavelength of the light transmission means by a control signal. A wavelength control method for calculating a value of an increment of the control signal (hereinafter referred to as an increment of the control signal) from the control signal, wherein the transmission wavelength of the light transmission means and a control signal corresponding to the transmission wavelength At least two or more sets of values are measured, and the values of the control signal are separated from the corresponding transmission wavelength by a predetermined wavelength interval (hereinafter, referred to as a target wavelength).
The value of the control signal corresponding to each target wavelength (hereinafter, referred to as a target control signal value) is obtained by approximation from the at least two sets of the transmission wavelength and the value of the control signal, and the value of the control signal is The increment value for each measured control signal is obtained by taking the difference between the values of the target control signals, and the transmission wavelength of the light transmitting means is controlled based on the obtained increment value. Wavelength control method. 8. An optical communication system using the wavelength control method according to claim 7. 9. The wavelength control method according to claim 7, wherein from the measured increment value for each control signal,
The wavelength control method is characterized in that the relationship between the control signal and the increment value is approximated by an appropriate function, and the wavelength control means uses the approximation function to calculate the increment of an arbitrary control signal from the control signal. . 10. An optical communication system using the wavelength control method according to claim 9. 11. The wavelength control method according to claim 7, wherein a numerical table made up of a plurality of numerical values consisting of the measured control signal and its increment value is arranged in ascending or descending order with respect to the value of the control signal. Then, the value of each increment is quantized with a predetermined value as a unit, and in the numerical table subjected to the quantization operation, a portion where the increment value continuously takes the same value (hereinafter referred to as a continuous portion). ), A plurality of consecutive parts, each of the successive parts is replaced by a set of numerical values in which the value of the control signal is minimum or maximum, and the wavelength control means refers to the replaced numerical table. A wavelength control method, wherein the increment of the control signal is calculated from the control signal. 12. An optical communication system using the wavelength control method according to claim 11. 13. The wavelength control method according to claim 7, wherein a numerical table made up of a plurality of numerical values consisting of the measured control signal and its increment value is arranged in ascending or descending order with respect to the value of the control signal. However, a pair of control signal values that are adjacent to each other in the above numerical table and whose absolute value of the difference between the corresponding increment values is a predetermined value or more (hereinafter,
They are called control signal 1 and control signal 2. ),
The value of the control signal following the control signals 1 and 2 is defined as a control signal 3, and the transmission wavelengths of the light transmitting means (hereinafter, referred to as transmission wavelengths 1 to 3) corresponding to the control signals 1 to 3 are measured. When the wavelength between the transmission wavelengths 1 and 2 is the intermediate wavelength 1 and the wavelength between the transmission wavelengths 2 and 3 is the intermediate wavelength 2, the control corresponding to the intermediate wavelengths 1 and 2 Signal value (below,
Called intermediate control signal values 1 and 2. ) Is obtained by approximation from the measured transmission wavelength and the value of the corresponding control signal, and for the set of numerical values of the intermediate wavelength and the intermediate control signal value, like the measurement point, corresponds to the increment of the intermediate control signal value. A target control signal value (hereinafter, referred to as an intermediate target control signal value) is obtained, and a set of numerical values including the intermediate control signal value and the intermediate target control signal value is added to the mathematical table. Wavelength control method. 14. An optical communication system using the wavelength control method according to claim 13.