JPH03259632A - Optical heterodyne polarized wave diversity receiver - Google Patents

Abstract

PURPOSE:To maintain a required S/N (reception sensitivity) stably by applying frequency conversion through the use of an oscillator and a mixer to one of intermediate frequency signals corresponding to two polarized wave components, synthesizing the result with an intermediate frequency component not subject to frequency conversion and inputting the result to a demodulation circuit. CONSTITUTION:A photoelectric conversion section 2 applies photoelectric conversion to a signal light together with a local oscillating light for each orthogonal polarized component and outputs two intermediate frequency signals having a frequency corresponding to the frequency of the signal light and the frequency of the local oscillation light according to each of the said polarized wave components. A mixer 6 mixes any of two intermediate frequency signals with an output signal of an oscillator 4. An adder 12 synthesizes output signals of 1st and 2nd band pass filters 8,10. A demodulation circuit 14 applies demodulation based on the signal from the adder 12. Thus, in the case of synthesizing the signal obtained for each polarized wave component at the intermediate frequency signal stage, an optical heterodyne polarized wave diversity receiver is realized in which the phase matching of the intermediate frequency signal is not required and the weight addition is not called for.

Description

[Detailed Description of the Invention] Table of Contents Overview Industrial Field of Application Conventional Technology Problems to be Solved by the Invention Means and Functions for Solving the Problems Example Effects of the Invention Signals Obtained for Each Polarization Component Regarding an optical heterodyne polarization diversity receiver that combines signals at an intermediate frequency signal stage, the main purpose of the present invention is to provide the above-mentioned receiver that does not require phase matching of intermediate frequency signals and does not require addition of weighting when performing the above combining, The signal light is converted into two intermediate frequency signals having a frequency corresponding to the difference between the frequency of the signal light and the frequency of the local light by optical-to-electrical conversion together with the local light for each polarized component whose plane of polarization is orthogonal to each other. an oscillator that oscillates at a predetermined oscillation frequency, a mixer that mixes either one of the two intermediate frequency signals with the output signal of the oscillator, and the two intermediate frequency signals first and second bandpass filters into which the other of the filters and the output signal of the mixer are respectively input, an adder that combines the output signals of the first and second bandpass filters, and a signal from the adder. and a demodulation circuit that performs demodulation based on.

INDUSTRIAL APPLICATION FIELD The present invention relates to an optical heterodyne polarization diversity receiver that combines signals obtained for each polarization component in an intermediate frequency signal stage.

Coherent optical transmission systems are suitable for long-distance transmission because they can improve reception sensitivity compared to the intensity modulation/direct detection systems currently in use. Further, since frequency selection can be performed relatively easily by electrical processing after optical detection, it is suitable for high-density FDM (frequency division multiplexing) transmission. Here, since the words "detection" and "demodulation" are likely to cause confusion, they are used in the specification of this application as "detection" and "demodulation".
The phrase "detection" or "optical detection" is used to mean the conversion of an optical signal to an electrical signal (intermediate frequency signal), and the phrase "demodulation" is used to mean the conversion from an intermediate frequency signal to a baseband signal.

When receiving by applying heterodyne detection, it is required that the polarization states of the signal light and the local light be the same when mixing the signal light and the local light. Shifts in polarization state lead to deterioration of reception sensitivity. For example, if the signal light and local light are linearly polarized waves and their polarization planes are orthogonal to each other, optical interference will occur and reception will not be possible. becomes. - Since a single mode fiber used in a boat does not have the ability to maintain the polarization state, the polarization state at the receiving end fluctuates over time due to environmental changes.

Therefore, in order to maintain the required receiving sensitivity, it is necessary to deal with variations in the polarization state at the receiving end.

2. Description of the Related Art A polarization diversity method is known as a technique for dealing with variations in polarization state at a receiving end. FIG. 17 shows an example of the configuration of a receiver to which this method is applied. Reference numeral 112 denotes an optical-to-electrical converter configured by combining a polarization separator, an optical coupler, a photodetector circuit, etc., and this converter converts the signal light into optical signals together with the local light for each polarized component whose polarization plane is perpendicular to each other. After electrical conversion, two intermediate frequency signals having a frequency corresponding to the difference between the frequency of the signal light and the frequency of the local light are output for each polarization component. These intermediate frequency signals are transmitted through variable gain AGC amplifiers 114.11.
6, passes through a bandpass filter 118120, and inputs to demodulation circuits 122 and 124. The demodulated signals are weighted and combined by, for example, the square law in an adder circuit 126, and then sent to a baseband circuit 128. 130 is an AGC (automatic gain control) circuit, and this circuit is connected to the demodulation circuit 122.
．． Detecting the power of the intermediate frequency signal input to 124,
AGC amplifier 114°11 so that the power is constant
6 gain control. If two series of intermediate frequency signals are added as they are, the signal may disappear, so in order to perform AFC (automatic frequency control), the IF signal extraction circuit 132
is provided. The frequency of the extracted signal is converted into, for example, a voltage signal by the frequency discrimination circuit 134, and this voltage signal is input to the FC circuit 136. The AFC circuit 136 controls the oscillation frequency (wavelength) of the local light source 138 so that the input signal level remains constant.

This type of receiver has the following drawbacks.

(2) Two series of intermediate frequency signal stage circuits from optical detection to demodulation are required, and their characteristics are required to be the same. Therefore, the circuit scale increases and the adjustment becomes complicated.

■ It is necessary to add the baseband signal after demodulating. Since the level of the signal input to this adder circuit changes greatly, an adder circuit with a large dynamic range is required. Furthermore, depending on the circuit configuration, weighting may require addition in order to maintain a required S/N ratio.

(2) An extraction circuit is required to extract an intermediate frequency signal for AFC that is independent of the polarization state from two intermediate frequency signals whose power changes depending on the polarization state. Its circuit configuration is extremely complex in practice.

(2) For the same reason as (2), the configuration of the AGC circuit becomes extremely complicated.

In order to solve some or all of the drawbacks of the conventional technology (Fig. 17) in which the signals obtained for each polarization component are combined in the baseband signal stage, the signals obtained for each polarization component are combined. A receiver has been proposed in which synthesis is performed at an intermediate frequency signal stage. This technique will be explained with reference to FIG. The two series of intermediate frequency signals from the optical-to-electrical converter 140 are weighted and added in an adder circuit 142, and the output signal is input to a demodulation circuit 148 via a bandpass filter 144 and an AGC amplifier 146, and is demodulated. The signal is baseband circuit 15
Sent to 0. 152 is an AGC circuit, 154 is a frequency discrimination circuit, 156 is an AFC circuit, and 158 is a local light source. When combining signals obtained for each polarization component at an intermediate frequency signal stage, two series of intermediate frequency signals may cancel each other out. For example, if two series of intermediate frequency signals have equal intensities and a phase shift of π, the signals will disappear if they are simply added. Therefore, it is necessary to perform phase detection of each intermediate frequency signal by the phase detection circuit 160, and adjust the phase of at least one intermediate frequency signal by the phase shifter 162k to perform phase matching.

Problems to be Solved by the Invention Regarding the phase shifter that was conventionally required when combining signals obtained for each polarization component at an intermediate frequency signal stage, it is difficult to realize a phase shifter with sufficient characteristics for practical use. be. That is,
Since the phase difference that occurs between two series of intermediate frequency signals is caused by variations in the polarization state in the optical transmission path, it is impossible to identify the degree of polarization state variation, including sudden variations. , the phase shifter is required to have a characteristic that allows unlimited phase shifting, making it difficult to realize a phase shifter. Also, considering that one intermediate frequency signal may disappear,
There are also difficulties in realizing the phase detection circuit.

Also, since the two series of intermediate frequency signals are in the same band,
Even if phase matching is performed and the signals are simply added, the S/N ratio will fluctuate in accordance with the fluctuation in the power ratio of the two series of intermediate frequency signals due to the fluctuation in the polarization state. In order to prevent this fluctuation in the S/N ratio, it is effective to add weights that relatively increase the gain of sequences with a large power ratio and relatively decrease the gain of sequences with a small power ratio. In order to add weights, it is necessary to distinguish between changes in signal power due to changes in the polarization state and changes in signal power due to changes in the power of the signal light itself. It becomes necessary.

The present invention was created in view of the above circumstances, and when the signals obtained for each polarization component are synthesized at the intermediate frequency signal stage, phase matching of the intermediate frequency signals is not required, and weighting does not require addition. The purpose is to provide an unnecessary optical heterodyne polarization diversity receiver.

It is also an object of the present invention to adapt this receiver to various modulation schemes.

Purposes other than these will become clear from the description below.

Means and Effects for Solving the Problems One of the intermediate frequency signals corresponding to the two polarization components is up-converted or down-converted by using an oscillator and a mixer to perform frequency conversion. The signal and the unconverted intermediate frequency signal are combined and input to the demodulation circuit. By taking such measures, it is possible to prevent signal loss, which is a concern when simply synthesizing intermediate frequency signals, and thus eliminates the need for phase matching of intermediate frequency signals. In addition, since the frequency-converted signal band and the non-frequency-converted intermediate frequency signal band are different, weighting at the time of synthesis or prior to the bottom can be done without processing the required S/
It becomes possible to maintain a stable N ratio (reception sensitivity).

FIG. 1 is a diagram showing the basic configuration of the present invention. Reference numeral 2 denotes an optical-to-electrical converter, which performs optical-to-electrical conversion of the signal light for each polarization component whose polarization planes are perpendicular to each other, together with the local light, and converts the frequency to the difference between the frequency of the signal light and the frequency of the local light. Two intermediate frequency signals are output for each of the polarization components. 4 is an oscillator, which oscillates at a predetermined oscillation frequency. 6 is a mixer;
Either one of the two intermediate frequency signals and the output signal of the oscillator 4 are mixed. 8.10 are the first and second respectively
The other of the two intermediate frequency signals and the output signal of the mixer 6 are input to these filters, respectively. 12 is an adder which combines the output signals of the first and second band pass filters 8°IO. 14
is a demodulation circuit, which performs demodulation based on the signal from the adder 12.

Now, if the center frequency of the intermediate frequency signal is fc and the oscillation frequency of the oscillator 4 is Δf, then the output signal of the mixer 6 has a center frequency of fc + Δ, as shown in FIG. 2(a).
A signal whose center frequency is at fc and a signal whose center frequency is fcΔf. Therefore, by setting the center frequency of the passband of the first bandpass filter 8 to fc and the center frequency of the passband of the second bandpass filter 10 to fe; Δf, an intermediate frequency having a center frequency at f is obtained. Signal and fc +Δ
The adder 12 can synthesize the up-converted signal having the center frequency at f. Furthermore, by setting the center frequency of the passband of the first bandpass filter 8 to fc and the center frequency of the passband of the second bandpass filter 10 to fc +Δf, an intermediate frequency signal having a center frequency at fc can be obtained. The adder 12 can synthesize the down-converted signal having the center frequency at f+Δf. Upconversion is suitable when fc is low, and downconversion is suitable when fc is high.

In order to amplify or down-convert an intermediate frequency signal whose center frequency is fc by Δf, an image signal can also be used as follows. That is, Figure 2 et al.)
As shown in , when the oscillation frequency of the oscillator 4 is 2fc+Δf, the output signal of the mixer 6 is a signal having a center frequency at 3fc+Δf and a signal having a center frequency at f,+Δf. By setting, up-conversion of Δf is realized. Similarly, as shown in FIG. 2(C), when the oscillation frequency of the oscillator 4 is 2fc+Δf, the output signal of the mixer 6 is 3
Since a signal has a center frequency between fo and Δf and a signal has a center frequency at fc+Δf, down-conversion of Δf is realized by setting the pass band. Hereinafter, the case where an image signal is used in a good manner as shown in FIG. 2(a) may be referred to as 1 and the normal case. The following explanation is for the normal case of up-conversion unless otherwise specified.

The intermediate frequency signal output from the optical-to-electrical converter 82 for each two polarization components is expressed by the following equation.

Ex=S(t) Acos [2πfet+φ(t)
:] ・(1)E,=S(t) (1
−A2) “2cos [2πfct+φ(1)−θ:
...(2) Here, s (B is the component proportional to the electric field of the signal light; hereinafter, "S (t) J is omitted and expressed as rsJ"), A is the electric field component of the X polarization, f is the center frequency,
φ(1) is a modulation signal component, and θ is a polarization phase fluctuation component. It is assumed that the fluctuations in A and θ are sufficiently slow relative to the modulation speed of the signal.

If the two intermediate frequency signals expressed by equations (1) and (2) are simply added, A=■ir2. When θ=π, the intermediate frequency signals cancel each other out and disappear. So, for example, Ex
Shift the frequency towards. The frequency shifted signal is expressed as:

Ex −5Acos [2yr(fc+Δf)t+
φ(t) ]・(3) Since the signals of E and Ey are separated on the frequency axis, there is no signal loss when they are synthesized. Before signal synthesis, a bandpass filter is used to remove high-frequency sidelobes. The synthesized signal is expressed by the following equation.

E (t)=SAcos [2π(fc+Δf)t+
φ(t)]+5(1-A2)”2CO6[2πfct
10 φ (1)-θ ]・(4) FIG. 3 shows an example of the spectral arrangement of each signal.

The vertical axis is the spectral density P (f), and the flu axis is the frequency f
It is. ■, ■ are intermediate frequency signals output from the optical-to-electrical converter 2, ■ is an output signal of the mixer 6, ■ is an output signal of the first bandpass filter 8, and ■ is the output of the second bandpass filter 10. The signal ■ is for the output signal of the adder 12.

Next, examples of the configuration and operation of a demodulator suitable for various modulation methods will be explained.

(1) In case of ASK method (φ(t) −〇) An example of the configuration of the demodulation circuit in this case is shown in FIG.

This demodulation circuit distributes the input signal equally and outputs 3dB
It is configured to include a coupler 16 and a multiplier 18 that multiplies the output signal of the 3 dB coupler 16. When this demodulation circuit is used, the output V is obtained as follows: V=E (t)XE (t). Therefore, V=(1/2)S2A2+(1/2)S2(1-A')
+(1/2)S”A”cos [:4 π(fc+Δ
f)t]+(1/2)S2(1-A2)CO5[4πL
t+2θ]+52A(1-A2)”2cos [2π
Δft-θ]+32A(1-A2)""(os[4
πfct+2πΔft+θ] (5). Usually, the value of fc is 2 times to 4 times the bit rate B of the transmission signal.
It is set to double. For example, if fc=2B (5)
Looking at the signal spectrum bands of each term in the equation, the first and second terms are DC-B, the third term is 3B+Δf~5B+Δf1, and the fourth term is 3B~5B.

The fifth term is Δf-B~Δf+B, and the sixth term is 3B↓2Δf~
The distribution will be 5B+2Δf. Here, in order to extract only the first term of equation (5) using a low-pass filter, the frequency shift amount Δf needs to satisfy the following equation, especially in relation to the fifth term.

Δ f ≧2 B ... (6) (
6) By setting the oscillation frequency of the oscillator 4 to satisfy the formula, when the output V is passed through an appropriate low-pass filter, the output is PFm'/2S2A2+'/2S2(1-A2) S2...
- From (7) and t, a baseband signal without polarization dependence can be obtained. In this case, it is desirable that the mixer 6 satisfies the following requirements.

■ It must have an output characteristic that outputs a signal proportional to the square of the input signal.

■ The input signal band is, for example, at least fc + Δf
+B=2B+28+B=5B, so it must be a broadband device that meets this.

■ Since addition processing is also performed within the mixer, there should be no difference in output characteristics for the two input signals.

By using such a mixer, the device of the present invention can be applied to the ASK system.

(2) In the case of DPSK method (S is constant, φ-0 or π) An example of the configuration of the demodulation circuit in this case is shown in FIG.

This demodulation circuit distributes the input signal equally and outputs 3dB
A coupler 20, a T delay circuit 22 that delays one of the output signals of the 3 dB coupler 20 by a time T corresponding to one time slot of the transmission signal, and a T delay circuit 22 that delays the other output signal of the 3 dB coupler 20 and the output signal of the T delay circuit 22. The multiplier 24 is configured to perform multiplication. When using this demodulation circuit,
As the output V, V=E (t) xE (t+T
) is obtained. Therefore, V= '/, s2^"cos [2π(fe+Δf)T
+φ(t+T)-φ(0]+'/1s2(t-^')c
os [2πfeT+φ(t+T)−φ(t)]+1/
2S2^2cos [2rr (fc+Δf) (2t
+7>+φ(t+T)+φ(t)〕+'/ 2S" (
1-A2) cos [2yr f c (2t+T)
+φ(t+T)+φD)+2θ]+l/232^(1-
person') ””cos [2w (fc+Δf)t+2
rr fe(t+T)+φ(t+T)+φ(1)+θ] +I/2S2^(1-A") "2CO8[-2rrΔ
f t+2 rr f cT+φ(t+T)−φ(1)
+θ] +'/2S"A(1-^") "cos [:2 π(
fc+Δf) (t+T)+2 πfct+φ(t+T
) + φ (1) 10θ] +'/2S"A (1-^') "'cos [2rr
(fc+Δf)T+2 yrΔft+φ(t+T)−φ
(1)-θ] ...(8). In order to maximize the 1x modulation efficiency of the demodulation circuit (so-called phase detection efficiency of a delayed detector), fc should be selected as follows.

fc=l8nB (n is a natural number greater than or equal to 3) - (9) Also, for the same reason, fe+Δf='/21B (J is a natural number greater than or equal to 3>-at
It is good to set it to i. Here, if n and 1 are even or odd, the respective output relationships will be reversed,
Signal is not played properly. Taking the above frequency selection into consideration, the output voltage V is expressed as follows.

L S'(-1) CO5Cφ(t+T)-φ(t
)]+'/2s'A2(-1) cos(4;r
(fe+Δf)t+φ(t+T)+φ(t)]+'/2
S"(1-A2)(-1)"cos C4rrfct+
φ(t+T)+φ(t) +2θ〕+S”A (1-A
')"" (-1) cos [2rr (2L+Δ
f) t + φ (T) 1φ (1) + θ] +'/zs”A(1-A2) ””(-1)I”cos
[2yrΔ"-φ(t+T)+φ(1)-θ]+゛8SA(1-A')"'(-1)'cos [
2rrΔft+φ(t+T)φ(1)-〇〕
...0001) Looking at the signal spectrum of each term in the equation, the first term (required baseband signal) is DC-B, and the second term is (f-1)B ~ (f+1
)B, the third term is (n -1)B ~ (n-f-1)B,
The fourth term is 18(n+j!-2)B='/2 (n-,
R+2) B, the fifth term and the sixth term are '/2 (jl'
n2) B~'/2 (A n+2)B.

In order for the band of the signal below the second term to not overlap with the baseband signal, n>2. i>2. R-n>4, that is, n
> 2 and f2 > 4 + n. For example, (n, ji(3,7>, (n,
A) -(4,8) is good. Therefore, R=n-2k (k is a natural number of 2 or more). Therefore, L - ('/2)nB (n is a natural number of 3 or more>・aV
The center frequency is set to satisfy Δf=kB(k
is a natural number greater than or equal to 2)...03) Set the oscillation frequency of the oscillator 4 to satisfy the following: Therefore, remove the terms after the second term in equation 01) with an appropriate low-pass filter. The output is LPF(V)・52(-1) cos Cφ(t
+T)-φ(t) : ] ·--ai) can be obtained. In this way, even when the DPSK method is applied, a polarization-independent baseband signal can be obtained. In this case, a mixer with substantially the same characteristics as in the ASK system is used.

(3) In case of CPFSK method (S is constant) FIG. 6 shows an example of the configuration of a demodulation circuit that can be used in this case.

This demodulation circuit distributes the input signal equally and outputs 3dB
A coupler 26, a τ delay circuit 28 that delays one of the output signals of the 3 dB coupler 26 by a predetermined delay time τ, and a multiplier 30 that multiplies the other output signal of the 3 dB coupler 26 by the output signal of the r delay circuit 28. It is prepared and configured. C.P.
When the FSK method is applied, the above predetermined delay time T
is set as follows according to the modulation index m.

Here, the modulation index m becomes 1 when the amount of frequency deviation corresponds to the bit rate of the transmission signal, and this modulation index m is usually set to a value of 0.5 or more.

τ=1/(2mB) ...αω In this case, the demodulation circuit shown in FIG. 6 has a frequency discrimination function,
The discrimination curve is shown at 34 in FIG. In FIG. 8, numeral 36 indicates the spectrum of the intermediate frequency signal, in which fS is the frequency at the space time and fm is the frequency at the mark time. spectrum 3
6 and the frequency discrimination curve 34, in order to maximize the demodulation efficiency, it is sufficient to set the center frequency of the intermediate frequency signal so as to satisfy the following equation.

fe=(2n+1)/4τ=(2n+1)mB/2 (n is a natural number)...αω
In addition, in order to make the baseband signal obtained by demodulating the upconverted signal and the baseband signal obtained by demodulating the intermediate frequency signal that was not upconverted, the upconverted signal is Make sure that the center frequency of satisfies the following equation.

fc+Δf=(2n+1)mB/2-i-2kmB (k is a natural number)...07) An example of the spectrum of the up-converted signal satisfying the αω expression is shown at 35 in FIG. In order for the center frequency of the up-converted signal to satisfy the G-force equation, the following equation must be satisfied:
Set the oscillation frequency of the oscillator 4 to .

Δ f = 2 kmB (k is a natural number satisfying km≧1) ...0II
DHere, the reason why the condition of km≧1 is added is as follows. That is, as is clear from the previous description, if Δf is not 2B or more, the baseband signal will be degraded by the beat noise between the intermediate frequency signals corresponding to the two polarization components, so km = 1 is satisfied. The frequency shift amount is set as follows.

FIG. 7 shows another configuration example of a demodulation circuit that can be used when the CPFSK method is applied.

This demodulation circuit includes, in place of the 3dB coupler 26, a 90° hybrid 3dB coupler 32 that equally distributes the input signal and outputs it with a 90° phase difference in the configuration of the demodulation circuit shown in FIG. It is composed of In this case, the frequency discrimination curve is as shown at 38 in FIG.
The frequency discrimination curve 34 of the demodulation circuit shown in FIG. 6 is translated by mB/2 on the frequency axis. In this case, the conditions for maximizing the demodulation efficiency are changed to the formula 00 and are as follows.

fc=nmB (n is a natural number)...α Note that the conditional expression α for the frequency shift amount Δf remains unchanged.

Next, a case will be described in which an image signal is used as the frequency-shifted signal. When frequency shifting the intermediate frequency signal by the mixer 6, the oscillation frequency of the oscillator 4 is changed to 2.
As described above, even if it is set to fc+Δf or 2fc−Δf, a signal according to equation (3) can be obtained (FIG. 2(b) Ic)).

Assuming that Ex expressed by equation (1) is frequency shifted as in the normal case, the signal is expressed by the following equation.

Ex = 5Acos [2π(fc+Δf) -φ(t) E ・ca This formula is shown in Figure 2(b)
This represents a signal obtained by up-converting an intermediate frequency signal using an image signal as explained in . Figure 2 (
The same applies to the case of down-conversion according to the explanation of C), so a description of down-conversion will be omitted. (
4) From the formula and the previous explanation, the ASK method is the same as the normal case, but the DPSK method and CPFS
It is clear that phase inversion needs to be considered for the K method.

(4) When using an image signal in the DPSK system In this case as well, as in the normal case, V' −E (t)xE (
t+T) is detected and expanded as follows.

V'='/2S2A''CO6[2π(fc+Δf)T
-φ(t+T)+φ(t)]+'/2S2(1-A')
cos [:2πfcT+φ(t+T)−φ(T)]+
...
...(21) Therefore, the conditions for maximizing the demodulation efficiency of the demodulation circuit are as in the normal case, using equation (9) and 0
0, so if n and 1 are even and odd, the same argument as in the normal case will be valid. Therefore, D.P.
When using an image signal in the SK method, by setting the center frequency of the intermediate frequency signal and the oscillation frequency of the oscillator 4 so as to satisfy equations 02) and α, a baseband signal independent of polarization can be obtained. Obtainable.

(5) When using an image signal in the CPFSK system In this case, the conditions for maximizing the demodulation efficiency are given as 00 as in the normal case. However, considering equation (A), since the high and low frequencies of the up-converted signal are reversed, how can it be up-converted to the baseband signal obtained by demodulating the up-converted signal? The conditions for the baseband signal obtained by demodulating the Q intermediate frequency signal to be the same are as follows, in place of the OD equation.

fc+△f=(2n=,1)mB/2+(2k
-1) mB (k is a natural number) (22) An example of the spectrum of an up-converted signal that satisfies equation (22) is shown at 37 in FIG. Equation (22) or △
f −(2k −1) mB −(23
). Since the value of Δf is required to be 2B or more as in the normal case, k in equation (23) is a natural number that satisfies (2k-1)m≧2.

Regarding the S/N ratio, it is clear from the operating principle explained so far that there is no risk of deterioration compared to the prior art. Furthermore, since the frequency domains of the two systems of signals input to the adder 12k are separated, if the demodulation circuit is equipped with a multiplier as explained above, weighting based on the square characteristic of this multiplier can be applied. As a result of this action, there is no possibility that the S/N ratio will fluctuate due to fluctuations in the signal ratio of the two series due to fluctuations in the polarization state.

Example The present invention will be explained in detail below.

When implementing this invention, a wideband mixer is required in order to receive a wideband signal. It is effective to apply this invention to the system.

FIG. 10 is a block diagram of an optical heterodyne polarization diversity receiver showing an embodiment of the present invention.

40 is a local light source configured using, for example, a semiconductor laser, and in this local light source 40, the frequency of laser oscillation can be adjusted by adjusting the bias current of the semiconductor laser, etc. .

The optical-electrical converter 2 has, for example, the following configuration. That is, in this example, the optical-to-electrical converter 2 includes first and second polarization separators 42 and 44 that separate the signal light and the local light into polarization components whose polarization planes are orthogonal to each other and output them; First and second optical couplers 46° 48 add and distribute polarized components having the same plane of polarization from the separator and output them, respectively, and perform optical-to-electrical conversion of the light from these optical couplers, respectively. first and second outputting intermediate frequency signals;
The optical detection circuits 50 and 52 are configured. By using constant polarization fiber couplers as the optical couplers 46 and 48, it is possible to specify the polarization state of the light input to each of the optical detection circuits 50 and 52k.

FIG. 11 shows an example of the configuration of the optical detection circuits 50 and 52. In this example, the optical detection circuits 50, 52 each have a single PIN
A light receiving element 56 such as a photodiode is provided, and the light receiving element 56 is configured to receive one of the lights distributed by the first and second optical couplers 46 and 48, respectively. A reverse bias is applied to the light receiving element 56 as usual, and the photocurrent generated in the light receiving element 56 flows through the load resistor 58. The intermediate frequency signal generated as a potential change at the connection point between the light receiving element 56 and the load resistor 58 is transmitted to the front end amplifier 60.
The signal is amplified by and output from the optical detection circuit 50.52. If optical configuration allows, optocoupler 46.4
8 may be received by a single light receiving element 56.

As another example of the optical detection circuit 50, 52, a configuration example of a double-balanced optical detection circuit (DBOR) is shown in FIG. 1st
The configuration shown in FIG. 2(a) includes two light receiving elements 62.6 having substantially the same characteristics, each consisting of a PIN photodiode or the like.
The photocurrent changes occurring in step 4 are taken out as voltage changes, amplified by front-end amplifiers 66 and 68, and then subtracted by a subtracter 70. In this configuration, if the optical path lengths from the optical couplers of the lights incident on the light receiving elements 62 and 64 are set to be equal, as a result of the optical phase reversal in the optical couplers, the signal components input to the light receiving elements 62 and 64 will be The phase is opposite, and the intensity noise components of the local light are in phase. Therefore, the signal components are added, the intensity noise components are canceled out, and the influence of the intensity noise of the local light can be suppressed. Furthermore, since both of the lights distributed by the optical coupler can be used without waste, this is advantageous in improving reception sensitivity. As shown in FIG. 12(b), light-receiving elements 62 and 64 having the same characteristics may be connected in series, and the potential change at the connection point may be amplified and output from the front-end amplifier 72k.

In the configuration of the optical-to-electrical converter 2 shown in FIG. Optical detection is performed for each wave component, but if simplification of the configuration is given priority over maintaining high reception sensitivity,
Either the local light is circularly polarized without polarization separation, or the local light is linearly polarized with a plane of polarization tilted at 45 degrees with respect to the plane of polarization of the signal light that has been polarized. Then, these circularly polarized waves or linearly polarized waves may be added to the polarized wave components of the signal light for optical detection. Further, in this example, the signal light and the local light are polarized and then added and distributed, but the signal light and the local light may be added and distributed and then the polarized light is separated.

In this embodiment, a variable gain amplifier 74 amplifies a signal input to the demodulation circuit 14, and a gain of the variable gain amplifier 74 is controlled by detecting the power of the signal input to the demodulation circuit 14 so that the power is constant. An automatic gain control circuit 76 is provided to perform AGC control.

When the power of the signals after being added by the adder 12 is measured, a value independent of the polarization state is obtained by adding the powers of the intermediate frequency signals corresponding to both polarization components. Since this value is proportional to the power of the received signal light, the input level to the demodulation circuit 14 can be kept constant by performing the above-mentioned AGC control. In this case, an amplifier 74 with a wide band is required, but the number of amplifiers is reduced compared to the AGC control described in FIG. 17, and the control becomes easier. In the case of the ASK method, the demodulated signal itself is
However, in order to perform reliable AGC control, it is desirable to compensate for mark rate fluctuations in the same way as normal AGC control performed on baseband signals.

In the case of the DPSK method and the CPFSK method, the power of the intermediate frequency signal does not vary depending on the mark rate, so the above compensation is not necessary. When the dynamic range of the multiplier in the demodulation circuit 14 is high, AGC control can be performed on the demodulated signal, but generally, the dynamic range of the multiplier to which two signals with widely varying amplitudes are input is Since there are limitations on the number of signals, it is desirable to perform AGC control on the synthesized intermediate frequency signal as in this embodiment. When AGC control is performed before the demodulation circuit 14 in this way, the demodulation circuit 1
A small dynamic resonator of the 4 multiplier becomes sufficient.

Further, this embodiment includes a frequency discriminator 78 that discriminates the frequency of the signal input to the demodulation circuit 14, and an automatic frequency control circuit 80 that controls the frequency of the local light so that the output of the frequency discriminator is constant. As a result, AF
C control is performed.

By performing AFC control, it is possible to obtain an intermediate frequency signal whose center frequency is constant regardless of frequency fluctuations of signal light and local light.

For example, assume that a CPFSK method (modulation index m=1) with a modulation rate of I Gb/s is applied. In this case, the signal light is obtained, for example, by directly modulating a semiconductor laser on the transmitting side. When this signal light is input to the receiver shown in FIG. 10, it is first divided by the polarization separator 42k into two polarization components whose polarization planes are orthogonal. Each polarized wave component is transmitted through optical couplers 46 and 48.
, they are combined with local light having the same plane of polarization, and optically detected by optical detection circuits 50 and 52. The center frequency of the intermediate frequency signal obtained by optical detection is calculated according to the 00 formula, for example, 2. It is set to 5 GHz (when n=2). The intermediate frequency signal from one of the optical detection circuits 50 is input to the first bandpass filter 8 . The center frequency of the pass band of the first band pass filter 8 is made to correspond to the center frequency of the intermediate frequency signal. It is set to 5GHz. The intermediate frequency signal from the other optical detection circuit 52 is
The signal is, for example, up-converted through the mixer 6 and input to the second band-pass filter 10. The center frequency of the pass band of the second band pass filter 10 is, for example, 4. 5GHz
is set to . 1st R and 2nd bandpass filter 8
．． For example, the passband width of 10 is 2. It is set to 5GHz. In order to prevent sidebands from interfering with the baseband band of the up-converted signal when up-converting the intermediate frequency signal, an appropriate band-pass filter is used to filter the signal before up-converting, that is, upstream of the mixer 6. It is also effective to limit the band.

Two optical detection circuits 50.52k The first and second bandpass filters when the equally divided polarization components are input 8.
It is preferable to adjust the power of the up-converted signal input to the second band-pass filter 10 so that the output powers of the filters 10 are equal.

In reality, it is effective to adjust the outputs of the demodulation circuits to be equal in order to compensate for the characteristics of the mixer.

Signals from the first and second bandpass filters 8 and 10 are combined by an adder 12, then subjected to AGC control and input to a demodulation circuit 14. Power detection in the automatic gain control circuit 76 can be performed using, for example, a square characteristic of a Schottky barrier diode or the like.

As the demodulation circuit 14, for example, one having the configuration shown in FIG. 6 is used. The delay time τ is set to 500 ps, and in this case the frequency giving the zero point of the frequency discrimination curve is 2
．． 5GHz, 3. 5G) Iz, 4゜5G
Hz, 5. 5GHz, etc. Therefore, for example, if the center frequency is 2. The same frequency discrimination characteristics are given to a 5 GHz signal and a signal whose center frequency is 4°5 GHz. The demodulated signal is output in the form of the sum of the demodulated signal based on the up-converted signal and the demodulated signal based on the intermediate frequency signal that was not up-converted, so it is possible to obtain an output that is independent of the polarization state. . The demodulated signal is input to the baseband circuit 54,
Processing such as identification processing is performed here.

Regarding AFC control, since the signals input to the frequency discriminator 78 are separated on the frequency axis, it is necessary to deal with this. When the frequency of the signal light and/or local light changes, the frequency change of the amplified and converted signal and the frequency change of the non-amplified intermediate frequency signal are usually in the same direction. That is, when the frequency of the up-converted signal increases, the frequency of the intermediate frequency signal that has not been up-converted also increases. On the other hand, when an image signal is used, the frequency change of the up-converted signal and the frequency change of the non-up-converted intermediate frequency signal are in opposite directions.

That is, when the frequency of the up-converted signal increases, the frequency of the non-up-converted intermediate frequency signal decreases. Therefore, in a normal case, the frequency discriminator 78, as shown in FIG. 13(a),
A frequency discrimination characteristic is used in which the frequency discrimination output V is zero at frequencies fc and fc +Δf, and the signs of the slopes of the frequency discrimination curves are the same.

In this case, when down-converting,
A frequency discrimination characteristic is used in which the frequency discrimination output is zero at frequencies fe and fc-Δf, and the signs of the slopes of the frequency discrimination curves are the same. on the other hand,
When using an image signal, the frequency discriminator 78 selects the frequencies fe and fc +Δf (for up-conversion) or the frequencies fc and fc -Δf (for down-conversion), as shown in FIG. 13(b). In this case, a frequency discrimination characteristic is used in which the frequency discrimination output is zero and the slope of the frequency discrimination curve has a reverse sign.

In this embodiment, since the AGC-controlled signal is input to the frequency discriminator 78, there is no need to consider fluctuations in the amplitude of the input signal to the frequency discriminator 78. When performing AFC control based on
It is desirable to use a frequency discriminator in which the zero point of the frequency discrimination characteristic does not change even if the amplitude of the input signal changes.

Such a frequency discriminator is realized, for example, by providing a characteristic such that the output voltage is the square of the input signal.

A Costas loop frequency discriminator is a frequency discriminator 78 suitable for obtaining the output characteristics shown in FIG. An example of its configuration is shown in FIG.

This Costas loop type frequency discriminator consists of a first 3 dB coupler 82 that equally distributes the input signal and outputs the same, and this 3 dB
The second 3 outputs equally distribute one of the output signals of the coupler.
A first delay circuit 8 that delays one of the output signals of the dB coupler 84 and the second 3dB coupler 84 by a predetermined delay time τ1.
6 and the other of the output signals of the second 3 dB coupler 84 and the first
A first multiplier 8 that multiplies the output signal of the delay circuit 86 of
8, a 90° hybrid 3dB coupler 90 that equally distributes the other output signal of the first 3dB coupler 82 and outputs it with a phase difference of 90°; a second delay circuit 92 that delays, a second multiplier 94 that multiplies the output signal of the second delay circuit 92 by the other of the output signals of the 90° hybrid 3 dB coupler 90, and first and second multipliers. vessel 88.9
and a third multiplier 96 that multiplies the output signal of No. 4.

Since the CPFSK method is applied in this example,
τ1 and τ2 are 1/(2
mB). FIG. 15 shows an example of the relative relationship between the frequency discrimination curve obtained in this case and the spectrum of each signal. 5mB/2 (2゜5G)Iz ) and 9mB/
2 (4,5 GHz) (corresponding to Fig. 13(a)), and these frequencies are the center frequency of the intermediate frequency signal that is not up-converted and the center frequency of the up-converted signal, respectively. The center frequency of . In this way, by using the Costas loop frequency discriminator, AFC control in the CPFSK system becomes possible, and accurate AFC control can be performed regardless of mark rate fluctuations.

When using a Costas loop frequency discriminator,
When the ASK method or the DPSK method is applied, the delay time rl+ of the first and second delay circuits 86.92
τ2 is set to 1/B.

FIG. 16 shows a configuration example of the frequency discriminator 78 that can be used when the modulation index is large. This frequency discriminator includes a 90° hybrid 3 dB coupler 98 that equally distributes an input signal and outputs it with a phase difference of 90'', a delay circuit 100 that delays one of the output signals of this coupler by a predetermined delay time, and The multiplier 102 multiplies the other output signal of the coupler by the output signal of the delay circuit 100.In this embodiment, by setting the delay time of the delay circuit 100 to ins, the Costas loop It is possible to achieve the same frequency discrimination characteristics as when using a type frequency discriminator.

In addition to the example described above, the frequency discriminator 78 can also be configured by using a filter or the like to set a frequency discrimination curve for each band.

When using an image signal, set the oscillation frequency of the oscillator 4 to 8 GHz (=2 fc + 3 GHz),
Set the center frequency of the intermediate frequency signal to 5. 5GHz (=8
-fc).

Effects of the invention The effects of this invention are clear from the description of the examples etc.
If we clarify the parts related to the basic structure for confirmation, it will be as follows. That is, according to the present invention, when the signals obtained for each polarization component are bottomed out at the intermediate frequency signal stage, there is no need for phase matching of the intermediate frequency signals, and addition is unnecessary for weighting, using optical heterodyne polarization diversity reception. This has the effect of making it possible to provide equipment. As a result, the configuration of the receiver is simplified, making it possible to provide a low-cost and highly reliable receiver, which greatly contributes to the development of optical transmission technology.

[Brief explanation of drawings]

Fig. 1 is a diagram showing the basic configuration of the optical heterodyne polarization diversity receiver of the present invention, Fig. 2 is an explanatory diagram of frequency conversion, Fig. 3 is a diagram illustrating an example of the spectrum of each signal, and Fig. 4 is a diagram showing an example of the spectrum of each signal. 6 to 6 are diagrams showing configuration examples of a demodulation circuit applicable to the ASK system, DPSK system, and CPFSK system, respectively. FIG. 7 is a diagram showing another configuration example of a demodulation circuit applicable to the CPFSK system. Figure 8 is a diagram for explaining the frequency discrimination characteristic of the demodulation circuit applied to the CPFSK system and an example of the setting of fe, and Figure 9 is a diagram for explaining the frequency discrimination characteristic of the demodulation circuit shown in Figure 6 and an example of the setting of Δf. FIG. 10 is a block diagram of an optical heterodyne polarization diversity receiver showing an embodiment of the present invention, FIG. 11 is a diagram showing an example of the configuration of a photodetection circuit, and FIG. 12 is a diagram for explaining the optical detection circuit. Other circuit configuration examples (double balanced type)
Figure 13 is a diagram showing an example of the output characteristics of a frequency discriminator for AFC control, Figure 14 is a diagram showing a configuration example of a frequency discriminator for AFC control (Costas loop type), Figure 15 is a diagram showing an example of the configuration of a frequency discriminator for AFC control. converts the frequency discriminator shown in Fig. 14 into CPFS
FIG. 16 is a diagram showing an example of the relative relationship between the frequency discrimination curve and the spectrum of each signal when applied to the K method, FIG. 16 is a diagram showing another example of the configuration of a frequency discriminator for AFC control, FIGS. FIG. 18 is an explanatory diagram of the prior art. 12... Adder, 14... Demodulation circuit.

Claims (1)

[Claims] 1. Optical-to-electrical conversion of the signal light and the local light for each polarization component whose polarization planes are orthogonal is performed to generate two frequencies corresponding to the difference between the frequency of the signal light and the frequency of the local light. an optical-to-electrical converter (2) that outputs an intermediate frequency signal for each of the polarization components; an oscillator (4) that oscillates at a predetermined oscillation frequency; and one of the two intermediate frequency signals and the oscillator (4). ), and first and second bandpass filters (8, 10) into which the other of the two intermediate frequency signals and the output signal of the mixer (6) are input, respectively.
), an adder (12) that combines the output signals of the first and second bandpass filters (8, 10), and a demodulation circuit (14) that performs demodulation based on the signal from the adder (12). An optical heterodyne polarization diversity receiver comprising: 2. In the receiver according to claim 1, when the center frequency of the intermediate frequency signal is f_c and the oscillation frequency of the oscillator (4) is Δf, the passage of the first bandpass filter (8) The center frequency of the band is f_
c, and the center frequency of the passband of the second bandpass filter (10) is f_c+Δf. 3. The optical heterodyne polarization diversity receiver according to claim 2, wherein the center frequency of the passband of the second bandpass filter (10) is f_c-Δf instead of f_c+Δf. . 4. In the receiver according to claim 1, when the center frequency of the intermediate frequency signal is f_c and the oscillation frequency of the oscillator (4) is 2f_c+Δf, the passage of the first bandpass filter (8) An optical heterodyne polarization diversity receiver characterized in that the center frequency of the band is f_c, and the center frequency of the pass band of the second band-pass filter (10) is f_c+Δf. 5. In the receiver according to claim 4, when the oscillation frequency of the oscillator (4) is set to 2f_c-Δf instead of 2f_c+Δf, the center frequency of the passband of the second band-pass filter (10) is f_c+Δf
An optical heterodyne polarization diversity receiver characterized in that f_c-Δf instead of f_c−Δf. 6. In the receiver according to claim 2 or 3, the signal light modulation method is an ASK method, and the demodulation circuit (14) includes a 3 dB coupler (16) that equally distributes and outputs the input signal, and a 3 dB coupler (16) that equally distributes and outputs the input signal. It is equipped with a multiplier (18) that multiplies the output signal of the 3 dB coupler (16), and when the bit rate of the transmission signal is B, the oscillation frequency of the oscillator (4) is adjusted so that Δf≧2B is satisfied. An optical heterodyne polarization diversity receiver characterized in that: 7. In the receiver according to claim 4 or 5, the signal light modulation method is an ASK method, and the demodulation circuit (14) includes a 3 dB coupler (16) that equally distributes and outputs the input signal, and a 3 dB coupler (16) that equally distributes and outputs the input signal. It is equipped with a multiplier (18) that multiplies the output signal of the 3 dB coupler (16), and when the bit rate of the transmission signal is B, the oscillation frequency of the oscillator (4) is adjusted so that Δf≧2B is satisfied. An optical heterodyne polarization diversity receiver characterized in that: 8. In the receiver according to claim 2 or 3, the signal light modulation method is a DPSK method, and the demodulation circuit (14)
is a 3dB coupler (20
), and one of the output signals of the 3 dB coupler (20) is delayed by a time T corresponding to one time slot of the transmission signal.
It includes a delay circuit (22) and a multiplier (24) that multiplies the other output signal of the 3 dB coupler (20) by the output signal of the T delay circuit (22), and adjusts the bit rate of the transmission signal. When B, f_c=(
The center frequency is set to satisfy 1/2)nB (n is a natural number of 3 or more), and the oscillation frequency of the oscillator (4) is set to satisfy Δf=kB (k is a natural number of 2 or more). An optical heterodyne polarization diversity receiver characterized in that: 9. In the receiver according to claim 4 or 5, the signal light modulation method is a DPSK method, and the demodulation circuit (14)
is a 3dB coupler (20
), and one of the output signals of the 3 dB coupler (20) is delayed by a time T corresponding to one time slot of the transmission signal.
It includes a delay circuit (22) and a multiplier (24) that multiplies the other output signal of the 3 dB coupler (20) by the output signal of the T delay circuit (22), and adjusts the bit rate of the transmission signal. When B, f_c=(
The center frequency is set to satisfy 1/2)nB (n is a natural number of 3 or more), and the oscillation frequency of the oscillator (4) is set to satisfy Δf=kB (k is a natural number of 2 or more). An optical heterodyne polarization diversity receiver characterized in that: 10. In the receiver according to claim 2 or 3, the modulation method of the signal light is a C-PFSK method, and the demodulation circuit (
14) includes a 3 dB coupler (26) that equally distributes and outputs the input signal, a τ delay circuit (28) that delays one of the output signals of the 3 dB coupler (26) by a predetermined delay time τ, and the 3 dB coupler ( 26) and the output signal of the τ delay circuit (28), and when the bit rate of the transmission signal is B and the modulation index is m, , the predetermined delay time is set to satisfy τ=1/(2mB), and f_c
The center frequency of the intermediate frequency signal is set so as to satisfy = (2n+1) mB/2 (n is a natural number), and the center frequency of the above oscillator ( 4) An optical heterodyne polarization diversity receiver, characterized in that the oscillation frequency of 4) is set. 11. The receiver according to claim 10, wherein the 3 dB
In place of the coupler (26), a 90° hybrid 3 that equally distributes the input signal and outputs it with a 90° phase difference
with dB coupler (32), f_c=(2n+1)mB
An optical heterodyne polarization diversity receiver, characterized in that the center frequency of the intermediate frequency signal is set to satisfy f_c=nmB (n is a natural number) instead of /2 (n is a natural number). 12. In the receiver according to claim 4 or 5, the signal light modulation method is a CPFSK method, and the demodulation circuit (1
4) is a 3dB coupler (
26), a τ delay circuit (28) that delays one of the output signals of the 3 dB coupler by a predetermined delay time τ, and a τ delay circuit (28) that delays the other output signal of the 3 dB coupler (26) and the τ delay circuit (28).
), and when the bit rate of the transmission signal is B and the modulation index is m, the above equation is set to satisfy τ=1/(2mB). A predetermined delay time is set, f_c
The center frequency of the intermediate frequency signal is set to satisfy = (2n+1)mB/2 (n is a natural number), and Δf = (2k-1)mB (k satisfies (2k-1)m≧2). 1. An optical heterodyne polarization diversity receiver, characterized in that the oscillation frequency of the oscillator (4) is set so as to satisfy (a natural number). 13. The receiver according to claim 12, wherein the 3 dB
In place of the coupler (26), a 90° hybrid 3 that equally distributes the input signal and outputs it with a 90° phase difference
with dB coupler (32), f_c=(2n+1)mB
An optical heterodyne polarization diversity receiver, characterized in that the center frequency of the intermediate frequency signal is set to satisfy f_c=nmB (n is a natural number) instead of /2 (n is a natural number). 14. The receiver according to any one of claims 1 to 13, further comprising: a variable gain amplifier (74) for amplifying a signal input to the demodulation circuit (14); and a power of the signal input to the demodulation circuit (14). The variable gain amplifier (7) detects the power and makes the power constant.
4) An optical heterodyne polarization diversity receiver comprising an automatic gain control circuit (76) for controlling the gain. 15. The receiver according to any one of claims 1 to 14, further comprising a frequency discriminator (78) for frequency discriminating the signal input to the demodulation circuit (14), and an output of the frequency discriminator (78) being constant. an automatic frequency control circuit (80) that controls the frequency of the local light so that
An optical heterodyne polarization diversity receiver comprising: 16. The receiver according to claim 6, 8, 10 or 11, comprising the variable gain amplifier (74) and automatic gain control circuit (76) according to claim 14 and the frequency discriminator (78) according to claim 15. ) and an automatic frequency control circuit (80), the frequency discriminator (78) has frequencies f_c and f_c
An optical heterodyne polarization diversity receiver having a frequency discrimination characteristic in which the frequency discrimination output is zero and the sign of the slope of the frequency discrimination curve is the same at +Δf or frequencies f_c and f_c−Δf. 17. The receiver according to claim 7, 9, 12 or 13, comprising the variable gain amplifier (74) and automatic gain control circuit (76) according to claim 14 and the frequency discriminator (78) according to claim 15. ) and an automatic frequency control circuit (80), and the frequency discriminator (7 to 8) has frequencies f_c and f_
At c+Δf or frequencies f_c and f_c−Δf,
An optical heterodyne polarization diversity receiver characterized by having frequency discrimination characteristics in which the frequency discrimination output is zero and the sign of the slope of the frequency discrimination curve is opposite. 18. The receiver according to claim 16 or 17, wherein the frequency discriminator (78) includes a first 3 dB coupler (82) that equally distributes and outputs the input signal;
A second 3 dB coupler (84) that equally distributes and outputs one of the output signals of the coupler (82), and a first that delays one of the output signals of the second 3 dB coupler (84) by a predetermined delay time τ_1. a delay circuit (86), the other of the output signals of the second 3 dB coupler (84) and the first delay circuit (86);
a first multiplier (88) that multiplies the output signal of 6);
90° which equally distributes the other output signal of the first 3dB coupler (82) and outputs it with a phase difference of 90°.
A second delay circuit (92) that delays one of the output signals of the hybrid 3dB coupler (90) and the 90° hybrid 3dB coupler (90) by a predetermined delay time τ_2, and the output of the 90° hybrid 3dB coupler (90). a second multiplier (94) that multiplies the other signal by the output signal of the second delay circuit (92); and a second multiplier (94) that multiplies the output signals of the first and second multipliers (88, 94). An optical heterodyne polarization diversity receiver characterized in that it is a Costas loop frequency discriminator comprising a third multiplier (96). 19. In the receiver according to claim 18, the claim cited in claim 16 or 17 is any one of claims 6 to 9, and the predetermined delay times τ_1 and τ_2 are An optical heterodyne polarization diversity receiver characterized by a 1/B polarization diversity receiver. 20. In the receiver according to claim 18, the claim cited in claim 16 or 17 is any one of claims 10 to 13, and the predetermined delay times τ_1 and τ_2 are 1/(2mB).An optical heterodyne polarization diversity receiver. 21. The receiver according to claim 16 or 17, wherein the frequency discriminator (78) includes a 90° hybrid 3dB coupler (98) that equally distributes the input signal and outputs a 90° phase difference; The 90° hybrid 3d
A delay circuit (100) that delays one of the output signals of the B coupler (98) by a predetermined delay time, and the output signal of the delay circuit (100) is multiplied by the other output signal of the 90° hybrid 3 dB coupler (98). 1. An optical heterodyne polarization diversity receiver, comprising: a multiplier (102). 22. The receiver according to any one of claims 1 to 21, wherein the optical-to-electrical converter (2) separates the signal light and the local light into polarization components whose polarization planes are orthogonal to each other and outputs the separated signals. 1
and a second polarization separator (42, 44), and the polarization components having the same polarization plane from the first and second polarization separators (42, 44) are added together and distributed, respectively. first and second optical couplers (46, 48) to output, and a first optical coupler (46, 48) to perform optical-to-electrical conversion of the light from the first and second optical couplers (46, 48), respectively, and output an intermediate frequency signal. and a second optical detection circuit (50, 52). 23. The receiver according to claim 22, wherein the first and second photodetector circuits (50, 52) each include a single light receiving element (56), and the light receiving element (56)
and a second optical coupler (46, 48). 24. The receiver according to claim 22, wherein the first and second optical detection circuits (50, 52) each include a pair of optical receivers (62, 64) and are configured in a double-balanced type. , the pair of light receivers (62, 64) are connected to the first and second light receivers (62, 64).
An optical heterodyne polarization diversity receiver, wherein the optical heterodyne polarization diversity receiver is configured to receive both of the lights distributed by the optical couplers (46, 48).