Light Receiving Automatic Gain Control Circuit

This invention relates to a light receiving circuit used for repeaters and terminals, etc., in an optical transmission system.

Repeaters are used in optical transmission systems in order to transmit light signals to the remote terminals. A repeater converts the received light signal into an electrical signal and converts it again into a light signal after waveform equalization and shaping, and finally transmits it to the next repeater. In the terminal station, on the other hand, the received light signal is converted into an electrical signal and thereafter demodulation is carried out.

Automatic gain control (AGC) is carried out at repeaters and terminals so that an electrical signal of constant level can be obtained even if the level of the received light signal changes. FIG. 1 is an ordinary light receiving circuit having an automatic gain control circuit. In this FIG, 11 is a light signal; 12 is an avalanche photo diode (APD); 13 is an amplifier; 14 is a level detector; 15 is the output terminal; 16 is an AGC circuit.

The following methods are currently proposed for executing AGC using the circuit shown in FIG. 1.

(a) Only the multiplication factor M of APD 12 is controled. For example, when the light input level is doubled, M of the APD is reduced to a half (1/2).

(b) The same as method (a) for low input levels with: M of the APD being fixed for high input levels and with the gain of amplifier 13 being controlled.

(c) The same as method (a) for low input levels as is proposed in the published unexamined Japanese patent application No. 53-58748. The gain amplifier 13 is changed step by step for high input levels and simultaneously M of the APD 12 is also controlled.

The methods listed above will contribute to providing an electrical signal of a constant level as an output even when the level of a received light signal changes. FIGS. 2 (1) to (3) show the relation between the light input level P, the multiplication factor M of the APD and the gain G of the amplifier in the methods (a) to (c) given above. In the same figures, the solid line indicates M, the broken line indicates G and the chained line indicates the optimum value of M, explained later. M.sub.min is the minimum value of the applicable multiplication factor of the APD. When M<M.sub.min, which is not practical, the response speed of APD becomes very low due to an increase of the capacitance.

On the other hand, the signal to noise ratio (SNR) of the output signal of amplifier 13 is basically given by the following equation.

Where,

I.sub.o : APD output current for unit light input (when M=1)

P: APD light input power

M: APD multiplication factor

N.sub.s : APD shot noise current power for unit light input (when M=1)

x: Excessive noise figure of the APD

Nth: Converted input noise current power of the next stage amplifier of APD

The equation (1) teaches that SNR depends on P and M, and the optimum value M.sub.o for M gives the best SNR. From the equation (1), the dependency of M.sub.o on the light input level is expressed as follows.

Therefore,

Usually, x is selected to be a value ranging from 0.3 to 1.

The gain G of the amplifier corresponding to M.sub.o is automatically determined from the condition that the output of the amplifier is maintained at a constant level and the determined gain is the optimum value. Here, since there is a difference between the M which keeps an electrical signal at a constant level and said optimum value M.sub.o, the SNR deteriorates if only an electrical signal level is kept at a constant level. FIG. 3 (1), (2) and (3) show such conditions. Namely, the figures show the relation between light input and SNR corresponding to the methods (a), (b) and (c). The solid line in the figure shows the ideal values, while the broken lines 32, 33, 34 respectively show the values of the methods (a), (b) and (c). Explained below is the reason why the solid line 31 and broken lines 32, 33, and 34 cross at the point P.sub.0, and the solid line 31 and the broken lines 33 and 34 cross at the point P.sub.3. As shown in FIG. 2 (1), (2), (3), the value of M crosses the optimum value M.sub.o at the points P.sub. 0 and P.sub.3. At the point P.sub.0, M becomes equal to M.sub.o (M=M.sub.o) by the setting, while at the point P.sub.3, M becomes also equal to M.sub.o. In a method being employed currently in order to improve the SNR, a value of M is not lowered to M.sub.min but instead is always kept at a value which is larger than than the value M.sub.1 (>M.sub.min). Thereby, before the SNR deviates largely from the value SNR.sub.o for the condition M=M.sub.o, M is limited by M.sub.1 and the SNR is improved. This is shown in FIG. 4 with an example of the method (b). In the same figure, the solid lines 41, 44 and 47 show the ideal conditions, while the broken lines 42, 45, 48 show the conditions of the method (b), and the chained lines 43, 46, and 49 show the condition where M is larger than M.sub.1. However, the method where M is kept larger than M.sub.1 also has the following disadvantage. Namely, since the minimum value of M becomes M.sub.1 (>M.sub.min), the additional gain variation width required is as much as (M.sub.1 /M.sub.min) in order to obtain the desired dynamic range. As can be understood from FIG. 4, when M>M.sub.min, the amplifier gain width required only ranges from G1 to G2, but when M>M.sub.1, it must range from G1 to G.sub.3 (G.sub.2 >G.sub.3). As proposed in the unexamined Japanese patent application No. 53-90802, it is possible to simultaneously control the M of the APD and the gain of the amplifier in combination so that M of the APD always satisfies the equation (3) as shown in FIG. 2 (4) for variations of the light input level, but it has a disadvantage in that the control circuit is complicated.

With consideration to the abovementioned existing method, it is an object of the present invention to provide a light receiving circuit which assures a large light input dynamic range, based on the idea that the optimum values for M and amplifier gain for obtaining the optimum SNR are determined by the polygonal line approximation method, without allowing the SNR to largely deviate from the optimum value and without widening the desired variable gain range of the amplifier. This object is attained by a light receiving circuit wherein the amplifier output is kept at a constant by automatic gain control of the multiplication factor of an avalanche photo diode which converts a light signal into an electrical signal and automatic gain control of the amplifier which amplifies the output of the avalanche photo diode, thus characterized in that the circuit is provided with a first comparing means which compares a voltage proportional to the voltage of the multiplication factor control terminal of the avalanche photo diode and a first reference voltage, a second comparating means which compares the output of the amplifier and a second reference voltage, a third comparing means which compares a third reference voltage and a control voltage for automatic gain control of the amplifier, a high voltage generator circuit which supplies an output to the multiplication factor control terminal of the avalanche photo diode, and a switch device which selectively changes over the outputs of the first and second comparing means outputs of first and second comparing means are connected to the switch device, the output of the switch device is connected to the input of the high voltage generator circuit, output of the second comparing means is coupled to the control voltage input terminal for automatic gain control of the amplifier and the output of the third comparing means is coupled to the first reference voltage generator circuit, and thereby a voltage output which gives a multiplication factor which approximates (by a polygonal line) the multiplication factor of an avalanche photo diode satisfying the optimum signal to noise ratio is obtained at the multiplication factor control terminal of said avalanche photo diode, while a voltage output which makes the output voltage of the amplifier constant in accordance with the multiplication factor can be obtained at said control voltage input terminal for automatic gain control of the amplifier.

FIG. 1 is a block diagram for explaining a light receiving circuit;

FIG. 2 indicates the relation between the multiplication factor M of APD and the gain G of the amplifier as a function of light input for various conventional light receiving circuit methods;

FIG. 3 indicates the relation between light input and SNR in three methods of FIG. 2;

FIG. 4 indicates the relation between light input vs SNR, M and G when the value of M is kept larger than M.sub.1 a method used in conventional light receiving circuits;

FIG. 5 is the schematic diagram of an embodiment of the light receiving circuit of the present invention;

FIG. 6 indicates the voltages, M and G as a function of light input for the circuit shown in FIG. 5;

FIG. 7 indicates the relation between SNR, M and G as a function of light input for the circuit shown in FIG. 5;

FIG. 8 indicates the relation between V.sub.APD and M of an APD;

FIG. 9 indicates the relation between V.sub.AGC and G of an amplifier;

FIG. 10 indicates the relation between the input voltage V.sub.7 and the output voltage V.sub.APD of the high voltage generator circuit;

FIG. 11 shows another embodiment of the present invention;

FIG. 12 explains the of the circuit of FIG. 11;

FIG. 13; shows another embodiment of the present invention;

FIG. 14 explains of the circuit of FIG. 13.

The preferred embodiment of the present invention will be explained by referring to FIG. 5 to FIG. 14. FIG. 5 is a circuit diagram of the embodiment of the present invention. A light signal 11 is applied to the APD 12, the an APD output of which is connected is amplified by the amplifier 13 and output to the output terminal 15. The output of the amplifier 13 is also connected to the level detector 14 and the output voltage V.sub.l of the level detector 14 is applied to the inverting input of the second comparator 52 consisting of a differential amplifier. To the non-inverting input of the second comparator 52, a second reference voltage V.sub.ref2 is applied. The output voltage V.sub.2 of the second comparator 52 is connected to the AGC control terminal input of the amplifier 13 via the resistor 58, supplying the AGC control voltage V.sub.AGC thereto. The output of the second comparator 52 passes a switch device which includes the diode 56, supplying an input voltage V.sub.7 to the high voltage generator circuit 54. The output voltage V.sub.APD of the high voltage generator circuit 54 applied as an input to the APD 12, and simultaneously V.sub.APD is divided into a voltage V.sub.6 via the resistor 59. The voltage V.sub.6 is being applied to the inverting input of the first comparator 51, consisting of a differential amplifier. To the non-inverting input of the first comparator 51, a voltage V.sub.4 obtained by dividing a first reference voltage V.sub.ref1 is applied. The output of the first comparator 51 is applied as an input to the high voltage generator circuit 54 after passing a switch device which includes the diode 55. The voltage V.sub.AGC is supplied to the non-inverting input of the third comparator 53, consisting of a differential amplifier, while a third reference voltage V.sub.ref3 is applied to the inverting input. The output voltage V.sub.5 of the third comparator 53 determines a voltage V.sub.4 in combination with the first reference voltage V.sub.ref1.

The operations of the above circuit will now be explained by referring to FIG. 6. Starting with a low light input level, M is adjusted to the optimum value and the gain G of amplifier 13 is at the maximum. When the light input gradually increases, the voltage V.sub.l rises as shown in the figure. As a result, the output voltage V.sub.2 of the second comparator is gradually reduced. Since the voltage V.sub.2 is positive, it is applied to the high voltage generator circuit 54, passing the diode 56, and thereby a voltage V.sub.APD is obtained as the output. The voltage V.sub.APD is also reduced as the voltage V.sub.2 reduces. Thus, M of the APD is also reduced. The voltage V.sub.APD falls until the voltage V.sub.6 divided from the voltage V.sub.APD reaches the voltage V.sub.4 divided from the first reference voltage V.sub.ref1. When the light input level is given as P.sub.2 ', the voltage V.sub.6 reaches the voltage V.sub.4, and the output of the first comparator becomes positive and thereby turns diode 55 ON, forming a loop through the first comparator 51 and the high voltage generator circuit 54, to keep a V.sub.APD at a constant. Therefore, the value of M becomes a constant. As was explained above, the voltage V.sub.2 is positive until the light input level reaches the point P.sub.2 ' in FIG. 6 and is clamped by the diode 57. However, when the light input increases further, causing the voltage V.sub.l to rise and the voltage V.sub.2 to fall to a negative voltage, the voltage V.sub.AGC is reduced and thereby the gain G of the amplifier 13 also falls. When the voltage V.sub.2 further and the light input level reaches P.sub.3, the third reference voltage V.sub.ref3 is reached. Thus, the output voltage V.sub.5 of the third comparator 53 changes from a positive to negative, causing the voltage V.sub.3 to change to a negative. When the light input level reaches the point P.sub.4, the output voltage V.sub.5 is saturated, and the voltage V.sub.3 is lowered until the voltage V.sub.5 reaches a negative saturation level. As a result the voltage V.sub.4 falls accordingly until it reaches a value corresponding to the saturation level of the voltage V.sub.5. Thereby, the V.sub.APD and the value of M fall in the same way, and when M reaches a constant value corresponding to the saturation level of V.sub.5, they are fixed by this value. When the value of M is fixed at a constant value, the V.sub.AGC is further reduced and the gain G is also lowered. In case the light input level is within the range from P.sub.3 to P.sub.4, the gain G falls since the magnitude of V.sub.AGC falls, but it is as small as 1/(gain of comparator 53) as compared with the variation of M, and therefore it is indicated as a constant value for simplification of the explanation. FIG. 7 uses solid lines 71, 73 and 75 to show the optimum values of SNR, M and G as a function of light input, and it also shows the approximate values obtained by the light receiving circuit of the above-described embodiment of the present invention with broken lines 72, 74 and 76. FIG. 8 shows a graph indicating an example of the relation between V.sub.APD and M of APD 12. FIG. 9 is a graph indicating an example of the relation between V.sub.AGC and the gain G of amplifier 13. FIG. 10 shows a graph indicating the relation between the input voltage V.sub.7 and output voltage V.sub.APD of the high voltage generator circuit 54.

In the above embodiment, the multiplication factor M of the APD approximates the optimum value M.sub.0 with a polygonal line having three turning points (the gain G, automatically approximates the ideal condition), but it is also possible to obtain a modification of the embodiment which better approximates the optimum value of the SNR by additionally providing in parallel a plurality of the third comparators illustrated in FIG. 11 and by suitably setting the reference voltages corresponding to the third reference voltage.

Namely, the comparators 53-1, 53-2 to 53-n are provided in parallel to the comparator 53, and the reference voltages V.sub.ref4, V.sub.ref5 to V.sub.refn are applied to the inverting inputs while V.sub.AGC is applied to the non-inverting inputs. The outputs of the comparators 53-1 to 53-n are connected with the resistors R.sub.4, R.sub.5 to R.sub.n, diodes D.sub.4 to D.sub.n, and diodes D.sub.4-1 to D.sub.n-1. Here, the voltages V.sub.ref3, V.sub.ref4, V.sub.ref5 to V.sub.refn are related as follows V.sub.ref3 >V.sub.ref4 >V.sub.rev5 . . . >V.sub.refn, while the resistors R.sub.3, R.sub.4, R.sub.5 to R.sub.n are related as follows R.sub.3 >R.sub.4 >R.sub.5 .about.>R.sub.n. The operation will be explained below.

The steps up to saturation of the output voltage of the comparator 53 in FIG. 11 are the same as those of the circuit shown in FIG. 5. When the value of V.sub.AGC falls lower than the reference voltage V.sub.ref4 as shown in FIG. 12, the output voltage V.sub.5-1 of the comparator 53-1 becomes negative and thereby the voltage V.sub.4 is also lowered. Since the voltage V.sub.4 is lowered until the output voltage V.sub.5-1 is saturated, the value of V.sub.APD is also lowered in accordance with the voltage V.sub.4. As a result, M is also lowered until the output voltage V.sub.5-1 is saturated. M maintains a constant value until the voltage V.sub.AGC becomes lower than the reference voltage V.sub.ref5 after the output voltage V.sub.5-1 is saturated and thereafter it is further lowered in the same way as mentioned above. The chained line in FIG. 12 indicates the optimum values.

FIG. 13 shows another embodiment of the present invention. The circuit indicated has the same functions as those of the circuit shown in FIG. 5. The only difference from the circuit of FIG. 5 is that the comparator 51 and associated elements diode 55, and resistor 59 are removed, and the output of the high voltage generator circuit 54 is connected to a resistor which is connected to the reference power supply V.sub.ref1 via the diode D.sub.1 and another resistor.

The circuit operation will be explained by referring to FIG. 14. When the light input level is low, the voltage V.sub.l is lower than the reference voltage V.sub.ref2. Therefore, the output voltage V.sub.2 of the comparator 52 is positive and is applied to the high voltage generator circuit 54 as the voltage V.sub.7 via the diode 56. When the light input level becomes high, the voltage V.sub.7 is lowered and also the voltage V.sub.APD becomes low. When the light input level reaches P.sub.2 ', the voltage V.sub.l becomes equal to the reference voltage V.sub.ref2, the output voltage V.sub.2 becomes negative, the diode 56 turns OFF and the voltage V.sub.APD becomes equal to the voltage V.sub.4, the voltage V.sub.APD being clamped to V.sub.4. Thereby, M becomes constant and the gain G is lowered since the voltage V.sub.AGC becomes low. When the light input exceeds the point P.sub.3, the voltage V.sub.AGC becomes lower than the reference voltage V.sub.ref3. Thus, the output voltage V.sub.5 becomes negative and the voltage V.sub.4 becomes low. Thereby, the voltage V.sub.APD also becomes low, according to the voltage V.sub.4, and M falls again. When the light input reaches P.sub.4, the output voltage V.sub.5 is saturated and thereafter M becomes constant. The voltage V.sub.AGC becomes low and the gain G also becomes small. Thus, M can approximate the optimum values indicated by the chained line.

As will be obvious from the above explanation, the present invention provides a light receiving circuit with a simple structure which can assure a wide light input dynamic range without allowing the SNR to deviate largely from the optimum value and without widening the desired variable gain width of the amplifier.