Apparatus and method for reception of optical signal

[0002] 1. Field of the Invention

[0003] The present invention relates to an optical signal receiving apparatus and method.

[0004] Along with the development of multimedia technology, optical communication using optical fibers has been rapidly spreading. Particularly, while optical communication hithertofore had mainly spread at the trunk line side, in the future optical communication will mainly spread at the subscriber side for realizing “fiber to the home” (FTTH) systems. The “subscriber side” referred to here includes not only individual subscribers (users), but also local area networks (LANs) etc.

[0005] In for example a passive optical network (PON) transmission system, the subscriber side is constituted by an optical splitter (star coupler) connected to the office side equipment via a common optical fiber transmission line and a plurality of units of subscriber side equipment individually connected to this optical splitter via individual optical fiber transmission lines. Here, downstream optical transmission from the office side equipment to each subscriber side equipment is carried out by continuously transmitting cell signals, while upstream optical transmission from the plurality of units of subscriber side equipment to the office side equipment is carried out by using time slots allotted to the units of subscriber side equipment in advance and transmitting the cell signals of each subscriber in a burst-like manner in a time division mode in units of cells each having a fixed bit length.

[0006] Looking at the latter upstream optical transmission, the office side equipment receives the bursts of cell signals from the optical transmission apparatuses of the units of subscriber side equipment. The present invention relates to an optical signal receiving apparatus which receives the burst-like cell signals in the office side equipment. This optical signal receiving apparatus receives cell signals having different levels of received light for every time slot due to the difference of transmission loss of the individual optical fiber transmission lines. Therefore, the automatic gain control (AGC) for this becomes important subjects.

[0007] 2. Description of the Related Art

[0008] As will be explained in detail below referring to the drawings, there are the following problems in the related art.

[0009] Since the optical signal receiving apparatus receives the cell signals having different amplitude levels in bursts for every subscriber, the optical signal receiving apparatus has to instantaneously determine the optimum detection state for each amplitude of the incoming cell signals, that is, the optimum gain.

[0010] However, the optical signal receiving apparatus must be given a considerably large range of variation of gain control in order to cope with burst-like cell signals having a wide dynamic range. As a result, the instantaneous response of the optical signal receiving apparatus to the burst-like cell signals is deteriorated. In order to deal with this deterioration of the instantaneous response, it is necessary to secure a transient time region (period) until the gain stabilizes at the optimum value.

[0011] In the state where no cell signal has yet arrived, for example, the optical signal receiving apparatus waits on standby at the maximum gain. When a cell signal arrives during this standby state, the optical signal receiving apparatus first amplifies the input signal with its maximum gain and then lowers the gain if the level of the amplified signal is larger than the optimum value. It then waits for stabilization of the gain. Usually, at least 10 bits worth of time is necessary until the gain is stabilized. When the dynamic range of the cell signals becomes larger, the range of variation of the gain is enlarged and therefore the number of bits must be further increased. That is, the transient response region is increased.

[0012] Due to the above explained control of the gain at every time of reception of signal, there is sometimes an overshoot at each rising edge of the amplified signal from the later explained variable gain amplifier circuit. During the occurrence of this overshoot, the signal is not the inherent amplified signal, so it is difficult to correctly discriminate between the logics “1” and “0” at the following discrimination circuit.

[0013] The region of above difficulty of discrimination is the follow-up period up to the stabilization to the optimum gain, that is, a training period. In the related art, each cell signal has a data region for the training added to its front position as a header. A data region added as a header in this way does not contribute at all to the region of the cell signals for transfer of data as inherent information. In the end, therefore, there is a problem of a reduction of the transmission efficiency of the transmission system as a whole.

[0014] Accordingly, in consideration with the above problems, an object of the present invention is to provide an apparatus and method for detection of an optical signal capable of causing the gain to quickly stabilize at the optimum gain without accompanying transient response (overshoot) at the time of reception of each cell signal.

[0015] To attain the above object, the present invention provides an optical signal receiving apparatus capable of substantially instantaneously and stably discriminating the logic from the header of a burst-like optical signal. This apparatus comprises a pre-amplifier circuit for amplifying an output from a photodiode receiving cell signals, a dummy side variable gain amplifying unit for performing the gain control by feedback control with respect to the output from the pre-amplifier circuit and outputting the gain at that time as gain information, a memory unit for holding the gain information, and a signal side variable gain amplifying unit for performing the gain control by feedfoward control by using the gain set by the gain information and outputting the amplified signal.

[0016] The above object and features of the present invention will be more apparent from the following description of the preferred embodiments with reference to the accompanying drawings, wherein:

[0017] FIG. 1 is a view of the basic configuration of the present invention;

[0018] FIG. 2 is a view for explaining the principle of the present invention;

[0019] FIG. 3 is a step diagram showing a method for reception of an optical signal according to the present invention;

[0020] FIG. 4 is a view more concretely showing amplifying units 21 and 23 shown in FIG. 1;

[0021] FIG. 5 is a view of a further concrete first embodiment of the circuit configuration shown in FIG. 4;

[0022] FIG. 6 is a view of a second embodiment according to the present invention;

[0023] FIG. 7 is a view for explaining a discrimination level generating means 51, in which (A) and (B) are input waveform and output waveform diagrams of a variable gain amplifier circuit 44;

[0024] FIG. 8 is a first part of a view of a third embodiment according to the present invention;

[0025] FIG. 9 is a second part of a view of the third embodiment according to the present invention;

[0026] FIG. 10 is a waveform diagram for explaining a DC level deviation of differential outputs from the variable gain amplifier circuit;

[0027] FIG. 11 is a waveform diagram for explaining a low frequency level variation;

[0028] FIG. 12 is a first part of a view of a fourth embodiment according to the present invention;

[0029] FIG. 13 is a second part of the view of the fourth embodiment according to the present invention;

[0030] FIG. 14 is a first part of a view of a modified fourth embodiment according to the present invention;

[0031] FIG. 15 is a second part of the view of the modified fourth embodiment according to the present invention;

[0032] FIG. 16 is a first part of a view of a fifth embodiment according to the present invention;

[0033] FIG. 17 is a second part of the view of the fifth embodiment according to the present invention;

[0034] FIG. 18 is a view of an example of the variable gain amplifier circuit;

[0035] FIG. 19 is a view schematically showing an example of the transmission system to which the present invention is applied;

[0036] FIG. 20 is a view of the configuration of an optical signal receiving apparatus of the related art;

[0037] FIG. 21 is a view of an example of the waveform of an input signal S_inapplied to an AGC stage;

[0038] FIG. 22 is a waveform diagram of signals of principal parts in FIG. 20, in which (A) and (B) are views of an input waveform to and output waveform from a variable gain amplifier circuit 14, and (C) is a view of the output waveform of a discrimination circuit 19; and

[0039] FIG. 23 is a view for explaining the problems of the related art, in which (A) is a view of the cell signals CL, (B) is a view of the gain, and (C) is a view of an amplified signal Sa.

[0040] Before describing the embodiments of the present invention, the related art and the disadvantages therein will be described with reference to the related figures.

[0041] FIG. 19 is a view schematically showing an example of a transmission system to which the present invention is applied, that is, the PON transmission system explained before.

[0042] In the figure, 1 denotes office side equipment which constitutes a trunk line side in cooperation with the other office side equipment 1 shown at the right side in the figure. On the other hand, the equipment 1 constitutes a subscriber side in cooperation with a plurality of units of subscriber side equipment (#1, #2 . . . #n) 5 shown at the left side in the figure. The present invention relates to the latter subscriber side.

[0043] At the subscriber side, a downstream optical transmission is performed from the office side equipment 1 to each subscriber side equipment 5, while the reverse upstream optical transmission is performed from each subscriber side equipment 5 to the office side equipment 1. The present invention relates to the latter upstream optical transmission.

[0044] This upstream optical transmission is carried out by transmitting cell signals CL from an optical transmission unit (indicated by electrical/optical converter “E/O” in the figure) 6 in each subscriber side equipment 5 and a wavelength division multiplexer (WDM) 7 provided with a filtering function for preventing mutual interference with the downstream optical transmission. Cell signals CL pass through individual optical fiber transmission lines 8, the optical splitter (star coupler SC) 4, and a common optical fiber transmission line 9 common to the plurality of units of subscriber side equipment 5 to reach the office side equipment 1. On this common optical fiber transmission line 9, cell signals CL from the units of subscriber side equipment 5 are transferred in bursts. Note that, in the downstream transmission from the office side equipment 1 to the subscriber side equipment 5, continuous cell signal trains are sent through the transmission lines 9 and 8.

[0045] In the upstream optical transmission, the burst-like cell signals CL reaching the office side equipment 1 are received at the receiving apparatus (indicated by the optical/electrical converter “O/E”) 2 via the wavelength division multiplexer (WDM) 3 in the office side equipment 1. There, the logics “1” and “0” of the data contained in each cell signal CL is discriminated. The present invention will describe this optical receiving unit 2. This will be referred to as the “optical signal receiving apparatus”. In this apparatus, the optical cell signals are received in bursts.

[0046] FIG. 20 is a view of the configuration of an optical signal receiving apparatus of the related art. Optical cell signals CL from the subscriber side are first received at a photodiode 11 located at an input stage of the optical signal receiving apparatus 10, converted to electrical cell signals there, and further pass through a pre-amplifier circuit 12. Automatic gain control is applied to the amplitude of each cell signal at an intermediate AGC stage resulting in cell signals having a substantially constant amplitude. The discrimination circuit 19 located at an output stage of the optical signal receiving apparatus 10 discriminates the logics “1” and “0” forming each cell signal and provides a complementary digital output (“1”/“0”). Note that this digital output is given to for example a latter retiming circuit (not illustrated) for establishing bit synchronization.

[0047] FIG. 21 is a view of an example of the waveform of the input signal S_inapplied to the AGC stage. Cell signals from the units of subscriber equipment 5 (#1, #2 . . . #n) shown in FIG. 19 have different transmission losses as mentioned above, therefore the magnitude of amplitude of each is different. In this way, even if the discrimination circuit 19 discriminates the logics “1” and “0” of the input signals S_in, each having different amplitudes, by a fixed discrimination level (threshold) from a discrimination level generating circuit 18, correct discrimination cannot be carried out.

[0048] Therefore, in FIG. 20, the input signals S_infrom the pre-amplifier circuit 12 are all made to have the same amplitude for all of input signals from the subscribers so as to enable correct discrimination. This is done by the AGC stage.

[0049] The AGC stage, as shown in FIG. 20, is constituted by a variable gain amplifier circuit 14 receiving the input signal S_in, a peak detection circuit 15 forming a feedback loop with respect to this circuit 14, a gain control amplifier circuit 16, and a gain reference generating circuit 17.

[0050] The peak detection circuit 15 constantly monitors the amplified signal Sa from the variable gain amplifier circuit 14 and detects the peak value thereof. In order to make this detected peak value P converge to the gain reference Rg given from the gain reference generating circuit 17, a gain control amplifier circuit 16 is provided. This circuit 16 controls the gain of the variable gain amplifier circuit 14 based on the peak value P and the gain reference Rg. An AGC feedback loop is formed here.

[0051] FIG. 22 shows the waveforms of signals of principal parts in FIG. 20, in which (A) and (B) are views of the input waveform and output waveform of the variable gain amplifier circuit 14 and (C) is a view of the output waveform of the discrimination circuit 19. Note that the waveforms of the parts <1>, <2>, . . . <9> in FIG. 20 are indicated by the corresponding numbers <1>, <2>, . . . <9> in FIG. 22.

[0052] The waveform <1> in (A) of FIG. 22 is the waveform of the input signal S_inand the inverted output waveform of the photodiode 11. A DC level generating circuit 13 supplies the level <2> substantially equal to the “0” level of this waveform <1> to the variable gain amplifier circuit.

[0053] The variable gain amplifier circuit 14 receiving the inputs shown in the above <1> and <2> outputs the complementary amplified signal Sa shown in (B) of the same figure as the waveforms <3> and <4>. The peak value P of the amplified signal Sa having the waveform <3> is detected at the peak detection circuit 15. This is the waveform <5>. This peak value P is controlled via the variable gain control amplifier circuit 16 so that the difference Δp from the gain reference Rg from the gain reference generating circuit 17 indicated by the waveform <6> always converges to 0. Namely, AGC feedback control is carried out so that the peak value P coincides with the gain reference Rg. Thus, the input signals S_inall become the amplified signal Sa maintained at a constant amplitude irrespective of from which subscriber the signal is sent.

[0054] The logics “1” and “0” contained in the amplified signal Sa are discriminated at the discrimination circuit 19. The result becomes as shown in (C) of the same figure. Note that an alternating pattern of “1” and “0” is shown as one example.

[0055] FIG. 23 is a view for explaining the problems of the related art mentioned above, and in which (A) is a view of the cell signals CL, (B) is a view of the gain, and (C) is a view of the amplified signal Sa. Namely, (A) shows the input waveform to the photodiode 11, (B) a gain of the variable gain amplifier circuit 14, and (C) the output waveform of the circuit 14. Note that (A) is equivalent to FIG. 21 explained above.

[0056] As shown in (A) of FIG. 23, the optical signal receiving apparatus 10 receives cell signals CL having different amplitude levels for every subscriber in bursts, therefore the apparatus 10 is required to instantaneously determine the optimum state of reception for each amplitude of incoming cell signals CL, that is, the optimum gain.

[0057] As explained above, however, in order to handle burst-like cell signals having a wide dynamic range, the range of variation of the gain control in the optical signal receiving apparatus 10 must be made considerably large. As a result, the instantaneous response with respect to the burst-like cell signals of the optical signal receiving apparatus 10 is deteriorated. In order to deal with such a deterioration of the instantaneous response, it is necessary to secure a transient time region (period) until the gain is stabilized to the optimum value.

[0058] Referring to (B) of FIG. 23, in the state where no cell signal has yet arrived, for example, the optical signal receiving apparatus 10 waits on standby at the maximum gain (“maximum gain” in the figure). When a cell signal CL arrives in this standby state, the optical signal receiving apparatus 10 first amplifies the input signal S_inwith its maximum gain and then lowers the gain if the level of the amplified signal Sa is larger than the optimum value. Then, it waits for stabilization of the gain. Usually, at least 10 bits worth of time is necessary until the gain is stabilized. When the dynamic range of the cell signals becomes larger, the range of variation of the gain is enlarged and therefore the number of bits must be further increased. That is, the transient response region is increased. Note that, in (B) of FIG. 23, in a state where no cell signal CL has yet arrived, the case of waiting on stand-by at a maximum gain is shown, but conversely it is also possible to wait on stand-by at the minimum gain. In any case, each time the reception of a cell signal is completed, the gain is set again to the maximum or minimum gain (refer to “reset” of (B) in the figure).

[0059] Since the gain is controlled at the time of each signal reception as explained above, there is sometimes an overshoot (OS), shown in (C) of FIG. 23, at each rising edge of the amplified signal Sa from the variable gain amplifier circuit 14. During the period when this overshoot occurs, the signal is not the inherent amplified signal Sa, so it is difficult to correctly discriminate between the logics “1” and “0” at the following discrimination circuit 19 (“discrimination difficulty” in the figure).

[0060] The region of the above discrimination difficulty is a follow-up period until stabilization to the optimum gain, that is, a training period. In the related art, each cell signal CL has a data region for the training added to its front position as a header. The data region added as the header in this way does not contribute at all to the region of the cell signals CL for transfer of data as inherent information. In the end, there is the aforementioned problem of a reduction of the transmission efficiency of the entire transmission system.

[0061] Accordingly, the present invention provides an apparatus and method for reception of an optical signal capable of quickly stabilizing the gain to the optimum gain without accompanying transient response (overshoot) at the time of reception of each cell signal.

[0062] FIG. 1 is a view of the basic configuration of the present invention. Note that similar constituent elements are indicated by the same reference numerals or symbols throughout all of the figures. Accordingly, the constituent elements introduced by the optical signal receiving apparatus 20 according to the present invention shown in the figure are a dummy side variable gain amplifying unit 21, a memory unit 22, and a signal side variable gain amplifying unit 23.

[0063] This optical signal receiving apparatus 20 comprises the following constituent elements as a whole. Namely, it comprises:

[0064] a pre-amplifier circuit 12 for amplifying the output from a photodiode 11 receiving the cell signal CL transmitted from each subscriber;

[0065] a dummy side variable gain amplifying unit 21 for receiving the output, as an input signal S_in, from the pre-amplifier circuit 12, performing a gain control with respect to the output by feedback control and outputting the gain control signal, as the gain information Ig, when a dummy amplified signal having a predetermined output amplitude is generated;

[0066] a memory unit 22 for holding the gain information Ig; and

[0067] a signal side variable gain amplifying unit 23 for receiving the output, as an input signal S_in, from the pre-amplifier circuit 12, performing a gain control with respect to the output by feedfoward control by using the gain set by the gain information Ig from the memory unit 22 and outputting the amplified signal Sa having the predetermined output amplitude.

[0068] FIG. 2 is a view explaining the principle of the present invention, in which (A) shows the cell signals, (B) shows the dummy side gain, (C) shows the signal side gain, and (D) shows the amplified signal. Note that, (A), (B), and (D) of the figure correspond to (A), (B), and (D) of FIG. 23 mentioned above.

[0069] Assume that the cell signals CL as shown in (A) of FIG. 2 are transmitted from the subscribers (#1, #2 . . . #n) at timings (time slots) assigned to respective subscribers. The amplitude of each cell signal CL is different due to individual transmission loss of each subscriber.

[0070] When such a cell signal CL is once amplified at the pre-amplifier 12 to obtain the input signal S_inand then the amplified signal Sa is produced which has the predetermined output amplitude at the variable gain amplifying circuit 14 of the related art, as shown in (C) of FIG. 23, an overshoot OS occurs making correct discrimination of the logic difficult. Therefore, it is required to quickly stabilize the gain to the predetermined optimum gain.

[0071] (B) of FIG. 2 shows the variation of the gain Gd in the dummy side variable gain amplifying unit 21 shown in FIG. 1. This is exactly the same as the variation of the gain in the optical signal receiving apparatus 10 of the related art explained by referring to (B) of FIG. 23. Here, the reason for use of the term “dummy side” is that the amplified signal (dummy amplified signal) produced in this variable gain amplifying unit 21 is not used as the signal to be discriminated in the following discrimination circuit 19, i.e., it is not utilized as an inherent amplified signal.

[0072] However, the gain information Ig concerning the gain Gd obtained at this dummy side variable gain amplifying unit 21, particularly, a stable gain Gd obtained near the end of each cell signal CL, is effectively utilized as the gain information Ig. For example, for the cell signal CL from the subscriber #1 of (B) of FIG. 2, the stable gain Gd is picked up as indicated by the broken line circle in (B) in the latter half of the cell signal and stored and held in an area corresponding to the subscriber #1 in the memory unit 22 as the gain information Ig.

[0073] In the PON transmission system, however, the office side equipment 1 has predetermined sequence information. This sequence information determines the sequence of from which subscriber (#1, #2 . . . #n) and at which timing (time slot) the cell signal CL is to be transmitted.

[0074] Then, as shown in (A) of FIG. 2, it is known from the sequence information that the optical signal receiving apparatus 20 receives the cell signal CL from for example the subscriber #1 at the timing shown at the left end in the figure and receives the next cell signal CL from the subscriber #1 at the timing shown at the right end in the figure. Therefore, since it is seen that the cell signal CL from the subscriber #1 is received at the timing shown at the right end in the figure, as shown in (C) of the figure, at the start of the timing, the stabilized gain Gd shown in (B) of the figure learned at the time of reception of the cell signal CL received immediately before, is read from the memory unit 22 as the gain information Ig. This Gd is set as the gain of the signal side variable gain amplifying unit 23 (refer to “set” of (C) of the figure).

[0075] Thus, as shown in (C) of FIG. 2, the gain for obtaining the amplified signal Sa having the predetermined output amplitude in the optical signal receiving apparatus 20 becomes substantially the optimum gain from the start of each burst. Here, looking at the same subscriber, note is taken of the fact that the optimum gain obtained at immediately preceding burst timing does not change so much from the optimum gain which will be obtained at the current burst timing.

[0076] As a result, the obtained amplified signal Sa becomes as shown in (D) of FIG. 2. As seen here, the problem of the overshoot OS appearing at the start of each burst shown in (C) of FIG. 23 is solved.

[0077] The above principle of the present invention can be applied to a method of receiving an optical signal.

[0078] FIG. 3 is a step diagram of the method of receiving an optical signal according to the present invention. This includes the following three steps:

[0079] Step S1: Reception of the optical cell signal CL at the photodiode 11 and conversion to an electrical input signal S_in.

[0080] Step S2: Amplification of the input signal S_inwith a variable gain Gd by the feedback control, production of a dummy amplified signal having a predetermined output amplitude, and storage of gain information Ig indicating the gain.

[0081] Step S3: Amplification of the subsequently received input signal S_inwith the gain Gd indicated by the gain information Ig by feedfoward control and generation of an amplified signal Sa having a predetermined output amplitude.

[0082] The optical signal receiving apparatus 20 further has a discrimination circuit 19 which receives as its input the amplified signal Sa output from the signal side variable gain amplifying unit 23 as shown in FIG. 1, discriminates the logics “1” and “0” of this amplified signal Sa, and produces a digital output.

[0083] FIG. 4 is a view showing the amplifying units 21 and 23 shown in FIG. 1 in more detail. In the figure, the dummy side variable gain amplifying unit 21 comprises a dummy side variable gain amplifier circuit 31 which receives and amplifies the input signal S_inand a dummy side gain control amplifier circuit 41 for forming a feedback loop FB with respect to this dummy side variable gain amplifier circuit 31. It outputs the gain Gd given from this dummy side gain control amplifier circuit 41 to the dummy side variable gain amplifier circuit 31 to the memory unit 22 as the gain information Ig.

[0084] On the other hand, the signal side variable gain amplifying unit 23 comprises a variable gain amplifier circuit 33 which receives and amplifies the input signal S_inand a gain setting unit 43 which reads the gain information Ig held in the memory unit 22 and sets the gain Gs by a feedfoward (FF) control with respect to this variable gain amplifier circuit 33.

[0085] FIG. 5 is a view of a more specific first embodiment of the circuit configuration shown in FIG. 4.

[0086] First, looking at the dummy side variable gain amplifying unit 21, the dummy side variable gain amplifier circuit 31 of FIG. 4 and the dummy side gain control amplifier circuit 41 are disposed at positions indicated by reference numerals 31 and 41 in FIG. 5, respectively. As apparent when compared with the above FIG. 20, the dummy side variable gain amplifying unit 21 has exactly the same configuration as that of the AGC stage in the optical signal receiving apparatus 10 of the related art. Note that the dummy side variable gain amplifying unit 21 according to the present invention is configured to supply the gain information Ig to the signal side variable gain amplifying unit 23 side.

[0087] Next, looking at the signal side variable gain amplifying unit 23, the variable gain amplifier circuit 33 and gain setting unit 43 of FIG. 4 are disposed at positions indicated by reference numerals 33 and 43 in FIG. 5, respectively.

[0088] The variable gain amplifier circuit 33 is constituted by the variable gain amplifier circuit 44 comprising a differential amplifier circuit as illustrated. Further, the gain setting unit 43 comprises an A/D converter 45, a processor 46, and a D/A converter 47.

[0089] In more detail, the signal side variable gain amplifying unit 23 includes the memory unit 22 and, at the same time, has an A/D converter 45 for converting analog gain information Ig to digital gain information Ig, a processor 46 for writing the digital gain information Ig from this A/D converter 45 into the memory unit 22, and a D/A converter 47 for converting the gain information read from the memory unit 22 via this processor 46 to analog gain information.

[0090] The processor 46 accesses the memory unit 22 whenever a cell signal CL is received and, at the same time, reads the gain information Ig held immediately before for the related subscriber when receiving a cell signal CL from each subscriber according to the above sequence information held in it or the above sequence information given from the outside and sets the gain of the circuit 33 in the signal side variable gain amplifying unit 23.

[0091] FIG. 6 is a view of a second embodiment according to the present invention. The difference from the above first embodiment resides in that an automatic discrimination (threshold) level control means 51 for giving a discrimination level to discriminate the logics “1” and “0” of the amplified signal Sa, is added to the discrimination circuit 19. The automatic discrimination level control means 51 provides not a fixed discrimination level, but a variable discrimination level. More specifically, as shown in FIG. 6, it comprises a peak detection circuit 52 and a bottom detection circuit 53 for respectively detecting the maximum level and minimum level of the amplified signal Sa output from the signal side variable gain amplifying unit 23 and a voltage dividing circuit 54 for producing a substantially intermediate voltage of the output voltages from the peak detection circuit 52 and the bottom detection circuit 53 and using the voltage as the variable discrimination level. These form an ATC (automatic threshold control) circuit as a whole and enable a further improvement of the precision of the logic discrimination operation. This will be explained using FIG. 7.

[0092] FIG. 7 is a view for explaining the automatic discrimination level control means 51. (A) and (B) are views of the input waveform and output waveform of the variable gain amplifier circuit 44. Referring to (A) of the figure, there is sometimes a discrepancy in the amplitude of the input waveform (S_in) of the variable gain amplifier circuit 44 between the previous burst timing and current burst timing. This may be caused for example by an abrupt change of the loss of the transmission line and an abrupt change of the ambient temperature.

[0093] As shown in (B) of the figure, there is a discrepancy between the set output amplitude in the figure (output amplitude of the amplified signal Sa) set according to the previous gain information Ig and the actual output amplitude (current output amplitude). In the example of the figure, it shifts to a higher level. Needless to say a similar operation is performed even in a case where the amplitude shifts to a lower level.

[0094] In such a case, when discriminating the logic with the fixed discrimination level as shown in the first embodiment (FIG. 5), the level deviates (ΔTH) from the ideal discrimination level which is the center value of the amplitude (between peak and bottom). In the example of the figure, the ideal discrimination level is the discrimination level TH shifted upward by the amount of ΔTH.

[0095] In order to produce this ideal discrimination level TH, the peak detection level corresponding to the above peak and the bottom detection level corresponding to the above bottom are found at the peak detection circuit 52 and the bottom detection circuit 53 shown in FIG. 6, respectively. A predetermined divided value in voltage between the peak detection level and the bottom detection level is found at the voltage division circuit 54 to obtain the ideal discrimination level. The dividing value is preferably ½.

[0096] Note that at the point of time when the amplified signal Sa is output from the signal side variable gain amplifier circuit 44, the signal is sufficiently amplified. Also, the dynamic range of amplitude is small. Therefore, the error of the peak detection value is small. Further, this peak detection can be sufficiently started up in about 1 bit's worth of time, so there is almost no effect in lowering the transmission efficiency.

[0097] Further, in this second embodiment, while a pre-amplifier circuit of wide dynamic range (DR) is used as the pre-amplifier circuit 12, it is also possible to use a usual pre-amplifier circuit as in the first embodiment (FIG. 5).

[0098] Furthermore, common to all of the embodiments, provision is made of a DC level generating circuit 13 for applying a DC level substantially equal to an output level corresponding to the logic “0” from the pre-amplifier circuit 12 at one of the differential inputs of the dummy side variable gain amplifier circuit 14.

[0099] Further, provision is made of a DC level generating circuit 13 for applying a DC level substantially equal to an output level corresponding to the logic “0” from the pre-amplifier circuit 12 at one of differential inputs of the variable gain amplifier circuit 44.

[0100] Preferably, a single circuit is used commonly for these two DC level generating circuits.

[0101] FIG. 8 and FIG. 9 are first and second parts of a view of a third embodiment according to the present invention.

[0102] In the embodiment, a modification of the DC level generating circuit (13) is shown. The DC level generating circuit supplies a fixed reference voltage at one of differential inputs of the differential amplifier circuits forming the variable gain amplifier circuits 14 and 44. However, if the reference voltage is fixed, there is sometimes an inconvenience due to an offset peculiar to the differential amplifier circuit and error of the DC level generating circuit per se. This will be explained by referring to FIG. 10.

[0103] FIG. 10 is a waveform diagram for explaining the DC discrepancy of the differential outputs from the variable gain amplifier circuit. In a similar waveform diagram shown in FIG. 22, the level of the logic “0” output from one of the differential outputs of the gain amplifier circuit (14 and 44) and the level of the logic “0” output from the other coincide with each other. However, due to the offset or error, a DC discrepancy is sometimes produced between the above two logic “0” levels. This is indicated by ΔDC in the figure.

[0104] The third embodiment has the following configuration in order to minimize this DC discrepancy ADC to zero.

[0105] Provision is made of a dummy side DC feedback unit 61 connected to one of the differential inputs of the dummy side variable gain amplifier circuit 14. This is used for making the output level corresponding to the logic “0” from one of the differential outputs of this dummy side variable gain amplifier circuit 14 and the output level corresponding to the logic “0” from the other of the differential outputs of this dummy side variable gain amplifier circuit 14 coincide.

[0106] Further, provision is made of a DC feedback unit 63 connected to one of the differential inputs of the variable gain amplifier circuit 44. This is used for making the output level corresponding to the logic “0” from one of the differential outputs of this variable gain amplifier circuit 44 and the output level corresponding to the logic “0” from the other of the differential outputs of this variable gain amplifier circuit 44 coincide.

[0107] Both of the dummy side DC feedback unit 61 and the DC feedback unit 63 have the same configuration. In the figure, as one example, a configuration comprising bottom detection circuits (64, 67), peak detection circuits (65, 68), and DC control amplifier circuits (66, 69) for performing DC feedback so as to make the difference between outputs from these circuits zero is shown.

[0108] Note that, in FIG. 8, the bottom detection circuit 53 for the ATC circuit shown in FIG. 6 is commonly used for the bottom detection circuit 67 so as to simplify the circuit. Needless to say another bottom detection circuit can be used when the design of the time constant etc. is different.

[0109] Looking at the pre-amplifier circuit 12 in the third embodiment, an example where a wide dynamic range (DR) pre-amplifier circuit is also adopted is shown. The wide DR pre-amplifier circuit 55 avoids a saturation of the circuit 55 by turning on the transistor 56 connected to the input of the circuit 55 and by-passing the current when an excess input current is generated from the photodiode 11. Therefore, it can operate in a wide input dynamic range.

[0110] Next, the problem of low frequency level variation will be studied.

[0111] FIG. 11 is a waveform diagram for explaining the low frequency level variation. The explanation heretofore was predicated on the base level of the burst optical signal, that is, the base level of the input signal S_in, being constant. However, the base level is not constant but is sometimes accompanied by a low frequency level variation as shown in B of FIG. 11. This is because, for example, a low frequency monitor signal is sometimes sent from the office side equipment 1 so as to monitor the state etc. of each subscriber or measure a distance up to each subscriber. This is mixed with the burst optical signal and causes a variation of the level of the base level portion.

[0112] In order to deal with such a low frequency level variation, in the fourth embodiment, the configuration shown in FIGS. 12 and 13 is adopted.

[0113] FIG. 12 and FIG. 13 are first and second parts of a view of the fourth embodiment according to the present invention.

[0114] This fourth embodiment is characterized in that a base level detection circuit is introduced in place of the DC level generating circuit 13 of the first embodiment. This detects an output level corresponding to the logic “0” from the pre-amplifier circuit 12 and gives this to one of the differential inputs of the dummy side variable gain amplifier circuit 14. Then, it gives an output level corresponding to the logic “0” to one of the differential inputs of the variable gain amplifier circuit 44.

[0115] For this reason, it brings such an advantage that the base level detection circuit can effectively detect also the low frequency level variation. This will be explained in further detail.

[0116] The dummy side variable gain amplifier circuit 14 comprises a differential amplifier circuit receiving the input signal S_inat one of differential inputs. The input stage of this dummy side variable gain amplifier circuit 14 is provided with a base level detection circuit 81 which can detect the low frequency level variation contained in the base level of the input signal S_inand apply this to the other of the differential inputs of the differential amplifier circuit.

[0117] Further, the variable gain amplifier circuit 44 comprises a differential amplifier circuit receiving the input signal S_inat one of its differential inputs. The input stage of this variable gain amplifier circuit 44 is provided with a base level detection circuit 81 which can detect the low frequency level variation contained in the base level of the input signal S_inand apply this to the other of the differential inputs of the differential amplifier circuit.

[0118] Preferably, above one circuit is commonly used for the two base level detection circuits as illustrated. Note that the input signal S_inis the output of the pre-amplifier circuit 12, so has a waveform obtained by vertically reversing the waveform shown in FIG. 11. Accordingly, the base level detection circuit 81 is realized as for example a peak detection circuit detecting the peak of the input signal S_in. Note that, as mentioned above, the base level detection circuit gives an “0” level of the input signal S_inin place of the DC level generating circuit of the first embodiment, therefore effectively operates even in a case of no low frequency monitor signal.

[0119] Thus, a low frequency level variation (B of FIG. 11) is extracted from the base level detection circuit 81. If the variation level is used as each reference voltage of the variable gain amplifier circuits 14 and 44, the low frequency level variation component can be eliminated from the input signal S_in.

[0120] When the low frequency level variation is due to the low frequency monitor signal and it is necessary to extract this, the low frequency monitor signal is output through a sample holding circuit 82 and low-pass filter circuit 83 shown in FIG. 13. Note that, there is a possibility of disturbance of the signal waveform in the header of each cell signal (packet), therefore the signal is preferably input to the sample holding circuit (82) and low-pass filter circuit 83 at the latter half of each packet at which the signal waveform is stabilized as much as possible. When the disturbance of the waveform in the header is small, the sample holding circuit may be eliminated.

[0121] Further, the present embodiment shows an example of a configuration provided with a dummy side offset cancel feedback unit 71 and an offset cancel feedback unit 73. Note that these offset cancel feedback units may be used in the other embodiments as well except the third embodiment.

[0122] The dummy side offset cancel feedback unit 71 is connected to the offset cancel control input of the dummy side variable gain amplifier circuit 14 and used for making the output level corresponding to the logic “0” from one of the differential outputs of this dummy side variable gain amplifier circuit 14 and the output level corresponding to the logic “0” from the other of the differential outputs of this dummy side variable gain amplifier circuit 14 coincide.

[0123] Further, the offset cancel feedback unit 73 is connected to the gain control input of the variable gain amplifier circuit 14 and used for making the output level corresponding to the logic “0” from one of the differential outputs of this variable gain amplifier circuit 44 and the output level corresponding to the logic “0” from the other of the differential outputs of this variable gain amplifier circuit 44 coincide.

[0124] Note that the offset cancel control input will be shown in detail in the later FIG. 18.

[0125] In some cases, a DC discrepancy ΔDC similar to that shown in FIG. 10 is produced due to the output error of the base level detection circuit 81 and the offset peculiar to the variable gain amplifier circuits (14, 44). The offset cancel feedback units (71, 73) are introduced so as to reduce this to zero.

[0126] Both of the dummy side offset cancel feedback unit 71 and the offset cancel feedback unit 73 have the same configuration. In the figure, as one example, a configuration is shown comprising bottom detection circuits (74, 77), peak detection circuits (75, 78), and DC control amplifier circuits (76, 79) performing DC feedback so as to reduce the difference between outputs from these circuits to zero.

[0127] FIG. 14 and FIG. 15 are first and second parts of a view of a modified fourth embodiment according to the present invention.

[0128] In this modified fourth embodiment, the output of the DC control amplifier circuit 79 shown in FIG. 12 is fed back to one of the differential inputs of the differential amplifier circuit forming the variable gain amplifier circuit 44 by introducing a first resistor R1 and a second resistor R2 as shown in FIG. 14. Similarly. the output of the DC control amplifier circuit 76 shown in FIG. 13 is fed back to one differential input of the differential amplifier circuit forming the variable gain amplifier circuit 14 by introducing a third resistor R3 and a fourth resistor R4 as shown in FIG. 15.

[0129] FIG. 16 and FIG. 17 are first and second parts of a view of a fifth embodiment according to the present invention.

[0130] The characteristic feature of the fifth embodiment resides in the provision of a low frequency level variation eliminating means 91 for absorbing the low frequency level variation contained in the base level of the input signal S_inbefore the input signal S_inis input to the dummy side variable gain amplifying unit 31 and the signal side variable gain amplifying unit 33.

[0131] This low frequency level variation eliminating means 91 receives the input signal S_infrom the pre-amplifier circuit 12 directly or via the differential amplifier circuit 96 performing linear amplification as illustrated and negatively feeds back the low frequency level variation (B) to the input of the pre-amplifier circuit 12.

[0132] This low frequency level variation eliminating means 91 comprises a low frequency level variation detection unit 92 provided on the output side of the pre-amplifier circuit 12 and a first voltage controlled current source 93 directly connected to the photodiode 11 according to the illustrated example, and a control voltage is applied, in accordance with the low frequency level variation detected by the low frequency level variation detection unit 92, to the first voltage controlled current source 93 to draw a current in accordance with the low frequency level variation from the input side of the pre-amplifier circuit 12.

[0133] The low frequency level variation detection unit 92 is constituted by, as one example, a bottom detection circuit 97, a peak detection circuit 98, and a DC control amplifier circuit 99 for applying feedback so as to reduce the difference between the outputs from these circuits to zero.

[0134] Further, provision is made of a second voltage controlled current source 94 controlled by the control voltage in accordance with the low frequency level variation detected by the low frequency level variation detection unit 92 and a current/voltage transforming circuit 95 for generating a voltage proportional to the drawn current by this second voltage controlled current source 94. The low frequency level variation is extracted via a low-pass filter circuit 100 by this current/voltage transforming circuit 95. This low frequency level variation is just the low frequency monitor signal mentioned above.

[0135] FIG. 18 is a view of an example of the variable gain amplifier circuit. The above variable gain amplifier circuits (14, 44) can be constituted as shown in the figure. It comprises a variable gain amplifying unit 110 as a main part and an offset adjustment unit 120 and bias unit 130 attached to this. Note that the offset adjustment unit 120 is unnecessary for the first, second, third and the modified fourth embodiments shown in FIG. 14 and FIG. 15.

[0136] Referring to the variable gain amplifier circuit 14 shown in FIG. 12 as a representative case, the offset cancel control input receiving the output from the DC control amplifier circuit 79 is indicated as the offset cancel control input (+, −) 121 in FIG. 18.

[0137] The inherent gain control input from the gain control amplifier circuit 16 is indicated as the gain control inputs 111 and 112 of FIG. 18.

[0138] The differential inputs of the variable gain amplifier circuit 14 are indicated as signal inputs 113 and 114.

[0139] The main part of the variable gain amplifying unit 110 comprises a large gain differential transistor pair 115 and a small gain differential transistor pair 116.

[0140] The final amplified signal is taken out as Sa shown at the center of FIG. 18.

[0141] In the end, by controlling the amount of current (i in the figure) to be drawn from each source side of the differential transistor pairs 115 and 116 and controlling the amount of current (I in the figure) to be drawn from each drain side thereof, the gain can be set to an intended value.

[0142] As explained above, according to the present invention, when cell signals CL transmitted from the subscribers in bursts in units of packets are received at the office side equipment, the logic can be substantially instantaneously and stably discriminated from the header for any cell signal CL. Therefore, the training bit train added to the header of each packet in the related art becomes unnecessary, so the transmission efficiency of the entire transmission system is rapidly improved.