LIGHT RECEIVER HAVING GEIGER-MODE AVALANCHE PHOTODIODES AND METHOD FOR READING OUT

Abstract

A light receiver (50) is provided having a plurality of avalanche photodiode elements (10) that are each biased by a bias above a breakdown voltage and that are thus operated in a Geiger mode to trigger a Geiger current on light reception, wherein the avalanche photodiode elements (10) have a first connector (20, 22, 28a-b) and a second connector (20, 22, 28a-b); and wherein a first signal tapping circuit (12) for reading out the avalanche photodiode elements is connected to one of the connectors (20, 22, 28a-b). In this respect, a second signal tapping circuit (12) for reading out the avalanche photodiode elements (10) is connected to the other connector (20, 22, 28a-b).

Claims

1. A light receiver having a plurality of avalanche photodiode elements that are each biased by a bias above a breakdown voltage and that are thus operated in a Geiger mode to trigger a Geiger current on light reception, wherein the avalanche photodiode elements have a first connector and a second connector; wherein a first signal tapping circuit for reading out the avalanche photodiode elements is connected to one of the first and second connectors, and wherein a second signal tapping circuit for reading out the avalanche photodiode elements is connected to the other one of the first and second connectors.

2. The light receiver in accordance with claim 1, wherein at least one of the first signal tapping circuit and the second signal tapping circuit has an active coupling element, the active coupling element having an input that is connected to one of the first and second connectors and having an output; wherein the coupling element maps a Geiger current of the respective avalanche photodiode element at the input to a measured current corresponding in progression and level to the Geiger current in that the input for the Geiger current forms a virtual short circuit against a potential and the output is decoupled from the input.

3. The light receiver in accordance with claim 1, wherein the first connector is arranged on the anode side and the second connector is arranged on the cathode side of the respective avalanche photodiode element.

4. The light receiver in accordance with claim 1, wherein the avalanche photodiode elements are biased via the first connector and via the second connector.

5. The light receiver in accordance with claim 4, wherein a charge unit is respectively arranged between the avalanche photodiode element and the first connector or the second connector and the charge unit is bridged by a capacitor connected in parallel.

6. The light receiver in accordance with claim 1, wherein the avalanche photodiode elements have a third connector; with the first connector being configured for a capacitive decoupling of the Geiger current; and with the avalanche photodiode elements being biased via the second connector and the third connector.

7. The light receiver in accordance with claim 1, wherein the avalanche photodiode elements have a third connector and a fourth connector and a first charge unit is respectively arranged between the avalanche photodiode element and the third connector and a second charge unit is arranged between the avalanche photodiode element and the fourth connector; and wherein the avalanche photodiode elements are biased via the third connector and via the fourth connector.

8. The light receiver in accordance with claim 7, wherein the first charge unit is bridged to the first connector by a first capacitor connected in parallel and the second charge unit is bridged to the second connector by a second capacitor connected in parallel.

9. The light receiver in accordance with claim 1, wherein the avalanche photodiode elements form a matrix arrangement; wherein a plurality of first signal tapping circuits are each connected to the first connectors of the avalanche photodiode elements of a column and a plurality of second signal tapping circuits are each connected to the second connectors of the avalanche photodiode elements of a row.

10. The light receiver in accordance with claim 1, wherein the avalanche photodiode elements form a linear arrangement divisible into groups; wherein a plurality of first signal tapping circuits are each connected to the first connectors of the avalanche photodiode elements of a group and a plurality of second signal tapping circuits are each connected to the second connectors of an avalanche photodiode element from each group.

11. The light receiver in accordance with claim 1, wherein avalanche photodiode elements are connected in parallel and generate a common signal.

12. An optoelectronic sensor having at least one light receiver having a plurality of avalanche photodiode elements that are each biased by a bias above a breakdown voltage and that are thus operated in a Geiger mode to trigger a Geiger current on light reception, wherein the avalanche photodiode elements have a first connector and a second connector; wherein a first signal tapping circuit for reading out the avalanche photodiode elements is connected to one of the first and second connectors, and wherein a second signal tapping circuit for reading out the avalanche photodiode elements is connected to the other one of the first and second connectors, wherein the sensor is configured for distance measurement in accordance with a time of flight process and/or is configured for data transmission.

13. A method for reading out avalanche photodiode elements that are each biased by a bias above a breakdown voltage and are thus operated in a Geiger mode, wherein the avalanche photodiode elements trigger a Geiger current on light reception; wherein the avalanche photodiodes are read out by a first signal tapping circuit that is connected to a first connector of the avalanche photodiode elements; and wherein the avalanche photodiode elements are also read out by a second signal tapping circuit that is connected to a second connector of the avalanche photodiode elements.

14. The method in accordance with claim 13, wherein an input of an active coupling element for the Geiger current in the first signal tapping circuit and/or in the second signal tapping circuit and connected to one of the first and seconds connectors is virtually short circuited against a potential such that the Geiger current flows into the coupling element practically without change and is mapped there onto a measurement current corresponding to the Geiger current in progression and level at an output of the coupling element, wherein the output is decoupled from the input.

15. The method in accordance with claim 13, wherein the avalanche photodiode elements form a matrix arrangement and each avalanche photodiode element is read out, on the one hand, via a first signal tapping circuit associated with its column and, on the other hand, is read out via a second signal tapping circuit associated with its row.

Description

[0028] The invention will be explained in more detail in the following also with respect to further features and advantages by way of example with reference to embodiments and to the enclosed drawing. The Figures of the drawing show in:

[0029] FIG. 1 a representation of a signal tapping circuit for an avalanche photodiode element with an active coupling element;

[0030] FIG. 2 a comparison of the frequency response of a signal circuit with an active coupling element and a conventional signal tapping circuit;

[0031] FIG. 3 a schematic circuit arrangement of a matrix of avalanche photodiode elements having three connectors and a respective double signal tapping for the reading out in rows and columns;

[0032] FIG. 4 an alternative embodiment of the circuit arrangement in accordance with FIG. 3 of avalanche photodiode elements having two connectors;

[0033] FIG. 5 an alternative embodiment of the circuit arrangement in accordance with FIG. 3 of avalanche photodiode elements having four connectors;

[0034] FIG. 6 a block diagram of a logic for associating a detection event to an avalanche photodiode element;

[0035] FIG. 7 an exemplary circuit arrangement for the logic in accordance with FIG. 6;

[0036] FIG. 8 an exemplary signal progression at different points of the circuit arrangement in accordance with FIG. 7;

[0037] FIG. 9 an alternative linear arrangement of avalanche photodiode elements with a double signal tapping for the spatial association of detection events;

[0038] FIG. 10 a matrix representation of avalanche photodiode elements for the explanation of crosstalk effects; and

[0039] FIG. 11 a circuit detail of a matrix of avalanche photodiode elements for the further explanation of crosstalk effects.

[0040] FIG. 1 shows as a starting point a representation of an avalanche photodiode element 10 having a signal tapping circuit 12 that is optimized for sensitivity and bandwidth. The invention relates to an arrangement of a plurality of such avalanche photodiode elements each having two signal taps that are preferably each configured as in FIG. 1 and will then be explained further below with reference to FIGS. 3 to 11.

[0041] The avalanche photodiode element 10 is shown as an equivalent circuit diagram. The actual design of the semiconductor component is considered as known and is not shown. A corresponding component can, for example, be manufactured in a CMOS process. The breakdown voltage of the avalanche photodiode elements 10 is considerably smaller than in conventional avalanche photodiodes and amounts, for example, to at most 50 V or 30 V.

[0042] The avalanche photodiode element 10 shows the behavior of a diode 14, on the one hand. It additionally has a capacitance that is represented by the capacitor 16 connected in parallel. An avalanche effect is triggered by at least one incident photon, with this process acting like a switch 18. In standby mode, a voltage above the breakdown voltage is present above the diode 14 between a connector 20 and a connector 22. If then an incident photon generates a charge carrier pair, it so-to-say closes the switch 28 such that the avalanche photodiode element 10 is flooded with charge carriers and a so-called Geiger current flows. New charge carriers, however, only arise as long as the electrical field remains strong enough. If the capacitor 16 is discharged so much that the breakdown voltage is fallen below, the avalanche self-quenches (“passive quenching”). The capacitor 16 is then charged from the connectors 20, 22 via a resistor 24 until a voltage above the breakdown voltage is again applied to the diode 14. There are alternative embodiments in which the avalanche is recognized from outside and a discharge is thereupon triggered below the breakdown voltage.

[0043] The output signal rises rapidly and independently of the intensity of the triggering light to a maximum value during the avalanche and then falls again after the quenching of the avalanche. The time constant of the drop that indicates a dead time of the avalanche photodiode element 10 is typically in the range of some nanoseconds up to some tens of nanoseconds. The dead time is not an absolute dead time for, as soon as the bias is large enough to support an avalanche, the output signal can also rise again; however, not to the same degree as from the standby state. The amplification factor amounts to up to 10.sup.6 and substantially results from the maximum number of charge carriers that can be recruited by the avalanche in the avalanche photodiode element 10.

[0044] The function of the signal tapping circuit 12 is to lead off a measurement signal from the Geiger current during an avalanche effect, utilizing the full current flow where possible, and indeed such that high frequency portions are also maintained and a high signal-to-noise ratio is achieved. The signal decoupling preferably takes place capacitively over a coupling capacitor 26. In the embodiment shown, the avalanche photodiode element 10 has its own connector 28 for reading out a measurement signal, with this connector 28 being connected via the coupling capacitor 26. Alternatives to such an avalanche photodiode element 10 will also be presented further below having more or fewer than three connectors 20, 22, 28.

[0045] Further avalanche photodiode elements not shown in FIG. 1 form, from the viewpoint of the avalanche photodiode element 10 shown, parasitic capacitances that can be increased by other parasitic effects. The parasitic capacitance 30 is the sum of the typically numerous further avalanche photodiode elements in practical applications and can therefore be much larger than the capacitance of the associated coupling capacitor 26. The parasitic capacitance 30 behaves like a low pass that blocks high frequency signals.

[0046] The signal tapping circuit 12 should have a signal resistance that is as small as possible to achieve high speeds or bandwidths and should simultaneously have a large signal resistance for a high sensitivity. To meet these contradictory demands, the signal tapping circuit 12 utilizes an active circuit solution having an active coupling element 32 that is configured as a bipolar NPN transistor in the base circuit in FIG. 1. Other active elements, in particular other transistors (FETs), a different polarity (PNP) or a different circuit (emitter circuit) are conceivable. In addition, a plurality of transistors can also be used instead of a single-stage circuit.

[0047] The active coupling element 32 now has a plurality of substantial advantages for the signal tapping. On the one hand, it provides practically no resistance for the Geiger current capacitively decoupled as a current pulse, i.e. it forms a virtual short circuit. This is practically not fully achieved, but less than one ohm is by all means possible. This has the result that there are no relevant voltage changes over the parasitic capacitance 30 despite the avalanche and that there are thus no charge transfer procedures and no current flow. Almost the total Geiger current that flows through the coupling capacitance 26 is therefore available to the active coupling element 32. Without the virtual short circuit, in contrast, considerably appreciable portions of the Geiger current would be lost in the parasitic capacitance 30 and the fast, high frequency signal portions would above all be suppressed.

[0048] Secondly, the active coupling element 32 at the output side generates a measurement current whose time progression and level corresponds to the Geiger current. In this respect, the active coupling element 32 can also directly vary, in particular amplify, the measurement current with respect to the Geiger current by its transfer function. Practically the complete current of the avalanche effect is available at the coupling element 32 for this purpose. The measurement current is subsequently tappable at the output side for further processing as the detection result. The measurement current is fed from a power source of the active coupling element 32 and not from the avalanche photodiode element 10.

[0049] In this respect, thirdly, the input circuit and the output circuit are decoupled from one another. The further processing of the measurement current therefore has, within the limits of a real decoupling, no effects on the Geiger current. Practically any desired subsequent stages are therefore possible which, unlike conventional signal tappings, have no unfavorable repercussions for the Geiger current.

[0050] In the specific example of FIG. 1, the emitter of the active coupling element 32 forms the input 34 to which the Geiger current is supplied by the connector 28 of the avalanche photodiode element 10. The base is connected to ground or generally to a fixed potential and is thus virtually short circuited. The collector forms the output 36 at which the measurement current is provided. There is also a constant power source 39 between the input 34 and a supply voltage −U.sub.2 in the emitter circuit of the transistor. A DC current that sets the operating point of the transistor thereby flows at times outside avalanche effects. The constant power source 38 can alternatively be realized passively by a resistor 40 or as a combination of both, as shown in FIG. 1. The measurement current is tapped at the output 36. A simple measurement impedance 42 is only one example of any desired measurement circuits that can have further active and/or passive elements. These measurement circuits practically have no repercussion on the actual measurement due to the decoupling by means of the signal tapping circuit 12.

[0051] FIG. 2 illustrates the high frequency behavior of the signal tapping circuit in accordance with FIG. 1 as the result of a test arrangement in which a laser radiates a signal onto a light receiver having a plurality of avalanche photo elements 10, with the laser light being modulated in its amplitude by a varying frequency of 100 kHz to 3 GHz. In this respect, the upper progression 44 corresponds to the result with the signal tapping circuit 12 and the lower progression 46 corresponds, for comparison, to a conventional signal tapping. The signal tapping circuit 12 shows a very flat frequency response with a bandwidth of 2 GHz and a clear signal drop only on the other side of 2 to 3 GHz. This is a multiplication with respect to the lower progression 46 where the measurement signal already breaks down at moderate frequencies long before the GHz range.

[0052] FIG. 3 shows a schematic circuit arrangement of a light receiver 50 having a plurality of m×n avalanche photodiode elements 10.sub.11 . . . 10.sub.mn that are arranged in a matrix with m columns and n rows. The individual avalanche photodiode elements 10.sub.11 . . . 10.sub.mn can correspond to FIG. 1 in their design even though they are shown simplified here. In addition, the position of the connector 28 and of the resistor 24 at the diode 14 have been swapped over in FIG. 3 with respect to FIG. 1 to illustrate that their interconnection is possible at the anode side and at the cathode side. The number of avalanche photodiode elements 10.sub.11 . . . 10.sub.t, varies in dependence on the component and on the target demands and can amount to several thousand and more. In this respect, avalanche photodiode elements can also be electrically interconnected group-wise in a different manner than in the illustration. The individual matrix positions are then therefore occupied by a plurality of avalanche photodiode elements that deliver a common sum signal. This enables statistical methods that compensate the fact that an individual Geiger mode avalanche photodiode element can, on the one hand, already be incorrectly triggered by a signal external light photon and can, on the other hand not be activated for some time after each triggering.

[0053] The detection signal of the avalanche photodiode elements 10.sub.11 . . . 10, is in each case tapped twice at the connector 28 over a respective row line 52.sub.1 . . . 52.sub.n and at signal tapping circuits 12a1 . . . 12an associated with the rows as well as at the connector 22 over a respective column lead 54.sub.1 . . . 54.sub.m and signal tapping circuits 12b1 . . . 12bm associated with the columns. Which avalanche photodiode element 10.sub.11 . . . 10.sub.mn has triggered a detection event can be determined by this type of reading out. The light receiver 50 thus becomes spatially resolving.

[0054] The avalanche photodiode elements 10.sub.11 . . . 10.sub.mn are biased over the voltage sources VCC6 and VCC1 and their pre-resistors R111 . . . R1n1. The function of the active coupling elements 32 in the signal tapping circuits 12a1 . . . 12an, 12b1 . . . 12bm have already been explained in connection with FIG. 1. They ensure an input resistance that is as small as possible and that ensures that the voltages hardly change on an avalanche event. Larger circuit capacitances that are unavoidable with the plurality of interconnected avalanche photodiode elements 10.sub.11 . . . 10.sub.mn thus also have hardly any effect on the received signal because there are practically no charge transfer procedures, because the signal current is available almost completely, and because the signal shows very good high frequency properties.

[0055] The bias current of the active coupling element 32 results in the signal tapping circuits 12a1 . . . 12an of the rows from the voltage source VCC4, less U.sub.be, and from the respective resistor R411 . . . R41n. The voltage source VCC5 supplies the transistor stage. The signal voltage which is a negative pulse that is superposed on a positive DC voltage/bias voltage can be decoupled at the resistor R511 . . . R51n. If the diode D11, for example of the avalanche photodiode element 10.sub.11, now breaks down, a negative current pulse is coupled onto the row signal line via the capacitor C11 and is supplied to the emitter of the transistor T511. U.sub.be thus rises and the collector current will increase accordingly. The output signal U_Y1 will output a negative voltage pulse.

[0056] The signal tapping circuits 12b1 . . . 12bm of the columns are connected via the anodes of their respective diodes 14 to the column lines 54.sub.1 . . . 54.sub.m. On an avalanche effect, a positive current pulse is therefore coupled onto the emitter T311 . . . T3m1 via the coupling capacitors C211 . . . C2m1 and is tapped at the collector or at R311 . . . R3m1. The operating point of the transistors T311 . . . T3m1 is here fixed via VCC2, less U.sub.be, and via the resistor R211 . . . R2m1 and the supply takes place via VCC3.

[0057] In the embodiment in accordance with FIG. 3, avalanche photodiode elements 10.sub.11 . . . 10.sub.mn are used having three connectors 20, 22, 28. FIG. 4 shows an alternative embodiment having two connectors 20, 22. The signal tapping thus takes place at the connectors via which the bias is also applied. For this purpose, the typically high-ohmic charge resistor R11 . . . Rmn is bridged in an AC current aspect by a respective small capacitor C11 . . . Cmn in a parallel branch to conduct fast current pulses on an avalanche effect at low ohms to the signal tapping circuits 12a1 . . . 12an, 12b1 . . . 12bm. The processes correspond overall to those in the embodiment in accordance with FIG. 3.

[0058] FIG. 5 shows a further alternative embodiment in which the avalanche photodiode elements 10.sub.11 . . . 10.sub.mn have four connectors 20, 22, 28a-b. For this purpose, the charge resistor 24a-b is divided into two portions R11, R110 . . . Rmn, Rmn0, in particular by half, that are connected at the cathode or at the anode of the diode D11 . . . Dmn to supply the bias +VCC6 or −VCC1. The signals are decoupled via a respective capacitor C11, C110 . . . Cmn, Cmn0 at the cathode or at the anode. Advantages can particularly result with a symmetrical or 50% division, for example, with respect to time of flight differences that are as small as possible between the row lines 51.sub.1 . . . 52.sub.n and the column lines 51.sub.1 . . . 54.sub.m.

[0059] FIG. 6 shows a block diagram of a logic 56 for associating a detection event with an avalanche photodiode element 10.sub.11 . . . 10.sub.mn. It is the object of this logic 56 to carry out this association by linking the Y signals with the X signals. In this respect, the input signals can be pre-processed, amplified here by way of example.

[0060] FIG. 7 shows an exemplary simple circuit arrangement for the logic 56. In this respect, the signals X.sub.m and Y.sub.m are linked with one another via an AND logic. In the bias state, a LOW potential is applied to both inputs of the AND logic. This is achieved for the X signal line with a leak resistor R1 to ground in combination with a capacitor C1 connected upstream. Since the input signals on the Y signal line are negative, an inverter is also arranged here. So that a LOW potential is also applied to the input of the AND gate here at bias, the input of the inverter is drawn to a HIGH potential via a voltage splitter. With corresponding adaptations, it is also possible to use an OR gate, a NAND gate or a NOR gate instead of an AND logic.

[0061] FIG. 8 shows exemplary signal progressions at different points of the circuit arrangement in accordance with FIG. 7, namely at the top the voltage U_R2 of the Y signal; in the middle the voltage U_R1 of the X signal, and at the bottom the output voltage U_out. In the bias state, the voltage U_R2 is HIGH and the voltage U_R1 is LOW. The output voltage is HIGH exactly when both U_R2 changes to LOW and U_R1 changes to HIGH. In the example of FIG. 8, the voltage progression U_R2 is somewhat delayed with respect to U_R1 for an exemplary illustration. This has the consequence that the output voltage pulse at U_out is a little shorter than the input pulses. The times of flight of the two signal lines X or Y should generally be very similar or are equal in the ideal case to counteract such a reduction of the pulse width.

[0062] FIG. 9 shows an alternative linear arrangement of avalanche photodiode elements 10.sub.11 . . . 10.sub.mn. The double signal tapping also enables a spatial association of detection events here. The first signal tap divides the avalanche photodiode elements 10.sub.11 . . . 10.sub.mn into groups while the second signal tap respectively reads out the ith avalanche photodiode element 10.sub.11 . . . 10.sub.mn of a group. The circuit topology thus basically corresponds to FIGS. 3 to 5 and the variants shown there are also possible. Only the geometrical arrangement of the avalanche photodiode elements 10.sub.11 . . . 10.sub.mn is different. A linear arrangement is only one example. In addition, the circuit topology, that is the association of the two signal taps, can also be varied.

[0063] In a matrix arrangement having numerous avalanche photodiode elements 10.sub.11 . . . 10.sub.mn connected in parallel, problems typically occur due to crosstalk effects. The crosstalk is greatly reduced by the particularly low ohmic or virtually short circuited signal tapping circuit 12 because there are only very small voltage changes or signal voltages in the matrix lines. This will be conclusively explained with reference to FIGS. 10 and 11.

[0064] FIG. 10 shows the matrix arrangement of the light receiver 50, with numerous elements being omitted for reasons of clarity, the avalanche photodiode elements being reduced to a number of 4×4, and only being shown as capacitors in accordance with their properties relevant to the crosstalk. If now, for example, D42 triggers, the measurement signal has to be measured at U_Y2 and in the opposite polarity at U_X4. R_Y and R_X represent the differential input resistances of the signal tapping circuits. If now R_Y2 is not very low ohm, a specific coupling of U_Y2 to U_X3 will occur over the row D32. Signal portions are consequently coupled to U_X1 and U_X2 over the corresponding cells D12 and D22. It must be noted in this respect that the polarity of the crosstalk signals is opposite to U_X4. Signal portions are likewise coupled to U_Y3 over the cell D33. They are, however, damped twice and are therefore very weak. There is rather an inversely phased coupling over cell D43 here. If, in contrast, the differential input resistances are very low ohmic or virtually short circuited, as in a signal tapping circuit in accordance with FIG. 1, the described interference effects are substantially reduced. This observation only serves for illustration and is incomplete since there are a number of further coupling paths.

[0065] FIG. 11 shows a circuit section for the triggering diode D42. The negative and positive pulses, shown by minus and plus signs in a circle here, are supplied from D42 to the resistor R_Y2 or R_X4. Major crosstalk terms for R_Y3 should now be looked at. The resistances of the individual avalanche photodiode elements are shown as complex resistances Z. The strongest crosstalk path runs inversely phased to the signal at R_Y2 over only the one resistor Z_D43. In addition, there are a number of parallel crosstalk paths over a respective two resistors in the longitudinal direction that deflect R_Y3 in the same phase as R_Y2. These crosstalk paths, however, include voltage splitters with Z_D12, Z_D22 or Z_D32 in the signal path and R_X1,R_X2 or R_X3 against ground and are thus much weaker. This again illustrates how advantageous resistances R_X and R_Y that are as small as possible are for a small crosstalk.