Method for operating a photodiode

Abstract

A method (200) for operating a photodiode includes the steps of: applying a voltage consisting of the sum of a constant first voltage (300) and a second voltage (301) having a fixed frequency to the photodiode, and detecting the output signal of the photodiode. Subsequently, the spectral composition of the output signal is determined, and at least one coefficient of a nonlinear term of the photodiode's gain (302) is determined based on the spectral composition. Then, a matched first voltage is applied to the photodiode, wherein the first voltage (300) is matched based on the coefficient. A first coordinate axis (101) shows the voltage applied to the photodiode, and a second coordinate axis (102) shows the gain of the photodiode. A first characteristic curve (103) shows the gain at a first temperature. The gain (302) has a term containing twice the fixed frequency ω0. The output signal (302) can also have higher-order frequency components. The determined coefficients allow the gain at the current temperature to be inferred. The first voltage is matched in such a way that the gain remains constant. By adjusting the gain in this way to keep it constant, the photodiode can provide particularly reliable and comparable measurement data. Such an adjusted photodiode can be part of a car's lidar system.

Description

Technical Field

[0001] This invention relates to a method for operating a photodiode. Background Technology

[0002] Avalanche photodiodes (APDs) are used to detect electromagnetic radiation from the infrared spectrum up to the gamma-ray range. APDs are used in various fields, such as lidar systems, medical technology, and high-energy physics. APDs fully utilize the photoelectric effect to generate electrons from incident electromagnetic radiation. These primary electrons migrate to regions with very high electric field strengths, where they trigger an avalanche effect through impact ionization, which constitutes the gain of the APD. Of particular interest is the use of APDs as detectors in lidar systems in the automotive field. Here, short light pulses are emitted from a transmitter into a scene; the reflection of these short light pulses is recorded by the APD, and the distance is determined by the light's propagation time.

[0003] Temperature dependence of gain can hinder the reliable use of APDs in the automotive field. Temperature variations affect the structure of the space charge region and the electron occupancy density, both of which influence gain. When compensation for temperature dependence of gain is insufficient, the signal-to-noise ratio of the APD can deteriorate significantly. Typically, to compensate, the voltage applied to the APD (approximately ~1 to 1.5 V / K) can be matched for different temperatures. This requires accurate knowledge of both the APD's temperature and its temperature characteristics. However, in real-world applications, uncertainties regarding the precise temperature of the APD can exist. Furthermore, variations in temperature characteristics may occur between different batches of APDs. Therefore, only approximate compensation for the temperature characteristics of the APD can be achieved. To minimize temperature effects, the gain must be measured for each APD based on the applied voltage and temperature.

[0004] A bias circuit for applying a bias voltage to an APD is known from EP 1 006 591 A2. A temperature sensor detects the temperature surrounding the APD. A memory contains voltage data corresponding to multiple temperatures within a pre-fixed temperature range. The bias voltage is adjusted based on the voltage data to compensate for temperature-induced variations in the APD's gain.

[0005] EP 2 056 126 A1 discloses a method for detecting light pulses reflected from an object. Here, a light pulse with a known intensity and duration is emitted in the direction of the object, and the reflected signal of the light pulse is detected. An amplifying sensor amplifies the reflected signal. The gain of the sensor is controlled to be adjusted so that the gain at each time point of the detection can be determined.

[0006] An APD gain control circuit is known from EP 0 856 943 A2. This circuit includes a bias generator for changing the bias voltage on an APD with variable gain in response to a bias control value, wherein the bias control value is generated by a control device. The control device receives the output signal of the APD and determines the system noise for different bias control values. This system noise is compared to a threshold to determine an optimized bias voltage for optimizing the gain of the APD.

[0007] The methods described in the prior art have the following drawbacks: they cannot, or can only insufficiently, compensate for changes in the gain of the APD based on temperature. Although EP 1 006 591 A2 describes a temperature compensation based on temperature measurement, this method is not based on an accurate understanding of the temperature and temperature characteristics of the APD, because it only measures the temperature around the APD. Summary of the Invention

[0008] The object of the present invention is to provide an improved method for operating a photodiode. Furthermore, an apparatus for performing the method is to be provided. This object is achieved by a method for operating a photodiode having the features of the independent claims and an apparatus for performing the method. Advantageous extensions are given in the dependent claims.

[0009] A method for operating a photodiode includes the following steps: In a first step, a voltage consisting of the sum of a constant first voltage and a second voltage having a fixed frequency is applied to the photodiode. In a second step, an output signal of the photodiode is detected. In a third step, the spectral composition of the output signal is determined. In a fourth step, at least one coefficient of a nonlinear term of the photodiode's gain is determined based on the spectral composition. In a fifth step, a matched first voltage is applied to the photodiode, wherein the first voltage is matched based on the coefficient.

[0010] Advantageously, temperature-induced gain variations can be compensated by applying a matched first voltage. However, it is not necessary to know the temperature and / or temperature characteristics of the photodiode. Instead, a test signal is applied to the photodiode by applying a second voltage in the first method step to determine at least one coefficient of the nonlinear term of the gain. This allows the current gain to be deduced, and thus enables the application of the matched first voltage, which can also be referred to as the cutoff voltage, to adjust the gain. That is, the method enables the monitoring and adjustment of the photodiode's gain.

[0011] In one implementation, the first voltage is matched such that the gain remains constant. Advantageously, when the photodiode is operated with a constant gain, it can provide particularly reliable measurements. Temperature dependence of the gain can be compensated for without knowledge of the photodiode's temperature and / or temperature characteristics.

[0012] In one embodiment, the power of the electromagnetic radiation incident on the photodiode is constant during the execution of the method. Advantageously, the determination of the coefficient of the nonlinear term of the photodiode's gain is unaffected by variations in the incident intensity of the electromagnetic radiation. In one embodiment, this can be ensured, for example, by selecting the measurement duration for detecting the output signal such that the power of the electromagnetic radiation incident on the photodiode is constant during that measurement duration. In an alternative embodiment, the method is performed during the dark phase of the photodiode. That is, in this case, only the dark current of the photodiode is amplified and modulated by means of the applied test signal.

[0013] In one embodiment, the method is performed while the lidar system with photodiodes is in operation. Here, in one embodiment, the pulse frequency of the optical pulses in the lidar system can be selected such that it is possible to distinguish between changes in the output signal caused by the optical pulses and changes in the output signal caused by the applied second voltage. Advantageously, when the changes in the output signal caused by the applied second voltage do not superimpose with the changes in the output signal caused by the optical pulses, the coefficients of the nonlinear term of the gain can be determined more effectively.

[0014] In principle, the frequency of the second voltage can be within a wide frequency range, for example, from 0Hz to 100MHz. In one embodiment, the frequency of the second voltage is greater than the image refresh rate of the lidar system and less than the upper limit frequency of the bandwidth of the lidar system's amplifier. In one embodiment, the frequency of the second voltage is in the range between 100Hz and 20MHz.

[0015] In one implementation, the spectral composition is obtained by Fourier analysis or by demodulation. Advantageously, Fourier analysis and demodulation provide simple methods for obtaining the spectral composition of the output signal.

[0016] In one implementation, the photodiode is part of the vehicle's lidar system and performs the method while the vehicle is in motion. Advantageously, the photodiode in the lidar system can provide reliable measurement data while the vehicle is in motion because the photodiode's gain is adjusted. This improves the efficiency of the lidar system. In the automotive field, lidar systems are used, for example, in autonomous vehicles.

[0017] In one embodiment, the method steps are repeated. Advantageously, repeating the method can compensate for changes in the photodiode gain due to rapid temperature variations. For example, the method can be repeated at discrete time intervals or, alternatively, quasi-continuously.

[0018] An apparatus for performing the method includes a photodiode with a connection for applying a voltage, a detection device for detecting the output signal of the photodiode, an analysis device for determining the spectral composition of the output signal, an analysis processing device for determining at least one coefficient of a nonlinear term of the photodiode's gain based on the spectral composition, and an adjustment device for matching a first voltage based on the coefficient. In one embodiment, the photodiode is an avalanche photodiode. Attached Figure Description

[0019] The above-described features, characteristics, and advantages of the present invention, as well as the ways in which these features, characteristics, and advantages are realized, will become clearer and more explicit in conjunction with the following description of the embodiments, which are further elaborated in conjunction with the accompanying drawings. Here, the corresponding schematic diagrams show:

[0020] Figure 1 An exemplary temperature characteristic curve of a photodiode is shown;

[0021] Figure 2 The method steps for operating the photodiode are shown;

[0022] Figure 3 The diagram illustrates the change in gain of a photodiode produced by applying a voltage with a fixed frequency.

[0023] Figure 4 The spectrum of the output signal of the photodiode is shown;

[0024] Figure 5 The device used to perform this method is shown. Detailed Implementation

[0025] Figure 1 An exemplary temperature characteristic curve of a photodiode is schematically shown. For example, the photodiode could be an avalanche photodiode (APD). An APD fully utilizes the photoelectric effect to generate electrons from incident electromagnetic radiation. These primary electrons migrate to a region with a very high electric field strength, where they are triggered by impact ionization, resulting in an avalanche effect that constitutes a gain. This form of gain is internal gain. For example, the photodiode could also be a Zener diode, in which the avalanche effect occurs at least partially, depending on the voltage applied to the photodiode.

[0026] exist Figure 1 In the diagram, the first coordinate axis 101 shows the applied voltage. The second coordinate axis 102 shows the gain of the photodiode. The correlation of gain is illustrated for four different temperatures. The first characteristic curve 103 shows the gain at the first temperature. The second characteristic curve 104 shows the gain at the second temperature. The third characteristic curve 105 shows the gain at the third temperature. The fourth characteristic curve 106 shows the gain at the fourth temperature. Figure 1 As shown, with a fixed voltage of 107V, the gain of the photodiode varies at different temperatures. Furthermore, in... Figure 1 In the diagram, the tangent line 108 on characteristic curves 103, 104, 105, and 106 indicates that for a fixed voltage 107, characteristic curves 103, 104, 105, and 106 have different slopes. Furthermore, for a fixed voltage 107, the curvatures of characteristic curves 103, 104, 105, and 106 are also different. The concept upon which this invention is based is that if the curvature of the characteristic curves can be known, changes in gain can be compensated for by temperature variations without requiring detailed information about the temperature of the photodiode itself. Therefore, it is necessary to characterize the nonlinear performance of the gain.

[0027] To describe the nonlinear properties of the gain, it can be represented in a first-order approximation using linear and nonlinear (quadratic) terms of the applied voltage:

[0028] G(T,U)=G0(T)U(t)+G1(T)U 2 (t) (1)

[0029] Here, G is the gain, T is the temperature, U is the voltage applied to the photodiode, t is time, G0 is the coefficient of the linear term, and G1 is the coefficient of the nonlinear term. For the applied voltage U, it is now possible to apply a time-dependent voltage.

[0030] U(t) = U0 + U A sin(ω0t) (2)

[0031] Here, U0 is a constant first voltage, U A It is the amplitude of the second voltage with a fixed frequency ω0.

[0032] The gain is derived from equations (1) and (2):

[0033] G(T,U)=G0(T)U0+G1(T)U0 2 +U A G0sin(ω0t)+2G1(T)U0U A sin(ω0t)+G1(T)U A2 cos(2ω0t) / 2 (3)

[0034] Figure 2 The method steps 201, 202, 203, 204, and 205 of the method 200 for operating a photodiode are schematically shown.

[0035] In the first method step 201, a voltage consisting of the sum of a constant first voltage and a second voltage having a fixed frequency is applied to the photodiode. In the second method step 202, the output signal of the photodiode is detected. In the third method step 203, the spectral composition of the output signal is determined. In the fourth method step 204, at least one coefficient of the nonlinear term of the photodiode's gain is determined based on the spectral composition. In the fifth method step 205, a matched first voltage is applied to the photodiode, wherein the first voltage is matched based on the coefficient. Optionally, method steps 201, 202, 203, 204, and 205 of method 200 can be repeated.

[0036] The first voltage, U0, can also be called the cutoff voltage. The second voltage has an amplitude U. A And a fixed frequency ω0. The second voltage can also be referred to as the test signal.

[0037] Figure 3 The diagram schematically illustrates the change in the gain 302 of the photodiode produced by applying a second voltage 301 with a fixed frequency ω0. Figure 3 In the diagram, the first coordinate axis 101 again shows the applied voltage. The second coordinate axis 102 shows the gain of the photodiode. For example, Figure 3 Only the first characteristic curve 103 of the gain is shown.

[0038] In addition, Figure 3 The diagram shows the applied first voltage 300 and the applied voltage with amplitude U. A The time curve of the second voltage 301 at a fixed frequency ω0 is shown. A change in gain 302 occurs along with the applied second voltage 301. If, when a test signal is applied, i.e., when the second voltage 301 is applied, the photodiode is exposed to incident electromagnetic radiation of constant power, the output signal can be measured, which is also altered by the changing gain 302. For example, the output signal can be the photocurrent of the photodiode.

[0039] As can be seen from equation (3), gain 302 has a term containing twice the fixed frequency ω0. For this reason, in the third method step 203, the spectral composition of the output signal, i.e., the spectrum, is determined. In addition to twice the fixed frequency ω0, the output signal can also have higher-order frequency components. Integer and even multiples of the fixed frequency ω0 are possible.

[0040] Figure 4 A schematic spectrum of the test signal and the output signal is shown. The third axis 401 shows the frequency. The fourth axis 402 shows the amplitude of a single frequency component. That is, the spectrum is, exemplarily, the amplitude spectrum.

[0041] The spectrum of the test signal has only frequency component 403, which has a fixed frequency ω0. The output signal has a first frequency component 404 with a fixed frequency ω0, as well as other frequency components 405, 406, and 407. The second frequency component 405 has a frequency of 2ω0. The third frequency component 406 has a frequency of 4ω0. The fourth frequency component 405 has a frequency of 6ω0. The output signal can also have other integer and even multiples of the fixed frequency ω0.

[0042] The spectral composition can be obtained through Fourier analysis or demodulation. In the case of Fourier analysis, the spectrum is obtained using Fourier transform. For example, the amplitude spectrum of the components containing the frequency components of the output signal can be obtained. In the case of demodulation, the envelope at a fixed frequency is obtained. This frequency should correspond to an integer and even multiple of the fixed frequency of the second voltage.

[0043] exist Figure 2 In the fourth method step 204 of method 200, at least one coefficient of the nonlinear term of the gain of the photodiode is determined based on the spectral composition. Here, at least coefficient G1 is determined. Similarly, coefficients of higher-order terms of the gain can also be determined. However, method 200 is described below only for the case where only the coefficient G1 of the power term is determined. The coefficient of the power term can be determined from the amplitude ratio of the first frequency component 404 to the second frequency component 405. Coefficient G1 allows the gain G to be inferred at the current temperature. Based on the determined coefficient of the nonlinear term of the gain, a first voltage U0 can now be matched to match the gain G. For this reason, in the fifth method step 205 of method 200, the matched first voltage is applied to the photodiode. For example, the first voltage can be matched such that the gain remains constant. By adjusting the gain in this way to keep it constant, the photodiode can provide particularly reliable and comparable measurement data.

[0044] able to execute in this way Figure 2Method 200 ensures that the power of electromagnetic radiation incident on the photodiode is constant during its execution. Therefore, the output signal of the photodiode is unaffected by variations in the incident intensity of the electromagnetic radiation. In one embodiment, this can be guaranteed, for example, by selecting the measurement duration for detecting the output signal such that the power of the electromagnetic radiation incident on the photodiode remains constant during that measurement duration. Alternatively, the method is performed during the dark phase of the photodiode. That is, in this case, only the dark current of the photodiode is amplified and modulated by the applied test signal.

[0045] Figure 5 Schematic illustration of the use of execution Figure 2 Method 200 equipment 500.

[0046] The device 500 includes a photodiode 501 with a connection for applying a voltage, a detection device 502 for detecting the output signal of the photodiode, an analysis device 503 for determining the spectral composition of the output signal, an analysis processing device 504 for determining at least one coefficient of the nonlinear term of the photodiode's gain based on the spectral composition, and an adjustment device 505 for matching a first voltage based on the coefficient. In one embodiment, the photodiode is an avalanche photodiode.

[0047] exist Figure 2 In method 200, the detection device detects the output signal in step 202 of the second method and provides the output signal to the analysis device 503. The analysis device 503 determines the spectral composition of the output signal and provides the determined spectral composition to the analysis processing device 504. The analysis processing device 504 determines at least one coefficient of the nonlinear term of the gain and provides the at least one coefficient to the adjustment device 505. The adjustment device 505 matches a first voltage, which is applied to the photodiode to match the gain.

[0048] For example, device 500 can be part of a lidar system. Figure 2 Method 200 can also be executed while the lidar system is running. The objective here is to select the pulse frequency of the lidar system's optical pulses in such a way that it is possible to distinguish between changes in the output signal caused by the optical pulses and changes in the output signal caused by the applied second voltage.

[0049] In principle, the frequency of the second voltage can be within a wide frequency range, such as from 0Hz to 100MHz. For the purpose of a LiDAR system, it is desirable that the frequency of the second voltage is greater than the LiDAR system's image refresh rate and less than the upper limit of the bandwidth of the LiDAR system's amplifier. For example, the frequency of the second voltage can be in the range between 100Hz and 20MHz. For example, the amplitude of the second voltage can be 1V. This value is merely an exemplary order of magnitude of the second voltage amplitude and should not be interpreted restrictively. For example, the first voltage can be 100V. The exemplary example of a 100V value for the first voltage should also only represent a typical order of magnitude and should not be interpreted restrictively.

[0050] Photodiodes can also be part of a car's lidar system. Figure 2 Method 200 can also be executed while the vehicle is in operation. Simultaneously, the method can also be executed while the vehicle's lidar system is in operation. Advantageously, the photodiode of the lidar system can provide reliable measurement data while the vehicle is in operation because the gain of the photodiode is adjusted.

Claims

1. A method (200) for operating a photodiode (501), the method having the following steps: applying a voltage to the photodiode (501) as a bias voltage, the voltage consisting of a constant first voltage (300) and a second voltage (301) having a fixed frequency, detecting an output signal of the photodiode (501), determining a spectral composition of the output signal, determining at least one coefficient of a non-linear term of a gain (302) of the photodiode (501) based on the spectral composition, wherein the gain (302) being represented in a first order approximation by means of linear and non-linear terms of the applied voltage, and wherein at least only the coefficient of the quadratic term is determined, applying a matched first voltage to the photodiode (501), wherein the first voltage (300) is matched based on the coefficient, wherein the method (200) is performed while a laser radar system having the photodiode (501) is operated, and the pulse frequency of light pulses of the laser radar system is chosen such that changes of the output signal due to the light pulses and changes of the output signal due to the applied second voltage (301) can be distinguished from each other.

2. The method (200) according to claim 1, wherein the matching of the first voltage (300) is performed such that the gain remains constant.

3. The method (200) according to claim 1 or 2, the power of electromagnetic radiation incident on the photodiode (501) is constant while the method (200) is performed.

4. The method (200) according to claim 3, the measurement duration for detecting the output signal is chosen such that the power of electromagnetic radiation incident on the photodiode (501) is constant over the measurement duration.

5. The method (200) according to claim 3, wherein the method (200) is performed in a dark phase of the photodiode (501).

6. The method (200) according to claim 1 or 2, wherein the frequency of the second voltage (301) is greater than the image refresh rate of the laser radar system and smaller than the upper limit frequency of the bandwidth of an amplifier of the laser radar system.

7. The method (200) according to claim 6, wherein the frequency of the second voltage (301) lies in a range between 100 Hz and 20 MHz.

8. The method (200) according to any one of the preceding claims, wherein the spectral composition is determined by Fourier analysis or by demodulation.

9. The method (200) according to any one of the preceding claims, wherein the photodiode (501) is part of a laser radar system of a car, wherein the method (200) is performed while the car is operated.

10. The method (200) according to any one of the preceding claims, wherein, the method steps (201, 202, 203, 204, 205) of the method (200) can be repeated.

11. A device (500) for performing the method (200) according to any one of claims 1 to 10, wherein The device (500) has a photodiode (501) with connections for applying a voltage, detection means (502) for detecting an output signal of the photodiode (501), analysis means (503) for ascertaining a spectral composition of the output signal, analysis processing means (504) for ascertaining at least one coefficient of a non-linear term of a gain (302) of the photodiode (501) on the basis of the spectral composition, and adjustment means (505) for matching the first voltage on the basis of the coefficient.

12. The device (500) according to claim 11, wherein The photodiode (501) is an avalanche photodiode.

Images