signal processing device

Abstract

To solve the problem that pertains to errors and limitations occurring in measurement by a pump-probe method.SOLUTION: A signal processing device 10 comprises a multiplier 21 and an automatic gain controller 17. The multiplier 21 multiplies a noise that appears, when the signal intensity of a probe signal that is a signal obtained by irradiating a sample with probe light Pr and the signal intensity of a reference signal obtained from reference light R are not balanced, as the difference between the signal intensity of the probe signal and the signal intensity of the reference signal, by the reference signal so as to obtain an error signal. The automatic gain controller 17 automatically adjusts, by the error signal, a gain for controlling the probe signal so as to correct the difference in signal intensity between he probe signal and the reference signal.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present disclosure relates to a signal processing device. [Background technology]

[0002] The pump-probe method is known as one of the optical measurement techniques. In this measurement, two superimposed beams of light, pump light and probe light, are incident on a sample to probe its structure. At this time, the sample responds in a specific way to the pump light. The probe light is intensity-modulated to reflect the response to the pump light, and by measuring this intensity change, the response of the material to the pump light can be observed.

[0003] Technologies related to the pump-probe method include an optical measurement device and method that measure changes in the intensity of probe light resulting from a response to a stimulus applied to a sample with a higher signal-to-noise ratio than when detecting changes in the phase of a detection signal (see Patent Document 1).

[0004] Furthermore, there is a technique relating to an interferometer that is useful for measuring, for example, temporally resolved optical nonlinearity as an optically induced change in the optical properties of a material sample (see Patent Document 2).

[0005] There is also a technology related to an optical measurement device that effectively removes noise in optical measurements and detects minute signals with high sensitivity (see Patent Document 3). In this technology, a balanced detector has an output terminal that outputs an electrical signal that represents the difference between the detection probe light and the reference probe light, and a lock-in amplifier detects a desired signal at the lock-in frequency.

[0006] Furthermore, there is a technology relating to a system having a configuration in which optical characteristics or electrical characteristics are changed using a variable gain amplifier (see Patent Document 4). [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Publication No. 2018-197705 [Patent Document 2] Japanese Patent Application Publication No. 11-257914 [Patent Document 3] Re-table 2015 / 030202 publication [Patent Document 4] Japanese Patent Application Laid-Open No. 2015-108626 Summary of the Invention [Problem to be solved by the invention]

[0008] In the pump-probe method, the probe light or the signal detected by the probe light is required to have very small fluctuations in the measurement of the sample. Here, the signal-to-noise ratio (S / N ratio) of the pump-probe method is dominated by the fluctuations (noise) of the probe light intensity. The signal in the pump-probe method is expressed by the ratio of the probe light intensity change, and the ratio of the intensity change is typically 10 -5 From 10 -3 For example, if the signal is 10 -4 When the noise of the probe light is 10 -4 (i.e., a fluctuation in intensity of 1 / 10,000), then S / N=1.

[0009] Given this background, it is desirable to reduce noise as much as possible in pump-probe measurements. A well-known method for reducing probe noise is the balanced detection method, in which a reference signal, which replicates the noise, is subtracted from the probe signal. The balanced detection method used in pump-probe measurements involves automatic balancing to compensate for the intensity balance between the typical probe and reference signals against noise. However, automatic balancing in the balanced detection method involves issues related to errors and limitations that arise in pump-probe measurements, such as response speed.

[0010] The present disclosure has been made in consideration of the above circumstances, and aims to provide a signal processing device that can solve problems related to errors and limitations that arise in measurements using the pump-probe method. [Means for solving the problem]

[0011] In order to achieve the above object, the signal processing device according to the present disclosure includes: a multiplier that obtains an error signal by multiplying a reference signal by noise that appears as a difference between the signal strength of a probe signal, which is obtained by irradiating a sample with probe light, and the signal strength of a reference signal obtained from reference light when the signal strengths of the probe signal and the reference signal are not balanced; and an automatic gain controller that automatically adjusts a gain for controlling at least one of the probe signal and the reference signal using the error signal obtained by the multiplier so as to correct the difference in signal strength between the probe signal and the reference signal. [Effects of the Invention]

[0012] The signal processing device of the present disclosure has the advantage of being able to solve problems related to errors and limitations that arise in pump-probe measurements. [Brief explanation of the drawings]

[0013] [Figure 1] Schematic illustration of the pump-probe method. [Figure 2] This figure shows the basic configuration of the apparatus for implementing the general balance detection method in the pump-probe technique. [Figure 3] FIG. 1 is a diagram showing the configuration of a device that performs automatic balance control in a balance detection method. [Figure 4] 10 is a graph showing an example of amplitude versus time of a signal in each process in measurement using automatic balancing. [Figure 5] FIG. 10 is a diagram showing an example of the ratio of DC offset to the intensity modulation depth of a signal in a noise-free situation. [Figure 6]1 is a diagram illustrating a configuration of a signal processing device according to an embodiment of the present invention. [Figure 7] FIG. 10 is a diagram illustrating a configuration of a signal processing device according to a modified example. DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings.

[0015] First, a technology that is a premise of the technology in the embodiments of the present disclosure will be described.

[0016] Figure 1 is a schematic diagram of the pump-probe technique. In pump-probe measurements, the sample is excited by pump light, and this excitation is observed as a change in the probe light. The pump-probe technique obtains a signal from the sample's response to the pump light stimulus, enabling super-resolution measurements that capture minute spatial structures. The pump-probe technique can be applied to a variety of measurements and instruments. For example, by amplifying small changes due to pump light stimulus with a strong probe light, it can be used to measure single-molecule absorption and stimulated Raman scattering molecular vibrations. Furthermore, applying it to photothermal microscopy and stimulated emission microscopy can improve spatial resolution. Furthermore, by observing how the sample relaxes after stimulation, it can be used to measure transient absorption.

[0017] Figure 2 shows the basic configuration of an apparatus for implementing a general balanced detection method in the pump-probe technique. As shown in Figure 2, the apparatus comprises an intensity modulator 11, mirrors 12A and 12D, a mirror or beam combiner 12B, an optical filter 12C, a beam splitter 13, a photodetector 14 (14A and 14B), and a subtractor 15. In this apparatus, a sample S is excited by a pump light Pu, and the sample S is observed by a probe light Pr.

[0018] Intensity modulator 11 intensity-modulates pump light Pu. Mirrors 12A and 12D reflect pump light Pu or probe light Pr. Mirror or beam combiner 12B irradiates sample S with pump light Pu without interfering with probe light Pr irradiated onto the sample. Probe light Pr passes through mirror or beam combiner 12B and is irradiated onto sample S.

[0019] The beam splitter 13 splits the probe light Pr into light directed toward the sample S for observing the sample S and reference light R that does not pass through the sample S. The photodetector 14A detects the probe light Pr that has been transmitted through or reflected or scattered by the sample S irradiated with the pump light Pu as a photodetection signal. The optical filter 12C prevents the pump light Pu from entering the photodetector 14A. Hereinafter, this photodetection signal will be referred to as a probe signal. The photodetector 14B detects a reference light detection signal (hereinafter referred to as a reference signal) from the reference light R.

[0020] The subtractor 15 subtracts the reference signal detected by the photodetector 14B from the probe signal detected by the photodetector 14A. The sample-derived signal contained in the probe signal is detected from the subtracted output using lock-in detection (synchronous detection) at the modulation frequency of the pump light Pu or a bandpass filter around the modulation frequency of the pump light Pu. Because the noise contained in the reference light R originates from the same source as the noise in the probe light Pr, as long as the intensities of the probe signal and the reference signal are balanced (matched), the noise can be reduced by subtracting the probe signal from the reference signal. In other words, if the intensities of the probe signal and the reference signal are unbalanced, the greater the degree of imbalance, the more difficult it becomes to reduce the noise by subtracting the probe signal from the reference signal. Therefore, it is important to match the intensities of the probe signal and the reference signal, i.e., to correct them to achieve balance. This is because maximum noise reduction is achieved when the average intensities of the probe signal and the reference signal are equal. However, in imaging (morphological observation using a microscope) where the measurement position of the sample changes, the transmittance of the probe light Pr changes during measurement, causing a dynamic imbalance, so automatic balance control is required to automatically correct the imbalance.

[0021] Fig. 3 is a diagram showing the configuration of a device that performs automatic balance control in the balance detection method. In addition to the basic device configuration shown in Fig. 2, the device that performs automatic balance control has a variable gain amplifier 16 and an automatic gain controller 17. In automatic balance, the automatic gain controller 17 corrects the DC offset after calculating the difference between the probe signal and the reference signal so that it becomes zero.

[0022] FIG. 4 is a graph showing an example of the amplitude of a signal over time in a measurement using automatic balancing. In FIG. 4, (a) is a graph showing the probe signal that has passed through the optical receiver 14A. (b) is a graph showing the probe signal that has passed through the variable gain amplifier 16, with the solid line and dotted line representing the signal before correction and the gain-controlled corrected signal, respectively. (c) is a graph showing the reference signal that has passed through the optical receiver 14B. (d) is the signal that is output after difference calculation, with the graph showing the signal before correction shown on the bottom and the graph showing the signal after correction shown on the top. Looking at (d), it is clear that a DC offset has occurred between the signal before correction and the signal after correction.

[0023] However, the automatic balancing of the balance detection method has problems related to the errors and limitations that arise in the pump-probe measurement described above. Specifically, there are various problems, such as (1) DC offset problems, (2) dynamic range problems, and (3) automatic balancing response speed problems. Each of these problems is explained in detail below.

[0024] (1) The DC offset problem will be explained. The DC offset of the difference signal is derived from the difference between the intensity of the probe signal and the intensity of the reference signal. This DC offset of the difference signal is used to control the automatic balance. However, if the DC offset of the circuit elements is added to the DC offset derived from the difference in optical intensity, an error will be introduced into the optimal gain.

[0025] Circuit elements that generate DC offset include amplifiers, photodiode dark current, and variable gain amplifiers. Amplifiers are nearly constant and can be corrected using variable resistors, etc. Photodiode dark current is temperature sensitive and unstable. Variable gain amplifiers have DC offsets that depend on the set gain. Gain control balances the probe signal and reference signal based on the DC level after the difference, so the optimal feedback target point fluctuates for each light intensity ratio. In other words, DC offset fluctuates during feedback control, resulting in instability.

[0026] (2) The dynamic range issue will be explained. The dynamic range is expressed as the ratio of the noise of a circuit element to the maximum output value. Figure 5 shows an example of the ratio of the DC offset and the intensity modulation depth of a signal in a noise-free situation. The circuit element can change the signal intensity by 10 times. 3 From 10 5 It is necessary to ensure that noise is smaller than the signal while preventing saturation even with DC components of photodetection signals that are approximately 1,000 to 100,000 times larger. This requires circuit elements to have a large dynamic range. When attenuating the signal input to a certain element to prevent it from saturating with a signal that contains a large DC component, the noise generated by the element remains constant regardless of the attenuation of the input signal. Therefore, in such cases, the relative contribution of noise becomes large (the signal-to-noise ratio becomes small). As described above, it is difficult to ensure a dynamic range that does not saturate even with signals that contain a large DC component while keeping noise smaller than the signal, and ensuring a dynamic range is particularly difficult with variable gain amplifiers.

[0027] (3) The issue of the response speed of the auto-balance is explained. The required modulation speed of the pump light Pu is "movement of the measurement point on the sample < auto-balance < pump light intensity modulation." The speed of the measurement point on the sample must be slower than the auto-balance speed, allowing the auto-balance to track changes in transmittance. Furthermore, the auto-balance speed must be slower than the pump light intensity modulation speed, allowing the probe signal to be observed even after balance correction. If the auto-balance speed is faster than the pump light intensity modulation speed, the amplitude of the probe signal is canceled out by the correction, effectively eliminating the probe signal. Furthermore, the pump light intensity modulation may not be fast enough. The pump light intensity modulation is limited by the response speed of the sample in a photothermal microscope, the optical modulator, and the pulse repetition frequency in the case of a pulsed light source. Furthermore, the speed of the measurement point on the sample is limited by the response speed of the auto-balance.

[0028] By applying the method of this embodiment to the above problems of the conventional methods, all of problems (1) to (3) can be solved simultaneously. In the method of this embodiment, the noise itself that appears when the balance between the signal strength of the probe signal and the signal strength of the reference signal is lost (when there is an imbalance) is used as the error signal for auto-balancing, rather than the DC offset after difference. When the light source noise contained in the probe signal of the probe light Pr that appears during output when there is an imbalance is multiplied by the light source noise (replica) contained in the reference signal of the reference light, only the same noise becomes a DC component out of the AC component of the noise. This DC component is used as the error signal for auto-balancing. The variable gain amplifier 16 uses the error signal to control the probe signal so as to correct the difference in signal strength between the probe signal and the reference signal.

[0029] We will explain why each problem can be solved. (1) Regarding the DC offset problem, the noise appearing in the light source is an AC component, so it is unrelated to the DC offset of the circuit elements. Therefore, there is no problem in removing the DC component appearing in the circuit elements. (2) Regarding the dynamic range problem, the noise and the signal level from the sample are close, so the dynamic range is not consumed. The large DC component in the light detection signal, which was necessary in the conventional method and consumes a large amount of dynamic range, can be removed. (3) Regarding the autobalance response speed problem, because the noise is uncorrelated with the sample signal, the sample signal is not included in the error signal. Therefore, with autobalance using the error signal, even if the autobalance response speed is increased, the sample signal will not be canceled out by the autobalance, so the autobalance response speed can be increased.

[0030] 6 is a diagram showing the configuration of a signal processing device 10 of this embodiment. In addition to the configuration of a device that performs automatic balance control, the signal processing device 10 includes band-pass filters 18 (18A and 18B), low-noise amplifiers 19 (19A and 19B), high-pass filters 20 (20A and 20B), and a multiplier 21. The band-pass filter 18A is an example of the first band-pass filter of the present disclosure, and the band-pass filter 18B is an example of the second band-pass filter of the present disclosure. The low-noise amplifier 19A is an example of the first amplifier of the present disclosure, and the low-noise amplifier 19B is an example of the second amplifier of the present disclosure. The high-pass filter 20A is an example of the first high-pass filter of the present disclosure, and the high-pass filter 20B is an example of the second high-pass filter of the present disclosure.

[0031] The bandpass filter 18 will now be described. The bandpass filter 18 removes unnecessary components from the photodetection signal. The unnecessary components are the DC component of the signal and other components not required for measurement. The bandpass filter 18A removes unnecessary components from the probe signal. The bandpass filter 18B removes unnecessary components from the reference signal.

[0032] The low-noise amplifier 19 will now be described. The low-noise amplifier 19 amplifies the signal with a higher signal-to-noise ratio than the variable gain amplifier 16. Because unnecessary components are removed by the band-pass filter 18, the signal and noise due to light source fluctuations in the required band can be amplified without saturation. The low-noise amplifier 19A amplifies the probe signal. The low-noise amplifier 19B amplifies the reference signal.

[0033] The band-pass filters 18A and 18B and the low-noise amplifiers 19A and 19B are components that mainly contribute to solving the dynamic range problem (2).

[0034] The probe signal is amplified by the low-noise amplifier 19 A, and then amplified by the variable gain amplifier 16 before being input to the high-pass filter 20 .

[0035] The high-pass filter 20 will now be described. The high-pass filter 20 removes DC offset components of circuit elements. This means that unstable DC offsets from the variable gain amplifier 16 and the like are no longer relevant to control. The high-pass filter 20A removes DC offsets from the probe signal. The high-pass filter 20B removes DC offsets from the reference signal. The high-pass filter 20 is mainly configured to contribute to solving the (1) DC offset problem.

[0036] The probe signal output from high-pass filter 20A and the reference signal output from high-pass filter 20B are input to difference calculator 15, and difference calculator 15 outputs noise that is the difference between the probe signal and the reference signal when there is an imbalance.

[0037] The multiplier 21 receives the differential noise output from the differentiator 15 and the reference signal output from the high-pass filter 20B. The multiplier 21 multiplies the differential noise by the reference signal to obtain an error signal. The automatic gain controller 17 automatically adjusts the gain for controlling the probe signal based on the error signal obtained by the multiplier 21 so as to correct the signal intensity error between the probe signal and the reference signal. The variable gain amplifier 16 controls the intensity of the probe signal according to the gain adjusted by the automatic gain controller 17. The automatic gain controller 17 and the multiplier 21 are mainly configured to solve the problem of (3) the response speed of automatic balancing. This makes it possible to achieve automatic balancing with a response speed that exceeds the modulation speed of the pump light.

[0038] As described above, the signal processing device 10 according to the embodiment of the present disclosure can solve the problems related to errors and limitations that occur in measurements using the pump-probe method.

[0039] The present disclosure is not limited to the above-described embodiment, and various modifications and applications are possible without departing from the spirit and scope of the present invention.

[0040] For example, the variable gain amplifier 16 and automatic gain controller 17 may be provided on the reference signal side instead of the probe signal side. In this case, the reference signal is amplified by the low-noise amplifier 19B, then amplified by the variable gain amplifier 16, and input to the high-pass filter 20. The automatic gain controller 17 controls the reference signal using the error signal obtained by the multiplier 21, thereby correcting the signal strength error between the probe signal and the reference signal.

[0041] Furthermore, a variable gain amplifier may be provided not only on the probe signal side but also on the reference signal side. This allows the frequency characteristics of the probe signal and the reference signal to be kept nearly identical, resulting in efficient noise reduction across a wide bandwidth. Furthermore, an automatic gain control amplifier 22 may be inserted into the reference signal immediately before the multiplier. The automatic gain control amplifier 22 maintains a constant intensity of the reference signal input to the multiplier. This allows the signal processing device 10 to operate stably even when the state of the light source changes and the level of noise fluctuates. Figure 7 shows a modified configuration of the signal processing device 10 in which a variable gain amplifier and automatic gain control amplifier 22 are provided on the reference signal side. The probe signal side is provided with a variable gain amplifier 16A and an automatic gain controller 17, while the reference signal side is provided with a variable gain amplifier 16B and an automatic gain control amplifier 22 before the multiplier. In this configuration, the signal is amplified by the variable gain amplifier 16B and then filtered by a high-pass filter 20B. A constant is assigned to the variable gain amplifier 16B. Furthermore, the automatic gain control amplifier 22 controls the reference signal output from the high-pass filter 20B to be input to the multiplier 21 so that the average signal strength is kept constant.

[0042] In the above embodiment, all configurations capable of solving all of the problems of (1) DC offset, (2) dynamic range, and (3) auto-balancing response speed have been described with reference to FIG. 6. However, the signal processing device 10 may have only a configuration that solves at least one of the three problems. For example, as a configuration that solves the problem of (3) auto-balancing response speed, the multiplier 21 and the automatic gain controller 17 in FIG. 6 may be included, but the band-pass filters 18A and 18B, the low-noise amplifiers 19A and 19B, and the high-pass filter 20 may be omitted. The band-pass filters 18A and 18B, the low-noise amplifiers 19A and 19B, and the high-pass filter 20 may be selectively applied. [Explanation of symbols]

[0043] 10. Signal Processing Device 11 Intensity Modulator 12A, 12D mirrors 12B Mirror or beam combiner 12C Optical Filter 13 Beam splitter 14(14A, 14B) Receiver 15 Differentiator 16(16A, 16B) Variable Gain Amplifier 17 Automatic Gain Controller 18(18A, 18B) Bandpass Filter 19(19A, 19B) Low noise amplifier 20(20A, 20B) High-pass filter 21 Multiplier 22 Automatic Gain Control Amplifier

Claims

1. A multiplier for obtaining an error signal by multiplying a reference signal obtained from a reference light by noise that appears as a difference between the signal strength of the probe signal and the signal strength of the reference signal when the signal strength of the probe signal is not balanced, the probe signal being obtained by irradiating a sample with a predetermined pump source in a pump-probe method to excite it and irradiating the sample with probe light, preventing the pump source from entering a photodetector using a predetermined filter and causing the probe light to enter the photodetector; and an automatic gain controller that automatically adjusts a gain for controlling at least one of the probe signal and the reference signal, based on the error signal obtained by the multiplier, so as to correct a signal strength difference between the probe signal and the reference signal; A signal processing device comprising:

2. a first high-pass filter and a second high-pass filter, which are high-pass filters for removing the effects of a DC offset in the circuit; The first high-pass filter removes a DC offset of noise in the probe signal; The signal processing device according to claim 1 , wherein the second high-pass filter removes a DC offset of noise in the reference signal.

3. a first band-pass filter that removes predetermined unnecessary components from the probe signal; a second band-pass filter that removes predetermined unnecessary components from the reference signal; a first amplifier that amplifies the probe signal that has passed through the first band-pass filter; a second amplifier that amplifies the reference signal that has passed through the second band-pass filter; a first variable gain amplifier that controls the strength of a signal passing through the first amplifier; Further comprising: the first amplifier and the second amplifier amplify a signal with a higher signal-to-noise ratio than the first variable gain amplifier; the first high-pass filter removes a DC offset of noise in the probe signal transmitted through the first variable gain amplifier; The signal processing device according to claim 2 , wherein the second high-pass filter removes a DC offset of noise in the reference signal transmitted through the second amplifier.

4. further including a second variable gain amplifier separate from the first variable gain amplifier; 4. The signal processing device according to claim 3, wherein the reference signal is filtered out by the second band-pass filter, amplified by the second amplifier, and filtered out by the second high-pass filter after being amplified by the second variable gain amplifier.

5. the automatic gain controller is a first automatic gain controller that controls the probe signal; further including an automatic gain control amplifier that controls the reference signal, separate from the first automatic gain controller; The signal processing device according to claim 4 , wherein the automatic gain control amplifier controls an average signal strength of the reference signal, from which the signal has been removed by the second high-pass filter, that is input to the multiplier.

Images