Calculation reconstruction spectrum detection device and method based on random structure

Abstract

The invention discloses a calculation reconstruction spectrum detection device and method based on a random structure. The device comprises an input system, a light splitting system, a random structure waveguide and a two-dimensional photoelectric detector, the input system is a light guide waveguide, the light splitting system is tightly connected with the light guide waveguide and splits light of the light guide waveguide, the random structure waveguide is connected with the light splitting system, and spectrum transmission efficiencies are different. After light is output from the random structure waveguide, interference is carried out on the two-dimensional photoelectric detector, and an interference pattern with spectral response is obtained. According to the invention, the incident light spectrum is reconstructed by using the interference pattern with different transmission efficiencies of input light with different wavelengths in a random structure and high wavelength sensitivity at the output end, and the sampling rate is improved by more than two orders of magnitude, so that higher spectral resolution and reconstruction precision are achieved. Compared with a traditional spectrograph, the spectrograph has the advantages of being few in device, compact in structure and small in size, the system complexity and cost of the spectrograph are effectively reduced, and the portability of the spectrograph is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of spectrometers, and in particular to a spectrum detection device and method based on random structure calculation reconstruction. Background Art

[0002] Spectrometers play a vital role in many applications, such as astronomical detection, biological and chemical characterization, and material analysis. Currently, there is increasing attention on miniaturized spectrometers due to their small size, light weight, and low energy consumption. As a result, two major categories of miniaturized spectrometers have emerged. The first category is based on on-chip dispersion, such as arrayed waveguide grating (AWG) and on-chip step grating spectrometers. The second category is based on on-chip Fourier transform, such as standing wave integrated Fourier transform spectrometers. Both types of spectrometers have relatively small sizes. However, the principles of these spectrometers are similar to those of conventional spectrometers. To achieve higher resolution, the size must be larger. This does not overcome the limitation between size and resolution, which imposes restrictions on the design of ultra-compact spectrometers.

[0003] In recent years, a new type of spectrometer, called reconstruction spectrometer, has emerged, which combines algorithms to overcome traditional limitations in resolution and size. Its working principle involves using a device or structure with a spectral response to establish a mapping relationship between the spectrum and the detected signal. This relationship is stored in a transfer matrix. By obtaining the transfer matrix through pre-calibration, the input spectrum can be reconstructed from the detected signal. Structures with wavelength-dependent properties, such as quantum dots, disordered scattering media, photonic crystal-based filter arrays, and even single nanowires, have been proposed and verified for reconstructing spectra. At present, reconstruction spectrometers have great development prospects.

[0004] The current computational reconstruction spectrometer is greatly limited by the number of detection channels. To achieve higher performance, a larger number of channels, such as 32 or 64, is required. First, this leads to an increase in device size, and the large number of waveguides makes the design difficult; second, the computational reconstruction spectrometer is essentially a filter, and since it is divided into multiple channels, it will cause energy loss, which is an important reason for the limitation of the computational reconstruction spectrometer and a problem that needs to be solved urgently. Summary of the invention

[0005] The present invention proposes a spectrum detection device and method based on random structure computational reconstruction. The present invention uses random structure and detects interference light field to obtain different spectral responses, uses spectrum reconstruction to obtain target spectrum, and reconstructs spectrum with high resolution and high speed. It does not use grating prism filter and other spectral elements, has few devices, compact spatial optical path and small volume, thus solving the defects of traditional spectrometers in that they have high precision requirements for optical components, large volume, expensive and inconvenient to carry, greatly increases spectrum sampling, and obtains high-precision spectrum with small structure.

[0006] The technical solution adopted by the present invention is:

[0007] A spectrum detection device based on random structure computational reconstruction comprises an input system, a spectroscopic system, a random structure waveguide and a two-dimensional photodetector. The input system is a light-guiding waveguide. The spectroscopic system is closely connected to the light-guiding waveguide and splits the light of the light-guiding waveguide. The random structure waveguide is connected to the spectroscopic system. The spectral transmission efficiency of the random structure waveguide is different. Light interferes at the output end of the random structure waveguide to obtain the light field intensity distribution on different pixels of the two-dimensional photodetector, thereby obtaining a high-sensitivity interference pattern with different spectral responses.

[0008] Furthermore, the optical splitting system includes a first-level optical splitting system and a second-level optical splitting system, the first-level optical splitting system is coupled from the input, the second-level optical splitting system is coupled from the first-level optical splitting system, and the random structure waveguide is connected to the second-level optical splitting system.

[0009] Furthermore, the input system is structured so that the light becomes stable light after propagation, so that the spot and power of the input light remain stable; the width of the single-mode waveguide satisfies that the light propagation is single-mode in the working band of the spectrometer, the thickness of the single-mode waveguide is the same as the width, and the length of the single-mode waveguide ensures that the light propagates stably as single-mode during the propagation process, thereby maintaining the stability of the input light field.

[0010] Furthermore, the material of the waveguide is a transparent material in the working band, and the material used is silicon nitride, silicon, silicon dioxide, lithium niobate, III-V group semiconductor compounds or polymer materials or other materials that can be used to make optical waveguides.

[0011] Furthermore, the random structure waveguide has different spectral responses, which is used to obtain an interference pattern with high wavelength sensitivity at the output end.

[0012] Furthermore, the length and width of the basic unit of the random structure both have random amounts, or one of them has a random amount, or part of the whole number has a random amount, and the obtained random structure has different spectral responses.

[0013] Furthermore, the detection method is to sample the interference light field generated between the randomly structured waveguides to obtain richer spectral information.

[0014] A spectrum detection method based on random structure computational reconstruction, comprising the following steps:

[0015] 1) Calibration: allowing a number of calibration lights of known wavelengths and intensities within a band to enter a device, wherein the device is a random structure computational reconstruction spectrum detection device as described in any one of claims 1 to 7, light interferes at the output end of the waveguide, and a two-dimensional photodetector collects interference light signals in a spatial region output from the optical waveguide, obtains light field intensity distributions on different pixels of the two-dimensional photodetector, and stores the light field intensity values ​​obtained at different wavelengths in a transmission matrix as a spectral response function;

[0016] 2) Collect the intensity distribution of the side scattered light field: When light of unknown spectrum enters the device, the light interferes from the output end of the optical waveguide to obtain the intensity distribution of the light field on the two-dimensional photodetector;

[0017] 3) Reconstruct spectrum: The calibrated transmission matrix and the light field distribution of the unknown light form a linear equation, and the reconstructed spectrum of the light to be measured is obtained by solving the linear equation.

[0018] Furthermore, the equation used in the spectrum reconstruction process is as follows:

[0019]

[0020] in: : Spectral component matrix of input light; : The transmission matrix of the device, the spectral response matrix of the device to be calibrated; : The light field intensity distribution matrix obtained on the detector;

[0021] Expanding this equation yields:

[0022] .

[0023] Further, the specific steps include:

[0024] (1) Calibrate and obtain the transmittance function Tn at different positions of the device. Tn is the transmittance function of the nth spatial spot channel. The wavelength S of the incident light is continuously adjusted to obtain m Tn samples with equal wavelength intervals. These Tn constitute the transmission matrix T(n,m) of the device, which describes the spectral response of the device.

[0025] (2) Measure the device response of the light to be measured. Introduce the unknown light into the single-mode waveguide. After propagating to the random optical waveguide area, the light is output at the output end and interferes to form a light spot. The intensity In of the light spot is obtained on the two-dimensional photodetector. In is the light field intensity at the nth pixel point on the two-dimensional photodetector, corresponding to the nth spatial spectrum channel.

[0026] (3) Solve the unknown spectrum. The unknown spectrum S is calculated by solving the linear equation system get.

[0027] .

[0028] The beneficial effects of the present invention are:

[0029] The present invention is based on the random structure calculation reconstruction spectrum detection device. When in use, the signal of different spectral response is obtained through the random structure, and the interference light field at the output end of the waveguide is detected, which greatly reduces the number of waveguides required. Since the optical waveguide is used as the main structure, the volume of the device is small, only a few square millimeters, and the volume of the spectrometer can be reduced. The interference correlation between the random structure waveguides used is low, so a wavelength interference pattern with low correlation can be obtained, which is conducive to calculation and reconstruction. In addition, the whole device has high integration and corresponding stability, which is conducive to subsequent detection. In addition, there is no specific requirement for the spectral response of each random structure waveguide, and the interference correlation is low. Therefore, the tolerance for processing errors is high, which is conducive to processing, and it also avoids the defects of traditional spectrometers that have high requirements for the precision of optical components, large volume, and expensive and inconvenient to carry. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A schematic structural diagram of a spectrum detection device based on random structure computational reconstruction provided by an embodiment of the present invention.

[0031] Figure 2 A random structure diagram provided by an embodiment of the present invention.

[0032] Figure 3 A schematic diagram of the correlation of the spectrometer provided by an embodiment of the present invention over the entire waveband.

[0033] Figure 4 Schematic diagram of correlation under broadband according to an embodiment of the present invention.

[0034] Figure 5 Schematic diagram of correlation under narrow band according to an embodiment of the present invention.

[0035] Figure 6 It is a schematic diagram of a broadband equally spaced spectrum and an incident spectrum reconstructed and restored according to an embodiment of the present invention.

[0036] In the figure, 1. input system; 2. first-level splitting system; 3. second-level splitting system; 4. random structure waveguide; 5. two-dimensional photodetector; 6. random structure length; 7. random structure width. DETAILED DESCRIPTION

[0037] The present invention is further described in detail below through specific embodiments:

[0038] The present invention is not limited to the following specific embodiments. It should be understood that the following specific embodiments are only exemplary illustrations and explanations of the present invention and do not limit the protection scope of the present invention. Any simple changes or modifications made by adopting the design structure and concept of the present invention will fall within the protection scope of the present invention.

[0039] A computationally reconstructed spectrum detection device based on random structure comprises an input system, a spectroscopic system (specifically comprising a first-level spectroscopic system and a second-level spectroscopic system), a random structure waveguide, and a two-dimensional photoelectric detector.

[0040] Preferably, the input system is a light-guiding waveguide (a single-mode waveguide is specifically used in this embodiment), the width of the single-mode waveguide satisfies that light propagation is single-mode in the working band of the spectrometer, the thickness of the single-mode waveguide is the same as the width, and the length of the single-mode waveguide can ensure that the light propagates stably in single-mode during the propagation process, thereby maintaining the stability of the input light field.

[0041] Preferably, the material of the single-mode waveguide is a transparent material or a low-absorption material in the working band. The single-mode waveguide material is selected as silicon nitride in the present invention, but is not limited to silicon, silicon dioxide, lithium niobate, III-V semiconductor compounds or polymer materials.

[0042] Preferably, the light splitting system is closely connected to the light guiding waveguide and splits the light of the light guiding waveguide.

[0043] Preferably, the first-stage splitting system is designed to satisfy the requirement of even distribution of light energy. The material of the first-stage splitting system is the same as that of the single-mode waveguide. The material selected in the present invention is silicon nitride. The material of the waveguide is a transparent material in the working band, and silicon nitride is used, but is not limited to the use of silicon, silicon dioxide, lithium niobate, III-V semiconductor compounds or polymer materials.

[0044] Preferably, the design of the second-stage optical splitting system does not need to follow the equal distribution of light energy, but it is necessary to ensure that each waveguide has energy input. The material of the second-stage optical splitting system is the same as that of the single-mode waveguide. The material of the present invention is silicon nitride. The material of the waveguide is a transparent material in the working band, and silicon nitride is used, but it is not limited to silicon, silicon dioxide, lithium niobate, III-V semiconductor compounds or polymer materials.

[0045] Preferably, the random structure waveguide is connected to the spectroscopic system. The spectral transmission efficiency of the random structure waveguide is different. The random structure waveguide is generated by simulation software. When generating different random structures, its random parameters are adjusted to a certain extent. Finally, a certain number of random structure waveguides with low interference correlation are selected.

[0046] Preferably, the photodetector is a two-dimensional detector to fully sample the interference light field behind the random structure.

[0047] A spectrum detection method based on random structure computational reconstruction, the steps of which include:

[0048] 1) Calibration: Allow several calibration lights of known wavelength and intensity within the band to enter the device and output them at the output end of the optical waveguide. The intensity distribution of the interference pattern is obtained through a two-dimensional photodetector. The intensity values ​​obtained at this series of different wavelengths are stored in the transmission matrix as a spectral response function.

[0049] 2) Collect the light field interference intensity distribution at the waveguide output end: When light of unknown spectrum enters the device, the light is output from the output end of the optical waveguide and interferes to obtain the light field intensity distribution on the two-dimensional photodetector;

[0050] 3) Reconstruct spectrum: The calibrated transmission matrix and the light field distribution of the unknown light form a linear equation, and the reconstructed spectrum of the light to be measured is obtained by solving the linear equation.

[0051] Preferably, the two-dimensional photodetector collects the optical signal output from the output end of the optical waveguide and interferes.

[0052] Preferably, the equation used in the spectrum reconstruction process is as follows:

[0053]

[0054] In the above formula:

[0055] : Spectral component matrix of input light;

[0056] : The transmission matrix of the device, the spectral response matrix of the device to be calibrated;

[0057] : The light field intensity distribution matrix obtained on the detector.

[0058] Expanding this equation yields:

[0059]

[0060] Preferably, the spectrum reconstruction process has the following detailed steps:

[0061] 1. Calibrate to obtain the transmittance function Tn at different positions of the device. Tn is the transmittance function of the nth spatial spot channel. The wavelength S of the incident light is continuously adjusted to obtain m equal-wavelength interval sampling Tn. These Tn constitute the transmission matrix T(n,m) of the device, which describes the spectral response of the device. T(n,m) is a two-dimensional matrix:

[0062]

[0063] 2. Measure the device response of the light to be measured. Introduce the unknown light into the single-mode waveguide. After propagating to the random optical waveguide area, the light will be output at the output end and interfere to form a light spot. The intensity In of the light spot can be obtained on the two-dimensional photodetector. In is the light field intensity at the nth pixel point on the two-dimensional photodetector, corresponding to the nth spatial spectrum channel. The light field intensity distribution I is a one-dimensional matrix with a length of n:

[0064]

[0065] 3. Solve the unknown spectrum. The unknown spectrum S is solved by solving the linear equation system Obtain. S is a one-dimensional matrix of length m:

[0066]

[0067] Preferably, m≥n.

[0068] like Figure 1 , showing the specific structure of a specific embodiment of the present invention. The defect scattering calculation reconstruction spectrum detection device of the present invention is mainly composed of an input system 1, a first-level spectroscopic system 2, a second-level spectroscopic system 3, a random structure waveguide 4, and a two-dimensional photodetector 5.

[0069] When in use, the input system uses a single-mode waveguide, and the coupling of the incident light is achieved through a pair of optical fiber lenses. After the incident light propagates through the single-mode waveguide, it stabilizes as a single-mode light, and then enters the first-level optical splitting system 2. The optical waveguide is a single-mode waveguide, and energy equalization is achieved, and then enters the second-level optical splitting system 3 to achieve further energy dispersion. At this stage, energy equalization does not need to be achieved, and it is sufficient to ensure that each waveguide has energy. Then it enters the random structure waveguide 4 to generate a spectral response signal with low interference correlation, and then interference is performed and the interference pattern is collected by the detector 5. After the device is calibrated, the light spot generated by the unknown light can be used to obtain the spectrum of the unknown light through the reconstruction algorithm.

[0070] In a specific embodiment, the materials of the input system, the first-stage spectroscopic system, the second-stage spectroscopic system and the random structure waveguide are all selected to be silicon nitride. In addition to silicon nitride, any material that remains transparent in the working band of the instrument can be used. For example, if used near the infrared band, the material of the waveguide can also be silicon. The cross-sectional size of the input system is 1μm×1μm, and the length of the waveguide is 200 μm. This is to maintain the stability of the input light in the working band of the device. The working band of the device is 1.52-1.56 μm. Within the working band, the input system can make all wavelengths of input light be converted into single mode after propagation by coupling into the single-mode waveguide, and maintain the stability of an input light spot during spectrometry, thereby achieving the stability of spectrometry.

[0071] The cross-sectional size of the first-stage splitting system is 1 μm×1 μm, and the length of the waveguide is 450 μm. After the light is introduced from the input system, it is evenly split at the first-stage splitting system and is stably transmitted before entering the second-stage splitting system.

[0072] The cross-sectional size of the second-level splitting system is 1 μm×1 μm, and the length of the waveguide is 300 μm. The light is coupled from the first-level splitting system to the second-level splitting system for further splitting, ensuring that all 10 waveguides have energy and are stably transmitted before entering the random structure.

[0073] like Figure 2 As shown, the random structure waveguide has a waveguide length of 300 μm, and each small structure is in the form of a rectangle, which is modulated by a random structure length of 6 and a random structure width of 7 to obtain an overall random structure waveguide with low interference correlation.

[0074] like Figure 3 As shown, each random structure waveguide will reflect a different spectral response with low interference correlation, and then interfere at the output end, increasing the spectral sensitivity of the entire structure.

[0075] In a specific embodiment, the input system, the first-stage light splitting system, the second-stage light splitting system and the random structure waveguide are made of silicon nitride with a thickness of 150 nm.

[0076] The waveguide is made by plasma enhanced chemical vapor deposition (PECVD), and the defect is made by reactive ion etching (RIE). The gas used in the reactive ion etching is and In dry etching, the active free radical molecules and atoms used in etching cause chemical reactions to form volatile substances, which cause the etched material to be stripped off. The free radicals are not affected by the electric field and are in thermal motion in all directions. The etching is isotropic, and it is easy to obtain the required random structure waveguide.

[0077] The spectrum reconstruction comprises the following steps:

[0078] In a specific embodiment, a continuously adjustable light source is used to calibrate the device, and the equation used during calibration is as follows:

[0079]

[0080] In the above formula:

[0081] : Spectral component matrix of input light;

[0082] : The transmission matrix of the device, the spectral response matrix of the device to be calibrated;

[0083] : The interference spot intensity matrix is ​​obtained on the detector.

[0084] Expanding this equation yields:

[0085]

[0086] During calibration, input monochromatic light of known wavelength and intensity The corresponding light intensity distribution can be obtained on the two-dimensional photodetector , these two known values ​​can be used to calibrate The components in the transmission matrix are used to obtain the entire transmission matrix of the device by inputting monochromatic light with equally spaced wavelengths within the working band of the device. m is the number of spectral components, n is the number of pixels used in the calculation of the two-dimensional photodetector, m ≥ n.

[0087] Finish After the transfer matrix is ​​calibrated, the unknown spectrum is input again. , the spot intensity distribution function is obtained on the sensor through scattering of the side defects of the device , by solving the equation , the components of the unknown spectrum can be obtained. Figure 4 This figure shows the correlation of the spectrometer at 1.52-1.56 μm. The half-width δλ at half-peak represents its theoretical resolution of 3.5 nm. Experiments were conducted for light with a wavelength range of 1.550-1.552 μm, such as Figure 5 As shown in the figure, it can be seen that it has a higher resolution of 0.4 nm in a small bandwidth range. Figure 6 As shown, a single peak reconstruction result of 1.52-1.56 μm can be obtained in the entire working band.

[0088] In summary, the present invention provides a computationally reconstructed spectral detection device and method based on random structures. The device includes an input system, a first-stage spectroscopic system, a second-stage spectroscopic system, a random structure waveguide, and a two-dimensional photodetector. The technical solution of the present invention utilizes the different transmission efficiencies of input lights of different wavelengths in the random structure and the interference pattern of high wavelength sensitivity at the output end to reconstruct the incident light spectrum. Compared with the existing random structure spectral chip, the sampling rate is increased by more than 2 orders of magnitude, so it has higher spectral resolution and reconstruction accuracy. Compared with traditional commercial spectrometers, it has the advantages of fewer components, compact structure, and small size. It effectively reduces the system complexity and cost of the spectrometer and improves the portability of the spectrometer.

[0089] The above embodiment is only one example to fully illustrate the principle and spirit of the present invention. Those skilled in the art can make various changes and modifications to the described method. The technical scope of the present invention is not limited to the contents of the specification, and the protection scope of the present invention is subject to the claims.

Claims

1. A spectrum detection device based on random structure computational reconstruction, characterized in that: The invention comprises an input system, a light splitting system, a random structure waveguide and a two-dimensional photodetector. The input system is a light guiding waveguide. The light splitting system is closely connected to the light guiding waveguide and splits the light of the light guiding waveguide. The random structure waveguide is connected to the light splitting system. The spectral transmission efficiency of the random structure waveguide is different. Light interferes at the output end of the random structure waveguide to obtain the light field intensity distribution on different pixels of the two-dimensional photodetector, thereby obtaining a high-sensitivity interference pattern with different spectral responses.

2. The random structure-based computational reconstruction spectrum detection device according to claim 1, characterized in that: The optical splitting system comprises a first-level optical splitting system and a second-level optical splitting system. The first-level optical splitting system is coupled from the input, the second-level optical splitting system is coupled from the first-level optical splitting system, and the random structure waveguide is connected to the second-level optical splitting system.

3. The random structure-based computational reconstruction spectrum detection device according to claim 1, characterized in that: The input system is structured so that the light becomes stable light after propagation, so that the spot and power of the input light remain stable; the width of the single-mode waveguide satisfies that the light propagation is single-mode in the working band of the spectrometer, the thickness of the single-mode waveguide is the same as the width, and the length of the single-mode waveguide ensures that the light propagates stably in single-mode during propagation, thereby maintaining the stability of the input light field.

4. The random structure-based computational reconstruction spectrum detection device according to claim 1, characterized in that: The material of the waveguide is a transparent material in the working band, and the material used is silicon nitride, silicon, silicon dioxide, lithium niobate, III-V group semiconductor compounds or polymer materials or other materials that can be used to make optical waveguides.

5. The random structure-based computational reconstruction spectrum detection device according to claim 1, characterized in that: The random structure waveguide has different spectral responses and is used to obtain an interference pattern with high wavelength sensitivity at the output end.

6. The random structure-based computational reconstruction spectrum detection device according to claim 1, characterized in that: The length and width of the basic unit of the random structure both have random amounts, or one of them has a random amount, or part of the whole number has a random amount, and the obtained random structure has different spectral responses.

7. The random structure-based computational reconstruction spectrum detection device according to claim 1, characterized in that: The detection method is to sample the interference light field generated between random structure waveguides to obtain richer spectral information.

8. A spectrum detection method based on random structure computational reconstruction, characterized in that the steps include: 1) Calibration: allowing a number of calibration lights of known wavelengths and intensities within a band to enter a device, wherein the device is a random structure computational reconstruction spectrum detection device as described in any one of claims 1 to 7, light interferes at the output end of the waveguide, and a two-dimensional photodetector collects interference light signals in a spatial region output from the optical waveguide, obtains light field intensity distributions on different pixels of the two-dimensional photodetector, and stores the light field intensity values ​​obtained at different wavelengths in a transmission matrix as a spectral response function; 2) Collect the intensity distribution of the side scattered light field: When light of unknown spectrum enters the device, the light interferes from the output end of the optical waveguide to obtain the intensity distribution of the light field on the two-dimensional photodetector; 3) Reconstruct spectrum: The calibrated transmission matrix and the light field distribution of the unknown light form a linear equation, and the reconstructed spectrum of the light to be measured is obtained by solving the linear equation.

9. The method for spectrum detection based on random structure computational reconstruction according to claim 8, characterized in that: The equation used in the spectrum reconstruction process is as follows: in: : Spectral component matrix of input light; : The transmission matrix of the device, the spectral response matrix of the device to be calibrated; : The light field intensity distribution matrix obtained on the detector; Expanding this equation yields: 。 10. The method for spectrum detection based on random structure computational reconstruction according to claim 9, characterized in that: The specific steps include: (1) Calibrate and obtain the transmittance function Tn at different positions of the device. Tn is the transmittance function of the nth spatial spot channel. The wavelength S of the incident light is continuously adjusted to obtain m Tn samples with equal wavelength intervals. These Tn constitute the transmission matrix T(n,m) of the device, which describes the spectral response of the device. (2) Measure the device response of the light to be measured. Introduce the unknown light into the single-mode waveguide. After propagating to the random optical waveguide area, the light is output at the output end and interferes to form a light spot. The intensity In of the light spot is obtained on the two-dimensional photodetector. In is the light field intensity at the nth pixel point on the two-dimensional photodetector, corresponding to the nth spatial spectrum channel. (3) Solve the unknown spectrum. The unknown spectrum S is calculated by solving the linear equation system get.

Images