Reconstruction method for multiple input spectra, and computational reconstruction type on-chip spectrometer

Abstract

A reconstruction method for multiple input spectra, and a computational reconstruction type on-chip spectrometer. The on-chip spectrometer used has multiple input ports (1) and output ports (3), and the core region is an arrayed waveguide containing thermo-optic tunable components, such that parallel detection for multiple spectra under test can be achieved, and simultaneously a reconstruction model is introduced to synchronously establish a mapping relationship between the spectra under test and measured patterns under multi-port input, thereby realizing a spectrometer system for efficient parallel detection for multiple spectra under test. High-resolution and large-window parallel detection for multiple spectra under test can be achieved by means of a single spectrometer, thereby significantly improving detection efficiency.

Description

A multi-input spectrum reconstruction method and a computing reconstruction type on-chip spectrometer TECHNICAL FIELD

[0001] The present application relates to a spectrum reconstruction method, in particular to a multi-input spectrum reconstruction method and a computing reconstruction type on-chip spectrometer for realizing multi-spectrum parallel detection. BACKGROUND

[0002] As an important scientific research means, the spectrometer has been widely used in agricultural detection, medical analysis, astronomical research and optical coherence tomography and other fields. Generally, in various application scenarios, it is necessary to realize parallel detection of multiple spectrum data while maintaining high resolution and large bandwidth.

[0003] For example, in the field of aerospace, when large-scale spectrum detection is performed using a large field of view telescope, it is necessary to divide the large aperture into multiple sub-views and connect them in parallel with hundreds of spectrometer systems through optical fibers. The increase in the number of spectrometers means higher detection efficiency. At the same time, it is necessary to maintain high resolution to distinguish finer features in order to reveal key information about the nature and composition of celestial bodies, and it is also necessary to have a larger working window to observe more celestial bodies or more features of the same celestial body.

[0004] However, the traditional spectrometer system faces a bottleneck due to its large size and expensive components. Notably, multi-target detection techniques often require array-based spectrometer systems, which exacerbate this bottleneck. In recent years, although on-chip integrated spectrometers have gained great attention due to their compact size and cost-effectiveness, they still rely on array stacking of functional units. This is often limited by large-scale optical systems, complex control circuits, and high costs. Furthermore, achieving high resolution and large bandwidth in current spectrum parallel detection has proven to be difficult. Therefore, it is of great significance to propose a new multi-spectrum parallel detection architecture to realize high resolution, large window, and multi-spectrum parallel detection within a single spectrometer.

[0005] SUMMARY

[0006] In view of the problems in the background art, the present application aims to provide a computing reconstruction type on-chip spectrometer that supports high resolution, large window, and multi-spectrum parallel detection. The present application proposes an on-chip spectrometer containing a thermo-optic tunable component, which has multiple input / output ports and a core area that realizes parallel detection of multiple spectra to be measured. A reconstruction method is introduced to simultaneously establish the mapping relationship between the spectra to be measured and the measured patterns under multi-port input, thereby realizing efficient parallel detection of multiple spectra to be measured. The present application can realize high resolution, large window, and parallel detection of multiple spectra to be measured with a single spectrometer, greatly improving the detection efficiency.

[0007] The technical scheme adopted by the present application is:

[0008] One is a multi-input spectrum reconstruction method

[0009] Step 1: Lay a thermo-optic tunable component in the array waveguide region of the on-chip spectrometer, specifically, lay a corresponding thermo-optic tunable component at each waveguide in the array waveguide region, so that the phase distribution of light on each waveguide is adjustable;

[0010] Step 2: Randomly activate the thermo-optic tunable component, and then calibrate the calibration matrix of the current on-chip spectrometer;

[0011] Step 3: Keep the same random activation state as step 2, and simultaneously input multiple to-be-measured spectra to the current on-chip spectrometer, and output a detection spectrum in the on-chip spectrometer;

[0012] Step 4: According to the calibration matrix of the current on-chip spectrometer and the detection spectrum, the multiple to-be-measured spectra are respectively optimized and reconstructed to obtain multiple reconstructed spectra.

[0013] The reconstruction method further comprises the following steps:

[0014] Step 5: If the accuracy of the current reconstructed spectrum does not reach the target accuracy, repeat steps 2-4 to reconstruct the spectrum again until the reconstructed spectrum with the optimal accuracy is obtained.

[0015] In step 2, the calibration matrix of the current on-chip spectrometer is calibrated, specifically:

[0016] Input multiple known spectra to the current on-chip spectrometer, measure the output spectrum; calculate the calibration matrix of the current on-chip spectrometer according to the multiple known spectra and the corresponding output spectrum.

[0017] In step 4, the optimization target of the kth to-be-measured spectrum is constructed, and the kth to-be-measured spectrum is iteratively optimized according to the constructed optimization target, so that the optimization target is minimized to obtain the reconstructed spectrum corresponding to the kth to-be-measured spectrum. The formula of the optimization target of the kth to-be-measured spectrum is as follows:

[0018] Wherein, is the reconstructed spectrum, O N×1 is the detection spectrum, D() represents the difference operation, K is the number of to-be-measured spectra, α 1,k , α 2,k and α 3,k are the first weight to the third weight; A N×M,k is the calibration matrix of the kth to-be-measured spectrum, S M×1,k is the kth to-be-measured spectrum, and || ||2 is the L2 norm; || ||1 is the L1 norm.

[0019] The thermo-optic tunable component comprises an electrode.

[0020] The accuracy of the reconstructed spectrum is specifically a relative reconstruction error ε r , and the calculation formula is as follows:

[0021] Wherein, S M×1,k is the kth to-be-detected spectrum, is the reconstructed spectrum, and || ||2 is an L2 norm.

[0022] The number of input waveguides of the on-chip spectrometer is greater than or equal to the number of to-be-detected spectra.

[0023] Two, a computing reconstruction type on-chip spectrometer for multi-spectral parallel detection

[0024] The computing reconstruction type on-chip spectrometer comprises an on-chip spectrometer comprising a thermo-optic tunable component, a spectrometer calibration module and a spectrum optimization reconstruction module.

[0025] The spectrometer calibration module is configured to randomly activate the thermo-optic tunable component and obtain a calibration matrix of the spectrometer in the activated state.

[0026] The on-chip spectrometer comprising the thermo-optic tunable component is configured to generate detection spectra corresponding to a plurality of to-be-detected spectra.

[0027] The spectrum optimization reconstruction module is configured to reconstruct the plurality of to-be-detected spectra according to the calibration matrix and the detection spectra, and obtain reconstructed spectra.

[0028] The computing reconstruction type on-chip spectrometer further comprises a spectrum accuracy optimization module, which is configured to calculate the accuracy of the reconstructed spectrum and optimize the reconstructed spectrum.

[0029] The present application has the beneficial effects that:

[0030] The present application first proposes a waveguide-based multi-spectral parallel detection spectrometer. For multi-port input spectra, the combination of the on-chip spectrometer comprising the thermo-optic tunable component and the reconstruction model allows a larger working window and improves the resolution of spectral detection. In addition, the present application can realize the simultaneous reconstruction of a plurality of to-be-detected spectra from multiple objects by means of a single spectrometer, greatly improving the efficiency of spectral detection. BRIEF DESCRIPTION OF DRAWINGS

[0031] Fig. 1 is a method flowchart of the present application.

[0032] Fig. 2 is an architecture schematic diagram of the present application.

[0033] Fig. 3 is a schematic diagram of a single random spectrum response generation unit of an embodiment of the present application.

[0034] Figure 4 is a schematic diagram of the spectrum of the multispectral parallel probe spectrometer in the embodiment for restoring the four-channel input spectrum.

[0035] In the figure: 1 - input port, 2 - single random spectral response generation unit, 3 - output port, 4 - input waveguide, 5 - output waveguide, 6 - first flat area, 7 - array waveguide area with heating electrodes, 8 - second flat area. DETAILED DESCRIPTION

[0036] The application will be further described below in conjunction with the drawings and embodiments.

[0037] As shown in Figure 1, the reconstruction method of the multi-input spectrum proposed by the application includes the following steps:

[0038] Step 1: lay electrodes in the array waveguide area of the on-chip spectrometer, specifically lay corresponding electrodes at each waveguide in the array waveguide area, so that the phase distribution of light on each waveguide is adjustable;

[0039] Step 2: randomly activate the electrodes to increase the disorder of the spectrum, and then calibrate the calibration matrix of the current on-chip spectrometer;

[0040] In which, the calibration matrix of the current on-chip spectrometer is specifically:

[0041] Input multiple known spectra to the current on-chip spectrometer, measure the output spectrum; calculate the calibration matrix of the current on-chip spectrometer according to the multiple known spectra and the corresponding output spectrum.

[0042] Step 3: activate the electrodes in the activation mode in step 2, input multiple to-be-measured spectra to the current on-chip spectrometer through the multiple input waveguides of the on-chip spectrometer, and output the detected spectrum in the output waveguide of the on-chip spectrometer; each output port of the on-chip spectrometer will have an output spectrum line (the horizontal coordinate is the wavelength and the vertical coordinate is the energy), and all the output ports finally synthesize a matrix (X axis is the port number, Y axis is the wavelength, and Z axis is the energy), which is the detected spectrum.

[0043] Step 4: according to the calibration matrix and the detected spectrum, the multiple to-be-measured spectra are respectively optimized and reconstructed to obtain multiple reconstructed spectra.

[0044] The optimization and reconstruction in step 4 can be simply understood as establishing the connection between the measured pattern and the to-be-measured spectrum array by using the mapping network in the spectrometer, and the mapping network is a three-dimensional matrix A. Mathematically, the detected signal O N is expressed as the integral of the spectral response matrix A of the architecture and the incident unknown spectrum array S k (λ), that is:

[0045] Here, λstart and λ stop are the start and stop wavelengths of the working wavelength range, respectively. n,k (λ) is the calibration matrix. k (λ) is the incident unknown multispectral array, K is the number of spectra to be measured, and k is the incident port order. n,k is the signal detected by all output channels after a single spectrum is input from the kth channel. N represents the number of random module tuning states. n is the sum of signals detected by all output channels after multiple spectra are simultaneously incident, i.e., the actual detected optical power. Each layer of the three-dimensional spectral response matrix A k represents the spectral mapping under a specific kth port input. After uniformly discretizing the unknown spectrum into M columns, the above equation can be rewritten as:

[0046] where A N×M,K is a three-dimensional matrix representing the dependence of the spectral response of the architecture on the tuning state / wavelength / incident port order. k Each row vector in A k corresponds to the spectral response at a specific sampling step, and each column vector in A N×M,K represents the temporal speckle of a single wavelength. Since A N×1 matrix and O M×1,K can be measured experimentally, the solution of S N×1 can be directly obtained.

[0047] However, in this case, the system is generally unconstrained, so converting the spectral reconstruction problem into the processing of the inverse matrix requires optimization regularization techniques to specify a unique solution. Here, it is assumed that the spectral values are positive and exhibit continuous smooth characteristics, which is reasonable according to the properties of the spectrum.

[0048] In order to realize the spectral reconstruction of multiple spectra with narrowband, broadband, or even complex characteristics, a solver is needed to solve the current multispectral reconstruction problem. The present application proposes a target function for the current multispectral reconstruction problem. For the optimization reconstruction of the kth to-be-measured spectrum, the formula of the optimization target is as follows: according to the constructed optimization target, the kth to-be-measured spectrum is iteratively optimized, so that the kth to-be-measured spectrum corresponding to the reconstructed spectrum is obtained after minimizing the optimization target:

[0049] where, is the reconstructed spectrum, O N×1 is the detected spectrum, minSM×1,k, SM×1,k≥0(·) is the global optimal value of the to-be-measured spectrum S M×1,k , the output power at k is minimized and non-negative. The first term is used to minimize the detected spectrum O N×1 and the calculated α 1,k·A N×M,k ·S M×1,k The difference between the first regularization term, the second regularization term to promote the sparsity of the to-be-tested spectrum S M×1,k The third difference regularization term to smooth S M×1,k Here, α 1,k , α 2,k And α 3,k The weight of the corresponding term, the appropriate parameters α 1,k , α 2,k And α 3,k Are determined by using the standard k-fold cross-validation technique. D() represents the difference operation, K is the number of to-be-tested spectra, α 1,k , α 2,k And α 3,k The first weight to the third weight; A N×M,k The calibration matrix of the kth to-be-tested spectrum, S M×1,k The kth to-be-tested spectrum, || ||2 is the L2 norm, and the square root of the sum of the absolute values of the vector elements; || ||1 is the L1 norm, which is the sum of the absolute values of each element in the vector.

[0050] Step 5: If the accuracy of the current reconstructed spectrum does not meet the target accuracy, repeat steps 2-4 to reconstruct the spectrum again until the reconstructed spectrum with the optimal accuracy is obtained. Experiments show that the more irregular the electrode activation is, the higher the accuracy of the reconstructed spectrum is.

[0051] The accuracy of the reconstructed spectrum is specifically the relative reconstruction error ε r , and the calculation formula is as follows:

[0052] Where, S M×1,k The kth to-be-tested spectrum, The reconstructed spectrum, || ||2 is the L2 norm, and the square root of the sum of the absolute values of the vector elements.

[0053] The application also provides a multi-spectral parallel detection calculation reconstruction type on-chip spectrometer, which comprises an on-chip spectrometer comprising electrodes, a spectrometer calibration module and a spectrum optimization reconstruction module; wherein the parallel detection of the spectrum refers to the technology of synchronously collecting multiple spectral information of the same light source or different light sources by using a specially designed optical system and a detector array in the process of spectral measurement or analysis.

[0054] The spectrometer calibration module is used for randomly activating the electrodes and obtaining the calibration matrix of the spectrometer in the activated state;

[0055] The on-chip spectrometer comprising electrodes is used for generating detection spectra corresponding to multiple to-be-tested spectra;

[0056] The spectrum optimization reconstruction module is configured to reconstruct the plurality of to-be-detected spectra according to the calibration matrix and the detection spectrum to obtain reconstructed spectra.

[0057] The computing reconstruction on-chip spectrometer further comprises a spectrum accuracy optimization module configured to calculate the accuracy of the reconstructed spectrum and optimize the reconstructed spectrum.

[0058] As shown in FIG. 2, the on-chip spectrometer comprises a plurality of input ports 1, a single random spectrum response generation unit 2 and a plurality of output ports 3 connected in sequence.

[0059] FIG. 3 shows the single random spectrum response generation unit in a specific embodiment. The core device is a multi-port arrayed waveguide grating, which comprises a plurality of input waveguides 4, a plurality of output waveguides 5, a first slab region 6, a second slab region 8 and an arrayed waveguide region 7 with electrodes. The waveguide length of the arrayed waveguide region 7 gradually increases with a constant difference ΔL. Each waveguide in the arrayed waveguide region 7 is paved with a thermo-optic tunable section with a length of L t , that is, an electrode. Specifically, the input light of the plurality of to-be-detected spectra is input from the plurality of input waveguides 4, and all the input light is diffracted when passing through the first slab region 6. Different input spectra do not interfere with each other when being diffracted, and the diffracted input light is recoupled into the arrayed waveguide region for transmission. Due to the difference in physical length, the arrayed waveguide region will change the phase distribution of the light on each waveguide. In addition, the phase distribution can be further arbitrarily controlled by randomly activating a certain number of heating electrodes. Then the output light emitted from each output position of the arrayed waveguide will be subjected to multi-beam interference in the second slab region 8, which will result in an unordered pattern along the output waveguide of the slab region, that is, the detection spectrum. Finally, the reconstruction model is used to complete the restoration of the plurality of to-be-detected spectra.

[0060] The specific embodiments of the present application are as follows:

[0061] The silicon nanowire optical waveguide based on silicon insulator material is selected: the core layer is silicon material Si, the thickness is 220 nm, and the shallow etching layer thickness is 150 nm; the lower cladding layer and the upper cladding layer are both silica SiO2, and the thicknesses are both 2 μm. The key parameters of the arrayed waveguide grating in the embodiment are shown in Table 1.

[0062] Table 1 shows the key parameters of the arrayed waveguide grating

[0063] In the embodiment, firstly, 30 heating states are determined, 30 heating electrodes are randomly selected in each state, and a voltage of 3V is added. An input port is fixed each time, and 32 output spectra in 30 heating states are recorded respectively. There are 32 input ports in total. This is taken as the A matrix of the random module. Then, 4 input ports are randomly selected to simultaneously input the to-be-detected spectrum, and the output power of the 32 output ports is recorded. The spectrum restoration is realized through the calculation reconstruction method, and 4 to-be-detected spectra are obtained. As shown in FIG. 4, the embodiment can simultaneously realize the restoration of 4 to-be-detected spectra, and the working bandwidth reaches 100nm, the resolution reaches 0.02nm, and the relative reconstruction error is less than 0.13.

[0064] The above embodiments are used to explain and illustrate the present application, but not to limit the present application. Any modification and change made to the present application within the spirit and protection scope of the claims of the present application falls within the protection scope of the present application.

Claims

1. A method of reconstruction of a multi-input spectrum, characterized in that, The method comprises the following steps: Step 1: laying a thermo-optic tunable component in the array waveguide region of the on-chip spectrometer, specifically laying a corresponding thermo-optic tunable component at each waveguide in the array waveguide region, so that the phase distribution of light on each waveguide is adjustable; Step 2: randomly activating the thermo-optic tunable component, and then calibrating the calibration matrix of the current on-chip spectrometer; Step 3: while maintaining the same random activation state as in Step 2, inputting multiple to-be-detected spectra into the current on-chip spectrometer, and outputting detection spectra in the on-chip spectrometer; Step 4: according to the calibration matrix of the current on-chip spectrometer and the detection spectra, respectively optimizing and reconstructing the multiple to-be-detected spectra to obtain multiple reconstructed spectra.

2. The method of claim 1, wherein, The reconstruction method further comprises the following steps: Step 5: if the accuracy of the current reconstructed spectrum does not meet the target accuracy, repeating Steps 2-4 to reconstruct the spectrum again until a reconstructed spectrum with optimal accuracy is obtained.

3. The method of claim 1, wherein, In Step 2, the calibration matrix of the current on-chip spectrometer is calibrated, specifically: Input multiple known spectra into the current on-chip spectrometer, measure the output spectra, and calculate the calibration matrix of the current on-chip spectrometer according to the multiple known spectra and the corresponding output spectra.

4. The method of claim 1, wherein, In step 4, an optimization target of the kth to-be-measured spectrum is constructed, and the kth to-be-measured spectrum is iteratively solved according to the constructed optimization target, so that the reconstruction spectrum corresponding to the kth to-be-measured spectrum is obtained after the optimization target is minimized, and the formula of the optimization target of the kth to-be-measured spectrum is specifically as follows: wherein To reconstruct the spectrum, O N×1 To detect the spectrum, D() represents the difference operation, K is the number of to-be-detected spectra, α 1,k , α 2,k , and α 3,k are respectively the first weight to the third weight; A N×M,k is the calibration matrix of the kth to-be-detected spectrum, S M×1,k is the kth to-be-detected spectrum, and || ||2 is the L2 norm; || ||1 is the L1 The norm.

5. The method of claim 1, wherein, The thermo-optic tunable component comprises an electrode.

6. The method of claim 2, wherein, The accuracy of the reconstructed spectrum is specifically a relative reconstruction error ε r , and the calculation formula is as follows: wherein S M×1,k is the kth measured spectrum, For the reconstructed spectrum, || ||2 is the L2 norm.

7. The method of claim 1, wherein, The number of input waveguides of the on-chip spectrometer is greater than or equal to the number of to-be-detected spectra.

8. A computationally reconstructed on-chip spectrometer for multi-spectral parallel detection, characterized in that, The method comprises an on-chip spectrometer comprising a thermo-optic tunable component, a spectrometer calibration module, and a spectrum optimization and reconstruction module; The spectrometer calibration module is configured to randomly activate the thermo-optic tunable component and obtain the calibration matrix of the spectrometer in the activated state; The on-chip spectrometer comprising a thermo-optic tunable component is configured to generate detection spectra corresponding to the multiple to-be-detected spectra; The spectrum optimization and reconstruction module is configured to reconstruct the multiple to-be-detected spectra according to the calibration matrix and the detection spectra to obtain reconstructed spectra.

9. The computationally reconstructed on-chip spectrometer with multi-spectral parallel detection of claim 8, wherein, The reconstructed on-chip spectrometer further comprises a spectrum accuracy optimization module, which is configured to calculate the accuracy of the reconstructed spectrum and optimize the reconstructed spectrum.

Images