Frequency multiplexing spectrum analysis chip based on modulator array

Abstract

The invention discloses a frequency multiplexing spectrum analysis chip based on a modulator array. The frequency multiplexing spectrum analysis chip comprises an input end main waveguide, a filter array, an intensity modulator array, a total beam combiner and an optical detector. According to the scheme, a low-correlation transmission matrix is constructed through a filter array, transmission spectrums of all channels are different, and a broadband is covered. Periodic intensity modulation is applied to the light passing through the filter at different frequencies, and the modulated light intensity is combined. The intensity of each frequency channel can be obtained by carrying out demodulation processes such as Fourier transform on the collected time sequence optical signals, spectrum reconstruction is carried out by utilizing an intensity sequence and a corresponding measurement matrix, and spectrum information of an incident end can be restored.

Description

Technical Field

[0001] This invention relates to the field of spectrometer technology, and more specifically to a frequency multiplexing spectral analysis chip based on a modulator array. Background Technology

[0002] Traditional spectrometers rely on mechanical components for spectral dispersion, resulting in poor stability and large size, which limits their application scenarios and makes reliable measurements difficult outside the laboratory. In recent years, with the help of optical chip technology, miniature spectral detection systems implemented on-chip have gradually developed, offering significant advantages over traditional spectrometers in terms of size, stability, and power consumption.

[0003] Among these approaches, spectral reconstruction using filter arrays has been extensively studied and developed. This involves reconstructing the incident spectrum by comparing the light intensity data acquired from different arrays with the spectral response of the filters. However, such approaches typically require splitting the light into individual array channels and using detector arrays for data acquisition, resulting in a relatively low overall signal-to-noise ratio. Therefore, we propose a frequency-multiplexed spectral analysis chip based on a modulator array. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention proposes a frequency multiplexing-based spectral analysis chip with advantages such as high signal-to-noise ratio, high resolution, and single detector.

[0005] This invention is achieved through the following techniques: A frequency multiplexing spectral analysis chip based on a modulator array, comprising: Input main waveguide, The filter array is composed of structures such as micro-ring resonant cavities or waveguide filters. Each filter is coupled to the main waveguide and outputs the filtered optical signal to the modulator array. The intensity modulator array consists of N independent Mach-Zehnder intensity modulation units, which are used to apply periodic intensity modulation signals of different frequencies to different channels. Each modulation unit includes: a 1×2 beam splitter, two waveguide arms, a 1×2 beam combiner, and a thermo-optical phase shifting structure, which is implemented by a heater located above the waveguide arm. A spatial optical structure beam combiner is used to converge the optical signals from all output terminals after periodic modulation at different frequencies through each channel of the modulator array in free space. A photodetector is used to receive the optical signal after it has been converged by the beam combiner and output a timing electrical signal. The frequency domain of the signal carries the intensity information of different channels. The electrode array is connected one-to-one with each heater in the modulator array to apply electrical signals of different frequencies to each modulator, thereby generating modulated beams encoded by different frequencies.

[0006] The spatial optical structure beam combiner includes: An array of end-face couplers located at the output of each modulation unit is used to couple optical signals within the waveguide to free space. Lenses are used to focus the emitted light beams from each end coupler onto the photosensitive surface of the detector.

[0007] The frequency multiplexing on-chip spectrometer based on modulator arrays described above uses a microfilter array in which the transmission spectra of each channel cover a wide band, and the transmission spectra of filters in different channels have low correlation.

[0008] The frequency-multiplexing on-chip spectrometer based on modulator arrays, Heaters are installed above the two arms of the waveguide of each Mach-Zehnder modulation unit. The electrode array controls the temperature change by applying a periodic modulation signal to each heater, thereby changing the effective refractive index of the waveguide through the thermo-optic effect and realizing the periodic modulation of the output light intensity.

[0009] A spectral reconstruction method based on the spectrometer includes the following steps: S1. Construct a frequency domain channel allocation matrix G(N, M) ∈ {0, 1}^(N×M), where each column corresponds to each modulation frequency and each row corresponds to the frequency in which a channel participates; S2. Based on the preset M modulation frequencies [f1, f2, ..., f...] M Generate a family of modulation functions U = [sin(f1t) + u1, sin(f2t) + u2, ..., sin(f...]. M t) +u M ] T The modulation signal F = G*U applied to the modulator array is determined according to the allocation matrix; S3. Input the optical signal to be measured into the main waveguide, apply the modulation function through the electrode array, and collect the time-series optical intensity signal I(t) on the detector; S4. Perform FFT on the acquired time-series optical intensity signal to obtain frequency domain information, resulting in M ​​discrete peaks I(f i ) =[ I(f1), I(f2),…, I(f M )], respectively representing [f1, f2, ..., f M The information carried by each frequency; S5. Pre-calibrate the system using a calibration light source: Irradiate the system with a tunable laser of known spectrum S0(λ), apply an electrical signal to sequentially open each channel, record the individual spectral response of each channel, and obtain the transmission matrix T(N, Q) after wavelength scanning, where N is the number of channels and Q is the spectral dimension. Where T... ijThis represents the normalized transmittance of the i-th channel at the j-th wavelength. The measurement matrix H = G'*T is obtained from the generator matrix, where ' represents the matrix transpose. S6. Solve the optimization problem: min ||H·S - I||2+ λ1||D·S||2+λ2||S||1, where S is the Q-dimensional spectral vector to be reconstructed, D is the first-order difference operator, and λ1 and λ2 are regularization coefficients, to obtain the optimal spectral estimate.

[0010] Compared with existing technologies, the beneficial effects of this invention are as follows: 1. The embodiments of the present invention use a silicon photonics platform, which has many advantages such as low loss, high integration density and CMOS compatibility.

[0011] 2. This invention constructs a low-correlation transmission spectrum using a filter array, and achieves multiplexing of information from different channels in the frequency domain by applying periodic modulations at varying frequencies. By adjusting the combination of periodic functions, the transmission spectrum can be combined in the frequency domain.

[0012] 3. This invention converges multiple outgoing beams onto a single detector, improving light energy utilization and signal-to-noise ratio. By applying periodic modulation combined with a signal demodulation step, the device's anti-interference and anti-noise performance are improved. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the structure of a spectrometer. The components include: 1. Main waveguide; 2. Filter array; 3. Modulator array; 4. Spatial optical structure beam combiner; 5. Detector; and 6. Electrode array.

[0014] Figure 2 These are top views (above) and waveguide cross-sections (below) of the Mach-Zehnder modulator array. The components include: beam splitter 7, two-arm waveguides 10, beam combiner 8, and heater 9.

[0015] Figure 3 This is a schematic diagram of a space optical beam combiner. It includes: end-face coupler array 13, lens 12, and detector 5.

[0016] Figure 4 This is a schematic diagram of the workflow of the present invention, including: input light → filter splitting → applying periodic modulation function to modulator array → spatial beam combining → detector acquisition → Fourier transform → matrix calibration → regularized reconstruction.

[0017] Figure 5(a) shows the simulation results, specifically the optical intensity timing signal of a certain channel.

[0018] Figure 5(b) shows the simulation results, specifically the timing signal of the light intensity after beam combining.

[0019] Figure 5(c) shows the simulation results, specifically the Fourier transform of the light intensity time-series signal.

[0020] Figure 5(d) shows the simulation results, a comparison of the spectral reconstruction results of the two peaks. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings. Example

[0022] like Figure 1 As shown, the on-chip spectrometer in this embodiment is fabricated using an SOI platform, with a waveguide height of 220 nm and a width of 450 nm. The filter array is implemented using a silicon-based microring resonator array, with the microring perimeter increasing nonlinearly from 40 μm to 436 μm. C i =40+3× i 1.01 The total number is 128; according to the formula FSR = λ 2 / ( n g L The FSR of adjacent microrings decreased from 20nm to 2nm, and the resonance peaks between channels had a low degree of overlap, which met the low correlation requirement.

[0023] Each microring's drop end is connected to a Mach-Zehnder modulator unit 3. For example... Figure 2 As shown, each modulator consists of a beam splitter 7, two-arm waveguides 10 (width 450nm), and a beam combiner 8. A TiN heater 9 (width 2μm, length 1000μm) is deposited 1μm above the upper arm.

[0024] When a periodic current signal is applied to the heater, the local temperature of the waveguide increases, the refractive index changes, and the phase difference between the two arms changes. After beam combining, the light intensity changes, thus achieving periodic control of the light intensity.

[0025] like Figure 3 As shown, the output terminals of each modulator are connected to end-face couplers, which couple the light into free space. After being focused by the convex lens 12, the light converges to the space photodetector 11.

[0026] like Figure 4 As shown, the spectral reconstruction process is as follows: S1. Construct the frequency domain channel allocation matrix G(128, 100) ∈ {0, 1}^(128×100), where 128 corresponds to the number of micro-ring filters and 100 represents the number of modulation frequencies used.

[0027] S2. Generate a family of modulation functions U=[sin(200*2pi*t)+u1, sin(400*2pi*t)+u2, ..., sin(20000*2pi*t)+u2] based on 100 preset modulation frequencies [200Hz, 400Hz, ..., 20000Hz]. 100 ] T The modulation signal F = G*U applied to the modulator array is determined based on the allocation matrix. S3. Input the optical signal to be tested into the main waveguide (1), apply the modulation function through the electrode array (6), and collect the time-series optical intensity signal I(t) on the detector (11).

[0028] S4. Perform FFT on the acquired time-series light intensity signal to obtain frequency domain information, resulting in 100 peaks I(f i ) =[ I(f1), I(f2),…, I(f M )], respectively representing [f1, f2, ..., f 100 The information carried by each frequency.

[0029] S5. Pre-calibrate the system using a calibration light source: Illuminate the system with a tunable laser of known spectrum S0(λ), apply an electrical signal to sequentially open each channel, record the individual spectral response of each channel, and obtain the transmission matrix T(128, 1000) after wavelength scanning, where 128 is the number of filter channels and 1000 is the spectral dimension. Where T... ij Let represent the normalized transmittance of the i-th channel at the j-th wavelength. The measurement matrix H(100, 1000) = G'*T is obtained from the generator matrix, where ' represents the matrix transpose.

[0030] S6. Solve the optimization problem using the CVX toolbox: min ||H·S - I||2+ λ1||D·S||2+λ2||S||1, where S is the 1000-dimensional spectral vector to be reconstructed, D is the first-order difference operator, and λ1 and λ2 are regularization coefficients, to obtain the optimal spectral estimate.

[0031] Figure 5a An example is given that the periodic modulation function of a certain filter channel is applied. Figure 5b The middle part shows the timing signal acquired after beam combining. Figure 5c The corresponding number of discrete peaks at different frequencies can be obtained from this. Figure 5d The spectral signal of the two peaks was reconstructed, and the two peaks could be distinguished.

[0032] The embodiments described above can be further combined or substituted, and the implementation schemes are merely descriptions of embodiments of the present invention, not limitations on the concept and scope of the present invention. Various changes and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the design spirit of the present invention are all within the protection scope of the present invention. The protection scope of the present invention is given by the appended claims and any equivalents.

Claims

1. A frequency multiplexing spectral analysis chip based on a modulator array, characterized in that, Include: Input main waveguide (1), The filter array (2) is composed of structures such as micro-ring resonant cavities or waveguide filters. Each filter is coupled to the main waveguide (1) and outputs the filtered optical signal to the modulator array. The intensity modulator array (3) consists of N independent Mach-Zehnder intensity modulation units, which are used to apply periodic intensity modulation signals of different frequencies in different channels. Each modulation unit includes: a 1×2 beam splitter (7), two waveguide arms (10), a 1×2 beam combiner (8), and a thermo-optical phase shifting structure, which is implemented by a heater (9) located above the waveguide arm. The spatial optical structure beam combiner (4) is used to converge the optical signals from all the output ends of the modulator array (3) after periodic modulation at different frequencies in each channel in free space. The photodetector (11) is used to receive the optical signal after it has been converged by the beam combiner (4) and output a timing electrical signal. The frequency domain of the signal carries the intensity information of different channels. The electrode array (6) is connected one-to-one with each heater (9) in the modulator array (3) to apply electrical signals of different frequencies to each modulator, thereby generating modulated beams encoded by different frequencies.

2. The frequency multiplexing spectral analysis chip based on a modulator array according to claim 1, characterized in that, The spatial optical structure beam combiner (4) includes: The end-face coupler array (13) set at the output end of each modulation unit is used to couple the optical signal in the waveguide to free space. Lens (12) is used to focus the outgoing light beams from each end coupler onto the photosensitive surface of the detector (11).

3. The frequency-multiplexed on-chip spectrometer based on a modulator array according to claim 1, characterized in that, The transmission spectra of each channel in the microfilter array used cover a wide band, and the transmission spectra of filters in different channels have low correlation.

4. The frequency-multiplexed on-chip spectrometer based on a modulator array according to claim 1, characterized in that: A heater (9) is placed above the two-arm waveguide (10) of each Mach-Zehnder modulation unit. The electrode array (6) controls the temperature change by applying a periodic modulation signal to each heater (9), and changes the effective refractive index of the waveguide through the thermo-optic effect to achieve periodic modulation of the output light intensity.

5. A spectral reconstruction method based on the spectrometer described in claims 1-4, characterized in that, Includes the following steps: S1. Construct a frequency domain channel allocation matrix G(N, M) ∈ {0, 1}^(N×M), where each column corresponds to each modulation frequency and each row corresponds to the frequency in which a channel participates; S2. Based on the preset M modulation frequencies [f1, f2, ..., f...] M Generate a family of modulation functions U = [sin(f1t) + u1, sin(f2t) + u2, ..., sin(f...]. M t) +u M ] T The modulation signal F = G*U applied to the modulator array is determined according to the allocation matrix; S3. Input the optical signal to be measured into the main waveguide (1), apply the modulation function through the electrode array (6), and collect the time-series optical intensity signal I(t) on the detector (11); S4. Perform FFT on the acquired time-series optical intensity signal to obtain frequency domain information, resulting in M ​​discrete peaks I(f i ) =[ I(f1), I(f2),…, I(f M )], respectively representing [f1, f2, ..., f M The information carried by each frequency; S5. Pre-calibrate the system using a calibration light source: Irradiate the system with a tunable laser of known spectrum S0(λ), apply an electrical signal to sequentially open each channel, record the individual spectral response of each channel, and obtain the transmission matrix T(N, Q) after wavelength scanning, where N is the number of channels and Q is the spectral dimension. Where T... ij This represents the normalized transmittance of the i-th channel at the j-th wavelength. The measurement matrix H = G'*T is obtained from the generator matrix, where ' represents the matrix transpose. S6. Solve the optimization problem: min ||H·S - I||2+ λ1||D·S||2+λ2||S||1, where S is the Q-dimensional spectral vector to be reconstructed, D is the first-order difference operator, and λ1 and λ2 are regularization coefficients, to obtain the optimal spectral estimate.

Images