High-resolution coherent spectrum reconstruction method based on phase noise power spectrum analysis

Abstract

The invention relates to the technical field of spectral analysis, in particular to a phase noise power spectrum analysis-based high-resolution coherent spectrum reconstruction method, which comprises a coherent spectrometer system, the coherent spectrometer system comprises a light path, a photoelectric conversion part and a data processing part, the light path part comprises linear chirp local oscillator light and a to-be-measured optical signal, and the photoelectric conversion part comprises a phase noise power spectrum; the linear chirp local oscillator light and the optical signal to be measured are respectively connected with an optical isolator, and the two optical isolators are communicated with the optical fiber coupler. The photoelectric conversion part comprises a balanced photoelectric detector, the upstream of the balanced photoelectric detector is communicated with the photoelectric coupler, and the downstream of the balanced photoelectric detector is communicated with the low-pass filter; and the data processing part comprises data acquisition, time domain Fourier transform, time domain spectrum superposition and spectrum reconstruction of a signal to be measured. Compared with a traditional coherent spectrometer system, the structure of the coherent spectrometer system is more simplified, and the limitation that the spectral resolution of a traditional coherent spectrometer depends on the frequency and bandwidth of a narrow-band filter is broken through.

Description

Technical Field

[0001] The invention relates to the technical field of spectrum analysis, in particular to a high-resolution coherent spectrum reconstruction method based on phase noise power spectrum analysis. Background Art

[0002] Spectral analysis technology is widely used in the fields of fiber optic sensing, optical metrology and optical communication. Its spectral resolution, dynamic range and other indicators directly determine the sensing limit, channel spacing, laser line width, side mode suppression ratio and other technical indicators of fiber optic sensing. With the rapid development of dense wavelength division multiplexing, high-speed optical communication and optical frequency comb, the traditional spectrometer based on the grating diffraction principle (resolution is usually around 1.25GHz) can no longer meet the requirements of high resolution, large dynamic range and fast scanning, and has become one of the important factors restricting the development of optical sensing, optical metrology and optical communication technology. There are two main technical directions of modern high-resolution spectrometers: Brillouin spectrometer (BOSA) based on stimulated Brillouin scattering and coherent spectrometer (COSA) based on heterodyne coherence. Among them, BOSA is more sensitive to environmental vibration and has poor robustness. COSA is based on the beat frequency processing of the optical signal under test (SUT) and the linearly swept local oscillator optical signal (LO), and reconstructs the SUT spectrum according to the mapping relationship between LO frequency and time. It has good robustness and high resolution.

[0003] COSA needs to perform power statistics within a limited bandwidth on the filtered SUT and LO beat frequency signals to obtain the SUT relative optical power of the corresponding frequency. This process is greatly affected by the phase noise of the SUT and LO. In 2002, Agilent Laboratories in the United States systematically studied the mutual constraints between the scanning rate, resolution and power uncertainty of COSA. The study showed that the resolution and power uncertainty of COSA are directly related to the frequency and bandwidth of the bandpass filter, and the resolution is inversely proportional to the power uncertainty, while the power uncertainty is proportional to the scanning rate. The mutual constraints of the three have limited the development of COSA.

[0004] Take monochrome SUT as an example. Figure 1 As shown in Figure 1, the spectrum reconstruction principle of COSA is based on the fact that the time domain and frequency domain signals of the SUT and LO beat frequency have the same linear frequency modulation characteristics. By bandpass filtering the beat frequency signal, we can obtain Figure 2 A pair of symmetrical spectrums matched with bandpass filters and time domain spectrum with the same characteristics are shown. The wavelength of the SUT can be determined based on the LO wavelength corresponding to the symmetric center after filtering the time domain beat signal, and the relative optical power of the SUT can be determined by integrating the amplitude of the limited bandwidth time domain signal of the bandpass filter.

[0005] For non-monochromatic SUTs, the spectral resolution of COSA based on bandpass filtering depends on the frequency and bandwidth of the bandpass filter, which is specifically expressed as 2f0 +Δf, where f 0 is the center frequency of the filter, and Δf is the filter bandwidth. To obtain higher spectral resolution, the filter frequency and bandwidth need to be reduced as much as possible, but the low-frequency part of the beat signal is affected by the phase noise of the SUT and LO, the signal amplitude changes significantly with the noise, and the power uncertainty increases. The relative optical power of the SUT cannot be accurately measured, resulting in a reduction in the dynamic range of the COSA, and the narrowband filter frequency cannot be simply reduced to obtain high resolution.

[0006] In 2016, Feng Kunpeng and others from Harbin Institute of Technology achieved a 6MHz resolution COSA based on a narrowband filter. In 2018, Wu Weidong from Harbin Institute of Technology further improved the resolution of COSA to 5MHz using the Wiener convolution algorithm. The above studies show that coherent spectroscopy technology has good development and application prospects in the field of ultra-high resolution spectroscopy, but there are still problems such as low scanning rate, power uncertainty and mutual constraints on resolution. Summary of the invention

[0007] The purpose of the present application is to provide a high-resolution coherent spectrum reconstruction method based on phase noise power spectrum analysis, aiming to solve the above-mentioned problems in the prior art.

[0008] The present application provides a high-resolution coherent spectrum reconstruction method based on phase noise power spectrum analysis, including a coherent spectrometer system, wherein the coherent spectrometer system includes an optical path, photoelectric conversion and data processing, wherein the optical path part includes a linearly chirped local oscillator light and a light signal to be measured, wherein the linearly chirped local oscillator light and the light signal to be measured are each connected to an optical isolator, and the two optical isolators are connected to an optical fiber coupler; the photoelectric conversion part includes a balanced photodetector, wherein the upstream of the balanced photodetector is connected to the photocoupler, and the downstream of the balanced photodetector is connected to the low-pass filter; the data processing part includes data acquisition, time-domain Fourier transform, time-domain spectrum superposition and spectrum reconstruction of the signal to be measured, wherein the original data obtained after multiple frequency sweeps of the local oscillator light are subjected to time-domain Fourier transform, and the spectrum of the signal to be measured is reconstructed by time-domain spectrum superposition.

[0009] Furthermore, the coherent spectrometer system has the following working process: the local oscillator light and the optical signal to be measured pass through an optical isolator to prevent the reflected light from affecting the light source, and generate interference behavior through an optical fiber coupler. Assume that the initial phase of the local oscillator light is The angular frequency is ω LO =ω 0 +γt, γ is the chirp rate of the local oscillator light, ω 0 is the initial angular frequency of the local oscillation light, and the amplitude is E LO ,but Similarly, let the initial phase of the optical signal to be measured be The angular frequency is ω SUT, the amplitude is E SUT ,but The local oscillator light and the optical signal to be measured undergo heterodyne interference in the fiber coupler to form a beat frequency signal. The two output ports of the fiber coupler are connected to a balanced photodetector. The AC part of the current signal measured by the balanced photodetector can be expressed as in, is the phase difference between the local oscillator light and the optical signal to be measured.

[0010] Furthermore, a low-pass filter is used to perform a time domain Fourier transform on the beat signal according to the different degrees to which each frequency difference Δω of the beat signal is affected by phase noise. The phase noise frequency distribution of the low-frequency difference part is wide and the amplitude is large, while the phase noise frequency amplitude of the high-frequency difference part is relatively small.

[0011] Furthermore, the spectrum of each frequency difference is integrated to obtain the relative power of each frequency difference of the optical signal to be measured, thereby realizing the reconstruction of the spectrum of the optical signal to be measured.

[0012] Furthermore, a distributed feedback laser with a direct modulation linewidth of 100 MHz is used as the local oscillator light.

[0013] The beneficial effects of the present invention are as follows: the coherent spectrometer system of the present invention has a simpler structure than the traditional one, and no narrow-band filter is needed to pre-process the beat frequency signal. The bandpass filtering scheme adopted by the traditional coherent spectrometer is replaced by a phase noise analysis scheme; the spectral resolution is further improved to the width limit of the local oscillator light adopted, breaking through the limitation that the spectral resolution of the traditional coherent spectrometer depends on the frequency and bandwidth of the narrow-band filter. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 Linear frequency sweep characteristics of the coherent beat signal between SUT and LO: (a) time domain; (b) frequency domain.

[0015] Figure 2 The beat frequency signals of SUT and LO processed by Gaussian bandpass filter: (a) spectrum; (b) time domain spectrum.

[0016] Figure 3 This is a block diagram of the coherent spectrometer system based on phase noise recording.

[0017] Figure 4 is the time domain signal of the beat frequency signal with different initial phase differences.

[0018] Figure 5 Detection results of the reflection spectrum of a fiber Bragg grating with a bandwidth of 0.08 nm: (a) coherent spectrometer based on phase noise recording; (b) traditional grating spectrometer.

[0019] Figure 6The spectrum reconstruction results of ultra-narrow linewidth lasers are obtained by a coherent spectrometer system based on phase noise recording. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] like Figure 3-6 The high-resolution coherent spectrum reconstruction method based on phase noise power spectrum analysis shown includes a coherent spectrometer system, such as Figure 3 As shown, the coherent spectrometer system includes an optical path, photoelectric conversion and data processing. The optical path part includes a linearly chirped local oscillator light and a light signal to be measured. The linearly chirped local oscillator light and the light signal to be measured are each connected to an optical isolator, and the two optical isolators are connected to an optical fiber coupler; the photoelectric conversion part includes a balanced photodetector, the upstream of the balanced photodetector is connected to the photocoupler, and the downstream of the balanced photodetector is connected to the low-pass filter; the data processing part includes data acquisition, time-domain Fourier transform, time-domain spectrum superposition and spectrum reconstruction of the signal to be measured, the original data obtained after multiple frequency sweeps of the local oscillator light is subjected to time-domain Fourier transform, and the spectrum of the signal to be measured is reconstructed by time-domain spectrum superposition.

[0022] The working process of the coherent spectrometer system is as follows: the local oscillator light and the optical signal to be measured pass through an optical isolator to prevent the influence of reflected light on the light source, and generate interference behavior through an optical fiber coupler. Assume that the initial phase of the local oscillator light is The angular frequency is ω LO =ω 0 +γt, γ is the chirp rate of the local oscillator light, ω 0 is the initial angular frequency of the local oscillation light, and the amplitude is E LO ,but Similarly, let the initial phase of the optical signal to be measured be The angular frequency is ω SUT , the amplitude is E SUT ,but The local oscillator light and the optical signal to be measured undergo heterodyne interference in the fiber coupler to form a beat frequency signal. The two output ports of the fiber coupler are connected to a balanced photodetector. The AC part of the current signal measured by the balanced photodetector can be expressed as in, is the phase difference between the local oscillator light and the optical signal to be measured.

[0023] Since the initial phases of the local oscillator light and the optical signal to be measured are random, the amplitude uncertainty of the beat frequency signal measured by the balanced photodetector is caused. Figure 4 As shown, the low frequency difference part of the beat signal (Δω=ω SUT -ω LO ≈0) The amplitude uncertainty is significantly affected by random phase noise, while the high-frequency difference has little effect. Based on this phenomenon, the traditional coherent spectrometer solution adopts Figure 4 The bandpass filtering scheme of the proposed method filters out the low-frequency part with high amplitude uncertainty. According to the symmetrical frequency characteristics of the beat frequency signal, the local oscillator frequency corresponding to the center of the double peak of the envelope after filtering is the reconstruction frequency of the optical signal to be measured, and the energy integral within the filter bandwidth is the relative optical power of the optical signal to be measured. The frequency resolution of this scheme depends on the filter frequency and bandwidth, but the filter frequency and bandwidth are inversely proportional to the power uncertainty, which limits the performance of the coherent spectrometer.

[0024] Low-pass filtering is used to perform time-domain Fourier transform on the beat signal according to the different degrees to which each frequency difference Δω of the beat signal is affected by phase noise. The phase noise frequency distribution of the low-frequency difference part is wide and the amplitude is large, while the phase noise frequency amplitude of the high-frequency difference part is relatively small.

[0025] The spectrum of each frequency difference is integrated to obtain the relative power of each frequency difference (wavelength) of the optical signal to be measured, thereby realizing the reconstruction of the spectrum of the optical signal to be measured.

[0026] A distributed feedback laser with a direct modulation linewidth of 100MHz is used as the local oscillator to detect the reflected light of the fiber Bragg grating. The results are as follows: Figure 5 As shown, compared with traditional spectrometers, more spectral details can be obtained and the spectral resolution is further improved.

[0027] In order to further confirm the spectral resolution limit of the coherent spectrometer based on phase noise recording, the coherent spectrometer system was used to reconstruct the spectrum of a 100kHz ultra-narrow linewidth laser as the optical signal to be measured. The results are shown in Figure 2. Figure 6 The measured reconstructed optical signal line width is about 100MHz, which is consistent with the local oscillator line width.

[0028] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A high-resolution coherent spectrum reconstruction method based on phase noise power spectrum analysis, characterized in that: The invention comprises a coherent spectrometer system, which comprises an optical path, photoelectric conversion and data processing. The optical path part comprises a linearly chirped local oscillator light and a light signal to be measured, the linearly chirped local oscillator light and the light signal to be measured are respectively connected to an optical isolator, and the two optical isolators are connected to an optical fiber coupler; the photoelectric conversion part comprises a balanced photodetector, the upstream of the balanced photodetector is connected to the photoelectric coupler, and the downstream of the balanced photodetector is connected to a low-pass filter; the data processing part comprises data acquisition, time-domain Fourier transform, time-domain spectrum superposition and spectrum reconstruction of the signal to be measured, the original data obtained after multiple frequency sweeps of the local oscillator light are subjected to time-domain Fourier transform, and the spectrum of the signal to be measured is reconstructed by time-domain spectrum superposition.

2. The high-resolution coherent spectrum reconstruction method based on phase noise power spectrum analysis according to claim 1 is characterized in that: The working process of the coherent spectrometer system is as follows: the local oscillator light and the optical signal to be measured pass through an optical isolator to prevent the influence of reflected light on the light source, and generate interference behavior through an optical fiber coupler. Assume that the initial phase of the local oscillator light is The angular frequency is ω LO =ω0+γt, γ is the chirp rate of the local oscillator light, ω0 is the initial angular frequency of the local oscillator light, and the amplitude is E LO ,but Similarly, let the initial phase of the optical signal to be measured be The angular frequency is ω SUT , the amplitude is E SUT ,but The local oscillator light and the optical signal to be measured undergo heterodyne interference in the fiber coupler to form a beat frequency signal. The two output ports of the fiber coupler are connected to a balanced photodetector. The AC part of the current signal measured by the balanced photodetector can be expressed as in, is the phase difference between the local oscillator light and the optical signal to be measured.

3. The high-resolution coherent spectrum reconstruction method based on phase noise power spectrum analysis according to claim 1 is characterized in that: Low-pass filtering is used to perform time-domain Fourier transform on the beat signal according to the different degrees to which each frequency difference Δω of the beat signal is affected by phase noise. The phase noise frequency distribution of the low-frequency difference part is wide and the amplitude is large, while the phase noise frequency amplitude of the high-frequency difference part is relatively small.

4. The high-resolution coherent spectrum reconstruction method based on phase noise power spectrum analysis according to claim 3 is characterized in that: The spectrum of each frequency difference is integrated to obtain the relative power of each frequency difference of the optical signal to be measured, thereby realizing the reconstruction of the spectrum of the optical signal to be measured.

5. The high-resolution coherent spectrum reconstruction method based on phase noise power spectrum analysis according to claim 1 is characterized in that: A distributed feedback laser with a direct modulation linewidth of 100 MHz is used as the local oscillator.

Images