Integrated Silicon Photonics OFDR System for High-Resolution Distributed Measurements Based on Rayleigh Backscattering

Optical frequency domain reflectometry (OFDR) has high spatial resolution and measurement accuracy, driving its popularity in various fields. Integration of OFDR technology has made it accessible, cost-effective and deployable in many applications, including battery management and photonic integrated circuits. An integrated OFDR system based on Rayleigh backscattering and silicon photonics technology on an SOI platform has been developed for the first time. The system's simplified configuration was simulated, fabricated and characterized in detail, achieving an experimental spatial resolution of 8.28 μm, matching the theoretical level. This system shows high potential for sensing, monitoring and detection where precise spatial information is crucial. OFDR's accessibility and high performance in distributed measurements make it a promising technology for future advancements.

various fields, and in recent years, it has seen increasing use in photonic integrated circuits measurements.Beyond assessing the propagation loss and group refractive index of a waveguide, a fibre-based OFDR system is employed to investigate and analyze the wavelength-dependent transmittance and reflectance of elemental devices equipped with grating couplers [9].For both test element groups and large-scale integrated chips, a fibrebased OFDR system has been demonstrated the effectiveness to independently assess each component based on the spatial distribution of reflections acquired in a single measurement, and the capability to significantly streamlines the process and enhances accuracy [10].
Additionally, we have conducted a comprehensive review of available sensing technologies utilizing optical fibre sensors for battery monitoring [11], it highlighted the significance of employing a fibre-based OFDR system with high precision in both temporal and spatial resolution for the development of advanced and intelligent batteries.To fulfil this requirement, we have successfully demonstrated a fibre-based OFDR system with a world record experimental spatial resolution of 12.1 μm [12], while a remarkable temperature uncertainty of 0.13 °C [12] and a strain accuracy of 0.51 με [3] are separately achieved in our previous endeavours.These results serve as the foundational outcomes that prompted us to delve further into the exploration of cutting-edge sensing technology for batteries and guide us in enhancing the capabilities of OFDR systems by facilitating the integration with the battery management system (BMS).
However, the size and weight of fibre-based OFDR systems make them impractical for integration with BMS, impeding their development in frontier fields.As such, there is an urgent need for a reliable and high-performing integrated OFDR system to facilitate progress in these areas.
Silicon photonics, as a high-index-contrast technology, offers an excellent opportunity for the integration of OFDR.Researchers have made some attempts to achieve the integrated OFDR using silicon photonics technology.Bru [13] developed an integrated OFDR system on a silicon nitride (Si 3 N 4 ) platform.The system utilized a sweep-wavelength homodyne interferometric detection method to characterize arrayed waveguide gratings, using the reflecting light for its operation.The achieved experimental spatial resolution was 18.5 ± 1.7 μm, with a system size of approximately 5.5 × 3.5 mm 2 .The key features of  [13] AND THIS STUDY both [13] and this study are outlined and compared in Table I.Shishkin [14] demonstrated an integrated Michelson interferometer on a silicon-on-insulator (SOI) platform, theoretically proposed the potential for it to be used in an integrated OFDR system on an SOI platform.Nevertheless, it remains a prototype study that has not been implemented to date.
The current literature lacks an integrated OFDR system based on Rayleigh backscattering on an SOI platform, which is addressed in our work by successfully demonstrating such a system utilizing silicon photonics technology.
In this study, we have demonstrated an integrated OFDR system using silicon photonics technology on an SOI platform.The integration of the system on an SOI platform offers advantages such as high integration density, reduced power consumption, and compatibility with standard complementary metal-oxidesemiconductor (CMOS) technology.A coherent OFDR with the heterodyne detection method is employed due to its simple operating configuration.The simulated and fabricated integrated OFDR are in good agreement, and the system achieves a remarkable ultimate spatial resolution of 8.28 μm.
This paper is organized as follows: In Section II, we introduce the design of the integrated OFDR and present the simulation results.Section III describes the fabrication process flow of the integrated OFDR and discusses the results of the designto-fabrication verification and the distributed measurement.Finally, Section IV summarizes the contributions of this work and highlights potential directions for future research.

A. Configuration
Fig. 1(a) illustrates the functional blocks of a typical coherent OFDR system, which comprises the laser source module for providing a linearly-tuned light source, the main interferometer module for sensing the measurands by connecting to the fibre under test (FUT), and the data acquisition and processing module for converting optical signals to electrical signals.To compensate for the nonlinearity in laser frequency tuning, an auxiliary interferometer module is often included in the system.In this work, we demonstrate an integrated OFDR that replaces the fibre-based interferometer modules with on-chip photonic circuits based on silicon photonics technology.Fig. 1(b) shows the one-to-one correspondence between the fibre components and the on-chip devices, enabling the realization of a compact and highly integrated OFDR system.
The choice of the multimode interference (MMI) coupler in the design of the integrated OFDR demonstrated in this work is significant for several reasons.First, the MMI coupler offers wide optical bandwidth and low insertion loss, making it an attractive choice for power splitting.Second, its fabricationtolerant power splitting capability is advantageous for maintaining a consistent power distribution across different devices.Third, the use of the MMI coupler simplifies the design, making it easier to fabricate and operate.Specifically, in the demonstrated integrated OFDR, the MMI coupler replaces both the optical coupler and the optical circulator used in the fibre-based OFDR.The local oscillator (LO) and the optical path difference (OPD) are implemented using a single-mode waveguide and a multimode waveguide, respectively.Finally, the device under test (DUT) is designed as a defect waveguide to simulate a FUT with two defect locations.
The simulated and fabricated integrated OFDR is realized on an SOI platform consisting of a 700 μm crystalline silicon substrate, a 2 μm silica buried oxide layer, a 220 nm crystalline silicon core layer, and a 1 μm silica top cladding layer.All waveguides and devices are rib-designed with an etching depth of 120 nm, and support the transverse-electric (TE) mode with an effective group refractive index of 3.73.
To maximize the signal-to-noise ratio (SNR) of the integrated OFDR system, and to increase the visibility of the interference fringes of the main interferometer module, an optical power budget allocation is assessed due to the integrated OFDR system exhibiting higher power loss compared to the fibre-based OFDR system.Consequently, the integrated OFDR design employs MMI0 and MMI1 as 2 × 2 MMI couplers with a power splitting ratio of 70:30, while MMI2 through MMI5 require a 50:50 power splitting ratio and utilize 1 × 2 MMI couplers.The LO and OPD lengths are designed to be 202.70 μm and 1690.32 μm, respectively.Two defect points are designed along the DUT, each with an etching depth of 70 nm and a width of 315 nm.The distance between the centres of the two defect points is 20.20 μm.

B. Simulation
In the demonstrated integrated OFDR, the MMI couplers play a crucial role as on-chip devices, and their performance is of paramount importance.An MMI coupler usually comprises three sections: the input waveguides, the multimode interference region, and the output waveguides.The multimode interference region is typically a one-dimensional waveguide with an expanded section and perfectly reflecting walls.The light is coupled into the multimode interference region from the input waveguides, and multiple modes are excited in the multimode interference region.These modes propagate simultaneously and interfere with each other, and then redistribute across the output waveguides according to a specific distribution pattern.Selfimaging is achieved in an MMI coupler when the input field profile is reproduced in single or multiple images at periodic intervals along the propagation direction of the multimode waveguide.Assuming that the modal fields are zero at the sidewall, the modal field within the multimode interference region can be expressed as where x is the lateral direction; z is the direction of propagation; M is the total number of modes; m is the mode number; c m and ψ m are the field excitation coefficients and the normalized modal field distribution of the m mode, respectively; L is the length of the multimode interference region and L π is the beat length over which a π phase shift is realized between the first two guided modes.
In an MMI coupler, the propagation constant of the m mode is given by where k 0 is the free space wave number; n e is the effective refractive index of the multimode interference region; λ 0 is the free space wavelength and W e is the effective width of the multimode interference region.Therefore, the beat length can be defined as For the traditional 1 × 2 and 2 × 2 MMI couplers, power splitting ratios can be achieved according to their imaging properties.As shown in Fig. 2(a), the input waveguide is at the centre of MMI width in a 1 × 2 MMI coupler, the self-imaging point is given at L self −imaging = 3L π /4 and the 2-fold imaging point is at half of L self −imaging .Different from the 1 × 2 MMI coupler, the input waveguides are off centre and often positioned at 1/6 of MMI width in a 2 × 2 MMI coupler, as shown in Fig. 2(b), the mirror-imaging point is given at L mirror−imaging = 3L π , the self-imaging point requires double L mirror−imaging , the 2-fold imaging point can be found at L mirror−imaging /2 and 3L mirror−imaging /2.Therefore, 50:50 power splitting ratio can be achieved by positioning the two output waveguides at the 2-fold imaging point while 70:30 power splitting ratio can be obtained by changing the device dimension and geometry.
The positions and widths of the input and output waveguides, as well as the length of the multimode interference region, are accurately optimized using Lumerical FDTD, a widely used tool for designing and simulating photonic devices and systems.Tapers are added to the input and output waveguides to improve coupling efficiency.
The simulation results of the 70:30 2 × 2 MMI coupler, which is used as a splitter, are shown in Fig. 3.The taper width is 1.5 μm, and the taper length is 20 μm, with a separation of 1.7 μm.The required power splitting ratio of 70:30 can be achieved at a multimode interference region length of 64.5 μm, with a fixed width of 6 μm.When the input signal is linearly tuned from 1500 nm to 1600 nm with an amplitude of 1, the transmissions of output1 and output2 are shown in Fig. 3(b).The transmission ratio at the central wavelength of 1550 nm is 0.69:0.31,which meets the requirement.Fig. 3(c) shows the electric field distribution when the MMI coupler is operating at 1550 nm.
The 50:50 1 × 2 MMI coupler can be achieved by positioning the two output waveguides at the 2-fold imaging point of a traditional 1 × 2 MMI coupler.To achieve this, tapers with the same size as the ones added in the 70:30 2 × 2 MMI coupler are added to the input and output waveguides, and the separation between tapers is set to 1.64 μm.The width of the multimode interference region is 6 μm, and the 2-fold imaging point is at a distance of 32.7 μm from the input waveguide taper.Simulation results in Fig. 4(b) show the electric field distribution of MMI2∼MMI5 at 1550 nm.When the 50:50 1 × 2 MMI coupler works as a splitter, as depicted in Fig. 4(c), the transmissions of output1 and output2 are nearly equal, and the transmission ratio at the central wavelength of 1550 nm is 0.49:0.50,which corresponds to a 3 dB power splitting ratio.As a reciprocal device, the transmission of each port remains the same when the 1 × 2 MMI coupler functions as either a circulator or a combiner.Fig. 4(d) presents the simulation result of the combiner situation, where the original output1 and output2 in Fig. 4(a) become the new input1 and input2, respectively, and the original input is now the new output, all with an amplitude of 1.

III. EXPERIMENT
The integrated OFDR presented in this study is fabricated on a single-side polished SOI wafer using a specific process flow, which is depicted in Fig. 5.
The fabrication process starts with the SOI substrate described in Section II-A.A resist patterning is prepared to create the grating couplers and defect points along the DUT, followed by a shallow silicon etch of 70 nm to form these structures.After stripping the grating resist, a silicon dioxide hard mask layer of 180 nm is deposited on the silicon top layer.Then, a resist patterning is employed for the hard mask etching of rib waveguides and components.An intermediate silicon etch of 120 nm is performed after the resist is stripped from the rib devices.Finally, a 1 μm thick silica top cladding is deposited to complete the fabrication process.
The experimental measurements are carried out using a threedimensional adjustable probe station equipped with a microscope and three 3-axis micropositioners to attach optical fibres for coupling the laser into and out of the chip.A polarization controller (PC) is used to ensure the excitation of the fundamental TE mode.A linearly tuned laser from a Santec TSL-550 laser source is employed as the input signal for all measurements.The output power of the integrated photonic circuit is recorded by a Santec MPM-212 power meter, while the beat signals generated from the integrated OFDR are detected by the Thorlabs PDB450C photodetectors, followed by the DAQ (Tektronix MSO64).

A. Power Splitting Ratio Measurement
Before characterizing the integrated OFDR, it is necessary to verify the power splitting ratios of the designed 1 × 2 and 2 × 2 MMI couplers.This is done by coupling an input laser into several sets of cascaded MMI couplers, of which the same output port is intended to be measured.The power of the output laser through each set of MMI couplers can be expressed as where P in is the power of the input laser, L n is the insertion loss of the n set of MMI couplers where the number of the cascaded MMI couplers is n, T is the transmission of the measured output port, C is the coupling efficiency of the grating coupler.For n = 1, 2, 3, (4) can be represented as following Therefore, the coupling efficiency of the grating coupler and the transmission of the measured output port can be Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.
For the power splitting ratio verifying measurements, the experimental conditions are designed to be consistent with the simulation situations.The tunable laser source (TLS) used has a wavelength range of 100 nm (1500∼1600 nm) with a step of 0.01 nm and an output power of 10 dBm.Three sets of cascaded MMI couplers are tested for each output port of the designed 1 × 2 and 2 × 2 MMI couplers.The results of the measurements are presented in Fig. 6.
Fig. 6(a) displays the power of the laser as it passes through the PC for the entire wavelength range.It can be observed that less than 0.8 dB of power is lost due to the adjustment of the PC.In Fig. 6(b), the calculated coupling efficiency spectrum of the grating coupler is presented.The grating coupler exhibits a 1 dB bandwidth of 48 nm, with a coupling efficiency ranging from −7.45 to −6.45 dB.The central wavelength range of the grating coupler is between 1527 nm and 1575 nm.
The 70:30 2 × 2 MMI coupler, as depicted in Fig. 3(a), undergoes measurement to obtain the transmission characteristics of output1 and output2, as presented in Fig. 6(c).The transmission of output1 ranges from 0.62 to 0.72 over the central wavelength range, while the corresponding transmission of output2 varies from 0.25 to 0.34.At the wavelength of 1550 nm, the transmission ratio between output1 and output2 is measured as 0.66:0.31.
In Fig. 6(d), the measured transmissions of output1 and out-put2 of the 50:50 1 × 2 MMI coupler described in Fig. 4(a) are presented.The two transmission curves are symmetrical concerning T = 0.5 as the axis of symmetry.Within the central wavelength range of 1527∼1575 nm, the transmissions of output1 and output2 are 0.50∼0.52 and 0.50∼0.48,respectively.At 1550 nm, the transmission ratio is measured to be 0.51:0.49.
The fabricated MMI couplers are observed under a microscope, and their images are shown in Fig. 6(e) and (f) for the 70:30 2 × 2 MMI coupler and the 50:50 1 × 2 MMI coupler, respectively.The power splitting ratio measurement results are consistent with the simulation results and within the fabrication tolerance, indicating that the integrated OFDR is expected to perform as designed.

B. Group Refractive Index Measurement
The group refractive index is a crucial factor in an OFDR system as it has a direct relationship with the beat frequency, which Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.can significantly influence spatial resolution.Therefore, it is essential to measure the group refractive index accurately before demonstrating the integrated OFDR.One effective approach for measuring the group refractive index is to use the free spectral range (FSR) of an unbalanced Mach-Zehnder interferometer (MZI), where the lengths of the two arms are different.The FSR of an unbalanced MZI can be defined as where λ is the central wavelength, ΔL is the length difference between the two arms of the MZI and n g is the group refractive index which can be furtherly calculated by To ensure accurate measurements in the demonstrated integrated OFDR, it is necessary to measure the group refractive index separately for the LO and the OPD, as they have different values.For this purpose, two unbalanced MZIs are designed, each with a specific length difference, 140 μm for the LO and 160 μm for the OPD.Each MZI has two equivalent output ports, allowing for a more precise measurement of the group refractive index.
In the experiment, the TLS operates in the wavelength range of 1500 nm to 1600 nm with a step of 0.01 nm.The output interference signals from the two output ports of the MZIs are recorded by the PD, as shown in Fig. 7(a) for the LO and Fig. 7(b) for the OPD.By analyzing the FSR of the MZIs, the group refractive index for each component is accurately determined.
The FSR, which is the difference in wavelength between two adjacent resonance modes of the MZI, is measured for both  11), the corresponding group refractive index is obtained, and the results are depicted in Fig. 7(e) and (f).
Within the central wavelength range of the grating coupler, the group refractive index for the LO is measured to range from 3.6393 to 3.8089, with a mean value of 3.7167.In the case of the LO-based MZI with a length difference of 140 μm, the beat frequency ranges from a minimum of 21.23 Hz to a maximum of 22.22 Hz, with a mean beat frequency of 21.68 Hz.Utilizing the mean group refractive index introduces a beat frequency error of less than 2.4%, deemed acceptable for the OFDR characterization experiment.
Similarly, the measured group refractive index for the OPD falls within the range of 3.6711 to 3.7884, with a mean value of 3.7230.For the OPD-based MZI with a length difference of 160 μm, the minimum, maximum, and mean beat frequencies are 24.47Hz, 25.26 Hz and 24.82 Hz, respectively.The associated beat frequency error resulting from using the mean group refractive index is less than 1.7%, which is also within acceptable limits.
Therefore, the mean group refractive index values are employed in subsequent measurements.

C. Integrated OFDR Characterization
The experimental setup, shown in Fig. 8(a), consists of a linear frequency tuning laser coupled into the integrated OFDR via a PC.The laser is split at MMI0, with 70% of the light entering the main interferometer and the remaining 30% entering the auxiliary interferometer.MMI1 then splits the light into the LO and DUT channels (30% and 70%, respectively).The Rayleigh backscattered light from the DUT passes through MMI5 and interferes with the light from the LO at MMI2, and the resulting interference signal is detected by PD1.The auxiliary interferometer generates a beat signal through an unbalanced MZI formed Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.To ensure optimal performance in the distributed measurement, the wavelength tuning range of the TLS is restricted to the central wavelength range of the grating coupler.The TLS is set to operate from 1529 nm to 1572 nm, with a tuning range of 43 nm, in the characterization experiment.Compared to the fibre-based OFDR, the integrated OFDR suffers from higher power loss and lower SNR.To mitigate these issues, several methods are employed in the experiment, including: 1) Setting the output power of the TLS to the maximum working power of 10 dBm and using no optical attenuator, relying solely on the power splitting ratios of the optimized MMI couplers for power budget allocation.2) Tuning the wavelength of the TLS at the highest available speed of 100 nm/s, which increases the beat frequency of the auxiliary interferometer and compensates for the tuning nonlinearity of the TLS. 3) Adjusting the PC to ensure that the TE mode is excited and transmitted.4) Adjusting the angle of the fibre positioner to match the fibre coupling angle of the grating coupler, the optimized angle is determined to be 11°from the vertical.5) Selecting a PD with a gain of 10 5 that is appropriate for the experiment, as well as a bandwidth of 4 MHz that satisfies the Nyquist sampling theorem.6) Setting the DAQ to operate at a suitable sample rate of 31.25 MHz and using an analogue-to-digital converting resolution of 12 bits.7) Implementing the equal frequency resampling method [12] to effectively compensate for the nonlinear frequency tuning noise, which can improve the accuracy of the measurement.8) To facilitate location and verification in the distributed measurement, the DUT is designed with several strong scattering positions.The detailed design of the DUT is shown in Fig. 9(a), which is vertically stretched for clarity.The start position of the DUT is at z 0 , which is the side edge with two ports of the MMI5.The other side edge of the MMI5, at position z 1 , forms a strong scattering surface at the near end of the DUT.Positions z 2 and z 3 are the central positions of the two designed defect points, respectively, and they cause stronger Rayleigh backscattering compared to other non-defect parts along the DUT.Position z 4 is the start position of the grating coupler, and it also forms a strong scattering surface at the far end of the DUT.The end position of the DUT is at z 5 , where the backscattering is weak because the light has already been coupled out from the on-chip system at the grating coupler.A characterization experiment is conducted under the conditions discussed above.Fig. 9(b) presents the spectrum distribution along the whole DUT introduced in Fig. 9(a) in the spatial domain, the performance of the demonstrated integrated OFDR is described as below: 1) The Spatial Resolution: The theoretical spatial resolution (Δz) of OFDR is determined by the frequency tuning range (ΔF ) and the group refractive index (n g ), it is defined as where c is the speed of light in space.Therefore, in the characterization experiment, the theoretical spatial resolution of the integrated OFDR is 7.51 μm.
The experimental spatial resolution of OFDR is the full width at the half maximum of the peaks of the spectrum distribution in the spatial domain.As shown in the inset of Fig. 9(b), an experimental spatial resolution of 8.28 μm is achieved at position z 3 , it reaches the theoretical spatial resolution level.
2) The Design Verification: In the demonstrated integrated OFDR, the auxiliary interferometer is an unbalanced MZI with a designed fixed length difference of 1517.60 μm, its interference fringes have a fixed beat frequency based on the heterodyne detection, the relationship between the beat frequency (f B ) and the length difference (ΔL) can be given by Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.

TABLE II THE DESIGNED AND EXPERIMENTAL LOCATIONS ALONG THE DUT
where γ is the frequency tuning speed, n g is the group refractive index and c is the speed of light in space.
During the characterization experiment, the beat frequency of the auxiliary interferometer is calculated by performing a fast Fourier transform (FFT) on the signal received by PD2.The calculated beat frequency is 235.94Hz, corresponding to a length difference of 1520.99 μm.Although there is a difference between the experiment and the design due to the random frequency distribution of the interference signal caused by the nonlinear frequency tuning in the actual TLS, the difference is only 3.39 μm, which is much smaller than the experimental spatial resolution.This result indicates that the integrated OFDR is fabricated as designed and works as expected.
3) The Performance: Just like the auxiliary interferometer, the beat frequency of a given location on the DUT is directly proportional to its position, as described by (13).Thus, the various beat frequencies extracted from the frequency spectrum of the PD1 signal correspond to distinct locations along the DUT.Notably, because the Rayleigh backscattered light travels twice in the main interferometer, the location of a given position is half the length difference between it and the LO.
The design geometry of the main interferometer of the integrated OFDR determines the length difference between the LO and position z 0 to be 223.88μm.Thus, position z 0 is designed to be located at 111.94 μm, which is half of the length difference.Using this principle, the designed locations of positions z 0 to z 5 along the DUT are determined.Table II presents the designed locations along with their corresponding experimental values and the error, which is calculated as the experimental value minus the designed value.
Prior to the characterization experiment, the distance between positions z 2 and z 3 is measured using a microscope, resulting in a distance of 19.88 μm, as depicted in Fig. 8(a).The designed value for this distance is 20.20 μm as marked in Fig. 1(b), while the experimental value is 22.64 μm.The discrepancy between the measured and designed values can be attributed to fabrication and human errors.Meanwhile, the difference between the measured and experimental values is due to inaccurate estimation of beat frequencies and limitations in the experimental spatial resolution.The errors obtained during the experiment range from −1.25 μm to 2.01 μm, all of which fall within the experimental spatial resolution.This indicates that the demonstrated integrated OFDR is capable of performing accurately and reliably under the given conditions.

IV. CONCLUSION
In conclusion, the paper presents a significant advancement in the field of OFDR with the first real integrated OFDR based on Rayleigh backscattering achieved on the SOI platform using silicon photonics technology.The use of MMI couplers with different power splitting ratios greatly simplifies the design of the integrated system, making it easy to fabricate and reliable to operate.The paper provides a clear fabrication process flow and verifies that the fabricated integrated OFDR is consistent with the original design.The distributed measurement of the DUT demonstrates that the integrated OFDR performs perfectly, achieving an experimental spatial resolution of 8.28 μm, which is close to its theoretical level.The realization of this integrated OFDR on SOI platform opens up new possibilities for OFDR in the frontier area, and has the potential to significantly impact many fields, including civil engineering, geo-hydrology, composite materials, chemical sensing, biomedical imaging, electromagnetic field, aerospace and defense, smart infrastructure, robotics and automation.

Fig. 4 .
Fig. 4. Simulation results of 50:50 1 × 2 MMI coupler.(a) The top-down view; (b) the electric field distribution at 1550 nm; (c) the transmissions under the situations of splitter and circulator during 1500∼1600 nm; (d) the transmission under the situation of combiner during 1500∼1600 nm.

Fig. 7 .
Fig. 7. Results of group refractive index measurement.(a) The output power of the LO-based MZI; (b) the output power of the OPD-based MZI; (c) the FSR of the LO-based MZI; (d) the FSR of the OPD-based MZI; (e) the group refractive index of the LO; (f) the group refractive index of the OPD.

Fig. 8 .
Fig. 8. Experimental setup of integrated OFDR characterization.(a) The schematic diagram; (b) the deployment.The integrated OFDR chip is placed on the self-designed three-dimensional adjustable probe station.
by MMI3 and MMI4, which is detected by PD2.The signals from both PDs are recorded by the DAQ.Fig. 8(b) illustrates the layout of the experimental setup.

Fig. 9 .
Fig. 9. Performance of the demonstrated integrated OFDR.(a) The geometry of the whole DUT; (b) the spectrum distribution along the whole DUT in the spatial domain.

TABLE I THE
KEY FEATURES OF THE WORKS PRESENTED IN REF.