Methods and apparatus for ofdr interrogator monitoring and optimization

REFERENCE TO RELATED APPLICATIONS This application claims priority to and the benefit of U.S. Provisional Patent Application No. 62/355,957, entitled METHODS AND APPARATUS FOR OFDR INTERROGATOR MONITORING AND OPTIMAZATION, filed June 29, 2016, which is incorporated herein by reference in its entirety.

The technology described in this application relates to Optical Frequency Domain Reflectometry (OFDR) measurements used for optical fiber geometry sensing, and to data processing techniques that improve the accuracy and reliability of those OFDR measurements.

Optical strain sensing is a useful technique for measuring the physical deformation of a waveguide, e.g., caused by tension, compression, or temperature changes in the optical fiber. A multi-core optical fiber consists of several independent waveguides embedded within a single fiber. By interpreting the optical response of the cores, typically using swept wavelength interferometry in the form of optical frequency domain reflectometry (OFDR) measurements, a continuous measurement of the strain along the length of the cores can be derived. With knowledge of the relative positions of the cores along the length of the fiber, these independent strain signals may be combined to obtain a measurement of the strain profile imposed on the multi-core optical fiber. Fiber strain profile refers to the measurement of bend strain, twist strain, and/or axial strain imposed along the length of the fiber with high (e.g., less than 50 micrometer) sample resolution.

Previous patents have described OFDR-based shape sensing using multi-core optical fibers (see, for example, U.S. Patent Nos. 7,781,724 and 8,773,650, which are incorporated by reference). Some applications of OFDR-based shape sensing fibers require a high degree of confidence in the accuracy and reliability of the shape sensing output. A non-limiting exemplary application is a robotic arm used in surgery or other environments.

There are three basic elements in an OFDR measurement system: light, a medium (e.g., a fiber waveguide) that the light traverses, and a receiver that detects the light and converts the light into an electrical signal. Each of these basic elements contributes to the accuracy of the measurements that are performed. An exemplary fiber optic shape sensing system is shown in FIG. 1 and includes a tunable laser 1 that is swept to provide light at different frequencies or wavelengths to an optical network 2 that is coupled to a fiber optic sensor/device under test (DUT) 3 (also referred to herein as "sensor 3" or "DUT 3"). Each scan by the tunable laser through a tuning range of wavelengths or frequencies generates a set of OFDR measurement data. The optical network 2 includes detection, acquisition, and control electronics 4 that include a detector that converts the light information into an electrical signal, an analog-to-digital converter that converts the analog electrical signal into a digital electrical signal, and a field programmable gate array (EFGA) that processes the acquired data and controls the acquisition. The detection, acquisition, and control electronics 4 provides an output to a processor 5 for further processing, such as calculating the fiber geometry, and ultimately outputting information from the fiber sensor/DUT 3. The processor 5 may additionally or alternatively include an FPGA or a GPU.

In the exemplary fiber optic shape sensing system shown in FIG. 1, shape measurements depend on several factors, including the ability to accurately detect light reflected from a medium, which is a multi-core fiber optic sensor. OFDR-based fiber optic shape sensing detects point-to-point length changes in each core of the fiber sensor, and the accuracy and noise of the OFDR measurement system depend on the repeated transmission of light into the fiber optic sensor and the repeatable detection of the reflected light for each measurement. Detection of reflected light at different frequencies in the optical frequency domain and time domain (in OFDR the laser is swept or scanned through a measurement range of wavelengths or frequencies) is important to achieve accurate fiber optic shape measurements. OFDR scan-to-scan differences or inter-scan differences that are not the result of physical changes in the sensing fiber can introduce errors into the OFDR measurement data if they are not reduced or corrected. Exemplary error sources include variations in the laser tuning speed, variations in the optical output power, changes in the optical path length of the interferometer, delay shifts between cores, and delays of electrical signals in the OFDR channels.

Assuming that the relative delay of the detection circuitry between the cores is constant, the phase change between the cores can be interpreted as due to physical changes made to the sensing fiber. However, the inventors have discovered that these delay relationships change over time, change with temperature, and/or are affected by failures in the interrogation device. Delay shifts introduced in the data acquisition and processing introduce errors into the OFDR measurements that are erroneously interpreted as physical changes to the sensing fiber geometry. In addition, changes in the optical power level between laser scans can also result in OFDR measurement errors. Such changes in the optical power level can result in inconsistent signal-to-noise ratios (SNRs) and/or varying SNRs between the upsweep and downsweep of the laser over the optical frequency range over which the OFDR measurements are made.

An exemplary embodiment includes an optical measurement system for measuring a fiber optic sensor including a plurality of optical cores. A tunable laser sweeps laser light over a first measurement range of wavelengths and an optical amplifier amplifies the swept laser light. An optical network provides the amplified swept laser light to the fiber optic sensor and outputs reflected light from the fiber optic sensor associated with each optical core of the plurality of optical cores. Detection circuitry detects and converts the output reflected light from the fiber optic sensor into a corresponding electrical signal. Data processing circuitry controls the gain of the optical amplifier to control the power of the swept laser light.

The optical amplifier may be implemented, for example, with an Erbium-doped fiber amplifier (EDFA) connected to the output of a pump laser source and to a portion of the laser light from an optical splitter that is coupled to the tunable laser.

An exemplary optical measurement system is an optical frequency domain reflectometry (OFDR) interrogation system. The optical network includes a laser monitor interferometer coupled to detection circuitry and a measurement interferometer coupled to the optical fiber sensor and the detection circuitry. The tunable laser continuously sweeps over a first measurement range of wavelengths such that the detection circuitry acquires OFDR measurement data from the optical fiber sensor during down- and up-sweeps of the tunable laser.

In an exemplary implementation, the data processing circuitry controls the gain of the optical amplifier to compensate for laser power imbalances or fluctuations during upsweep and downsweep of the tunable laser, corrects the gain of the optical amplifier at a plurality of different frequencies within the first measurement range of wavelengths, controls the gain of the optical amplifier to maintain a substantially constant laser power level across the first measurement range of wavelengths, and controls the gain of the optical amplifier to a first gain for the upsweep of the tunable laser and a second different gain for the downsweep of the tunable laser.

Another exemplary embodiment includes an optical frequency domain reflectometry (OFDR) interrogation system for measuring a fiber optic sensor including a plurality of optical cores. A tunable laser sweeps a laser light over a first measurement range of wavelengths to generate a swept laser output signal. A modulator adds a known signal to the swept laser output signal. An optical interference network provides the swept laser light to the fiber optic sensor and outputs reflected light from the fiber optic sensor associated with each optical core of the plurality of optical cores corresponding to sensor measurement data. Detection circuitry detects and converts the output reflected light from the fiber optic sensor into a corresponding electrical signal. Data processing circuitry processes the sensor measurement data acquired during the sweep of the tunable laser through the first measurement range of wavelengths based on the applied known signal.

The data processing circuitry may be configured to determine an error from a delay caused by one or more of the optical interference network, the fiber optic sensor, and the detection circuitry based on the applied known signal.

An exemplary implementation includes a laser driver, and a modulator is coupled to the output of the laser driver. The modulator includes a controller coupled to a digital-to-analog converter that drives a voltage-controlled oscillator, and a filter that filters the output from the voltage-controlled oscillator to generate a known signal. Another exemplary modulator includes a numerically controlled oscillator that generates a binary signal with a most significant bit used to provide a clock signal, and a filter that filters the clock signal to generate a known signal. The fiber may have N optical cores, where N is a positive integer greater than 3, and the modulator includes a numerically controlled oscillator that generates N phase signals and N-1 phase difference signals corresponding to the N optical cores. A phase error in the sensor measurement data may be based on the N-1 phase difference signals.

The tunable laser sweep may include an upsweep, in which the laser light wavelength increases from a minimum wavelength to a maximum wavelength within the first measurement range of wavelengths, and a downsweep, in which the laser light wavelength decreases from a maximum wavelength to a minimum wavelength within the first measurement range of wavelengths. The laser sweep includes a turnaround portion transitioning between the upsweep and the downsweep. The modulator may be controlled to add a known signal to the swept laser output signal during the turnaround portion. Alternatively, the modulator may be controlled to add a known signal to the swept laser output signal at a wavelength outside the first measurement range of wavelengths.

A still further exemplary embodiment includes an OFDR interrogation system having a tunable laser that sweeps over a first measurement range of wavelengths including an upsweep of the tunable laser where the laser light wavelength increases from a minimum wavelength to a maximum wavelength within the first measurement range of wavelengths, and a downsweep of the tunable laser where the laser light wavelength decreases from a maximum wavelength to a minimum wavelength within the first measurement range of wavelengths. The laser sweep includes a turnaround portion transitioning between the upsweep and the downsweep. An optical interference network provides the amplified swept laser light to a fiber optic sensor and outputs reflected light from the fiber optic sensor associated with each optical core of the plurality of optical cores corresponding to sensor measurement data. Detection circuitry detects and converts the output reflected light from the fiber optic sensor to a corresponding electrical signal. Data processing circuitry processes sensor measurement data acquired during the upsweep and downsweep of the tunable laser within the first measurement range of wavelengths and performs additional operations during the turnaround portion of the laser sweep.

For example, the additional operation may be performing additional measurements at swept laser wavelengths other than the swept laser wavelengths within the predetermined range, in which the laser sweep speed for some of the other swept laser wavelengths may be slower than the laser sweep speed for sweeping the tunable laser within the first measurement range of wavelengths.

Other exemplary additional operations include balancing the power levels of the swept laser light in the upsweep and downsweep, performing scattering-based OFDR measurements for optical fiber sensors, and performing in-system checks or adjustments in response to system dynamics.

If the optical fiber sensor includes a fiber Bragg grating that provides optical reflection within the wavelengths of the first measurement range, another exemplary additional operation includes performing a scattering-based OFDR measurement that is separated in wavelength from the grating reflection.

<figref num="1">1 illustrates an exemplary OFDR measurement system.</figref>

<figref num="2">1 is a graph showing rotation about a point in a laser sweep or scan.</figref>

<figref num="3">1 illustrates an exemplary OFDR measurement system in accordance with an exemplary embodiment.</figref>

<figref num="4">2 illustrates an exemplary OFDR measurement system in greater detail in accordance with an exemplary embodiment.</figref>

<figref num="5">1 is a graph showing an example descending hydrogen cyanide (HCN) gas cell power level versus an ascending hydrogen cyanide (HCN) gas cell power level in an OFDR system without correction.</figref>

<figref num="6A">1 is a graph illustrating an exemplary laser sweep.</figref><figref num="6B">1 is a graph showing an exemplary pump laser power.</figref>

<figref num="7">1 is a graph showing an example descending hydrogen cyanide (HCN) gas cell power level versus an ascending hydrogen cyanide (HCN) gas cell power level in an OFDR system with correction.</figref>

<figref num="8">1 is a flow chart illustrating an exemplary procedure for implementing an EDFA in an OFDR measurement system and for adjusting pump laser gain to balance the laser power output for up and down laser sweeps.</figref>

<figref num="9">FIG. 2 is a diagram of a laser drive system.</figref>

<figref num="10">FIG. 13 is a diagram illustrating the application of modulation to the laser driving system in FIG.</figref>

<figref num="11">1 illustrates a first exemplary laser modulator approach.</figref>

<figref num="12">2 illustrates a second exemplary laser modulator.</figref>

<figref num="13">1 is an apparatus for modulation and measurement in accordance with an exemplary embodiment;</figref>

<figref num="14">1 is an exemplary apparatus for phase calculation by mixing with the same signal used to modulate a laser in accordance with an exemplary embodiment.</figref>

<figref num="15">1 is a flow chart illustrating an exemplary procedure for performing delay monitoring using laser diode ripple injection.</figref>

<figref num="16A">1 is a graph showing out-of-band laser modulation before linearization.</figref><figref num="16B">13 is a graph showing out-of-band laser modulation after linearization.</figref>

<figref num="17A">1 is a graph showing laser modulation at laser tuning turnaround.</figref><figref num="17B">1 is a graph showing laser modulation at laser tuning turnaround.</figref>

<figref num="18">11 is a graph illustrating an extended turnaround point for taking additional measurements according to an example embodiment.</figref>

<figref num="19">1 is a graph showing an exemplary out-of-band slow laser sweep portion along with an exemplary faster sweep portion in accordance with an exemplary embodiment.</figref>

<figref num="20A">1 is a graph showing a fast sweep portion.</figref><figref num="20B">13 is a graph showing a slow out-of-band sweep portion.</figref>

<figref num="21">11 is a flow chart illustrating an exemplary procedure for performing additional measurements utilizing laser sweep edges and turnarounds.</figref>

<figref num="22">1 illustrates an exemplary OFDR measurement system in accordance with an exemplary embodiment.</figref>

<figref num="23">1 illustrates an exemplary use of a fiber optic shape sensing system in a robotic surgical arm.</figref>

The following description sets forth specific details such as specific embodiments for purposes of explanation and not limitation. However, it will be understood by those skilled in the art that other embodiments may be utilized apart from these specific details. In some cases, detailed descriptions of well-known methods, interfaces, circuits, and devices are omitted so as not to obscure the description with unnecessary detail. Individual blocks are shown in the figures corresponding to various nodes. Those skilled in the art will understand that the functions of those blocks may be implemented using individual hardware circuits, using software programs and data in conjunction with a suitably programmed digital microprocessor or general-purpose computer, and/or using application specific integrated circuits (ASICs), and/or using one or more (one or more) digital signal processors (DSPs). The software program instructions and data may be stored in a non-transitory computer-readable storage medium, and when the instructions are executed by a computer or other suitable processor controller, the computer or processor performs the functions associated with those instructions.

Thus, for example, it will be understood by those skilled in the art that the figures herein may represent conceptual diagrams of example circuits or other functional units. Similarly, it will be understood that any flow charts, state diagrams, pseudocode, and the like may be substantially embodied in a computer-readable medium and thus represent various processes that may be performed by a computer or processor, whether or not such a computer or processor is explicitly depicted.

The functionality of the various illustrated elements may be provided through the use of hardware, such as circuitry and/or hardware capable of executing software in the form of coded instructions stored on a computer-readable medium, and thus such functionality and illustrated functional blocks should be understood as being either hardware-implemented and/or computer-implemented, and thus machine-implemented.

With respect to a hardware implementation, the functional blocks may include or contain hardware (e.g., digital or analog) circuitry, including, but not limited to, digital signal processor (DSP) hardware, reduced instruction set processors, application specific integrated circuits (ASICs) and/or field programmable gate arrays (FPGAs), and (where appropriate) state machines capable of performing such functions.

With respect to computer implementations, a computer is generally understood to include one or more processors or one or more controllers, and the terms computer, processor, and controller may be used interchangeably. When provided by a computer, processor, or controller, the functionality may be provided by a single dedicated computer or processor or controller, by a single shared computer or processor or controller, or by multiple individual computers or processors or controllers, some of which may be shared or distributed. Moreover, the term "processor" or "controller" may refer to other hardware capable of performing such functions and/or executing software, such as the exemplary hardware described above.

The techniques described in this application determine and reduce or eliminate sources of errors that affect optical fiber measurements.

Gain Boosting and Gain Balancing in OFDR Systems As described in the introduction, FIG. 1 is an exemplary OFDR measurement system. Unamplified light swept over a range of frequencies/wavelengths by a single tunable laser 1 is guided into an optical network 2, and reflected light from a sensor or DUT 3 interferes with light traversing a reference path of the optical network 2. The resulting light is detected by detection and acquisition electronics 4, converted to digital form, and processed in a processor 5 to provide the desired OFDR measurements, e.g., the shape of the fiber sensor 3. In some cases, the OFDR measurements are performed in one laser sweep direction, where the optical frequency increases or decreases. Depending on the application requirements, such as update rate, sweep speed, and sweep range, it is not always appropriate to perform the desired measurements in only a single sweep direction of the laser. Many applications require a laser that is continuously swept with rising/increasing and falling/decreasing sweeps.

An example of a continuous laser sweep is illustrated in FIG. 2, which depicts the optical frequency of the laser as a function of time. The slope of the plot represents the sweep speed of the laser. For an up sweep, the laser is tuned from a lower optical frequency to a higher optical frequency. For a down sweep, the laser is tuned from a higher optical frequency to a lower optical frequency. In addition to these up and down sweeps, there is an additional portion of the sweep referred to in this application as the laser turnaround, which includes the portion of the laser sweep from the completion of one measurement to the start of the next measurement. The turnaround involves the continuation of the current sweep before the laser sweep speed slows down, eventually changing sweep direction, and then accelerating in the opposite direction until the desired sweep speed is reached.

When the fiber sensor 3 is under tension or compression, the fiber cores undergo some amount of elongation or shortening. The bending, twisting, and overall tension cause changes in the strain measured in the various fiber cores. A matrix can be formed to describe the relationship between the bending, twisting, and strain of the fiber sensor and the strain of each core. For example, if four cores in the fiber sensor are used to measure the geometry, the relationship between the strain of these four cores and the applied bending, twisting, and strain as a function of length is as follows:

<math num="1"><img file="JP7553140B2_D0001.tif" /></math>

Here, ε_i(z) is the strain measured in the core as a function of distance below the sensor, z,α is a constant relating the strain to bending (the "bend gain"), β is a constant relating the strain to twist (the "twist gain"), and γ_iis the core distance relative to the center of the fiber_iis the radial location of_iis a core relative to a reference core such as core 2 in Figures 1A to 1C._iis the angle location of B_x(z) is the bending in the XZ plane as a function of distance below the sensor (see Figure 2), and B_y(z) is the bending in the YZ plane as a function of distance, T(z) is the torsion of the sensor as a function of distance, and E(z) is the axial strain applied to the sensor as a function of distance.

Measurement of the amplitude and phase of light reflected along the length of a fiber sensor with high resolution and sensitivity may be achieved using Optical Frequency Domain Reflectometry (OFDR).

In the case of a multi-channel OFDR interrogation system, each channel corresponds to a DUT or a core of a multi-core fiber sensor. In a multi-channel OFDR interrogation system, it may be advantageous to add an amplifier to increase the power to each channel. An exemplary embodiment adds an erbium-doped fiber amplifier (EDFA) to the OFDR system to increase the power of the laser light coupled into the fiber sensor. However, the EDFA introduces a new variable, namely, amplifier gain, to the OFDR measurement system. When the pump laser in the EDFA is driven by a constant current source, the gain of the EDFA may vary depending on various factors, including the direction in which the swept laser is tuned, the instantaneous wavelength of the light being amplified, and the sweep speed or laser speed.

An exemplary embodiment of an added EDFA is shown in FIG. 3. A tunable laser 1 controlled by detection, acquisition, and control electronics 4 is shown with an exemplary wavelength sweep range of 1520 nm to 1560 nm. A pump laser 7 controlled by detection, acquisition, and control electronics 4 provides light at a specific wavelength, e.g., 980 nm, to an erbium-doped fiber amplifier (EDFA) 8. An optical splitter 6 splits the light from the tunable laser 1 into two paths: one to the EDFA 8 and one to a reference path of the optical network 2. The amplified light from the EDFA 8 is split into each OFDR measurement channel and guided to each core of a multi-core sensor fiber containing a fiber sensor 3.

FIG. 4 shows a detailed exemplary embodiment of an EDFA added to an OFDR-based interrogation system for an exemplary 6-core fiber sensor. Light from a frequency tunable laser 16, controlled in this example by a processor 22 rather than by data acquisition electronics 20, is split with a 90/10 coupler between the laser monitor interferometer 10 and the measurement interferometer 12. In the laser monitor interferometer 10, the light is split into three paths using a 3×3 coupler. The first path goes to a detector to monitor the laser power. The second path goes through a hydrogen cyanide (HCN) gas cell to the detector to provide an absolute wavelength reference. The final path goes through an isolator and another 3×3 coupler to two Faraday rotator mirrors (FRMs), one leg with a known delay difference from the other leg. The return signal from this interferometer forms the 1/Q signal. With a phase offset of 120 degrees, the I/Q signals are converted into quadrature signals and used to measure the change in optical frequency as the laser sweeps.

The light going to the measurement interferometer 12 is split using a 90/10 coupler between the reference and measurement branches of the interferometer 12. The light in the reference branch is split into six reference signals using cascaded couplers. The light in the measurement branch passes through an isolator and then a length of Erbium doped fiber, which in this example is wavelength division multiplexed. Light is pumped from a 980 nm pump laser 18, controlled by a processor 22 rather than by data acquisition electronics 20, which couples through a Wavelength Division Multiplexed (WDM) coupler. This combination of Erbium doped fiber and pump laser 18 amplifies the light in the measurement branch of the interferometer. The light passes through another isolator and then a polarization controller that is set to flip the light between two orthogonal (or nearly orthogonal) polarization states on subsequent scans. The light is then split into six measurement channels with cascaded couplers. The returning light is combined with six reference paths using 2x2 couplers. These combined signals then go through a polarizing beam splitter (PBS) to two detectors (S and P) for each channel (C, I, J, K, U, V) that are input to the data acquisition circuitry 20, for polarization diverse detection. A multicore fiber sensor 24 is formed using a multicore fiber optics (MI) optical fiber optics (MIS) optical fiber optics (OF ...

The recorded data is the reflected amplitude as a function of optical frequency for the two polarization states S and P for each optical fiber core being measured. The controller/data processor 22 linearizes this recorded data with respect to optical frequency using data from the laser monitor interferometer 10 so that it is represented in equal increments of optical frequency. The linearized data is Fourier transformed to the time domain to represent the amplitude and phase of the reflected light as a function of the optical delay along each fiber core. The S and P data from two successive orthogonal polarization scans are combined to compensate for birefringence in the fiber core and to form a scalar measure of the amplitude and phase of the reflected light from each core. This combined complex signal (amplitude and phase) is compared to the interferometer data recorded in the reference scan, and the resulting phase difference/change for each core is the measurement signal used to calculate the current shape of the fiber.

The derivative of the measured phase change is proportional to the strain in each core._iThe proportionality constant γ that relates the strain in_iis the strain-optical coefficient for the core. Equation 1 can be expressed as:

<math num="2"><img file="JP7553140B2_D0002.tif" /></math>Here, φ_i'(z) is the derivative of the measured phase change for core i as a function of distance below the fiber sensor 24.

The inverse of this equation is needed because the position of the fiber sensor is first found by measuring the phase change in each core, and then calculated by integrating Bx(z) and By(z) along the fiber while accounting for the twist and bend.

<math num="3"><img file="JP7553140B2_D0003.tif" /></math>Where:<img file="JP7553140B2_D0004.tif" />is known as the shape matrix.

Although the addition of a fiber amplifier in the measurement branch provides the benefit of increased power to the sensor or DUT, it also introduces errors into the OFDR measurement in the form of power fluctuations between the up and down sweeps of the laser. An example of these fluctuations is shown in the graph of FIG. 5, which illustrates the difference in power levels between the up and down laser sweeps due to the difference in amplification between the sweeps made over 10 nm at 2,937,500 GHz/s. The power levels detected using an HCN gas cell (not shown in FIG. 4) connected in place of the sensor or DUT show more than a 2 dB difference between the up and down laser sweep measurements. Note that in FIG. 5, the up laser sweep data (thick black line) is inverted for comparison with the down laser sweep data (thin black line). Both are displayed from higher to lower optical frequencies.

A change in pump laser power results in a gain change in the EDFA with a response time that is a function of the signal and pump powers and the fluorescence lifetime in the erbium-doped fiber. When the pump laser power is adjusted, there is a delay before the power at the output of the EDFA changes. This delay, along with the gain response of the EDFA as a function of wavelength, can affect the EDFA output when the tunable laser is swept. To compensate for how the gain of the EDFA changes as a function of wavelength, a gain flattening filter (GFF) at the output filter may be added. However, there is a gain difference that changes based on the sweep direction even at the same wavelength. The GFF also undergoes a temperature-dependent wavelength shift.

In a continuously swept OFDR system, OFDR measurement data is obtained when the optical frequency of the tunable laser is increased (laser upsweep) and when the optical frequency of the swept laser is decreased (laser downsweep). The light is amplified using an optical amplifier, whose gain is higher during the downsweep than during the upsweep. This gain difference can result in less optimal system performance and less accurate measurement results.

To correct for the unbalanced power during and between sweeps, the exemplary embodiment differs the pump laser power between predetermined points to maintain a substantially constant level of output power to the sensor fiber or device under test (DUT) in the OFDR system. In one exemplary embodiment, the amplifier gain is modulated between two states, one for the up sweep and one for the down sweep. This embodiment allows the power to be balanced in the two laser sweep directions. Furthermore, the amplifier gain may be modulated within the laser sweep with the addition of gain set points to flatten the power over the optical frequency range over which the OFDR measurement data is acquired. The exemplary embodiment also uses a turnaround to make adjustments to balance the laser power output between the up and down laser sweeps. One exemplary adjustment includes adjusting the current that drives the diode of the tunable laser over the entire sweep cycle, including the up, down, and turnaround portions.

In an exemplary embodiment, to compensate for the gain difference depending on the sweep direction, the data, acquisition, and control electronics 4 adjusts the amplifier pump laser power to a pre-calibrated level for the upsweep and downsweep. Specifically, the system is configured to include two power states for the pump laser: one for the downsweep and one for the upsweep. (The adjustments may alternatively be controlled by the processor 5). An example of this pump laser power switching is illustrated in Figures 6A and 6B.

6A and 6B, the down-sweep power is left unchanged and the up-sweep power is increased to match the down-sweep power. Because of the amplifier gain response delay, the pump laser is driven to a higher power level following completion of the down-sweep at the beginning of the laser turnaround. Setting the pump laser power level at the beginning of the turnaround allows enough time for the amplifier gain to adjust to the new level.

FIG. 7 compares the HCN power levels for the upsweep (thick line) and downsweep (thin line) after an adjustment is applied to the amplifier pump laser power, which reduces the average power difference between the up and down laser sweeps to less than 0.02 dB.

In addition to adjusting the pump laser power to compensate for sweep direction dependent gain, the pump laser power can be adjusted to include more than two set points during the course of the laser sweep cycle to further flatten the power to the DUT.

FIG. 8 is a flow chart illustrating an exemplary procedure for implementing an EDFA in an OFDR measurement system and for adjusting pump laser gain to balance the laser power output for the up and down laser sweeps. The power requirements of the EDFA are determined (step S1), and the EDFA is configured to meet the gain requirements, including the type of gain medium, length, and pumping direction (step S2). The placement of the EDFA in the optical network is determined, and the EDFA is added to the optical network (step S3). The tunable laser is configured for the desired sweep behavior (step S4). The gain difference between the up and down laser sweeps is determined (step S5). To balance the EDFA gain between the up and down laser sweeps, a pump laser power set point is added at a specific optical frequency (step S6). The pump laser power set point is adjusted to balance the EDFA output power (step S7).

In fiber optic sensing systems (modulating the laser output with a known signal), measuring delay is important. One delay that is measured is the delay change between the cores of a multicore fiber sensor. This delay change can be measured in terms of a phase shift or phase difference from a reference phase for each core. In OFDR systems, there are many delay paths that the optical signal and the corresponding electrical signal undergo before being detected by the OFDR detection and acquisition system. One exemplary delay, among many other delays, is associated with the reference path of the measurement interferometer through the detection and acquisition system to the photodiode.

Exemplary embodiments monitor the change in delay of these paths and provide measurements that can be used to signal erroneous data to correct for dynamic phase shifts in the reference paths of the optical network. More specifically, by injecting a known signal, e.g., a ripple signal having a known frequency, into the laser, the phase of each delay path at that frequency can be calculated. In one exemplary implementation, the calculation is done without performing a Fast Fourier Transform (FFT). Performing this measurement over a set of frequencies/wavelengths produces a phase response of each detection channel in the system.

In addition to adjusting the amplifier pump laser during laser turnaround, exemplary embodiments perform in-system testing during laser turnaround. In these exemplary embodiments, a tunable laser diode output in an optical (e.g., OFDR) measurement system is modulated to detect changes in optical and electrical delays in the system. Modulation in the following examples includes injecting a known signal, e.g., a ripple with a known frequency, into the tunable laser diode output. Other known signals or modulation techniques may be used.

Since a constant laser output is desired in an OFDR system, a control system and processing algorithm may be specially designed to reduce or compensate for any ripple present in the tunable laser output. A typical diode driver circuit for driving and maintaining a constant power output is illustrated in FIG. 9. The diode driver 30 that drives the laser diode 32 may be a current source controlled via an analog and/or digital closed loop control scheme. The driver 30 may be a look-up table of values stored in memory that is retrieved by a processor to adjust the current to a predetermined level at a particular point in the laser's sweep cycle in an open loop manner. This table may be periodically updated by an internal and/or external control process to maintain the laser power level over time.

Unwanted modulation in the laser power can introduce errors into the OFDR measurement in the form of broadband noise or as reflective events that are not actually present in the sensor or DUT. In contrast to conventional thinking, the exemplary embodiment intentionally injects ripple into the laser diode output signal with steps taken so as not to degrade the OFDR measurement.

10 depicts a laser diode driver 30 with a modulator 34, the output of which is combined by a combiner 36 and provided to a laser diode 32. The modulator 34 can be implemented in a variety of ways and can be combined with the diode driver 30 signal in a variety of ways, such as by sending the signals together or by adding the signals using an op amp configured as an adder.

An example of a modulator 34 is shown in FIG. 11 and includes a processor-controlled digital-to-analog converter (DAC) 40, a voltage-controlled oscillator (VCO) 42, and an active filter 44 that provides signal gain control and low-pass filtering. In this exemplary embodiment, the processor 5 selects the VCO 42 output signal frequency by setting the DAC 40 output voltage to a desired value. The VCO 42 output signal may then be used as a modulator by adding this signal to a laser diode drive signal. Depending on the desired signal characteristics, such as amplitude and noise level, additional filtering/amplification 44 may be applied. The exemplary processor 5 includes a field programmable gate array (FPGA), a microcontroller, a digital signal processor, or other processor. One disadvantage of this modulation approach is that an additional data acquisition channel is required to measure the frequency of the VCO output in order to measure the phase difference between the modulated signal and the detected optical signal.

Another exemplary embodiment of the modulator 34 is shown in FIG. 12 and includes a processor 5 having or controlling a numerically controlled oscillator (NCO) 46. The NCO 46 generates a digital output representing a sine wave at a desired frequency. From this digital output, a clock signal is generated that corresponds to the most significant bit (MSB) 48 of the NCO output. The digital clock signal output from the processor 5 is low pass filtered by a filter/shaper 50 to band limit the content of the digital clock signal. One exemplary filter/shaper 50 is an RC filter. In this exemplary embodiment, the filtered clock signal is a modulation signal that is added to the laser diode drive signal to inject ripple into the laser output. This technique requires minimal external components and has the advantage that the ripple signal (clock) is provided directly from the processor 5. As a result, the processor 5 knows the frequency of the ripple signal, so no additional measurements are required to determine the frequency of the modulation signal. These configurations are advantageous when making high-resolution phase difference measurements in OFDR systems.

Driving a laser diode with the addition of this type of modulator produces optical and electronic detection circuitry, e.g., a laser light carrying a known frequency ripple signal component all along the path with known amplitude modulation. By measuring the light traversing the reference path of the optical network, the amplitude modulated signal in each measurement channel in the OFDR system is detected in the data acquisition electronics 4. The phase difference of the detected ripple signal in each channel (each channel corresponds to a core in the multicore fiber sensor) from the ripple originally injected at the modulated frequency is measured, and this phase difference is a measure of the optical and electrical delay difference between the channels.

An exemplary modulation and measurement system is shown in FIG. 13. As an example, assume that the NCO 46 is set to 12.5 MHz and the MSB 48 is used as a clock signal that is sent to a filter/shaper 50 and combined at 36 with the laser output from the diode driver 30. The modulated driver output then drives a tunable laser 32 that is used to sweep the optical network 2. The optical network response for the N channels is detected by a photodiode 52 and the corresponding N electrical signals are conditioned (54) and converted by an ADC 56 into N digital signals corresponding to the N core ODFR measurement channels. The N channel digital signals are processed by a DSP or processor 58 to generate N different channel phases and calculate and output N-1 phase difference signals between each determined channel phase relative to one common channel phase. The common channel may be selected as one of the OFDR channels.

Data from the N channels is acquired by the processor 5 in the time domain, meaning that each point is sampled at a particular frequency, for example 200 MHz. To determine the phase of the injection ripple for each channel, a Fast Fourier Transform (FFT) is performed on the measurement data from each channel, without the need to perform data linearization. This differs from the processing for sensor/DUT OFDR measurements, where the OFDR measurement data must first be linearized or resampled based on the laser tuning speed during the course of the measurement. Once the FFT is performed on the nonlinearized measurement data, a peak is observed in the data at index 12,500, which corresponds to a 12.5 MHz modulated signal (in this example). To determine the phase difference between the N channels, the DSP 58 takes the FFT results and calculates the complex point at index 12,500 for each channel. The phases of the N-1 points (N-points) are calculated to determine the phase of the signal detected at the modulation frequency. One common channel (exemplary channel 0) is then subtracted from all N channels. These resulting N-1 phase differences from each of the N-1 channels relative to common channel 0 indicate the N-1 delay differences between the N cores in the fiber sensor. The N-1 delay differences between the N cores in the fiber can generate significant errors, for example in shape sensing applications, if not removed from the OFDR measurement.

By sweeping the NCO 46 over the frequency range of the DUT or sensor, the phase of each OFDR measurement channel at each frequency is calculated to obtain the phase response of the electronic acquisition and detection circuitry and the reference path of the optical network for each core measurement channel as a function of frequency. This phase response is used to correct for phase differences between each measurement channel. Additionally, monitoring this phase response for changes provides real-time monitoring of delay paths in the optical network and electronics, which may be used to provide feedback to detect measurement errors induced as a result of changes in the optical and electrical delay paths.

The inventors have realized that since the phase of only a single point of the FFT is required, the phase can be calculated without performing a full FFT by mixing the data from the N channels to baseband using the same frequency as the modulator, which in this example is the output of the NCO used to provide the modulating signal. Once at baseband, the DC term of each channel becomes the point at which the phase is calculated and compared to the common channel. The phase is calculated by accumulating the complex components of the baseband signal of each channel and calculating the phase of the resulting complex value. These operations can be performed, for example, in a DSP or FPGA.

An example of such DSP processing is shown in Figure 14. The N digital samples from the N core OFDR measurements are mixed with the output signal from the NCO 46 in a mixer 60. The mixer 60 multiplies each channel by the complex NCO 46 output (sine and cosine). The complex mixed signals are accumulated (summed) at 62 and converted from complex to polar form to determine the phase. These phase values represent the phase difference between the modulated signal and the signal detected for each channel. To obtain the phase difference between the channels, the phase of one common channel is subtracted from each of the N channels at 66 to determine N-1 phase differences between the channels.

FIG. 15 is a flow chart illustrating an exemplary procedure for performing delay monitoring using laser diode ripple injection. Electronics are added to the OFDR system to modulate the laser diode at one or more frequencies (step S10) and select the frequency(ies) of the injected ripple (step S11). The added ripple is located at a frequency or frequencies outside the measurement frequency range of the sensor to ensure that the added ripple does not interfere with the sensor measurement (step S12). The ripple signal is injected into the laser diode output (step S13) and OFDR data and phase measurements are taken for all monitored channels at the modulation frequency(ies) (step S14). The DSP calculates the phase difference of the modulated signal among the N channels and determines whether a change in optical or electrical delay has occurred (step S15).

The modulated laser light (which uses the laser sweep edges and turnarounds to perform additional measurements) can interfere with the sensor measurements, but it is also possible that the interference is caused by reflective events within the fiber sensor. The inventors have recognized that the effect of linearization on this modulated signal needs to be understood and accounted for. Linearization refers to the process of resampling the acquired measurement data based on the instantaneous tuning speed of the laser. The result of the linearization process is measurement data that is evenly spaced in optical frequency rather than in time.

Figure 16A shows how out-of-band modulation without linearization produces a strong observable peak on the sensor grating at the frequency of the modulation source. Linearization is not required to measure the modulated signal because this is not an OFDR measurement, but just a measurement of the amplitude modulated signal in a time domain measurement. Figure 16B illustrates the effect of linearization on such a signal. The resampling performed as part of the linearization process spreads this signal over a wide range of frequencies based on the tuning speed variations of the swept laser.

Modulating the laser power can interfere with the measurements being performed if the modulation frequency and timing of this signal are not taken into account. For example, if the modulated signal is too close to a sensor grating included along the fiber sensor, it may interfere with the OFDR grating measurement when the data is linearized. In OFDR-based measurements, the modulation of the laser diode can be performed at the same time as the OFDR measurement if the modulation frequency is outside the expected frequency range of the OFDR measurement. However, care must be taken to ensure that fiber sensor connector and termination reflections are minimized. Otherwise, these reflections can generate interference signals when combined with the modulated signal generating additional unwanted frequencies of modulation in the fiber sensor that can introduce errors into the OFDR measurement.

If grating measurements are being performed, higher grating strengths may corrupt the measurement of the modulated signal. The inventors have recognized several options to address this problem. One option is to limit the modulation frequency to frequencies outside the frequency of the grating fiber. The disadvantage of this option is the possibility of measurement errors, since the frequency where the sensor is actually located is not being measured. This disadvantage may be eliminated by moving the modulation from a frequency in the measurement region to the turnaround frequency. Depending on the amplifier response and the duration of the turnaround, the amplifier can be turned off completely or the amplifier gain can be reduced. Reducing the amplifier gain sufficiently can reduce the grating reflections below the noise floor, thereby allowing the modulation measurement to be performed without any measurable interference from the grating.

To avoid measurement interference, an exemplary embodiment performs a laser diode modulation based measurement during the turnaround points of the laser sweep, as illustrated in Figures 17A and 17B. Although the example in Figure 17B shows modulation occurring only during the turnaround time, there may be embodiments in which some modulation may occur outside of the turnaround time or continuously.

The inventors have recognized that these laser sweep turnaround points provide opportunities to make further corrections, perform validation, and make other dynamic adjustments. This continuous sweeping action of the laser can be further broken down into different sections, which are illustrated in the graph of FIG.

One exemplary embodiment uses a laser sweep turnaround to perform additional OFDR measurement data acquisition beyond the fiber sensor wavelength. For example, for a fiber sensor including a Bragg grating, the grating reflection can be designed or the fiber can be constrained to reflect light within a known wavelength range. The continuous laser sweep is configured such that the turnaround exceeds the wavelength range in which those OFDR measurements are obtained from the fiber sensor, i.e., beyond the fast rise and fast fall portions of the laser sweep shown in FIG. 19. This allows the OFDR measurement to be performed over the additional wavelength range in the turnaround portion of the laser sweep. Extending the turnaround to include the additional range also allows Rayleigh scattering-based OFDR measurement (different from Bragg grating-based OFDR measurement) to be performed.

In addition to acquiring data beyond the frequency range of the grating or the main OFDR measurement, out of band OFDR measurements can be taken at slower sweep speed portions of the laser sweep as illustrated in Figure 19. By slowing down the laser sweep speed in these portions, out of band OFDR measurements can be made over longer distances along the fiber sensor.

An exemplary extension of the OFDR measurement range is illustrated in Figures 20A and 20B. Figure 20A illustrates an exemplary fiber Bragg grating amplitude measurement at a faster sweep speed and shorter delay or length along the fiber sensor. Using a wider laser sweep or scan range results in higher resolution, but a faster sweep speed results in a shorter measurement range when the Nyquist rate (i.e., the minimum speed at which the signal can be sampled without introducing error, which is twice the highest frequency present in the signal) is placed at a shorter delay/distance from 0 ns. Figure 20B shows a graph of a slower, lower resolution sweep. As a result of using a slower sweep speed and an optical frequency beyond the range of the grating, reflection events beyond those of the sensing fiber can be measured.

21 is a flow chart illustrating an exemplary procedure for utilizing laser sweep edges and turnarounds to perform additional measurements. The laser is configured to perform a desired sweep behavior, including, for example, a desired sweep speed and power for performing additional measurements (step S20). Acquisition specific parameters, such as a desired number of measurement points and wavelength range of the measurements, and where to begin wavelength-specific data acquisition are determined (step S21). Data acquisition circuitry acquires and processes ODFR measurements over the specified wavelength range (step S22).

22 illustrates an example of an OFDR system including a number of configurations from above, such as a frequency-specific power level and a laser modulator, such as one of the examples above, that adjust the power level of the pump laser. According to an exemplary embodiment, a table of frequency-specific power levels stored in memory 35 is used by processing control circuitry in detection, acquisition, and control electronics 4 to control the pump laser power level. The output of the modulator 34 is coupled to the output of the diode driver circuit 30 to control the output power of the tunable laser 1, for example, for the reasons described in the previous paragraph, and/or to inject a modulation signal into the laser output, for example, for the reasons just described in this paragraph.

The above-described techniques have a wide variety of applications for improving the accuracy and reliability of optical network measurements. One non-limiting application for shape-sensing fibers coupled to OFDR measurement instruments that require a high degree of confidence in the accuracy and reliability of the shape-sensing output is a robotic arm used in surgery or other environments. Figure 23 shows an exemplary use of an optical fiber shape-sensing system for a robotic surgical arm in which one or more of the various technical configurations and/or embodiments described above may be used.

Although various embodiments have been shown and described in detail, the claims are not limited to any particular embodiment or example. None of the above description should be read as implying that a particular element, step, range, or function is essential to be included in the scope of the claims. The scope of legal protection is defined by the terms recited in the allowed claims and their equivalents. All structural and functional equivalents to the elements of the above-described preferred embodiments known to those of skill in the art are expressly incorporated by reference herein and are intended to be encompassed by the claims. Moreover, a device or method need not address each and every problem sought to be solved by the described technology to be encompassed by the claims of the present invention. No claim is intended to invoke Section 6 of 35 USC § 112 unless the term "means for" or "step for" is used. Moreover, no embodiment, configuration, component, or step in this specification is intended to be dedicated to the public, regardless of whether or not such embodiment, configuration, component, or step is recited in the claims.