This application relates to methods and apparatus for distributed fibre optic sensing and especially to Rayleigh backscatter based distributed fibre optic sensing.

Distributed fibre optic sensing is a known type of sensing where an optical fibre is deployed as a sensing fibre and interrogated with electromagnetic radiation to provide sensing of environmental stimuli affecting the sensing fibre along its length. By analysing the radiation backscattered from within the sensing fibre, the sensing fibre can effectively be divided into a plurality of discrete sensing portions which may be (but do not have to be) contiguous.

One class of distributed fibre optic sensors is based on illuminating the sensing fibre with coherent illuminating radiation and detecting illuminating radiation which has been Rayleigh backscattered from inherent scattering sites within the sensing fibre, i.e. radiation which has been elastically backscattered at the same frequency as the interrogating radiation. Within each discrete sensing portion of the sensing fibre, mechanical disturbances of the fibre, for instance, dynamic strains due to incident acoustic waves, cause a variation in effective optical path length of that section, which results in a variation in the properties of the radiation which is backscattered. This variation can be detected and analysed and used to give an indication of disturbance of the fibre at that sensing portion. Such a fibre optic sensor effectively acts as a linear sensing array of sensing portions of optical fibre which are responsive to disturbances such as acoustic stimuli, and thus such sensing is often referred to as distributed acoustic sensing (DAS), although the same principles can be applied to detect any stimulus that results in a variation in effective optical path length of the sensing fibre, such as temperature variations.

In one form of a Rayleigh backscatter based DAS system, based on the principles of coherent optical time domain reflectometry (COTDR), the sensing fibre is repeatedly interrogated, with each interrogation involving launching a single continuous pulse of coherent interrogating radiation into a first end of the sensing fibre and detecting the backscatter from the pulse. As the pulse propagates along the sensing fibre, different portions of the fibre are effectively illuminated by the propagating pulse at different times and there will be at least some backscatter due to inherent scattering sites within the optical fibre. The received backscatter may be analysed in different times bins based on the time after launch of the pulse to provide the different channels of the distributed fibre optic sensor corresponding to sensing portions of the sensing fibre.

The backscatter arriving back at the first end of the sensing fibre at any time will include contributions of backscatter from various different scattering sites from a portion of fibre illuminated by the pulse. As the interrogating radiation is coherent, the backscatter from the different scattering sites will interfere to provide an overall interference signal. The intensity of this backscatter interference signal will depend upon the extent to which the various contributions from the different scattering sites constructively or destructively interfere, which depends on the distribution of the inherent scattering sites within the fibre, which will vary effectively randomly along the length of the fibre.

Thus, the intensity of the backscatter received from any given interrogating pulse will exhibit a random variation from one sensing channel to the next but, in the absence of any environmental stimulus, the backscatter intensity from any given sensing channel should remain the same for each repeated interrogation (provided the characteristics of the interrogating pulse remains the same). However, a stimulus acting on the relevant sensing portion of the fibre can result in an optical path length change for that section of fibre, e.g. through stretching/compression of the relevant section of fibre and/or a refractive index modulation. A change in optical path length will vary the distribution of the scattering sites within the sensing portion and hence vary the degree of interference of all the individual scatter signals and thus result in a change in backscatter intensity of the overall interference signal. This change in intensity can be detected and used as an indication of a disturbance acting on the fibre, such as an incident acoustic wave.

The rate at which interrogation pulses can be introduced into the sensing fibre or “ping rate” is determined by the time taken for each pulse to propagate from the proximate end or launch end of the fibre, to the distal end and back again i.e. the round trip travel time of the radiation pulse in the fibre. This condition ensures that the backscatter radiation received at the detector can be spatially resolved based on the time taken to receive the backscatter radiation following launch. If pulses from two separate interrogations were propagating in the fibre at the same time, and the pulses had the same optical frequency as one another, the Rayleigh backscatter radiation from either pulse would also have the same frequency. Absent any other means of distinguishing the pulses, e.g. some coding, the processor would thus be unable to distinguish the backscatter radiation from a pulse of one interrogation from the backscatter radiation of the pulse of the other interrogation and thus the backscatter from the two pulses could not be spatially discriminated.

In other types of Rayleigh backscatter based DAS system, the interrogating radiation for each interrogation may take different forms. For instance, in some DAS systems each interrogation comprises two pulses, which may be at different optical frequencies to one another. The backscatter from each such interrogation will thus be an interference signal that includes a signal component at the difference frequency, which can provide advantages in processing. In some systems each interrogation may comprise a pulse with a time varying frequency, i.e. a frequency swept or chirped pulse. In each case however the same optical characteristics, e.g. launch frequencies, frequency range etc. are used for each interrogation in a series of interrogation. Thus the same issue arises that if backscatter were received from one interrogation of the series were received at the same time as backscatter from an earlier interrogation, with the same optical characteristics, then it would not be possible to discriminate between the backscatter from the two interrogations. Thus, the ping rate is set to allow a complete round trip travel time for one interrogation before launching another interrogation with the same optical characteristics.

The ping rate sets the sample rate at which the fibre can be interrogated, which thus defines the sample rate for the measurement signal from any sensing portion. As noted above, the distributed fibre optic sensor effectively detects any stimulus by monitoring any variation in the response of a given sensing portion between successive interrogations of the sensing fibre. The sample rate sets the limit of the maximum frequency of the stimuli which the fibre can reliably detect e.g. an acoustic vibration, e.g. according to the Nyquist limit.

Therefore, the longer the length of fibre, the lower the ping rate and thus the lower the maximum frequency of stimulus, e.g. incident acoustic wave, that can be detected. This therefore may mean that lengths of fibre over a certain length are unsuitable for detecting certain stimuli of interest. Furthermore, for some applications, even if a signal of interest can be detected at the relevant ping rate, a higher sampling rate may be preferable, e.g. for bandwidth reasons.

Embodiments of the present disclosure relate to methods and apparatus for distributed fibre optic sensing, especially for fibre optic distributed acoustic sensing, that at least mitigate some of the issues noted above.

Thus, according to an aspect, there is provided a distributed fibre optic sensing apparatus, comprising: an optical source operable to repeatedly interrogate an optical fibre with interrogating optical radiation having a launch frequency, the interrogating optical radiation having an optical power such that the interrogating optical radiation experiences spectral spreading when propagating in the optical fibre; a detector configured to receive a backscatter signal comprising radiation from the interrogating optical radiation which is Rayleigh backscattered from within the sensing fibre; and a narrowband filter configured to filter the backscatter signal, wherein the narrowband filter comprises a passband at the launch frequency; wherein the optical source is operable such that an interval between repeated interrogations is less than a round trip travel time for the interrogating optical radiation to a distal end of the optical fibre.

In some examples the interrogating optical radiation may have optical properties such that the interrogating optical radiation experiences spectral spreading to the extent that, any component of the backscatter signal at the launch frequency received from beyond a cut-off distance into the optical fibre, which is before the distal end of the optical fibre, is below a noise floor of the detector. The optical source may be operable such that an interval between repeated interrogations is at least equal to a round trip travel time for the interrogating optical radiation to the cut-off distance in the optical fibre.

In some examples, the apparatus may be operable to provide measurement signals from one or more sensing portions of the sensing fibre corresponding to an identified region of interest in a part of the optical fibre before the cut-off distance. The region of interest may extend from a first point into the optical fibre to a second point further into the optical fibre. The apparatus may be operable to repeatedly interrogate the optical fibre such that, for the identified region of interest, at any time when the backscatter signal may comprise a component arising from the region of interest for a current interrogation, any contribution to the backscatter signal from any previous interrogation will arise from beyond the cut-off distance. The apparatus may be configured such that an interval between repeated interrogations is at least equal to a round trip travel time for the interrogating optical radiation to the second point in the optical fibre.

In some instances, the first point is the optical fibre may be the launch end of the optical fibre. In other instance, the first point in the optical fibre is a first, non-zero, distance into the optical fibre. In some examples, the optical source may be configured to launch a new interrogation into the optical fibre before a time (tcuttoff) when backscatter from the previous interrogation will be received from beyond the cut-off distance. The optical source may launch the new interrogation earlier than this time (tcuttoff) by an amount of time which is at most equal to the round-trip travel time to the first distance.

The apparatus may further comprise a processor configured to processes the backscatter signal in one or more time bins corresponding to the region of interest to generate said measurement signals from one or more sensing portions of the sensing fibre corresponding to the identified region of interest. The processor may be configured so as to not output measurement signals for any sensing portions of the optical fibre outside of the identified region of interest.

In some examples the apparatus may be operable to controllably define at least one of the first point and the second point of the identified region of interest. That is the

In some example the apparatus may be configurable to control optical properties of the optical radiation so as to controllably vary the cut-off distance.

In some implementations the optical source may also be operable, in a second mode of operation, to repeatedly interrogate the optical fibre with interrogating optical radiation having an optical power which is below a non-linear threshold such that the interrogating optical radiation does not experience significant spectral spreading when propagating in the optical fibre. In the second mode of operation the interval between repeated interrogations may be equal to or greater than a round trip travel time for the interrogating optical radiation to a distal end of the optical fibre.

In another aspect there is provided a method of interrogating an optical fibre comprising: interrogating, repeatedly, an optical fibre with interrogating optical radiation at a launch frequency, the interrogating optical radiation having an optical power such that the interrogating optical radiation experiences spectral spreading when propagating in the optical fibre; receiving a backscatter signal comprising radiation from the interrogating optical radiation which is Rayleigh backscattered from within the sensing fibre; and filtering the backscatter signal with a narrowband filter, wherein the narrowband filter comprises a passband at the launch frequency; wherein an interval between repeated interrogations is less than a round trip travel time for the radiation pulse to complete a round trip propagation to the distal end of the optical fibre.

It should be noted that any of the features of any of the examples or embodiments described herein may be implemented in combination with any one or more of any of the other described features, unless such combination is explicitly ruled out or such features are clearly incompatible.

Embodiments, and feature of embodiments of the present disclosure, will now be described by way of example only with respect to the accompanying drawings, of which:

Figure 1 illustrates an example of a distributed fibre optic sensor system;

Figure 2 illustrates an example of a radiation pulse propagating within a sensing fibre;

Figure 3 illustrates another example of radiation pulses propagating within a sensing fibre;

Figure 4 illustrates the principle of spectral spreading for a radiation pulse propagating within a sensing fibre;

Figure 5 illustrates another example of radiation pulses propagating within a sensing fibre launched with a distributed sensing apparatus according to an embodiment;

Figure 6 illustrates a further example of radiation pulses propagating within a sensing fibre launched with a distributed sensing apparatus according to an embodiment.

Embodiments of the present disclosure relate to distributed fibre optic sensing apparatus with a ping rate which is not limited by the round-trip travel time of a radiation pulse over the length of the fibre. This can allow an increase in ping rate compared to the conventional approach, which can provide an increased sample rate and hence increase the maximum measurable frequency of the stimuli acting on the sensing fibre. Some embodiments relate to a distributed fibre optic sensing apparatus with improved sensing along a region of interest of an optical fibre.

Figure 1 shows a schematic of a distributed fibre optic sensing arrangement 100. A length of optical fibre 101 is removably connected at one end to an interrogator 102, which, in use, interrogates the optical fibre to provide sensing of environmental disturbances acting on the optical fibre. The optical fibre 101 will be referred to herein as a sensing optical fibre or just a sensing fibre. The measurement signals from interrogator 102 may, in some implementations, be passed to a signal processor 103, which may be co-located with, or integrated into, the interrogator or may be remote therefrom. Optionally there may also be a user interface/graphical display 104, which may be co-located with the signal processor or may be remote therefrom and in practice may be realised by an appropriately specified PC.

The sensing fibre 101 can be many kilometres in length and can, in some applications be tens of kilometres in length, say up to 40 km or more. For distributed fibre optic sensing, the sensing fibre 101 may be a standard, unmodified single mode optic fibre such as is routinely used in telecommunications applications without the need for deliberately introduced reflection sites such a fibre Bragg grating or the like. The sensing optical fibre will be protected by containing it with a cable structure which may contain more than one optical fibre. In embodiments of the disclosure the cable structure may be configured to provide a specific response to incident strains.

In use, the fibre optic cable comprising the sensing fibre 101 is deployed in an area of interest to be monitored. Depending on the particular use case, the sensing fibre may be deployed in a relatively permanent manner, e.g. being buried or otherwise secured in place. The interrogator 102 may be removably coupled to the sensing optical fibre 101 , and thus in some instances, if continuous monitoring is not required, the interrogator 102 may be removed from the sensing fibre 101 when sensing is not required, possibly leaving the sensing fibre in situ.

In operation, the interrogator 102 launches coherent electromagnetic radiation, e.g. optical radiation, into the sensing fibre. The electromagnetic radiation launched into the sensing fibre by the interrogator 102 will be referred to herein as interrogating radiation. The sensing fibre may, for instance, be repeatedly interrogated with optical radiation in a series of interrogations. In some examples a single pulse of optical radiation at a given launch frequency may be used for each interrogation, although in some embodiments each interrogation may comprise two (or more) pulses, in which case the optical pulses may have a frequency pattern as described in GB patent publication GB2,442,745 or optical characteristics such as described in WO2012/137022, the contents of which are hereby incorporated by reference thereto. In some examples the interrogating radiation could comprise at least one pulse of optical radiation with a time varying frequency, e.g. a frequency swept or chirped pulse. In some examples the interrogating radiation may be amplitude and/or phase modulated, e.g. according to some code, for instance as described in US2019/0025094.

Note that as used herein the term “optical” is not restricted to the visible spectrum and, for the avoidance of doubt, optical radiation as used herein includes infrared radiation and ultraviolet radiation. Any reference to “light” should also be construed accordingly.

The phenomenon of Rayleigh backscattering results in some fraction of the light input into the fibre being reflected back to the interrogator, where it is detected to provide an output signal which can be representative of acoustic disturbances in the vicinity of the fibre. The interrogator therefore conveniently comprises at least one laser 105 and may comprise at least one optical modulator 106 for producing interrogating radiation. The interrogator also comprises at least one photodetector 107 arranged to detect radiation which is Rayleigh backscattered from the intrinsic scattering sites within the sensing fibre 101. A Rayleigh backscatter DAS sensor is very useful, but systems based on Brillouin or Raman scattering are also known.

For a distributed fibre optic sensor, the backscatter from the sensing optical fibre 101 will depend on the distribution of the inherent scattering sites within the optical fibre, which will vary effectively randomly along the length of the fibre. Thus the backscatter intensity from any given interrogation will exhibit a random variation from one sensing portion to the next but, in the absence of any environmental stimulus, the backscatter intensity from any given sensing portion should remain the same for each repeated interrogation (provided the characteristics of the interrogating radiation, i.e. the frequency of the pulse or pulses, remains the same for each interrogation). However, an environmental stimulus acting on the relevant sensing portion of the fibre can result in an optical path length change for that section of fibre, e.g. through stretching/compression of the relevant section of fibre and/or a refractive index modulation. As the backscatter from the various scattering sites within the sensing portion of fibre will interfere to produce the resulting intensity, a change in optical path length will vary the degree of interference. The variation in distribution of the scattering sites will result in a variation in intensity of backscattered from an affected sensing portion, which can be detected and used as an indication of a disturbance acting on the fibre, such as an incident acoustic wave. Additionally or alternatively, if each interrogation comprises spatially separated pulses at different frequencies to one another, or the backscatter is mixed with a local oscillator signal at a different frequency, the change in optical path length for a sensing portion with result in a change in phase of a carrier signal at the difference frequency, which can be detected and used as an indication of the disturbance.

The signal from the photodetector may thus be processed by processing module 108 of the interrogator 102 to provide a measurement signal which is representative of disturbances acting on the sensing portions or channels of the fibre. Some processing may additionally or alternatively be done by signal processor 103. As noted, in some implementations, the processing may demodulate the returned signal based on a frequency difference between optical pulses of the interrogating radiation. The processing module 108 may, in some implementations, process the detected backscatter for example as described in any of GB2,442,745, W02012/137021 or WO2012/137022, depending on the form of the interrogating radiation. In some implementations the processing may determine a phase value from the backscattered light, e.g. the phase of a signal component at a defined carrier frequency. As described previously any changes in the effective optical path length within a given section of fibre, such as would be due to incident pressure waves causing strain on the fibre, can therefore be detected.

If the interrogation comprises a frequency swept or chirped pulse, the detected backscatter may be processed by using matched filtering or cross correlation so as to identify backscatter from different sensing portions of the sensing fibre, which may then be processed so as to determine any disturbances acting on that portion of the sensing fibre. Likewise, if the interrogation includes any coding, the received signal may be correlated with the known code, for example as described in LIS2019/0025094.

The form of the optical input and the method of detection allow a single continuous fibre to be spatially resolved into discrete longitudinal sensing portions. That is, a measurement signal indicative of disturbance at one sensing portion, i.e. indicative of an incident acoustic wave, can be provided substantially independently of a measurement signal for another sensing portion. Note that the term acoustic, as used herein, shall be taken to mean any type of pressure wave or mechanical disturbance or varying strain generated on the optical fibre and will, for instance, include seismic waves or the like. The term acoustic is intended to refer to the type of stimulus acting on the sensing fibre but is not used to imply any particular frequency limitation. As used in this specification the terms “distributed fibre optic sensing” and “distributed acoustic sensing” will be taken to mean sensing by optically interrogating an optical fibre to provide a plurality of discrete sensing portions distributed longitudinally along the fibre and the terms “distributed fibre optic sensor” and “distributed acoustic sensor” shall be interpreted accordingly.

Such a sensor may be seen as a fully distributed or intrinsic sensor, as it uses the intrinsic scattering process inherent in an optical fibre and thus distributes the sensing function throughout the whole of the optical fibre.

Such DAS sensors can be usefully employed in a range of applications to provide information about environmental disturbances acting on the sensing fibre for each of a plurality sensing portions. This can be used, for example, to detect the occurrence of events of interest and/or to allow some characterisation or analysis of the event.

Figure 2 illustrates how a single pulse of optical radiation propagates within a sensing fibre. The pulse is launched from a proximate end of the fibre or “launch end” Do. Lines 201 and 202 illustrate the leading edge and trailing edge of the pulse, respectively. Thus, at time to, an interrogator unit starts launching optical radiation into the fibre and continues to transmit continuous radiation until a time ti . The pulse duration is thus equal to ti-to. The pulse propagates within the fibre with a speed given by c/n, where c is the speed of light in a vacuum and n is the refractive index of the optical fibre. The gradient of the leading edge 201 and the trailing edge 202 therefore corresponds to c/n, and the spatial distance between the leading edge 201 and the trailing edge 202 defines the pulse width of Wi.

As the pulse propagates in the sensing fibre, any inherent scattering sites within the optical fibre may result in Rayleigh scatting and at least some radiation will be backscattered to propagate back towards the launch end of the fibre Do. The Rayleigh backscatter will propagate back to the detector at the launch end of the fibre, again with a speed given by c/n. Figure 2 illustrates a path for a backscatter signal 203, that propagates back to the launch end from a location within the sensing fibre. The backscatter signal 203 propagates with the same speed as the pulse, i.e. the backscatter signal path 203 has an equal but opposite gradient to the leading edge 201 and the trailing edge 202 of the forward propagating pulse, and such backscatter is received at a photodetector at the launch end of the fibre Do at time t2. The backscatter received at the launch end of the fibre Do at time t2 may comprise radiation which is scattered from the leading edge of the pulse at one point in time from the leading edge of the pulse 201 at a distance Du. into the sensing fibre. The backscatter received at time t2 may also comprise radiation scattered a short while later from the trailing edge of the pulse 202 from a distance DIT into the sensing fibre. In effect any backscatter from the leading edge of the pulse at the distance Du. will propagate backwards, as the pulse continues to propagate forward, until the backscatter from the leading edge meets the trailing edge. Any backscatter from the section of the fibre between Du. and DIT (which is equal to half the spatial width Wi of the pulse) which is generated at a point that coincides with the line 203 will be received back at the start Do of the sensing fibre at the same time, and thus will be coincident on the detector 107, and thus will interfere to provide the resultant signal detected by the detector.

In this example, the backscatter received at the launch end of the fibre at a time t2 after launch of an interrogating pulse will thus correspond to this section of the sensing fibre. By processing the returns from multiple different interrogations in the same time bins, the responses from the same section of fibre to repeated interrogations can be determined. Any stimulus which is acting on the fibre at this first distance Di will produce an optical path length change in this sensing channel of the fibre. The optical path length change will change the distribution of scattering sites within this portion of the fibre, which will change how the scattering from the various sites interfere, thus resulting in a change in the intensity pattern at the detector. The intensity pattern from each scattering from repeated interrogations can thus to processed to provide an indication of any stimulus acting on that part of the fibre.

Figure 3 illustrates an example of multiple interrogations of a sensing fibre, where each interrogation comprises a single continuous optical pulse. As discussed above, for a first interrogation the pulse may be launched into the launch end Do of the sensing fibre between times toA and tiA. Lines 201a and 202a illustrate the leading edge and trailing edge respectively of the first radiation pulse of a first interrogation. The first radiation pulse will propagate in the optical fibre with a speed c/n and as described above, produce a backscatter signal from various scattering sites within the sensing fibre. Eventually the pulse will reach the end of the fibre, at a distance D _x equal to the fibre length. Backscatter signal 203a which may propagate from the end of the fibre D _x will arrive at the detector at the launch end of the fibre Do at time txA. The round-trip travel time for a pulse to propagate to the end of the fibre D _x and back again therefore corresponds to txA-to.

In conventional COTDR distributed fibre sensing techniques, the period between successive interrogations (with the same optical characteristics, e.g. at the same optical frequency as one another) should be long enough to allow a round trip travel time for the interrogation to propagate to the distal end of the fibre and back again. This is to ensure that the backscatter received at the launch end of the fibre can be identified as originating from a known part of the fibre purely based on the time of arrival. If pulses from multiple interrogations were propagating within the fibre at a given time, the backscattered radiation from one interrogation from one part of the fibre could be received at the same time as backscatter from another interrogation from another part of the fibre. If the pulses were at the same optical frequency, i.e. wavelength as one another, the backscatter would also be at the same frequency.

Referring again to Figure 3, between times to and txA, backscatter radiation from the first interrogation may be propagating in the fibre. Only after time txA has expired i.e. the round-trip travel time for the first pulse to the end of the sensing fibre, does the interrogator unit introduce a second pulse into the fibre. Therefore, at time toB the interrogator unit launches a second interrogation comprising a pulse into the fibre (with a duration, until t , that matches that of the first pulse). Figure 3 shows the leading edge 201b and the trailing edge 202b of this second pulse of the second interrogation, and also a second backscatter signal 203b that may travel from the distal end of the fibre at Do and reach the photodetector at time txB.

It will be noted that figure 3 shows an example where each interrogation comprises a single continuous pulse. One skilled in the art will understand that the interrogating radiation may take other forms in other examples and may comprise more than one pulse. For instance, each interrogation could comprise a pulse pair of short pulses, spatially separated by a distance that defines the size of the sensing portion. In some examples the pulses of an individual pulse pair may be at different frequencies to one another to define a carrier frequency. Each interrogation however typically uses optical radiation with the same characteristics, e.g. where the interrogation comprises pulse pairs, the pulse pairs of each interrogation are the same. Thus, in the same way as discussed above there should only be radiation from one interrogation propagating in the sensing fibre at any time.

Likewise, as noted above in some implementations the interrogating radiation could comprise a frequency swept or frequency chirped pulse, or there may be some coding, e.g. in amplitude and/or phase, applied to the interrogating radiation. In some implementations matched filtering or correlation may be applied to the backscatter to identify the signal component from a given sensing portion of the fibre. As one skilled in the art will understand, however, the filtering or correlation relies on the backscatter signal at any given instant comprising only one signal component that matches or correlates. This again requires that after a first interrogation is launched, a second interrogation (with the same optical characteristics, i.e. the same frequency sweep or the same code) is not launched until after the round trip travel time to the end of the sensing fibre and back.

The rate at which interrogations may be introduced into the fibre is termed the ping rate P. As noted above, in conventional COTDR systems, the duration between interrogations having the same optical characteristics (e.g. pulse frequency) as one another is set to be at least as long as the round-trip travel time of optical radiation within the fibre. Where an interrogation comprises a single pulse of continuous frequency, a subsequent interrogation should not be introduced until all backscatter from the current interrogation has been received. For an interrogation with coding, the time should be sufficient to allow all backscatter from the start of the coded sequence to be received before launching another instance of the same coded sequence (but the second coded sequence may be launched whilst later parts of the first coded sequence are still propagating in the sensing fibre).

In the illustrated example of Figure 3, the ping rate P = t ₃ - to. Note that for clarity figure 3 illustrates a short gap between the time txA at which the last of any backscatter from the first interrogation would be received and the time toB at which the pulse of the second interrogation is launched. For the maximum ping rate this gap should be as short as possible and in some examples there may be no gap, i.e. toB = txA.

The ping rate P sets the sample rate of the distributed fibre optic sensing system and therefore the maximum signal frequency e.g. maximum frequency that can be resolved, i.e. measured, by the sensing system. As one skilled in the art will be familiar with, the maximum measurable signal frequency is limited by the Nyquist sampling theorem fmax < f _s /2, where f _ma x is the maximum measurable frequency and f _s is the sample rate. Purely as an example, for a fibre length of 40 km with a refractive index of n=1.5, the round-trip travel time of the pulse in the fibre is ~400ps, giving a maximum ping rate of around 2.5 kHz. The maximum signal frequency measurable by the distributed sensing system would thus be 1.25 kHz.

For a number of sensing applications, the stimuli of interest may be expected to have a relatively low maximum frequency component, and in such cases a sample rate of the order of 5kHz will be sufficient to sense the stimuli of interest acting on the fibre. However, in some applications the stimuli of interest may be of a higher frequency, for instance approaching the maximum frequency measurable by the distributed sensing system. In such examples, there is limited bandwidth available to sense the higher frequency signals. In such application therefore a higher sample rate, and hence ping rate, would be useful.

Also, as noted above, the ping rate is generally defined by the length of the sensing fibre. In some examples, it may be desirable to perform distributed sensing techniques on a relatively long optical fibre e.g. >100km. For example, an optical fibre may already be installed so that at least part of the optical fibre is deployed in an area of interest. The installation of an optical fibre into an area of interest is often the most costly and complex process in performing distributed fibre optic sensing. Therefore, it would be advantageous to perform distributed sensing on a sensing fibre already present in an area of interest e.g. a legacy telecommunications fibre buried in a seabed. However, due to the length of such a long fibre e.g. ~100km with a refractive index of n=1.5, this would lead to a sample rate of around 1 kHz and a maximum measurable signal frequency f _ma x ~ 500 Hz In such examples, this may not be acceptable as the stimuli of interest may be at a higher frequency.

It will be understood that for some relatively long sensing optical fibres, e.g. of the order of several hundreds of kilometres, there may a distance into the sensing fibre beyond which no significant backscatter may be received. That is, due to the propagation losses within the fibre as a pulse propagates in the fibre - which would also affect the backscatter propagating back to the launch end of the fibre - there may come a point in the sensing fibre, before the actual end of the fibre, beyond which no significant backscatter would be received, e.g. the intensity of any received backscatter may be below the noise floor of the distributed fibre optic sensor. In such a case the ping rate could be set based on the round-trip time to such a point of negligible backscatter return, rather than the total length of the sensing fibre. However, given that most optical fibres are designed to exhibit low propagation losses, the distance to such a point of negligible backscatter return may be quite long, requiring a relatively low ping rate.

It will also be understood that optical radiation, e.g. pulses, corresponding to different interrogations may be propagating in the fibre at the same time if the interrogations use optical radiation with different optical characteristics, for example if the different interrogations uses pulses at different optical frequencies/wavelengths. Thus a pulse of a first interrogation at a first wavelength could be propagating in the fibre at the same time as a pulse of a second interrogation at a second wavelength, and the backscatter from each pulse could be filtered by wavelength so as to distinguish the backscatter from each interrogation. Thus, following any interrogation at the first wavelength, the system would need to allow sufficient time for the round trip propagation to the end of the fibre and back before launching another interrogation at the first wavelength, but in the intervening time an interrogation at the second wavelength could be launched.

However, as mentioned above, the processing of a Rayleigh based fibre optic sensor may rely of the difference between backscatter returns from the same sensing portion in response to the interrogating radiation of the same optical characteristics. Thus, considering the example of interrogations at two different wavelengths, the system will process the returns at the first wavelength separately from the returns at the second wavelength. In effect there is a first series of interrogations, at the standard ping rate, and a second independent series of interrogations, at the same ping rate. Whilst the interrogations of the different series may be alternated and the measurement signals from the two series of interrogations may be processed into a single resultant signal, this may not provide the same bandwidth improvement as a single series of interrogations at double the ping rate. In addition, launching interrogating optical radiation with different optical characteristics, e.g. different wavelengths, can require additional components such as additional laser sources or modulators and wavelength division filters and separate processing channels, which can add to the costs and complexity of an interrogator unit

Embodiments of the present disclosure therefore relate to a distributed fibre optic sensing apparatus with a ping rate, for a series of interrogations of the fibre with optical radiation of the same optical characteristics, which is not limited by the round trip travel time of a radiation pulse to the distal end of fibre. The ping rate may thus be configured such that the interval between repeated interrogations of the series is less than a round trip travel time for interrogating radiation to complete a round trip propagation within the whole length of the optical fibre.

A distributed fibre optic sensing apparatus according to an embodiment of the present disclosure comprises an optical source configured to repeatedly interrogate an optical fibre with interrogating optical radiation comprising optical radiation having a launch frequency, the interrogating optical radiation having an optical power such that the interrogating optical radiation experiences spectral spreading when propagating in the optical fibre. The optical power of the interrogating optical radiation, i.e. the first pulse, is set sufficiently high i.e. above a threshold value, where the interrogating optical radiation, i.e. the first pulse, will experience spectral spreading as it propagates along the fibre.

As one skilled in the art will be familiar, if optical radiation is launched into an optical fibre with sufficient power, this can lead to non-linear effects within the optical fibre causing the optical radiation to undergo spectral spreading. That is, if the interrogating optical radiation comprises a single, relatively narrowband, pulse at a first frequency, and the optical power of the pulse is high enough, non-linear effects as the pulse propagates within the fibre can lead to spectral spreading, which result in the intensity of the radiation spreading across a number of frequencies. For conventional Rayleigh backscatter based distributed fibre optic sensing, such non-linear effects are undesirable and can be detrimental. Thus, conventionally, the maximum power for the interrogating radiation is limited by a threshold for the onset of non-linear effects.

In embodiments of the present disclosure, the spectral spreading of the radiation pulse is exploited to allow an increase in the interrogation ping rate for a given optical fibre length compared to conventional COTDR techniques. In particular, embodiments rely on the fact that the extent of the spectral spreading will increase with the propagation distance of the pulse into the fibre. This will mean that for a given point in the fibre, the pulse downstream of said point will experience spectral spreading to a greater extent than for a pulse upstream of said point. Thus, for a relatively narrowband pulse launched at a first optical frequency f 1 , as the pulse propagates the intensity at the launch frequency f1 will decrease due to the non-linear processes (in addition to any usual propagation losses), but there will an increase in intensity at other frequencies around f1. By applying narrowband filtering to the backscatter signal around the frequency f 1 , the other frequencies can be excluded. At a certain distance into the fibre, the effect of the spectral spreading will be such that no significant backscatter at the launch frequency f1 will be received back at the launch end of the sensing fibre, i.e. the backscatter at the frequency f1 may be below the noise floor of the system. The distance at which only negligible backscatter at the launch frequency f1 is received can be before the end of the sensing fibre and may be significantly less than the distance for negligible backscatter for pulses launched with an optical power just below the non-linear power threshold, as would be conventional. Counterintuitively, increasing the optical power may thus reduce the distance into the sensing fibre at which negligible backscatter at the launch frequency is received back at the launch end of the sensing fibre. The ping rate may therefore be set with regard to the return trip time to this distance into the sensing fibre, which may allow an increase in ping rate compared to the conventional approach.

Whilst the spectral spreading does, as described, reduce the intensity at the launch frequency (or frequencies) as the pulse propagates, for at least for the part of the sensing fibre near the launch end the effect of spectral spreading will be limited. Thus, for a part of the fibre extending from the launch end there will be sufficient energy at the launch frequency to provide a Rayleigh backscatter signal at this frequency which can be processed to provide a measurement signal, and, as mentioned, the unwanted components at other frequencies caused by spectral spreading can be removed by filtering.

Figure 4 illustrates this principle. Figure 4 illustrates an optical pulse 402 propagating along a sensing optical fibre 401. The pulse 402 may be launched into the fibre 401 at time to, with an optical power such that the radiation pulse 402 experiences spectral spreading when propagating in the optical fibre 401 . In this example the pulse is a single continuous pulse of optical radiation at a single launch frequency f 1 , that it is the pulse is launched as a narrowband pulse at a (centre) frequency f1.

The left side of figure 4 illustrates the pulse at a time ti relatively shortly after launch, at a position 403-1 within the sensing fibre which is at the beginning of the sensing fibre, i.e. relatively close to the launch end. Figure 4 also illustrates an example spectrum 404- 1 for the pulse at this point, i.e. an indication of the intensity at different frequencies. Whilst the pulse may have experienced some non-linear interaction leading to spectral spreading, the extent of the spectral spreading may be relatively limited at this point, so the spectrum 404-1 may still be that of a relatively narrowband pulse with significant intensity at the launch frequency f1 .. As the pulse propagates it continues to experience non-linear interactions and the effects of the spectral spreading accumulate. The right side of figure 4 illustrates that the pulse at a later time t2 when the pulses has propagated to a location 403-2 which is further along the sensing fibre, at this point the pulse has undergone a greater extent of spectral spreading. Figure 4 illustrates an example of a spectrum 404-2 for the pulse at this point which illustrates that the energy of the pulse has been spread into a range of wavelengths, and the intensity at the launch frequency is significantly reduced.

As noted above, this means that as the pulse propagates along the fibre, the energy of the pulse is increasingly spread into a range of different frequencies, but any unwanted backscatter due to these additional frequencies can be removed by applying narrowband filtering at the launch frequency f1 . This can mean that a certain distance into the sensing fibre, which will be referred to herein as a cut-off distance De, any backscatter received at the launch frequency f1 is effectively negligible, e.g. is below the noise floor of the sensor system or otherwise insufficient to significantly effect any backscatter at that frequency received from another interrogating pulse. Referring back to figure 1 , distributed fibre optic sensing apparatus according to embodiments of the present disclosure therefore comprises a filter 109, which in this example is a narrowband filter, where the narrowband filter comprises a passband at the launch frequency of the interrogating radiation. For interrogating radiation of the example of figure 4, where the interrogation comprises a single pulse at a frequency f 1 , the filter 109 may therefore comprise a narrowband filter with a passband centred substantially at f1.

Figure 1 illustrates that the filter 109 is an optical filter that forms part of the interrogator 102 and is arranged as part of an optical path for the backscattered radiation upstream of the photodetector. The filter 109 in this example thus does not form part of the path for the outgoing interrogating radiation. Various types of narrowband optical filter are known and could be implemented, as would be understood by one skilled in the art. It will be understood that other arrangements are possible however, for instance the filter could be external to the interrogator 102 so as to be used only when required, and/or the filter may be arranged in a part of the shared signal path for both the outgoing interrogating radiation and the received backscatter. Given the filter 109 has a passband at the launch frequency of the interrogating radiation, it will have no or only a very limited effect on the outgoing interrogating radiation.

In implementations where each interrogation comprises optical radiation which has a substantially constant narrowband optical frequency, e.g. each interrogation comprises a single continuous interrogating pulse or a coded sequence at a substantially constant frequency, there will thus effectively be just one relatively narrowband launch frequency. As mentioned the passband of the filter 109 may be relatively narrow and centred on that launch frequency. In some implementations the passband of the filter 109 may have a spectral width of the order of a few GHz or so. The spectral spreading experienced by the interrogating radiation will, however, be of a greater extent and thus the effect of the spectral spreading will be to spread the energy of the interrogating radiation to wavelengths outside the passband.

As noted above, in some implementations the launch frequency for the interrogating optical radiation may comprises different frequency components. For example, as discussed previously, each interrogation may comprise two relatively narrowband pulses at different optical frequencies to one another, e.g. f1 and f2, and thus the launch frequency may include the frequency components at f1 and f2. In other implementations, with a frequency swept pulse, the frequency of the interrogating radiation may vary, over a certain range, e.g. from a frequency f1 to a frequency f2. In any event these different launch frequency components can be within the passband of the filter 109.

For instance, where two pulses at different frequencies f1 and f2 to one another are used, the difference frequency is typically selected so as to be a relatively low frequency so as to provide benefits in processing. Thus, the frequency separation of the two pulses f1 - f2 may be of the order of 100 MHz or so, and thus both frequencies f1 and f2 will be within the passband of the filter 109. As both pulses propagate, spectral spreading will be experienced by both pulses which will spread the energy of the pulses outside of the passband. For frequency swept pulses, the range or extent of the frequency variation of the frequency sweep may also be within the passband of the filter 109, for instance of the order of 100 MHz or so. It will be noted that as a frequency swept pulses propagates in the fibre the spectral spreading will result in a spread of frequencies that will reduce the degree to which the backscatter matches or correlates to the frequency sweep, and thus in such applications at least part of the filtering may be applied in the processing.

As used herein the term launch frequency will thus be taken to mean the frequency or frequencies at which the interrogating radiation is launched, and thus the launch frequency may refer to one or more defined frequencies at which pulses are launched or a frequency range over which the interrogating radiation is launched.

As mentioned, the narrowband filter 109 is configured to pass radiation at the launch frequency of the interrogating radiation and filter out other frequencies. For backscatter received from sensing portions of the sensing fibre which are relatively near to the launch end of the fibre, there will be significant backscatter received at the launch frequency, as the effect of spectral spreading will be limited as the interrogating radiation propagates through such parts of the sensing fibre. The filtered backscatter may thus be processed to form measurement signals from such sensing portions. As the pulse propagates further into the fibre, the effect of the spectral spreading will increase, but there may still be sufficient energy at the launch frequency to provide sufficient Rayleigh backscatter at the launch frequency to allow a measurement signal to be determined. Further into the fibre the interrogating radiation may reach the cut-off distance De, beyond which no significant backscatter at the launch frequency is received. As noted above, once sufficient time has elapsed to allow for the round-trip travel time to the cut-off distance De, another interrogation in the series, e.g. comprising a pulse at the same launch frequency, can be launched, even thought the prior pulse may still be propagating in the part of the sensing fibre beyond the cut-off distance De.

Thus, in a situation where a region of interest for sensing corresponds to only part of a sensing optical fibre, and there is a distal part of the sensing fibre which is not of interest, then if the interrogating radiation can be configured such that at least part of the distal end of the sensing fibre (which is not of interest) is beyond the cut-off distance De, the ping rate for sensing applied to the region which is of interest can be improved. As noted above, conventionally the ping rate would be set with regard to the length of the whole sensing fibre. However, with interrogating radiation having a cut-off distance which is before the end of the sensing fibre, a faster ping rate can be used.

As noted above the cut-off distance De may correspond to the distance at which the backscattered radiation at the launch frequency has fallen below the noise floor of a detector configured to receive backscatter radiation from the sensing fibre.

In general, the cut-off distance De may depend on the optical fibre used as the sensing fibre and may be controlled by controlling the characteristics of the interrogating radiation. The cut-off distance for the particular sensing fibre for interrogating radiation having a given set of characteristics may be determined through testing/experimentation, for example in a calibration process and/or an estimated cut-off distance could be used based on the type of sensing fibre and interrogating radiation.

In one example, a ping rate for a given region may be determined and the launch power of the interrogating radiation adjusted, e.g. by controlling a variable attenuation applied to the interrogating radiation before launch into the sensing fibre, so as to optimise SNR.

In one example the launch power of the optical radiation may initially be set to a first level, which may be below the non-linear threshold for the sensing fibre, and the backscatter level, i.e. the optical return, from a certain distance into the fibre, e.g. at the end of the region of interest may be observed . The launch power may then be increased, e.g. by decreasing the launch attenuation applied, until a reduction in intensity is observed at that distance. In some cases the launch power may be adjusted until the noise level at that distance is a minimum. In some embodiments the optimisation of the launch power may also take into account the noise at the beginning oft the region of interest. By observing the acoustic noise floor (and so the level of residual signal) at the front end of the fibre and adjusting the launch power to minimise this whilst still maintaining the noise floor at the end of the fibre an optimal configuration is achieved. It is an optimisation between launching more light to improve the front end of the fibre whilst not launching too much so as to significantly degrade the noise at the end of the region of interest.

In some embodiments, the cut-off distance De may be configurable in use. That is, the interrogator 102 may be configurable such that the characteristics of the interrogating radiation, e.g. optical pulse, may be altered, for example by controlling the optical power, to vary the cut-off distance De. The characteristics of the pulse may be varied by controlling the laser 105 and/or optical modulator 106, or via some other optical component to controllably adjust the cut-off distance De.

The ping rate of the distributed fibre optic sensing apparatus according to some embodiments of the present disclosure may therefore be determined by the cut-off distance De into the fibre, beyond which no significant backscatter at the launch frequency is received, e.g. at which the backscatter radiation at the central launch frequency has fallen below the noise floor of the detector. In some embodiments, the part of the sensing fibre extending from the launch end Do of the fibre to the cut-off distance De may correspond to the region of interest. That is, the region of interest may comprise the whole of a first part of the fibre, from the launch end to the cut-off distance. In practice, however, as the interrogating radiation approaches the cut-off distance, the amount of backscatter at the launch frequency will be relatively low, and near the noise floor of the system. It may therefore be desirable to set the cut-off distance a bit further into the sensing fibre than the end of a region of interest from which good quality measurement signals are desired. In this case the ping rate may therefore be limited by the round-trip travel time for a pulse to propagate to the cut-off distance De and back again to the detector at the launch end Do.

In one embodiment, the maximum ping rate is determined by P < c / 2nDc, where c is the speed of light in a vacuum, n is the refractive index of the optical fibre and De is the cut-off distance at which the backscattered radiation at the launch frequency has fallen below the noise floor of the detector. Thus, for example, for a fibre with a total length of 100km, but where the interrogating radiation is launched so that the cut-off distance De is, say, 40km, the ping rate could be set to be around 2.5 kHz (assuming n is about 1 .5). This is a much faster rate compared to the conventional approach, where the ping rate, based on the whole fibre length, would be around 1 kHz.

Figure 5 illustrates an example of radiation pulses propagating along an optical fibre launched using a distributed fibre optic sensing apparatus according to an embodiment. As described above, a first pulse of a first interrogation may be launched into an optical fibre between a time toA and time tiA, respectively, and the pulse propagates with a leading edge 101a and a trailing edge 101b. In this example the first pulse is launched into the optical fibre with an optical power such that the radiation pulse experiences spectral spreading when propagating in the optical fibre.

The ping rate P may be determined by the cut-off distance De at which the backscattered radiation at the central frequency has fallen below the noise floor of the detector. As such, the ping rate P is determined by the round-trip travel time for a radiation pulse to travel distance De into the fibre and back to the detector.

Therefore, referring to Figure 5, the first pulse travels along the optical fibre and radiation that may be backscattered from a scattering site at distance De into the fibre will arrive back at the detector at time tcA. The maximum ping rate is therefore set by P = tcA - to. The distance from the launch end of the fibre Do to the cut-off distance De may correspond to a first section of the optical fibre 503a. Measurement signals for one or more sensing portions within this first section may be determined and output from the interrogator. At time toB, a pulse of a second interrogation (at the same launch frequency as the first pulse) is launched into the optical fibre, with a pulse duration (the same as the first pulse) up to ti B- In this example the launch time toB of the second pulse corresponds to the time tcA at which any backscatter from the cut-off distance De would be received, but as discussed above there could, in some examples, be a short gap between tcA and Again the second pulse propagates with a leading edge 101b and a trailing edge 102b. The second pulse is also be launched into the optical fibre with an optical power such that the radiation pulse experiences spectral spreading when propagating in the optical fibre.

It can be seen that immediately after launch of the second pulse, at the time toB, there are two pluses propagating in the fibre. The first pulse, previously launched into the fibre, is propagating in a second section 503b of the sensing fibre, which is downstream of the cut-off distance De. The second pulse is propagating in the first section 503a, upstream of the cut-off distance De. The second section of the fibre 503b corresponds to the distal part of the sensing fibre from the cut-off distance De to the distal end Dx of the optical fibre. Backscatter radiation originating from the second section 503b, will not significantly impact on the processing of the backscatter from the first section 503a. Backscatter radiation from the second section 503b will either be filtered from the backscatter signal by the narrowband filter 109 or backscatter radiation at the central launch frequency will have fallen below the noise floor of the detector.

At time tcB, radiation from the second pulse that may be backscattered from a scattering site at the cut-off distance De will be incident upon the detector and (although not shown in figure 5) at this point a further interrogating pulses at the same launch frequency could be launched. In this embodiment there may be optical interrogating radiation from multiple interrogations propagating in the sensing fibre at the same time, but only a single pulse is propagating within a region of interest, which in this example is the first section 503a, at any one time.

Embodiments thus far have illustrated that distributed sensing apparatus according to the present disclosure can provide sensing data over a region of interest proximate the launch end of the sensing fibre. However, in some instances, there may be a proximal part of the sensing fibre which may not be of interest for sensing, i.e. the region of interest for sensing may start a certain distance into the sensing fibre. In such a case, it may be possible to allow additional improvement in the ping rate for a given cut-off distance or alternatively achieve a given ping rate with a longer cut-off distance, which may improve signal returns from the region of interest.

Figure 6 illustrates another example of radiation pulses propagating along an optical fibre, launched using a distributed fibre optic sensing apparatus according to an embodiment. Figure 6 illustrates two interrogations, a first pulse launched at a time toA and having a leading edge 101a and a trailing edge 102a, and a second pulse (of the same launch frequency), launched at a time toB and having a leading edge 101 b and a trailing edge 102b. Again, the interrogations are launched with an optical power that leads to spectral spreading such that there is a cut-off distance De beyond which no significant backscatter at the launch frequency is expected. Thus, as discussed with respect to figure 5, the sensing fibre can be seen as being divided into two portions, a first section 603a extending from the launch end Do of the fibre and the cut-off distance De, and a second section 603b extending from the cut-off distance De to the distal end of the fibre D _x . Figure 6 illustrates that backscatter 103a from the first pulse from the cut-off distance De may be received at a time tcA. As discussed above, the ping rate could therefore be set with regard to the round-trip travel time to this cut-off distance, i.e. the second interrogation could be launched only on or after the time tcA.

In this example however, the sensing region of interest does not include the initial part of the sensing fibre. In this example the sensing region of interest starts at a distance d1 into the sensing fibre and extends to a distance d2 into the fibre. That is, there a first sub-section 604a, between the launch end of the fibre at Do and the beginning of the region of interest at distance d1. The region of interest between d1 and d2 is thus a second sub-section 604b.

Following launch of the first radiation pulse of the first interrogation at the time toA, this pulse may initially propagate along the first sub-section 604a from the launch end Do of the fibre, to the start of the region of interest at first distance di. As noted, in the example of Figure 6, the first sub-section 604a may not itself be part of a region of interest, and thus measurement signals from this region may not be of interest. The first radiation pulse will then enter the second sub-section 604b, which, as noted comprises a region of interest between first distance di and second distance d2. Any radiation which is backscattered from first distance di will propagate as a first backscatter signal 601a and will be incident on the detector at the launch end Do at time t2A. Any radiation which is backscattered from second distance d2 will propagate as a second backscatter signal 602a and will be incident on the detector at the launch end Do at time t3A.

The interval between t2A and tsA therefore corresponds to the time interval at which backscatter radiation may be received from the second sub-section 604b i.e. the region of interest. During this time interval, the backscatter may be processed in various time bins to provide a measurement signal for one or more sensing portions within this region of interest, and thus no significant radiation at the launch frequency should be received from any other interrogation. Therefore, a pulse of the second interrogation should not be launched into the fibre before the time tsA, as the backscatter radiation could not be spatially discriminated as being from one interrogation or the other. In one embodiment the ping rate may therefore be determined based on the condition that only a single pulse should be propagating in a part of the fibre up to the end of the region of interest at a given time, i.e. there should only be a single pulse within the combined first and second sub-sections 604a and 604b.

In the example discussed above with reference figure 5, the second interrogation is only launched after a round trip time to the cut-off distance De, so that the second interrogation is only launched at a time tcA when there is no significant backscatter from the first interrogation at the launch frequency. In the example of figure 6 however, where an initial section 604a of the sensing fibre is not part of the region of interest, the second interrogation may be launched before this point in time, whilst backscatter from before the cut-off distance De is still being received from the first interrogation.

Figure 6 illustrates that the second pulse of the second interrogation is launched into the fibre at time toB, which in this example coincides with the time tsA (but in general it may be any time at or after the time tsA when the last of the backscatter from the first interrogation from the region of interest is received). As the second pulse propagates over the first sub-section 604a, the pulse will encounter scattering sites and backscatter radiation will be scattered back the detector. Whilst the second pulse is propagating in the first sub-section 604a, backscatter may be received from the first pulse from part of the fibre which is beyond the distance d2 of the end of the region 604b of interest, but which is before the cut-off distance De. Thus, the initial part 603a of the sensing fibre may include a third sub-portion 604c which extends from the distance d2 at the end of the region of interest to the cut-off distance De.

As noted previously, it may be advantageous for the cut-off distance De to be some distance after the end of the region of interest, so that the spectrum of the interrogating radiation still has sufficient power at the launch frequency at the end of the region of interest to provide a detectable backscatter signal with good SNR.

In a period between the time toB (of launch of the second interrogation) and the time tcA (when backscatter from the first interrogation comes from the cut-off distance), backscatter radiation received from the second pulse originating from the first subsection 604a may be coincident at the detector with backscatter radiation originating from the first pulse at a scattering site within the third sub-portion 604c. For example, the backscatter signal 603 may be received at the detector, which will comprise at least some significant backscatter at the launch frequency from the first pulse as well as the significant backscatter from the second pulse. The backscatter radiation at the launch frequency from both the first pulse and the second pulse will be passed by the narrowband filter. The detector will therefore not be able to spatially discriminate the backscatter as originating from the first pulse or the second pulse.

However, in the illustrated example of Figure 6, neither the first sub-portion 604a nor the third sub-portion 604c are part of the region of interest and thus the inability to discriminate is not a problem. Therefore, any interference that occurs between the backscatter signals that originate from the first sub-portion 604a (from the second interrogation) and the third portion 604c (from the first interrogation) is not of concern for the sensing operation in the embodiment of Figure 6. Provided that, by the time t2B that backscatter from the region of interest 604b from the second pulse starts to be received, there is no significant backscatter at the launch frequency from the first pulse, the backscatter within the time window t2A to t3A can be processed as originating solely from the second pulse from the region of interest.

The second interrogation is thus launched such that backscatter received from the region of interest i.e. second sub-portion 604b, from the second interrogation is only received at a time when any backscatter from the first interrogation is received from the cut-off distance or beyond. Figure 6 thus illustrates that the time tcA a backscatter signal 103a may be received at the detector from a scattering site at the cut-off distance De and that only afterwards, at the time t2B does backscatter from the second interrogation from the region of interest start be received. Note that figure 6 illustrates that the time t2B at which backscatter from the second interrogation starts to be received from the region of interest occurs a short time after the time tcA where backscatter from the first interrogation comes from the cut-off distance, but in some implementations these times could be simultaneous, i.e. t2B could be equal to tcA.

During the time period from t2B to t3B backscatter from the second interrogation from the region of interest is received. Any backscatter from the first interrogation during this period is from beyond the cut-off distance De and thus comprises components at other frequencies that are filtered by the narrowband filer. The amount of backscatter at the launch frequency from the first interrogation is negligible, e.g. below the noise floor of the system. During this period from t2B to t3B backscatter the backscatter can thus be processed in time bins as originating from the second interrogation only.

In the example of figure 6 the ping rate is again controlled such that, for a region of interest, any significant backscatter received at the launch end of the sensing fibre at a given launch frequency arises from a single interrogation only.

As noted above, if the region of interest starts at or very near the launch end of the fibre, i.e. at the distance Do, then the maximum ping rate would be determined based on the round trip travel time for a pulse to propagate to the cut-off distance De and back again to the detector i.e. P’ = tcA - to., in a similar fashion as described with respect to figure 5, in where the maximum ping rate would be governed by P < c / 2nDc. However, if the initial part of the sensing fibre is not part of the region of interest, i.e. there is first sub-section 604a from which measurement signals are not required, a subsequent interrogation (at the same launch frequency as the first) can be launched whilst there is still some significant backscatter from the first interrogation at the launch frequency. This can allow a faster ping rate for a given cut-off distance, or alternatively allow the cut-off distance to be further after the end of the region of interest, which may improve SNR from the region of interest.

The ping rate P may be configured such that a single interrogation is generating detectable backscatter radiation during propagation over the region of interest. Referring to Figure 6, the ping rate P can therefore set such that the interval between repeated interrogations is equal to the round trip travel time to the cut-off distance De, i.e. the period tcA - toA less an amount which is, at most, equal to the round trip travel time to the start of the region of interest, e.g. t2B - However, the interval between successive interrogations should not be lower than the round-trip travel time to the end of the region of interest, i.e. to d2 and back, the time period t3A - toA.

With these conditions, the pulse of the second interrogation can be launched into the fibre at time toB, as described above. The ping rate P is therefore configured as P = t()B — toA.

As illustrated, this allows a ping rate P that would be faster than the ping rate P’ based purely on a round trip travel time to the cut-off distance De. This therefore leads to an increased sample rate and thus a higher maximum measurable signal frequency. In one embodiment, the maximum ping rate may be given by P < c / 2n(D-d), where c, n and D have the same definition as described above and d is the distance from the launch end of the fibre to the start of the region of interest, provided that the ping rate is not greater than c I 2n(d2), where d2 is the distance from the launch end of the fibre to the end of the region of interest.

It will be appreciated from figure 6 that a ping rate equal to P could be achieved by setting the cut-off distance De to be equal to the distance d2 to the end of the region of interest. As discussed, however, it may be beneficial for the cut-off distance De to be further into the sensing fibre than the end of a region of interest from which measurement signals are to be obtained, so as to ensure that there is sufficient backscatter at the launch frequency from across the whole extent of the region of interest.

In some implementations a region of interest may be predetermined and may, for instance, depend on the deployment of the optical fibre used as the sensing fibre. For instance, it may be wished to perform distributed fibre optic sensing, e.g. DAS, in a first location and an existing fibre could run partly through the location. Purely by way of example the location could include part of some infrastructure, such a railway track, road, bridge, tunnel, dam etc. or some other linear structure or assets such a pipeline or power cable. Additionally, or alternatively the location could be in the vicinity of some geological formation of interest, e.g. an oil or gas reservoir, water reservoir, sequestration reservoir etc. An optical fibre may be deployed that runs partly through the location it is wished to monitor but also partly in some other locations. Purely as an example a long optical fibre, say of the order of 100km or more may be deployed such that a region of interest corresponds to a part of the fibre that extends from say 10km to 30km into the sensing fibre. As noted above, if the ping rate were set for the round-trip time for the whole length of the sensing fibre, thus would limit the ping rate and hence the maximum signal frequency that could be reliably detected. In an embodiment of the present disclosure the region of interest could be set to be the region of 10-30km into the fibre. The interrogator may be configured so that the cut-off distance De is greater than 30km, by an amount that provides good SNR over the region of interest, but which is significantly shorter than the 100km length, e.g. of the order of 35km or 40km or so.

In some implementations the region of interest may be configurable in use. For instance, sensing may be performed in a first mode over a first length of the sensing fibre. In some instances, the whole of the sensing fibre may be monitored in the first mode, in which case, in some examples the first mode may involve launching interrogations at an optical power below a non-linear threshold. If a signal of interest is detected at one location of the sensing fibre, the system may operate in a second mode with a region of interest based around the location at which the signal(s) of interest are detected. In the second mode the interrogator could operate with interrogating radiation with an optical power above the non-linear threshold so as to provide a cut-off distance De which may be shorter than any cut-off distance in the first mode so as to allow for a faster ping rate in the second mode compared to the first mode. Embodiments of the present disclosure therefore provide a distributed sensing apparatus with a ping rate, which is not limited by the round-trip travel time of a radiation pulse within the optical fibre. This allows an increased ping rate compared to the conventional approach, which thus allows for a faster sample rate and hence a greater maximum measurable frequency for stimuli acting on the fibre. This can improve the sensitivity of the sensor system for higher frequency signals.

Distributed sensing apparatus according to an embodiment provides a more versatile sensing apparatus. As the ping rate is limited by the round trip travel time of a radiation pulse within an optical fibre for conventional COTDR sensing apparatus, this means that for optical fibres beyond a threshold length e.g. >100km, the sample rate may be too low to provide sensing data for a given stimuli of interest. However, using a distributed sensing apparatus according to the present disclosure, an optical fibre of any length may be interrogated to provide sensing data, because the ping rate is not limited by the length of the fibre.

Providing a more versatile distributed sensing apparatus may provide cost saving measures in certain applications. A substantial cost associated optical fibre distributed sensing is generally the installation of a suitable optical fibre into an area of interest. However, in some applications, e.g. sub-sea cables, an optical fibre may already be installed in an area of interest, but said fibre may be very long e.g. >100km. With conventional COTDR methods such cables would not be suitable for sensing. However, using distributed sensing apparatus according to an embodiment, said cables would be suitable for sensing.

Distributed sensing apparatus according to an embodiment may also provide improved sensing over a particular region of interest. For example, the portion of the fibre proximate the launch end of the fibre may not be of interest and a region downstream of this proximate portion may be the region of interest. This therefore results in a shorter region of interest and therefore an increased ping rate.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. The word “comprising” does not exclude the presence of elements or steps other than those listed in a claim, “a” or “an” does not exclude a plurality, and a single processor or other unit may fulfil the functions of several units recited in the claims. Any reference signs in the claims shall not be construed so as to limit their scope.