DEVICE FOR MEASURING SIGNALS PROVIDED BY A NETWORK OF FIBER OPTIC LASER RESONATOR SENSORS

Description

technical field

[0001] The invention relates to the general field of fiber optic laser cavities (FOLCs). It falls within the domain of sensor networks based on fiber optic laser cavities and concerns the associated interrogation architectures. In general, this invention is of particular interest for systems requiring a very large number of sensors.

[0002] It is of particular interest to fiber optic laser cavity-based sensors used for the realization of SONAR acoustic antennas in optical technology. Previous technique

[0003] Optical sensors based on fiber optic laser cavities, or CLFOs, have been studied for about ten years. CLFOs are fiber optic lasers with a DFB or DBR structure (Distributed Feedback or Distributed Bragg Reflector, according to the Anglo-Saxon terminology). These lasers can be used to create highly sensitive sensors by converting the quantity to be measured into an elongation. The variation in the length of the CLFO cavity, induced by the variation in the measured quantity, results in a variation in the emission frequency of the laser wave produced by the CLFO.

[0004] Typically, in a detection chain, each CLFO is associated with a detection channel allowing the extraction of variations in its optical frequency over time.

[0005] There is currently a need to create measurement systems comprising a large number of optical sensors, generally CLFOs, forming a network.

[0006] The implementation of a network comprising a large number of sensors is currently based on the parallel use of several elementary systems. Each elementary system, constituting a detection channel, has a finite number of sensors. Therefore, paralleling N elementary systems implies considering and processing N times the number of measurement channels of each elementary system.

[0007] In these elementary systems, CLFOs are optically pumped (at 980 nm or 1480 nm) in series on the same optical fiber and multiplexed by wavelength to form an elementary system. However, the maximum number of CLFOs connected to a single elementary system generally does not exceed a few dozen (up to approximately 40 in the 1525 nm–1570 nm CLFO emission band, for example). This limitation on the number of CLFOs stems primarily from their intrinsic properties (channel spacing, sensitivity to back-reflection, gain profile width, etc.).

[0008] To increase the performance of the sensor network, it is partly necessary to increase the number of CLFOs in an acoustic detection antenna system. The naturally used method is to multiply the number of detection channels by duplicating the elementary architecture as many times as necessary.

[0009] On the other hand, processing the data from this detection chain with a very large number of detection channels becomes increasingly difficult. To utilize all these detection channels, the natural method used for the measurement chain of this device is to process the data within N times the number of detection channels of each elementary system, using N elementary measurement channels.

[0010] However, this method significantly reduces the system's compactness. Furthermore, this measurement chain solution, with its N independent measurement channels, considerably increases the system's power consumption and the cost of the interrogation chain. From a system perspective, it is therefore unsatisfactory. Consequently, for SONAR applications, where there is a real need for a compact interrogation system and low power consumption, such a solution is not feasible.

[0011] Another known solution is to combine the different measurement channels of each elementary system into a measurement chain with a single measurement channel, by performing time multiplexing whose operation is based on the implementation of an interferometric method which allows the optical frequency of each CLFO forming a given system to be measured without having to multiply the measurement channels.

[0012] Devices capable of implementing this method have a functional structure primarily based on the combined use of an optical switch and a phase-modulated interferometer. Their structure also includes an optical receiver comprising a demodulator (33) and a photodiode, which converts the optical signal delivered by the interferometer into an electrical signal, the intensity of which is given by the following general expression: i = i 0 1 + V cos ϕ c t + Δϕ t where i0 represents an average intensity, and where V represents the visibility of the interference fringes. The quantity V is a dimensionless quantity defined by a ratio of the optical amplitudes of the two arms of the interferometer. ϕc(t) represents the modulating carrier wave, while Δϕ(t) represents the signal to be measured. The frequency variation of the CLFO is indeed contained within the phase variation of this function.

[0013] The implementation of such a measurement chain can be envisaged using interferometers employing external phase modulation measurement methods based on the principles known by the Anglo-Saxon names "Phase Generated Carrier" (PGC) or "Phase-Stepping", the basic diagrams of which are presented in figures 1 et 2 .

[0014] The so-called "Phase Generated Carrier" or PGC ( figure 1 These methods rely on sinusoidal phase modulation on one of the two output fibers (defining the interferometer arms) 11 of an optical coupler 12 via a piezoelectric phase modulator (PZT) 13, and on demodulation of the signal using a demodulation device 14 comprising a photodiode at its input. The current i at the photodiode output is thus used to determine the phase ΔΦ carrying the desired measurement information.

[0015] The electrical signal output from the photodiode (of the photoelectric detector) takes the following form: i = i 0 1 + V cos ϕ c sin ω c t + Δϕ where ϕ c is the amplitude of the phase modulation and ω c = 2πf c is the angular frequency of the phase modulation.

[0016] This signal can be developed in the following form: i = i 0 1 + V cos ϕ c sin ω c t cos Δϕ − sin ϕ c sin ω c t sin Δϕ so that by decomposing the previous expression into Bessel functions, we find that the spectrum of the modulated signal (i.e. the electric current at the output of the photodiode) consists of a fundamental frequency at ω c (carrier) as well as its harmonics on which the signal Δϕ appears in the modulation sidebands.

[0017] PGC methods use this characteristic to extract, by known processing methods not described here, the components of Δϕ in phase and in quadrature by detecting on an even harmonic and on an odd harmonic, and thus obtain the signal Δϕ.

[0018] The so-called "Phase-Stepping" methods ( figure 2 The methods rely on step-by-step phase modulation (n steps) on one of the outputs (interferometer arm) 21 of an optical coupler 22 via an electro-optical modulator 23, and on step-by-step demodulation 24 of the signal. The currents (i1, ... in), measured for each step, are thus used to determine the phase ΔΦ, which carries the desired measurement information.

[0019] The electrical signal at the output of the photodiode is written i n = i 0 1 + V cos ϕ n + Δϕ with ϕ n = n π 2 and with n representing the number of steps.

[0020] Consequently, considering N steps, and assuming that the phase shift to be measured Δϕ does not vary during a cycle of N steps, we obtain a system of N equations with unknowns I₀, V, and Δϕ, from which Δϕ can be determined. Clearly, the number N is chosen here to be greater than 3, the number of equations, in other words, steps. However, a number N equal to 4 is preferred to ensure demodulation that is robust to phase errors generated, for example, by a phase bias induced by the modulator.

[0021] The use of devices implementing these two known methods, based on a principle of phase modulation, induces a number of constraints.

[0022] To multiplex a given number of detection channels on the same measurement chain, in order to limit the number of measurement channels, it should be noted that these implementations are very rarely considered or described in the specialized literature relating to the technical field considered here.

[0023] Furthermore, the association of a phase-modulated interferometer implementing one or the other of the methods described above, with an optical switch, induces limitations in the processed signals, in terms of bandwidth and dynamics in particular, which make the performance obtained by such devices insufficient for SONAR applications.

[0024] Indeed, as is known, the bandwidth B occupied by a modulated signal, expressed as Δϕ = ϕ s sin(2πf st) , where ϕ s represents the instantaneous amplitude of the signal to be measured and fs represents the instantaneous frequency of this signal, is defined by Carson's rule and is given by the following relation: B = 2 f s 1 + ϕ s

[0025] This relationship allows us to define the minimum bandwidth B required for the demodulation device, based on the system requirements. This bandwidth is an important characteristic of the useful signal. In particular, its width must not exceed the bandwidth of the demodulation system to prevent distortion of the useful signal during demodulation.

[0026] This bandwidth must then be compared to the available bandwidth (defined by the demodulation technique used), which depends on the waveform and the carrier frequency. Therefore, if one wants to be able to demodulate a large possible signal dynamic range, it is necessary to extend this demodulation bandwidth as much as possible.

[0027] This limitation imposed by Carson's rule constitutes one of the major drawbacks of the PGC method, which requires the use of twice as much bandwidth to demodulate the same signal dynamics as the "Phase-Stepping" method.

[0028] Moreover, the use of an optical switch with switching frequency f SW does indeed result, in a known way, in the frequency domain, in a periodization of the spectrum of the optical signal carried by each of the switched channels.

[0029] Consequently, in order not to restrict the bandwidth, it is necessary that there be no overlap between the different spectra. This implies that the carrier frequency fp and the switch frequency f SW satisfy the condition f SW > 2f p .

[0030] Finally, in order to respect Shannon's theorem, since the signal is phase modulated, it is necessary that f SW = 4f p .

[0031] In this case of using a phase-stepping interferometer, the use of an optical switch implies taking samples for the N channels at each phase step, resulting in a significant decrease in the carrier frequency. The constraints imposed by the use of an optical switch are then added to the constraints imposed by Carson's rule to limit the dynamic range and bandwidth of devices based on these methods, as well as the dynamic range of the resulting signal.

[0032] EP 2 833 102 discloses a high-resolution thermopile infrared sensor array comprising several parallel signal processing channels for signals from a sensor array, and a digital port for serial output of signals.

[0033] AD Kersey et al., “Fiber grating sensors”, Journal of Lightwave Technology, vol. 15, no, 8, August 1, 1997, pages 1442-1463, USA discloses a review of Bragg grating sensor developments and in particular mentions the use of a frequency-translational interferometer (e.g., the Michelson interferometer).

[0034] WO 2019046881 A1 discloses a fiber optic sensor arrangement that includes a plurality of fiber optic-based sensor elements, the sensor elements being configured to modify an associated optical carrier signal based on changes in a quantity detected at a location of the sensor element and a phase modulation arrangement to modulate the phase of each optical carrier signal in accordance with respective uncorrelated pseudo-random binary sequence signals.

[0035] EP 0 895 072 A2 discloses a high-resolution multiplexed measurement device for frequency variations in fiber optic laser multimode acoustic sensors.

[0036] WO2011 / 039501 A2 discloses a distributed acoustic detection method in which the derivative or rate of change of a signal backscattered by a fiber is measured. The phase change or derivative measured according to this method has a much smaller amplitude than the signal itself if the difference between the times at which the signal is measured is less than the period of the signal being measured, thus resulting in lower sensitivity. Description of the invention

[0037] One aim of the invention is to propose, in the context of sonar applications in particular, a means of using a large number of CLFOs while limiting the number of measurement channels in the measurement chain used to process the signal from the detection chain, while maintaining system performance compatible with the targeted SONAR applications.

[0038] To this end the invention relates to a sonar signal detection device comprising N detection channels each having M optical fiber laser cavities, said detection channels being configured to detect and convert acoustic signals into optical signals, as well as a unique measurement chain, combining the N detection channels by time multiplexing.

[0039] According to the invention, the device is characterized by a measurement chain which combines an optical switch and interferometric measurement means configured to measure, for each of the N detection channels, the optical frequency of the signal delivered by each of the M optical fiber laser cavities.

[0040] The said measurement chain comprises: an optical switch Nx1 allowing selection, at each interferometric measurement, of the optical fiber containing the M wavelength-multiplexed optical fiber laser cavities forming a single detection channel; a frequency-translation interferometer configured to transform the frequency-modulated optical signal delivered by the M optical fiber laser cavities forming a single detection channel into an intensity-modulated optical signal; an optical demultiplexer allowing separation of the M optical signals forming said intensity-modulated optical signal and delivering each of the M optical signals on a separate output allowing independent photoelectric conversions; an optoelectronic conversion stage configured to generate, from the signal resulting from the beat of the two waves of different frequencies generated in the interferometer, an electrical signal, analog and then digital;a time demultiplexer allowing the reconstruction of the different time signals from the different detection channels; a phase demodulator allowing the processing of the signals reconstructed by the time demultiplexer from the optical signals delivered by each of the fiber optic laser cavities.

[0041] According to various provisions, each of which may be considered separately or in combination with one or more other provisions, the device according to the invention may have different technical characteristics attached to its constituent elements. Thus, According to a first arrangement, the frequency-translation interferometer includes means for frequency-translating, relative to each other, the optical signals passing through each of the arms of the interferometer; According to another arrangement, the frequency-translation means are placed in the longer arm of the interferometer; According to another arrangement, the frequency-translation means are placed in the shorter arm of the interferometer; According to another arrangement, the frequency-translation means are composed of two frequency translators performing translations of distinct frequencies, each placed in one of the arms of the interferometer; According to another arrangement, the frequency-translation means are acousto-optic modulators.

[0042] According to another provision, the means of frequency translation are optical elements integrated into an optical fiber; According to another arrangement, the frequency-translational interferometer is a Michelson interferometer; According to another arrangement, the frequency-translational interferometer includes Faraday mirrors configured and arranged to make the system independent of variations in polarization of the different signals by the optical fiber laser cavities; According to another arrangement, each of the arms of the interferometer is formed by an optical fiber of a different size, the optical fiber forming the longer arm having a lower temperature sensitivity than the fiber of the shorter arm.

[0043] According to another arrangement, the time demultiplexer allows, or is configured to, generate a control signal for the Nx1 optical switch.

[0044] According to another arrangement, the phase demodulator enables, or is configured to enable, the generation of a frequency translation control signal for the interferometer.

[0045] Advantageously, the frequency translation command allows a carrier frequency value fp to be defined.

[0046] The features and advantages of the invention will be better appreciated from the following description, which is based on the attached figures which show: [ Fig 1 ] There figure 1 As previously discussed, a schematic illustration of a first known principle ("Phase Generated Carrier") for acquiring measurements using CLFOs, allowing the number of measurement channels to be limited; Fig 2 ] There figure 2 As previously discussed, a schematic illustration of a second known principle ("phase stepping") for acquiring measurements using CLFOs, which allows limiting the number of measurement channels; Fig 3 ] There figure 3 , a schematic representation of the principle of the device according to the invention; [ Fig 4 ] There figure 4 , a detailed schematic representation showing the different stages of the device according to the invention; [ Fig 5 ] There figure 5 , a schematic timing diagram illustrating the time constraints which condition the performance required of the different elements constituting the device according to the invention.

[0047] It should be noted that, in the attached figures, the same functional or structural element preferably bears the same reference symbol.

[0048] The general structure of a detection device comprising the measurement chain according to the invention is illustrated by the schematic diagram of the figure 3 The receiving device 30 according to the invention mainly comprises: an optical switch Nx1, 31, with N inputs and one output, whose inputs are connected to the N detection channels forming the sensor network considered, each detection channel comprising a laser pumping device 37 associated with a wavelength division multiplexing device 36 ("Wavelength Division Multiplexing" or WDM according to the Anglo-Saxon designation) connected to an optical fiber 34 having a number of fiber optic laser cavities 35, or CFLOs, each tuned to one of the wavelengths carried by the fiber 34; an optical processing module 32, comprising a frequency translation interferometer, whose input is connected, by optical fiber, to the output of the switch 31 and which delivers M output signals, each signal resulting from the wave produced by one of the M CFLOs which constitute a detection channel;an optoelectronic module 33 responsible for converting the light waves delivered by the optical processing module 32 into electrical signals which are then phase-demodulated. ;

[0049] The device according to the invention thus integrates an optical switch Nx1, 31, with N input channels and one output channel, allowing selection of the light wave carried by one of the N optical fibers 34 forming the N detection channels of a sensor network, as well as a frequency translation interferometer.

[0050] Advantageously, the association of a switch 31 allowing the number of measurement channels required to be reduced to a single measurement channel to process the N detection channels of the sensor network and a frequency translation interferometer allowing the dynamics and bandwidth necessary to process, without alteration, the signals carried by the light wave conveyed by each optical fiber, makes it possible to create sensor networks 35 comprising a large number of CLFO optical fiber laser cavities and exhibiting technical characteristics and features in terms of compactness, cost and energy consumption, allowing them to be used in applications which do not generally allow their use; SONAR applications in particular.

[0051] The operating principle of the device according to the invention is described in the following text, with reference to the illustration of the figure 4 which details the constitution of the optical processing module 32 and that of the optoelectronic module, or demodulation module, 33.

[0052] As illustrated by the figure 4 , the optical processing module 32 of the device according to the invention includes a frequency translation interferometer 41, consisting of an interferometer associated with frequency translation elements introducing a frequency shift between the optical waves propagating in the two arms of the interferometer, so as to produce a frequency beat at the level of the signal transmitted to the receiving stage 43, or analog stage, of the demodulation module 33 (a photodiode for example).

[0053] From a functional point of view, the frequency translation interferometer 41 operates as described below.

[0054] The incident optical wave E 0 of frequency v 0 is distributed to the two arms of the interferometer by means of a 50 / 50 coupler.

[0055] The part of the signal E 0 which passes through the first arm of the interferometer, passes through a first frequency translator module, which induces a first shift f 1 of the signal frequency, then undergoes a delay τ before passing through the frequency translator module again after reflection on the end of the channel and undergoing a second shift f 1.

[0056] The delay τ can, for example, be induced by passing through a coil of optical fiber of length L and a second pass, after reflection at the end of the interferometer arm, through said coil. It is then expressed as: τ t = 2 nL / c where n represents the refractive index of the optical fiber used, of length L, and c the speed of light in a vacuum.

[0057] Consequently, the optical input signal E0 has the following general expression, as a function of time t: E 0 ¯ = E 0 e 2 iπν 0 t The signal E1 thus formed has the following expression: E 1 ¯ = E 1 e i 2 π ν 0 + 2 f 1 t + 2 π ν 0 + f 1 τ where v 0 + 2f 1 is the shifted frequency of the optical wave E 1 , the frequency shift f 1 being induced by the first frequency translation module 41.

[0058] Similarly, the part of the signal E 0 which passes through the second arm of the interferometer, passes through a second frequency translator module, which induces a second shift f 2. It does not undergo any delay.

[0059] Consequently, the signal E2 thus formed has the following expression: E 2 ¯ = E 2 e i 2 π ν 0 + 2 f 2 t where v 0 + 2f 2 is the shifted frequency of the optical wave E 2 , the frequency shift f 2 being induced by the second frequency translation modulus.

[0060] The intensity of the resulting optical wave E3 = E1 + E2, resulting from the recombination of waves E1 and E2 at the 50 / 50 coupler, is converted by the optical receiver 43, a photodiode for example, into a current I which has the general expression: I = E 3 ¯ E 3 ¯ ∗ = I 0 + 2 I 1 I 2 cos 2 π ν 0 + 2 f 1 − ν 0 − 2 f 2 t + 2 π ν 0 + f 1 τ where I0 = I1 + I2 represents the non-coherent part of the intensity, that is, the sum of the squared magnitudes of the amplitudes E1 and E2. The coherent part of the intensity is represented by the double product E1.E2.

[0061] Consequently, the intensity of the electrical signal at the output of the photodiode is expressed as: i t = E 3 ¯ E 3 ¯ ∗ = i 0 1 + V cos 4 π f 1 − f 2 t + Δϕ t where the phase term Δϕ is given by the relation: Δϕ t = 2 π ν 0 t + f 1 τ

[0062] The measured phase variation Δϕ is therefore expressed in such a way that the frequency considered is the center frequency v 0 shifted by f 1.

[0063] The phase term Δϕ corresponds to a phase modulation on a carrier frequency fp = 2 · (f1 - f2). The modulated signal therefore has a maximum listening bandwidth equal to: 2 ⋅ f p = 4 ⋅ f 1 − f 2

[0064] Combining an optical switch 31 with a frequency-translational interferometer 41 to separate the light waves from the different detection channels, which are associated with the various CLFO sensors 35, represents an optimal solution. It allows for the best bandwidth / carrier frequency ratio, resulting in improved dynamics. This means the measurement chain has high sensitivity while maintaining high saturation levels, allowing the system to easily adapt to variations in the measured flux and optimize its detections. For a length of optical fiber reel fixed in each arm of the interferometer (i.e.(a fixed delay τ), the PGC method requires a frequency carrier twice as high to achieve performance equivalent to combining an optical switch with a frequency-translation interferometer, while the "Phase-Stepping" method requires a frequency carrier n times higher, n being the number of steps used.

[0065] The optical processing module 32 also includes a wavelength demultiplexing stage 42 (optical demultiplexer) which has the function of separating the different wavelengths constituting the light wave carried by a detection channel and of transmitting separately to the receiving module 33 the M light waves carried by the detection channel considered, each light wave corresponding to the laser signal delivered by a CLFO 35.

[0066] From a structural point of view, the receiving module 33 is the demodulation stage and conventionally includes an opto-digital signal conversion stage 43, preferably made up of photodiodes, whose role is to convert the light waves transmitted by the optical processing module into electrical signals, as well as an analog-to-digital conversion (ADC) module 44 responsible for digitizing these signals.

[0067] It also includes a time-division multiplexing device 45 whose function is to manage the sequencing of the switching of the different detection channels and thus to reconstruct the different time signals, the switching being carried out by the optical switch 31.

[0068] Advantageously, the time-division multiplexing device 45 generates the switching command 47 which drives the optical switch 31.

[0069] It also includes a phase demodulator 46 which exploits the time-reconstructed optical signals and delivers the Δϕ signal after demodulation for each of the time-multiplexed detection channels.

[0070] Advantageously, the phase demodulator 46 generates the RF control signals 48 and 49, of respective frequencies f1 and f2, which drive the frequency translation modules of the interferometer 41, thus allowing the value of the frequency carrier fp to be played with and the performance of the measurement chain to be improved.

[0071] The operation of the various elements of module 33, which is known from other sources, is not detailed here.

[0072] From a functional point of view, the optical switch 31 has the function of sequentially selecting each of the N optical fibers, each fiber constituting a measurement channel grouping M CLFO sensors.

[0073] According to the invention, the switching frequency f SW of the optical switch 31 is defined by the time demultiplexing device 45 taking into account the frequency fp of the RF carrier wave defined by the phase demodulator device 46, as well as the frequency of the phase-modulated signal fs and the amplitude of the phase modulation ϕ s of the optical signal which carries the acoustic information to be measured.

[0074] Indeed, as previously mentioned, Carson's rule allows us to estimate the bandwidth B occupied by a phase-modulated signal, according to the following relationship: B = 2 f s ⋅ 1 + ϕ s where fs is the frequency of the signal considered and ϕs the amplitude of the phase modulation.

[0075] Insofar as the signal in question is carried by an RF carrier wave, the bandwidth B of the modulated signal is, moreover, limited by the frequency fp of the carrier, according to the relation: B = 2 ⋅ f p a signal modulated by the frequency translation interferometer 41 having in principle a bandwidth limited to 1.5 times the frequency of the modulation carrier.

[0076] Consequently, in order to correctly demodulate the carrier wave, it is necessary, since, as previously stated, the modulated signal at fp occupies a maximum bandwidth equal to B = 2 · fp, to sample the light wave with a sampling frequency fsample at least equal, according to Shannon's rule, to: f éch = 4 ⋅ f p

[0077] From a functional point of view, fp is generally between a few tens of kHz and a few MHz, so here we choose a frequency fp preferably equal to 100kHz.

[0078] Therefore, since channel switching constitutes sampling of the wave carried by each of the channels with a frequency f SW, guaranteeing condition

[14] for each of the N detection channels implies that the switching frequency f SW of the optical switch satisfies, at a minimum, the following relationship: f SW = N ⋅ f éch = N ⋅ 4 ⋅ f p

[0079] It should also be noted that the choice of the frequency fp of the RF carrier wave depends on the delays related to the optical components, defined largely by the difference in length between the two arms of the frequency translation interferometer, and on the delays related to the electronic components, fixed largely by the intrinsic properties of the devices constituting the optoelectronic module, which condition the switching frequency of the optical switch f SW, a frequency which itself conditions the frequency fp.

[0080] However, as illustrated by the figure 5The acquisition of a detection channel, after sending the switching command, can only begin after a certain delay, which is induced by a delay dependent on the switch and the processing time by the optical (t opt) and electrical (t elec) parts. The time t opt ​​is related to the length of the delay line within the interferometer, defining the bandwidth dynamics, and the time t elec is related to the bandwidth and therefore to the carrier.

[0081] As described previously, these optical time t opt ​​and electrical time t elec are therefore correlated and decreasing one of these two parameters will tend to increase the other, leading to the consideration of a compromise to optimize the measurement chain.

[0082] More specifically, the processing time of the different detection channels is mainly limited by the sum of the switching delay τ 1 of switch 31, the optical signal transit delay τ 2 on the delay line of interferometer 41 and the delay τ 3 defined in part by the rise time of the analog stage 43.

[0083] The rise time of the analog stage 43 of module 33 corresponds mainly to the delay induced by the amplification stage which follows the detector.

[0084] As is known, this time depends directly on the bandwidth f BP of the analog stage 43 and consequently on the choice of the carrier frequency fp.

[0085] Thus, for a single-stage low-pass filter, the delay τ3 is given, in practice, by the relation: τ 3 = τ elec ≈ 0 , 35 f BP

[0086] The choice of the carrier frequency in turn conditions the switching frequency of the switch, which must reach a switching frequency related to the carrier frequency, knowing that the higher the carrier frequency, the faster the switching must be (see relation

[15] ).

[0087] Furthermore, since the interferometer gain / analog amplification stage bandwidth relationship 43, responsible for the delay τ 3, is constant, it is necessary to consider the input level of the demodulation stage 33.

[0088] Furthermore, the bandwidth B of this amplification stage determines the current noise limit in of the demodulation stage 33, which is directly related to the bandwidth according to the following relationship: i n ∝ B

[0089] Furthermore, the delay τ2 introduced by the delay line of the optical processing module 32 relates the gain G of the interferometer to the phase variation Δϕ, induced by the shifted frequency Δf = v0 + f1, by the relation: Δϕ = G × Δf = 2 π × τ opt × Δf where τ opt is defined by the following relation: τ 2 = τ opt = 2 × nL c n here represents the refractive index of the optical fiber and c the speed of light in a vacuum; L represents the length of the delay line within the interferometer (length ratio between the two arms of the interferometer) and is in practice between 1m and 100m.

[0090] Considering the points presented above, the length L must be chosen such that the noise from the laser fiber cavities (LFCs) characterizing the signal to be measured is dominant over the demodulation noise induced by the optoelectronic module 33, in order to effectively exploit the result. Adjusting this length allows us to frame the dynamic range.

[0091] In other words, as the carrier frequency increases, the bandwidth increases (see relation

[13] ) and therefore τelec decreases (see relation

[16] ). Simultaneously, the noise of the demodulation module 33 increases with the increase in bandwidth (see relation

[17] ), and consequently L must be chosen to be larger, and therefore the delay τopt increases (see relation

[18] ). The best compromises are then chosen according to the system requirements, with the objective of minimizing τelec + τopt.

[0092] As can be seen from the preceding text, the association, according to the invention, of an optical switch with a frequency-translation interferometer thus advantageously makes it possible to form a unique measurement chain which, associated with a detection chain comprising an optical sensor network consisting of a plurality of detection channels, each consisting of an optical fiber on which is arranged a plurality of laser fiber optic cavities (LFOCs), makes it possible to realize a device for receiving and measuring the various incident optical signals from the detection chain exhibiting higher levels of performance, both in terms of bandwidth and dynamics, for a high material compactness and low power consumption, compared to other known measurement chains, such as devices implementing PGC or "Phase-Stepping" type interferometers.

[0093] It is thus possible, for example, to implement the single measurement channel according to the invention to make a linear sonar antenna having a small cross-section compared to conventional linear antennas and comprising a large number of optical fiber laser cavity sensors (CLFOs) distributed in groups of M CLFOs placed in series on optical fibers to form detection channels connected to a common measurement channel according to the invention, ensuring the phase demodulation of the M signals carried by each of the detection channels.

[0094] Such a structure advantageously allows for the pooling of measurement resources to form a single chain, whereas in a standard setup, these resources are duplicated, with an independent measurement chain connected to each detection channel and independently performing phase demodulation of the signals carried by that detection channel. This streamlining of measurement resources advantageously results in a compact antenna signal reception and measurement system with reduced manufacturing costs and power consumption, while guaranteeing signal dynamics and bandwidth performance compatible with the expected performance of such an antenna.

[0095] From the point of view of implementation, the different modules constituting the measurement chain according to the invention can be made up of different elements.

[0096] Thus, according to a preferred embodiment of the device according to the invention, the frequency translator(s) included within the frequency translation interferometer 41 of the optical processing module 32 can be made up of acousto-optic modulators.

[0097] Alternatively, frequency translation can be achieved through integrated optics, i.e. via very small optical elements or components included in portions of optical fiber and made by partial modifications of the optical fiber (e.g. through laser-matter interactions between a fabrication laser and the portion of optical fiber), e.g. by serrodyne frequency translation technique.

[0098] Furthermore, the frequency translation interferometer 41 of the optical processing module 32 can be manufactured according to the Michelson interferometry model.

[0099] Furthermore, in one particular embodiment, the Michelson interferometer used can be a mirror interferometer employing Faraday mirrors. Such a device advantageously renders the interferometer's operation insensitive to variations in the polarization of the optical signals from the CLFO sensors transmitted to the two arms of the interferometer.

[0100] Alternatively, the frequency-translational interferometer module can be fabricated according to the polarization-maintaining Mach-Zehnder interferometry model.

[0101] Advantageously, the longer arm of the interferometer is made of a fiber with a low temperature sensitivity compared to that of the fiber of the short arm, allowing the device, including the measurement chain and subjected to its environment, to passively compensate for temperature variations which normally induce variations in path difference (or delay τ), and consequently a non-zero mean drift of much greater amplitude than that of the useful signal for difficult exploitation of the measurement.

[0102] Thus, too, the optical switch 31 can consist of an electro-optical switch.

[0103] Alternatively, the optical switch 31 can be implemented using semiconductor optical amplifiers.

Claims

1. Device for detecting sonor signals, comprising N detection pathways, each having M fibre optic laser cavities (35), said detection pathways being configured to detect and convert the acoustic signals into optical signals, as well as a single measurement chain, combining the N detection pathways by time-division multiplexing; characterised in that said measurement chain combines an optical switch and interferometric measurement means, configured to measure, for each of the N detection pathways, the optical frequency of the signal delivered by each of the M fibre optic laser cavities, said measurement chain comprising: - an optical switch N×1 (31) making it possible to select, at each interferometric measurement, the optical fibre comprising the M fibre optic laser cavities multiplexed in a wavelength forming one same detection pathway; - a frequency translation interferometer (41), configured to transform the frequency modulated optical signal delivered by the M fibre optic laser cavities forming one same detection pathway in an intensity modulated optical signal; - an optical demultiplexer (42) making it possible to separate the M optical signals forming said intensity modulated optical signal and each delivering M optical signals over a specific outlet enabling independent photoelectric conversions; - an optoelectronic conversion stage (43-44), configured to generate, from the signal resulting from the beat of the two different frequency waves generated in the interferometer (41), an electric signal, analogue then digital; - a time-division demultiplexer (45) making it possible to reconstitute the different time signals coming from the different detection pathways; - a phase demodulator (46) making it possible to process the signals reconstituted by the time-division demultiplexer (45) from the optical signals delivered by each of the fibre optic laser cavities (35).

2. Device according to claim 1, wherein said interferometer (41) comprises means to translate into frequency, against one another, the optical signals crossing through each of the arms of the interferometer (41).

3. Device according to claim 2, wherein said frequency translation means are placed in the longest arm of the interferometer (41).

4. Device according to claim 2, wherein said frequency translation means are placed in the shortest arm of the interferometer (41).

5. Device according to claim 2, wherein said frequency transition means are composed of two frequency translators performing distinct frequency translations, each placed in one of the arms of the interferometer (41).

6. Device according to any one of claims 2 to 5, wherein said frequency translation means are acousto-optical modulators.

7. Device according to any one of claims 2 to 5, wherein said frequency translation means are optical elements integrated in an optical fibre.

8. Device according to any one of claims 1 to 7, wherein said interferometer (41) is a Michelson interferometer.

9. Device according to any one of claims 1 to 8, wherein said interferometer (41) comprises Faraday mirrors, configured and arranged to make the system independent from the polarised variations of the different signals by the fibre optic laser cavities.

10. Device according to any one of claims 1 to 9, wherein each of the arms of said interferometer (41) is formed by an optical fibre of a different size, the optical fibre forming the longest arm having a sensitivity to the temperature lower than the fibre of the shortest arm.

11. Device according to any one of the preceding claims, wherein the time-division demultiplexer (45) makes it possible to generate a control signal of the optical switch N×1 (31).

12. Device according to any one of the preceding claims, wherein the phase demodulator (46) makes it possible to generate a control signal of the interferometer.