Coherent Optical Receiver

Abstract

The coherent optical receiver comprises a local oscillator, a polarization diversity actuator configured to modify the optical signal output by the local oscillator, and a 2x2 coupler combining the optical signal output by the polarization diversity actuator and the modulated optical signal received from the coherent optical transmitter. The local oscillator therefore provides a boost effect to the amplitude shift keying modulated optical signal. The coherent optical receiver comprises a control unit that performs a domain switching procedure acting on the ellipse principal axis orientation and / or the ellipse phase shift for coarse control of the polarization diversity actuator, and a phase refinement procedure acting on a controlled error signal incident on the phase of the optical signal output by the polarization diversity actuator for fine control of the polarization diversity actuator.

Description

[Technical field]

[0001] This invention relates generally to optical communications, and more particularly to coherent optical receivers for optical communications. [Background technology]

[0002] Optical communications, such as Passive Optical Networks (PON), are increasingly being used to provide network access (typically Internet access) to homes and office gateways or data centers in the context of Fiber To The Home (FTTH) technology deployments. Optical communications are also used to provide backhaul for mobile infrastructure, for example in the context of the introduction of 3G (third generation) or 4G (fourth generation) mobile technologies, which typically use point-to-point mechanisms.

[0003] With the rise of 5G (5th generation) mobile technology, fronthaul is coming and the demand in terms of data rate capabilities is increasing significantly, see Non-Patent Document 1. In such a framework, fronthaul is achieved by moving the processing functions that were performed at or near the base station in 3G or 4G mobile technology to the upstream of the mobile infrastructure. This is called the split option in Non-Patent Document 2 (for more details, see Section 11 and Table A-1 of Non-Patent Document 2).

[0004] 5G mobile technology therefore has more extensive requirements compared to the FTTH requirements that have driven the deployment of optical access technology to date. 5G mobile technology therefore leads to, among other things, higher nominal data rates, reduced latency and higher density deployment. This not only means higher nominal data rates for optical access, but also expected split ratios higher than 1:128 compared to the typical FTTH split ratio of 1:64 for current PON systems.

[0005] To address this trend, coherent optical communication is envisioned, a technique that enables large-capacity data transmission over optical fibers using modulation of the amplitude and phase of light. Digital signal processing (DSP) is typically used to achieve robust performance of coherent communication over optical fibers.

[0006] Coherent optical communication is known because it has revolutionized the metro and core network sector, namely by offering good optical receiver sensitivity and good wavelength selectivity, as well as the ability to capture the full diversity of channels and to obtain high spectral efficiency.

[0007] For the principle of coherent optical communication used in metro and core network fields, please refer to Non-Patent Document 3. In this document, a coherent optical receiver utilizing phase and polarization diversity is shown. Two phase diversity homodyne receivers based on 90 degree optical hybrid power splitters to extract the in-phase and quadrature components of the incident coherent optical signal are arranged in a polarization diversity configuration together with a common local oscillator (see FIG. 9 of Non-Patent Document 3 for more details). A DSP mechanism needs to be used additionally to extract the modulated symbols by digitally processing the phase and polarization diversity outputs to stably recover the complex signal amplitude despite the fluctuations of the carrier phase and polarization signal states. To do this, the DSP mechanism comprises an anti-aliasing filter, a four-channel analog-to-digital converter (ADC), a frequency-domain equalizer (FDE) that allows symbol rate extraction to drive the ADC inversely, an adaptive finite impulse response (FIR) equalizer that allows clock phase recovery, and a carrier phase estimator (see FIG. 18 of non-patent document 3 for more details).

[0008] One major problem with coherent optical communication is that transmission over long-distance optical fibers means that the polarization state changes slow but large enough to be detectable at the coherent optical receiver. To ensure transmission continuity, two alternative approaches are used to avoid signal fading due to mismatch between the state of polarization of the signal received over the optical fiber and the state of polarization of the local oscillator. One approach is to compensate for the change in the state of polarization induced by the long-distance optical fiber using complex digital signal processing with a full diversity polarization receiver. Another approach is to track the state of polarization using an optical phase-locked loop (OPLL), for example in cases where full polarization diversity cannot be restored. Both of these approaches lead to complex analog and / or digital mechanisms with at least two analog branches. See for example "PLL: A Novel Method for Tracking the State of Polarization" by ...

[0009] From the above description, it is clear that the inherent theoretical hardware implementation applied in the metro and core network field does not meet the essential requirements in the network access field, such as low design complexity, low power consumption, low manufacturing cost and complexity, low installation cost and complexity, and small size. In fact, the number of devices installed in the network access field is much larger than that installed in the metro and core network field, especially to meet the geographical density expectations of base stations in 5G mobile technology, and furthermore, the locations where such devices are installed are often particularly limited in terms of space (home devices, base station installation locations, etc.). Furthermore, optical transmission in the metro and core network field suffers from noise due to signal amplification used to cover long distances from point to point. On the contrary, optical transmission in the network access field operates in a relatively short distance, but the signal strength (the amount of photons received by the optical receiver) is low considering the splitting ratio, and therefore the sensitivity of the optical receiver is the main concern. Therefore, coherent optical communication technology as used in the metro and core network field is not suitable for 5G mobile technology.

[0010] In addition to the scope of 5G mobile technology, coherent optical communication may be suitable for FTTH deployment. In fact, the growing demand for high-speed Internet has led to higher split ratios of current PON systems in addition to higher data rates. However, for the deployment of base stations in 5G mobile technology, optical network units (ONUs) as used in such PON systems are required to have low design complexity, low power consumption, low manufacturing cost and complexity, low installation cost and complexity, and small size, since they are deployed in or near the user's premises. Furthermore, for the 5G mobile technology context, optical communication in the metro and core network fields suffers from noise due to signal amplification used to cover point-to-point long distances. Conversely, optical communication in the network access field operates over a relatively short distance, but the signal strength (the amount of photons received by the optical receiver) is low when considering the split ratio, and therefore the sensitivity of the optical receiver is the main concern. Therefore, coherent optical communication technology as used in the metro and core network fields is also not suitable for FTTH deployment.

[0011] It is therefore desirable to provide a cost effective solution for a coherent optical receiver suited to access network considerations. Furthermore, it is desirable to provide a solution that is as simple as possible. [Prior art documents] [Non-patent literature]

[0012] [Non-Patent Document 1] the International Mobile Telecommunications (IMT) recommendations ITU-R M.2083 “IMT Vision: Framework and overall objectives of the future development of IMT for 2020 and beyond”, released in September 2015 [Non-Patent Document 2] the specifications 3GPP TR 38.801 V 14.0.0 “Study on new radio access technology: Radio access architecture and interfaces”, released in March 2017 [Non-Patent Document 3] “Fundamentals of Coherent Optical Fiber Communications”, K. Kikuchi, Journal of Lightwave Technology, vol. 34, no.1, released in January 2016 [Non-Patent Document 4] “Endless Polarization Control Systems for Coherent Optics”, Noe et al, IEEE Journal of Lightwave Technology, vol. 6, no.7, pp. 1199-1207, released in July 1988 Summary of the Invention

[0013] To this end, a coherent optical receiver intended to receive an amplitude shift keying modulated optical signal from a coherent optical transmitter is provided, the receiver comprising a local oscillator and an ellipse having a major axis orientation φ l and the elliptical phase shift ψ la polarization diversity actuator configured to modify the optical signal output by the local oscillator to form an optical signal having an elliptical polarization of φ; a 2×2 coupler, one input of which is targeted to receive an amplitude shift keying modulated optical signal received from a coherent optical transmitter and the other input of which receives another optical signal output by the set formed by the local oscillator and the polarization diversity actuator, so as to enable the local oscillator to provide a boost effect to the amplitude shift keying modulated optical signal received from the coherent optical transmitter, the 2×2 coupler further having one output connected to the set formed by a photodiode and a direct current filter placed after the photodiode, which removes a continuous component of the analog electrical signal output by the photodiode; a transimpedance amplifier for converting the current electrical signal output by the direct current filter into a voltage electrical signal; and a domain switching procedure for determining whether the ellipse main axis orientation φ is satisfied when the voltage electrical signal output by the transimpedance amplifier is equal to or smaller than a predetermined first threshold THdsp corresponding to a predetermined percentage of the theoretical maximum achievable by the boost effect. l and / or elliptical phase shift ψ l and a control unit in the form of an electronic circuitry configured to command a configuration change of the polarization diversity actuator by changing the phase of the local oscillator and the polarization diversity actuator. Furthermore, the electronic circuitry is further configured to perform a phase refinement procedure to insert a controlled error signal in the phase of the optical signal output by the set formed by the local oscillator and the polarization diversity actuator and to adjust the static component of the controlled error signal towards the maximum voltage electrical signal output by the transimpedance amplifier. Therefore, by using such a single-branch coherent optical receiver, neither complex digital signal processing nor optical phase-locked loops are used, thus providing a cost-effective solution for access network considerations. Moreover, the phase refinement procedure makes it possible to optimize the coarse configuration provided by the domain switching procedure.

[0014] In certain embodiments, during the domain switching procedure, the elliptical main axis orientation φ l The modification of is performed by applying a predefined shift of π / 4, resulting in an elliptical phase shift ψ l The modification of is performed by applying a predefined shift of π / 2. Hence, restoration of the effective boost effect (due to the boost term provided by the local oscillator) is easily achieved.

[0015] In a particular embodiment, the polarization diversity actuator has an elliptical major axis orientation φ l and allows the change of the elliptical phase shift ψ l The present invention includes a variable wave plate that allows for the change of configuration. Therefore, configuration changes are easily implemented with low complexity and cost.

[0016] In a particular embodiment, the electronic circuitry is configured to perform an automatic configuration of the standby period between successive executions of the domain switching procedure, which includes monitoring the evolution of the electrical signal output by the transimpedance amplifier first continuously during the domain switching procedure, tracking the history of the variations related to the polarization state by the monitoring, monitoring the period between successive commanded changes in the ellipse parameters, and defining the standby period between successive executions of the domain switching procedure to be shorter than the stabilization period between successive crossings of the predetermined first threshold value THdsp when the evolution of the polarization state variations is stabilized. Thus, the energy of the coherent optical receiver can be preserved during the standby period between successive executions of the domain switching procedure, the standby period being defined automatically.

[0017] In a particular embodiment, the electronic circuitry is configured to perform an automatic configuration of the standby period between successive executions of the phase refinement procedure. The automatic configuration includes that monitoring of the evolution of the electrical signal output by the transimpedance amplifier is first performed continuously in the range of the phase refinement procedure, monitoring a history of the period between successive crossings of the predetermined second threshold THprp, and defining the standby period between successive executions of the phase refinement procedure to be shorter than the stabilization period between successive crossings of the predetermined second threshold THprp. Thus, the energy of the coherent optical receiver can be preserved during the standby period between successive executions of the phase refinement procedure, the standby period being defined automatically.

[0018] In certain embodiments, the controlled error signal is a modulated signal so that it can be easily tracked in the voltage electrical signal output by the transimpedance amplifier.

[0019] In a particular embodiment, the controlled error signal is a modulation on the phase of the optical signal output by the set formed by the local oscillator and the polarization diversity actuator, and has the following form: φ0+Δφ0cos(Ωt+φ) where φ0 is the static component of the controlled error signal. Therefore, the controlled error signal can be easily generated.

[0020] In a particular embodiment, during the phase refinement procedure, the electronic circuitry is configured to monitor the output variation of the transimpedance amplifier at a frequency defined by Ω and adjust φ towards reaching a maximum voltage electrical signal output by the transimpedance amplifier, where φ is φ max is equal to.

[0021] In certain embodiments, the electronic circuitry comprises a φ max The elliptical phase shift ψ is used to compensate for lBy varying φ and resetting φ, the dynamics of the modulator used to generate the error signal is therefore preserved.

[0022] In certain embodiments, the electronic circuitry determines whether a maximum voltage electrical signal at the output of the transimpedance amplifier is in accordance with an ellipse major axis orientation φ l allows the modification of the elliptical phase shift ψ l is configured to transfer the mitigation of the static component φ0 if the polarization diversity actuator does not vary by more than a predetermined third threshold THmax over a period corresponding to the bandwidth of a variable waveplate used in the polarization diversity actuator to allow for the change of φ0. Thus, the transfer of the mitigation of the static component φ0 occurs once stability is reached.

[0023] In a particular embodiment, the electronic circuitry is configured to monitor the evolution of the phase detuning in order to detect a continuous monotonic evolution of the phase detuning and adjust the wavelength configuration of the local oscillator to compensate for the detected continuous monotonic evolution of the phase detuning, so that refinement of the phase of the optical signal output by the set formed by the local oscillator and the polarization diversity actuator is not triggered in time due to wavelength drift.

[0024] In certain embodiments, once a wavelength adjustment has been performed, the electronic circuitry is configured to restart the phase refinement procedure so that the new optimization easily takes the wavelength adjustment into account.

[0025] In a particular embodiment, the electronic circuitry is configured to restart the phase refinement procedure at time intervals during a local oscillator reconfiguration transient, such that phase refinement is performed smoothly during the local oscillator reconfiguration transient.

[0026] In a particular embodiment, the coherent optical receiver further comprises a temperature sensor capturing the temperature of the local oscillator and a look-up table establishing a relationship between temperature drift and wavelength drift, the electronic circuitry being configured to monitor the temperature evolution captured by the temperature sensor, to retrieve from the look-up table a wavelength drift corresponding to the temperature drift indicated by the temperature evolution, and to superimpose on the controlled error signal an opposite contribution of the phase evolution caused by the wavelength drift. Therefore, the wavelength drift effect on the optical signal phase is easily managed, and the refinement of the phase of the optical signal output by the set formed by the local oscillator and the polarization diversity actuator is not triggered in time due to the wavelength drift.

[0027] It is further proposed an optical network unit intended for use in a passive optical network, the optical network unit comprising a coherent optical receiver for receiving an amplitude shift keying modulated optical signal transmitted by a coherent optical transmitter included in an optical line terminal of the passive optical network. Thus, a low complexity and cost optical network unit solution is provided for the passive optical network infrastructure.

[0028] The characteristics of the invention will emerge more clearly on reading the following description of at least one example of embodiment, the said description being made with reference to the accompanying drawings, in which: [Brief description of the drawings]

[0029] [Figure 1A] 1 is a schematic diagram of the architecture of an optical communication system of the passive optical network type in which the invention can be implemented; [Figure 1B] 1 is a schematic diagram of the architecture of an optical communication system of point-to-point type in which the present invention can be implemented; [Diagram 2] FIG. 2 is a schematic diagram of a coherent optical receiver mechanism that can be used in the optical communication system of FIG. 1A or FIG. 1B. [Diagram 3] 3 is a schematic diagram of a mechanism of a polarization diversity actuator of the coherent optical receiver of FIG. 2. [Figure 4] 3 is a schematic diagram of the mechanism of a value decision block of the coherent optical receiver of FIG. 2. [Diagram 5] 3 is a schematic diagram of an algorithm executed by the control unit of the coherent optical receiver of FIG. 2 to adjust the configuration of the coherent optical receiver in a domain switching procedure; [Figure 6] FIG. 4 is a schematic diagram of an algorithm executed by the control unit to adjust the configuration of the coherent optical receiver in a phase refinement procedure. [Figure 7] 3 is a schematic diagram of an algorithm executed by the control unit to adjust the wavelength of the local oscillator of the coherent optical receiver of FIG. 2; [Figure 8] FIG. 2 is a schematic diagram of a particular temperature sensor-based mechanism of a coherent optical receiver. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0030] It should be noted that wavelength and frequency are linked to each other through a direct inverse relationship, and the two terms are used interchangeably by those skilled in the art since they refer to the same concept.

[0031] The present invention can be applied to implement a low-complexity coherent optical receiver intended to receive and detect an optical signal transmitted by a coherent optical transmitter. More specifically, the coherent optical receiver is suitable for receiving an amplitude shift keying (ASK) modulated optical signal. For example, the amplitude shift keying (ASK) modulated optical signal is a non-return-to-zero on-off keying (NRZ-OOK) modulated optical signal. In the following description of Figures 1A and 1B, an example of a context in which the present invention is useful is introduced.

[0032] FIG. 1A shows a schematic representation of the architecture of an optical communication system 100 of the Passive Optical Network (PON) type in which the invention can be implemented.

[0033] The optical communication system 100 in FIG. 1A includes a master device 110, typically an OLT (Optical Line Terminal) device, and multiple slave devices 141, 142, 143, typically ONU (Optical Network Unit) devices.

[0034] The optical communication system 100 in FIG. 1A may further include at least one spectral splitter device 130 and / or at least one power splitter.

[0035] Each spectrum splitter device 130 includes a pair of optical bandpass filters set for each PON, each intended to filter a respective wavelength band, and thus enables wavelength division multiplexing (WDM). Instead of using a spectrum splitter device 130, an equivalent mechanism can be obtained by applying a filtering film to the receiving diode.

[0036] Each power splitter can increase the number of slave devices that can be connected to the master device 110 by dividing the input signal power by the number of outputs going to the slave devices connected to the master device 110. Therefore, each output of the power splitter device transmits the same optical information that was received as the input signal, and the power splitter device only affects the signal power.

[0037] The slave devices 141, 142, 143 are interconnected to the master device 110 using optical fibers 120 via at least one spectral splitter device 130 and / or at least one power splitter.

[0038] In the context of a PON, the ONUs are typically intended to be located at end-user premises for FTTH services, and the OLT enables the ONUs to access a metropolitan or core network (not shown). Such PONs may also be used for mobile network infrastructure services.

[0039] FIG. 1B shows a schematic representation of the architecture of an optical communication system of the point-to-point type in which the invention can be implemented.

[0040] 1B includes a master device 110 and a slave device 140. The master device 110 and the slave device 140 are interconnected using an optical fiber 120.

[0041] In a mobile network fronthauling infrastructure, the slave device 140 is typically located at a remote radio head location and the master device 110 is located at a baseband unit (BBU) location. Such a mechanism may also be used for a mobile network backhauling infrastructure between the BBU and a core network access terminal.

[0042] In both the arrangements of Figures 1A and 1B, optical transmission of information from master device 110 to one or more slave devices is referred to as downlink transmission, and optical transmission in the reverse direction is referred to as uplink transmission.

[0043] 1A and 1B, it is desirable to place low-complexity and cost-effective optical receivers at least in the slave devices 140, 141, 142, 143, and potentially in the master device 110. Moreover, it is desirable for the optical receivers to be highly sensitive to handle potential low signal strengths.

[0044] Figure 2 shows a schematic diagram of a coherent optical receiver 200 that can be used in the optical communication system 100 of Figure 1A or 1B. The coherent optical receiver 200 outputs a signal 202 to be processed for demodulation.

[0045] According to the arrangement shown in Fig. 2, the coherent optical receiver 200 includes a 2x2 coupler 220. For example, the 2x2 coupler 220 is a fused fiber coupler or a directional coupler.

[0046] An optical signal transmitted by a coherent optical transmitter is input to one input 221 of the 2×2 coupler 220. Another optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211 is input to the other input 222 of the 2×2 coupler 220.

[0047] The local oscillator LO 210 is a laser diode or a mechanism that includes a laser diode. A vertical cavity surface emitting laser (VCSEL) may also be used as the local oscillator.

[0048] One output 223 of the 2x2 coupler 220 is connected to a set formed by a photodiode 230 followed by a DC (direct current) filter 240 of the coherent optical receiver 200. The DC filter 240 is a mechanism for removing the continuous component of the analog electrical signal output by the photodiode 230. The other output 224 of the 2x2 coupler 220 is left unused. The other output 224 of the 2x2 coupler 220 can be used for improved detection or other purposes. By doing this, a single analog branch coherent optical receiver is provided.

[0049] In general, a polarized optical signal E(t) can be described in the Jones representation as follows:

number

number

[0050] Therefore, at the output 223 of the 2×2 coupler 220, the 2×2 coupler 220 provides a composite signal C(t) that is:

number

[0051] The optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211 is not modulated for the transmission of information and has an amplitude E 0l Therefore, this means that the signal amplitude E 0l This means that the variability in does not contain significant information by itself.

[0052] E 0l =E 0s , ω l=ω s , φ l =φ s , and

number

[0053] Therefore, considering an amplitude shift keyed optical signal, and further considering the effect of DC filter 240 on the output of photodiode 230, the current i(t) output by DC filter 240 can be expressed as:

number

number

[0054] In the above equation for i(t), the following terms:

number

number

[0055] Control of the polarization diversity actuator PDA 211 therefore allows for adjustment of the boost effect. To control the polarization diversity actuator PDA 211, the coherent optical receiver 200 further comprises a control unit 280 in the form of an electronic circuit, which is present at the output of the DC filter 240 and which loops back to the polarization diversity actuator PDA 211 via line 261. The control unit 280 further also controls the local oscillator LO 210 and may therefore be looped back to the local oscillator LO 210 via at least one line, for example line 271.

[0056] The electronic circuitry forming the control unit 280 may be a chip or a chipset, such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit). In a variant, the electronic circuitry forming the control unit 280 may be a processor (microcontroller, DSP (Digital Signal Processor) etc.) associated with at least one memory. After power-up, the processor is capable of reading instructions from the memory and executing these instructions to implement the functions of the control unit 280. In this case, the functions of the control unit 280 are in software form, which may be stored on a non-transitory storage medium, such as an SD (Secure Digital) card.

[0057] Transimpedance amplifier TIA 245 resides between DC filter 240 and control unit 280 to convert the input current (output by DC filter 240) to an output voltage. Control unit 280 preferably loops back to transimpedance amplifier TIA 245 via line 251 for electrical gain control.

[0058] It should be understood that the control unit 280 does not perform tracking of the state of polarization. In the present case, it is not necessary to know exactly the state of polarization of the optical signal received from the coherent optical transmitter, nor the state of polarization of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211. Within the scope of the present invention, the control unit 280 determines whether the state of polarization of the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211 is suitable for obtaining an actionable boost term, and if not (without knowing which relevant state of polarization led to such a situation), the control unit 280 adjusts the configuration of the polarization diversity actuator PDA 211 in response. There is no need to determine the exact state of polarization of the optical signal, a single analog branch is sufficient to do this, which reduces the analog complexity. Furthermore, no optical phase-locked loop is required, and no complex digital signal processing is required, which further reduces the design complexity. Actionable boost terms are those boost terms that are greater than or equal to a predefined threshold percentage (e.g., 20%) of their theoretical maximum value, which is the maximum value of the power of the optical signal output by the local oscillator LO 210.

[0059] From the above formula, E 0s '=E 0l 'is (E 0s =E 0l (meaning that), but with further Δφ=(2q+1)π,

number

number

[0060] From the above formula, E 0s '=E 0l 'is (E 0s =E 0l (meaning that), but with further Δφ=(2q+1)π,

number

number

[0061] This therefore means that in these two cases, the optical signal received by the coherent optical transmitter and the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211 are in the same polarization state but have a phase shift of π. Considering the conditions faced in these cases, they are considered to be unlikely to occur, since they may be faced at one sample between consecutive samples in the modulation symbol and can be easily restored during the demodulation operation. Nevertheless, by providing a solution when these cases are faced, as detailed herein in at least one particular embodiment, the coherent optical receiving performance is improved.

[0062] More importantly, the local oscillator LO 210 allows to boost the optical signal coming from the optical fiber, unless the following conditions are met:

number

[0063] Apart from these circumstances, the local oscillator LO 210 allows to increase the sensitivity of the optical receiver and, in particular, to increase the transmission data rate together with an improved splitting ratio across the PON. It does not seem necessary to provide comprehensive systematic polarization tracking and polarization conversion (either in an analog or digital manner) in order to obtain an effective boost signal at the symbol granularity.

[0064] To avoid the above situation, the polarization diversity actuator PDA 211 is controlled by the control unit 280 to modify the optical signal as required when output by the local oscillator LO 210. In other words, by controlling the polarization diversity actuator PDA 211, it is possible to avoid the above situation when it is faced. In fact, even a small change in the state of polarization of the signal incident on the input 222 of the 2x2 coupler 220 produces a sufficient difference such that at least one of the boost terms is not equal to zero. In fact, the boost effect is obtained when at least one of these boost terms is not equal to zero and furthermore these terms do not compensate each other.

[0065] The main function of the control unit 280 is therefore to ensure that the electrical signal i(t) does not disappear to zero while the optical signal received from the coherent optical transmitter contains useful information, and therefore that the boost effect assumed to be provided by the local oscillator LO 210 is maintained as far as possible above the predefined threshold percentage mentioned above.

[0066] When the control unit 280 detects that the electrical signal output by the transimpedance amplifier TIA 245 is below a predefined threshold THdsp, the control unit 280 controls the polarization diversity actuator PDA 211 to change the state of polarization of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211. As a result, a boost effect is obtained in cases where the optical signal received from the coherent optical transmitter contains useful information, because the sensitive change of the state of polarization makes it possible to avoid facing a situation where the electrical signal i(t) disappears to zero even though the optical signal received from the coherent optical transmitter contains useful information. In the worst case, it can be demonstrated that the contribution associated with the boost term is larger than that of the optical signal received from the coherent optical transmitter as soon as the state of polarization of both optical signals entering the 2×2 coupler 220 shifts by a value that is a small percentage (for example, 5%) of the distance between adjacent optical related states of polarization.

[0067] After several kilometers in the optical fiber, the state of polarization of the optical signal initially transmitted by the coherent optical transmitter is randomly distributed on the Poincaré sphere. Therefore, the polarization state of the optical signal when received by the coherent optical receiver 200 is very likely to be elliptical. Therefore, the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211 is controlled to be elliptical polarized. Therefore, in order to change the state of polarization of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211, if the electrical signal i(t) falls into the zero region even though an optical signal containing useful information is received from the coherent optical transmitter, the control unit 280 controls the polarization diversity actuator PDA 211 to adjust the angle φ of the orientation of the main axis of the polarization ellipse of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211. land the elliptical phase shift ψ l , which allows the boost term to take effect and the photocurrent to be recovered with high sensitivity. In a particular embodiment, the control unit 280 commands the polarization diversity actuator PDA 211 to change at least one ellipse parameter from among the parameters φ of the orientation of the main axis of the polarization ellipse of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211. l and the elliptical phase shift ψ l to change at least one ellipse parameter from among

[0068] In a particular embodiment, the predefined shift targets a stationary extremum of the electrical signal i(t). In this particular embodiment, the predefined shift is l (and therefore Δφ) for π / 4, and the elliptical phase shift ψ l (and therefore π / 2 for Δψ).

[0069] Therefore, the current i(t) output by the DC filter 240 is prevented from vanishing to zero in a simple and cost-effective manner. The procedure disclosed above provides a coarse control of the change in the state of polarization without tracking the state of polarization, and is referred to herein as a "domain switching procedure". The domain switching procedure can be performed continuously at intervals in time, for example at regular time intervals. For example, the domain switching procedure is performed periodically on a time cycle basis, and its duration is defined to be lower than the time constant of the polarization state variation. The magnitude of the time constant of the polarization state variation is typically about 1 millisecond or a few milliseconds, but may vary according to the optical fiber deployment on the field. For example, the variation of the polarization state depends on the vibrations caused by the optical transmission and varies depending on whether the optical fiber deployment on the field is in the air, deep underground, or at a short distance from the earth's surface.

[0070] In a particular embodiment, an automatic configuration of the standby period between successive executions of the domain switching procedure is implemented. First, monitoring of the evolution of the electrical signal output by the transimpedance amplifier TIA 245 is performed continuously in the scope of the domain switching procedure. Hence, the history of the variations related to the polarization state is tracked and the period between successive changes in the ellipse parameters commanded by the control unit 280 is monitored. Once the evolution of the polarization state variations is stabilized, the control unit 280 defines the standby period between successive executions of the domain switching procedure to be shorter (e.g., half or with a predefined time margin) than the stabilization period between successive crossings of the predefined threshold THdps (i.e., triggering the change of at least one ellipse parameter commanded by the control unit 280). Thus, the time interval between successive executions of the domain switching procedure does not have to be specifically configured by the operator with respect to the effective deployment of the optical fiber on the field.

[0071] Further details regarding the domain switching procedure are disclosed below with respect to FIG.

[0072] The precise control of the change in polarization state is achieved by varying the phase φ of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211. l To do this, a modulated signal is introduced by the polarization diversity actuator PDA 211 under the control of the control unit 280 in a "phase refinement procedure".

[0073] The phase refinement procedure is preferably performed as a result of the modification of at least one ellipse parameter during the execution of the domain switching procedure. The phase refinement procedure can be performed independently of the domain switching procedure. The phase refinement procedure can be performed upon detecting that the evolution of the electrical signal output by the transimpedance amplifier TIA 245 exceeds a predefined threshold value THprp. In one variant, the phase refinement procedure can be performed at intervals, for example at regular intervals.

[0074] For the domain switching procedure, the standby period between successive executions of the phase refinement procedure can be automatically defined. First, monitoring of the evolution of the electrical signal output by the transimpedance amplifier TIA 245 is performed continuously in the range of the phase refinement procedure. The control unit 280 tracks the history of the period between successive crossings of the predefined threshold THprp. Upon stabilization, the control unit 280 defines the standby period between successive executions of the phase refinement procedure to be shorter (e.g., half or with a predefined time margin) than the stabilized period between successive crossings of the predefined threshold THprp.

[0075] Phase φ l The determination of how must be adjusted can be achieved by a controlled error signal introduced by the phase modulator PM 301 in the polarization diversity actuator PDA 211, as will be described in more detail below, and more particularly with respect to Figures 3 and 6. Indeed, introducing an error signal as a stimulus to the signal phase implies a perceptible effect on the current i(t) output by the DC filter 240, thereby determining the phase φ that results in the maximum signal strength. l It becomes possible to discover the value of

[0076] In a particular embodiment, the controlled error signal is a modulation on the phase of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211, and can be expressed as follows: φ0+Δφ0cos(Ωt+φ)

[0077] This results in φ l =π / 2 and ψ l Considering the case of a local oscillator with circular polarization of =π / 2 leads to the following expression for the current i(t) output by DC filter 240: E 0s 'E ol 'sin(φ s )cos(2πΔνt+Δφ+φ0+Δφ0cos(Ωt+φ)+ψ s )

[0078] Therefore, φ0 is the phase φ l t, while Δφ allows oscillation about this constant component, which results in perceptible fluctuations in the current i(t) output by DC filter 240. φ fixes the time origin of the controlled error signal and is of concern only for synchronous detection, as disclosed below.

[0079] Returning to FIG. 2, the control unit 280 comprises, in a particular embodiment, a value decision block VDB 250, a polarization diversity manager PDM 260, and a detuning manager DM 270.

[0080] The value decision block VDB 250 analyzes the electrical signal output by the transimpedance amplifier TIA 245. The value decision block VDB 250 analyzes the voltage output by the transimpedance amplifier TIA 245. The value decision block VDB 250 compares the value of the voltage output by the transimpedance amplifier TIA 245 with the theoretical maximum achievable by the boost effect, i.e., a predefined threshold percentage (e.g., 20%) of the theoretical maximum achievable by the boost term. The theoretical maximum is defined according to the gain of the local oscillator LO 210, the conversion rate R of incident photons to electrons by the photodiode 230, and the gain of the transimpedance amplifier TIA 245. If the voltage output by the transimpedance amplifier TIA 245 falls below a predefined threshold percentage of the theoretical maximum achievable by the boost term, the value decision block VDB 250 determines the angle φ of the orientation of the main axis of the polarization ellipse of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211. l and the elliptical phase shift ψ l , operative to change the configuration of the polarization diversity actuator PDA 211 in order to change at least one ellipse parameter from among

[0081] The polarization diversity manager PDM 260 controls the polarization diversity actuator PDA 211 via line 261. The polarization diversity manager PDM 260 receives from the value decision block VDB 250 via line 252 the angle φ of the orientation of the main axis of the polarization ellipse of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211. l and the elliptical phase shift ψ linto appropriate commands (e.g., drive voltage changes) suitable for the polarization diversity actuator PDA 211. The polarization diversity manager PDM 260 in turn controls the polarization diversity actuator PDA 211 via line 261.

[0082] The detuning manager DM 270 controls the local oscillator LO 210 via line 271 and the polarization diversity actuator PDA 211 via line 272. The detuning manager DM 270 converts instructions received from the value decision block VDB 250 via line 253 for changing the wavelength of the optical signal output by the local oscillator LO 210 into appropriate commands suitable for the local oscillator LO 210. The detuning manager DM 270 consequently controls the local oscillator LO 210 via line 271. For example, the detuning manager DM 270 adjusts the Peltier current if the local oscillator LO 210 comprises a Peltier module or adjusts the drive voltage if the local oscillator LO 210 comprises an optical ring resonator. The detuning manager DM 270 converts instructions received from the value decision block VDB 250 via line 253 (or another line dedicated to it) into appropriate commands suitable for the phase modulator PM 301 to modify the parameters of the controlled error signal introduced by the phase modulator PM 301 (φ0 and Δφ0 in the above formula).

[0083] Furthermore, the value decision block VDB 250 can provide instructions via line 251 to the transimpedance amplifier TIA 245 for modifying the electrical gain of the transimpedance amplifier TIA 245 .

[0084] Fig. 3 shows a schematic representation of a particular embodiment of the polarization diversity actuator PDA 211. The particular embodiment of Fig. 3 is suitable for using a local oscillator providing a linearly polarized optical signal, as is commonly found in off-the-shelf products, however, the particular embodiment of Fig. 3 is also suitable for using a local oscillator providing a circular or elliptical polarization.

[0085] Low-cost lasers currently used in the network access field are typically linearly polarized semiconductor lasers. Such lasers can be used as local oscillators LO 210. A way to convert linear polarization to elliptical polarization is to place a waveplate (also called a wavelength plate) at the output of the laser. Waveplates are based on birefringent materials with two orthogonal axes of different velocities, which induces a phase shift between the projection of a given field on the two axes of the birefringent material. By injecting a linearly polarized optical signal inside, an elliptically polarized optical signal is output. However, it can be taken into account that the waveplate combines both the polarization ellipse and the phase shift, which means that the phase shift between the optical signal output by the local oscillator LO 210 and the optical signal received from the coherent optical transmitter can be influenced by the waveplate as well. Therefore, in cases where the ellipse has to be independently controlled using a waveplate, a compensating phase shift can be applied to both axes of the optical signal output by the local oscillator LO 210 to counter the waveplate phase shift thus introduced.

[0086] A variable waveplate can be used to control the ellipse of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211. A variable waveplate is a specific waveplate that can control its phase shift between the slow (exceptional) and fast (normal) axes. Liquid crystal variable retarders and pocket cells are examples of variable waveplates. A tunable highly birefringent photonic liquid crystal fiber can alternatively be used. The phase difference is then adjusted over a specified range by changing the driving voltage.

[0087] 3, the optical signal emitted by the local oscillator LO 210 is incident on the phase modulator PM 301 described above, which generates a controlled error signal. The phase modulator PM 301 can be an electrically driven optical modulator based on a Pockels cell, or a liquid crystal modulator, or a variable length waveguide, or a waveguide with a variable optical index.

[0088] At the output of the phase modulator PM 301, the polarization diversity actuator PDA 211 further comprises a variable wave plate VW 302, here denoted by ψ l and therefore, as a result, l and ψ s It is used to reduce the difference between

[0089] The polarization diversity actuator PDA 211 may further comprise a controller CTRL 303 configured to interpret instructions received from the control unit 280, more specifically from the polarization diversity manager PDM 260 or the detuning manager DM 270, and to convert these instructions into commands (such as appropriate drive voltages) for the variable wave plate VW 302 and the phase modulator PM 301, respectively.

[0090] In one variant, the control unit 280, more specifically the polarization diversity manager PDM 260 and the detuning manager DM 270, send commands (such as appropriate driving voltages) directly to the variable wave plate VW 302 and the phase modulator PM 301. In this case, the control unit 280, more specifically the polarization diversity manager PDM 260, on the one hand, controls φ l (and consequently the change in Δφ) and / or ψ l (and consequently the change in Δψ) on the one hand and with commands to be applied to the polarization diversity actuator PDA 211, more particularly to the variable wave plate VW 302, on the other hand; furthermore the control unit 280, and more particularly the polarization diversity manager PDM 260, may be provided with a look-up table LUT that links the phase modulation parameters on the one hand and with commands to be applied to the polarization diversity actuator PDA 211, more particularly to the phase modulator PM 301, on the other hand.

[0091] FIG. 4 shows a schematic representation of a particular embodiment of the value decision block VDB 250 .

[0092] The value decision block VDB 250 comprises a level detector LD 402. The level detector LD 402 detects the voltage output by the transimpedance amplifier TIA 245 and performs a conversion into digital data that can be more easily handled for further analysis.

[0093] The value decision block VDB 250 further comprises an analyzer AN 403. The analyzer AN 403 receives the digital data from the level detector LD 402. The analyzer AN 403 compares the value of the voltage output by the transimpedance amplifier TIA 245 with a predefined threshold percentage (e.g., 20%) of the theoretical maximum achievable by the boost term. The analyzer AN 403 then makes a decision as to whether at least one action must be selected to ensure that the actionable boost term is met.

[0094] The value decision block VDB 250 further comprises a manager MGR 404 which receives instructions from the analyzer AN 403 regarding the action to be taken. The manager MGR 404 consequently provides polarization diversity related instructions to the polarization diversity manager PDM 260 via line 252 and / or electrical gain control instructions to the transimpedance amplifier TIA 245 via line 251 and / or phase tuning related instructions to the detuning manager DM 270 via line 253 and / or wavelength tuning related instructions to the detuning manager DM 270 via line 254.

[0095] In a preferred embodiment, the value decision block VDB 250 further comprises a squaring module SQ 401, which squares the electrical signal received from the transimpedance amplifier TIA 245 (i.e., the squaring module SQ 401 multiplies the input electrical signal by itself). The squaring module SQ 401 simplifies the analysis performed by the analyzer AN 403 by ensuring that only positive values ​​are handled (which is easier to compare the absolute electrical signal to a threshold). To do this, the squaring module SQ 401 is, for example, an analog amplifier whose both inputs receive the electrical signal received at the input of the value decision block VDB 250. The use of the squaring module SQ 401 further advantageously enhances the contrast of the boost term for the optical signal received from the coherent optical transmitter.

[0096] Instead, by working without the squaring module SQ 401, it is possible to perform a NRZ-OOK-like demodulation based on the minimum negative and maximum positive levels.

[0097] The value decision block VDB 250 outputs signals 202 to be processed for demodulation. These signals may be the outputs of the squaring module SQ 401. These signals may simply be the outputs of the transimpedance amplifier TIA 245.

[0098] 5 shows a schematic representation of an algorithm executed by the control unit 280 to configure the coherent optical receiver 200 within the domain switching procedure. The control unit 280 continuously receives an analog electrical signal from the transimpedance amplifier TIA 245.

[0099] In step S501, the control unit 280 performs signal level detection, as already described above, on the electrical signal output by the transimpedance amplifier TIA 245. Step S501 outputs a digital representation of the analog electrical signal (voltage) output by the transimpedance amplifier TIA 245 according to the sampling rate. The level detection preferably includes integration or averaging over a predetermined number of samples.

[0100] In step S502, the control unit 280 performs a comparison of the level detected in step S501 with a predefined threshold percentage (e.g., 20%) of the theoretical maximum signal level achievable by the boost term. As already mentioned, the theoretical maximum is defined according to the gain of the local oscillator LO 210, the conversion rate R of incident photons to electrons by the photodiode 230, and the gain of the transimpedance amplifier TIA 245. The conversion rate R of incident photons to electrons by the photodiode 230 is known to the control unit 280 by pre-configuration (e.g., at the factory). If the gain of the local oscillator LO 210 is not defined (i.e., controlled) by the control unit 280, the gain of the local oscillator LO 210 is known to the control unit 280 by pre-configuration (e.g., at the factory). If the gain of the transimpedance amplifier TIA 245 is not defined (i.e., controlled) by the control unit 280, the gain of the transimpedance amplifier TIA 245 is known to the control unit 280 by pre-configuration (e.g., at the factory). Alternatively, the theoretical maximum signal level achievable by the boost term is known to the control unit 280 by pre-configuration (e.g., at the factory). Still alternatively, the applicable threshold THdsp is known to the control unit 280 by pre-configuration (e.g., at the factory). In this latter case, the pre-configured applicable threshold THdsp maintains a pre-defined threshold percentage (e.g., 20%) of the theoretical maximum signal level achievable by the boost term, except that the effective value of the pre-defined threshold percentage (e.g., 20%) is not dynamically determined by the control unit 280. The pre-defined threshold percentage may change over time. For example, the control unit 280 may be instructed by an application layer of the coherent optical receiver 200 or a demodulator of the coherent optical receiver 200 that this change over time means that the signal-to-noise ratio of the electrical signal should be enhanced or else degraded.

[0101] In step S503, the control unit 280 checks the comparison that the level detected in step S501 is greater than or equal to the threshold value THdsp. If the level detected in step S501 is lower than the threshold value THdsp, step S504 is executed. If the level detected in step S501 is greater than or equal to the threshold value THdsp, step S501 is repeated. A waiting or standby time can be inserted before executing step S501 again. The duration of the waiting or standby time is less than the time required for the polarization of the carried optical signal to change by a predefined percentage in the polarization space, as observed for the optical fiber. For example, the duration of the waiting or standby time is 100 milliseconds.

[0102] In step S504, the control unit 280 commands a configuration change of the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211, and the configuration change changes the ellipse principal axis orientation φ l and / or the elliptical phase shift ψ l The at least one configuration change is selected from the above changes.

[0103] Then, step S501 is repeated. Preferably, no waiting time is applied at this point in time for the control unit 280 to determine whether another elliptical modification of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211 has to be performed.

[0104] As already mentioned, the domain switching procedure allows coarse control of the change in polarization state without effectively tracking the polarization state, fine control is provided by the phase refinement procedure disclosed below.

[0105] FIG. 6 illustrates diagrammatically the algorithm executed by the control unit 280 for adjusting the configuration of the coherent optical receiver 200 within the phase refinement procedure.

[0106] In step S601, the control unit 280 configures the polarization diversity actuator PDA 211 to set modulation parameters that define the controlled error signal. More specifically, the control unit 280 configures the phase modulator PM 301 with respect to the modulation parameters φ0 and Δφ0. With respect to the parameter Ω, this is set to correspond to a cycling period that is longer than the polarization time constant (on the order of milliseconds) but shorter than typical wavelength chirp or fast phase noise timescales (on the order of milliseconds to microseconds). A modulation of several hundred kHz can then be used.

[0107] First, φ0 is set to 0 and Δφ0 is set to a variation in Δφ corresponding to a predefined percentage (e.g., half or one-third of the threshold) of the threshold THprp that triggers a change in the ellipse parameters in the domain switching procedure, thereby limiting the probability that the photocurrent output of the DC filter 240 will drop to that threshold due to the phase refinement procedure.

[0108] In step S602, the control unit 280 tracks the photocurrent variations at the frequency Ω / 2π. More precisely, the control unit 280 tracks the voltage electrical signal variations at the output of the transimpedance amplifier TIA 245. The control unit 280 preferably performs integration over a predefined amount of data communication samples (e.g., hundreds of samples, such as 256 or 512, depending on the coding and interleaving robustness), which means that several tracking samples are obtained within a period of 2π / Ω (e.g., 8 or 16 samples, or several tens, or even 2 or 56 samples, depending on the noise introduced). The data communication samples should be understood as samples originating from the modulated optical signal transmitted by the coherent optical transmitter, and the tracking samples should be understood as samples originating from the modulated optical signal generated by the phase modulator PM 301.

[0109] According to a particular embodiment relying on synchronous detection, the control unit 280 monitors the evolution of the tracking samples acquired during the 2π / Ω period in view of the shape of the controlled error signal during said 2π / Ω period. In view of the definition of Ω, the signal propagation time in the coherent optical receiver 200 can be neglected, and the fluctuations of the controlled error signal and the implied fluctuations at the output of the transimpedance amplifier TIA 245 are considered to occur on the same time basis. Therefore, the control unit 280 analyzes the output of the transimpedance amplifier TIA 245 in parallel with the shape of the contribution of cos(Ωt+φ) of the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211.

[0110] According to the starting position of the phase modulation caused by the phase modulator PM 301 and its amplitude, the photocurrent output of the DC filter 240, and consequently the analog electrical signal from the transimpedance amplifier TIA 245, increases or decreases synchronously or exceeds a maximum value (peak) relative to the shape of the controlled error signal amplitude. In other words, the control unit 280 analyzes whether the analog electrical signal from the transimpedance amplifier TIA 245 increases or decreases synchronously or exceeds a maximum value (peak) in view of the controlled error signal amplitude fluctuating as a cosine wave between φ0-Δφ0 and φ0+Δφ0.

[0111] In step S603, the control unit 280 checks whether the analog electrical signal from the transimpedance amplifier TIA 245 exceeds a maximum value (peak). If so, step S605 is performed, otherwise step S604 is performed.

[0112] In step S604, the control unit 280 modifies the modulation parameter φ0 to at least start the modulation closer to the above-mentioned maximum value, and determines the value φ that results in the above-mentioned maximum value. maxThe control unit 280 increases or decreases the parameter φ0 by a predetermined step value, for example Δφ0, thereby moving the modulated signal towards the above-mentioned maximum value (peak). Typically, in synchronous detection, if the analog electrical signal from the transimpedance amplifier TIA 245 decreases and at the same time the cosine wave shape of the controlled error signal amplitude increases or vice versa, the modulation parameter φ0 must be decreased, and if the analog electrical signal from the transimpedance amplifier TIA 245 increases (or decreases) at the same time as the cosine wave shape of the controlled error signal amplitude, the modulation parameter φ0 must be increased. Then, step S603 is repeated for further monitoring variations at the frequency Ω / 2π.

[0113] In step S605, the control unit 280 sets the modulation parameter φ0 to φ max The phase is then fixed to φ as the starting point for a new phase refinement procedure. The phase is therefore adjusted to optimize the signal strength at the output of the transimpedance amplifier TIA 245. The phase refinement procedure then ends, the modulation parameter φ remains unchanged, and Δφ is set to 0 until the phase refinement procedure is re-initiated. When the phase refinement procedure is restarted, the control unit 280 fixes the modulation parameter φ to φ as the starting point for a new phase refinement procedure. max It is preferable to maintain the temperature at 0.5°C.

[0114] φ max is obtained and the polarization diversity actuator PDA 211 is configured accordingly, the maximum value of the photocurrent output by the DC filter 240, and consequently any fluctuations in the analog signal from the transimpedance amplifier TIA 245, result from changes in the axis of the polarization state of the optical signal. Therefore, with this phase refinement procedure, any phase detuning is mitigated independently of its characteristics (whether it is wavelength detuning, polarization change or phase shift) and, if several phase detuning phenomena occur simultaneously, they are mitigated simultaneously. Therefore, an optimal signal strength is obtained at the output of the transimpedance amplifier TIA 245, taking into account the orientation of the axis of the polarization state of the optical signal.

[0115] In a particular embodiment of step S605, the control unit 280 calculates φ max and reset the modulation parameter φ0 (fix φ0 to 0), l 3. In fact, the elliptical phase detuning Δψ between the optical signal output by the set formed by the local oscillator LO 210 and the polarization diversity actuator PDA 211 and the optical signal received from the coherent optical transmitter acts similarly to the wavelength detuning Δν and the phase detuning Δφ in both the polarization-induced contributions at the output of the DC filter 240. Therefore, any of the above detuning phenomena can be compensated by adjusting at least one of the others. Therefore, in the arrangement of FIG. 3, the mitigation of the static component of the phase detuning is transferred from the phase modulator PM 301 to the variable waveplate VW 302. This allows the dynamics of the phase modulator PM 301 to be preserved. In a more particular embodiment, the transfer of the mitigation of the static component of the phase detuning is triggered by the control unit 280 when the maximum value of the voltage at the output of the transimpedance amplifier TIA 245 does not vary beyond a predefined threshold THmax for a period corresponding to the bandwidth of the variable waveplate VW 302.

[0116] In a particular embodiment, the control unit 280 monitors the evolution of the phase detuning to detect the wavelength detuning and adjusts the wavelength configuration of the local oscillator LO 210 accordingly. The processing resources and complexity required to perform such monitoring are limited compared to the processing resources to track the polarization states. This aspect is presented below with respect to FIG. 7.

[0117] FIG. 7 illustrates diagrammatically the algorithm executed by the control unit 280 for adjusting the wavelength of the local oscillator LO 210 .

[0118] In step S701, the control unit 280 monitors the evolution of the phase detuning to detect the wavelength detuning. Therefore, in step S702, the control unit 280 checks whether a continuous and monotonic evolution of the phase detuning is present at the output of the transimpedance amplifier TIA 245. In fact, the wavelength drift of the laser is caused by the adiabatic evolution of the laser cavity. This can be caused by automodulation gain, thermal effects, etc. Adiabatic evolution often refers to a phenomenon having a time constant of the order of a few milliseconds to a few seconds or even more. Phase noise and frequency or wavelength detuning are distinguished from each other with reference to a time scale, i.e. the symbol duration. For this reason, any instability that is longer than the symbol duration is considered as a wavelength drift, whereas any instability that is shorter than the symbol duration or has a time constant of the order of the symbol duration is considered as a phase noise.

[0119] If the control unit 280 detects a continuous and monotonic evolution of the phase detuning, step S703 is executed, otherwise step S701 is repeated. A waiting period may be applied before repeating step S701.

[0120] In step S703, the control unit 280 adjusts the wavelength of the local oscillator LO 210 to compensate for the continuous and monotonic evolution of the detected phase detuning. Then, step S701 is repeated. A waiting period can be applied before repeating step S701.

[0121] Once the wavelength adjustment is made, the control unit 280 preferably restarts the phase refinement procedure. In a more particular embodiment, the control unit 280 preferably restarts the phase refinement procedure at time intervals (e.g., periodic time intervals) during the reconfiguration transient period (slow response) of the local oscillator LO 210. Thus, during the time required for the local oscillator LO 210 to effectively adapt the wavelength, the control unit 280 compensates for the transient wavelength evolution with overmodulation of the controlled error signal.

[0122] In a particular embodiment, when the domain switching procedure is initiated and when a phase refinement procedure is in progress, the control unit 280 stops and resets the phase refinement procedure. l and / or elliptical phase shift ψ l is accomplished. Therefore, the domain switching procedure has a higher priority than the phase refinement procedure, especially during the initialization of the coherent optical receiver 200.

[0123] FIG. 8 illustrates a schematic representation of a particular temperature sensor-based mechanism of the coherent optical receiver 200 .

[0124] In fact, the presence of a slow evolution of the wavelength detuning is often due to the evolution of the temperature of the laser chip. This occurs when there is no Peltier system to properly regulate the temperature of the laser chip. Therefore, in a particular embodiment, the local oscillator LO 210 is coupled with a temperature sensor TS 801 that captures the temperature of the local oscillator LO 210, typically the laser chip of the local oscillator LO 210. The temperature data is provided to the detuning manager DM 270, for example, via a link 802. Furthermore, the control unit 280 includes a look-up table that establishes a relationship between temperature drift and wavelength drift. The control unit 280 monitors the evolution of the temperature captured by the temperature sensor. According to the temperature drift indicated by the evolution of the temperature captured by the temperature sensor TS 801, the control unit 280 retrieves the corresponding wavelength drift from a look-up table and subtracts the effect of said wavelength drift against other effects affecting the phase by superimposing on the controlled error signal a contribution that is the opposite of the phase evolution caused by the wavelength drift, i.e., 2πrΔνt, for example over the [0,2π] domain. By doing this, the dynamics of the phase modulator PM 301 is preserved.

Claims

1. 1. A coherent optical receiver intended to receive an amplitude shift keying modulated optical signal from a coherent optical transmitter, comprising: A local oscillator; Ellipse principal axis orientation φ ｌ and the elliptical phase shift ψ ｌ a polarization diversity actuator configured to modify the optical signal output by the local oscillator to form an optical signal having an elliptical polarization of a 2x2 coupler, one input of which receives the amplitude shift keying modulated optical signal received from the coherent optical transmitter and another input of which receives another optical signal output by a set formed by the local oscillator and the polarization diversity actuator, so as to enable the local oscillator to provide a boost effect to the amplitude shift keying modulated optical signal received from the coherent optical transmitter, the 2x2 coupler further having one output connected to a set formed by a photodiode and a DC current filter placed after the photodiode for removing a continuous component of the analog electrical signal output by the photodiode; a transimpedance amplifier that converts the current electrical signal output by the DC current filter into a voltage electrical signal; A domain switching procedure is performed, and when the voltage electrical signal output by the transimpedance amplifier is equal to or smaller than a first predetermined threshold value THdsp corresponding to a predetermined percentage of a theoretical maximum value achievable by the boost effect, the ellipse major axis orientation φ ｌ and / or the elliptical phase shift ψ ｌ a control unit in the form of electronic circuitry configured to command a configuration change of the polarization diversity actuator by changing Equipped with the electronic circuitry is further configured to perform a phase refinement procedure to insert a controlled error signal into the phase of the optical signal output by the set formed by the local oscillator and the polarization diversity actuator and to adjust a static component of the controlled error signal towards a maximum voltage electrical signal output by the transimpedance amplifier.

2. During the domain switching procedure, the elliptical principal axis orientation φ ｌ The modification of is performed by applying a predefined shift of π / 4, where the elliptical phase shift ψ ｌ 2. The coherent optical receiver of claim 1, wherein the modification of is performed by applying a predefined shift of π / 2.

3. The polarization diversity actuator has an elliptical principal axis orientation φ ｌ and the elliptical phase shift ψ ｌ 3. A coherent optical receiver as claimed in claim 1, further comprising a variable waveplate which allows for the variation of

4. The electronic circuitry is configured to perform an automatic configuration of a standby period between successive executions of the domain switching procedure, the automatic configuration comprising: Monitoring of the evolution of the voltage electrical signal output by the transimpedance amplifier is first performed continuously within the domain switching procedure, A history of variations associated with the polarization state is tracked by said monitoring, and a period between successive commanded changes in the ellipse parameters is monitored; When the evolution of the polarization state variations has stabilized, the standby period between successive executions of the domain switching procedure is defined to be shorter than a stabilization period between successive crossings of the predetermined first threshold value THdsp. The coherent optical receiver according to any one of claims 1 to 3, comprising:

5. The electronic circuitry is configured to perform an automatic configuration of a standby period between successive executions of the phase refinement procedure, the automatic configuration comprising: monitoring the evolution of the voltage electrical signal output by the transimpedance amplifier is first performed continuously within the phase refinement procedure; A history of the time between successive crossings of a second predetermined threshold THprp is monitored; the standby period between successive executions of the phase refinement procedure is defined to be shorter than a stabilization period between successive crossings of the second predetermined threshold value THprp. The coherent optical receiver according to any one of claims 1 to 4, comprising:

6. A coherent optical receiver according to any one of claims 1 to 5, wherein the controlled error signal is a modulated signal.

7. The controlled error signal is a modulation on the phase of the optical signal output by the set formed by the local oscillator and the polarization diversity actuator, and has the following form: f ０ +Df ０ cos (Ωt+φ) where φ ０ 7. The coherent optical receiver of claim 6, wherein: is a static component of the controlled error signal.

8. During the phase refinement procedure, the electronic circuitry monitors the output variation of the transimpedance amplifier at a frequency defined by Ω and adjusts φ toward the maximum voltage electrical signal output by the transimpedance amplifier. ０ where φ ０ is φ ｍａｘ 8. The coherent optical receiver of claim 7, wherein:

9. The electronic circuit unit has a diameter of φ ｍａｘ The elliptical phase shift ψ ｌ Change and then change φ ０ By resetting the static component φ ０ 9. A coherent optical receiver as claimed in claim 7 or 8, configured to forward a signal for mitigation of

10. The electronic circuitry is adapted to adjust the maximum voltage electrical signal at the output of the transimpedance amplifier relative to the ellipse major axis orientation φ ｌ and allows the elliptical phase shift ψ ｌ If the static component φ does not vary by more than a predetermined third threshold THmax over a period corresponding to a bandwidth of a variable wave plate used in the polarization diversity actuator to allow the change of ０ 10. The coherent optical receiver of claim 9, configured to forward a signal for mitigation of

11. 11. A coherent optical receiver according to claim 1, wherein the electronic circuitry is configured to monitor an evolution of the phase detuning to detect a continuous monotonic evolution of the phase detuning and to adjust a wavelength configuration of the local oscillator to compensate for the detected continuous monotonic evolution of the phase detuning.

12. The coherent optical receiver of claim 11 , wherein once a wavelength adjustment has been performed, the electronic circuitry is configured to restart the phase refinement procedure.

13. 13. The coherent optical receiver of claim 12, wherein the electronic circuitry is configured to restart the phase refinement procedure at time intervals during a local oscillator reconfiguration transient.

14. 14. A coherent optical receiver according to claim 1, further comprising a temperature sensor capturing a temperature of the local oscillator and a look-up table establishing a relationship between temperature drift and wavelength drift, the electronic circuitry being configured to monitor the temperature evolution captured by the temperature sensor, to look up from the look-up table a wavelength drift corresponding to a temperature drift indicated by the temperature evolution, and to superimpose on the controlled error signal an opposite contribution of a phase evolution caused by the wavelength drift.

15. An optical network unit intended for use in a passive optical network, the optical network unit comprising a coherent optical receiver according to any one of claims 1 to 14, for receiving an amplitude shift keying modulated optical signal transmitted by a coherent optical transmitter included in an optical line terminal of the passive optical network.

Images