Robust coherent optical communication through fading channels using ultra-short pulses and dynamic coherent all-optical matched filter
Patent Information
- Application Number
- PCT/IL2025/050214
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2024-03-05
- Filing Date
- 2025-03-05
- Publication Date
- 2025-10-02
AI Technical Summary
Optical communication channels suffer from fading due to multipath interference, leading to power fluctuations and reduced link budget, especially in channels like long-range atmospheric and underwater communication, which affects high-order modulation schemes that rely on phase information.
Utilize Mode-Locked Lasers (MLL) for generating optical carriers and local oscillators, with dynamic all-optical coherent matched filters (DCOF) to adjust phase and amplitude, synchronize pulse repetition rates, and perform spectral phase manipulations to maximize temporal overlap and reduce power variations.
Enhances coherent communication through fading channels by reducing power fluctuations, improving link budget, and supporting high data rates, while maintaining phase information for high-order modulation schemes.
Smart Images

Figure IL2025050214_02102025_PF_FP_ABST
Abstract
Description
[0001] ROBUST COHERENT OPTICAL COMMUNICATION THROUGH FADING CHANNELS USING ULTRA-SHORT PULSES AND DYNAMIC COHERENT ALL-OPTICAL MATCHED
[0002] FILTER
[0003] Field of the Invention
[0004] The present invention relates to the field of optical communication. More particularly, the invention relates to a method for effectively performing coherent communication through fading channels.
[0005] Background of the Invention
[0006] Several optical communication channels, such as long-range atmospheric channels that are used to communicate between satellites and ground, may suffer from fading (where the magnitude and quality of propagating signals fluctuate as a function of time and distance). The multipath phenomenon generates the fading in such channels, such that coherent signals that are transmitted will reach the receiving antenna (optical or radio-frequency) at different times and, therefore, with phase difference. As a result of the phase difference, some signals will pass interference (which may be randomly constructive or distractive) that will deteriorate the optical beam. This interference causes sharp power degradation and great variance in the received power levels. As a result, in a channel that suffers from multipath, in order to keep a wide dynamic range that corresponds to the detectors in the receiver, the transmitted power should be significantly increased, and the analog front-end at the receiver needs to cope with fast varying and high dynamic range signals. For example, if the uncertainty is about 10 dB, the signal's transmitted power should be increased by 10 dB, and the receiver front-end should incorporate a high dynamic range high bandwidth automatic gain control circuitry. This constraint causes power waste. Signal fading occurs in multipath channels, e.g., free-space optics (FSO), causing substantial power variations in the received signal and hindering the link budget. Ultra-short pulses (UPS) mitigate the signal power variations compared to conventional transmission based on continuous wave (CW) signal carrier due to shortening the near temporal coherence. One of the existing solutions to the fading problem is to transmit the optical signals using a narrow pulsed laser, which has a very wide spectrum. The wide spectrum averages the fluctuations, such that the uncertainty regarding the received signal power is drastically reduced. However, averaging the fluctuations actually eliminates the effect of the signals' phase. Therefore, this solution is limited to communication that is based only on the signal's magnitude, and is not useful to modulation schemes with high spectral efficiency, such as high-order Quadrature Amplitude Modulation (QAM), where the phase information is critical.
[0007] Scintillation mitigation using ultra-short pulses (USP) has already been reported for intensity modulation and direct detection (IM / DD) systems. Nevertheless, it did not disclose how coherent modulation and detection can be done using the transmission of ultra-short pulses to enable robust optical communication through fading channels.
[0008] It is therefore an object of the present invention to provide a method for effectively performing coherent communication through fading channels, while decreasing the dynamic range of the signal's power levels and improving the optical link budget.
[0009] It is another object of the present invention to provide a method for effectively performing coherent communication through fading channels, while improving the optical link budget.
[0010] It is a further object of the present invention to provide a method for effectively performing coherent communication through fading channels, which supports high data rates.
[0011] Other objects and advantages of the invention will become apparent as the description proceeds.
[0012] Summary of the Invention A method for coherent transmission and reception of complex modulation constellations over an optical multipath fading channel, according to which a Mode-Locked Laser (MLL) is utilized for generating an optical carrier at the transmitter side (TX-MLL), for transmitting coherent signals over the multipath fading channel. A second MLL is utilized as a local oscillator (LO-MLL) at the receiver side. Spectral phase manipulations are performed by adjusting the phase of each combline of the LO-MLL, such that the overlapping temporal integral between the TX-MLL and the RX-MLL on the received signal will be maximized.
[0013] The modulating complex signals may be any QAM symbols. The transmitted signals may have dual or single polarization.
[0014] The optical multipath fading channel may be atmosphere, underwater, multimode fibers, multicore fibers, and biological tissues.
[0015] A TX-MLL ultra-short temporal coherence length may be utilized to reduce the statistically coherent interference between the different delayed replicas from the multipath fading channel, to thereby reduce the received optical power variations.
[0016] In one aspect, all-optical coherent match filtering adaptive phase and / or amplitude manipulations (phase / amplitude mask) of the different comblines of the LO-MLL are performed, to thereby generate temporal LO-MLL pulse optimization to maximize the temporal overlapping integral between the received distorted pulses of the TX-MLL waveform and the LO-MLL pulses.
[0017] A Dynamic Coherent All-Optical Matched Filter (DCOF) may be implemented for channel tracking that improves the receiver sensitivity, by continuously changing the phases and / or the amplitude of the LO-MLL comblines modes to maximize the temporal overlapping integral between the received distorted pulses of the TX-MLL waveform and the LO-MLL pulses. The pulse repetition rate (PRR) of the two pulse trains may be synchronized in parallel to the DCOF, according to an optimization parameter for both the DCOF optimization and / or the PRR synchronization.
[0018] The optimization parameter may be the detected analog power at the output of the coherent receiver, or other parameters extracted by the modem, selected from the group of:
[0019] - the received signal strength indicator (RSSI);
[0020] - the signal-to-noise ratio (SNR);
[0021] - the error-vector magnitude (EVM);
[0022] - the bit-error-rate (BER) and others.
[0023] The DCOF may be applied to the received signal of the Tx-MLL, rather than to the LO-MLL, to maximize its temporal overlapping integral with the LO-MLL pulses, thereby attaining a higher spectral resolution.
[0024] A phase encoder may be used to encode the signal at the Tx-MLL before transmission, and transmitting a pre-distorted signal, thereby attaining a higher spectral resolution without requiring using a feedback channel.
[0025] Whenever the transmitted signal at the Tx-MLL is encoded, the DCOF may be applied to the transmitted signal at the Tx-MLL side rather than to the LO-MLL, using a feedback channel, to maximize its temporal overlapping integral with the LO-MLL pulses, instead of encoding unmodulated comblines of the LO-MLL, thereby allowing a higher spectral resolution to be attained.
[0026] An independent spatial diversity adaptive optics may be added, including as receiver-side coherent beam combing, maximal combing ratio, wavefront corrections with deformable mirrors, or spatial light modulators, and waveform correction following spatial mode decomposition. The gain of the spatial diversity adaptive optics may be added to the existing gain, as the spatial domain, which is taken care of by the AO, and the temporal domain, which is taken care of by the DCOF, are orthogonal.
[0027] A coherent optical communication system, comprising: a) a pulsed coherent transmitter, which comprises: a.l) a Mode-Locked Laser (MLL) for generating an optical carrier at the transmitter side (TX-MLL), for transmitting pulsed coherent signals over the multipath fading channel; a.2) a first Spatial Light Modulator for equalizing the spectrum of the MLL to accommodate the desired spectral envelop shape and to resolve a certain number of comblines; a.3) a Dual-Parallel Mach-Zehnder Modulator having in-phase (I) and quadrature (Q) components representing Quadrature Amplitude Modulation (QAM) of two orthogonal polarizations, for modulating the pulsed coherent signals; a.4) an amplifier for amplifying the pulsed coherent signals, to be transmitted through an FSO channel; b) an all-optical matched filter coherent receiver implementing a DCOF which comprises: b.l) a local oscillator; b.2) a second Spatial Light Modulator for continuously performing Spectral Phase Encoding (SPE), to maximize the LO pulses beating with the received pulses; b.3) a DCOF microcontroller (pC) for locking the Pulse Repetition Rate (PRR) of the local oscillator (LO) MLL to the PRR of the received pulses; b.4) an ICR consisting of a hybrid mixer for outputting an in-pase (I) and quadrature (Q) components representing Quadrature Amplitude Modulation (QAM) of two orthogonal polarizations, a Beam-Splitter (BS) and Polarizing Beam-Splitter (PBS), for splitting the signal and the LO into two orthogonal states of polarization, respectively; and b.5) an analog-to-digital converter (ADC) followed by a DSP, for reconstructing the baseband spectrum.
[0028] Brief Description of the Drawings
[0029] The above and other characteristics and advantages of the invention will be better understood through the following illustrative and non-limitative detailed description of preferred embodiments thereof, with reference to the appended drawings, wherein:
[0030] Fig. 1 shows the temporal coherent function and coherence time;
[0031] Figs. 2a-2b illustrate scintillation mitigation, according to an embodiment of the invention;
[0032] Fig. 3 is a functional block diagram of a dynamic coherent all-optical matched filter system for free-space optics communication;
[0033] Fig. 4 illustrates MLLs Locking, according to an embodiment of the invention;
[0034] Fig. 5 shows a dynamic coherent all-optical matched filter detection operation;
[0035] Fig. 6 shows a dynamic coherent all-optical matched filter detection operation;
[0036] Fig. 7 shows the error vector magnitude (EVM) performance;
[0037] Figs. 8a shows the Bit-Error Ratio (BER) performance; and
[0038] Fig. 8b shows the data of Fig. 8a, displayed as a histogram.
[0039] Detailed Description of the Present Invention
[0040] The present invention provides a method for effectively performing coherent communication through fading channels, while decreasing the dynamic range of the signal's power levels and improving the optical link budget. In the proposed system and method, a robust coherent optical communications (coherent transmission and reception of complex constellations) through optical multipath and dispersive (fading) channels, such as the atmosphere, underwater medium, multimode fibers (fibers with various paths, or modes, in which light can travel through optical fibers), multicore fibers (optical fibers that contain multiple cores or light guiding cores within a single strand of optical fiber), and biological tissues is enabled with pulsed transmission and Dynamic All-Optical Coherent Matched Filter (DCOF). The proposed system and method generate both the optical signal carrier and the Local Oscillator (LO) from Mode-Locked Laser (MLL) sources. The Dynamic All-Optical Coherent Matched Filter is implemented on the receiver side by optimizing the phase and / or amplitude of the LO MLL by performing phase manipulations, according to the channel distortion. In parallel, the Pulse Repetition Rate (PRR) of the LO MLL is synchronized to the PRR of the signal carrier MLL. The optimization parameter can be either the detected analog power at the output of the coherent receiver or other parameters extracted by the modem, such as the Received Signal Strength Indicator (RSSI), the signal-to-noise ratio (SNR), the Error Vector Magnitude (EVM - a metric to quantify the combination of all signal impairments in a system, mainly for devices that use digital modulation, which can be represented through a constellation diagram. EVM is calculated by finding the ideal constellation location for each received symbol, where the RMS of all error vector magnitudes between the received symbol locations and their closest ideal constellation locations constitute the EVM value), the Bit-Error Ratio (BER), etc. The concept and the method proposed here can incorporate polarization diversity and any optical modulation that uses the phase and / or the amplitude of the optical field, including high-order modulation schemes.
[0041] In addition, the method inherently mitigates chromatic dispersion up to a single pulse duration, which is known to affect FSO. Spatial diversity adaptive optics, such as waveform sensing and correction, or receiver-side Coherent Beam Combing (CBC), where the receiver's aperture is tiled to many sub-apertures, which are then coupled to fibers and coherently combined, can be accommodated on top of the proposed method synergetically to increase the receiver sensitivity further. Another notable merit of the proposed method is the inherent estimation of the channel's response, allowing the estimation of atmospheric parameters in the FSO system. Another application of such systems is clock distribution. We propose two flavors of receiver implementation based on a dynamic all-optical matched filter. In the first flavor, the LO is adjusted to maximize its interaction with the received signal. The received signal is adjusted to optimize its interaction with the LO in the second flavor.
[0042] The proposed system and method enable robust coherent optical communications through multipath and dispersive channels with pulsed transmission and dynamic all-optical coherent matched filter (DCOF). In the proposed system and method, both the signal carrier and the Local Oscillator (LO) are generated from Mode-Locked Laser (MLL - a type of laser source laser which emits ultrashort pulses, based mode locking techniques for generating ultrashort pulses in lasers) sources.
[0043] A dynamic all-optical matched filter is implemented on the receiver side by optimizing the phase and / or amplitude of the LO MLL according to the level of channel distortion. In parallel, the pulse repetition rate (PRR) of the LO MLL is synchronized to the PRR of the signal carrier MLL.
[0044] In the first implementation, the LO is adjusted to maximize its interaction with the received signal. This implementation uses the same phase encoding as the LO described in international patent application No. WO 2021 / 009754 (which has originally been used for encryption), but for the purpose of adjusting the phases of the different modes to correspond to the phases of different signals that passed the fading channel (since each mode is received with a different phase, due to the multipath phenomenon which caused different portions of the optical beam to arrive at different times). This is done by using a different phase modulator per each mode, in order to match the phases of different modes in the LO to the equivalent phases of the received signal.
[0045] In the time domain, the laser pulse is spread due to the fact that in this implementation, the phases are unlocked. Therefore, matching in the time domain causes the spread pulses of the adjusted LO to correspond to the spread pulses of the faded received signals. After matching, for each spread pulse of each mode of the received signal, a corresponding spread pulse is selected in the LO. By doing so, the phase mask of the received signal is accurately matched to the phase mask of the LO, such that in the time domain, the two waveforms overlap.
[0046] This overlap is very important to perform effective coherent detection, since the value of the overlapping integral between the two spread pulsed of each mode is maximized. This way, coherent detection with a small variance is obtained and the coherent data can be reconstructed, with performance that is similar to direct detection. Also, the proposed method allows using both amplitude and phase polarizations.
[0047] In the second implementation, the received signal is adjusted to optimize its interaction with the LO, which remains mode-locked.
[0048] Other implementations are also possible, provided that all the used manipulations (adjustments) will be done such that the overlapping integral on the received signal will be maximized.
[0049] The adaptive adjustment of the LO pulses must be done in a rate which is substantially higher than the changes in the fading signal.
[0050] Fig. 1 shows the temporal coherent function and coherence time. The Generalized Temporal Coherence Function (GTCF) is obtained for a 40 GHz Pulse Repetition Rate (PRR) or 25 picoseconds Pulse Repetition Interval (PRI - the time interval between pulses) mode-locked laser (MLL) with a varying number of comblines. The term generalized is used in conjunction with the temporal coherence function to emphasize that it applies to the E-field of the MLL after being modulated. The GTCF is provided by the normalized autocorrelation function (ACF). The GTCF is computed for the optical pulses train, modulated at 32.78 GBaud with Quadrature Phase Shift Keying (QPSK) symbols stream, that was shaped with Root-Raised Cosine (RRC - a transmit / receive filter in a digital communication system to perform matched filtering) zero-lnterSymbol Interference (ISI - a form of distortion of a signal in which one symbol interferes with subsequent symbols) filter. For a continuous wave (CW) transmission, provided by a single combline and plotted with dotted curve 10, the coherence time, i.e., the Full Width at Half Maximum (FWHM - a statistical measure used to describe the width of a normal distribution or Gaussian distribution) of the GTCF, is 15 picoseconds. The coherence time reduces linearly while increasing the number of comblines. For 41 equalized comblines comprising a flat emission spectrum of 1.67 THz, the coherence time is around 360 femtoseconds.
[0051] Figs. 2a-2b illustrate scintillation mitigation, according to an embodiment of the invention. The power variations were recorded with a Direct Detection (DD) receiver (can only detect the amplitudes of the signals) for a Ultrashort Pulse (USP) transmission (an electromagnetic pulse whose time duration is of the order of a picosecond or less) through a turbulent atmosphere. The free-space optical (FSO) propagation is modeled as a multipath interference channel with the maximum allowed delay between the ballistic beamlet (a small beam of light) and most meandering beamlet, rmax, was varying between 4% to 48% of 25 picoseconds Pulse Repetition Interval (PRI), resembling 5 km of mild-to-severe atmospheric propagation. This model addresses the regime where rmax » A.c, the central wavelength of the Mode-Locked Laser (MLL). Consequently, four different channel types were evaluated, represented by Fig. 3a dashed lines. A constant power signal, with different numbers of comblines, was transmitted through these channels. Each data point represents a statistical computation of the scintillation loss extracted from 100 Monte-Carlo (a broad class of computational algorithms that rely on repeated random sampling to obtain numerical results) channel realizations. The scintillation loss is the reciprocal of the scintillation transmittance, computed for 99.9% availability on the Complementary Cumulative Distribution Function (CCDF - a statistical power measurement that provides a deep understanding of signal behavior) curve. Expressly, 99.9% of the realizations of this channel and per a specific number of comblines yielded a detected power with a better scintillation loss stated. It can be seen that 41 comblines transmission improves the link budget (a calculation that quantitatively assesses whether a communications link will perform successfully) with 10.5 dB compared to a Continuous Wave (CW) transmission. Fig. 2b provides additional insight into the availability improvement via the Probability Density Function (PDF - an expression used in statistics that defines the probability that some outcome will occur), it can be observed that increasing the number of comblines to 41 reduces scattering, leading to a more concentrated distribution around the 13 dB average scintillation loss. The narrowing of the PDF due to Ultra-Short Pulse (USP) transmission of 41 comblines under turbulent channel indicates that less power is required to achieve 99.9% availability. Therefore, the link budget can be dramatically improved by 10.5 dB compared to a CW conventional transmission.
[0052] Fig. 3 is a functional block diagram of dynamic coherent all-optical matched filter system for free-space optics communication. The concept system for robust coherent Free-Space Optical (FSO) communication using a Dynamic Coherent All-Optical Matched Filter (DCOF) is depicted. The transmitter comprises a Mode-Locked Laser (MLL) that generates an optical carrier of Ultra-Short Pulse (USP). A first Spatial Light Modulator (SLM - a device that can control the intensity, phase, or polarization of light in a spatially varying manner) device equalizes the MLL's spectrum to accommodate the desired spectral envelop shape and to resolve a certain number of comblines, to distinguish between different frequencies in the signal). The USP train then interacts inside a Dual-Parallel Mach-Zehnder Modulator (DP- MZM - two MZMs in parallel) with four data streams: in-phase (I) and quadrature (Q) components representing Quadrature Amplitude Modulation (QAM) of two orthogonal polarizations.
[0053] The modulated pulsed signal is then amplified by an Erbium Doped Fiber Amplifier (EDFA - an optical amplifier with erbium ions added to the core of the optical fiber) and transmitted through an FSO channel. The receiver includes an Integrated Coherent Receiver (ICR - monolithically integrated coherent receiver chip which can be used to receive high speed coherent optical signals) and implementation of the DCOF scheme. As part of the ICR frontend, beamsplitter (BS) and polarizing beamsplitter (PBS) split the signal and the LO into two orthogonal states of polarization, respectively. The DCOF microcontroller (pC) provides two major functions: locking the Pulse Repetition Rate (PRR) of the local oscillator (LO) MLL to the PRR of the received pulses. This is done by driving either the LO MLL Piezoelectric Transducer (PZT - a device that uses the piezoelectric effect to measure changes in pressure, acceleration, temperature, strain, or force by converting them to an electrical charge), gain current, or a Saturable Absorber (SA - an optical component with a certain absorption loss for light, which is reduced at high optical intensities). The Spectral Phase Encoding (SPE) is also done by the DCOF pC continuously to maximize the LO pulses beating with the received pulses. The SPE is done using a second SLM device, which should operate at MHz rates for adjusting tens of comblines faster than the dynamics of the atmosphere. For performance monitoring, Signal-To-Noise-Ratio (SNR) and Bit-Error Ratio (BER) are extracted from the Digital Signal Processor (DSP). The DSP considered in this scheme is a conventional coherent communication modem. Inset (a) represents the electrical analog baseband information signal at the output of the Digital-to-Analog Converter (DAC), a 33 GBd quadrature phase shift keying (Q.PSK) signal, shaped with a Root-Raised Cosine (RRC) filer. Insets (b) and (c) represent the transmitted and received 1.67 THz spectra, respectively, comprising 41 comblines at 40 GHz PRR. The received spectrum suffers from heavy fading, pronounced as a random notches pattern, caused by multipath interference. The reconstructed baseband spectrum, at the outputs of the analog-to-digital converter (ADC), is shown in Inset (d), after operation of the DCOF, indicating minimal impairments although the aggressive channel's distortion. Insets (e)-(g) demonstrate preliminary experimental results of the transmitter subsystem. The raw emission (e), the equalized (f), and the modulated (g) spectra are measured with 150 MHz resolution bandwidth (RBW) Optical Spectrum Analyzer (OSA).
[0054] In order to reduce the received optical power variations, a TX-MLL short temporal coherence length is utilized to reduce the statistically coherent interference between the different delayed replicas from the multipath fading channel.
[0055] In order to maximize the temporal overlapping integral between the received distorted pulses of the TX-MLL waveform and the LO-MLL pulses, all-optical coherent match filtering adaptive phase and / or amplitude manipulations (phase / amplitude mask) of the different comblines of the LO-MLL is performed, In one embodiment, channel tracking that improves the receiver sensitivity is performed by implementing a dynamic coherent all-optical matched filter (DCOF), for continuously changing the phases and / or the amplitude of the LO-MLL comblines modes to maximize the temporal overlapping integral between the received distorted pulses of the TX-MLL waveform and the LO-MLL pulses.
[0056] In one embodiment, in parallel to the DCOF, the pulse repetition rate (PRR) of the two pulse trains is synchronized according to an optimization parameter for both the DCOF optimization and / or the PRR synchronization.
[0057] In one embodiment, a higher spectral resolution without requiring using a feedback channel is attained by using a phase encoder to encode the signal at the Tx-MLL before transmission, and transmitting a pre-distorted signal.
[0058] A higher spectral resolution is attained by applying the DCOF to the transmitted signal at the Tx-MLL side rather than to the LO-MLL, whenever the transmitted signal at the Tx-MLL is encoded. This is done by using a feedback channel, to maximize its temporal overlapping integral with the LO-MLL pulses, instead of encoding unmodulated comblines of the LO-MLL,
[0059] An independent spatial diversity adaptive optics is added, such as receiver-side coherent beam combing, maximal combing ratio, wavefront corrections with deformable mirrors, or spatial light modulators, and waveform correction following spatial mode decomposition.
[0060] The receiver sensitivity is further increased by adding the gain of the spatial diversity adaptive optics to the existing gain.
[0061] Fig. 4 illustrates MLLs Locking, according to an embodiment of the invention. Experimental measurements and characterization of the two mode-locked lasers (MLLs), Tx-MLL and the LO-MLL, locking using beating envelop detection and Proportional-lntegral-Derivative (PID) controller (feedback-based control loop mechanism commonly used to manage processes that require continuous control and automatic adjustment). For the sake of this experiment, a static channel was emulated with a spectral phase encoder based on a liquid crystal on silicon (LCoS) spatial light modulator (SLM). In Fig. 4a, three stages of the dynamic coherent all-optical matched filter (DCOF) and MLLs pulse locking are provided. When the DCOF is disabled, and the MLLs are unlocked, the beating envelope (frequency of the "envelope" or general appearance of the wave) generates distorted pulses where each spans the entire Pulse Repetition Interval (PRI). This is indicated by the solid line 11. The curve 12 is generated after successfully operating the DCOF, producing a beating envelope that coincides with the theoretical results obtained for a fully compensated pulse. In addition, the pulse repetition rate (PRR) mismatch (APRR) is pronounced. This measurement captures an instantaneous APRR of about 900 Hz. Once the MLL locking is enabled, a constant output is generated, indicated by the dotted curve 13. The PID set-point is tuned to reach the peak power. In Fig. 4b, a locked MLLs frequency drift is calculated for a 200-second snapshot. Throughout this acquisition, and for many hours of continuous operations, no "glitches" were observed in the locking. Such indication cannot be obtained from an oscilloscope but rather from the information module that provides erroneous bit count. The maximal deviation in the APRR is below 40 Hz, which is E-9 of the 40 GHz native PRRs, indicating excellent locking. In Fig. 4c, the Allan variance (AVAR - a measure of frequency stability in clocks, oscillators and amplifiers) is calculated. A ten-second averaging interval indicates a frequency deviation of approximately 10-15 Hz.
[0062] Fig. 5 shows the dynamic coherent all-optical matched filter (DCOF) resultant power and the local oscillator (LO) waveform evolution. The main chart on the left depicts the power detected at the output of the coherent receiver relative to the transmitted power. The loss comprises the scintillation loss and the DD penalty. The DD penalty was recorded during the optimization of the DCOF for a 41 comblines system. A Greedy algorithm was applied and hence, the power keeps monotonically increasing, asymptotically approaching the DD bound, which is determined in the scintillation loss. Figs. 5a-5c represent the temporal evolution of the local oscillator (LO)'s waveforms. In each waveform, the same three pulses are captured. Line 14 is the modulated received pulses, and line 15 is the LO pulses. Fig. 5a plots the waveform after minimal optimizations 1 / 4. Fig. 5b plots the waveform after 21 / 41 comblines were optimized. Fig. 5c plots the final waveform after 41 / 41 comblines were optimized. The algorithm ends up with less than 2 dB of DD penalty, which means that about 65% of the received power is successfully exploited for coherent detection. The DD bound indicates an asymptotic measure for power collection per a given channel.
[0063] Fig. 6 shows a dynamic coherent all-optical matched filter detection operation. The performance of the DD penalty for 100 Monte Carlo fading channels with rmax =12 ps. The mean values, standard deviation, and relative standard deviation compared to CW were calculated and displayed in the legend. The standard deviation decreases while the number of comblines increases, indicating improved detection stability. For 41 comblines, the average DD penalty is 2.41 dB with low variations.
[0064] Fig. 7 shows the Error Vector Magnitude (EVM) performance. The EVM was recorded for 100 Monte Carlo fading channels with rmax = 12 ps. The mean values, standard deviation, and relative standard deviation compared to CW were calculated and displayed in the legend. The relative standard deviation decreases while increasing the number of comblines, indicating improved detection stability. The EVM stability improvement of 41 comblines exceeds 11 dB compared to CW.
[0065] Figs. 8a shows the bit-error ratio (BER) performance, recorded for 100 Monte-Carlo fading channels with rmax = 12 ps. The mean values and availability were calculated and displayed in the legend. The measurement limit, denoted by dotted line 16 determined by the number of bits transmitted in each iteration, is 3.05e"5. Values below this threshold mean that no errors were recorded in this acquisition. A dashed horizontal line 17 denotes the preforward error correction (FEC) BER threshold.
[0066] Fig. 8b shows the same data, displayed as a histogram. The BER threshold, indicated by dashed line 18, is turned vertically. The BER of the CW transmission is equally distributed on both sides of the BER threshold, indicating poor availability of 57%. On the other end, it is clearly noticed that the 41 comblines transmission shifts the BER scattering to the left, with 95% availability recorded. The above examples and description have of course been provided only for the purpose of illustration, and are not intended to limit the invention in any way. As will be appreciated by the skilled person, the invention can be carried out in a great variety of ways, employing more than one technique from those described above, all without exceeding the scope of the claims.
Claims
Claims1. A method for coherent transmission and reception of modulated signals having complex modulation constellations, over an optical multipath fading channel, comprising: a) utilizing a Mode-Locked Laser (MLL) for generating an optical carrier at the transmitter side (TX-MLL), for transmitting coherent signals over said multipath fading channel; b) utilizing an MLL as a local oscillator (LO-MLL) at the receiver side; and c) performing phase manipulations by adjusting the phase of each mode of said LO-ML, such that the overlapping temporal integral between the TX-MLL and the RX-MLL on the received signal will be maximized.
2. A method according to claim 1, wherein the modulated complex signals are any QAM symbols.
3. A method according to claim 1, wherein the coherent transmitted signals have dual or single polarization.
4. A method according to claim 1, wherein the optical multipath fading channel are atmosphere, underwater medium, multimode fibers, multicore fibers, and biological tissues.
5. A method according to claim 1, wherein a TX-MLL short temporal coherence length is utilized to reduce the statistically coherent interference between the different delayed replicas from the multipath fading channel, to thereby reduce the received optical power variations.
6. A method according to claim 1, comprising performing all-optical coherent match filtering adaptive phase and / or amplitude manipulations (phase / amplitude mask) of the different comblines of the LO-MLL, thereby generating temporal LO-MLLpulse optimization to maximize the temporal overlapping integral between the received distorted pulses of the TX-MLL waveform and the LO-MLL pulses.
7. A method according to claim 1, wherein a dynamic coherent all-optical matched filter (DCOF) is implemented for channel tracking that improves the receiver sensitivity, by continuously changing the phases and / or the amplitude of the LO- MLL comblines modes to maximize the temporal overlapping integral between the received distorted pulses of the TX-MLL waveform and the LO-MLL pulses.
8. A method according to claim 1, wherein in parallel to the DCOF, synchronizing the pulse repetition rate (PRR) of the two pulse trains according to an optimization parameter for both the DCOF optimization and / or the PRR synchronization.
9. A method according to claim 1, wherein the optimization parameter is the detected analog power at the output of the coherent receiver, or other parameters extracted by the modem, selected from the group of:- the received signal strength indicator (RSSI);- the signal-to-noise ratio (SNR);- the error-vector magnitude (EVM);- the bit-error-rate (BER) and others.
10. A method according to claim 1-9, wherein the DCOF is applied to the received signal of the Tx-MLL, rather than to the LO-MLL, to maximize its temporal overlapping integral with the LO-MLL pulses, thereby attaining a higher spectral resolution.
11. A method according to claim 1, further comprising using a phase encoder to encode the signal at the Tx-MLL before transmission, and transmitting a pre-distorted signal, thereby attaining a higher spectral resolution without requiring using a feedback channel.
12. A method according to claim 1, wherein whenever the transmitted signal at the Tx- MLL is encoded, the DCOF is applied to the transmitted signal at the Tx-MLL side rather than to the LO-MLL, using a feedback channel, to maximize its temporal overlapping integral with the LO-MLL pulses, instead of encoding unmodulated comblines of the LO-MLL, thereby allowing a higher spectral resolution to be attained.
13. A method according to claim 1, further comprising adding an independent spatial diversity adaptive optics, such as receiver-side coherent beam combing, maximal combing ratio, wavefront corrections with deformable mirrors, or spatial light modulators, and waveform correction following spatial mode decomposition.
14. A method according to claim 1, further comprising adding the gain of the spatial diversity adaptive optics to the existing gain, to further increase the receiver sensitivity.
15. A coherent optical communication system, comprising: a) a pulsed coherent transmitter, which comprises: a.l) a Mode-Locked Laser (MLL) for generating an optical carrier at the transmitter side (TX-MLL), for transmitting pulsed coherent signals over said multipath fading channel; a.2) a first Spatial Light Modulator for equalizing the spectrum of said MLL to accommodate the desired spectral envelop shape and to resolve a certain number of comblines; a.3) a Dual-Parallel Mach-Zehnder Modulator having in-phase (I) and quadrature (Q) components representing Quadrature Amplitude Modulation (QAM) of two orthogonal polarizations, for modulating said pulsed coherent signals; a.4) an amplifier for amplifying said pulsed coherent signals, to be transmitted through an FSO channel;b) an all-optical matched filter coherent receiver implementing a DCOF which comprises: b.l) a local oscillator; b.2) a second Spatial Light Modulator for continuously performing Spectral Phase Encoding (SPE), to maximize the LO pulses beating with the received pulses; b.3) a DCOF microcontroller (pC) for locking the Pulse Repetition Rate (PRR) of said local oscillator (LO) MLL to the PRR of the received pulses; b.4) an ICR consisting of a hybrid mixer for outputting an in-pase (I) and quadrature (Q) components representing Quadrature Amplitude Modulation (QAM) of two orthogonal polarizations, a Beam-Splitter (BS) and Polarizing Beam-Splitter (PBS), for splitting the signal and the LO into two orthogonal states of polarization, respectively; and b.5) an analog-to-digital converter (ADC) followed by a DSP, for reconstructing the baseband spectrum.