A fiber sensing and communication fusion method based on spectrum shaping

Abstract

The application discloses a kind of fiber sensing integration methods based on spectrum shaping, method includes: to the continuous communication signal is spectrum shaping, in the spectrum of communication signal, low-energy trap wave region is actively structured;Generate linear frequency modulation pulse as sensing signal, and the spectrum of sensing signal is accurately mapped to trap wave region, so that sensing signal and communication signal are orthogonal in frequency domain;After the spectrum shaping of communication signal and sensing signal are superimposed in electric domain, obtain composite electrical signal;Composite electrical signal is modulated to optical carrier, forms sensing integration optical signal and injects optical fiber;Receive the forward optical signal after transmission by optical fiber, recover communication signal by direct detection;Receive the back Rayleigh scattering echo returned by optical fiber, realize the compression of sensing pulse after the matching filter of echo and local light coherent detection, which realizes the integration of high-speed data transmission and high-precision distributed sensing in optical network.

Description

Technical Field

[0001] This invention belongs to the field of optical fiber sensing technology, and particularly relates to an optical fiber sensing fusion method based on spectrum shaping. Background Technology

[0002] Optical fiber communication and distributed optical fiber sensing technology are two important branches of the optical information field. With the rapid development of 5G / 6G, data centers, and smart infrastructure, simultaneously achieving high-speed data transmission and distributed environmental sensing (i.e., sensor fusion) on the same optical fiber link has become an important direction for reducing deployment costs and improving resource utilization. Intensity Modulation Direct Detection (IMDD) systems are widely used in optical interconnects due to their simple structure and low power consumption, while Distributed Acoustic Sensing (DAS) technology utilizes backscattering Rayleigh scattering in optical fibers to achieve long-distance, high-resolution vibration monitoring. Integrating these two technologies on the same optical fiber and the same signal carrier holds promise for building intelligent optical networks that combine communication and sensing capabilities.

[0003] Existing inductive-sense fusion schemes mainly employ two types of techniques. The first is the power superposition method, which involves directly superimposing the communication signal and high-energy sensing pulses in the electrical or optical domain using a coupler and then injecting the result into an optical fiber. The receiving end then separates the two signals through filtering or digital signal processing. The second is the time-division multiplexing method, which reserves time gaps in the communication data stream. During these gaps, the communication signal is temporarily reduced or shut off, and a high-peak sensing pulse is inserted, thereby isolating the communication and sensing functions in time. In addition, some schemes also employ high-pass filters or complex interference cancellation algorithms (such as the A1 / A2 algorithm) at the receiving end to suppress inductive-sense interference.

[0004] However, the aforementioned existing methods have significant drawbacks. The power superposition method causes overlap between the communication signal and the sensing signal in the time, frequency, or power domains, leading to severe mutual interference, increased communication error rate, and decreased sensing signal-to-noise ratio. The time-segmentation method causes a significant drop in the system's average power due to the discontinuous transmission of the communication signal, resulting in an extremely high peak-to-average power ratio (PAPR). This forces the Mach-Zehnder modulator (MZM) into the nonlinear saturation region and induces transient gain distortion in the erbium-doped fiber amplifier (EDFA), causing communication signal distortion. Post-processing algorithms at the receiver (such as high-pass filtering and A1 / A2) not only increase system complexity and power consumption but also fail to fundamentally address the disruption of communication linearity caused by high-peak sensing pulses. Summary of the Invention

[0005] In view of this, the present invention provides a fiber optic sensing fusion method based on spectrum shaping to solve the problems of mutual interference between communication and sensing signals and nonlinear distortion caused by excessively high peak-to-average power ratio of the system.

[0006] To achieve the above objectives, the technical solution of this invention is to provide a fiber optic sensing fusion method based on spectrum shaping, comprising: performing spectrum shaping on a continuous communication signal to actively construct a low-energy notch region in the spectrum of the communication signal; generating a linear frequency modulated pulse as a sensing signal and accurately mapping the spectrum of the sensing signal onto the notch region so that the sensing signal and the communication signal are orthogonal in the frequency domain; superimposing the spectrum-shaped communication signal and the sensing signal in the electrical domain to obtain a composite electrical signal with a continuous power envelope; modulating the composite electrical signal onto an optical carrier to form a sensing-integrated optical signal and injecting it into an optical fiber; receiving the forward optical signal transmitted through the optical fiber, recovering the communication signal through direct detection, and demodulating the communication data; receiving the backscattered Rayleigh echo returned from the optical fiber, performing pulse compression and subsequent demodulation on the echo through coherent detection and matched filtering to recover the distributed sensing information.

[0007] In one embodiment, the spectral shaping of the continuous communication signal includes: processing the communication signal using minimum operating numbers and encoding to generate a notch region near DC.

[0008] In one embodiment, before injecting the integrated sensing optical signal into the optical fiber, the integrated sensing optical signal is frequency-shifted by an acousto-optic modulator (AOM) to move the notch region to a location in the optical fiber link that is intended to interfere with the signal.

[0009] In one embodiment, generating a linear frequency modulated pulse as a sensing signal includes: determining the time domain width and peak power of the linear frequency modulated pulse according to a preset sensing detection energy requirement; wherein, by increasing the time domain width to reduce the peak power, the power fluctuation range of the composite electrical signal applied to the modulator is within the linear operating range of the modulator.

[0010] In one embodiment, a composite electrical signal with a continuous power envelope has no actively inserted time gaps on the time axis, and its peak-to-average power ratio is lower than a preset threshold.

[0011] In one embodiment, the demodulated communication data includes performing a baseband drift cancellation operation on the directly detected electrical signal to remove DC offset.

[0012] In one embodiment, the pulse compression of the echo through coherent detection and matched filtering includes: performing cross-correlation operation between the echo signal and a locally stored linear frequency modulated pulse template to compress the wide pulse into a narrow pulse; wherein, the larger the sweep bandwidth, the narrower the compressed pulse width.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: By performing spectral shaping on the communication signal to construct a low-energy notch region, and generating a linear frequency-modulated pulse as the sensing signal to precisely fill this notch region, the communication and sensing signals achieve orthogonal superposition in the frequency domain, fundamentally eliminating mutual interference between them. This improves the signal-to-noise ratio of the sensing signal while ensuring a low bit error rate for the communication signal. Simultaneously, the communication and sensing signals coexist continuously in the time domain without actively inserted time gaps. The composite electrical signal has a stable power envelope and a peak-to-average power ratio far lower than traditional time-division multiplexing schemes, avoiding the nonlinear saturation of the Mach-Zehnder modulator and the transient gain of the erbium-doped fiber amplifier caused by drastic power jumps. Distortion is eliminated, maintaining the linearity of the optical link. Furthermore, mutual interference suppression is mainly achieved at the transmitting end through waveform co-design, eliminating the need to deploy complex interference cancellation algorithms at the receiving end, significantly reducing the complexity and power consumption of digital signal processing at the receiving end. Moreover, by increasing the time domain width of the linear frequency modulated pulse to reduce its peak power, the sensing signal can be injected into the optical fiber with low peak power. Pulse compression is then achieved with the help of matched filtering at the receiving end. This satisfies the requirements of distributed acoustic sensing for high spatial resolution and long detection distance, while also being compatible with the stringent requirements of high-speed continuous communication for low peak-to-average power ratio, realizing the integrated fusion of high-speed data transmission and high-precision distributed sensing in optical networks. Attached Figure Description

[0014] Figure 1 This is a schematic diagram showing the superposition of existing communication signals and sensor signals; Figure 2 This is a schematic diagram of the existing time segmentation method; Figure 3 A flowchart illustrating the steps of the fiber optic sensing fusion method based on spectrum shaping provided in the first embodiment of the present invention; Figure 4 This is a schematic diagram of the MRDS encoding principle of the transmitting end provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of the MRDS decoding principle of the receiving end provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the integration of spectral-shaping-based inductive signals according to an embodiment of the present invention; Figure 7 This is a schematic diagram illustrating the principle of LFM sensor signal generation and matched filtering in an embodiment of the present invention. Figure 8 This is a schematic diagram of the overall architecture of an embodiment of the present invention; Figure 9 This is a schematic diagram of the power spectral density of the communication signal after spectral shaping, according to an embodiment of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0016] It should be noted that all directional indications in the embodiments of this application are only used to explain the relative positional relationship and movement of each component in a specific posture. If the specific posture changes, the directional indications will also change accordingly.

[0017] Furthermore, the use of terms such as "first" and "second" in this application is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed in this application.

[0018] In existing research and engineering implementations, the following two traditional solutions are mainly used to address the problem of coexistence of communication and sensing signals in the same optical fiber: The first type of scheme is the power superposition method: it directly superimposes power in the frequency domain, relying on algorithms to hard-disassemble the components, which has extremely high complexity. Figure 1 As shown, this scheme uses a power coupler at the transmitting end to directly superimpose continuous PAM4 communication baseband signals with high-energy, narrow-pulse-width coherent sensing pulses in the electrical or optical domains. These signals are simultaneously injected into the optical fiber through the same light source or the same fiber optic channel. At the receiving end, the communication and sensing information are extracted separately through filtering and demodulation. Its implementation is relatively simple, the system structure is intuitive, and it is easy to introduce sensing functionality into existing communication links. Therefore, it was widely adopted in early explorations of communication-sensing fusion.

[0019] However, because the communication signal and the sensing signal are not orthogonally superimposed at the transmitting end, this method has the following problems: No spectrum shaping capability: Uncontrollable interference; Due to the overlap between communication signals and sensing signals in the time, frequency, or power domains, mutual interference between signals is inevitably introduced after superposition. In this way, low-frequency components in the communication signal and carrier beat frequency noise caused by multipath reflections overlap with the spectrum of the sensing signal, resulting in an increase in the communication bit error rate and a decrease in the sensing signal-to-noise ratio. Moreover, this interference changes with the link state and is difficult to suppress stably.

[0020] The sensor signal cannot be compressed: it is difficult to achieve both long distance and high resolution; in order to obtain sufficient echo energy, it is usually necessary to increase the sensor signal power, but this will further aggravate the interference to the communication signal, making it difficult for the system to simultaneously achieve long-distance detection and communication stability.

[0021] Interference suppression relies on complex post-processing, resulting in high system complexity. The wideband spectrum and randomness of high-speed communication signals can cause background noise superposition in the sensor echo signal, thereby reducing the signal-to-noise ratio and spatial resolution accuracy of the sensing system. To alleviate these problems, existing systems typically require the introduction of high-pass filtering (HPF), digital cancellation, or multi-stage signal processing algorithms (such as A1 / A2 type suppression algorithms) at the receiver, increasing implementation complexity and power consumption. Furthermore, real-time implementation in high-speed IMDD systems is challenging.

[0022] The second type of method is time segmentation (time division multiplexing): gaps are reserved in the communication stream to temporarily reduce the communication signal power and inject short-duration high-peak sensing pulses, thereby achieving time isolation between communication and sensing, such as... Figure 2 As shown, its basic operating mode is as follows: On the time axis, the communication signal is transmitted discontinuously; within a predetermined communication gap, the communication signal power is temporarily stopped or significantly reduced, and a short-duration, high-peak-power sensing pulse signal is injected. The communication signal and the sensing signal alternate in the time domain, thus avoiding overlap between the two at the same moment. The core idea of ​​this method is to achieve the multiplexing of communication and sensing functions under the same optical fiber and light source conditions through time isolation, avoiding direct superposition in the frequency domain or power domain.

[0023] However, from the perspective of signal statistical characteristics, the peak-to-average power ratio (PAPR) is defined as the ratio of the instantaneous peak power of the signal to the average power over the entire observation period, and its mathematical expression is: ; In this segmented transmission structure, although the peak power of the communication signal remains constant within the effective time slot, the presence of time-domain gaps significantly reduces the system's duty cycle on the overall time axis. This high discontinuity in the time domain drastically lowers the overall statistical average power of the signal (i.e., the denominator decreases), thus significantly increasing the PAPR of the synthesized signal. A high PAPR places extremely high demands on the linearity of subsequent active devices such as power amplifiers, increasing the complexity and cost of system implementation.

[0024] In addition, in IMDD systems, both the Mach-Zehnder modulator (MZM) and its driving circuit have a limited linear operating range. When the peak signal power is too high, the modulator is forced to operate in the nonlinear region or even the saturation region, thus introducing significant nonlinear distortion. For multilevel modulation formats such as PAM4, nonlinear distortion directly disrupts the amplitude spacing between symbols, leading to a significant increase in the bit error rate. Since this nonlinearity originates from the drastic fluctuations in the signal power envelope itself, it is difficult to completely eliminate through simple equalization or digital compensation. Furthermore, in systems containing erbium-doped fiber amplifiers (EDFAs), the EDFAs are highly sensitive to changes in input power. When the signal power jumps drastically between the communication segment and the sensing pulse, it easily triggers gain competition and dynamic gain adjustment processes. This process will then lead to significant envelope distortion in the communication signal, such as power droop, overshoot, or enhanced transient noise, thereby further deteriorating the communication signal quality.

[0025] Based on this, the present invention provides a fiber optic sensing fusion method based on spectrum shaping, such as... Figure 3 As shown, it includes the following steps: S101 performs spectrum shaping on continuous communication signals, actively constructing low-energy notch regions in the spectrum of the communication signals. First, the continuous communication signal to be transmitted is acquired, and then its spectrum is shaped to actively create a low-energy notch region in its amplitude-frequency characteristics. The notch region is characterized by a significantly reduced power spectral density in the communication signal, reserving spectral space for subsequent sensing signals. Spectrum shaping can be achieved through digital encoding, filtering, or pre-distortion, with the aim of adjusting the power spectral distribution of the communication signal so that it no longer uniformly fills the entire available bandwidth, but instead forms an energy dip in the target frequency band.

[0026] S102 generates a linear frequency modulated pulse as a sensing signal and accurately maps the spectrum of the sensing signal onto the notch region so that the sensing signal and the communication signal are orthogonal in the frequency domain. A linearly frequency-modulated (LFM) pulse signal (i.e., frequency changes linearly with time) is generated as the sensing signal. This LFM pulse has waveform characteristics of a large time-domain width and relatively low instantaneous peak power. Furthermore, through parameter design (such as center frequency and frequency sweep range), the spectral energy of this LFM pulse is made to fall precisely within the notch region constructed in step S101 in the frequency domain. As a result, the sensing signal and the communication signal are orthogonally distributed in the frequency domain, and no significant co-channel interference occurs when they are transmitted simultaneously.

[0027] S103 superimposes the spectrum-shaped communication signal and the sensing signal in the electrical domain to obtain a composite electrical signal with a continuous power envelope. Since the energy of the communication signal remains continuous outside the notch region, while the energy of the sensing signal is concentrated within the notch region, and the time-domain envelope of the linear frequency modulation pulse is relatively flat, the composite electrical signal obtained after superposition has a continuous power envelope on the entire time axis. That is, there is no drastic power drop or spike caused by signal turn-off or sudden change. This continuous power envelope characteristic is beneficial for maintaining the linear working state in the subsequent electro-optic modulation process.

[0028] S104 modulates the composite electrical signal onto the optical carrier to form an integrated sensing optical signal and injects it into the optical fiber; The obtained composite electrical signal is loaded into an electro-optic modulator (e.g., a Mach-Zehnder modulator) to modulate the intensity or phase of the continuous light wave, thereby converting the composite signal in the electrical domain to the optical domain, forming a combined communication and sensing optical signal. This optical signal simultaneously carries communication information and sensing detection information. Subsequently, the optical signal is injected into the fiber under test / transmission through an optical circulator, causing it to generate a backscattered Rayleigh echo while transmitting forward.

[0029] S105 receives the forward optical signal transmitted through optical fiber, recovers the communication signal through direct detection, and demodulates the communication data. At the far end or receiving end of the optical fiber, the integrated sensing and communication optical signal transmitted forward via the optical fiber is received. A direct detection method is used, that is, the optical signal is converted into an electrical signal using a photodetector. The sensing signal is then filtered out to obtain the communication signal. Since the communication signal and the sensing signal are orthogonal in the frequency domain, they can be mutually filtered to obtain their respective baseband signals, ensuring that the recovery of the communication signal is not significantly interfered with by the sensing signal. Conventional demodulation processing (such as sampling, equalization, and decision-making) is then performed on the detected electrical signal to reconstruct the communication data from the transmitting end.

[0030] S106 receives the backscattered Rayleigh echo returned from the optical fiber, performs pulse compression on the echo through coherent detection and matched filtering, and then demodulates to recover the distributed sensing information. At the input end (or the same end) of the optical fiber, backscattered Rayleigh echo signals returned from various points along the fiber are received. These echoes carry modulation information about external physical quantities such as vibrations and sound waves along the fiber. Using coherent detection, the echoes are mixed with the local oscillator light to extract a complex signal containing phase and amplitude. Subsequently, matched filtering is performed on this complex signal, and cross-correlation is performed using the transmitted linear frequency modulated pulse signal as a reference template. This achieves pulse compression, concentrating the wide pulse energy into a narrow time span to recover high spatial resolution distributed sensing information (such as vibration location, frequency, and intensity).

[0031] like Figures 4 to 6As shown, in some embodiments, spectrum shaping of continuous communication signals includes processing the communication signals using Minimum Operating Digital Sum (MRDS) coding to generate a notch region near DC.

[0032] Spectrum shaping of communication signals is achieved using Minimum Running Digital Sum (MRDS) encoding. The core principle of this encoding method is to adjust the polarity distribution of symbols within the communication data block so that the Running Digital Sum (RDS) of the signal in the time domain approaches zero, thereby generating a significant spectral notch near the DC frequency in the frequency domain.

[0033] Specifically, for the input PAM4 symbol sequence {x i}, its total number of operations Defined as: To implement notch filtering, the algorithm encodes data blocks of length N (e.g., N=16). First, the sum of the entire block is calculated. (i=1 to N), then find an optimal reversal point K such that the reversed sequence satisfies the charge balance condition: ; By adjusting the polarity distribution within the data block, the sum of the running numbers in the time domain can be made close to zero, thus generating a notch filter near the 0-frequency range in the frequency domain.

[0034] MRDS coding does not require additional training sequences or redundant overhead; it constructs a low-energy notch region in the center of the baseband spectrum of the communication signal simply by dynamically adjusting the symbol polarity. The width and depth of this notch region can be controlled by the coding block length N and the inversion point selection strategy, thereby providing a controllable spectral cavity for the subsequent embedding of sensing signals. In some implementations, before injecting the integrated sensing optical signal into the optical fiber, the integrated sensing optical signal is frequency-shifted by an acousto-optic modulator (AOM) to move the notch region to a location in the optical fiber link that is intended to interfere with the signal.

[0035] After the integrated sensing optical signal is formed in step S104 and before it is injected into the optical fiber, the optical signal undergoes frequency shifting processing. Specifically, an acousto-optic modulator (AOM) is used as the frequency shifting device. The driving frequency of the AOM is adjusted. The entire spectrum of the integrated sensing optical signal (including the spectrum of the communication signal and the constructed notch region therein) is shifted from the original optical carrier frequency. Move to + . The size is flexibly set according to the location of strong interference frequency bands (such as Brillouin scattering noise or low-frequency mechanical noise) that actually exist in the fiber optic link.

[0036] After AOM relocation, the notch region originally generated near DC by MRDS encoding is shifted to the center of the target frequency, thus precisely covering the strong noise band in the link, thereby achieving physical-level spectrum avoidance. In this way, this method can actively avoid interference at any frequency location in the fiber optic link, and is not limited to noise suppression near DC.

[0037] It should be noted that AOM frequency shifting is an additional step independent of MRDS encoding, and the two can be used in combination: first, a notch filter is constructed near DC using MRDS encoding, and then the entire spectrum (including the notch filter) is shifted to the desired frequency band using AOM. This combined approach significantly enhances the adaptability to different fiber optic link environments.

[0038] like Figure 7 As shown, in some embodiments, generating a linear frequency modulated pulse as a sensing signal includes: determining the time domain width and peak power of the linear frequency modulated pulse according to a preset sensing detection energy requirement; Specifically, by increasing the time domain width to reduce peak power, the power fluctuation range of the composite electrical signal applied to the modulator is kept within the linear operating range of the modulator.

[0039] First, a sensing energy value is preset based on the detection energy requirements of the distributed fiber optic sensing system. Based on this preset energy value, the time-domain width and peak power of the linear frequency modulated pulse are determined.

[0040] The complex envelope of a linear frequency modulated pulse is represented as: ; Where K is the chirp rate, The pulse time domain width, B represents the sweep bandwidth.

[0041] Pulse energy E and peak power Time domain width The relationship between them is: ; For a given preset energy requirement E, the field of view width With peak power Mutual constraint: the wider the time domain width, the lower the required peak power. Therefore, this implementation significantly reduces the peak power by increasing the time domain width of the linear frequency modulated pulse (i.e., using a wide pulse instead of the traditional narrow pulse).

[0042] Since the linear frequency modulated pulse is subsequently superimposed on the spectrum-shaped communication signal in step S103, the power envelope of the resulting composite electrical signal will be affected by both the communication signal and the sensing signal. Through this design, the peak power of the sensing signal is controlled at a low level, thus limiting the power fluctuation range of the entire composite electrical signal to an acceptable range, preventing it from exceeding the linear operating range of the subsequent electro-optic modulator (e.g., a Mach-Zehnder modulator, MZM). In other words, the modulator always operates in the region where its input voltage and output optical power are linearly related, avoiding the modulator being forced into the nonlinear saturation region due to excessively high peak power of the sensing signal. This protects the amplitude spacing of multi-level modulation formats such as PAM4 and also avoids waveform distortion of the sensing signal.

[0043] In some implementations, the composite electrical signal with a continuous power envelope has no actively inserted time gaps on the time axis, and its peak-to-average power ratio (PAPR) is lower than a preset threshold.

[0044] Traditional time-division multiplexing (TDM) injects sensing pulses by actively inserting time gaps into the communication signal, resulting in discontinuous signal transmission along the time axis. This discontinuity significantly reduces the system's average power throughout the observation period, while the instantaneous peak power remains constant, thus producing an extremely high peak-to-average power ratio (PAPR). When time gaps exist in the signal, the average power decreases, and the PAPR increases sharply.

[0045] In contrast, the composite electrical signal in this embodiment has no actively inserted time gaps on the time axis. The communication signal and the sensing signal coexist continuously within the same time interval, without any artificial shutdown or power drop. Since the communication signal is continuously transmitted after spectrum shaping, and the sensing signal uses time-domain broadened linear frequency modulated pulses, the power envelope of the composite electrical signal formed by their superposition is continuous and stable across the entire time axis. Therefore, its average power is maintained, and there is no decrease in the denominator, resulting in a significantly lower PAPR (Power Approval Ratio) compared to traditional time-division multiplexing schemes.

[0046] Furthermore, the PAPR is designed to be below a preset threshold, which is determined based on the linear dynamic range of the subsequent electro-optic modulator (such as a Mach-Zehnder modulator) and the gain stability requirements of the power amplifier (such as an erbium-doped fiber amplifier, EDFA). Specifically, by selecting appropriate communication signal spectrum shaping parameters (such as MRDS coding block length and notch depth) and sensing signal parameters (such as LFM pulse time-domain width and peak power), the maximum instantaneous power of the composite electrical signal is ensured not to exceed the upper limit of the modulator's linear region, while the average power remains within the steady-state gain range of the EDFA, thereby guaranteeing that the PAPR is below the preset threshold (e.g., below 6 dB, with the specific value set according to the actual hardware characteristics).

[0047] In this way, this implementation eliminates the risk of nonlinear distortion caused by high PAPR from the waveform design perspective, and avoids problems such as modulator saturation and EDFA gain transient distortion caused by signal shutdown in traditional time-division multiplexing schemes, thus providing a guarantee for the linear transmission of the integrated sensing signal.

[0048] In some implementations, demodulating the communication data includes performing a baseband drift cancellation (BDC) operation on the directly detected electrical signal to remove the DC offset.

[0049] After the forward-transmitted integrated optical signal is directly detected by a photodetector, an electrical signal is obtained. This electrical signal may contain a DC offset component. Sources of DC offset include: dark current in the photodetector, zero-point drift in subsequent amplification circuits, and residual carrier components introduced when the system uses an acousto-optic modulator (AOM) for frequency shifting. DC offset can interfere with the decision-making of communication signals, especially for multi-level modulation formats such as PAM4, where DC drift can disrupt symbol amplitude spacing, leading to an increased bit error rate.

[0050] To eliminate DC offset, this implementation employs Baseband Drift Cancellation (BDC) operation. BDC is a low-complexity digital signal processing operation, and its processing logic is expressed as follows: ; in Let be the i-th symbol in the data block at the receiving end, and N be the data block length. This operation first calculates the average value (i.e., DC component) of all symbols within a data block, and then subtracts this average value from each symbol, thereby forcibly removing the DC offset from the signal.

[0051] BDC operations do not rely on prior channel information or require training sequences, and have extremely low computational cost, making them suitable for implementation in high-speed real-time systems. After BDC processing, the baseline of the communication signal is calibrated to zero level, and subsequent equalization (such as FFE) and symbol decision can be performed without DC bias, thereby accurately recovering the original communication data.

[0052] In this embodiment, BDC operation works in conjunction with the aforementioned MRDS encoding: MRDS constructs a spectral notch at the transmitting end, AOM may generate a frequency shift, and BDC calibrates the zero frequency at the receiving end, together ensuring high-quality demodulation of the communication signal within the integrated sensing framework.

[0053] In some implementations, pulse compression of the echo through coherent detection and matched filtering includes: performing cross-correlation calculations on the echo signal and a locally stored linear frequency modulated pulse template to compress the wide pulse into a narrow pulse; wherein, the larger the sweep bandwidth, the narrower the compressed pulse width.

[0054] At the sensor receiver, the complex envelope of the backscattered Rayleigh echo signal, denoted as r(t), is obtained through coherent detection. Simultaneously, a linear frequency modulated pulse template signal s(t), identical to that at the transmitter, is pre-stored locally, with the following expression: ; Where K is the linear frequency modulation slope, The pulse time domain width, B represents the sweep bandwidth.

[0055] The essence of matched filtering is to convolve (or equivalently perform cross-correlation) the received echo signal r(t) with the conjugate time-reversed signal of the local template. The output y(t) of the matched filter is: ; in, This is the complex conjugate of s(t). This operation essentially calculates the cross-correlation between r(t) and s(t).

[0056] According to the pulse compression principle, when r(t) contains a scattered signal component that matches s(t), the above cross-correlation operation will produce a sharp peak. The original width is... The long pulse is compressed into a narrow pulse, and there is a definite logical relationship between the width of the compressed pulse and the sweep bandwidth B: the larger the sweep bandwidth, the narrower the compressed pulse width. Therefore, by increasing the sweep bandwidth, a narrower compressed pulse can be obtained, thereby improving the spatial resolution of the system.

[0057] In this way, this implementation method uses matched filtering to achieve "wide pulse at the transmitting end and narrow pulse at the receiving end" sensing detection, which maintains low peak power to be compatible with communication signals, and restores high spatial resolution through signal processing at the receiving end.

[0058] In some implementations, after recovering the distributed sensing information, the vibration information along the optical fiber is calculated using a rotating vector summation and moving average filtering algorithm.

[0059] Step S106, after completing the matched filter pulse compression and obtaining the complex values ​​of the scattered signals at each location along the optical fiber, also includes post-processing the complex values ​​using a rotating vector summation and moving average filtering algorithm to accurately calculate the distributed vibration information.

[0060] Specifically, for each spatial sampling point on the optical fiber, the corresponding matched filter output is a complex signal. The phase change of this complex signal reflects the dynamic modulation of the fiber length by external vibrations or acoustic disturbances. To extract the vibration frequency and intensity from a series of complex signals in time sequence, this implementation first performs a rotating vector summation: the complex signals corresponding to the same location point within multiple consecutive transmission pulse periods are considered as rotating vectors. By coherently superimposing these vectors, the periodic vibration signal components are enhanced while suppressing incoherent random noise. Subsequently, a moving average filter is used to smooth the summed signal, further filtering out high-frequency residual noise, to obtain the envelope and phase change curves of the vibration signal. Finally, by performing spectral analysis (such as Fast Fourier Transform) or phase demodulation on these curves, the vibration frequency distribution and relative intensity information at each location point along the optical fiber can be obtained.

[0061] The aforementioned combined algorithm of rotating vector summation and moving average filtering (MRAS) effectively improves the signal-to-noise ratio and detection sensitivity of the distributed acoustic sensing system while maintaining the high spatial resolution obtained by matched filtering, enabling accurate identification and localization of weak vibration signals.

[0062] In some implementations, the suppression of mutual interference between communication signals and sensing signals is mainly achieved at the transmitting end through spectrum shaping and waveform co-design, rather than relying on interference cancellation algorithms at the receiving end.

[0063] Traditional integrated sensing solutions often deploy complex high-pass filters, digital cancellation, or multi-stage interference suppression algorithms (such as the A1 / A2 algorithm) at the receiver to reduce mutual interference caused by direct superposition or time-division gap insertion at the transmitter. However, these receiver post-processing methods increase system power consumption and hardware complexity, and are difficult to implement in real time in high-speed IMDD systems.

[0064] This implementation achieves mutual interference suppression through active collaborative design at the transmitting end: on the one hand, a low-energy notch region is constructed in the communication signal spectrum using spectrum shaping (such as MRDS coding), reserving frequency domain space for the sensing signal from the signal source; on the other hand, a time-domain broadened, low-peak-power linear frequency-modulated pulse is selected as the sensing signal, and its spectrum is precisely filled into the notch region, ensuring that the two signals are orthogonal in the frequency domain before entering the optical fiber. This "orthogonal first, then superimposed" design ensures that the communication and sensing signals generate almost no co-channel interference during optical domain transmission. The receiving end only needs conventional filtering or simple baseband drift cancellation to recover their respective information, without the need for complex interference cancellation algorithms. This significantly reduces the burden of digital signal processing at the receiving end, while improving the system's real-time performance and energy efficiency.

[0065] In some implementations, the communication signal is a four-level pulse amplitude modulation (PAM4) signal.

[0066] The PAM4 signal maps every two bits to four different amplitude levels, achieving a bit rate twice that of traditional non-return-to-zero (NRZ) modulation at the same symbol rate. It is currently the standard modulation format for 400GE / 800GE optical interconnects. In this embodiment, after spectral shaping (such as MRDS encoding) of the PAM4 signal, its multi-level amplitude characteristics are still maintained; and because the sensing signal is confined within the notch region, the amplitude spacing of the PAM4 signal is not disrupted, thus ensuring low bit error rate transmission.

[0067] In some implementations, the composite electrical signal is modulated onto the optical carrier using a Mach-Zehnder modulator (MZM), and the inductive-synthetic optical signal is amplified by an erbium-doped fiber amplifier (EDFA) and then injected into the optical fiber through a circulator.

[0068] The obtained composite electrical signal is amplified by an arbitrary waveform generator (AWG) and an electrical amplifier, and then applied to the RF port of the MZM. The DC bias point of the MZM is set at the midpoint of the linear operating region (such as the quadrature bias point) so that the output optical power varies linearly with the input voltage. Since the composite electrical signal of this method has a continuous power envelope and the peak power is controlled within the linear range of the modulator, the MZM can complete the electro-optical conversion without distortion.

[0069] The modulated, integrated sensing optical signal is first amplified by an erbium-doped fiber amplifier (EDFA) to compensate for subsequent fiber transmission and circulator insertion loss. The amplified optical signal is input through port 1 of the circulator, and port 2 is connected to the sensing fiber. The signal is injected into the fiber from port 2 and transmitted forward. The backscattered Rayleigh echo returns from the fiber, enters port 3 of the circulator via port 2, and is output to the sensing receiver module. This circulator configuration enables unidirectional transmission and echo reception of the optical signal on the same fiber, simplifying the system structure.

[0070] like Figure 8 As shown, to verify the feasibility of the integrated sensing solution, a PAM4 short-range optical transmission and distributed vibration sensing integrated system is used as an example to explain its signal characteristics and processing flow in detail: Communication signal generation: First, the original bit sequence is mapped into multi-level pulse amplitude modulation (e.g., PAM4) symbols to form a continuous communication baseband signal. The symbol rate of the communication signal is set according to the system's transmission bandwidth requirements.

[0071] Communication signal spectrum shaping: Minimum Running Digital Sum (MRDS) coding is used to encode communication symbols in blocks. Based on a preset data block length, MRDS coding actively constructs a low-energy notch region near the low frequencies or DC of the communication signal's baseband spectrum. The width and depth of the notch region are determined by the data block length to ensure sufficient frequency domain orthogonality space for the sensing signal. Its spectral characteristics are as follows: Figure 9 As shown, from Figure 9 It is evident that a significant energy dip occurs near the zero frequency.

[0072] Sensing signal generation: A linear frequency modulated (LFM) pulse sequence is generated as the sensing signal. To avoid communication interference, the sweep bandwidth of the sensing signal is limited to within the notch region width of the communication signal, so that the spectrum of the sensing signal can be accurately mapped and nested within the notch region; the pulse time domain width, repetition period, and peak power of the sensing signal are configured according to preset values ​​of sensing detection distance, spatial resolution, and energy requirements.

[0073] Electrical domain superposition: The communication signal after MRDS encoding and shaping is linearly superimposed with the LFM pulse sequence in the digital domain to form a composite electrical signal. Since the two are orthogonal in the frequency domain, the energy envelope of the composite electrical signal in the time domain remains continuous and stable without any inserted time gaps, thereby avoiding nonlinear distortion caused by time domain switching.

[0074] Electro-optic modulation and transmission: The composite electrical signal is amplified by an arbitrary waveform generator and an electrical amplifier, driving an optical modulator (such as a Mach-Zehnder modulator, MZM) to modulate the electrical signal onto an optical carrier. The optical modulator is preferably biased in the linear operating region (such as the quadrature point) to ensure that the composite signal does not undergo clipping distortion. The modulated, integrated sensing optical signal is then boosted to the target power by an optical amplifier and injected into the distributed sensing fiber under test through an optical directional device (such as an optical circulator).

[0075] Communication Reception and Mediation: At the far end of the optical fiber (or via an optical splitter), the forward optical signal is converted into an electrical signal by a photodetector. At the receiving end, the sensor signal located within the notch region is first filtered out using a digital or analog filter. The filtered signal is then resampled, matched, and baseband drift cancellation (BDC) is performed to eliminate low-frequency link impairments. After BDC processing, the signal is compensated by an equalizer and then decoded using inverse MRDS to recover the original bitstream. Because the sensor signal is entirely within the notch of the communication signal and has been filtered and suppressed, the bit error rate of the communication branch can be kept below the ideal forward error correction (FEC) threshold.

[0076] Sensing reception and modulation: The backscattered Rayleigh echo generated in the optical fiber returns along the original path, is extracted by the optical directional device, and coherently mixed with the local optical oscillator in a coherent receiver, outputting two sensing electrical signals, one in phase and one quadrature (I / Q). Sensor digital signal processing (DSP) first uses a local LFM pulse template corresponding to the transmitter to perform matched filtering (pulse compression operation) on the received I / Q signals to extract high spatial resolution pulse echoes. Subsequently, phase demodulation (such as rotating vector summation and moving average filtering) is performed on the compressed pulses within multiple consecutive pulse periods, ultimately demodulating the vibration frequency and intensity information at various locations along the optical fiber.

[0077] The above examples fully demonstrate that the spectrum shaping orthogonalization scheme proposed in this invention can simultaneously realize high-speed PAM4 communication and high-resolution distributed acoustic sensing on the same optical fiber, without interference between the two. The power envelope of the system is stable, the modulator always operates in the linear region, and the signal processing at the receiving end is simple and efficient.

[0078] The specific embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made in accordance with the technical concept of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A fiber optic sensing fusion method based on spectrum shaping, characterized in that, include: Spectrum shaping is performed on continuous communication signals to actively construct low-energy notch regions in the spectrum of the communication signals. A linear frequency modulated pulse is generated as a sensing signal, and the spectrum of the sensing signal is accurately mapped onto the notch region so that the sensing signal and the communication signal are orthogonal in the frequency domain. The communication signal after spectrum shaping is superimposed with the sensing signal in the electrical domain to obtain a composite electrical signal with a continuous power envelope. The composite electrical signal is modulated onto the optical carrier to form an integrated sensing optical signal, which is then injected into the optical fiber. It receives the forward optical signal transmitted through optical fiber, recovers the communication signal through direct detection, and demodulates the communication data. The backscattered Rayleigh echo returned from the optical fiber is received, and pulse compression is performed on the echo through coherent detection and matched filtering. The distributed sensing information is then demodulated to recover the signal.

2. The fiber optic inductive fusion method based on spectrum shaping as described in claim 1, characterized in that: The spectrum shaping of continuous communication signals includes: processing the communication signals using minimum operating numbers and encoding to generate a notch region near DC.

3. The fiber optic inductive fusion method based on spectrum shaping as described in claim 1, characterized in that: Before injecting the integrated sensing optical signal into the optical fiber, the frequency of the integrated sensing optical signal is shifted by an acousto-optic modulator (AOM) to move the notch region to a specified interference frequency band in the optical fiber link.

4. The fiber optic inductive fusion method based on spectrum shaping as described in claim 1, characterized in that: The generation of the linear frequency modulated pulse as a sensing signal includes: determining the time domain width and peak power of the linear frequency modulated pulse according to the preset sensing detection energy requirements; Specifically, by increasing the time domain width to reduce peak power, the power fluctuation range of the composite electrical signal applied to the modulator is kept within the linear operating range of the modulator.

5. The fiber optic inductive fusion method based on spectrum shaping as described in claim 1, characterized in that: A composite electrical signal with a continuous power envelope has no actively inserted time gaps on the time axis, and its peak-to-average power ratio is lower than a preset threshold.

6. The fiber optic inductive fusion method based on spectrum shaping as described in claim 1, characterized in that: The demodulated communication data includes: performing a baseband drift cancellation operation on the directly detected electrical signal to remove DC offset.

7. The fiber optic inductive fusion method based on spectrum shaping as described in claim 1, characterized in that: The pulse compression of the echo through coherent detection and matched filtering includes: performing cross-correlation operation between the echo signal and a locally stored linear frequency modulated pulse template to compress the wide pulse into a narrow pulse; wherein, the larger the sweep bandwidth, the narrower the compressed pulse width.

Images