Space optical communication code pattern switching method and system based on error vector magnitude feedback

Abstract

The application discloses a spatial optical communication code type switching method and system based on error vector magnitude feedback. The method is applied to a coherent optical communication scene, a receiving end of the system carries out real-time coherent digital signal processing on a received signal based on FPGA, calculates an error vector magnitude and generates a feedback control signal, and a transmitting end of the system receives the feedback control signal based on FPGA and carries out real-time switching between different coherent modulation formats. The application introduces a feedback control mechanism based on the error vector magnitude, and can realize adaptive switching of BPSK and QPSK and other modulation formats without interrupting communication, so as to improve the stability and transmission efficiency of the spatial optical communication system under time-varying channel conditions.

Description

Technical Field

[0001] This invention belongs to the field of free-space optical communication, and specifically relates to a spatial optical communication code switching method and system based on error vector amplitude feedback. Background Technology

[0002] Free-space optical communication, with its advantages of large bandwidth, low latency, flexible deployment, and unregulated spectrum resources, is of great value in applications such as long-distance space-to-ground communication, inter-building links in cities, and emergency communication in complex environments. However, free-space optical links are inevitably affected by atmospheric channels, especially beam attenuation and flicker caused by atmospheric turbulence, which leads to drastic fluctuations in received light intensity over time. This significantly degrades the link's signal-to-noise ratio and bit error rate, severely restricting the system's reliability and availability.

[0003] To address the above challenges, existing technologies have proposed several solutions:

[0004] Receiver-side adaptive compensation: By detecting and compensating for wavefront distortion in real time at the receiver, the wavefront quality reaching the receiving surface is improved, thereby reducing scintillation and distortion. This type of solution typically requires complex adaptive optics hardware, resulting in higher system cost and size.

[0005] End-to-end pre-distortion compensation: This method utilizes channel reciprocity to detect channel distortion at the transmitting end and perform pre-compensation using additional beacon light. This approach requires additional auxiliary transmission links, increasing system complexity and cost.

[0006] RF backup link: Millimeter wave multi-link redundancy ensures the probability of communication interruption under poor channel quality conditions, but this method cannot achieve the high bandwidth characteristics of FSO and will also bring additional power consumption and hardware costs.

[0007] Adaptive adjustment of transmitted signal parameters at the physical layer (e.g., transmit power, symbol rate, coding rate, modulation format, etc.) allows direct control over optical signal characteristics, effectively adapting to random fluctuations in the atmospheric environment. Modulation format switching technology leverages the differences in channel condition tolerance among different modulation schemes. Higher-order modulation (e.g., QPSK) is used to improve spectral efficiency under favorable channel conditions, while switching to lower-order modulation (e.g., BPSK) to ensure link reliability under deteriorating channel conditions. Compared to intensity modulation / direct detection (IM / DD), coherent modulation and coherent reception technologies offer higher receiver sensitivity and spectral efficiency, and can suppress phase disturbances and amplitude fluctuations caused by atmospheric turbulence to a certain extent. Therefore, they are more beneficial for improving the overall adaptability and communication quality of the system under complex atmospheric channel conditions.

[0008] However, existing technologies generally lack a truly real-time feedback mechanism operating in a closed loop at the physical layer. One type of scheme only employs compatible modulation and demodulation structures at the transmitting or receiving end, facilitating switching between different modulation formats, but without introducing continuous monitoring and feedback control of link quality. Another type of scheme, while measuring or estimating link quality, fails to form a high-speed closed loop capable of directly driving rapid switching of physical layer modulation formats. In real-world atmospheric turbulence environments, atmospheric channels exhibit significant randomness and time-varying characteristics. Especially under complex weather conditions, received light intensity and phase can fluctuate significantly over short timescales. Systems using only fixed modulation formats often experience significant deterioration in bit error rate under adverse channel conditions. On the other hand, using low-order modulation to ensure reliability in the long term inevitably sacrifices spectral efficiency, making it difficult to simultaneously achieve both link reliability and spectral efficiency.

[0009] Furthermore, existing link quality assessment methods largely rely on indirect measurements such as received power and statistical parameters like signal-to-noise ratio and bit error rate, resulting in limited response speeds. Therefore, there is an urgent need for a free-space coherent optical communication method and system that is simple in structure, low in power consumption, and suitable for hardware implementation, capable of real-time assessment of atmospheric channel quality at the receiving end. Summary of the Invention

[0010] To address the issues of signal amplitude and phase fluctuations, increased bit error rate, and the inability of fixed modulation schemes to simultaneously achieve reliability and spectral efficiency in existing free-space optical communication (FSO) links caused by random atmospheric turbulence, this invention proposes a spatial optical communication code switching method and system based on error vector amplitude feedback. This invention dynamically switches between modulation formats such as BPSK and QPSK based on the real-time estimated error vector amplitude (EVM) at the receiver without interrupting communication, achieving closed-loop adaptive control of the physical layer modulation scheme, thereby improving link stability and effective throughput. The error vector amplitude used in this invention serves as a typical demodulation domain indicator, directly corresponding to the constellation aggregation state in coherent demodulation, thus providing a more sensitive and accurate characterization of link quality. Based on the above evaluation results, adaptive modulation format switching is driven at the physical layer, simultaneously ensuring link reliability and spectral efficiency without significantly increasing system complexity.

[0011] The technical solution of the present invention is as follows:

[0012] I. A Spatial Optical Communication Code Switching Method Based on Error Vector Amplitude Feedback

[0013] For the transmitting FPGA of a free-space coherent optical communication system, it receives a feedback control signal transmitted by the receiving FPGA. The transmitting FPGA switches the modulation format and generates a modulation signal according to the feedback control signal.

[0014] For the receiving FPGA of a free-space coherent optical communication system, after processing the I / Q baseband data output by the ADC module, a baseband symbol sequence is obtained. Then, several windowed error vector amplitudes between the baseband symbol sequence and the ideal reference constellation points corresponding to the current modulation mode are calculated. Based on the calculated windowed error vector amplitudes, a feedback control signal is generated and transmitted to the transmitting FPGA.

[0015] The calculation of several windowed error vector magnitudes between the baseband symbol sequence and the ideal reference constellation points corresponding to the current modulation mode specifically includes:

[0016] The baseband symbol segments of the baseband symbol sequence are obtained by using a sliding window. The error vector magnitude between the baseband symbol segments corresponding to different windows and the ideal reference constellation points corresponding to the current modulation mode is calculated and denoted as the windowed error vector magnitude.

[0017] The maximum permissible error vector amplitude of the high-order coherent modulation format that meets the preset communication performance requirements is determined and used as the target threshold; when the system is in the high-order coherent modulation format and the corresponding error vector amplitude is greater than the target threshold, a modulation control command to switch to the low-order coherent modulation format is generated and recorded as a feedback control signal.

[0018] The ideal channel conditions corresponding to the high-order coherent modulation format are determined when the preset communication performance requirements are met, and the error vector amplitude corresponding to the low-order coherent modulation format is obtained under the ideal channel conditions and used as the cut-in threshold; when the system is in the low-order coherent modulation format and the corresponding error vector amplitude is less than the cut-in threshold, a modulation control command to switch to the high-order coherent modulation format is generated.

[0019] When the error vector amplitude is between the target threshold and the cut-in threshold, the system is in the hysteresis interval and maintains the current modulation format unchanged.

[0020] II. A Spatial Optical Communication Code Switching System Based on Error Vector Amplitude Feedback

[0021] The system includes a transmitting unit and a receiving unit. The transmitting unit includes a transmitting FPGA, and the receiving unit includes a receiving FPGA. The receiving FPGA generates a feedback control signal and transmits it to the transmitting FPGA. The transmitting FPGA switches the modulation format and generates a modulation signal according to the feedback control signal.

[0022] The transmitting FPGA includes a pseudo-random sequence generation module, a feedback signal receiving module, a switchable modulation mapping module, and a DAC driver module. The feedback control signal serves as the input to the feedback signal receiving module. Both the pseudo-random sequence generation module and the feedback signal receiving module are connected to the switchable modulation mapping module, which is connected to the DAC driver module. The DAC driver module is connected to the DAC module.

[0023] The receiving FPGA includes an ADC sampling data receiving module, a digital signal processing module, a reference constellation point generation module, an EVM calculation module, a dual-threshold decision and fault-tolerant calculation module, and a feedback signal transmission module. The ADC sampling data receiving module is connected to the digital signal processing module, the digital signal processing module is connected to the reference constellation point generation module and the EVM calculation module, the reference constellation point generation module is connected to the EVM calculation module, the EVM calculation module is connected to the dual-threshold decision and fault-tolerant calculation module, and the dual-threshold decision and fault-tolerant calculation module is connected to the feedback signal transmission module.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] 1. Achieving true closed-loop adaptive modulation control at the physical layer: This invention evaluates the link quality in real time at the receiving end based on the coherent demodulation results, and directly drives the modulation format switching at the transmitting end through feedback control signals, forming a closed-loop control link at the physical layer. This overcomes the shortcomings of existing technologies that rely solely on forward configuration or upper-layer protocols and cannot respond in real time to rapidly changing atmospheric channels.

[0026] 2. Using Error Vector Amplitude as a Link Quality Metric for More Direct and Sensitive Characterization: This invention uses error vector amplitude as the core evaluation index. EVM has a direct correspondence with constellation aggregation state in the coherent demodulation stage. Compared with traditional methods based on statistical quantities such as received power, it can more accurately reflect the dispersion of the modulation constellation and instantaneous demodulation performance, thereby improving the accuracy and sensitivity of link quality assessment.

[0027] 3. Balancing reliability and spectral efficiency in turbulent atmospheric environments: This invention utilizes a dual-threshold decision strategy based on EVM to automatically switch to low-order modulation such as BPSK to ensure bit error rate performance when channel conditions deteriorate and EVM increases, and to switch to high-order modulation such as QPSK to improve spectral efficiency when channel conditions improve and EVM decreases. This achieves a dynamic optimization trade-off between link reliability and average throughput under random fluctuations caused by atmospheric turbulence.

[0028] 4. Dual thresholds and multi-window consistency decision improve handover stability: This invention sets upper and lower handover thresholds and hysteresis intervals, and combines multi-window result consistency and fault-tolerant counting mechanisms to trigger modulation handover only when the same decision condition is met in multiple consecutive time windows. This effectively suppresses misjudgments caused by short-term noise disturbances or instantaneous fluctuations, avoids frequent jittering of the modulation format near the threshold, and ensures a smooth and stable handover process.

[0029] 5. Simple structure and easy to implement in real time on FPGA: The core functions of this invention are all implemented in hardware logic within the receiving FPGA. With the configurable constellation mapping structure of the transmitting FPGA, BPSK / QPSK online switching can be completed without the need to introduce additional adaptive optics devices or auxiliary optical links. The hardware structure is simple, with low power consumption and latency, making it easy to integrate and deploy in engineering systems.

[0030] 6. Excellent scalability and versatility: This invention uses EVM as the unified link quality metric and control basis. In addition to BPSK / QPSK, it can be extended to higher-order modulation formats such as M-order PSK or QAM. By adjusting the threshold and mapping rules, it can be adapted to different system parameters, thus exhibiting excellent scalability and versatility. Attached Figure Description

[0031] Figure 1 This is a block diagram of a free-space coherent optical communication system.

[0032] Figure 2 This is a block diagram of the FPGA structure at the transmitting end.

[0033] Figure 3 This is a block diagram of the receiver FPGA.

[0034] Figure 4 This is a flowchart of the method of the present invention.

[0035] Figure 5 This is a schematic diagram of the EVM dual threshold proposed in this invention. Detailed Implementation

[0036] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0037] In this invention, the "high" and "low" orders of the coherent modulation formats are relative concepts. Different coherent modulation formats can be selected, such as BPSK, QPSK, 8PSK, M-PSK, and other phase-shift keying coherent modulation formats, as well as 16QAM, M-QAM, and other quadrature amplitude coherent modulation formats. Taking QPSK mode as a high-order coherent modulation format and BPSK mode as a low-order coherent modulation format as an example, the technical solution of this invention will be explained in detail.

[0038] like Figure 1 As shown, the spatial optical communication code switching system based on error vector amplitude feedback proposed in this invention includes a transmitting unit and a receiving unit.

[0039] The transmitting unit sequentially includes a tunable laser TL, a polarization controller PC, an IQ modulator IQM, an erbium-doped fiber amplifier EDFA, a bandpass filter BPF, and a first collimator Collimator 1. The transmitting FPGA drives the digital-to-analog converter DAC to output the electrical domain baseband signal to drive the IQ modulator. After BPSK / QPSK modulation, the optical carrier is amplified, filtered, and collimated before being coupled into the free space link.

[0040] Tunable laser: provides a continuous optical carrier with an operating wavelength of approximately 1550 nm;

[0041] IQ modulator: The baseband data stream from the transmitting FPGA is loaded onto the I and Q branches respectively to realize BPSK or QPSK modulation;

[0042] Transmitter FPGA: Generates a fixed-length (e.g., 512-bit) pseudo-random bit sequence and transmits it cyclically; Based on the modulation control signal fed back from the receiver, it decides whether to transmit data only in the I branch (BPSK) or to transmit data interleaved in both the I and Q branches (QPSK), and updates the constellation mapping and drive waveform accordingly.

[0043] Erbium-doped fiber amplifier (EDFA) and bandpass filter (BPF): to amplify the power and shape the spectrum of the modulated optical signal, and to suppress amplification noise and out-of-band interference;

[0044] Collimator: Couples the modulated optical signal in the optical fiber into free space to form the emitted beam.

[0045] The atmospheric channel and disturbance simulation module includes a variable optical attenuator (VOA) and a rotating phase plate (RPP).

[0046] Variable Optical Attenuator (VOA): Used to simulate static attenuation in free-space links, adjusting the received optical power to the target range to reproduce long-distance link loss conditions;

[0047] Rotating Phase Plate (RPP): By changing the rotation speed and rotation state, it simulates atmospheric turbulence of varying intensities, introducing phase disturbances to generate light intensity scintillation, thus achieving laboratory simulation of dynamic atmospheric channels. The receiving unit sequentially includes a second collimator (Collimator 2), an erbium-doped fiber amplifier (EDFA), a bandpass filter (BPF), an integrated coherent receiver (ICR), an radio frequency amplifier (RFA), an analog-to-digital converter (ADC), a receiving FPGA (FPGA2), and a host computer (Computer). The local oscillator laser (LO) is connected to the ICR via a polarization controller (PC) to form a coherent receiving structure. The receiving FPGA performs coherent demodulation of the I / Q signals and calculates the error vector amplitude. Based on a dual-threshold decision, it generates a modulation control feedback signal, which is returned to the transmitting FPGA1 via a feedback control link. This feedback controls the transmitting end to adaptively switch between BPSK and QPSK modulation formats. The host computer can interact with FPGA2 to read information such as the received signal or error vector amplitude.

[0048] like Figure 2 As shown, the transmitting FPGA includes a pseudo-random sequence generation module, a feedback signal receiving module, a switchable modulation mapping module, and a DAC driver module. The feedback control signal serves as the input to the feedback signal receiving module. Both the pseudo-random sequence generation module and the feedback signal receiving module are connected to the switchable modulation mapping module, which in turn is connected to the DAC driver module. The DAC driver module is also connected to the DAC module. The transmitting FPGA can control the DAC and the IQ modulation structure and update the constellation mapping, enabling online switching between BPSK and QPSK modulation formats.

[0049] The pseudo-random sequence generation module is used to generate pseudo-random bit sequences of fixed length and send them cyclically.

[0050] The feedback signal receiving module is used to receive feedback control signals from the receiving end and output the current modulation mode indication, which can be implemented in hardware through high and low level signals.

[0051] The switchable modulation mapping module selects between the BPSK mapping submodule and the QPSK mapping submodule according to the mode indication: when in BPSK mode, only the bit sequence is mapped to the I branch, and the Q branch is set to zero; when in QPSK mode, the bit sequence is interleaved and distributed to both the I and Q branches, thereby realizing online switching between different modulation formats.

[0052] The DAC driver module generates corresponding digital-to-analog conversion control signals based on the I and Q mapping results, driving the subsequent DAC to output I / Q analog baseband signals to drive the IQ modulator to complete BPSK / QPSK modulation of the optical carrier.

[0053] The modulated optical signal is sequentially amplified and spectrally shaped before being coupled into the free-space atmospheric channel by a collimator. In one implementation, different link loss conditions can be simulated using a variable optical attenuator, and dynamic phase perturbations can be introduced by rotating a phase plate to reproduce atmospheric turbulence environments of varying intensities.

[0054] like Figure 3 As shown, the receiving FPGA includes an ADC sampling data receiving module, a digital signal processing module, a reference constellation point generation module, an EVM calculation module, a dual-threshold decision and fault-tolerant calculation module, and a feedback signal transmission module. The ADC sampling data receiving module is connected to the digital signal processing module, which in turn is connected to the reference constellation point generation module and the EVM calculation module. The reference constellation point generation module is connected to the EVM calculation module, which is connected to the dual-threshold decision and fault-tolerant calculation module, which is connected to the feedback signal transmission module. The digital signal processing module internally includes DC baseline cancellation, normalization preprocessing, channel equalization, frequency offset compensation, clock recovery, and phase recovery.

[0055] Collimator: Recouples the returning beam propagating in free space into the optical fiber.

[0056] Erbium-doped fiber amplifier (EDFA) and bandpass filter (BPF): Amplify the received optical signal and control the noise bandwidth and suppress out-of-band noise through optical filtering.

[0057] Local Oscillator Laser Source and Integrated Coherent Receiver (ICR): The local oscillator laser provides coherent local oscillator light in the same band as the transmitted carrier. The ICR interferes the received signal with the local oscillator light in a 90° optical mixing structure, outputting I / Q electrical signals, which provide the basis for subsequent coherent digital signal processing.

[0058] Radio frequency amplifier (RFA): Performs bandwidth-controlled linear amplification of I / Q electrical signals to increase the signal amplitude to meet the input dynamic range requirements of analog-to-digital converters.

[0059] High-speed analog-to-digital converter (ADC): Under the clock and control of the FPGA at the receiving end, the amplified I / Q signal is synchronously sampled at a set sampling rate, and a complex baseband digital sequence is output.

[0060] This invention uses an ADC module for sampling at the receiving end and an FPGA connected to it for signal quality evaluation.

[0061] The ADC module is used to perform analog-to-digital conversion on the I / Q electrical signals received from coherent reception.

[0062] The digital signal processing module is used to perform filtering, timing, and carrier compensation on the I / Q sampled data, and outputs the compensated baseband symbol sequence.

[0063] The reference constellation point generation module generates the corresponding ideal constellation points based on the current modulation format.

[0064] The EVM calculation module calculates the error vector magnitude based on the baseband symbol sequence and reference constellation points within a preset time window.

[0065] The dual-threshold decision and fault-tolerant counting module compares the EVM with preset upper and lower thresholds and combines a multi-window consistency strategy to determine the modulation format, which can improve switching stability. In this way, modulation format switching is only triggered when the decision result continuously meets the conditions.

[0066] The feedback signal transmission module generates a feedback control signal (e.g., using a single-bit or few-bit control word) based on the decision result and sends it to the transmitting FPGA via the feedback link.

[0067] The host computer communication interface module is used to exchange data and parameters with the host computer.

[0068] like Figure 4 As shown, the spatial optical communication code switching method based on error vector amplitude feedback proposed in this invention includes:

[0069] like Figure 2 As shown, for the transmitting FPGA of a free-space coherent optical communication system, it receives a feedback control signal transmitted by the receiving FPGA. The transmitting FPGA switches the modulation format and generates a modulation signal of the corresponding modulation format according to the feedback control signal.

[0070] like Figure 3 As shown, for the receiving FPGA of a free-space coherent optical communication system, after performing ADC sampling and coherent digital signal processing on the I / Q baseband data output by the ADC module, a baseband symbol sequence is obtained. Then, several windowed error vector amplitudes between the baseband symbol sequence and the ideal reference constellation points corresponding to the current modulation mode are calculated. Based on the calculated windowed error vector amplitudes, a feedback control signal is generated and transmitted to the transmitting FPGA.

[0071] Calculate the magnitudes of several windowed error vectors between the baseband symbol sequence and the ideal reference constellation points corresponding to the current modulation mode, specifically including:

[0072] The baseband symbol segments of the baseband symbol sequence are obtained by using a sliding window. The error vector magnitude between the baseband symbol segments corresponding to different windows and the ideal reference constellation points corresponding to the current modulation mode is calculated and denoted as the windowed error vector magnitude.

[0073] The feedback control signal is generated based on the calculated windowed error vector magnitude, specifically including:

[0074] If the current modulation mode is QPSK mode and multiple consecutive windowing error vector magnitudes (EVM) are greater than the upper switching threshold (EVM) high, QPSK When the channel condition is deemed "deteriorated," a modulation format downgrade instruction is generated, which is a modulation control instruction to switch to BPSK mode and recorded as a feedback control signal. If the current modulation mode is BPSK mode and the amplitude of the windowed error vector EVM is less than the lower switching threshold EVM for multiple consecutive windowed errors, the signal is considered to be in the BPSK mode. low, BPSK If the channel condition is deemed "improved", a modulation format upgrade instruction is generated, which is a modulation control instruction to switch to QPSK mode; otherwise, the current modulation mode remains unchanged in the hysteresis interval, thus avoiding frequent jittering near the critical condition.

[0075] Figure 5 A schematic diagram of EVM with dual thresholds was plotted. The horizontal axis represents the atmospheric turbulence intensity index T, used to characterize the strength of atmospheric turbulence, and the vertical axis represents the error vector amplitude EVM. Different curves represent the trend of EVM with turbulence intensity under different modulation formats. (Upper threshold EVM) high, QPSK and lower switching threshold EVM low, BPSK The determination method is as follows:

[0076] Experiments were conducted to obtain QPSK and BPSK curves. A QPSK target threshold and a QPSK cut-in threshold were set, with the QPSK target threshold being greater than the QPSK cut-in threshold. The EVM value corresponding to the intersection of the QPSK curve and the QPSK target threshold was recorded as the upper switching threshold EVM. high, QPSK The atmospheric turbulence intensity index T corresponding to the intersection of the QPSK curve and the QPSK cut-off threshold is denoted as the calibration intensity, and the EVM value corresponding to the calibration intensity on the BPSK curve is denoted as the lower switching threshold EVM. low, BPSK .

[0077] In this invention, the upper switching threshold is the maximum acceptable error vector magnitude under the QPSK modulation format, used to determine whether the QPSK modulation format can still meet the target performance requirements; the lower switching threshold is used to determine whether the channel conditions have recovered to a level where it can re-enter the QPSK modulation format. Although the lower switching threshold is compared under BPSK operating conditions, its physical meaning is to determine whether the conditions for switching to the QPSK modulation format are met.

[0078] In this embodiment, the upper switching threshold is obtained through offline testing under different atmospheric turbulence conditions in the QPSK modulation format, and its value is selected as the maximum permissible error vector amplitude when the target performance requirements are met. The lower switching threshold is obtained by mapping the turbulence conditions that meet the QPSK margin requirements and reading the error vector amplitude corresponding to the BPSK modulation format under the same channel conditions, thereby forming a stable hysteresis interval between different modulation formats.

[0079] To further improve the stability of modulation format switching, the receiver FPGA introduces a fault-tolerant counting mechanism for multi-window result consistency judgment. The modulation format switching request is only confirmed when multiple consecutive time windows meet the threshold condition on the same side, thereby effectively suppressing misjudgments caused by short-term noise disturbances or instantaneous deep fading.

[0080] After the judgment result stabilizes, the receiving FPGA generates a single-bit or few-bit modulation control signal and transmits it to the transmitting FPGA via a wired or RF auxiliary link. Upon receiving the control signal, the transmitting FPGA updates the modulation mapping method and driver configuration without interrupting communication, enabling online switching between BPSK and QPSK modulation formats.

[0081] Through the above embodiments, the present invention can realize real-time closed-loop adaptive modulation control of the physical layer under atmospheric turbulence conditions, thereby improving the average spectral efficiency of the system while ensuring link reliability.

[0082] The following uses BPSK / QPSK as an example to illustrate the specific process of the method proposed in this invention:

[0083] S1: Pseudo-random sequence generation and modulation preparation

[0084] The transmitting FPGA generates a fixed-length pseudo-random bit sequence (e.g., 512 bits) and transmits it cyclically. In the initial state, a fixed modulation format (e.g., QPSK) can be selected, and the modulation mapping lookup table and related driving parameters are configured.

[0085] S2: BPSK / QPSK switchable modulation

[0086] When the feedback control signal is in "BPSK mode", the transmitting FPGA only maps the bit sequence to the I branch and the Q branch is 0, realizing two-phase phase shift keying; when the control signal is in "QPSK mode", the FPGA interleaves and distributes the bit sequence to the I and Q branches, and generates a four-phase phase shift keying signal through the IQ modulator.

[0087] S3: Atmospheric channel propagation

[0088] The modulated optical signal is amplified by EDFA and filtered by BPF, then enters free space through a collimator, and passes through channel simulation units such as VOA and RPP in sequence, before being coupled back to the optical fiber by the collimator at the receiving end.

[0089] S4: Coherent photoelectric conversion and sampling at the receiving end

[0090] At the receiving end, the local oscillator laser and the received signal are injected into the integrated coherent receiver to achieve coherent photoelectric conversion of I / Q channels. After the I / Q electrical signals are amplified by the RF amplifier, they are sent to the high-speed ADC. Under the clock and trigger control of the FPGA at the receiving end, they are synchronously sampled at a set sampling rate to obtain a complex baseband digital sequence.

[0091] S5: Error Vector Magnitude Calculation

[0092] The receiving FPGA performs digital signal processing on the sampled I / Q baseband sequence, including signal preprocessing, clock recovery, channel equalization, and carrier recovery, to obtain a baseband symbol sequence corresponding to the current modulation format. Based on the current modulation mode, it generates an ideal reference constellation point and calculates the following for each time window (e.g., each window contains N symbols): the error vector between the received symbol and the corresponding reference constellation point; the mean square value of the error vector amplitude; and the mean square value of the reference constellation amplitude. Based on this, the error vector amplitude EVM for that window is obtained and scaled and quantized in a fixed-point format within the FPGA for subsequent hardware decision-making.

[0093] S6: EVM-based dual-threshold decision

[0094] Preset switching threshold EVM low, BPSK (Determined by the low-order modulation format, corresponding to the high-order modulation format cut-in threshold) and the upper switching threshold EVM high, QPSK (Determined by the higher-order modulation format, corresponding to the target threshold of the higher-order modulation format), forming a hysteresis-based double threshold interval: when currently operating in QPSK mode and multiple consecutive windows (e.g., 5) satisfy EVM > EVM high, QPSK When the channel condition is deemed "deteriorated," a modulation control command to switch to BPSK is generated; when the current operation is in BPSK mode and the EVM is satisfied for multiple consecutive windows... <EVM low, BPSK When the channel condition is improved, a modulation control command to switch to QPSK is generated. When the EVM is in the hysteresis interval, the current modulation format is kept unchanged, thereby avoiding frequent jitter near the critical condition.

[0095] S7: Fault-tolerant counting and modulation control signal feedback

[0096] The receiving FPGA accumulates the decision results of each window through a fault-tolerant counter. The modulation format switching request is only confirmed when multiple consecutive sampling windows meet the threshold condition on the same side and the count reaches the set threshold.

[0097] After the judgment result stabilizes, the receiving FPGA generates a single-bit or few-bit modulation control signal, which is sent to the transmitting FPGA through a wired or optical feedback path, triggering the transmitting end to switch between BPSK / QPSK modes and update the corresponding constellation mapping and driver configuration.

[0098] S8: Demodulation and Error / Throughput Statistics

[0099] Based on the results of coherent demodulation and digital signal processing at the receiver, the decision of BPSK / QPSK symbols is completed and compared with the local pseudo-random sequence to calculate the real-time bit error rate. During the experimental and system optimization phases, performance indicators such as EVM, bit error rate, and effective throughput can be calculated in different time intervals to evaluate the performance advantages of EVM-based adaptive modulation over fixed modulation methods in atmospheric turbulence environments.

[0100] Each processing unit described above can be implemented in hardware logic within an FPGA, achieving microsecond-level response times while possessing low power consumption and low latency control characteristics. This makes it suitable for real-time adaptive modulation control of the physical layer in practical free-space coherent optical communication systems.

Claims

1. A spatial optical communication code switching method based on error vector amplitude feedback, characterized in that, include: For the transmitting FPGA of a free-space coherent optical communication system, it receives a feedback control signal transmitted by the receiving FPGA. The transmitting FPGA switches the modulation format and generates a modulation signal according to the feedback control signal. For the receiving FPGA of a free-space coherent optical communication system, after processing the I / Q baseband data output by the ADC module, a baseband symbol sequence is obtained. Then, several windowed error vector amplitudes between the baseband symbol sequence and the ideal reference constellation points corresponding to the current modulation mode are calculated. Based on the calculated windowed error vector amplitudes, a feedback control signal is generated and transmitted to the transmitting FPGA.

2. The spatial optical communication code switching method based on error vector amplitude feedback according to claim 1, characterized in that, The calculation of several windowed error vector magnitudes between the baseband symbol sequence and the ideal reference constellation points corresponding to the current modulation mode specifically includes: The baseband symbol segments of the baseband symbol sequence are obtained by using a sliding window. The error vector magnitude between the baseband symbol segments corresponding to different windows and the ideal reference constellation points corresponding to the current modulation mode is calculated and denoted as the windowed error vector magnitude.

3. The spatial optical communication code switching method based on error vector amplitude feedback according to claim 1, characterized in that, The step of generating a feedback control signal based on the calculated windowed error vector magnitude specifically includes: The maximum permissible error vector amplitude of the high-order coherent modulation format that meets the preset communication performance requirements is determined and used as the target threshold; when the high-order coherent modulation format is in use and the corresponding error vector amplitude is greater than the target threshold, a modulation control command to switch to a low-order coherent modulation format is generated and recorded as a feedback control signal. The ideal channel conditions corresponding to the high-order coherent modulation format are determined when the preset communication performance requirements are met. Under the ideal channel conditions, the error vector amplitude corresponding to the low-order coherent modulation format is obtained and used as the cut-in threshold. When the low-order coherent modulation format is in use and the corresponding error vector amplitude is less than the cut-in threshold, a modulation control command to switch to the high-order coherent modulation format is generated. When the error vector amplitude is between the target threshold and the cut-in threshold, the current modulation format remains unchanged.

4. A spatial optical communication code switching system based on error vector amplitude feedback, characterized in that, It includes a transmitting unit and a receiving unit. The transmitting unit includes a transmitting FPGA, and the receiving unit includes a receiving FPGA. The receiving FPGA generates a feedback control signal and transmits it to the transmitting FPGA. The transmitting FPGA switches the modulation format and generates a modulation signal according to the feedback control signal.

5. A spatial optical communication code switching system based on error vector amplitude feedback according to claim 4, characterized in that, The transmitting FPGA includes a pseudo-random sequence generation module, a feedback signal receiving module, a switchable modulation mapping module, and a DAC driver module. The feedback control signal serves as the input to the feedback signal receiving module. Both the pseudo-random sequence generation module and the feedback signal receiving module are connected to the switchable modulation mapping module, which is connected to the DAC driver module. The DAC driver module is connected to the DAC module.

6. A spatial optical communication code switching system based on error vector amplitude feedback according to claim 4, characterized in that, The receiving FPGA includes an ADC sampling data receiving module, a digital signal processing module, a reference constellation point generation module, an EVM calculation module, a dual-threshold decision and fault-tolerant calculation module, and a feedback signal transmission module. The ADC sampling data receiving module is connected to the digital signal processing module, the digital signal processing module is connected to the reference constellation point generation module and the EVM calculation module, the reference constellation point generation module is connected to the EVM calculation module, the EVM calculation module is connected to the dual-threshold decision and fault-tolerant calculation module, and the dual-threshold decision and fault-tolerant calculation module is connected to the feedback signal transmission module.

Images