FPGA-based airborne high-speed image storage control method and system

Abstract

This application provides an airborne high-speed image storage control method and system based on FPGA, relating to the field of aviation data link transmission technology. This application monitors the downlink transmission capability between the aircraft and the ground station using FPGA, dynamically adjusts the image data to be transmitted, and divides it into key frames and non-key frames, storing them in ferroelectric memory and solid-state memory respectively. In case of power failure, a protection program is activated to convert key frames into a persistent format and prioritize their transmission when bandwidth is sufficient. Simultaneously, as the link status continuously improves, selective retransmission of non-key frames is initiated. Combined with adaptive link adjustment, this achieves complete and reliable transmission of airborne image data under unstable link conditions. Through the collaborative mechanism of key frame protection and priority transmission, adaptive retransmission of non-key frames, and dynamic scheduling of link status, the integrity and reliability of image data transmission in complex airborne environments are ensured.

Description

Technical Field

[0001] This application relates to the field of aviation data link transmission technology, and in particular to an airborne high-speed image storage control method and system based on FPGA. Background Technology

[0002] In missions such as UAV remote sensing and aerial mapping, airborne equipment needs to continuously acquire and rapidly transmit massive amounts of image data. However, the wireless data downlink between the aircraft and the ground station often experiences drastic bandwidth fluctuations or even outages due to factors such as distance changes, obstructions, and electromagnetic interference. Furthermore, the airborne environment carries the risk of unstable power supply. This necessitates that the airborne storage and control system possess extremely high real-time performance, dynamically adapting to link changes and prioritizing the integrity and recoverability of critical data under limited resources and harsh environments, thereby achieving highly reliable and uninterrupted transmission of image data.

[0003] Currently, one existing solution to address this technical requirement is to adopt a software-defined storage control architecture based on a high-performance airborne processor. This solution monitors the link status through software programs running on the processor and employs a dual-storage media design, storing critical data in non-volatile memory and non-critical data in high-capacity memory. In the event of a power failure, the software program executes a data protection process.

[0004] However, this solution has obvious drawbacks. First, the inherent latency of its software processing causes the system to respond slowly when dealing with high-speed data streams and rapid changes in link status, making it difficult to meet real-time requirements. Second, the complex control logic continuously consumes a large amount of processor computing resources, which is prone to conflict with other onboard tasks and affects the overall system performance. Finally, the operational stability of this solution is highly dependent on the upper-level operating system, and it faces the risk of freezing or restarting in complex airborne environments, leading to data transmission interruptions. Summary of the Invention

[0005] The purpose of this application is to provide an airborne high-speed image storage control method and system based on FPGA, so as to solve the problems of high system response latency, large processor resource consumption and poor reliability in harsh airborne environments that exist in the prior art due to the use of software processing architecture.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides an airborne high-speed image storage control method based on FPGA, comprising:

[0007] The downlink status between the aircraft and the ground station is monitored to obtain the link transmission capability parameters, and the image data to be transmitted is dynamically adjusted through the FPGA according to the link transmission capability parameters.

[0008] The adjusted image data to be transmitted is divided into key image frames and non-key image frames, and the key image frames are scheduled to ferroelectric memory, while the non-key image frames are scheduled to solid-state memory.

[0009] During transmission, the power supply status is continuously monitored. When an abnormal power supply is detected, a data protection program is activated to convert the key image frames stored in the ferroelectric memory into a persistent storage format.

[0010] Based on the continuously monitored link transmission capacity parameters, when the transmission bandwidth reaches a preset threshold, key image frames that have undergone data protection processing are scheduled for priority transmission.

[0011] Based on the non-critical image frames stored in the solid-state memory, when the link transmission capability parameters are continuously improved, a selective retransmission procedure for the non-critical image frames is initiated.

[0012] By prioritizing the transmission of critical image frames and selectively retransmitting non-critical image frames, combined with adaptive link adjustment, complete transmission of airborne image data can be achieved under unstable link conditions.

[0013] Optionally, the selective retransmission procedure for non-critical image frames, based on those stored in the solid-state memory, when a continuous improvement in link transmission capability parameters is detected, includes:

[0014] Based on the non-critical image frames stored in the solid-state memory, the link monitoring module continuously acquires the changing trend of the link transmission capability parameters.

[0015] Based on the aforementioned trend, analyze the degree of improvement in the link transmission status. When the analysis results show that the link transmission capability parameters have maintained stable improvement for multiple consecutive monitoring cycles, initiate the selective retransmission procedure.

[0016] According to the selective retransmission procedure, non-critical image frames that need to be retransmitted are selected from the solid-state memory;

[0017] According to the preset importance assessment criteria, the non-critical image frames are organized into a retransmission data queue.

[0018] Non-critical image frames are transmitted through the downlink data transmission interface in the order of the retransmission data queue.

[0019] Optionally, analyzing the degree of improvement in link transmission status based on the changing trend includes:

[0020] Obtain a sequence of link transmission capability parameter values ​​collected over multiple monitoring periods, and calculate the parameter change values ​​between adjacent monitoring periods in the link transmission capability parameter value sequence to generate a parameter change value sequence.

[0021] The positive change count is performed on the parameter change value sequence, and the number of monitoring periods in which the parameter change value is consecutively positive is counted.

[0022] When the number of consecutive monitoring cycles in which the parameter change value is positive reaches a first predetermined number, the average value of the parameter change value within the consecutive monitoring cycles is calculated.

[0023] The average value is compared with a second predetermined threshold. If the average value is greater than or equal to the second predetermined threshold, it is determined that the degree of improvement in the link transmission status meets the stable improvement condition.

[0024] Optionally, the method of prioritizing the transmission of key image frames and selectively retransmitting non-key image frames, combined with link adaptive adjustment, to achieve complete airborne image data transmission under unstable link conditions includes:

[0025] A collaborative working framework is established for the priority transmission mechanism of the key image frames and the selective retransmission mechanism of the non-key image frames. Under the collaborative working framework, the runtime order of the priority transmission mechanism and the selective retransmission mechanism is dynamically coordinated based on the link status data.

[0026] By utilizing a link adaptive adjustment mechanism, the priority transmission mechanism is activated first when the link quality deteriorates, while the selective retransmission mechanism is temporarily suspended.

[0027] The priority transmission mechanism and the selective retransmission mechanism are run synchronously when the link quality improves, so as to achieve complete transmission of airborne image data under unstable link conditions.

[0028] Optionally, the step of continuously monitoring the power supply status during transmission and activating a data protection program when a power supply abnormality is detected, converting the key image frames stored in the ferroelectric memory into a persistent storage format, includes:

[0029] The output voltage parameters of the power supply system are continuously collected through a voltage sampling circuit, and the output current parameters are collected through a current sampling circuit.

[0030] The collected output voltage parameters and output current parameters are compared item by item with the preset voltage threshold range and current threshold range;

[0031] When the output voltage parameter or the output current parameter continuously exceeds the corresponding threshold range and reaches a predetermined duration, the data protection program is triggered.

[0032] After the data protection program is started, the current image data transmission process is interrupted and the transmission status information is saved. The interface controller of the ferroelectric memory is controlled by the FPGA to reorganize the data storage structure of the stored key image frames according to the physical characteristics of the ferroelectric memory.

[0033] The reorganized key image frames are written to the non-volatile storage area, completing the conversion to persistent storage format.

[0034] Optionally, the step of prioritizing the transmission of critical image frames that have undergone data protection processing when the transmission bandwidth reaches a preset threshold, based on the continuously monitored link transmission capacity parameters, includes:

[0035] Based on the continuously monitored link transmission capability parameters, bandwidth measurement data is obtained through the downlink status feedback channel;

[0036] The bandwidth measurement data is periodically compared and analyzed with a preset transmission threshold. When the analysis results show that the bandwidth measurement data exceeds the transmission threshold for multiple consecutive sampling periods, a priority transmission scheduling mechanism is activated.

[0037] According to the priority transmission scheduling mechanism, the key image frame, which has undergone data protection processing, is read from the non-volatile storage area of ​​the ferroelectric memory;

[0038] According to the transmission protocol specification, the key image frame is encapsulated into a transmission data unit of a specific format, and the transmission data unit is preferentially scheduled for transmission through the downlink data transmission interface.

[0039] Optionally, the monitoring of the downlink status between the aircraft and the ground station to obtain link transmission capability parameters, and the dynamic adjustment of the image data to be transmitted via FPGA based on the link transmission capability parameters, includes:

[0040] Link status data is obtained through the feedback channel of the downlink, and bandwidth and bit error rate parameters are extracted from the link status data.

[0041] The link transmission capacity evaluation coefficient is calculated based on the bandwidth index parameter and the bit error rate index parameter, and the compression ratio parameter of the image data is determined based on the link transmission capacity evaluation coefficient.

[0042] The compression parameters of the image encoder are configured via the FPGA, and the image data to be transmitted is encoded according to the compression ratio parameters.

[0043] Secondly, this application provides an airborne high-speed image storage control system based on FPGA, comprising:

[0044] The adjustment module is used to monitor the downlink status between the aircraft and the ground station to obtain the link transmission capability parameters, and dynamically adjust the image data to be transmitted through the FPGA according to the link transmission capability parameters.

[0045] The scheduling module is used to divide the adjusted image data to be transmitted into key image frames and non-key image frames, and schedule the key image frames to ferroelectric memory and the non-key image frames to solid-state memory.

[0046] The conversion module is used to continuously monitor the power supply status during transmission. When an abnormal power supply is detected, the data protection program is activated to convert the key image frames stored in the ferroelectric memory into a persistent storage format.

[0047] The transmission module is used to schedule the transmission of key image frames that have undergone data protection processing for priority transmission when the transmission bandwidth reaches a preset threshold, based on the continuously monitored link transmission capacity parameters.

[0048] The startup module is used to initiate a selective retransmission procedure for non-critical image frames based on the non-critical image frames stored in the solid-state memory when the link transmission capability parameters are continuously improved.

[0049] The second transmission module is used to cooperate with the priority transmission of the key image frames and the selective retransmission of the non-key image frames, combined with link adaptive adjustment, to achieve complete transmission of airborne image data under unstable link conditions.

[0050] Thirdly, this application provides an electronic device, comprising:

[0051] Memory, used to store computer programs;

[0052] A processor, used to execute the computer program to implement the steps of the FPGA-based airborne high-speed image storage control method as described in the first aspect above.

[0053] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the FPGA-based airborne high-speed image storage control method described in the first aspect above.

[0054] The FPGA-based airborne high-speed image storage control method provided in this application obtains link transmission capability parameters by monitoring the downlink status between the aircraft and the ground station. Based on these parameters, it dynamically adjusts the image data to be transmitted via the FPGA, enabling perception and rapid response to downlink transmission status. The method dynamically adapts to link fluctuations through hardware logic, providing a decision-making basis for data scheduling. By dividing the adjusted image data to be transmitted into critical and non-critical image frames, and scheduling critical image frames to ferroelectric memory and non-critical image frames to solid-state memory, it achieves hierarchical data storage management. High-value critical data is stored in power-loss-resistant non-volatile memory to ensure its absolute security, while non-critical data is stored in large-capacity memory to optimize storage resource costs and efficiency. Furthermore, by continuously monitoring the power supply status during transmission, a data protection program is activated when a power supply anomaly is detected. The system converts critical image frames stored in ferroelectric memory into a persistent storage format, providing an active protection mechanism against power outages and ensuring the security and recoverability of core data in extreme situations. By prioritizing the transmission of critical image frames after data protection processing based on continuously monitored link transmission capacity parameters, when the transmission bandwidth reaches a preset threshold, it ensures that high-priority data can be transmitted reliably when link conditions permit. Furthermore, by initiating a selective retransmission procedure for non-critical image frames stored in solid-state memory when continuous improvement in link transmission capacity parameters is detected, it intelligently initiates data retransmission operations during link quality recovery periods. Through the combination of priority transmission of critical image frames and selective retransmission of non-critical image frames, along with adaptive link adjustments, the system ultimately ensures the integrity and reliability of image data in complex and harsh airborne transmission environments.

[0055] Furthermore, the core weighted step quantifies the degree of improvement in link status by continuously monitoring the changing trends of link capability parameters. Specifically, it calculates the parameter changes over multiple consecutive monitoring periods, counts the number of periods with continuous positive improvement, and calculates the average improvement magnitude. When both the continuity and significance conditions are met, the link is determined to have entered a stable improvement state and a retransmission is triggered. Subsequently, non-critical image frames to be retransmitted are selected from the solid-state storage and organized into a queue based on preset importance criteria and transmitted in sequence. This mechanism can accurately identify the true recovery trend of the link status, avoid ineffective retransmissions during temporary or fluctuating quality improvements, and intelligently select the optimal time to start the retransmission process. By transmitting in an orderly manner according to importance, it ensures that limited retransmission bandwidth is prioritized for compensating more important non-critical data, thereby efficiently and reliably improving the final transmission integrity rate of the overall data. Attached Figure Description

[0056] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 A flowchart illustrating an airborne high-speed image storage control method based on FPGA provided in this application embodiment;

[0058] Figure 2 A flowchart illustrating a specific implementation of an airborne high-speed image storage control method based on FPGA, provided in this application embodiment;

[0059] Figure 3 A schematic diagram illustrating a specific implementation of an airborne high-speed image storage control method based on FPGA provided in this application embodiment;

[0060] Figure 4 This is a schematic diagram of an airborne high-speed image storage control system based on FPGA, provided as an embodiment of this application. Detailed Implementation

[0061] In applications such as UAV aerial remote sensing, existing software-defined storage control schemes based on high-performance airborne processors can achieve a certain degree of adaptive data transmission, but their core control logic relies on software processing, resulting in three inherent defects: First, the latency of software processing causes the system to be unable to respond to high-speed data streams and rapid fluctuations in the link, resulting in low data transmission efficiency; second, complex control algorithms continuously consume a large amount of processor resources, causing resource conflicts with other onboard tasks; and third, the stability of the entire system is constrained by the operating system, and it is prone to freezing or restarting in harsh airborne environments such as vibration and temperature differences, leading to a high risk of transmission interruption.

[0062] To address the aforementioned issues, this application proposes an FPGA-based airborne high-speed image storage control method. By combining hardware processing with intelligent scheduling strategies, it fundamentally improves the system's real-time performance, reliability, and resource utilization efficiency. The method first utilizes the FPGA to monitor the downlink status and dynamically adjust the transmission strategy. Then, image data is divided into key frames and non-key frames, stored in ferroelectric memory and solid-state memory respectively, achieving hierarchical storage and protection. When power supply is abnormal, key frames are automatically converted to a persistent format for protection and prioritized for transmission when the link recovers. Simultaneously, by continuously analyzing the link status change trend, selective retransmission of non-key frames is intelligently initiated after the link stability is determined to have improved. This solution uses FPGA hardware to replace software to implement core control functions, significantly reducing system latency and processor load. Furthermore, through hierarchical storage and adaptive retransmission mechanisms, it effectively overcomes the reliability deficiencies of software systems under harsh environments, ensuring the complete and reliable transmission of airborne image data under unstable link conditions.

[0063] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0064] The core of this application is to provide an airborne high-speed image storage control method based on FPGA, and a flowchart of one specific implementation is shown below. Figure 1 As shown, the method includes:

[0065] S101. Monitor the downlink status between the aircraft and the ground station to obtain the link transmission capability parameters, and dynamically adjust the image data to be transmitted through the FPGA according to the link transmission capability parameters.

[0066] Optionally, S101 may specifically include the following steps:

[0067] S1011. Obtain link status data through the feedback channel of the downlink, and extract bandwidth index parameters and bit error rate index parameters from the link status data;

[0068] S1012. Calculate the link transmission capacity evaluation coefficient based on the bandwidth index parameter and the bit error rate index parameter, and determine the compression ratio parameter of the image data based on the link transmission capacity evaluation coefficient.

[0069] S1013. Configure the compression parameters of the image encoder through the FPGA, and encode the image data to be transmitted according to the compression ratio parameters.

[0070] In the above scheme, the downlink data link between the aircraft and the ground station refers to the communication link through which the aircraft transmits data to the ground station, and is the core channel for airborne data backhaul. Link transmission capacity parameters refer to a set of key indicators reflecting the link's transmission performance, including bandwidth and bit error rate (BER) parameters. Bandwidth measures the amount of data that can be transmitted per unit time, while BER measures the proportion of erroneous bits in the transmitted data. FPGA refers to an integrated circuit with flexibly configurable hardware logic, possessing high-speed parallel processing capabilities, and can be used to adjust the data processing flow. Image data to be transmitted refers to the raw or pre-processed image data acquired by the imaging equipment onboard the aircraft and to be transmitted to the ground station. An image encoder is a device or module used to compress and encode image data, reducing the amount of image data through specific algorithms for easier transmission and storage. Compression ratio parameters measure the degree of image data compression, determining the ratio of compressed data to the original data. The link transmission capacity evaluation coefficient is a comprehensive coefficient calculated based on bandwidth and BER parameters, used to quantitatively evaluate the overall transmission capacity of the link and provide a basis for determining the image compression ratio.

[0071] In this embodiment, firstly, through S1011, a link status monitoring chip and a data acquisition algorithm are used to collect various status information such as signal strength, data transmission delay, and packet loss during the link transmission process via the feedback channel in the downlink data between the aircraft and the ground station, according to a preset sampling period. The collected information is temporarily stored by a data caching module and integrated according to a preset format to form structured link status data. Then, a parameter extraction algorithm is used to analyze the link status data, filtering out information fields directly related to the link transmission capability. Finally, a data parsing module extracts bandwidth index parameters and bit error rate index parameters from the information fields. For example, in a scenario where an aircraft is performing an in-flight mission, the feedback channel collects link status information such as signal strength and transmission delay every 100 milliseconds through the link status monitoring chip. After caching and integration, link status data is formed. Then, the parameter extraction algorithm filters and parses the data, finally extracting the current link's bandwidth index parameter as 2 Mbps and the bit error rate index parameter as 0.001 (i.e., 1 error bit in every 1000 bits of data).

[0072] Secondly, through S1012, based on the preset weighted summation calculation model, the bandwidth index parameter and bit error rate index parameter are substituted into the calculation to obtain the link transmission capability evaluation coefficient. Specifically, the Min-Max standardization algorithm is first used to process the bandwidth index parameter and bit error rate index parameter separately, with the formulas: Standardized bandwidth = Actual bandwidth / Preset maximum bandwidth, Standardized bit error rate = 1 - Actual bit error rate / Preset maximum bit error rate. Then, the standardized parameter values ​​are mapped to the 0-1 range to eliminate the influence of different parameter units. Next, the influence weights of the two parameters are preset according to the link transmission characteristics (e.g., bandwidth has a greater impact on transmission capability, so the weight is set to 0.6, and the bit error rate weight is set to 0.4). The weighted summation formula is used: Link transmission capability evaluation coefficient = Standardized bandwidth index parameter × Bandwidth weight + Standardized bit error rate index parameter × Bit error rate weight. The evaluation coefficient is then calculated, and then... The lookup table method calls the pre-stored table of the correspondence between evaluation coefficients and compression ratio parameters in the FPGA storage unit. Based on the calculated link transmission capacity evaluation coefficient, the image data compression ratio parameter adapted to the current link transmission capacity is determined. For example, in a scenario where an aircraft is performing an aerial mission, the link bandwidth is 2Mbps and the bit error rate is 0.001. First, the bandwidth is standardized using the Min-Max normalization algorithm. Assuming the preset maximum bandwidth is 2.5Mbps, the standardized bandwidth is calculated to be 2÷2.5=0.8. Then, the bit error rate is standardized. Assuming the preset maximum bit error rate is 0.01, the standardized bit error rate is calculated to be 1-0.001÷0.01=0.9. Substituting these values ​​into the weighted summation formula, the link transmission capacity evaluation coefficient is 0.8×0.6+0.9×0.4=0.84. Finally, the corresponding table is called to determine that the compression ratio parameter corresponding to the evaluation coefficient of 0.84 is 1:2.

[0073] Finally, through S1013, the FPGA's built-in hardware logic configuration module and high-speed data interface are used to send control signals to the image encoder to configure the encoder's compression parameters. The FPGA transmits the compression ratio parameters and the corresponding quantization parameters and prediction mode parameters to the image encoder's control register via a parallel data bus to complete the parameter configuration. Afterward, the image encoder uses an image compression algorithm to perform discrete wavelet transform, quantization, entropy coding, and other processing on the image data to be transmitted according to the compression ratio parameters, compressing the original image data to a data volume that is suitable for the current link's transmission capacity.

[0074] In practical applications, a certain type of aircraft carrying imaging equipment performs aerial observation missions. The acquired image data needs to be transmitted to the ground station via a downlink. During the transmission preparation phase, the link status monitoring function is first activated. Through the downlink feedback channel, a link status monitoring chip collects status information such as signal strength, transmission delay, and packet loss at a sampling period of 100 milliseconds / sample. The collected information is temporarily stored by a data caching module and integrated into structured link status data. Then, a feature-based parameter extraction algorithm is used to filter and parse the current link's bandwidth parameter (1.8 Mbps) and bit error rate parameter (0.002) from the link status data. Next, according to the preset link transmission capacity evaluation coefficient calculation method, the Min-Max normalization algorithm is used to process the bandwidth and bit error rate parameters. Assuming the preset maximum bandwidth is 2.5 Mbps, the actual bandwidth of 1.8 Mbps is substituted into the normalization formula, resulting in a normalized bandwidth of 1.8 ÷ 2.5 = 0.72. Assuming the preset maximum bit error rate is 0.01, the actual bit error rate of 0.002 is substituted into the normalization formula, yielding... The standardized bit error rate is calculated as 1 - 0.2 = 0.8. Then, a weighted summation formula is used to calculate the evaluation coefficient, with a preset bandwidth weight of 0.6 and a bit error rate weight of 0.4. First, 0.72 × 0.6 = 0.432 and 0.8 × 0.4 = 0.32 are calculated, then added together: 0.432 + 0.32 = 0.752, yielding a link transmission capacity evaluation coefficient of 0.752. The table of evaluation coefficients and compression ratios stored in the FPGA is then used to determine the compression ratio parameter corresponding to the evaluation coefficient of 0.752 as 1:3. Subsequently, the FPGA uses its built-in hardware... The component logic configuration module generates control signals and transmits quantization parameters and prediction mode parameters adapted to a 1:3 compression ratio to the control register of the onboard image encoder via a parallel data bus, thus completing the encoder parameter configuration. According to the configuration parameters, the image encoder uses an image compression algorithm to perform discrete wavelet transform, quantization, and entropy coding on the original image data with a resolution of 2048×1536 acquired by the imaging device, compressing the original image data to a data volume adapted to the current link transmission capacity, completing the dynamic adjustment of the image data to be transmitted, and preparing for subsequent data transmission.

[0075] The overall solution of S101 described above achieves precise control of the entire process of link status monitoring, parameter calculation, and image data adjustment through technologies such as dedicated monitoring chips, standardized algorithms, and FPGA hardware configuration. On the one hand, by using link status monitoring chips and parameter extraction algorithms, link bandwidth and bit error rate parameters can be accurately collected and extracted. Combined with Min-Max standardization and weighted summation algorithms, the link transmission capacity can be quantitatively evaluated, avoiding transmission problems caused by ambiguity in link status judgment. On the other hand, by utilizing the hardware configuration capabilities of FPGA and image compression algorithms, image compression parameters can be dynamically adjusted according to link capabilities, ensuring that image data can adapt to link transmission requirements while retaining key information to the greatest extent. This effectively solves the problems of easy interruption of data transmission and low image availability under unstable links. At the same time, it provides highly adaptable pre-processed data for subsequent data classification, storage, and priority transmission, significantly improving the reliability and adaptability of airborne image data transmission.

[0076] S102. Divide the adjusted image data to be transmitted into key image frames and non-key image frames, and schedule the key image frames to ferroelectric memory and the non-key image frames to solid-state memory.

[0077] Optionally, S102 may specifically include the following steps:

[0078] S1021. Analyze the content feature distribution of the image data to be transmitted, identify the range of image regions containing important information, mark the image regions containing important information as key image frames, and mark the remaining image regions as non-key image frames.

[0079] S1022. Using an FPGA-controlled data scheduler, the critical image frames are written into the ferroelectric memory, and the non-critical image frames are written into the solid-state memory.

[0080] In the above scheme, the adjusted image data to be transmitted refers to the image encoded data adapted to the current link transmission capability after the compression ratio is dynamically adjusted by the FPGA; the key image frame refers to the image frame containing information that is important to the mission objective; the non-key image frame refers to the image frame containing secondary or redundant information other than the key image frame; the content feature distribution refers to the distribution of information attributes, texture complexity, target proportion, and other features in different regions of the image data; important information refers to information related to the core objective of the mission; ferroelectric memory refers to non-volatile memory that uses ferroelectric materials as the storage medium; solid-state memory refers to memory based on solid-state electronic components such as flash memory chips; the data scheduler refers to the module used to control the allocation and transmission of data between different storage devices; and FPGA control refers to the generation of control signals through the logic control module built into the FPGA.

[0081] In this embodiment of the application, firstly, through S1021, an image content analysis algorithm is used to parse the adjusted image data to be transmitted frame by frame, extracting the content feature distribution of each frame of the image, including features such as target contours, texture complexity, and grayscale value change rate of each region of the image. Then, through a preset important information determination rule, the extracted content features are evaluated to identify the range of image regions containing important information. Finally, based on the identification results, the image frames containing the image regions containing important information are marked as key image frames, and the image frames containing the remaining image regions that do not contain important information or whose important information ratio is lower than a threshold are marked as non-key image frames.

[0082] Secondly, the FPGA's built-in logic control module generates a data scheduling instruction via S1022 and sends it to the data scheduler. The data scheduling instruction contains the identification information of key image frames and non-key image frames, as well as the corresponding storage device addresses. After receiving the instruction, the data scheduler first reads the data marked as key image frames through the high-speed data transmission interface, writes it into the designated storage block of the ferroelectric memory according to the ferroelectric memory address in the instruction, and ensures the accuracy of data writing through a CRC check algorithm. Then, it reads the data marked as non-key image frames, writes it into the designated storage block of the solid-state memory according to the solid-state memory address in the data scheduling instruction, and verifies the data integrity through a CRC check algorithm.

[0083] In practical applications, an aircraft equipped with imaging equipment performs aerial observation missions. First, the image content analysis module is activated. Using a lightweight target detection algorithm based on YOLO, 20 frames of 1920×1080 resolution coded image data (adjusted and compressed to suit link transmission capabilities) are analyzed frame by frame. The contour features, grayscale change rate, and target percentage (target region pixels / total image pixels) of the target region in each frame are extracted. According to the mission's preset rules, if an image contains a Class C target and its percentage is ≥5%, it is considered a critical image frame; otherwise, it is a non-critical image frame. Analysis revealed that four frames (Frame_003, Frame_007, Frame_012, and Frame_018) contain Class C targets. The target percentage in Frame_003 is calculated as ≥5% using the ratio of the target region pixels to the total image pixels (1920×1080). The remaining three frames also meet the ≥5% target percentage requirement. Therefore, these 4 frames are marked as critical image frames, and the remaining 16 image frames that do not contain Class C targets or whose target proportion is less than 5% are marked as non-critical image frames. Next, the logic control module built into the FPGA generates a data scheduling instruction, which specifies the identifiers of the 4 critical image frames and the designated addresses of the ferroelectric memory, and the identifiers of the 16 non-critical image frames and the designated addresses of the solid-state memory. The instruction is then sent to the data scheduler. After receiving the instruction through the SPI interface, the data scheduler first reads the data of Frame_003, writes it according to the designated address of the ferroelectric memory, and calculates the data check value through the CRC check algorithm. If the data matches the check value sent by the FPGA, the writing is confirmed to be successful. The writing of the other 3 critical image frames is completed in sequence. Then, the data of Frame_001 (non-critical image frame) is read and written according to the designated address of the solid-state memory. The CRC check is also completed. The writing of the remaining 15 non-critical image frames is completed in sequence, realizing the classified storage of the two types of image frames.

[0084] The overall solution of S102 described above achieves differentiated management of image data to be transmitted through image content analysis and classified storage. Employing a professional image content analysis algorithm, it accurately identifies key image frames containing important information, avoiding the omission of important data or misclassification of redundant data due to misjudgment, thus ensuring the accuracy of data classification. Using an FPGA-controlled data scheduler, key image frames are stored in ferroelectric memory with fast read / write speeds and strong anti-interference capabilities, while non-key image frames are stored in large-capacity, stable solid-state memory. This ensures both rapid access and secure storage of key data, while also making reasonable use of the advantages of different memory types, avoiding waste of storage resources. This lays the foundation for critical data protection in case of power failure and selective retransmission after link improvement, enhancing the targeting and reliability of airborne image data storage.

[0085] S103. During transmission, the power supply status is continuously monitored. When an abnormal power supply is detected, the data protection program is activated to convert the key image frames stored in the ferroelectric memory into a persistent storage format.

[0086] Optionally, S103 may specifically include the following steps:

[0087] S1031. The output voltage parameters of the power supply system are continuously collected through the voltage sampling circuit, and the output current parameters are collected through the current sampling circuit.

[0088] S1032. The collected output voltage parameters and output current parameters are compared item by item with the preset voltage threshold range and current threshold range;

[0089] S1033. When the output voltage parameter or the output current parameter continuously exceeds the corresponding threshold range and reaches a predetermined duration, the data protection program is triggered.

[0090] S1034. After starting the data protection program, interrupt the current image data transmission process and save the transmission status information. Control the interface controller of the ferroelectric memory through the FPGA to reorganize the stored key image frames according to the physical characteristics of the ferroelectric memory.

[0091] S1035. Write the reorganized key image frames into the non-volatile storage area to complete the conversion of persistent storage format.

[0092] In the above scheme, power supply status refers to the working status of the aircraft power supply system; power supply anomaly refers to the situation where the voltage or current output by the power supply system exceeds the normal operating range; data protection program refers to the program used to protect critical data from loss in abnormal scenarios; persistent storage format refers to the storage format that allows data to be preserved for a long time after the device is powered off; voltage sampling circuit refers to the circuit module used to collect the output voltage value of the power supply system, and the output voltage value refers to the voltage data collected by the circuit; current sampling circuit refers to the circuit module used to collect the output current value of the power supply system, and the output current parameter refers to the current data collected by the circuit; preset voltage threshold range and current threshold range refer to the preset range of voltage and current values ​​that the power supply system should be in when it is working normally; preset duration refers to the duration for which the parameters required to determine power supply anomaly exceed the threshold range; transmission status information refers to the progress, position, and other information of the current image data transmission; ferroelectric memory interface controller refers to the component that controls the data interaction between the ferroelectric memory and other modules; physical characteristics of ferroelectric memory refer to the inherent attributes of ferroelectric memory such as the way data is stored and the read / write speed; data storage structure refers to the way data is organized and arranged in the memory; non-volatile storage area refers to the storage area where data can still be retained after power failure.

[0093] In this embodiment, firstly, the output voltage parameters of the power supply system are continuously collected by the series voltage divider sampling circuit according to a preset period in S1031, and the output current parameters of the power supply system are simultaneously collected by the series shunt resistor sampling circuit. The collected parameters are then transmitted to the data processing module for temporary storage. For example, in a scenario where an aircraft is performing an aerial observation mission, the voltage sampling circuit collects the output voltage of the power supply system every 50 milliseconds, and the current sampling circuit collects the output current simultaneously. The collected voltage and current data are then sent to the data processing module of the FPGA.

[0094] Secondly, in S1032, the output voltage parameter and output current parameter are read from the temporary storage area of ​​the data processing module through a threshold range comparison algorithm. The output voltage parameter is compared with a preset voltage threshold range according to a preset logic, and the output current parameter is compared with a preset current threshold range item by item. The comparison result is fed back to the anomaly judgment module.

[0095] Then, in S1033, the anomaly determination module performs statistics on all comparison results. When the output voltage parameter or output current parameter is detected to continuously exceed the corresponding threshold range and the duration of the exceedance reaches the preset duration, an anomaly trigger signal is generated through continuous over-limit detection to trigger the data protection program.

[0096] Next, after the data protection program is started via S1034, an interrupt signal is sent to the data transmission module via the transmission interrupt control protocol to interrupt the current image data transmission process. At the same time, the transmission progress, current transmission frame identifier, and other information in the transmission module are read through the state snapshot storage technology, and the information is saved to the temporary storage area in the form of a data snapshot. Subsequently, the FPGA generates control signals through the built-in hardware logic programming module and sends them to the interface controller of the ferroelectric memory. The interface controller adopts a page-by-page data reorganization algorithm to reorganize the data of key image frames stored in the ferroelectric memory page by page according to the physical characteristics of the ferroelectric memory, and adjusts the data storage structure to adapt to the requirements of non-volatile storage.

[0097] Finally, via S1035, for the interface controller, the timing control write algorithm is used to write the reorganized key image frame data into the non-volatile storage area of ​​the ferroelectric memory according to the preset write timing and address. At the same time, the CRC check value is calculated and compared with the preset value to confirm the data integrity and ensure that the data is not lost after power failure, thus completing the conversion of persistent storage format.

[0098] In practical applications, an aircraft equipped with a power supply system performs aerial observation missions. First, a series voltage divider sampling circuit collects the output voltage parameters of the B-type power supply system every 50 milliseconds, while a series shunt resistor sampling circuit simultaneously collects the output current parameters. The collected data is transmitted to the FPGA's data processing module. Second, the data processing module compares the collected voltage parameters with a preset C value range (normal power supply voltage range) and the current parameters with a preset D value range (normal power supply current range). If any parameter exceeds the corresponding range, an anomaly is recorded. Then, when a voltage parameter is detected to be continuously above the set range for 6 consecutive seconds... When the value of C exceeds the predetermined duration of 300 milliseconds, the anomaly detection module triggers the data protection program. The data protection program then interrupts the current image transmission, saves the progress information of the current transmission to Frame_008, and the FPGA sends a control signal to the ferroelectric memory interface controller. Based on the page-based storage characteristic of the ferroelectric memory, the interface controller reassembles the three key image frames in the ferroelectric memory into a page-based data storage structure. Finally, the interface controller writes the reassembled key image frame data into the non-volatile storage area of ​​the ferroelectric memory. After writing, the data integrity is verified, completing the persistent storage format conversion.

[0099] The overall solution of S103 described above achieves timely protection of critical data by continuously monitoring the power supply status. It can capture abnormal power supply situations and avoid the loss of critical image frames due to power supply problems. By converting critical image frames into a persistent storage format, it ensures that critical data can still be retained after power failure. When transmission is interrupted, it saves transmission status information to facilitate subsequent transmission recovery, thus ensuring the safety and availability of airborne critical image data in scenarios with abnormal power supply.

[0100] S104. Based on the continuously monitored link transmission capability parameters, when the transmission bandwidth reaches a preset threshold, key image frames that have undergone data protection processing are scheduled for priority transmission.

[0101] Optionally, S104 may specifically include the following steps:

[0102] S1041. Based on the continuously monitored link transmission capability parameters, obtain bandwidth measurement data through the downlink status feedback channel;

[0103] S1042. Periodically compare and analyze the bandwidth measurement data with a preset transmission threshold. When the analysis results show that the bandwidth measurement data exceeds the transmission threshold for multiple consecutive sampling periods, start the priority transmission scheduling mechanism.

[0104] S1043. According to the priority transmission scheduling mechanism, the key image frame that has undergone data protection processing is read from the non-volatile storage area of ​​the ferroelectric memory;

[0105] S1044. The key image frame is encapsulated into a transmission data unit of a specific format according to the transmission protocol specification, and the transmission data unit is preferentially scheduled for transmission through the downlink data transmission interface.

[0106] In the above scheme, the link transmission capacity parameter refers to the set of performance indicators obtained by continuously monitoring the downlink status of the aircraft and ground station; the transmission bandwidth refers to the amount of data that the link can transmit per unit time; the bit error rate refers to the ratio of the number of erroneous bits in the transmitted data to the total number of bits; the preset threshold refers to the minimum transmission bandwidth value that is pre-set to determine that the link meets the conditions for priority transmission; the key image frame after data protection processing refers to the key image frame that is converted into a persistent storage format and stored in the non-volatile storage area of ​​the ferroelectric memory when the power supply is abnormal; the bandwidth measurement data refers to the specific data reflecting the current link transmission bandwidth obtained from the link status feedback; the downlink status feedback channel refers to the communication channel used to transmit back the link transmission status information; the transmission threshold refers to the preset bandwidth standard used to determine whether the link meets the requirements for priority transmission; the priority transmission scheduling mechanism refers to the rules and procedures for prioritizing the transmission of key data when the link conditions meet the standards; the transmission protocol specification refers to the standard that specifies the data transmission format, timing, and verification method; the transmission data unit of a specific format refers to the data block that can be directly transmitted through the link after being encapsulated according to the transmission protocol specification; and the data sending interface refers to the hardware interface used to send the transmission data unit to the downlink.

[0107] In this embodiment, firstly, through S1041, based on the continuous monitoring function of the link status monitoring module, the bandwidth measurement data in the link transmission capability parameters is obtained using the downlink status feedback channel and data acquisition technology. During the acquisition process, bit error rate data is acquired simultaneously to assist in judging link stability. The acquired bandwidth measurement data is then transmitted to the data processing module for temporary storage. For example, in a scenario where an aircraft is performing an aerial observation mission, the link status monitoring module collects bandwidth measurement data and bit error rate data every 100 milliseconds through the status feedback channel. In one instance, the acquired bandwidth measurement data is 4 Mbps, and the bit error rate data is in a low stability range. The data is then sent to the FPGA data processing unit.

[0108] Secondly, in step S1042, a periodic comparison algorithm is used to compare the temporarily stored bandwidth measurement data with the preset transmission threshold periodically. At the same time, the link stability is judged in conjunction with the bit error rate data. If the bandwidth measurement data exceeds the transmission threshold and the bit error rate remains stable for multiple consecutive sampling periods, it is determined that the link has the conditions for priority transmission, and the priority transmission scheduling mechanism is triggered. For example, if the preset transmission threshold is 3Mbps, the data processing module compares the collected bandwidth measurement data with 3Mbps every 100 milliseconds. The bandwidths collected in three consecutive sampling periods (a total of 300 milliseconds) are 4Mbps, 4.2Mbps, and 4.1Mbps, respectively, all of which exceed the transmission threshold and the bit error rate is stable. At this time, the priority transmission scheduling mechanism is started.

[0109] Next, according to the data reading rules in the priority transmission scheduling mechanism, S1043 causes the FPGA to generate a read control signal and sends the read control signal to the interface controller of the ferroelectric memory. The interface controller uses non-volatile region read technology to read key image frames that have undergone data protection processing from the non-volatile storage area of ​​the ferroelectric memory. During the read process, CRC check technology is used to verify data integrity to ensure that the read key image frames are undamaged.

[0110] Finally, according to the preset transmission protocol specifications, S1044 uses a data encapsulation algorithm to encapsulate the key image frames read by the data processing module into transmission data units of a specific format. During the encapsulation process, a frame header (containing frame identifier and length information) and a frame trailer (containing checksum) are added. After encapsulation, the FPGA generates a transmission control signal and uses priority scheduling technology to send the transmission data units to the link through the downlink data transmission interface, ensuring that the key image frames are transmitted before other data. For example, according to the transmission protocol specifications, each key image frame is encapsulated into a transmission data unit containing a 16-byte frame header and a 4-byte checksum. The FPGA controls the data transmission interface to send these transmission data units first, and other data is scheduled after the key image frames are transmitted, thus achieving priority transmission.

[0111] In practical applications, an aircraft carrying imaging equipment performs aerial observation missions. First, the link status monitoring module collects bandwidth measurement data and bit error rate data every 100 milliseconds through the downlink status feedback channel. The collected bandwidth data is transmitted to the FPGA data processing module. In one instance, the bandwidth was 3.8 Mbps, and the bit error rate was within a stable range. Second, the data processing module periodically compares the bandwidth data with a preset 3 Mbps transmission threshold. The bandwidths for three consecutive sampling periods (300 milliseconds) were 3.8 Mbps, 3.9 Mbps, and 4 Mbps, respectively, all exceeding the threshold and with stable bit error rates, thus initiating a priority transmission scheduling mechanism. Third, the FPGA sends a read control signal to the ferroelectric memory interface controller. The interface controller reads four data-protected key image frames from the non-volatile storage area, and each frame is checked for completeness using a CRC check after reading. Finally, the data processing module adds a frame header and checksum to each key image frame according to the transmission protocol specifications, encapsulating them into a specific format transmission data unit. The FPGA controls the data transmission interface to prioritize sending these transmission data units to the downlink, completing the priority transmission of the key image frames.

[0112] The overall solution of S104 described above achieves timely transmission of critical image frames by combining continuous monitoring and priority scheduling mechanisms for link transmission capability parameters; it can accurately judge link conditions based on bandwidth and bit error rate data, avoiding data loss caused by forced transmission when the link is unstable; through standardized encapsulation and priority transmission processes, it ensures that critical image frames are transmitted with priority when the link meets the standards, guaranteeing the timeliness of core data; the verification step when reading critical image frames further ensures data integrity, thus improving the timeliness and reliability of airborne critical image data transmission.

[0113] S105. Based on the non-critical image frames stored in the solid-state memory, when the link transmission capability parameters are continuously improved, a selective retransmission procedure for the non-critical image frames is initiated.

[0114] Optionally, S105 may specifically include the following steps:

[0115] Based on the non-critical image frames stored in the solid-state memory, the link monitoring module continuously acquires the changing trend of the link transmission capability parameters.

[0116] S1051. Based on the aforementioned trend, analyze the degree of improvement in the link transmission status. When the analysis results show that the link transmission capability parameters have maintained stable improvement for multiple consecutive monitoring cycles, initiate the selective retransmission procedure.

[0117] S1052. According to the selective retransmission procedure, select non-critical image frames that need to be retransmitted from the solid-state memory;

[0118] S1053. Based on the preset importance evaluation criteria, organize the non-critical image frames into a retransmission data queue.

[0119] S1054. Transmit non-critical image frames in the order of the retransmission data queue through the downlink data transmission interface.

[0120] In the above scheme, the link transmission capability parameters refer to the set of indicators reflecting the downlink transmission performance between the aircraft and the ground station; the link monitoring module refers to the functional module that continuously collects and processes the link transmission capability parameters; the change trend refers to the direction and magnitude of the change of the link transmission capability parameters over time; the degree of improvement in link transmission status refers to the degree to which the link transmission capability parameters change for the better; the selective retransmission procedure refers to the procedure that only retransmits non-critical image frames that have not been successfully transmitted or are of high importance; the importance assessment criterion refers to the rules for determining the retransmission priority of non-critical image frames; the retransmission data queue refers to the sequence of images to be transmitted formed by sorting the retransmission priority of non-critical image frames; and the downlink data transmission interface refers to the hardware interface used to send non-critical image frames to the ground station via the downlink.

[0121] In this embodiment, the link monitoring module first continuously collects link transmission capability parameters according to a preset cycle in step S1051, and analyzes the changing trend of the link transmission capability parameters using a parameter averaging method over multiple consecutive cycles. Then, the degree of improvement in the link transmission status is calculated based on the changing trend. Specifically, the average parameter value of the next cycle is compared with the average parameter value of the previous cycle. If the average bandwidth increases and the average bit error rate decreases, it is determined to be an improvement. When the average parameter values ​​of multiple consecutive monitoring cycles maintain a stable improvement state of increased bandwidth and decreased bit error rate, a retransmission start signal is generated to start the selective retransmission procedure.

[0122] Secondly, in step S1052, according to the preset data filtering rules in the selective retransmission procedure, the transmission status markers of all non-critical image frames in the solid-state memory are read, and then a marker matching algorithm is used to filter out the non-critical image frames marked as "not transmitted" or "transmission failed". The non-critical image frames are then identified as non-critical image frames that need to be retransmitted. For example, if 17 non-critical image frames are stored in the solid-state memory, of which 6 frames are marked as "not transmitted" and 2 frames are marked as "transmission failed", after filtering by the marker matching algorithm, these 8 frames are identified as non-critical image frames that need to be retransmitted.

[0123] Then, according to the preset importance evaluation criteria, the non-critical image frames that need to be retransmitted are scored one by one in S1053. The specific calculation method adopts the weighted summation formula, which is expressed as S=A×w1+B×w2, where S is the comprehensive importance score, A is the task relevance score, B is the texture complexity score, w1 is the relevance weight, and w2 is the complexity weight. After calculating the score of each frame, the non-critical image frames are sorted from high to low according to S to form a retransmission data queue. For example, if A=80 and B=70 for Frame_002, then S=80×0.6+70×0.4=76; if A=75 and B=65 for Frame_004, then S=75×0.6+65×0.4=71. The retransmission data queue is formed by sorting the frames according to S from high to low.

[0124] Finally, a communication connection is established between the data transmission module and the downlink data transmission interface via S1054. Then, a polling scheduling method based on queue order is adopted to read the non-critical image frame data in the retransmission data queue sequentially starting from the first frame, and transmit it to the ground station through the data transmission interface. At the same time, the reception confirmation signal from the ground station is received to ensure that the next frame is transmitted only after each frame is transmitted.

[0125] In practical applications, an aircraft equipped with imaging equipment performs aerial reconnaissance missions. After prioritizing the transmission of key image frames, the link monitoring module first collects link bandwidth and bit error rate every 100 milliseconds, calculating the average value every three cycles. The average bandwidth for the first three cycles is 1.5 Mbps, and the average bit error rate is 0.003; the average bandwidth for the second three cycles is 1.7 Mbps, and the average bit error rate is 0.0025; the average bandwidth for the third to fifth three cycles are 1.9 Mbps, 2.0 Mbps, and 2.1 Mbps respectively, with an average bit error rate of 0. The values ​​.002, 0.0018, and 0.0015 showed a stable improvement for five consecutive cycles, thus initiating the selective retransmission procedure. Subsequently, based on the selective retransmission procedure, the transmission status flags of 17 non-critical image frames in the solid-state memory were read, and 6 frames marked as "not transmitted" (Frame_002, Frame_003, Frame_006, Frame_008, Frame_010, Frame_012) and 2 frames marked as "transmission failed" (Frame_004, Frame_015), totaling 8 non-critical image frames that need to be retransmitted, were selected. Then, scores are assigned according to the preset importance evaluation criteria (target relevance 60% weight, texture complexity 40% weight), using the formula S = A × w1 + B × w2. Frame_002 has a target relevance score of 80 and a texture complexity score of 70, so S = 80 × 0.6 + 70 × 0.4 = 76. Frame_004 has a target relevance score of 75 and a texture complexity score of 65, so S = 75 × 0.6 + 65 × 0.4 = 71. The scores for the remaining 6 frames are calculated using the same method, and then sorted from highest to lowest score to form a retransmission data queue. Finally, the data transmission control module is connected to the downlink data transmission interface. The first frame of the queue, Frame_002, is sent first. After receiving the "reception successful" signal from the ground station, Frame_004 is sent, and so on, completing the transmission of 8 non-critical image frames. During this process, the reception status is confirmed after each frame transmission to ensure successful retransmission.

[0126] The overall solution of S105 described above achieves efficient utilization of link resources by initiating selective retransmission of non-critical image frames through continuous monitoring of link improvement status. It can accurately judge the link performance improvement trend and avoid resource waste caused by blind retransmission when the link is unstable. By filtering and sorting non-critical image frames according to their importance, it ensures that high-value non-critical data is transmitted first, thereby improving the overall data utilization rate. During the retransmission process, the reception confirmation mechanism is combined to ensure the reliability of non-critical image frame transmission, while not affecting the subsequent transmission of critical data. Overall, it optimizes the transmission efficiency and integrity of airborne image data.

[0127] S106. By prioritizing the transmission of the key image frames and selectively retransmitting the non-key image frames, combined with link adaptive adjustment, complete transmission of airborne image data is achieved under unstable link conditions.

[0128] Optionally, S106 may specifically include the following steps:

[0129] S1061. Establish a collaborative working framework for the priority transmission mechanism of the key image frames and the selective retransmission mechanism of the non-key image frames. Under the collaborative working framework, dynamically coordinate the runtime order of the priority transmission mechanism and the selective retransmission mechanism based on the link status data.

[0130] S1062. Using a link adaptive adjustment mechanism, the priority transmission mechanism is activated first when the link quality deteriorates, while the selective retransmission mechanism is temporarily suspended.

[0131] S1063. When the link quality improves, the priority transmission mechanism and the selective retransmission mechanism are run synchronously to achieve complete transmission of airborne image data under unstable link conditions.

[0132] In the above scheme, the collaborative working framework refers to the control structure that integrates the priority transmission mechanism for critical image frames and the selective retransmission mechanism for non-critical image frames; link status data refers to parameters reflecting downlink performance; runtime sequence refers to the time order rules for the start, pause, or synchronous operation of the priority transmission mechanism and the selective retransmission mechanism; the link adaptive adjustment mechanism refers to the functional modules that adjust the transmission mechanism according to the link status data, and can switch the mechanism's operating mode according to changes in link quality; link quality deterioration refers to a state in which the link bandwidth continuously decreases and the bit error rate continuously increases, exceeding the preset normal range; link quality improvement refers to a state in which the link bandwidth continuously increases and the bit error rate continuously decreases, stabilizing within the preset normal range; the priority transmission mechanism refers to the mechanism that schedules the priority transmission of critical image frames; the selective retransmission mechanism refers to the mechanism that selectively retransmits non-critical image frames; and complete transmission of airborne image data refers to the priority transmission of all critical image frames and the retransmission of non-critical image frames as needed, ultimately achieving the goal of transmitting all important data without loss.

[0133] In the embodiments of this application, such as Figure 2As shown, firstly, a collaborative working framework for the priority transmission mechanism and the selective retransmission mechanism is established using mechanism logic integration technology in S1061. This clarifies the rule that the two mechanisms cannot conflict when operating in the same link environment. The collaborative working framework receives link status data from the link monitoring module and dynamically adjusts the runtime sequence of the two mechanisms through status analysis logic: when the link status is good, the priority transmission mechanism and the selective retransmission mechanism run synchronously; when the link status is average, the priority transmission mechanism is scheduled to run first, and the selective retransmission mechanism is started after its completion, ensuring smooth and conflict-free cooperation between the two mechanisms.

[0134] Secondly, using the link adaptive adjustment mechanism in S1062, the link status data is monitored by the data acquisition unit. The data is divided using a link quality judgment algorithm. If, within three consecutive acquisition cycles, the bandwidth in the link status data is consistently below the lower limit of the "low bandwidth range" and the bit error rate is consistently above the preset upper limit of the bit error rate, it is determined that the link quality has deteriorated. Then, the link adaptive adjustment mechanism generates a control signal to activate the priority transmission mechanism, allowing the key image frames to occupy link resources first. At the same time, a pause signal is generated to temporarily suspend the operation of the selective retransmission mechanism to avoid non-critical data occupying limited link resources. For example, if the preset lower limit of the "low bandwidth range" is 1.2 Mbps and the upper limit of the bit error rate is 0.005, and the bandwidths collected in three consecutive acquisitions are 1.1 Mbps, 1.0 Mbps, and 0.9 Mbps, and the bit error rates are 0.006, 0.007, and 0.008, respectively, all exceeding the threshold, the link adaptive adjustment mechanism determines that the link quality has deteriorated, activates the priority transmission mechanism to transmit the key image frames, and temporarily suspends the selective retransmission mechanism to stop transmitting non-critical image frames.

[0135] Finally, via S1063, the link adaptive adjustment mechanism continuously monitors the link status data. If, for four consecutive acquisition cycles, the bandwidth in the link status data remains above the lower limit of the "medium bandwidth range" and the bit error rate remains below the preset lower limit, it is determined that the link quality has improved. The link adaptive adjustment mechanism generates a synchronization signal, allowing the priority transmission mechanism to continue operating to ensure the transmission of the newly added critical image frames. Simultaneously, the selective retransmission mechanism resumes operation, transmitting non-critical image frames according to the retransmission data queue. This achieves synchronous operation of the two mechanisms, ensuring complete transmission of airborne image data. In the scenario described above, the lower limit of the "medium bandwidth range" is preset to 1.8 Mbps and the lower limit of the bit error rate is 0.002. When the bandwidths collected in four consecutive measurements are 1.9 Mbps, 2.0 Mbps, 2.1 Mbps, and 2.2 Mbps, and the bit error rates are 0.0018, 0.0015, 0.0012, and 0.001, respectively, which meet the improvement standards, the link adaptive adjustment mechanism is triggered to run synchronously. This allows the priority transmission mechanism to transmit the newly generated key image frames, while the selective retransmission mechanism retransmits the incomplete non-key image frames according to the queue, thus achieving complete data transmission.

[0136] In practical applications, an aircraft carrying imaging equipment performs aerial observation missions. First, a collaborative working framework is established to integrate the operating rules of the priority transmission mechanism for critical image frames and the selective retransmission mechanism for non-critical image frames, clarifying that the two mechanisms cannot simultaneously occupy all link resources. The framework receives bandwidth and bit error rate data from the link monitoring module. When the current bandwidth is detected to be 9Mbps, the priority transmission mechanism is scheduled to run for 100 milliseconds, and then the selective retransmission mechanism is switched to run for 100 milliseconds, with the timing being adjusted alternately. When the bandwidth drops to 6Mbps, only the priority transmission mechanism is scheduled to run. The preset standard for link quality degradation is a bandwidth below 5Mbps and a bit error rate above 0.003. When the link status data shows a bandwidth of 4.5 Mbps and a bit error rate of 0.004, reaching the degradation standard, the link adaptive adjustment mechanism control framework prioritizes the priority transmission mechanism and simultaneously suspends the selective retransmission mechanism, allowing the critical image frame to occupy all link resources. Finally, the preset standard for link quality improvement is a bandwidth higher than 8 Mbps and a bit error rate lower than 0.002. When the link status data shows a bandwidth of 9.2 Mbps and a bit error rate of 0.0015, reaching the improvement standard, the link adaptive adjustment mechanism control framework simultaneously runs two mechanisms: first transmitting one critical image frame, then transmitting one non-critical image frame that needs to be retransmitted, alternating between the two to ensure the integrity of the airborne image data.

[0137] The overall solution of S106 described above achieves efficient cooperation between the two transmission mechanisms by establishing a collaborative working framework and a link adaptive adjustment mechanism. It can flexibly adjust the mechanism operation strategy according to the link status, avoid resource waste when the link quality deteriorates, ensure the priority transmission of the key image frames, and operate the two mechanisms simultaneously when the link quality improves to ensure the integrity of airborne image data. It adapts to unstable link conditions as a whole, ensuring that data transmission prioritizes key data without omitting non-key data, thereby improving the adaptability and reliability of airborne image data transmission.

[0138] The following is a complete example for steps 101 to 106, such as Figure 3 As shown, firstly, an aircraft carrying imaging equipment and a power supply system performs an aerial reconnaissance mission, requiring the transmission of reconnaissance images back to the ground station. During the mission, the aircraft first acquires the current link bandwidth of 1.8 Mbps and the bit error rate of 0.002 through the downlink feedback channel. To determine the link transmission capacity, the parameters are first processed using the Min-Max normalization algorithm (preset maximum bandwidth 2.5 Mbps, maximum bit error rate 0.01), calculating the normalized bandwidth = 1.8 ÷ 2.5 = 0.72 and the normalized bit error rate = 1 - (0.002 ÷ 0.01) = 0.8. Then, according to the weighting rule of "bandwidth weight 0.6, bit error rate weight 0.4", the link transmission capacity evaluation coefficient is calculated as 0.72 × 0.6 + 0.8 × 0.4 = 0.752. Based on this, the image compression ratio is determined to be 1:3, and the imaging equipment compresses the original image with a resolution of 2048 × 1536 according to this ratio.

[0139] After compression, the aircraft classified the 20 frames of 1920×1080 resolution images: Using a target detection algorithm, a frame (Frame_003) was identified as containing a reconnaissance target with a target pixel count of 1920×1080×6% = 116640 (out of a total of 2073600 pixels). The target percentage was 6% ≥ 5%, thus classifying it as a critical image frame. Of the remaining 19 frames, 16 had a target percentage < 5%, and 3 failed to transmit; all were marked as non-critical image frames. Subsequently, the aircraft wrote 4 critical frames to ferroelectric memory and 16 non-critical frames to solid-state memory, verifying the integrity of each frame after writing using a CRC check.

[0140] During transmission, the power supply system malfunctioned: the voltage sampling circuit sampled the voltage every 50 milliseconds, and after 6 consecutive samplings, the voltage was 24V (the preset normal range is 27V-33V), lasting for 50 milliseconds × 6 = 300 milliseconds, triggering the data protection program; the program interrupted the transmission, saved the progress "transmitted to Frame_008", and then, according to the physical characteristic of ferroelectric memory with 4KB per page, reassembled the 4 key frames (total capacity 10KB) into 3 data blocks (2 4KB and 1 2KB) and wrote them into the non-volatile storage area.

[0141] Subsequently, the link recovered, and the bandwidth was monitored to reach 4Mbps for three consecutive cycles (preset transmission threshold 3Mbps). The aircraft initiated priority transmission of key frames, encapsulating four key frames according to the protocol (including a 16-byte frame header and a 4-byte checksum) and sending them to the ground station first. After the key frame transmission was completed, the link bandwidth continued to rise to 2.1Mbps and the bit error rate dropped to 0.0015 (improving for five consecutive cycles). Non-key frame retransmission was then initiated: eight frames were selected from the 16 non-key frames to be retransmitted, and scored according to "target relevance 60% + texture complexity 40%". For example, Frame_002 had a target relevance score of 80 and a texture complexity score of 70, with a comprehensive score of 80 × 0.6 + 70 × 0.4 = 76 points; Frame_004 had a target relevance score of 75 and a texture complexity score of 65, with a comprehensive score of 75 × 0.6 + 65 × 0.4 = 71 points. The retransmission queue was formed by sorting the frames according to their scores, and the frames were sent one by one with confirmation of receipt.

[0142] When the link fluctuates again, the bandwidth drops to 1.0 Mbps and the bit error rate rises to 0.007, the aircraft suspends the retransmission of non-critical frames and prioritizes the transmission of newly added critical frames. Once the link recovers to a bandwidth of 2.2 Mbps and a bit error rate of 0.001, the newly added critical frames and the remaining non-critical frames are transmitted synchronously. Finally, all critical frames and non-critical frames are completely transmitted back to the ground station.

[0143] Figure 4 This is a schematic diagram illustrating a specific implementation of an airborne high-speed image storage control system based on FPGA, as provided in this application. Figure 4 The system may include:

[0144] The adjustment module 41 is used to monitor the downlink status between the aircraft and the ground station to obtain the link transmission capability parameters, and dynamically adjust the image data to be transmitted through the FPGA according to the link transmission capability parameters.

[0145] The scheduling module 42 is used to divide the adjusted image data to be transmitted into key image frames and non-key image frames, and schedule the key image frames to ferroelectric memory and the non-key image frames to solid-state memory.

[0146] The conversion module 43 is used to continuously monitor the power supply status during transmission. When an abnormal power supply is detected, the data protection program is activated to convert the key image frames stored in the ferroelectric memory into a persistent storage format.

[0147] Transmission module 44 is used to schedule the key image frames that have undergone data protection processing for priority transmission when the transmission bandwidth reaches a preset threshold, based on the continuously monitored link transmission capability parameters.

[0148] The startup module 45 is used to initiate a selective retransmission procedure for non-critical image frames based on the non-critical image frames stored in the solid-state memory when the link transmission capability parameters are continuously improved.

[0149] The second transmission module 46 is used to cooperate with the priority transmission of the key image frames and the selective retransmission of the non-key image frames, combined with link adaptive adjustment, to achieve complete transmission of airborne image data under unstable link conditions.

[0150] The FPGA-based airborne high-speed image storage control system of this application embodiment is used to implement the aforementioned FPGA-based airborne high-speed image storage control method. Therefore, the specific implementation of the FPGA-based airborne high-speed image storage control system can be found in the embodiment section of the FPGA-based airborne high-speed image storage control method above. The specific implementation can be referred to the description of the corresponding embodiments, and will not be repeated here.

[0151] This application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of any of the above-described FPGA-based airborne high-speed image storage control methods.

[0152] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of any of the above-described FPGA-based airborne high-speed image storage control methods.

[0153] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0154] Embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the embodiments of the FPGA-based airborne high-speed image storage control method described above.

[0155] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0156] The above provides a detailed description of the FPGA-based airborne high-speed image storage control method and system provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. An airborne high-speed image storage control method based on FPGA, characterized in that, include: The downlink status between the aircraft and the ground station is monitored to obtain the link transmission capability parameters, and the image data to be transmitted is dynamically adjusted through the FPGA according to the link transmission capability parameters. The adjusted image data to be transmitted is divided into key image frames and non-key image frames, and the key image frames are scheduled to ferroelectric memory, while the non-key image frames are scheduled to solid-state memory. During transmission, the power supply status is continuously monitored. When an abnormal power supply is detected, a data protection program is activated to convert the key image frames stored in the ferroelectric memory into a persistent storage format. Based on the continuously monitored link transmission capacity parameters, when the transmission bandwidth reaches a preset threshold, key image frames that have undergone data protection processing are scheduled for priority transmission. Based on the non-critical image frames stored in the solid-state memory, when the link transmission capability parameters are continuously improved, a selective retransmission procedure for the non-critical image frames is initiated. By prioritizing the transmission of critical image frames and selectively retransmitting non-critical image frames, combined with adaptive link adjustment, complete transmission of airborne image data can be achieved under unstable link conditions.

2. The method according to claim 1, characterized in that, Based on the non-critical image frames stored in the solid-state memory, when a continuous improvement in link transmission capability parameters is detected, the selective retransmission procedure for non-critical image frames is initiated, including: Based on the non-critical image frames stored in the solid-state memory, the link monitoring module continuously acquires the changing trend of the link transmission capability parameters. Based on the aforementioned trend, analyze the degree of improvement in the link transmission status. When the analysis results show that the link transmission capability parameters have maintained stable improvement for multiple consecutive monitoring cycles, initiate the selective retransmission procedure. According to the selective retransmission procedure, non-critical image frames that need to be retransmitted are selected from the solid-state memory; According to the preset importance assessment criteria, the non-critical image frames are organized into a retransmission data queue. Non-critical image frames are transmitted through the downlink data transmission interface in the order of the retransmission data queue.

3. The method according to claim 2, characterized in that, Based on the aforementioned trend, analyze the degree of improvement in the link transmission status, including: Obtain a sequence of link transmission capability parameter values ​​collected over multiple monitoring periods, and calculate the parameter change values ​​between adjacent monitoring periods in the link transmission capability parameter value sequence to generate a parameter change value sequence. The positive change count is performed on the parameter change value sequence, and the number of monitoring periods in which the parameter change value is consecutively positive is counted. When the number of consecutive monitoring cycles in which the parameter change value is positive reaches a first predetermined number, the average value of the parameter change value within the consecutive monitoring cycles is calculated. The average value is compared with a second predetermined threshold. If the average value is greater than or equal to the second predetermined threshold, it is determined that the degree of improvement in the link transmission status meets the stable improvement condition.

4. The method according to claim 1, characterized in that, By prioritizing the transmission of critical image frames and selectively retransmitting non-critical image frames, combined with adaptive link adjustment, complete airborne image data transmission under unstable link conditions is achieved, including: A collaborative working framework is established for the priority transmission mechanism of the key image frames and the selective retransmission mechanism of the non-key image frames. Under the collaborative working framework, the runtime order of the priority transmission mechanism and the selective retransmission mechanism is dynamically coordinated based on the link status data. By utilizing a link adaptive adjustment mechanism, the priority transmission mechanism is activated first when the link quality deteriorates, while the selective retransmission mechanism is temporarily suspended. The priority transmission mechanism and the selective retransmission mechanism are run synchronously when the link quality improves, so as to achieve complete transmission of airborne image data under unstable link conditions.

5. The method according to claim 1, characterized in that, During transmission, the power supply status is continuously monitored. When a power supply abnormality is detected, a data protection program is activated to convert the key image frames stored in the ferroelectric memory into a persistent storage format, including: The output voltage parameters of the power supply system are continuously collected through a voltage sampling circuit, and the output current parameters are collected through a current sampling circuit. The collected output voltage parameters and output current parameters are compared item by item with the preset voltage threshold range and current threshold range; When the output voltage parameter or the output current parameter continuously exceeds the corresponding threshold range and reaches a predetermined duration, the data protection program is triggered. After the data protection program is started, the current image data transmission process is interrupted and the transmission status information is saved. The interface controller of the ferroelectric memory is controlled by the FPGA to reorganize the data storage structure of the stored key image frames according to the physical characteristics of the ferroelectric memory. The reorganized key image frames are written to the non-volatile storage area, completing the conversion to persistent storage format.

6. The method according to claim 1, characterized in that, Based on the continuously monitored link transmission capacity parameters, when the transmission bandwidth reaches a preset threshold, priority transmission is scheduled for key image frames that have undergone data protection processing, including: Based on the continuously monitored link transmission capability parameters, bandwidth measurement data is obtained through the downlink status feedback channel; The bandwidth measurement data is periodically compared and analyzed with a preset transmission threshold. When the analysis results show that the bandwidth measurement data exceeds the transmission threshold for multiple consecutive sampling periods, a priority transmission scheduling mechanism is activated. According to the priority transmission scheduling mechanism, the key image frame, which has undergone data protection processing, is read from the non-volatile storage area of ​​the ferroelectric memory; According to the transmission protocol specification, the key image frame is encapsulated into a transmission data unit of a specific format, and the transmission data unit is preferentially scheduled for transmission through the downlink data transmission interface.

7. The method according to claim 1, characterized in that, The monitoring of the downlink status between the aircraft and the ground station to obtain link transmission capability parameters, and the dynamic adjustment of the image data to be transmitted via FPGA based on the link transmission capability parameters, includes: Link status data is obtained through the feedback channel of the downlink, and bandwidth and bit error rate parameters are extracted from the link status data. The link transmission capacity evaluation coefficient is calculated based on the bandwidth index parameter and the bit error rate index parameter, and the compression ratio parameter of the image data is determined based on the link transmission capacity evaluation coefficient. The compression parameters of the image encoder are configured via the FPGA, and the image data to be transmitted is encoded according to the compression ratio parameters.

8. An airborne high-speed image storage control system based on FPGA, characterized in that, include: The adjustment module is used to monitor the downlink status between the aircraft and the ground station to obtain the link transmission capability parameters, and dynamically adjust the image data to be transmitted through the FPGA according to the link transmission capability parameters. The scheduling module is used to divide the adjusted image data to be transmitted into key image frames and non-key image frames, and schedule the key image frames to ferroelectric memory and the non-key image frames to solid-state memory. The conversion module is used to continuously monitor the power supply status during transmission. When an abnormal power supply is detected, the data protection program is activated to convert the key image frames stored in the ferroelectric memory into a persistent storage format. The transmission module is used to schedule the transmission of key image frames that have undergone data protection processing for priority transmission when the transmission bandwidth reaches a preset threshold, based on the continuously monitored link transmission capacity parameters. The startup module is used to initiate a selective retransmission procedure for non-critical image frames based on the non-critical image frames stored in the solid-state memory when the link transmission capability parameters are continuously improved. The second transmission module is used to cooperate with the priority transmission of the key image frames and the selective retransmission of the non-key image frames, combined with link adaptive adjustment, to achieve complete transmission of airborne image data under unstable link conditions.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the FPGA-based airborne high-speed image storage control method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the implementation of the FPGA-based airborne high-speed image storage control method as described in any one of claims 1 to 7.

Images