EVTOL-oriented multipath transmission optimization and dynamic path selection method

Abstract

The invention relates to an eVTOL (electric vertical take-off and landing aircraft)-oriented multipath transmission optimization and dynamic path selection method, and aims to solve the problems of bandwidth, delay and packet loss of an eVTOL aircraft in high-throughput video data transmission. According to the method, key performance indexes such as bandwidth, delay and packet loss rate of a path are monitored in real time, and path selection is optimized in combination with flight state data, so that low-delay and high-bandwidth transmission of a panoramic video stream is ensured. An improved multi-path load balancing strategy is adopted, the flow is distributed according to the path score and the bandwidth margin, network congestion is avoided, and the data transmission stability is improved. By enhancing an MPQUIC protocol and optimizing multi-path cooperative transmission, a low-delay path is preferentially selected to transmit high-priority data. A redundant transmission mechanism ensures that data can still be reliably transmitted through other paths even if a part of paths lose packets. The method provides technical support for the construction of an immersive eVTOL driving experience platform, meets the requirements of low time delay, high bandwidth and high sense of reality, and has a wide application prospect.

Description

Technical Field

[0001] The present invention relates to communication network optimization technology, and in particular to a multi-path transmission optimization and dynamic path selection method for eVTOL (electric vertical take-off and landing) vehicles. The method aims to address bandwidth, latency, and packet loss issues during the transmission of high-throughput image data (such as high-definition video streaming and image transmission), improve the stability, real-time performance, and high throughput of data transmission, and is particularly suitable for the massive data transmission needs in dynamic flight environments. Background Art

[0002] With the development of eVTOL (electric vertical take-off and landing) technology, the communication needs of aircraft are gradually increasing, especially in the fields of low-altitude tourism and intelligent transportation. The rise of the low-altitude tourism market has put forward new requirements for eVTOL applications, especially in terms of immersive driving experience. eVTOL aircraft are typically equipped with multiple cameras and sensor systems for real-time video monitoring, navigation assistance, environmental perception, and augmented reality (AR) / virtual reality (VR) experiences. These multi-source data and video streams generate a large amount of high-throughput data, requiring real-time, low-latency, stable and efficient data transmission protocols to ensure a high-quality immersive experience.

[0003] Existing communication path selection methods typically rely on static network configurations or simplified path selection algorithms. These methods are unable to respond to dynamic changes during flight, particularly bandwidth fluctuations and signal attenuation, leading to unstable communication quality. In low-altitude tourism applications, aircraft must support real-time panoramic data capture and encoding, efficient multi-source data transmission, and augmented reality (AR) / virtual reality (VR) enhanced fusion technologies. Existing technologies fail to effectively meet these high-bandwidth, low-latency requirements.

[0004] Although existing multi-path transmission protocols support multi-path communication, there is still much room for optimization in terms of path selection, load balancing, and efficient data transmission. They fail to fully support the massive data transmission in eVTOL systems, especially real-time video streaming and panoramic image transmission in dynamic flight environments. In order to provide a higher sense of immersion, existing systems fail to fully utilize the bandwidth resources of available paths, resulting in bandwidth waste, network congestion, and affecting the interactive effects and experience smoothness in AR / VR scenarios. Therefore, there is an urgent need for a new technical solution that can dynamically optimize path selection according to flight status, support panoramic real-time acquisition and adaptive massive data transmission protocols, and achieve seamless integration of virtual and real scenes, thereby enhancing the authenticity and safety of the immersive eVTOL driving experience, while optimizing cost control capabilities, laying the foundation for large-scale implementation, and accelerating the transition of the low-altitude economy from the experimental stage to commercial application. Summary of the Invention

[0005] The purpose of the present invention is to propose a multi-path transmission optimization and dynamic path selection method for eVTOL, specifically targeting the high-throughput transmission requirements of image data (including high-definition images, video streams, etc.), and to solve the bandwidth, delay and packet loss problems of image data transmission in the existing technology, thereby improving the reliability, real-time performance and high throughput of data transmission, especially in dynamic flight environments, to ensure efficient and stable image data.

[0006] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] The present invention aims to provide a multi-path transmission optimization and dynamic path selection method for eVTOL, comprising the following steps:

[0008] S1, dynamic evaluation of path quality:

[0009] Path quality assessment considers the impact of bandwidth fluctuations, latency, and packet loss on communication quality during massive data transmission. High-throughput data, such as image and video streams, places particular demands on paths. Therefore, bandwidth availability and stability are a top priority when selecting a path. This ensures stable data flow over long periods of time and avoids bandwidth shortages or packet loss caused by network fluctuations.

[0010] Specifically, dynamic path quality assessment monitors key performance indicators (KPIs) such as bandwidth, latency, packet loss rate, signal-to-interference ratio (SIR), and Doppler shift across each transmission path in real time. Combined with a dynamic weighting mechanism and path stability assessment, it calculates a comprehensive score for each path and selects the best-performing transmission path to support subsequent steps. For high-throughput transmission of image data, in particular, it is crucial to ensure that the selected path provides sufficient bandwidth to support the transmission of large amounts of data, such as high-definition video streams.

[0011] S2, flight state fusion path optimization:

[0012] Utilize the real-time motion data (such as flight speed, acceleration, heading, etc.) collected by the eVTOL terminal to optimize the path selection. The steps include:

[0013] Trajectory prediction: Extended Kalman filter (EKF) is used to predict the future trajectory of the aircraft.

[0014] Carrier correction: Compensates for the aircraft's motion state to reduce errors in path selection.

[0015] Correct path score based on relative speed: Adjust the path score by calculating the relative speed between the aircraft and the base station.

[0016] When the flight status changes significantly, the system optimizes the path based on real-time flight data to ensure stable transmission of image data and video streams. Especially in high-speed flight or dynamic environments, the path selection can automatically adapt to massive data transmission scenarios with large bandwidth requirements.

[0017] S3, multipath load balancing:

[0018] An improved weighted least connections (WLC) strategy is used to distribute traffic based on the path's comprehensive score and real-time bandwidth headroom. This ensures that high-scoring paths carry more traffic and avoids path overload. Specifically, high-throughput data flows (such as video and image data) are prioritized over paths with greater bandwidth to ensure that these data are not affected by latency and packet loss. This strategy ensures load balancing across multiple paths and reduces network congestion.

[0019] S4, MPQUIC protocol enhancements:

[0020] To improve the efficiency of multipath transmission, this invention enhances the MPQUIC protocol to support dynamic adjustment of data flow across each path. Especially for high-throughput data (such as video streaming and image transmission), MPQUIC protocol optimization prioritizes paths with higher bandwidth and lower latency, ensuring efficient data transmission. Furthermore, the MPQUIC protocol extension field supports data priority setting. For important high-throughput data such as video streaming, it is given a higher transmission priority to ensure stable and efficient transmission of this data over the optimal path.

[0021] Specific operations include:

[0022] Dynamically adjust the data flow of each path based on the path quality score;

[0023] Expanding the MPQUIC protocol fields and adding a priority field to the protocol header ensures that high-priority data is transmitted over a stable, low-latency path;

[0024] Using the BBR-MP algorithm, the sending window of each path is adjusted according to bandwidth and RTT information to maximize bandwidth utilization.

[0025] S5, dynamic adjustment of redundant transmission:

[0026] To ensure data reliability, this invention employs keyframe replication and redundant path selection strategies. Even if packet loss occurs on certain paths, important data such as video streams can still be reliably transmitted. This redundant transmission strategy ensures the integrity of image and video data through multi-path redundant transmission. Especially in high-throughput transmission scenarios for high-definition video data, redundant paths must have sufficient bandwidth to support the simultaneous transmission of large amounts of data.

[0027] Specific methods include:

[0028] Key frame replication: Redundantly transmit key frames in the video stream through different paths to ensure that the receiving end receives the frame from other paths;

[0029] Redundant path selection: A greedy backoff algorithm is used to prioritize paths with higher stability scores for redundant transmission. If the redundant path load is too high, a suboptimal path is used to avoid network overload.

[0030] Redundant transmission can also ensure that redundant transmission does not cause excessive network load by optimizing redundant path selection and bandwidth management, while continuously ensuring real-time transmission of image data.

[0031] The multipath strategies described above (path quality assessment, load balancing, and redundant transmission) work together to form a complete multipath transmission optimization framework. Path quality assessment provides foundational data support, load balancing ensures the proper distribution of data traffic, and redundant transmission enhances system reliability. All these strategies work together to optimize the stability, reliability, and efficiency of data transmission.

[0032] The beneficial effects of the present invention are as follows:

[0033] (1) The present invention can effectively solve the problems of transmission interruption and quality fluctuation caused by path instability in the prior art by dynamically evaluating path quality and optimizing path selection, thereby improving the communication stability and reliability of the eVTOL aircraft during flight, especially in high-speed flight and dynamic environments, and ensuring the efficient transmission of images and video streams.

[0034] (2) Optimizing path selection based on the real-time status of the aircraft and the dynamic characteristics of the path, combined with a multi-path load balancing algorithm, not only improves bandwidth utilization, but also reduces network congestion and resource waste, ensuring stable transmission of high-throughput data (such as high-definition video streams).

[0035] (3) By extending the priority and path stability fields of the MPQUIC protocol, high-priority data (such as control signaling, image data, etc.) is preferentially transmitted through low-latency, high-bandwidth paths, thereby improving data transmission efficiency and optimizing transmission performance, especially in large data transmission scenarios.

[0036] (4) Redundant transmission and differentiated coding technologies are adopted, especially for high-throughput data such as images and video streams. Even in the case of packet loss on some paths, stable transmission of key data can be ensured, thereby improving the reliability of the system under high load. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a step flow chart;

[0038] Figure 2 This is a flow chart of the dynamic weight allocation mechanism;

[0039] Figure 3 This is the relationship diagram between path stability and comprehensive score;

[0040] Figure 4 It is a multi-path load balancing strategy diagram;

[0041] Figure 5 Enhanced structure diagram for MPQUIC protocol;

[0042] Figure 6 Dynamically adjust flow graphs for redundant transmissions. DETAILED DESCRIPTION

[0043] like Figure 1 As shown, the present invention discloses a multi-path transmission optimization and dynamic path selection method for eVTOL, including the following steps:

[0044] S1: Dynamic evaluation of path quality

[0045] Dynamic path quality assessment monitors key performance indicators (KPIs) such as bandwidth, latency, packet loss rate, signal-to-interference ratio (SIR), and Doppler shift across each transmission path in real time. It then combines a dynamic weight allocation mechanism with path performance evaluation to select the optimal transmission path. For high-throughput data, such as image and video streams, bandwidth availability and stability are particularly important when selecting a path to support stable transmission of high data volumes.

[0046] The specific implementation is as follows:

[0047] First, by calculating the volatility of delay and packet loss rate (the variance σ 2 ) quantifies the stability of a path, thereby predicting its impact on data transmission, especially the transmission of video streams. A higher stability score indicates a more stable path, making it suitable for long-term data transmission. The calculation formula is as follows:

[0048]

[0049] α and β represent dynamic adjustment coefficients, with different weight thresholds set based on the flight phase (e.g., takeoff and landing, cruise, etc.). A higher stability score indicates a more stable path, making it suitable for long-term data transmission. By quantifying stability, we can effectively avoid fluctuations in network transmission quality caused by path instability.

[0050] Then, the overall performance of each path is calculated through the comprehensive scoring formula. The real-time weight of each indicator (w i) is dynamically adjusted based on the sliding window algorithm. The comprehensive scoring formula is as follows:

[0051] Score=∑(w i Normalized (Metric i ))

[0052] Among them, Metric i It includes key performance indicators such as bandwidth, SIR, Doppler shift, and stability. By combining stability scoring with comprehensive scoring, the one-sidedness of a single indicator is avoided, and the requirements of multiple dimensions such as bandwidth, delay, and stability are balanced. In order to adapt to different flight phases and network environments, the weight of each indicator is dynamically adjusted. For example Figure 2 As shown in Figure 2, in complex scenarios, priority is given to increasing the weight of bandwidth to support the transmission of high-throughput image data. The calculation relationship between the stability score and the comprehensive score is as follows: Figure 3 shown.

[0053] S2: Flight state fusion path optimization

[0054] Path selection is optimized by collecting real-time motion data of the aircraft (such as flight speed, acceleration, heading, etc.) from the eVTOL end.

[0055] First, trajectory prediction and frequency shift compensation are performed. The extended Kalman filter (EKF) is used to predict the trajectory within the next 5 seconds and calculate the relative speed with the base station:

[0056] v rel =v eVTOL cos(θ)-v BS

[0057] Among them, v rel is the relative speed between the aircraft and the base station, v eVTOL is the speed of the aircraft, v BS is the velocity of the base station (usually static, with a velocity of zero, but there may also be dynamic base stations), and θ is the angle between the aircraft and the base station.

[0058] Based on the predicted relative velocity, the Doppler shift is further estimated:

[0059]

[0060] Where f0 is the signal's carrier frequency and c is the speed of light. The Doppler shift calculation influences path selection, ensuring that paths are chosen that compensate for the frequency offset, thereby minimizing signal attenuation and bit errors.

[0061] Then considering the relative velocity v of the aircraft relTo account for the impact of signal attenuation, the comprehensive score of each path is adjusted to modify the path weight. The formula is as follows:

[0062]

[0063] Among them, γ is a speed sensitivity coefficient, which controls the influence of relative speed on the path score. This coefficient is adjusted according to the specific needs of the aircraft (such as flight speed and communication requirements).

[0064] The system adjusts its routing in real time based on flight status changes, ensuring stable transmission of image data and video streams, especially during high-speed flight or in dynamic environments, and adapting to high-data-volume transmission requirements. Specifically, routing prioritizes base station paths aligned with the flight direction to minimize signal attenuation.

[0065] S3: Multipath Load Balancing

[0066] According to the comprehensive score of the path and the real-time bandwidth margin (B available,i ), uses an improved weighted least connection (WLC) strategy for traffic distribution, ensuring that high-scoring paths carry more traffic and avoid network overload. Especially in the transmission of high-throughput data streams (such as video streams and image data), paths with larger bandwidth are preferred for data transmission to ensure that image data is not affected by delays and packet loss. For example, Figure 4 The weight calculation formula is as follows:

[0067]

[0068] S4: MPQUIC protocol enhancements

[0069] By enhancing the MPQUIC protocol, the data flow of each path is dynamically adjusted to ensure that paths with higher bandwidth and lower latency are prioritized for high-throughput data transmission (such as video streaming and image transmission). Specifically, the BBR-MP (Bottleneck Bandwidth and RTT for Multi-path) algorithm will be used to adjust the sending window of each path to maximize bandwidth utilization and ensure that data can be transmitted stably and efficiently across multiple paths.

[0070] To avoid congestion caused by bursty traffic, the enhanced MPQUIC protocol uses packet pacing optimization on the basis of multipath load balancing (S13) to evenly distribute the sending time of data packets, ensuring that data packets are transmitted smoothly, reducing congestion caused by bursty traffic, and ensuring that high-priority data streams can be transmitted stably.

[0071] like Figure 5As shown, the MPQUIC protocol fields are expanded, and a 3-bit priority field is added to the MPQUIC header. Based on the importance of the data, the transmission content is divided into a core area (control signaling), a secondary area (video stream), and a background area (sensor data), each of which is assigned different priorities. High-priority data (such as control signaling) is preferentially transmitted over low-latency, high-bandwidth paths. The ground terminal can prioritize unpacking the core area data based on the field value. A 4-byte path stability flag is added to the ACK frame to record the path stability score in real time and feed this information back to the ground terminal. The ground terminal dynamically adjusts the path priority and weight based on the stability flag. For example, in S12: Flight State Fusion Path Optimization, when it is predicted that a path may be unstable in the future flight phase, the MPQUIC protocol will perform path switching in advance and, based on real-time path stability feedback, decide whether to start a redundant transmission path or switch to a backup path to ensure continuous data transmission and low latency.

[0072] S5: Dynamic adjustment of redundant transmission

[0073] To ensure data reliability, this invention employs keyframe replication and redundant path selection strategies, ensuring that even if packet loss occurs on certain paths, important data such as video streams can still be reliably transmitted. This redundant transmission strategy ensures the integrity of image and video data through multi-path redundant transmission. Especially in high-throughput transmission scenarios for high-definition video data, redundant paths require sufficient bandwidth to support the simultaneous transmission of large amounts of data.

[0074] The key frame replication strategy redundantly transmits key frames in a video stream across different paths. Even if data is lost on one path, the receiver can still receive the frame from other paths, thus ensuring the integrity of the video stream.

[0075] When selecting redundant paths, the present invention adopts a greedy-backoff algorithm, such as Figure 6 As shown in Figure 2, the algorithm prioritizes paths with higher stability scores for redundant transmission, ensuring the stability and low packet loss rate of redundant paths. If the path load exceeds 70%, the suboptimal path is enabled and marked as "temporarily redundant" to avoid network overload.

[0076] To avoid bandwidth waste caused by redundant transmission, the number of redundant copies is reduced when the packet loss rate for three consecutive packet loss windows is less than 1%, saving bandwidth resources. The size of the redundant copies is bound to the available bandwidth of the path, ensuring that redundant transmission does not exceed the path capacity, thereby avoiding congestion.

[0077] Key frames transmitted over redundant paths utilize differentiated encoding technologies, such as forward error correction (FEC), to further enhance robustness against packet loss. FEC allows the receiver to recover lost data using redundant data even if it is lost during transmission, improving data transmission reliability.

Claims

1. A multi-path transmission optimization and dynamic path selection method for eVTOL, characterized in that: The steps include: S1: Dynamic evaluation of path quality, especially during high-throughput image data transmission. This evaluation assesses path stability and overall performance based on key performance indicators (KPIs) such as bandwidth, latency, packet loss rate, signal-to-interference ratio, and Doppler shift. This ensures that the path can meet high bandwidth requirements and avoids bandwidth shortages or transmission quality fluctuations. S2: Flight state fusion path optimization, which optimizes path selection based on real-time flight data to ensure stable transmission of massive data such as images and video streams under high-speed flight or dynamic environments; S3: Multi-path load balancing dynamically allocates data traffic to ensure that high-throughput data is preferentially transmitted through high-bandwidth paths; S4: The MPQUIC protocol optimizes multi-path collaborative transmission. Combined with the path comprehensive scoring weight, it dynamically adjusts the data flow of each path. This prioritizes high-priority image data and video streams, ensuring preferential transmission of low-latency, high-bandwidth paths. The BBR-MP algorithm and pacing technology are used to optimize congestion control and flow control, and the MPQUIC protocol fields are expanded. S5: Dynamic adjustment of redundant transmission, through key frame replication and path redundancy selection strategy, ensures that image data can still be transmitted reliably even if packet loss occurs on certain paths, especially in scenarios with massive data transmission, ensuring efficient and stable data transmission.

2. The multi-path transmission optimization and dynamic path selection method for eVTOL according to claim 1, characterized in that: In step S1, path stability is evaluated by the volatility of delay and packet loss rate, where the volatility calculation formula is: Among them, α and β represent dynamic adjustment coefficients, and different weight thresholds are set according to the flight phase. 2 delay , σ 2 loss They represent the variance of delay and packet loss rate respectively. The higher the stability score, the more stable the path is, and it is suitable for long-term data transmission.

3. The multi-path transmission optimization and dynamic path selection method for eVTOL according to claim 1, characterized in that: In step S1, the comprehensive performance evaluation integrates multi-dimensional performance indicators such as bandwidth, delay, SIR, Doppler shift, and stability, and generates a total score through dynamic weight allocation to reflect the overall suitability of the path. The calculation formula is: Score=∑(w i ·Normalized(Metric i )) Among them, Metric i Represents a certain performance indicator, w i Indicates its weight, Normalized means normalizing it, and the weight of the indicator is adjusted through dynamic weight allocation to meet the needs of different flight scenarios.

4. The method for multi-path transmission optimization and dynamic path selection for eVTOL according to claim 1, characterized in that: In step S2, the path optimization is performed using the real-time motion data of the aircraft collected by the eVTOL terminal, including: a. Trajectory prediction; b.Carrier correction; c. Correct the path score based on the relative speed of the aircraft.

5. The trajectory prediction according to claim 4, characterized in that The extended Kalman filter is used to predict the trajectory in the next 5 seconds and calculate the relative speed v between the aircraft and the base station. rel , the formula is: in rel =in eVTOL ·cos(θ)-v BS Among them, v eVTOL is the speed of the aircraft, v BS is the speed of the base station, and θ is the angle between the aircraft and the base station.

6. The method of claim 4, wherein: The formula is: Score is a comprehensive score calculated by the dynamic evaluation of path quality in S1. γ is a speed sensitivity coefficient that controls the impact of relative speed on the path score. At high speeds, path selection prioritizes base station paths that are consistent with the flight direction to reduce signal attenuation.

7. The method for multi-path transmission optimization and dynamic path selection for eVTOL according to claim 1, characterized in that: In step S3, load balancing uses an improved weighted minimum connection number strategy to allocate data traffic based on the comprehensive path score and real-time bandwidth margin, ensuring that high-scoring paths carry more traffic and avoid network congestion.

8. The method for multi-path transmission optimization and dynamic path selection for eVTOL according to claim 1, characterized in that: In step S4, the MPQUIC protocol field is extended and a 3-bit priority field is added to its header. The transmission content is divided into core area, secondary area and background area according to the importance of the video data, and different priorities are assigned to each area. Low-latency, high-bandwidth paths are preferentially selected to transmit high-priority data. A 4-byte path stability mark is added to the protocol ACK frame to record the path stability score in real time and feed this information back to the ground end. The ground end dynamically adjusts the priority and weight of the path based on the stability mark.

9. The method for multi-path transmission optimization and dynamic path selection for eVTOL according to claim 1, characterized in that: In step S5, the path redundancy selection strategy adopts a greedy-backoff algorithm, which preferentially selects paths with higher stability scores for redundant transmission, ensuring the stability and low packet loss rate of the redundant paths to avoid network overload.

Images