Dynamic task scheduling method based on non-ground network and related equipment

Abstract

The invention provides a dynamic task scheduling method based on a non-ground network and related equipment. The method comprises the following steps: acquiring vehicle-mounted sensing data; judging whether the local processing total delay of the vehicle-mounted sensing data meets a constraint condition or not according to delay real-time constraint; if yes, determining to process the vehicle-mounted sensing data locally; if not, acquiring queuing waiting time according to the available queue buffer capacity of the service satellite terminal; judging whether the total processing delay of the service satellite terminal of the vehicle-mounted sensing data meets a constraint condition or not according to the queuing waiting time; if yes, generating a calculation service request according to the vehicle-mounted sensing data and sending the calculation service request to the service satellite terminal; and if not, discarding the current vehicle-mounted sensing data. By means of the non-ground network, internet access is provided from the air, on-demand connection of the non-ground network can be achieved under the condition that ground infrastructures cannot meet requirements, and effective supplement is formed for the ground network.

Description

Technical Field

[0001] The present application relates to the fields of communication technology and vehicle edge computing, and in particular to a dynamic task scheduling method based on a non-terrestrial network and related equipment. Background Art

[0002] In the field of intelligent transportation, with the development of autonomous driving technology, in-vehicle edge terminals need to process large amounts of sensor data to achieve safe driving. However, due to energy and capacity limitations, in-vehicle edge terminals often need to offload some tasks to external cloud computing environments for processing.

[0003] Currently, communication networks for external cloud computing environments mainly rely on the construction of ground communication infrastructure. However, the deployment of these infrastructures in remote areas (such as mountainous areas) faces the impact of harsh weather and terrain conditions, making the construction and operation of the infrastructure difficult and inconvenient for promotion.

[0004] The integration of terrestrial and non-terrestrial (T-NT) communication networks is a new trend in the development of future 6G communication technology. This strategy enhances remote communication and processing capabilities. Incorporating air and space platforms, such as high-altitude platforms or satellite platforms, provides the infrastructure for connecting nodes in remote areas. In non-terrestrial networks, low-Earth orbit satellites form a dense network of interconnected nodes around the Earth. Low-Earth orbit satellites in non-terrestrial networks can serve as space computing nodes, aggregating, storing, and processing data from ground terminals. However, how to effectively schedule tasks to rationally utilize the computing power of non-terrestrial networks remains a current challenge.

[0005] Therefore, there is an urgent need to provide a dynamic task scheduling method and related equipment based on non-terrestrial networks. Summary of the Invention

[0006] In order to overcome the above-mentioned shortcomings, the present application provides a dynamic task scheduling method and related equipment based on a non-terrestrial network.

[0007] The purpose of this application is achieved by the following technical solutions:

[0008] In a first aspect, the present application provides a dynamic task scheduling method based on a non-terrestrial network, the method being applied to an on-board edge terminal of a ground vehicle, the method comprising:

[0009] Acquire vehicle-mounted sensor data at a fixed inter-frame rate using a vehicle-mounted sensor, and determine whether a total local processing delay of the vehicle-mounted sensor data satisfies a constraint condition;

[0010] If the total local processing delay satisfies the constraint condition, determining to locally process the onboard sensor data; otherwise, obtaining a queue waiting time of the serving satellite terminal, and determining whether the total remote processing delay of the onboard sensor data satisfies the constraint condition based on the queue waiting time, where the total remote processing delay is the total processing delay of the onboard sensor data by the serving satellite terminal;

[0011] If the total remote processing delay meets the constraint condition, a computing service request is generated according to the vehicle-mounted sensor data and sent to the service satellite terminal, and feedback information returned by the service satellite terminal is received; otherwise, the vehicle-mounted sensor data is discarded.

[0012] This embodiment provides the following advantages: by hierarchically determining whether local and remote processing delays meet constraints, the processing of onboard sensor data is ensured. The onboard edge terminal prioritizes local processing to reduce remote transmission delays. When local processing is insufficient, it promptly switches to remote processing, maximizing the real-time nature of data processing. Dynamically determining and discarding data that fails to meet delay constraints avoids delayed responses due to processing outdated data, reducing latency risks. This approach is suitable for intelligent transportation and autonomous driving scenarios with high real-time requirements. The onboard edge terminal and satellite terminal rationally allocate computing tasks based on their respective capabilities. The onboard edge terminal processes data that meets local delay constraints, fully utilizing its computing power. Onboard sensor data frames exceeding local processing capacity are sent to the satellite terminal for processing, avoiding local task overload and improving computing resource utilization. Furthermore, the satellite terminal uses the available queue buffer capacity to determine the feasibility of remote processing, avoiding blindly sending data and causing satellite overload or resource waste, thereby improving satellite computing efficiency. Providing both local and remote processing options creates a multi-level redundant processing mechanism. Remote processing serves as a backup when local processing fails, enhancing the ability to cope with complex operating conditions and emergencies.

[0013] In some embodiments, the serving satellite terminal is obtained from a plurality of candidate satellite terminals according to a satellite selection strategy, wherein the satellite selection strategy includes:

[0014] Using the vehicle-mounted edge terminal as a transmitter and each candidate satellite terminal as a receiver, obtaining a signal-to-noise ratio with each candidate satellite terminal based on the effective isotropically radiated power of the transmitter, the ratio of the receiving antenna gain to the noise temperature, the path loss, the Boltzmann constant, and the bandwidth, and selecting the candidate satellite terminal corresponding to the maximum signal-to-noise ratio as the serving satellite terminal; or

[0015] The candidate satellite terminals are randomly sorted and their signal-to-noise ratios with the vehicle-mounted edge terminal are determined one by one to see if they are greater than a preset threshold. If so, the candidate satellite terminal is selected as the serving satellite terminal. Otherwise, the next one is determined until a qualified terminal is found.

[0016] The beneficial effects of this embodiment are: in the maximum signal-to-noise ratio strategy, by selecting the satellite terminal with the largest signal-to-noise ratio, the signal quality of the communication link is optimized (the higher the signal-to-noise ratio, the less noise interference the signal suffers, and the higher the accuracy and reliability of data transmission); at the same time, the high-quality communication link can support a higher data transmission rate, reduce data transmission time, and improve the efficiency of data interaction between the vehicle-mounted edge terminal and the satellite terminal. In the random selection strategy, the random sorting and threshold judgment method simplifies the selection process, reduces the computational complexity and selection time, and can avoid the situation where some satellite terminals are overloaded while other satellite terminals are idle, achieve load balancing, and improve the overall utilization of the satellite network; in the case of dynamic changes in the satellite network, the random selection strategy can adapt quickly to ensure that the vehicle-mounted edge terminal can find a satellite terminal that meets the communication quality requirements in a timely manner. In summary, the maximum signal-to-noise ratio strategy is suitable for scenarios with extremely high communication quality requirements, while the random selection strategy is suitable for scenarios with high real-time requirements. The combination of the two strategies can cope with different application scenarios and needs.

[0017] In some embodiments, determining whether the total delay of remote processing of the vehicle-mounted sensor data satisfies a constraint condition based on the queue waiting time includes:

[0018] Obtain upload delay, download delay and satellite task processing delay between the terminal and the service satellite;

[0019] Obtaining a total remote processing delay according to a preset relationship between the total remote processing delay and the queue waiting time, the upload delay, the download delay, and the satellite task processing delay;

[0020] When the total remote processing delay is less than the maximum allowed delay in the constraint condition, it is considered that the constraint condition is satisfied; otherwise, it is considered that the constraint condition is not satisfied.

[0021] The beneficial effect of this embodiment is that by accurately judging whether the total delay of remote processing meets the constraint conditions, unnecessary data transmission and processing is avoided. If it is judged that the conditions are not met, the data can be discarded in advance to avoid wasting computing resources, thereby improving the overall processing efficiency.

[0022] In some embodiments, when the total remote processing delay is less than the maximum allowed delay in the constraint condition, the method further comprises:

[0023] If no feedback information is received from the service satellite terminal regarding the computing service request within the time period corresponding to the constraint condition, the disconnected state from the service satellite terminal is maintained within the preset cooling time period, and the service satellite terminal is reconnected after the preset cooling time period; if feedback information is received from the service satellite terminal regarding the computing service request, the queuing waiting time is updated according to the feedback information.

[0024] The beneficial effects of this embodiment are: through the cooling mechanism, frequent requests are avoided when the network conditions are poor; the reconnection mechanism ensures that the communication link between the on-board edge terminal and the service satellite terminal can continue to work normally, avoiding communication interruption due to short-term problems; by receiving feedback information and updating the queue waiting time, the on-board edge terminal can understand the status of the service satellite terminal in real time, so as to better adapt to the dynamically changing network environment; dynamic updating of the queue waiting time can reduce delay misjudgment caused by outdated information, and ensure the accuracy and timeliness of task scheduling.

[0025] In some embodiments, when the total remote processing delay is less than the maximum allowed delay in the constraint condition, the method further comprises:

[0026] Obtaining an interruption probability according to the total remote processing delay and the maximum allowed delay;

[0027] An interruption strategy including a plurality of interruption periods is acquired and executed according to the interruption probability, wherein the interruption strategy is used to disconnect the service satellite terminal in each interruption period and reconnect the service satellite terminal after the interruption period.

[0028] The beneficial effect of this embodiment is that by actively interrupting some data upload requests, satellite terminal overload is avoided in advance, ensuring that the satellite can efficiently process the remaining tasks.

[0029] In a second aspect, the present application further provides a dynamic task scheduling method based on a non-terrestrial network, the method being applied to a service satellite terminal, the method comprising:

[0030] In response to a computing service request carrying a computing task sent by an onboard edge terminal of any ground vehicle, computing the onboard sensor data in the computing service request, and sending feedback information including the computing result back to the onboard edge terminal;

[0031] Among them, the computing service request is generated according to the vehicle-mounted sensor data after the vehicle-mounted edge terminal determines that the total local processing delay of the vehicle-mounted sensor data does not meet the constraint conditions and the total remote processing delay of the vehicle-mounted sensor data meets the constraint conditions.

[0032] In some embodiments, when the actual total remote processing delay is greater than the maximum allowable delay in the constraint condition due to the queue waiting time, the computing task corresponding to the computing service request is no longer loaded within a preset cooling time.

[0033] In a third aspect, the present application also provides an electronic device comprising one or more processors and a memory; one or more programs are stored in the memory and configured to be executed by the one or more processors to perform any of the methods described above.

[0034] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of any of the above-described methods when the computer program is executed by at least one processor.

[0035] In a fifth aspect, the present application further provides a computer program product, which includes a computer program, and when the computer program is executed by at least one processor, it implements the steps of any of the methods described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The present application is further described below with reference to the accompanying drawings and implementation methods.

[0037] Figure 1 This is a schematic diagram of a dynamic task scheduling scenario of this application.

[0038] Figure 2 This is a flow chart of a dynamic task scheduling method based on a non-terrestrial network provided in an embodiment of the present application.

[0039] Figure 3 This is a flowchart of a constraint condition judgment provided in an embodiment of the present application.

[0040] Figure 4 This is a partial flow chart of a dynamic task scheduling method based on a non-terrestrial network provided in an embodiment of the present application.

[0041] Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] Below, the present application is further described in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the various embodiments described below or the various technical features can be arbitrarily combined to form new embodiments. The implementation procedures of the present application will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific implementation procedures, and the various details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for the purpose of illustrating the present application, and are not intended to limit the scope of protection of the present application.

[0043] The following briefly describes the application scenarios and related terms of the embodiments of the present application to facilitate understanding by those skilled in the art.

[0044] See also Figure 1 , Figure 1 This is a schematic diagram of a dynamic task scheduling scenario for this application. The scenario includes n ground vehicle terminals, which collect data at a fixed inter-frame rate r through onboard sensors. The collected data can be processed onboard or onboard the vehicle using a loading strategy.

[0045] The constraints include a delay real-time constraint δ. Each frame contains a constant computational load C and needs to be processed within the delay real-time constraint δ. Each frame can be processed on the edge terminal or loaded to the LEO satellite for processing. Since LEO satellite communications are less restricted and can reasonably carry more computing power than vehicle-mounted edge terminals (with computing power recorded as C), the computational power of each frame can be processed on the edge terminal. GV ) A more powerful computing platform, with computing power recorded as C LEO , we think C LEO >>C GV Loading data onto the onboard platform can reduce overall processing delays.

[0046] In the related art, taking the vehicle edge computing method based on satellites and roadside units disclosed in CN202410354526.3 as an example, a vehicle edge computing method based on satellites and roadside units is adopted. By dynamically selecting a computing resource allocation strategy, part of the computing tasks are offloaded to nearby roadside units or satellite edge computing nodes for processing. The wide-area coverage of the satellite and the proximal high-speed and low-latency characteristics of the roadside unit are utilized to obtain the optimal correlation factor, bandwidth, computing frequency resources and transmission power, and optimize computing power allocation and wireless channels. In other words, it mainly focuses on the urgency of the task and selects the corresponding information processing end according to the urgency, without considering the impact of the available queue buffer capacity of the satellite terminal on task processing.

[0047] The service satellite terminals mentioned in this application include air and space nodes such as unmanned aerial vehicles, high-altitude platforms, and satellites. To address the above technical issues, this application proposes a dynamic task scheduling method and related equipment based on a non-terrestrial network. With the help of the non-terrestrial network, Internet access is provided from the air. When the ground infrastructure cannot meet the requirements, the non-terrestrial network can achieve on-demand connection, effectively supplementing the ground network. The following will first describe the method, followed by a description of the equipment, etc.

[0048] Example 1

[0049] See also Figure 2 , Figure 2 This is a flow chart of a dynamic task scheduling method based on a non-terrestrial network provided in an embodiment of the present application.

[0050] The method is applied to an on-board edge terminal of a ground vehicle, and the method includes:

[0051] S101, acquiring vehicle-mounted sensor data at a fixed inter-frame rate using a vehicle-mounted sensor, and determining whether a total local processing delay of the vehicle-mounted sensor data satisfies a constraint condition;

[0052] The vehicle edge terminal acquires sensor data from onboard sensors at a fixed inter-frame rate. This fixed rate ensures continuous and stable data acquisition, providing reliable data for subsequent processing. The vehicle edge terminal determines whether the total local processing latency of the collected data meets pre-set constraints. This total local processing latency covers the time required for data processing within the terminal, including data storage and preliminary calculations. Constraints indicate an upper limit on latency, ensuring timely data processing and meeting real-time requirements during vehicle operation.

[0053] S102: If the total local processing delay satisfies the constraint condition, determining to locally process the vehicle-mounted sensor data; otherwise, obtaining a queue waiting time of the serving satellite terminal and determining, based on the queue waiting time, whether the total remote processing delay of the vehicle-mounted sensor data satisfies the constraint condition, where the total remote processing delay is the total processing delay of the vehicle-mounted sensor data by the serving satellite terminal;

[0054] If the total local processing delay meets the constraint conditions, the data will be processed locally (S102), giving full play to the computing power of the vehicle-mounted edge terminal itself and reducing unnecessary remote transmission consumption.

[0055] If the total local processing delay does not meet the constraints, the on-board edge terminal uses the available queue buffer capacity of the serving satellite terminal to calculate the queue waiting time. This capacity can be considered to reflect the current task load of the satellite terminal and its ability to accept new tasks.

[0056] The total remote processing delay can be related to upload delay, download delay, satellite task processing delay, and queue waiting time, which will be explained in detail below. Since queue waiting time is related to the current load and buffer capacity of the satellite terminal, for ease of calculation, in this embodiment, queue waiting time can be used as a strong correlate of the total remote processing delay. The queue waiting time is used to determine whether the total remote processing delay meets the constraints and whether the task of transmitting the vehicle-mounted sensor data to the serving satellite terminal can be completed within the allowed delay. This simplifies the calculation and improves processing efficiency.

[0057] S103: If the total remote processing delay meets the constraint condition, generate a computing service request based on the vehicle-mounted sensor data and send it to the service satellite terminal, and receive feedback information returned by the service satellite terminal; otherwise, discard the vehicle-mounted sensor data.

[0058] If the total remote processing latency meets the constraints, a computing service request is generated and sent to the serving satellite terminal, leveraging the satellite's powerful computing power to process the complex task and compensate for insufficient local computing power. If the total remote processing latency also fails to meet the constraints, the onboard sensor data is discarded and S101 is re-executed to avoid system overload or data backlog. This mechanism ensures real-time data processing by abandoning processing when latency constraints cannot be met, ensuring overall performance and stability.

[0059] The technical solution provided in this embodiment can dynamically adapt to the characteristics of non-terrestrial networks. Considering the dynamic nature of satellite networks (satellite positions and loads constantly change), real-time access to information such as the available satellite queue buffer capacity allows for adjustments to the task scheduling strategy (local processing or satellite processing) for each piece of onboard sensor data acquired at a fixed inter-frame rate, enabling rapid response and adaptation to dynamic network environments.

[0060] There is a feedback mechanism between the on-board edge terminal and the serving satellite terminal. The serving satellite terminal needs to feedback the current queue status and processing capacity information to the on-board edge terminal, so that the on-board edge terminal can dynamically adjust subsequent task scheduling decisions, ensuring that task scheduling decisions are based on the latest network status, and improving the overall efficiency and reliability of this method.

[0061] This ensures that onboard sensor data is processed completely by hierarchically determining whether local and remote processing latency constraints are met. The onboard edge terminal prioritizes local processing to reduce remote transmission latency. When local processing is insufficient, it promptly switches to remote processing, maximizing real-time data processing. Dynamically identifying and discarding data that fails to meet latency constraints avoids delayed responses due to processing outdated data and reduces latency risks. This approach is suitable for intelligent transportation and autonomous driving scenarios with strict real-time requirements. The onboard edge terminal and satellite terminal rationally allocate computing tasks based on their respective capabilities. The onboard edge terminal processes data that meets local latency constraints, fully utilizing its computing power. Onboard sensor data frames exceeding local processing capacity are sent to the satellite terminal for processing, avoiding local task overload and improving computing resource utilization. Furthermore, based on the satellite terminal's available queue buffer capacity, queue waiting time is used to determine the feasibility of remote processing, avoiding overloading the satellite or wasting resources by blindly sending data, thereby improving the efficiency of satellite computing power. Providing both local and remote processing options creates a multi-level redundant processing mechanism. Remote processing serves as a backup when local processing fails, enhancing the ability to cope with complex operating conditions and emergencies.

[0062] In some embodiments, the serving satellite terminal is obtained from a plurality of candidate satellite terminals according to a satellite selection strategy, wherein the satellite selection strategy includes:

[0063] A maximum signal-to-noise ratio strategy is to use the vehicle-mounted edge terminal as the transmitter and each candidate satellite terminal as the receiver. The signal-to-noise ratio between the vehicle-mounted edge terminal and each candidate satellite terminal is obtained based on the transmitter's effective isotropic radiated power, the ratio of the receiving antenna gain to the noise temperature, the path loss, the Boltzmann constant, and the bandwidth. The candidate satellite terminal with the maximum signal-to-noise ratio is used as the serving satellite terminal. Alternatively,

[0064] The random selection strategy randomly sorts the candidate satellite terminals and determines whether their signal-to-noise ratio with the vehicle-mounted edge terminal is greater than a preset threshold. If so, the candidate satellite terminal is selected as the serving satellite terminal. Otherwise, the next one is selected until a qualified terminal is found.

[0065] The maximum signal-to-noise ratio strategy calculates the signal-to-noise ratios of all candidate satellite terminals and selects the satellite terminal with the highest signal-to-noise ratio as the serving satellite terminal. This strategy ensures that the satellite with the best signal quality is selected for connection, thereby improving data transmission reliability and efficiency.

[0066] The random selection strategy simplifies the selection process while ensuring a certain level of communication quality, making it suitable for scenarios with high real-time requirements. Furthermore, if the signal-to-noise ratio (SNR) between each candidate satellite terminal and the vehicle-mounted edge terminal is less than a preset threshold, the strategy automatically switches to the maximum SNR strategy, selecting the candidate satellite terminal with the maximum SNR as the serving satellite terminal.

[0067] Therefore, the maximum signal-to-noise ratio strategy ensures optimal signal quality for the communication link by selecting the satellite terminal with the highest signal-to-noise ratio (the higher the signal-to-noise ratio, the less noise interference the signal suffers, and the higher the accuracy and reliability of data transmission). Furthermore, high-quality communication links can support higher data rates, reduce data transmission time, and improve the efficiency of data exchange between the vehicle-mounted edge terminal and the satellite terminal. In the random selection strategy, random sorting and threshold judgment simplify the selection process, reducing computational complexity and selection time. It can also prevent some satellite terminals from being overloaded while others are idle, achieving load balancing and improving the overall utilization of the satellite network. In the case of dynamic changes in the satellite network, the random selection strategy can quickly adapt, ensuring that the vehicle-mounted edge terminal can promptly find a satellite terminal that meets the communication quality requirements. In summary, the maximum signal-to-noise ratio strategy is suitable for scenarios with extremely high communication quality requirements, while the random selection strategy is suitable for scenarios with high real-time requirements. The combination of these two strategies can address different application scenarios and needs.

[0068] See also Figure 3 , Figure 3 This is a flowchart of a constraint condition judgment provided in an embodiment of the present application.

[0069] In some embodiments, determining whether the total delay of remote processing of the vehicle-mounted sensor data satisfies a constraint condition based on the queue waiting time includes:

[0070] S201, obtaining upload delay, download delay and satellite task processing delay between the terminal and the serving satellite;

[0071] Upload and download latency can be calculated based on the signal-to-noise ratio (SNR) and bandwidth of the communication link. A higher SNR and a larger bandwidth result in shorter upload latency. Satellite mission processing latency can be calculated based on the mission's computational load and the satellite's computing power. A greater mission load or weaker satellite computing power results in longer processing latency.

[0072] S202, obtaining a total remote processing delay according to a preset relationship between the total remote processing delay and the queue waiting time, the upload delay, the download delay, and the satellite task processing delay;

[0073] S220: When the total remote processing delay is less than the maximum allowed delay in the constraint condition, it is considered that the constraint condition is satisfied; otherwise, it is considered that the constraint condition is not satisfied.

[0074] Therefore, by accurately judging whether the total delay of remote processing meets the constraint conditions, unnecessary data transmission and processing is avoided. If it is judged that the conditions are not met, the data can be discarded in advance to avoid wasting computing resources, thereby improving the overall processing efficiency.

[0075] In some embodiments, the method further includes: if no feedback information is received from the service satellite terminal regarding the computing service request within the time period corresponding to the constraint condition, then maintaining a disconnected state from the service satellite terminal within a preset cooling time period, and reconnecting to the service satellite terminal after the preset cooling time period; if feedback information is received from the service satellite terminal regarding the computing service request, updating the queuing waiting time according to the feedback information.

[0076] It can be understood that the on-board edge terminal starts timing after sending the computing service request, and begins to monitor whether it can receive feedback information from the service satellite terminal within the time corresponding to the constraint conditions. If no feedback information is received within the time, the on-board edge terminal will execute the preset cooling mechanism. This application does not limit the preset value of the time. During the preset cooling time, the on-board edge terminal remains disconnected from the current service satellite terminal, avoiding frequent requests, thereby reducing the burden on the satellite terminal and preventing network congestion. After the cooling time is over, the on-board edge terminal automatically reconnects to the service satellite terminal, ensuring that the on-board edge terminal can continue to try to establish a valid connection without manual intervention.

[0077] If the onboard edge terminal successfully receives feedback from the serving satellite terminal within the time limit specified by the constraints, it indicates that the current communication link is functioning properly and the serving satellite terminal is capable of processing the task. The onboard edge terminal updates the queue waiting time based on the received feedback. The feedback information can be considered to contain the current queue status and load information of the serving satellite terminal. The onboard edge terminal uses this information to recalculate the queue waiting time to more accurately estimate the total remote processing delay of subsequent tasks.

[0078] Therefore, the cooling mechanism avoids frequent requests when the network conditions are poor; the reconnection mechanism ensures that the communication link between the on-board edge terminal and the service satellite terminal can continue to work normally, avoiding communication interruptions due to short-term problems; by receiving feedback information and updating the queue waiting time, the on-board edge terminal can understand the status of the service satellite terminal in real time, so as to better adapt to the dynamically changing network environment; dynamic updating of the queue waiting time can reduce delays and misjudgments caused by outdated information, and ensure the accuracy and timeliness of task scheduling.

[0079] See also Figure 4 , Figure 4 This is a partial flow chart of a dynamic task scheduling method based on a non-terrestrial network provided in an embodiment of the present application.

[0080] In some embodiments, when the total remote processing delay is less than the maximum allowed delay in the constraint condition, the method further comprises:

[0081] S104, obtaining an interruption probability according to the total remote processing delay and the maximum allowable delay;

[0082] S105 , obtaining and executing an interruption strategy including multiple interruption periods according to the interruption probability, wherein the interruption strategy is used to disconnect the service satellite terminal during each interruption period and reconnect the service satellite terminal after the interruption period.

[0083] It can be assumed that as the total remote processing latency approaches the maximum allowable latency, the probability of interruption increases accordingly. Based on the calculated interruption probability, an interruption strategy is generated, consisting of multiple interruption periods. During each interruption period, the onboard edge terminal proactively disconnects from the serving satellite terminal and automatically reconnects after the interruption period. This proactive interruption of some data upload requests prevents satellite terminal overload in advance, ensuring that the satellite can efficiently process the remaining tasks.

[0084] As an example, a dynamic task scheduling method based on a non-terrestrial network is provided. The method is applied to an on-board edge terminal of a ground vehicle, and the method includes:

[0085] Using the vehicle-mounted edge terminal as a transmitter and each candidate satellite terminal as a receiver, obtaining a signal-to-noise ratio with each candidate satellite terminal based on the effective isotropically radiated power of the transmitter, the ratio of the receiving antenna gain to the noise temperature, the path loss, the Boltzmann constant, and the bandwidth, and selecting the candidate satellite terminal corresponding to the maximum signal-to-noise ratio as the serving satellite terminal; or

[0086] The candidate satellite terminals are randomly sorted and their signal-to-noise ratios with the vehicle-mounted edge terminal are determined one by one to see if they are greater than a preset threshold. If so, the candidate satellite terminal is selected as the serving satellite terminal. Otherwise, the next one is determined until a qualified terminal is found.

[0087] Acquire vehicle-mounted sensor data at a fixed inter-frame rate using a vehicle-mounted sensor, and determine whether a total local processing delay of the vehicle-mounted sensor data satisfies a constraint condition;

[0088] If the total local processing delay satisfies the constraint condition, the vehicle-mounted sensor data is processed locally; otherwise, the queuing waiting time of the serving satellite terminal is obtained, and the upload delay, download delay, and satellite task processing delay between the serving satellite terminal and the vehicle-mounted sensor data are obtained;

[0089] Obtaining a total remote processing delay according to a preset relationship between the total remote processing delay and the queue waiting time, the upload delay, the download delay, and the satellite task processing delay;

[0090] When the total remote processing delay is less than the maximum allowed delay in the constraint condition, it is considered that the constraint condition is satisfied; otherwise, it is considered that the constraint condition is not satisfied.

[0091] If the total remote processing delay satisfies the constraint condition, a computing service request is generated based on the on-board sensor data and sent to the service satellite terminal, and feedback information returned by the service satellite terminal is received; if no feedback information is received from the service satellite terminal for the computing service request within the time period corresponding to the constraint condition, the disconnection state with the service satellite terminal is maintained for a preset cooling time period, and the service satellite terminal is reconnected after the preset cooling time period; if feedback information is received from the service satellite terminal for the computing service request, the queue waiting time is updated according to the feedback information; alternatively, an interruption probability is obtained based on the total remote processing delay and the maximum allowable delay; an interruption strategy including multiple interruption periods is obtained and executed based on the interruption probability, wherein the interruption strategy is used to disconnect from the service satellite terminal during each interruption period and reconnect to the service satellite terminal after the interruption period;

[0092] If the total remote processing delay does not meet the constraint condition, the onboard sensor data is discarded.

[0093] This technical solution provides a dynamic task scheduling method based on non-terrestrial networks, applicable to vehicle-mounted edge terminals. The vehicle-mounted terminal acts as the transmitter, and each candidate satellite terminal acts as the receiver. The serving satellite is selected by calculating the signal-to-noise ratio (SNR) of each candidate satellite. This SNR calculation takes into account factors such as the transmitter's effective isotropically radiated power, the ratio of the receiving antenna gain to noise temperature, path loss, the Boltzmann constant, and bandwidth. The satellite with the highest SNR is selected as the serving satellite and serves as the serving satellite terminal. Alternatively, the candidate satellites can be randomly sorted and the SNR is determined individually to see if it exceeds a preset threshold to select the serving satellite and serve as the serving satellite terminal.

[0094] Next, the onboard terminal acquires sensor data at a fixed inter-frame rate using the onboard sensors and determines whether the local processing delay of the data meets the constraints. If so, local processing is performed; otherwise, remote processing begins. During remote processing, the queue waiting time is calculated based on the queue buffer capacity of the service satellite, along with the upload delay, download delay, and satellite task processing delay. The total remote processing delay is calculated based on these delay parameters. If the total remote processing delay is less than the maximum allowable delay, a computing service request is generated and sent to the service satellite terminal. If no feedback is received within the specified time, a cooling mechanism is triggered, and reconnection is initiated after a period of time. If feedback is received, the queue waiting time is updated. The interruption probability can also be calculated based on the total remote processing delay and the maximum allowable delay, and an interruption strategy is implemented. If the total remote processing delay does not meet the constraints, the sensor data is discarded.

[0095] This technical solution first improves communication efficiency by optimizing satellite selection and rationally allocating tasks. This not only selects satellites with high signal quality but also avoids unnecessary data transmission and remote processing overhead. Secondly, a cooling mechanism, reconnection strategy, and interruption strategy enhance the reliability of the method's execution, preventing resource waste and system overload. Furthermore, by reducing resource waste and dynamically adjusting task scheduling strategies, resource utilization efficiency is improved.

[0096] In order to further enhance the understanding of the technical solution provided by this application, an example is provided.

[0097] Among them, the position of the vehicle edge terminal is determined by its latitude y v and longitude x v Determine the position of the satellite terminal by its altitude h s and the latitude y projected on the Earth's surface s and longitude x s The position of a satellite in its orbit can be accurately determined based on the data from the service provider. Satellite communication service providers provide real-time and up-to-date satellite trajectories, based on which the position and velocity of satellite nodes in space can be calculated. The distance d between the ground terminal (vehicle-mounted edge terminal) and the communication satellite (satellite terminal) is calculated as follows:

[0098]

[0099] where R E is the radius of the Earth, which is a constant. α is the angle between the ground terminal and the satellite, observed with the Earth's center as the reference point, and can be calculated using Formula 2:

[0100] cos(α)=cos(y s )cos(y v )cos(x v -x s )+sin(yv )sin(y s ) (Formula 2)

[0101] The ground-to-satellite connection is established in the high-capacity millimeter wave (mmWave) band, achieving efficient data rates with a signal-to-noise ratio γ between transmitter i and receiver j under line-of-sight (LOS) conditions. i,j Expressed as:

[0102] γ i,j =EIRP i +(G j / T)-PL i,j -kB (Formula 3)

[0103] EIRP i is the effective isotropic radiated power at the transmitter, in W; (G j / T) is the ratio of the receiving antenna gain to the noise temperature; PL i,j is the path loss; k is the Boltzmann constant; B is the bandwidth, in Hz. For the ground-to-satellite channel, the path loss includes several attenuation components, mainly the scintillation loss PL caused by the change in atmospheric refractive index. s , and atmospheric absorption loss PL due to air and water vapor attenuation g , in summary, the path attenuation is PL = FSPL + PL s +PL g Where FSPL is the free space path loss, which corresponds to the following formula:

[0104] FSPL=92.45+20log(f c )+20log(d)

[0105] d is the free space distance in km; f c is the carrier frequency in GHz.

[0106] The upload delay t can be calculated by the signal-to-noise ratio UL and download delay t DL The estimation methods include delay estimation method based on correlation analysis, delay estimation method based on high-order cumulant, delay estimation method based on characteristic structure analysis and delay estimation method based on cost function. It is assumed that both the service satellite terminal and the vehicle-mounted edge terminal have a first-in-first-out queue to process related computing tasks, and the task processing time is C / C respectively. LEO and C / C GV , which is used to indicate the overall delay of local task processing phase, t GV It can be expressed as:

[0107]

[0108] in is the queue and processing delay (queue waiting time) of the vehicle edge terminal, C / C GV It is the time taken by the vehicle edge terminal to perform computing tasks.

[0109] Combined with the above, the vehicle-mounted edge terminal uploads the task to the service satellite terminal and completes the processing. The overall latency for the vehicle-mounted edge terminal to obtain the processing result, including the upload and download latency as well as the task processing latency, can be expressed as:

[0110]

[0111] where t UL and t DL is the upload and download delay, is the queue and processing delay of the service satellite terminal, C / C GV is the time taken by the service satellite terminal to perform computing tasks. LEO is the overall latency that indicates the total delay in remote processing.

[0112] It can be understood that the entire task needs to meet the processing delay t d <δ constraint. d is the total delay of task processing, including the delay of uploading, downloading and task processing. Δ is the upper limit of the delay of the real-time constraint, that is, the task must be completed within this time range.

[0113] As another example, based on the previous example, when the vehicle-mounted edge terminal faces data processing, if the vehicle-mounted edge terminal has sufficient computing resources, local processing is the most reasonable, which can avoid the additional delay and potential communication overhead caused by transmitting data to the remote server. When the local processing delay cannot meet the constraint of the application real-time requirement δ, the step of loading the data to the service satellite terminal for processing is triggered. Based on the local occupancy of its queue, the vehicle-mounted edge terminal can estimate and the onboard delay t GV :If t GV <δ, the data will be processed locally, otherwise it will be loaded to the serving satellite terminal for processing, or discarded.

[0114] As another example, based on the previous example, when loaded into the serving satellite terminal for processing, the strategy for the onboard edge terminal to connect to the serving satellite terminal may be:

[0115] In the maximum signal-to-noise ratio strategy, each vehicle-mounted edge terminal selects the satellite terminal with the highest signal-to-noise ratio γ for connection. γ is calculated using Formula 3.

[0116] Random selection strategy: each vehicle-mounted edge terminal randomly selects a connectable satellite terminal to connect to, as long as the signal-to-noise ratio of the selected satellite is greater than a pre-set threshold. The superscript S represents the signal-to-noise ratio threshold.

[0117] In both strategies, each vehicle-mounted edge terminal remains connected to the selected satellite until it is out of its coverage range, i.e., as long as the signal-to-noise ratio γ is greater than the threshold γ th Due to the highly dynamic nature of the network serving the satellite terminals, this example can also utilize a feedback mechanism to predict the current load on the satellite.

[0118] Considering that data scheduling and abandonment strategies can reduce latency during loading computational tasks, a feedback mechanism can be implemented between each vehicle-mounted edge terminal and its connected serving satellite terminal to monitor the current load of the system. This feedback mechanism includes the status and occupancy of the serving satellite terminal queue, so that ground vehicles can estimate the waiting time in the queue of low earth orbit satellites. This time is inversely proportional to the available queue buffer capacity.

[0119] The link delay can be calculated, and the distance d to the service satellite terminal is also known, so the total delay t for data loading can be obtained. LEO If t LEO <δ, data is loaded to the service satellite terminal. On the contrary, (e.g. t LEO >>δ) indicates that the computing power of the service satellite terminal has been overloaded and no further calculations can be performed. At this time, the data will be discarded. If a calculation is loaded, then at t o During this time, the service satellite terminal does not receive the second load calculation to reduce the burden on the service satellite terminal and maintain the consistency of the queue to prevent errors caused by multiple thread competition. o The value of is in the predefined interval (1, ) is evenly distributed within Is the preset system value. If the vehicle-mounted edge terminal does not receive feedback from the service satellite terminal within the corresponding time, it is considered that the recent data loading request has been abandoned, and the connection between them is disconnected, and at t o Then make the next attempt. If the requested feedback is received, it can be considered that the feedback information includes The value of , updates the vehicle edge terminal value.

[0120] Alternatively, when t LEO When <δ, the data upload request is still interrupted probabilistically to prevent the satellite's computing power from being overloaded.

[0121] p drop =(tLEO / δ) σ

[0122] Wherein, σ is a set value, which is greater than 0, for example, 0.8, 1 or 1.2. drop is the outage probability. In this case, congestion can be alleviated and latency improved by dropping data before the satellite queue overflows. In specific applications, the state of the local queue can be estimated by the serving satellite terminal.

[0123] Example 2

[0124] This embodiment provides a dynamic task scheduling method based on a non-terrestrial network. The specific embodiments and technical effects achieved by the method are consistent with those described in the above embodiment 1, and some contents are not repeated here. The method is applied to a service satellite terminal, and the method includes:

[0125] In response to a computing service request carrying a computing task sent by an onboard edge terminal of any ground vehicle, computing the onboard sensor data in the computing service request, and sending feedback information including the computing result back to the onboard edge terminal;

[0126] Among them, the computing service request is generated according to the vehicle-mounted sensor data after the vehicle-mounted edge terminal determines that the total local processing delay of the vehicle-mounted sensor data does not meet the constraint conditions and the total remote processing delay of the vehicle-mounted sensor data meets the constraint conditions.

[0127] In some embodiments, the method further comprises:

[0128] When the actual total remote processing delay is greater than the maximum allowable delay in the constraint condition due to the queue waiting time, the computing task corresponding to the computing service request will no longer be loaded within the preset cooling time.

[0129] Example 3

[0130] This embodiment provides an electronic device, and its specific embodiments and technical effects are consistent with the embodiments described in the above embodiments, and some contents are not repeated here.

[0131] The electronic device includes a memory and at least one processor, the memory stores a computer program, and the at least one processor is configured to implement the method as described in any one of the method embodiments when executing the computer program.

[0132] See also Figure 5 , Figure 5 The electronic device 10 may include at least one memory 11, at least one processor 12, and a bus 13 connecting different platform systems.

[0133] The memory 11 may include (computer) readable media in the form of volatile memory, such as a random access memory (RAM) 111 and / or a cache memory 112 , and may further include a read-only memory (ROM) 113 .

[0134] The memory 11 also stores a computer program, which can be executed by the processor 12 so that the processor 12 implements the steps of any of the above methods.

[0135] The memory 11 may also include a utility 114 having at least one program module 115, such program module 115 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include an implementation of a network environment.

[0136] Accordingly, the processor 12 may execute the aforementioned computer program and may execute the utility 114 .

[0137] The processor 12 may be implemented as one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs), or other electronic components.

[0138] Bus 13 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures.

[0139] The electronic device 10 may also communicate with one or more external devices, such as a keyboard, a pointing device, a Bluetooth device, or the like, as well as one or more devices capable of interacting with the electronic device 10, and / or any device that enables the electronic device 10 to communicate with one or more other computing devices (e.g., a router, a modem, etc.). Such communication may be performed via an input / output interface 14. Furthermore, the electronic device 10 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 15. The network adapter 15 may communicate with other modules of the electronic device 10 via the bus 13. It should be understood that, although not shown in the figures, other hardware and / or software modules may be used in conjunction with the electronic device 10 in actual applications, including but not limited to microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms.

[0140] Example 4

[0141] This embodiment provides a computer-readable storage medium, the specific embodiments of which are consistent with the embodiments and technical effects described in the above method embodiments, and some contents are not repeated here. The computer program is stored on the computer-readable storage medium, and when the computer program is executed by at least one processor, the steps of any of the above methods are implemented.

[0142] Example 5

[0143] This embodiment provides a computer program product, the specific embodiments of which are consistent with the embodiments and technical effects described in the above method embodiments, and some contents are not repeated here. The computer program product includes a computer program that, when executed by at least one processor, implements the steps of any of the above methods.

[0144] It should be noted that, in the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: the existence of A alone, the existence of A and B at the same time, and the existence of B alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can represent: a, b, c, a and b, a and c, b and c or a and b and c, where a, b and c can be single or multiple. It is worth noting that "at least one" can also be interpreted as "one or more items".

[0145] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are configured to distinguish similar objects and are not necessarily configured to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0146] This application is explained from the perspectives of purpose of use, effectiveness, progress and novelty, and has complied with the functional enhancement and use requirements emphasized by the Patent Law. The above description and drawings of this application are only preferred embodiments of this application and are not intended to limit this application. Therefore, all structures, devices, features, etc. that are similar or identical to those of this application, that is, all equivalent replacements or modifications made in accordance with the scope of the patent application of this application, should fall within the scope of protection of the patent application of this application.

Claims

1. A dynamic task scheduling method based on non-terrestrial network, characterized in that: The method is applied to an on-board edge terminal of a ground vehicle, and the method includes: Acquire vehicle-mounted sensor data at a fixed inter-frame rate using a vehicle-mounted sensor, and determine whether a total local processing delay of the vehicle-mounted sensor data satisfies a constraint condition; If the total local processing delay satisfies the constraint condition, determining to locally process the onboard sensor data; otherwise, obtaining a queue waiting time of the serving satellite terminal, and determining whether the total remote processing delay of the onboard sensor data satisfies the constraint condition based on the queue waiting time, where the total remote processing delay is the total processing delay of the onboard sensor data by the serving satellite terminal; If the total remote processing delay meets the constraint condition, a computing service request is generated according to the vehicle-mounted sensor data and sent to the service satellite terminal, and feedback information returned by the service satellite terminal is received; otherwise, the vehicle-mounted sensor data is discarded.

2. The dynamic task scheduling method according to claim 1, characterized in that: The serving satellite terminal is obtained from a plurality of candidate satellite terminals according to a satellite selection strategy, wherein the satellite selection strategy includes: Using the vehicle-mounted edge terminal as a transmitter and each candidate satellite terminal as a receiver, obtaining a signal-to-noise ratio with each candidate satellite terminal based on the effective isotropically radiated power of the transmitter, the ratio of the receiving antenna gain to the noise temperature, the path loss, the Boltzmann constant, and the bandwidth, and selecting the candidate satellite terminal corresponding to the maximum signal-to-noise ratio as the serving satellite terminal; or The candidate satellite terminals are randomly sorted and their signal-to-noise ratios with the vehicle-mounted edge terminal are determined one by one to see if they are greater than a preset threshold. If so, the candidate satellite terminal is selected as the serving satellite terminal. Otherwise, the next one is determined until a qualified terminal is found.

3. The dynamic task scheduling method according to claim 1, characterized in that: The determining, based on the queue waiting time, whether the total delay of remote processing of the vehicle-mounted sensor data satisfies a constraint condition includes: Obtain upload delay, download delay and satellite task processing delay between the terminal and the service satellite; Obtaining a total remote processing delay according to a preset relationship between the total remote processing delay and the queue waiting time, the upload delay, the download delay, and the satellite task processing delay; When the total remote processing delay is less than the maximum allowed delay in the constraint condition, it is considered that the constraint condition is satisfied; otherwise, it is considered that the constraint condition is not satisfied.

4. The dynamic task scheduling method according to claim 3, characterized in that: When the total remote processing delay is less than the maximum allowed delay in the constraint condition, the method further includes: If no feedback information is received from the service satellite terminal regarding the computing service request within the time period corresponding to the constraint condition, the disconnected state from the service satellite terminal is maintained within the preset cooling time period, and the service satellite terminal is reconnected after the preset cooling time period; if feedback information is received from the service satellite terminal regarding the computing service request, the queuing waiting time is updated according to the feedback information.

5. The dynamic task scheduling method according to claim 3, characterized in that: When the total remote processing delay is less than the maximum allowed delay in the constraint condition, the method further includes: Obtaining an interruption probability according to the total remote processing delay and the maximum allowed delay; An interruption strategy including a plurality of interruption periods is acquired and executed according to the interruption probability, wherein the interruption strategy is used to disconnect the service satellite terminal in each interruption period and reconnect the service satellite terminal after the interruption period.

6. A dynamic task scheduling method based on a non-terrestrial network, characterized in that: The method is applied to a service satellite terminal, and the method comprises: In response to a computing service request carrying a computing task sent by an onboard edge terminal of any ground vehicle, computing the onboard sensor data in the computing service request, and sending feedback information including the computing result back to the onboard edge terminal; Among them, the computing service request is generated according to the vehicle-mounted sensor data after the vehicle-mounted edge terminal determines that the total local processing delay of the vehicle-mounted sensor data does not meet the constraint conditions and the total remote processing delay of the vehicle-mounted sensor data meets the constraint conditions.

7. The dynamic task scheduling method according to claim 6, characterized in that: The method further comprises: When the actual total remote processing delay is greater than the maximum allowable delay in the constraint condition due to the queue waiting time, the computing task corresponding to the computing service request will no longer be loaded within the preset cooling time.

8. An electronic device, characterized in that: The electronic device includes a memory and at least one processor, the memory stores a computer program, and the processor is configured to execute the computer program so that the electronic device can implement the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by at least one processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, which implements the steps of the method according to any one of claims 1 to 7 when executed by at least one processor.

Images