Delay optimization method and system for automatic driving Internet of Vehicles

Abstract

The invention discloses an automatic driving Internet of Vehicles time delay optimization method and system, and the method comprises the steps: determining a high-time-delay traffic flow according to the arrival rate and the processing rate of a data package at a cloud end in each preset time period in the past, and a preset time delay threshold value; according to the arrival rate and the processing rate of the data packet at the edge node in each previous preset time period, and a first preset load threshold, determining an overload traffic flow; according to the position of the vehicle, the average speed and the task route, calculating the vehicle flow entering the range of the current edge node in each preset time period in the future; when the calculated traffic flow is greater than or equal to the high-delay traffic flow, processing a data packet corresponding to the traffic flow at an edge node; and when the calculated traffic flow is greater than or equal to the overload traffic flow, forwarding a data packet corresponding to the traffic flow of the overload part to the target edge node for processing. Through the method, the data processing time delay, the operation cost and the risk of sensitive data leakage can be reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of automatic driving Internet of Vehicles, in particular to an automatic driving Internet of Vehicles time delay optimization method and system. BACKGROUND

[0002] Currently, there are two mainstream schemes for automatic driving Internet of Vehicles. Scheme one, automatic driving Internet of Vehicles has no edge computing node, all vehicle end data is reported to the cloud, and the cloud server is used to perform related processing tasks. Scheme two, an edge computing node is deployed, and the edge node and the cloud server are used to cooperatively predict to improve the cache hit rate at the edge node.

[0003] The problem of scheme one is that all vehicle end data needs to be uploaded to the cloud, and a large amount of data transmission will cause the data transmission speed to slow down, increase the data processing time delay, and is not friendly to the scene with high real-time requirement. Data transmission occupies a large amount of network bandwidth, increasing the operation cost. In the process of uploading all data to the cloud for processing, there is a risk of sensitive data leakage or interception. The problem of scheme two is that the bearing capacity of a single edge node for vehicle end data is not considered, which easily leads to unbalanced load of each edge node, and further causes the increase of data processing time delay at part of the edge nodes. SUMMARY

[0004] The present application provides an automatic driving Internet of Vehicles time delay optimization method and system, which can solve the technical problems of large data processing time delay, high operation cost, and risk of sensitive data leakage in the current automatic driving Internet of Vehicles technology.

[0005] To achieve the above-mentioned purpose, in a first aspect, the present application provides an automatic driving Internet of Vehicles time delay optimization method, which comprises: According to the arrival rate and processing rate of data packets at the cloud in each past preset time period and a preset time delay threshold, a high time delay vehicle flow is determined.

[0006] According to the arrival rate and processing rate of data packets at the edge node in each past preset time period and a first preset load threshold, an overload vehicle flow is determined.

[0007] According to the position, average speed and task route of the vehicle, the vehicle flow entering the range of the current edge node in each future preset time period is calculated.

[0008] When the calculated vehicle flow is greater than or equal to the high time delay vehicle flow, the processing of the data packets corresponding to the vehicle flow is performed at the edge node.

[0009] When the calculated vehicle flow is greater than or equal to the overload vehicle flow, the data packets corresponding to the overload part of the vehicle flow are forwarded to the target edge node for processing.

[0010] Further, in an embodiment, the determining the high-latency vehicle flow according to the data packet arrival rate and processing rate at the cloud in each preset time period in the past and the preset latency threshold comprises: calculating the processing latency of the same type of data packet in the corresponding preset time period according to the data packet arrival rate and processing rate at the cloud in each preset time period in the past.

[0011] taking the vehicle flow corresponding to the processing latency greater than the preset latency threshold as the high-latency vehicle flow.

[0012] Further, in an embodiment, the determining the overload vehicle flow according to the data packet arrival rate and processing rate at the edge node in each preset time period in the past and the first preset load threshold comprises: calculating the load of the corresponding edge node in the corresponding preset time period according to the data packet arrival rate and processing rate at each edge node in each preset time period in the past.

[0013] taking the vehicle flow corresponding to the load greater than the first preset load threshold as the overload vehicle flow.

[0014] Further, in an embodiment, the forwarding the data packets corresponding to the vehicle flow of the overload part to the target edge node comprises: each edge node is connected through a physical link to form a tree topology.

[0015] fixing the data packets of the same data stream to the same physical link based on the MAC address or IP address of the target edge node, and forwarding the data packets to the target edge node through the physical link.

[0016] Further, in an embodiment, the determining the target edge node comprises: taking the current edge node as the root node through the tree topology, and sequentially finding the child edge nodes of the current edge node in a pre-order traversal manner.

[0017] taking the first child edge node whose load is lower than the second preset load threshold as the target edge node.

[0018] Further, in an embodiment, the calculating the vehicle flow entering the range of the current edge node in each preset time period in the future according to the position, average speed, and task route of the vehicle comprises: calculating the remaining path of the vehicle according to the current position and task route of the vehicle.

[0019] calculating the required time according to the remaining path and average speed.

[0020] determining the vehicle flow entering the range of the current edge node in each preset time period in the future according to the required time.

[0021] Further, in an embodiment, when the calculated vehicle flow is less than the high-latency vehicle flow, processing of data packets corresponding to the vehicle flow is performed at the cloud.

[0022] Further, in an embodiment, when the calculated vehicle flow is less than the overload vehicle flow, processing of data packets corresponding to the vehicle flow is performed at the current edge node.

[0023] Further, in an embodiment, the arrival rate and processing rate of data packets at the cloud and the arrival rate and processing rate of data packets at the edge node are obtained from a log system.

[0024] In a second aspect, based on the above-mentioned automatic driving vehicle networking latency optimization method, the present application provides an automatic driving vehicle networking latency optimization system, the system comprising: a high-latency vehicle flow module configured to determine a high-latency vehicle flow according to the arrival rate and processing rate of data packets at the cloud in each past preset time period and a preset latency threshold.

[0025] an overload vehicle flow module configured to determine an overload vehicle flow according to the arrival rate and processing rate of data packets at the edge node in each past preset time period and a first preset load threshold.

[0026] a calculation module configured to calculate a vehicle flow entering a range of the current edge node in each future preset time period according to the position, average speed and task route of the vehicle.

[0027] a processing module configured to perform processing of data packets corresponding to the vehicle flow at the edge node when the calculated vehicle flow is greater than or equal to the high-latency vehicle flow, and perform processing of data packets corresponding to an overload part of the vehicle flow at the target edge node.

[0028] a forwarding module configured to forward data packets corresponding to the overload part of the vehicle flow to the target edge node when the calculated vehicle flow is greater than or equal to the overload vehicle flow.

[0029] The technical scheme provided by the embodiments of the present application has the following beneficial effects: This application determines high-latency traffic flow based on the arrival and processing rates of data packets at the cloud in previous preset time periods, as well as a preset latency threshold. It also determines overloaded traffic flow based on the arrival and processing rates of data packets at the edge nodes in previous preset time periods, as well as a first preset load threshold. Based on the vehicle's location, average speed, and task route, it calculates the traffic flow entering the current edge node's range in future preset time periods. When the calculated traffic flow is greater than or equal to the high-latency traffic flow, the data packets corresponding to the traffic flow are processed at the edge node. This avoids increased data processing latency caused by uploading all vehicle-side data to the cloud, reduces the bandwidth consumption of the data transmission network, thereby reducing the operating cost of the Internet of Vehicles (IoV) and also reduces the risk of sensitive information leakage when data is uploaded to the cloud to a certain extent. Secondly, when the calculated traffic flow is greater than or equal to the overloaded traffic flow, the data packets corresponding to the overloaded traffic flow are forwarded to the target edge node for processing. This method takes into account the edge node's ability to process vehicle-side data. Once the current edge node is overloaded, other edge nodes are scheduled to share the traffic flow, effectively avoiding the problem of unbalanced load among edge nodes caused by the overload of a single node, thereby reducing the data processing latency at a single edge node. Attached Figure Description

[0030] Figure 1 This is a flowchart of an autonomous driving vehicle-to-everything (V2X) latency optimization method according to an embodiment of this application.

[0031] Figure 2 This is a block diagram of an autonomous driving vehicle-to-everything (V2X) latency optimization system according to an embodiment of this application.

[0032] Figure 3 This is a comparison chart of the latency of traditional processing methods, full-edge unloading methods, and the method of this application.

[0033] Figure 4 This is a comparison chart of the maximum load on edge nodes between the full edge unloading method and the method in this application. Detailed Implementation

[0034] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0035] First, some of the technical terms used in this application will be explained to help those skilled in the art understand this application.

[0036] Vehicle-to-everything (V2X) technology is a network that connects vehicles to each other, vehicles to infrastructure, vehicles to pedestrians, and other related devices via wireless communication. By integrating vehicle sensors, communication modules, cloud computing, and big data technologies, it enables information exchange and collaboration between vehicles, as well as intelligent interaction between vehicles and the external environment, thereby improving traffic safety, efficiency, and comfort.

[0037] Cache hit rate: This is an important metric for measuring the performance of a caching system, reflecting the proportion of data stored in the cache that is successfully accessed. Specifically, the cache hit rate is the ratio of the number of requests for data that exists in the cache and is successfully accessed to the total number of data requests within a certain period.

[0038] The cloud refers to a collection of high-performance computing and storage resources located at the center of a network. It has powerful computing and storage capabilities and can handle large-scale data analysis and complex computing tasks.

[0039] Edge nodes are computing devices located at the edge of a network, typically close to data sources (such as vehicles) or user terminals. Edge nodes can perform data processing and analysis locally, reducing the need for data transmission to the cloud, thereby reducing latency and bandwidth consumption.

[0040] MAC address: Media Access Control address, is a unique identifier used to identify network devices.

[0041] An IP address is an address used to uniquely identify a device on the Internet or a local area network (LAN). It allows communication and data transfer between devices. Unlike MAC addresses, IP addresses are network layer identifiers and are primarily used for routing and network layer communication.

[0042] AP bridge: Access Point, a network configuration technology used to connect multiple wireless access points to form a larger wireless network. This technology is typically used to extend the coverage of a wireless network or to enable seamless roaming and data transmission between multiple wireless access points.

[0043] WDS: Wireless Distribution System, is a wireless network technology used to connect multiple wireless access points (APs) to form a larger wireless network. It allows data transmission between the access points. WDS is primarily used to extend the coverage of wireless networks, enable seamless roaming, and improve network load balancing capabilities.

[0044] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0045] In a first aspect, embodiments of this application provide a method for optimizing latency in autonomous vehicle networking.

[0046] In one embodiment, see Figure 1 As shown, the above-mentioned method for optimizing vehicle-to-everything (V2X) latency in autonomous driving includes: S1. Determine the high-latency traffic flow based on the arrival rate and processing rate of data packets in the cloud during previous preset time periods, as well as the preset latency threshold.

[0047] S2. Determine the overloaded vehicle flow rate based on the arrival rate and processing rate of data packets at the edge nodes in previous preset time periods, as well as the first preset load threshold.

[0048] S3. Based on the vehicle's location, average speed, and mission route, calculate the traffic flow entering the current edge node within each preset time period in the future.

[0049] S4. When the calculated traffic flow is greater than or equal to the high-latency traffic flow, the data packets corresponding to that traffic flow are processed at the edge node.

[0050] S5. When the calculated traffic flow is greater than or equal to the overloaded traffic flow, the data packets corresponding to the overloaded traffic flow are forwarded to the target edge node for processing.

[0051] In this embodiment, by processing data packets corresponding to high-latency traffic flow at the edge nodes and forwarding data packets corresponding to overloaded traffic flow to the target edge nodes for processing, the data processing latency can be effectively reduced, the load pressure on the cloud can be alleviated, and the problem of increased data processing latency due to overload of a single edge node can be avoided. This improves the overall performance and stability of the system and enhances the real-time performance and reliability of autonomous vehicle networking.

[0052] Furthermore, in one embodiment, step S1 above determines the high-latency traffic flow based on the arrival rate and processing rate of data packets in the cloud during previous preset time periods, as well as a preset latency threshold. The specific steps are as follows: S101. Based on the arrival rate and processing rate of data packets at the cloud in previous preset time periods, calculate the processing delay of the same type of data packets in the corresponding preset time period. The processing delay can be calculated according to the following formula (1): (1), in, Indicates processing delay. Represents the coefficient. This represents the rate at which data packets arrive at the cloud within a preset time period. This indicates the processing rate of data packets in the cloud within a preset time period.

[0053] In this embodiment, the data packets arriving at the cloud within a preset time period are the same type of data packets uploaded by the vehicles. The arrival rate and processing rate of the data packets at the cloud are obtained from the log system.

[0054] S102. The traffic flow corresponding to the processing delay that is greater than the preset delay threshold is taken as the high delay traffic flow.

[0055] In this embodiment, by calculating the processing latency of the same type of data packets in the cloud and identifying high-latency traffic flow, high-latency risk prediction is achieved, providing a basis for traffic processing and scheduling.

[0056] Furthermore, in one embodiment, step S2 above, which determines the overloaded vehicle traffic based on the arrival rate and processing rate of data packets at the edge nodes within previous preset time periods and a first preset load threshold, specifically involves the following steps: S201. Based on the arrival rate and processing rate of data packets at each edge node in previous preset time periods, calculate the load of the corresponding edge node in the corresponding preset time period. The load can be calculated using the following formula (2): (2), in, Indicates the first time period within the preset time period The load of each edge node Indicates the first time period within the preset time period The arrival rate of data packets at each edge node. Indicates the first time period within the preset time period The processing rate of data packets at each edge node.

[0057] In this embodiment, the data packets arriving at the edge node within a preset time period are all data packets uploaded by the vehicle. The arrival rate and processing rate of data packets at the edge node are obtained from the log system.

[0058] S202. The traffic flow corresponding to the load that is greater than the first preset load threshold is taken as the overload traffic flow.

[0059] In this embodiment, by calculating the load of each edge node within a preset time period and determining the overload traffic flow based on the first preset load threshold, accurate monitoring and evaluation of the load status of edge nodes are achieved. This enables timely detection of which edge nodes are at risk of overload due to excessive traffic flow, thus providing a basis for subsequent traffic scheduling and load balancing strategies.

[0060] Furthermore, in one embodiment, step S3 above, which calculates the traffic flow entering the current edge node within each preset time period based on the vehicle's location, average speed, and task route, specifically involves the following steps: S301. Calculate the vehicle's remaining path based on its current location and mission route.

[0061] S302. Calculate the required time based on the remaining path and average speed.

[0062] S303. Determine the traffic flow entering the current edge node within each preset time period in the future, based on the required time. The coverage area of ​​the current edge node can be set as an area with a radius of 1 kilometer centered on that edge node.

[0063] In this embodiment, by acquiring the vehicle's location, average speed, and task route in real time, the system accurately calculates the time required for a vehicle to enter the edge node's range, thereby predicting traffic flow in future preset time periods. This accurate predictive capability enables the system to plan resource allocation in advance, rationally arrange the computing and storage resources of edge nodes, and optimize traffic scheduling strategies. When the calculated traffic flow is greater than or equal to the high-latency traffic flow determined in step S1, the data packets corresponding to this traffic flow are processed at the edge node. When the calculated traffic flow is greater than or equal to the overloaded traffic flow determined in step S2, the data packets corresponding to the overloaded portion of the traffic flow are forwarded to the target edge node for processing.

[0064] Furthermore, in one embodiment, step S5 above, which forwards the data packets corresponding to the traffic flow of the overloaded portion to the target edge node, is performed as follows: Each edge node is connected by a physical link, forming a tree-like topology.

[0065] Based on the MAC address or IP address of the target edge node, data packets of the same data stream are fixed to the same physical link, and the data packets are forwarded to the target edge node through the physical link.

[0066] In this embodiment, each edge node is connected via physical links and forms a WDS through AP bridging. The system has a tree topology (supporting up to 3 hops). Using this tree topology, with the current edge node from step S3 as the root node, its child edge nodes are searched sequentially using a preorder traversal. The first child edge node whose load is below a second preset load threshold is selected as the target edge node. Based on the target edge node's MAC address or IP address, data packets of the same data stream are fixed to the same physical link, and the data packets are forwarded to the target edge node through this physical link. This method forwards data packets corresponding to overloaded traffic flow to the target edge node, achieving traffic scheduling between edge nodes and thus avoiding the problem of increased latency caused by the overload of a single edge node.

[0067] Furthermore, in one embodiment, when the traffic flow calculated in step S3 is less than the high-latency traffic flow, the data packets corresponding to the traffic flow are processed in the cloud.

[0068] When the calculated traffic flow is less than the overloaded traffic flow, the data packets corresponding to the traffic flow are processed at the current edge node.

[0069] Secondly, based on the above-described embodiments of the autonomous driving vehicle-to-everything (V2X) latency optimization method, an embodiment of an autonomous driving V2X latency optimization system is provided. See also... Figure 2 As shown, the above system includes a high-latency traffic flow module, an overload traffic flow module, a calculation module, a processing module, and a forwarding module, specifically: The high-latency traffic flow module is used to determine high-latency traffic flow based on the arrival rate and processing rate of data packets in the cloud during previous preset time periods, as well as preset latency thresholds.

[0070] The overload traffic flow module is used to determine the overload traffic flow based on the arrival rate and processing rate of data packets at the edge nodes in previous preset time periods, as well as a first preset load threshold.

[0071] The calculation module is used to calculate the traffic flow entering the current edge node within each preset time period in the future, based on the vehicle's location, average speed, and task route.

[0072] The processing module is used to process the data packets corresponding to the calculated traffic flow at the edge node when the calculated traffic flow is greater than or equal to the high-latency traffic flow, and is also used to process the data packets corresponding to the traffic flow of the overloaded part at the target edge node.

[0073] The forwarding module is used to forward the data packets corresponding to the overloaded traffic flow to the target edge node when the calculated traffic flow is greater than or equal to the overloaded traffic flow.

[0074] To verify the effectiveness of the proposed autonomous driving vehicle-to-everything (V2X) latency optimization method, an urban road scenario was generated using SUMO (Simulation of Urban Mobility, an open-source microscopic traffic simulation platform), and a V2X communication environment was built using a network simulation platform. Within a 2 km × 2 km urban area, two edge nodes and one cloud center were deployed, simulating 10 autonomous vehicles to simulate the arrival and processing of road cleaning tasks by autonomous sanitation vehicles. A fleet configuration was used to test sudden traffic surges. The proposed method (hereinafter referred to as Hybrid-Sched) was compared with traditional processing methods (hereinafter referred to as Local Only) (i.e., data uploaded to the cloud for processing) and all-edge offloading methods (hereinafter referred to as All Edge). The performance evaluation metric was the average end-to-end latency. With maximum node load rate .

[0075] Simulation results are as follows Figure 3 and Figure 4 As shown, Figure 3 This is a comparison chart showing the latency of traditional processing methods, full-edge unloading methods, and the method described in this application. Figure 4 This is a comparison chart of the maximum load on edge nodes between the full-edge unloading method and the method described in this application. From Figure 3 As can be seen from this, when the vehicle density is 10 vehicles: the Local Only method corresponds to The time is approximately 46.2ms for the All Edge method. It takes approximately 30.4ms, while the corresponding method in this application (Hybrid-Sched) is... It takes approximately 19.7ms. Figure 4 In the diagram, Node group1 represents the All Edge method, Node group2 represents the method of this application (Hybrid-Sched), Nodeindex1 represents the first edge node, and Nodeindex2 represents the second edge node. It can be concluded that the method of this application not only reduces the latency of edge nodes but also improves the load balancing among edge nodes.

[0076] This application provides an innovative method and system for optimizing latency in autonomous vehicle-to-everything (V2X) networks. This solution achieves significant improvements and optimizations in several key aspects, effectively solving many problems existing in the prior art. Specifically, by accurately calculating the load and traffic flow of cloud and edge nodes, this application can identify high latency and overload risks in advance and rationally schedule the processing location of data packets accordingly. This process not only significantly reduces data processing latency but also greatly improves the system's response speed and real-time performance, ensuring the efficient operation of autonomous vehicle-to-everything (V2X) networks in various complex scenarios.

[0077] Meanwhile, this application optimizes the traffic scheduling strategy, achieving load balancing among edge nodes. In this way, the system can utilize resources more efficiently, avoiding performance bottlenecks caused by overload of a single edge node, thereby enhancing system reliability and stability. Furthermore, this application reduces operating costs by decreasing the amount of data transmitted during cloud processing, while simultaneously improving data security and reducing the risk of sensitive data leakage.

[0078] In summary, the method of this application not only significantly improves the performance and economic benefits of autonomous vehicle networks, but also provides strong support for the efficient and stable operation of autonomous vehicle networks.

[0079] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0080] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0081] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0082] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0083] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0084] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0085] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for optimizing latency in autonomous vehicle-to-everything (V2X) communication, characterized in that, The method includes: Based on the arrival and processing rates of data packets in the cloud during previous preset time periods, as well as preset latency thresholds, the high-latency traffic flow is determined. The overloaded vehicle flow rate is determined based on the arrival rate and processing rate of data packets at the edge nodes in previous preset time periods, as well as the first preset load threshold. Based on the vehicle's location, average speed, and mission route, calculate the traffic flow entering the current edge node within each preset time period in the future; When the calculated traffic flow is greater than or equal to the high-latency traffic flow, the data packets corresponding to the traffic flow are processed at the edge node; When the calculated traffic flow is greater than or equal to the overloaded traffic flow, the data packets corresponding to the overloaded traffic flow are forwarded to the target edge node for processing.

2. The method for optimizing latency in autonomous vehicle networking as described in claim 1, characterized in that, The process of determining high-latency traffic flow based on the arrival and processing rates of data packets in the cloud during previous preset time periods, and a preset latency threshold, includes: Based on the arrival rate and processing rate of data packets in the cloud during previous preset time periods, calculate the processing latency of the same type of data packets within the corresponding preset time period. Traffic flow with processing delays exceeding a preset delay threshold is considered high-latency traffic flow.

3. The method for optimizing latency in autonomous vehicle networking as described in claim 1, characterized in that, The step of determining the overloaded vehicle traffic based on the arrival rate and processing rate of data packets at the edge nodes in previous preset time periods, and a first preset load threshold, includes: Based on the arrival rate and processing rate of data packets at each edge node in previous preset time periods, calculate the load of the corresponding edge node in the corresponding preset time period. The traffic flow corresponding to a load greater than the first preset load threshold is taken as the overload traffic flow.

4. The method for optimizing latency in autonomous vehicle networking as described in claim 3, characterized in that, The method for forwarding the data packets corresponding to the overloaded traffic flow to the target edge node is as follows: Each edge node is connected by a physical link, forming a tree-like topology; Based on the MAC address or IP address of the target edge node, data packets of the same data stream are fixed to the same physical link, and the data packets are forwarded to the target edge node through the physical link.

5. The method for optimizing latency in autonomous vehicle networking as described in claim 4, characterized in that, The target edge node is determined as follows: Using the tree topology, with the current edge node as the root node, its child edge nodes are searched sequentially in a preorder traversal manner. The first child edge node whose load is lower than the second preset load threshold is selected as the target edge node.

6. The method for optimizing latency in autonomous vehicle networking as described in claim 1, characterized in that, The calculation of traffic flow entering the current edge node within each preset time period based on vehicle location, average speed, and task route includes: Calculate the vehicle's remaining path based on its current location and mission route; Calculate the required time based on the remaining path and average speed; The required timeframe determines the traffic flow entering the current edge node within each preset time period in the future.

7. The method for optimizing latency in autonomous vehicle networking as described in claim 1, characterized in that, When the calculated traffic flow is less than the high-latency traffic flow, the data packets corresponding to the traffic flow are processed in the cloud.

8. The method for optimizing latency in autonomous vehicle networking as described in claim 1, characterized in that, When the calculated traffic flow is less than the overloaded traffic flow, the data packets corresponding to the traffic flow are processed at the current edge node.

9. The method for optimizing latency in autonomous vehicle networking as described in claim 1, characterized in that, The arrival and processing rates of data packets at the cloud and at the edge nodes are obtained from the log system.

10. A latency optimization system based on the latency optimization method for autonomous vehicle networking according to any one of claims 1-9, characterized in that, The system includes: The high-latency traffic flow module is used to determine high-latency traffic flow based on the arrival rate and processing rate of data packets in the cloud during previous preset time periods, as well as preset latency thresholds. The overload vehicle flow module is used to determine the overload vehicle flow based on the arrival rate and processing rate of data packets at the edge node in previous preset time periods, as well as a first preset load threshold. The calculation module is used to calculate the traffic flow entering the current edge node within each preset time period in the future, based on the vehicle's location, average speed, and task route. The processing module is used to process the data packets corresponding to the traffic flow at the edge node when the calculated traffic flow is greater than or equal to the high-latency traffic flow, and is also used to process the data packets corresponding to the traffic flow of the overloaded part at the target edge node. The forwarding module is used to forward the data packets corresponding to the overloaded traffic flow to the target edge node when the calculated traffic flow is greater than or equal to the overloaded traffic flow.

Images