ARINC664 network fault tolerance and delay optimization method

Abstract

The invention discloses an ARINC664 network fault tolerance and delay optimization method, which comprises the following steps of: S1) managing data traffic by adopting a virtual link in an ARINC664 network, and realizing the fault tolerance of the network through a redundant path; s2) carrying out modeling on the basis of a three-dimensional polynomial differential system, and describing traffic characteristics of a virtual link in the ARINC664 network and a congestion phenomenon under a high-load condition; s3) controlling the flow and the data transmission jitter in the ARINC664 network by using a Darboux polynomial and a first integral; and S4) adjusting the traffic weight of the virtual link in real time by constructing an inverse Jacobi multiplier, and optimizing the traffic control on the redundant path.The method can effectively reduce the network jitter, shorten the transmission delay and enhance the fault-tolerant capability of the system under high-load and complex conditions, and provides more reliable technical guarantee for a modern aviation communication system.

Description

Technical Field

[0001] The present invention relates to a network fault tolerance and delay optimization method, in particular to an ARINC664 network fault tolerance and delay optimization method. Background Art

[0002] With the rapid development of aerospace technology, the complexity and diversity of aircraft are increasing, and the requirements for flight safety and efficiency are also increasing. The operation of modern aerospace systems depends on network communication systems with high reliability, low latency and strong fault tolerance. These systems undertake key tasks such as flight data transmission, control signal exchange and status monitoring. Especially in commercial aviation, military aviation and unmanned aerial vehicle systems, ensuring the real-time and accuracy of flight control data plays an irreplaceable role in ensuring flight safety.

[0003] At present, the ARINC664 Part 7 (Avionic Full-Duplex Switched Ethernet, AFDX for short) specification, which is widely used in the aviation field, is an important application standard of Ethernet technology in aviation communications. AFDX manages data traffic through virtual links (VL) and ensures the fault tolerance and reliability of the network through redundant paths. However, the current ARINC664 network optimization methods mainly focus on hardware design improvements and protocol optimization. Although these methods can improve system performance to a certain extent, they still have great limitations when facing complex flight conditions and sudden network anomalies. For example, data congestion under high load conditions may cause a significant increase in network jitter and delay, which in turn affects the real-time response of the flight control system. In addition, the load distribution of redundant paths often relies on fixed rules and is difficult to dynamically adapt to complex operating environments. To this end, a more scientific, efficient and dynamically adjustable network optimization method is urgently needed to meet the high standards of modern aviation communication systems for real-time, stability and fault tolerance. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an ARINC664 network fault tolerance and delay optimization method, which can effectively reduce network jitter, shorten transmission delay, and enhance the fault tolerance of the system under high load and complex conditions, and provide more reliable technical guarantee for modern aviation communication systems.

[0005] The technical solution adopted by the present invention to solve the above technical problems is to provide an ARINC664 network fault tolerance and delay optimization method, comprising the following steps: S1) using virtual links to manage data traffic in the ARINC664 network, and realizing network fault tolerance through redundant paths; S2) modeling based on a three-dimensional polynomial differential system to describe the traffic characteristics of virtual links in the ARINC664 network and congestion phenomena under high load conditions; S3) using Darboux polynomials and first integrals to control the traffic and data transmission jitter in the ARINC664 network; S4) by constructing inverse Jacobi multipliers, adjusting the virtual link traffic weights in real time, and optimizing the traffic control on redundant paths.

[0006] Furthermore, the step S2 is modeled as follows: in, Indicates the data flow transmission rate or bandwidth occupancy of the virtual link; Represents the transmission delay between network nodes; Indicates the system's fault tolerance or the degree of dynamic load distribution of redundant paths; A gain or attenuation factor that represents the data flow rate; Describe the damping or amplification factors associated with transmission delays; and Measures the sensitivity of load distribution across redundant paths.

[0007] Furthermore, the algorithm in step S3 is designed based on the first integral theory and Hilbert's finite basis theorem, and is actually calculated and verified in combination with software Mathematica and Singular. The Darboux polynomial finally obtained is: The corresponding cofactor is: ; The first integral determined is: ; 0.1 ≤ ≤ 1.0,0.1 ≤ ≤ 0.5.

[0008] Furthermore, the transmission delay in the ARINC664 network is composed of the delays of the sending end, network transmission and receiving end, and the total delay is: ;in To delay sending, is the transmission delay, The step S3 is performed by first integrating Control jitter, allocate traffic weights to VLs, control jitter ≤ 1 ms, maximum delay ≤ 10 ms, and bandwidth utilization ≥ 80%.

[0009] Furthermore, the jitter formula in the ARINC664 network is: in Indicates the frame size, bytes range from 64 to 1500; is the network bandwidth, ranging from 100 Mbps to 1 Gbps; It is the minimum frame interval, with a duration of 20~00 microseconds.

[0010] Furthermore, the inverse Jacobi multiplier constructed in step S3 is: ; Step S3 combines the traffic weight with the system load. When the traffic of a virtual link exceeds the threshold, the inverse Jacobi multiplier is used to adjust the traffic weight.

[0011] Furthermore, step S1 obtains the real-time activity status of each virtual link, including bandwidth utilization, packet loss rate and node status, and dynamically assigns different priorities to packets; step S3 allocates traffic to normal nodes according to the current load balancing strategy, and when a node failure is detected, reallocates the traffic to available redundant paths.

[0012] Furthermore, the step S1 assigns high priority to flight control instructions and key sensor data, and assigns low priority to status monitoring data. Compared with the prior art, the present invention has the following beneficial effects: the ARINC664 network fault tolerance and delay optimization method provided by the present invention, combined with the integrability theory of mathematical systems, introduces mathematical tools such as Darboux polynomials and first integrals, which can not only provide accurate theoretical support for the flow control of virtual links, but also optimize the dynamic load distribution of redundant paths through inverse Jacobi multipliers. This method makes up for the shortcomings of traditional optimization methods, can effectively reduce network jitter, shorten transmission delays, and enhance the fault tolerance of the system under high load and complex conditions, and provide more reliable technical guarantees for modern aviation communication systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flow chart of ARINC664 network fault tolerance and delay optimization of the present invention; Figure 2 A schematic diagram of flow optimization of a virtual link of the present invention; Figure 3 A schematic diagram of jitter generated by data packets transmitted through virtual links in the AFDX network of the present invention; Figure 4 It is a schematic diagram of fault tolerance optimization of redundant paths of the present invention. DETAILED DESCRIPTION

[0014] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0015] Figure 1 This is a flow chart of the ARINC664 network fault tolerance and delay optimization of the present invention.

[0016] See also Figure 1 The present invention aims to solve the delay and fault tolerance optimization problems in aviation communication networks and provides theoretical support through the construction of mathematical models. In view of the network congestion, jitter and delay problems that occur in the current AFDX network under high load, the model is established in combination with the following practical problems: a. Network behavior simulation under high load conditions: The model describes the traffic characteristics of the virtual link (VL) and the congestion phenomenon under high load conditions through the dynamic characteristics of the three-dimensional nonlinear system, and provides optimization strategies.

[0017] b. Realistic aviation network requirements: Based on the specific performance bottlenecks of the AFDX network of a certain type of aircraft (for example, the problem of excessive jitter in flight control data transmission of a certain type of aircraft), the model provides an optimization method for dynamic load distribution.

[0018] c. Application of Darboux polynomials and first integrals: Darboux polynomial theory is used to analyze the bandwidth allocation problem of virtual links, while first integrals are used to describe system state changes and optimize transmission paths to reduce delays and improve fault tolerance.

[0019] d. Dynamic adjustment of key parameters: Through the model constructed by inverse Jacobi multipliers, the virtual link traffic weight is adjusted in real time to ensure the stability and certainty of data flow transmission.

[0020] Specifically, the present invention is modeled based on a three-dimensional polynomial differential system, and the specific form is as follows: in, : Indicates the data flow transmission rate or bandwidth occupancy of the virtual link (VL), which is used to characterize the dynamic behavior of traffic.

[0021] : Indicates the transmission delay between network nodes, reflecting the average time it takes for a data packet to travel from the source node to the destination node.

[0022] : Indicates the system's fault tolerance or the degree of dynamic load distribution of redundant paths, and is used to measure the stability of the system in the event of node failure.

[0023] : Indicates the gain or attenuation factor of the data flow rate, reflecting the speed of traffic growth or suppression under high load conditions.

[0024] : Describes the damping or amplification factor associated with transmission delay, reflecting the sensitivity of propagation time between network nodes.

[0025] and : Measures the sensitivity of redundant path load distribution and determines the priority and dynamic adjustment capability of critical path data flow.

[0026] The algorithm is designed based on the first integration theory and Hilbert's finite basis theorem, and the software Mathematica and Singular are used for actual operation and verification. Finally, the Darboux polynomial of the model obtained by the present invention is: The corresponding cofactor is: The mathematical model constructed using the Darboux polynomial can effectively analyze the bandwidth control and fault-tolerant design of virtual links in AFDX networks, and provide mathematical support for the deterministic transmission of data streams.

[0027] The present invention can determine a first integral of the model through the above Darboux polynomial: The square integral of the model can be further determined: This allows the state changes of the system to be described.

[0028] At the same time construct the inverse Jacobi multiplier: Used to achieve the following optimization effects: a. Real-time dynamic adjustment: The inverse Jacobi multiplier captures changes in system status and combines key parameters such as traffic weight with system load to dynamically optimize the allocation of virtual link bandwidth.

[0029] b. Enhanced fault tolerance and stability: When the traffic of a virtual link exceeds the threshold, the inverse Jacobi multiplier automatically adjusts the weight to ensure the reliability and real-time performance of the network when the load changes.

[0030] c. Mathematical logic support: Inverse Jacobi multipliers satisfy in, The specific form is: It represents a vector field Function The directional derivative of exist The rate of change under the action of. Where P, , are functions, respectively exist , , Directional weight.

[0031] The specific form is: Represents a vector field The divergence describes the "degree of divergence" of the vector field at a certain point.

[0032] Therefore, the inverse Jacobi multiplier transforms the complex dynamic optimization problem into a constrained control problem, and the system performance can be optimized by weight adjustment.

[0033] The following is a specific embodiment of the virtual link (VL) traffic optimization of the present invention in combination with the application in the AFDX network. Figure 2 As shown in the figure; In the AFDX network, data packets are transmitted through virtual links (VLs), and the transmission path must meet strict control requirements for jitter and delay, such as Figure 3 As shown in Figure 1, the data flow of each virtual link (VL) must be transmitted within the specified bandwidth allocation interval (BAG), and the jitter must be controlled within the maximum allowable range to ensure that the timing characteristics of the data packet are maintained. For example, when the jitter is 0, the data packet is sent at a fixed interval (such as F1); when the jitter is between 0 and the maximum allowable jitter, the data packet is transmitted within this time window (such as F2). If the jitter reaches the maximum value, it means that the transmission of the data packet has been delayed to the end of the window (such as F3). The invention can provide dynamic constraints for the flow control of VL by using the cofactor constructed by Darboux polynomial, so as to realize deterministic transmission of the system under high load conditions and optimize network performance.

[0034] For example, the jitter formula in the AFDX network is: in Indicates the frame size (octet), which is adjusted according to different traffic requirements, preferably 64 to 1500 bytes; is the network bandwidth (bps), 100 Mbps ~ 1 Gbps or higher; and The minimum frame interval is preferably 20~00 microseconds. Control objectives: jitter ≤ 1 ms, maximum delay ≤ 10 ms, bandwidth utilization ≥ 80%.

[0035] The present invention uses the Darboux polynomial constraint system flow control formula: Jitter and transmission delay can be adjusted in real time to improve system performance.

[0036] In order to ensure the fault tolerance of the system, the AFDX network usually sets redundant paths. The present invention optimizes the flow control on the redundant paths by inverse Jacobi multipliers to ensure that data transmission is not affected when a node fails, as shown in FIG4 .

[0037] For example, using the inverse Jacobi multiplier to optimize the load distribution of redundant traffic: The load can be reasonably distributed among different VL paths, reducing network load and ensuring the reliability of data transmission on critical paths.

[0038] The transmission delay in the AFDX network of the present invention is composed of the delays of the sending end, network transmission and receiving end. The total delay can be expressed as: ; in , and They are send, transmit and receive delays respectively.

[0039] By first scoring Control jitter and reasonably allocate the traffic weight of VL to ensure that jitter is within an acceptable range. The construction form is: ; The transmission priority can be dynamically adjusted according to data traffic demand to ensure that jitter is minimized.

[0040] 1) (coupling coefficient between x and y), the specific value needs to be adjusted according to load conditions and network requirements This parameter determines the coupling relationship between x and y, that is, x changes with the change of y. It directly affects the stability of the system, especially the control of jitter.

[0041] Adjustment range: According to the stability requirements of the system, It should be kept in a moderate range. Too large a value may cause the system to be overly sensitive and unstable, while too small a value may cause the system to be slow to respond to flow changes. The preferred range is: 0.1≤ ≤ 1.0.

[0042] 2) (quadratic coefficient of y), the specific value needs to be adjusted dynamically according to different load conditions This parameter affects the self-influence of the y variable, i.e., affects the nonlinear variation of the delay. The larger it is, the more significant the impact of changes in y itself on the system will be, thus affecting the stability and jitter of the traffic.

[0043] Adjustment range: In order to ensure that the system jitter is not too large, It should not be too large to prevent the system from being overly sensitive and causing uncontrollable fluctuations; the preferred range is: 0.1 ≤ ≤ 0.5.

[0044] In the actual implementation process, the present invention loads the parameters of the mathematical model through the FDX interface module to dynamically optimize the control strategy of the AFDX network. The specific implementation steps are as follows: 1. Dynamic flow regulation of FDX control interface: Call the FdxCmdTxVLControl interface to implement dynamic control of virtual link (VL) traffic.

[0045] Working principle: This interface dynamically adjusts the sending rate of VL according to the current network load, and ensures stable traffic without overload by inputting model parameters such as bandwidth occupancy and priority weight.

[0046] Practical application: When the load is high, the rate of low-priority data streams is reduced to free up bandwidth to ensure real-time transmission of mission-critical data streams.

[0047] 2. Packet priority management mechanism: Use the FdxCmdTxQueueControl interface to adjust the queuing order of data packets according to the priority parameters provided by the model.

[0048] Priority Assignment: High-priority tasks: such as flight control instructions and critical sensor data, are transmitted first.

[0049] Low priority tasks: such as status monitoring data, are allowed to be delayed to reduce bandwidth usage.

[0050] Operational effect: By dynamically assigning priorities, it reduces delays and prevents network congestion, especially under conditions of tight system resources.

[0051] 3. Dynamic parameter update support: The FDX module supports updating model parameters such as transmission rate cap, packet size limit, queue weight, etc. through an external configuration interface.

[0052] Implementation method: The module periodically collects network operation status, combines the optimal control strategy calculated by the mathematical model, and updates relevant parameters.

[0053] The model constructed by the inverse Jacobi multiplier of the embodiment of the present invention provides a strong technical support for the fault tolerance of the AFDX network, which is specifically implemented as follows: 1. Real-time activity status acquisition: Call the FdxCmdRxVLGetActivity interface to obtain the real-time activity status of each virtual link (VL), including bandwidth utilization, packet loss rate, and node status.

[0054] Function: Monitor network operation in real time and detect potential fault signs, such as abnormal data flow or path congestion.

[0055] 2. Dynamic load distribution strategy: Combined with the mathematical model constructed by the inverse Jacobi multiplier, the traffic distribution on the redundant path is dynamically adjusted. The adjustment method is as follows: Normal nodes: Distribute traffic reasonably according to the current load balancing strategy.

[0056] Failed Node: When a node failure is detected, traffic is redistributed to available redundant paths.

[0057] Mathematical support: The inverse Jacobi multiplier ensures that the load distribution still meets the global optimal conditions when the node status changes, avoiding data loss caused by single point failure.

[0058] The present invention uses an inverse Jacobi multiplier model to dynamically optimize the load distribution strategy of redundant paths and calculate the load weight of key paths.

[0059] Optimization goal: maximize the real-time transmission of critical data streams and minimize the additional burden on redundant paths.

[0060] Actual effect: Significantly reduce latency and jitter under high load conditions, and quickly restore data transmission in the event of a node failure.

[0061] In summary, the ARINC664 network fault tolerance and delay optimization method provided by the present invention has the following advantages and effects: 1. Improve system real-time performance: The introduction of Darboux polynomial and first integral effectively reduces the data transmission jitter of the AFDX network and improves the real-time performance of data transmission.

[0062] 2. Enhance fault tolerance: Inverse Jacobi multipliers optimize redundant paths so that the system can still transmit data reliably in the event of a failure.

[0063] 3. Simplify system design: Provide engineers with simple flow control and optimization tools, reducing complex parameter adjustments during the design process.

[0064] Although the present invention has been disclosed as above with preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art may make some modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be based on the definition of the claims.

Claims

1. A method for optimizing fault tolerance and delay in an ARINC664 network, characterized in that: The steps include: S1) Use virtual links to manage data traffic in ARINC664 networks and achieve network fault tolerance through redundant paths; S2) A three-dimensional polynomial differential system is used as the basis for modeling to describe the traffic characteristics of virtual links in ARINC664 networks and the congestion phenomenon under high load conditions; S3) Control of flow and data transmission jitter in ARINC664 networks using Darboux polynomials and first integrals; S4) By constructing the inverse Jacobi multiplier, the virtual link traffic weight is adjusted in real time to optimize the traffic control on the redundant path.

2. The ARINC664 network fault tolerance and delay optimization method according to claim 1, characterized in that: The step S2 is modeled as follows: in, Indicates the data flow transmission rate or bandwidth occupancy of the virtual link; Represents the transmission delay between network nodes; Indicates the system's fault tolerance or the degree of dynamic load distribution of redundant paths; A gain or attenuation factor that represents the data flow rate; Describe the damping or amplification factors associated with transmission delays; and Measures the sensitivity of load distribution across redundant paths.

3. The ARINC664 network fault tolerance and delay optimization method according to claim 1, characterized in that: The step S3 designs an algorithm based on the first integration theory and Hilbert's finite basis theorem, and performs actual calculations and verifications in combination with software Mathematica and Singular. The Darboux polynomial finally obtained is: The corresponding cofactor is: ； The first integral determined is: ; 0.1 ≤ ≤ 1.0,0.1 ≤ ≤ 0.

5.

4. The ARINC664 network fault tolerance and delay optimization method according to claim 3, characterized in that: The transmission delay in the ARINC664 network is composed of the delays of the sender, network transmission and receiver, and the total delay is: ; in To delay sending, is the transmission delay, for receiving delays; The step S3 is performed by first integrating Control jitter, allocate traffic weights to VLs, control jitter ≤ 1 ms, maximum delay ≤ 10 ms, and bandwidth utilization ≥ 80%.

5. The ARINC664 network fault tolerance and delay optimization method according to claim 4, characterized in that: The jitter formula in the ARINC664 network is: in Indicates the frame size, bytes range from 64 to 1500; is the network bandwidth, ranging from 100 Mbps to 1 Gbps; It is the minimum frame interval, with a duration of 20~00 microseconds.

6. The ARINC664 network fault tolerance and delay optimization method according to claim 3, characterized in that: The inverse Jacobi multiplier constructed in step S3 is: ; The step S3 combines the traffic weight with the system load, and when the traffic of a certain virtual link exceeds a threshold, the inverse Jacobi multiplier is used to adjust the traffic weight.

7. The ARINC664 network fault tolerance and delay optimization method according to claim 6, characterized in that: The step S1 obtains the real-time activity status of each virtual link, including bandwidth utilization, packet loss rate and node status, and dynamically assigns different priorities to packets; the step S3 allocates traffic to normal nodes according to the current load balancing strategy, and reallocates traffic to available redundant paths when a node failure is detected.

8. The ARINC664 network fault tolerance and delay optimization method according to claim 7, characterized in that: The step S1 assigns a high priority to the flight control instructions and key sensor data, and a low priority to the status monitoring data.

Images