A redundant data elimination method for a spaceborne time-triggered network
By using a Bloom filter to detect and remove redundant data in a spaceborne time-triggered network, the problem of data transmission conflicts in traditional networks is solved, achieving efficient data redundancy elimination and improving the real-time performance and reliability of network device scheduling.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI AEROSPACE COMP TECH INST
- Filing Date
- 2023-03-20
- Publication Date
- 2026-05-01
AI Technical Summary
Traditional spaceborne Ethernet networks suffer from data transmission conflicts, which affect the reliability and real-time performance of data transmission. Redundant data needs to be eliminated in multi-channel data transmission to improve the real-time performance of network device scheduling.
A Bloom filter is used to detect data redundancy entering the receiving system from different networks, and redundant data packets are quickly eliminated through hardware calculation. Multi-bit arrays and random hash functions are used to determine the redundancy of data frames, ensuring that only data frames with different content are retained.
It effectively eliminates redundant dual-mode or multi-mode data, improves the real-time performance and reliability of the onboard time-triggered network, and reduces transmission latency.
Smart Images

Figure CN116261164B_ABST
Abstract
Description
A method for eliminating redundant data in spaceborne time-triggered networks Technical Field
[0001] This invention belongs to the technical field of satellite communication, and particularly relates to a method for eliminating redundant data in a spaceborne time-triggered network. Background Technology
[0002] With the rapid development of my country's aerospace industry, the demand for data interaction between satellite-borne units is constantly increasing, which puts forward higher requirements for on-board information exchange capabilities. Multi-connection and interoperability data interaction has become a development trend.
[0003] Against this backdrop, building a distributed Ethernet network based on multiple onboard platforms is an effective way to solve these problems. However, traditional Ethernet networks suffer from data transmission conflicts, which seriously affect the reliability and real-time performance of onboard data transmission.
[0004] Time-triggered Ethernet (TTE) is a new type of real-time industrial control Ethernet with high real-time performance, high resource utilization, and high fault tolerance, and it has been widely used in the industrial field. As a distributed network, TTE boasts communication rates exceeding 1Gbps and incorporates time attributes and various fault-tolerant designs to ensure reliable transmission of time-critical messages, exhibiting numerous superior characteristics such as high determinism and high real-time performance.
[0005] This invention proposes a method for eliminating redundancy in spaceborne time-triggered networks. To ensure the reliability of the spaceborne network, multi-channel data redundancy needs to be introduced, while redundant data must be eliminated from the output of the end system. Summary of the Invention
[0006] The purpose of this invention is to provide a method for eliminating redundant data in a spaceborne time-triggered network, which can effectively eliminate dual-mode or multi-mode redundancy and improve the real-time performance of network device scheduling through rapid hardware calculation.
[0007] To solve the above problems, the technical solution of the present invention is as follows:
[0008] A method for eliminating redundant data in a spaceborne time-triggered network includes:
[0009] Based on the redundant backup structure of the spaceborne time-triggered network, the source system transmits service data frames to the receiving system simultaneously on multiple networks. The receiving system then performs redundant data elimination on the received data frames before outputting them.
[0010] The process of eliminating redundant data in the received data frames includes:
[0011] The Bloom filter is used to detect data redundancy entering the receiving system from different networks and to remove redundant data packets.
[0012] According to an embodiment of the present invention, the step of detecting data redundancy entering the receiving system from different networks using a Bloom filter further includes:
[0013] A Bloom filter consists of an m-bit array and k different random hash functions, where m and k are set according to the packet size and the collision probability of the hash functions.
[0014] Initialize an m-bit array;
[0015] The received data frame is input into the Bloom filter, the output value of the data frame after being processed by k random hash functions is calculated, and 1 is written into the corresponding position of the bit array according to the mapping relationship between the random hash function and the bit array.
[0016] For a data frame, if there is a non-1 value in the mapping bits of its k random hash functions, then the data frame has no redundancy; otherwise, the data frame has redundancy and needs to be discarded.
[0017] According to one embodiment of the present invention, after the Bloom filter completes one data frame detection, it sets the m-bit bit array to 0.
[0018] Because the present invention adopts the above technical solution, it has the following advantages and positive effects compared with the prior art:
[0019] In one embodiment of the present invention, a method for eliminating redundant data in a spaceborne time-triggered network sets every bit of the multi-bit array in the Bloom filter to 0 under initial conditions to mark hash function values. For dual-path redundant input in the end system, the Bloom filter first calculates the hash values of one data packet and sets the corresponding bits of the multi-bit array to 1. Then, it calculates the hash values for the next data packet in the same way. If a corresponding bit in the multi-bit array is 1, it indicates that the two data packets have the same hash value and are likely identical in content; therefore, the later data packet is discarded. If all hash value bits in the multi-bit array corresponding to the second data packet are 0, it indicates that the two data packets have different hash values and must have different content; therefore, both data packets are retained and sent to subsequent modules in the end system. This method effectively eliminates dual-mode or multi-mode redundancy and improves the real-time performance of network device scheduling through rapid hardware calculation. Attached Figure Description
[0020] Figure 1 is a schematic diagram of the redundant structure of the spaceborne time-triggered network in one embodiment of the present invention;
[0021] Figure 2 is a schematic diagram of the principle of a Bloom filter in one embodiment of the present invention. Detailed Implementation
[0022] The following detailed description, in conjunction with the accompanying drawings and specific embodiments, provides a method for eliminating redundant data in a spaceborne time-triggered network according to the present invention. The advantages and features of the present invention will become clearer from the following description and claims.
[0023] Referring to Figure 1, the spaceborne time-triggered network employs two redundant networks for data transmission, improving network communication reliability and reducing transmission latency. A redundant backup network structure is formed through switches; cascading switches allows for network expansion, achieving a multi-hop (two to three hops) network architecture. The source and destination systems are connected through two or more independent networks, with redundancy mechanisms further enhancing network communication reliability. Both networks (Network A and Network B) operate simultaneously. The source system transmits service data frames on both networks concurrently, and the receiving system receives data frames from both networks, then uses a redundancy management mechanism to eliminate redundancy in the received data frames.
[0024] To eliminate redundant data in received data frames, this embodiment uses a Bloom filter to detect data redundancy entering the receiving system from different networks and removes redundant data packets. This Bloom filter consists of a multi-bit array and multiple hash functions. The hash functions map data of arbitrary length to a fixed range. Since the range of input values is larger than the range of output values, different input data may result in the same output data.
[0025] Referring to Figure 2, the multi-bit array in the Bloom filter is initially set to 0 for each bit to mark the hash function value. For the dual-path redundant input in the end system, the Bloom filter first calculates the hash values of one data packet and sets the corresponding bits in the multi-bit array to 1. Then, it calculates the hash values for the next data packet in the same way. If a corresponding bit in the multi-bit array is 1, it means the two data packets have the same hash value and are likely identical in content; the later data packet is discarded. If all hash value bits in the multi-bit array corresponding to the second data packet are 0, it means the two data packets have different hash values and must have different content; both data packets are retained and sent to subsequent modules in the end system. After the query is completed, a reset operation is performed on the multi-bit array, setting all bits to 0. If a third path redundancy exists, the same query operation is performed on the third data packet followed by a reset operation.
[0026] In practical applications, a Bloom filter consists of an m-bit array (each bit initially set to 0) and k different random hash functions, where m and k can be set according to the data packet size and the collision probability of the hash functions. The characteristic of a random hash function is that regardless of the length of the input value, the length of the output value is fixed; it is always a number between (0, m-1) and uniformly randomly distributed within that range. The output of the hash function corresponds to the index of the bit array in the filter. Because the input length of the hash function is unlimited, while the output is distributed within a certain range, different inputs can lead to the same hash function output, a situation known as a hash collision.
[0027] For a data packet entering the switch, calculate its output value after being processed by k random hash functions, and store 1 in the corresponding position of the filter array. For a given data packet, if there is a non-1 value in the mapping bits of the k random hash functions, then the data packet has no redundancy; otherwise, if all mapping bits are 1, then the data packet has redundancy and can be discarded.
[0028] Because hash collisions can occur, the Bloom filter has a certain probability of misclassifying different data packets as the same data packet. To avoid an increased probability of hash collisions due to too many data packets, the array in the Bloom filter is reset after each data frame detection, setting all bits to 0.
[0029] During the functional testing of the aforementioned redundant data elimination method for the spaceborne time-triggered network, with normal network connectivity and synchronization, two redundant messages were sent from different switches at the same time slot. Signal capture at the destination port's phosphor showed only one frame, indicating that the redundant TT message had been removed, and the redundant TT message packet removal function on the end system was normal. Disconnecting the network cable from one of the end nodes and capturing packets again at the test port showed that the data still existed, demonstrating that a failure in one channel during dual-channel redundancy would not affect normal data communication. When the two channel data are different, for example, if one data packet is incomplete, the Bloom filter treats the two channel data as different data packets and retains both.
[0030] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, if these changes fall within the scope of the claims of the present invention and their equivalents, they shall still fall within the protection scope of the present invention.
Claims
1. A method for eliminating redundant data in a spaceborne time-triggered network, characterized in that, include: Based on the redundant backup structure of the spaceborne time-triggered network, the source system simultaneously transmits service data frames to the receiving system on multiple networks. The receiving system performs redundant data elimination on the received data frames before outputting them. Redundancy elimination of the received data frames includes: detecting data redundancy entering the receiving system from different networks using a Bloom filter and removing redundant data packets. The detection of data redundancy entering the receiving system from different networks using a Bloom filter further includes: the Bloom filter contains an m-bit array and k different random hash functions, where m and k are set according to the data packet size and the collision probability of the hash functions; initializing the m-bit array; inputting the received data frame into the Bloom filter, calculating the output value of the data frame after being processed by the k random hash functions, and writing 1 at the corresponding position in the bit array according to the mapping relationship between the random hash functions and the bit array; for a data frame, if there is a non-1 value in the mapping bits of its k random hash functions, then the data frame has no redundancy; otherwise, the data frame has redundancy and needs to be removed; after the Bloom filter completes one data frame detection, it sets the m-bit array to 0.
Citation Information
Patent Citations
Redundancy elimination method and system for edge computing task unloading
CN116208610A