Message Packet Synchronization in TCP Data Streams
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Monitoring communication between two control units using the TCP transport protocol is challenging due to the lack of marked start and end of message packets, requiring synchronization that can take a long time, especially when 'Magic Cookies' are sent infrequently.
Innovation Solution
A method that mirrors the data stream, searches for known identifiers, interprets the following data item as a payload length specification, and checks for additional identifiers to reliably synchronize with the data stream without waiting for synchronization markers, allowing for real-time monitoring of message packets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If Magic Cookies are used to mark message packet boundaries in TCP data streams, then synchronization can be achieved, but the wait time for synchronization becomes very long due to infrequent sending
Solution Approach 1:
The patent extracts the synchronization problem from the Magic Cookie mechanism and creates a separate, dedicated synchronization mechanism using known identifiers. Instead of relying on infrequent Magic Cookies sent by control units, the system proactively injects synchronization markers into the data stream, separating the synchronization function from the normal message transmission protocol.
Solution Approach 2:
The system performs preliminary synchronization by continuously injecting known identifiers at regular intervals into the TCP data stream before actual message packet transmission begins. This preliminary action ensures that the monitoring system is already synchronized and ready to capture message packets immediately, eliminating the need to wait for Magic Cookies.
2Productivity
If TCP transport protocol is used for communication between control units, then continuous data stream transmission is achieved, but message packet boundaries are not marked requiring complex synchronization
Solution Approach 1:
The patent introduces an intermediary synchronization mechanism that operates between the TCP data stream and the message packet structure. Known identifiers act as mediators that bridge the gap between the continuous TCP stream and the discrete message packets, providing clear boundaries without requiring changes to the TCP protocol itself or the existing message format.
Solution Approach 2:
The system creates a copy of known identifiers and inserts them into the TCP data stream at regular intervals. These copied identifiers serve as synchronization markers that the monitoring system can recognize and use to establish message packet boundaries, simplifying the synchronization process without affecting the original message transmission.
3Speed
If monitoring is implemented without synchronization markers, then real-time monitoring is possible, but reliable message packet identification cannot be achieved
Solution Approach 1:
The system performs preliminary synchronization by continuously injecting known identifiers into the data stream before monitoring begins. This ensures that when real-time monitoring starts, the system is already synchronized with the message packet boundaries, allowing immediate and reliable identification without sacrificing monitoring speed.
Solution Approach 2:
Known identifiers are injected into the TCP data stream at periodic intervals, creating regular synchronization points. This periodic action ensures that the monitoring system can reliably identify message packet boundaries at predictable intervals, maintaining both real-time monitoring capability and identification reliability.
Data Source
AI summary
A method for monitoring message packets that are exchanged between at least two control units. The message packets are concatenated in a data stream and each have an identifier, a payload, and a length specification of the payload described by a data item of predefined word size. The at least two control units are connected by a distributor. The distributor is connected by a first distributor port to a first of the at least two control units, is connected by a second distributor port to a second of the at least two control units, and is connected by a third distributor port to a computer system. The data stream flows through the first and distributor port for communication between the first node and the second node. The computer system has a memory, and information on the respective identifiers of the message packets is stored in the memory.