Head-Node Data Bus Delay Measurement Without Node Collisions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for time synchronization in vehicle data bus systems, particularly using IEEE 802.1AS with 10Mbit/s Ethernet in MultiDrop mode, face issues due to collisions and resets of state machines among bus nodes, leading to unreliable synchronization and potential network disruptions.
Innovation Solution
A method that individually measures propagation delays between a head node and each bus node by putting all other nodes into a sleep state during the measurement cycle, using PTP-based time synchronization adapted for MultiDrop mode, and employing sleep messages and wake-up signals to ensure only one active node at a time, with error handling and optimized wake-up timing to maintain protocol compliance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If PTP-based time synchronization is used in MultiDrop mode with 10Mbit/s Ethernet, then time synchronization can be achieved, but state machines on uninvolved bus nodes are reset causing measurement disruptions
Solution Approach 1:
The patent segments the bus node population into three groups: the active bus node being measured, the head node performing measurements, and inactive bus nodes put into sleep state. This segmentation prevents measurement messages from disrupting inactive nodes' state machines while maintaining the ability to measure all nodes sequentially.
Solution Approach 2:
The patent applies preliminary action by placing all bus nodes except the target node into a sleep state before initiating propagation delay measurements. This preliminary positioning of nodes in a controlled state prevents them from receiving and processing measurement messages that would otherwise reset their state machines.
2Measurement precision
If measurement messages are exchanged between head node and bus nodes, then propagation delay can be determined, but other bus nodes receive and react to messages causing collisions
Solution Approach 1:
The patent divides the bus node population into active and inactive segments during measurement cycles. Only the target bus node and head node exchange measurement messages while other nodes remain in sleep state, eliminating message collisions and interference that would occur in a fully active multi-drop configuration.
Solution Approach 2:
The patent implements periodic measurement cycles where each bus node is sequentially selected as the active node for propagation delay measurement. During each cycle, the target node is awakened, measured, and then returned to sleep state before the next node is measured, creating a structured periodic pattern that prevents interference.
3Speed
If all bus nodes remain in awake state for measurements, then quick response to messages is possible, but synchronization reliability decreases due to node interference
Solution Approach 1:
The patent prepares the measurement environment by placing non-target nodes in sleep state before measurements begin. This preliminary action ensures that when measurement messages are exchanged between the head node and target node, no other nodes will interfere, thus maintaining both fast response and high reliability.
Solution Approach 2:
The patent dynamically changes the operational state of bus nodes based on their role in the current measurement cycle. The target node transitions from sleep to awake state for measurement, while other nodes remain in sleep state. This dynamic state management allows the system to optimize between response speed and measurement reliability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention is concerned with a method for determining a respective propagation delay between a head node (20) on one side and a plurality of other bus nodes (21) of a data bus system (17) on the other side, the determination of the respective propagation delay comprising an exchange of predetermined measurement messages (30, 33, 34) and, wherein the method comprises that the head node (20) successively performs a respective measurement cycle (40), comprising that for each measurement cycle (40) another one of the bus nodes (21) is selected, and each remaining one of the bus nodes (21) is switched from an awake state (71), in which the bus node (21) responds to measurement messages (30, 33, 34), to a predetermined sleep state, in which the bus node (21) remains unresponsive to measurement messages (30, 33, 34), and the propagation delay measurement between the head node (20) and the selected bus node (21) is performed by exchanging the measurement messages (30, 33, 34), and each remaining bus node (21) is activated from the sleep state by means of at least one wake-up signal (50).