Bus Clock Calibration Using Data Frame Edge Spacing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication systems with subscriber nodes using imprecise clock generators face challenges in calibration, leading to bandwidth inefficiency and synchronization issues, particularly in systems where special calibration messages are required, causing delays and potential failure in establishing communication.
Innovation Solution
The number of system clocks in a bit time is freely configurable, allowing calibration using any message on the data bus, with a clock divider ratio that can be fractionally rational, enabling calibration with any edges in data frames, and a restricted operating mode to prevent interference during calibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If special calibration messages are transmitted on the data bus, then clock generators can be calibrated to the system clock, but bandwidth is consumed and cannot be used for user data transmission
Solution Approach 1:
The invention makes regular data frames serve dual purposes: transmitting user data and providing calibration information for clock synchronization. The calibration information is embedded within the existing data frame structure, allowing the same communication resource to fulfill both data transmission and clock calibration functions simultaneously, thereby eliminating the need for separate calibration messages.
Solution Approach 2:
The invention merges the calibration function with regular data transmission by embedding calibration information within standard data frames. This combination allows the calibration process to occur alongside normal communication operations, maximizing bandwidth utilization while maintaining clock synchronization precision.
2Measurement precision
If special calibration messages are used for calibration, then clock synchronization can be achieved, but communication setup time increases and may fail in single-master systems
Solution Approach 1:
The invention prepares calibration information to be embedded within regular data frames that are already part of the communication protocol. By integrating calibration data into the normal data flow from the outset, the system eliminates the need for separate calibration phases, allowing synchronization to occur naturally as part of ongoing communication operations.
Solution Approach 2:
The invention enables continuous calibration by embedding calibration information in every regular data frame transmission. This continuous availability of calibration data allows nodes to continuously refine their clock synchronization without interrupting normal communication operations, thereby eliminating setup delays and enabling immediate communication in single-master systems.
3Ease of manufacture
If imprecise clock generators are used in subscriber nodes, then system cost is reduced, but calibration complexity increases
Solution Approach 1:
The invention enables subscriber nodes to automatically extract and process calibration information from regular data frames without requiring external calibration equipment or complex calibration procedures. The nodes self-calibrate by utilizing the embedded calibration data within the existing communication protocol, thereby maintaining low system cost while managing calibration complexity through automated processes.
Solution Approach 2:
The invention makes the existing data frame structure serve the additional function of carrying calibration information. By embedding calibration data within standard frames that are already part of the communication protocol, the system avoids adding separate calibration hardware or protocols, thereby maintaining simplicity while enabling calibration of imprecise clock generators.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
The present invention relates to a communication system (1) comprising a data bus (2) and several subscriber nodes (3, 4) connected to the data bus (2), and to a method for operating such a communication system (1). Data frames (10) for data transmission each comprise a data field (14) and/or a control field (13) and a test sum field (15). A subscriber (3) has a high-precision clock generator (5) (such as precise quartz, MEMS resonator, and the like), and the remaining subscribers (4) have a clock generator (7) (such as an RC oscillator or imprecise quartz) having lower precision. During the operation of the communication system (1), calibration messages are present at the data bus (2), and the subscribers (4) having the imprecise clock generator (7) receive a calibration message. The subscribers (4) having the imprecise clock generator (7) calibrate the imprecise clock generator (7) for a system clock pulse (6) of the data bus (2), while taking information contained in the received calibration message into consideration. In order to expedite the calibration of such a communication system (1), and simultaneously save bandwidth on the data bus (2), it is proposed to use arbitrary messages transmitted via the data bus (2) as calibration messages for calibrating the system (1). For the calibration, a spacing (NOPN) between an edge (40) in the data field (14) or in the control field (13) of a received data frame (10) and another edge (41) in a field (13; 14; 15) of the received data frame (10) is measured in oscillator periods (8). The number (NB) of bits between said two edges (40, 41) is counted, and a clock pulse divider (9), which generates a system clock pulse (6) of a data bus protocol control system of the subscriber (4) to be calibrated, is to be set such that a system clock pulse period is NOPN/NB NTQ oscillator periods (8) long.