Memory Bank Data Training via Noise Profile Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data training methods in memory devices do not account for core noise, leading to inefficiencies as operating speeds increase, particularly above 5 Gbps, resulting in significant variations in data window variations and potential errors in data transmission.
Innovation Solution
A method that involves enabling memory banks in different states, performing data training operations for each state, generating a noise profile, statistically analyzing it to select a reference enabling state, and using this state for optimized data training to compensate for skew and core noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If conventional data training method is used without considering core noise, then data training can be performed when banks are disabled, but data window variation increases significantly at high operating speeds
Solution Approach 1:
The patent performs preliminary data training operations when memory banks are in disabled state to generate a noise profile that characterizes core noise characteristics. This preliminary action allows the system to capture noise characteristics before actual data transmission begins, enabling subsequent compensation during active operation. The noise profile is generated in advance and stored for later use during data transmission.
2Measurement precision
If data training is performed for all enabling states of memory banks, then noise profile accuracy improves, but training time and complexity increase
Solution Approach 1:
The patent performs data training operations for multiple enabling states of memory banks (more than the minimum required), generating a comprehensive noise profile that covers various operational conditions. By performing training for all possible enabling states, the system obtains a more accurate and complete noise characterization, which can then be used to select the optimal compensation parameters for any given operational state.
3Device complexity
If skew compensation is not applied, then data training is simpler, but phase difference between clock and data signals causes errors
Solution Approach 1:
The patent implements a feedback mechanism where the noise profile generated from preliminary training operations is used to determine optimal skew compensation parameters. The system continuously monitors noise characteristics and adjusts compensation settings accordingly. This feedback loop ensures that skew compensation is dynamically optimized based on actual noise conditions, improving data transmission reliability while maintaining manageable complexity through automated parameter selection.
Data Source
AI summary
In one embodiment, a method of performing data training in a system including a memory controller and at least a first memory device including a group of memory banks is disclosed. The method includes providing a plurality of enabling states for the group of memory banks, wherein each enabling state is different and for each enabling state a set of the memory banks of the group is enabled and any remaining of the memory banks of the group are not enabled. The method further includes performing a first data training procedure that includes a series of first data training operations for the first memory device, each data training operation being performed for a different one of the plurality of enabling states, generating a noise profile based on the series of first data training operations, statistically analyzing the noise profile to select a reference enabling state of the group of memory banks, and performing a second data training procedure for the first memory device using the reference enabling state. As a result, the operating speed and reliability of the system including the memory device may be improved.