Memory Receiver Training for Per-Symbol Voltage and Timing Tuning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face challenges in efficiently communicating data between host devices and memory devices due to variations in signal characteristics across different unit intervals and symbols, leading to errors in transmission.
Innovation Solution
The implementation of a training procedure that tunes receivers by adjusting reference voltages and sampling times on an individual basis, using baseline training operations and individual training operations to compensate for variations in signal characteristics, allowing for independent configuration of reference voltages and sampling times across different unit intervals and symbols.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If individual training operations are performed for each reference voltage and sampling time, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The training procedure is segmented into multiple distinct phases: baseline training operations that establish initial reference voltages and sampling times, followed by individual training operations that fine-tune each parameter independently. This segmentation allows the system to achieve high precision through iterative refinement without overwhelming complexity in a single step.
Solution Approach 2:
Baseline training operations are performed as a preliminary action before individual training operations. This preliminary step establishes initial reference voltages and sampling times that provide a foundation for subsequent fine-tuning, reducing the complexity of individual operations by starting from a pre-configured state rather than from scratch.
2Reliability
If reference voltages and sampling times are tuned individually for each unit interval, then reliability is improved, but loss of time increases
Solution Approach 1:
The training procedure employs periodic action through iterative loops that alternate between baseline training operations and individual training operations. Each iteration refines the receiver settings further, with the period decreasing as the system converges to optimal values. This periodic refinement achieves high reliability through multiple passes while managing time loss by reducing adjustment magnitudes in later iterations.
Solution Approach 2:
The system implements feedback mechanisms where each training operation uses results from previous operations to guide subsequent adjustments. Error rates and signal quality metrics from baseline training inform the individual training operations, which in turn provide feedback for further refinement. This feedback loop ensures reliability improvements while minimizing time loss by avoiding redundant adjustments.
Data Source
AI summary
Systems, apparatuses, and methods for training procedures on reference voltages and sampling times associated with symbols communicated with a memory device are described. The training procedures may be configured to compensate for variations that may occur in different symbols of a signal. For example, an individual training operation may be performed for each reference voltage within a first unit interval. These individual training operations may allow a reference voltage of the first unit interval to be positionable independent of other reference voltages in the same unit interval or in different unit intervals. In another example, an individual training operation may be performed for the sampling time associated with a reference voltage. These individual training operations may allow a sampling time associated with a reference voltage in the first unit interval to be positionable independent of other sampling times in the same unit interval or in different unit intervals.