Multi-level Error Signal Reception in GDDR Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing DDR DRAM systems face challenges in efficiently receiving and processing multi-level error signals, particularly in graphics double data rate (GDDR) memory, where separate retraining of the WCK-to-ERR signal skew is time-consuming and disruptive to system operations.
Innovation Solution
A method and circuitry for robust multi-level signal reception in GDDR memory systems, involving oversampling, deserialization, and symbol alignment to determine the start of a symbol period and filter error symbols without the need for phase training, using a small amount of digital circuitry and tolerating wide signal skew variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If separate retraining of the WCK-to-ERR signal skew is performed, then signal reception accuracy is improved, but system operation is disrupted and time is lost
Solution Approach 1:
The patent performs preliminary alignment of the multi-level error signal by detecting the start of a symbol period and adjusting the sampling phase accordingly. This preliminary action ensures that subsequent signal reception does not require disruptive retraining, thus improving signal reception accuracy while minimizing time loss.
Solution Approach 2:
The system performs self-alignment by automatically detecting signal transitions and determining symbol period starts without external intervention. The error signal circuitry autonomously adjusts its sampling phase based on detected transitions, eliminating the need for manual or periodic retraining operations.
2Speed
If multi-level signaling is employed, then data transmission rate is improved, but link training complexity increases
Solution Approach 1:
The patent changes the parameter of signal levels from traditional two-level to multi-level signaling, thereby increasing data transmission rate. The oversampling and symbol alignment techniques are specifically designed to handle the increased complexity of multi-level signals, maintaining reliable reception despite the higher data rates.
Solution Approach 2:
The patent introduces a time dimension by oversampling the error signal at multiple phases. Instead of attempting to decode multi-level signals at a single sampling point, the system captures multiple samples across different time phases and uses these to determine symbol period starts and align sampling, effectively adding temporal dimensionality to simplify multi-level signal reception.
3Reliability
If oversampling and symbol alignment are implemented, then signal reception robustness is improved, but circuit complexity increases
Solution Approach 1:
The patent segments the error signal reception process into distinct functional blocks: an input circuit that performs oversampling at multiple phases, a filter and alignment circuit that detects symbol period starts, and a filtering stage that processes aligned samples. This segmentation allows each block to perform a specific function efficiently, improving overall robustness while keeping individual circuit complexities manageable.
Data Source
AI summary
A method for receiving a multi-level error signal having more than two logic levels includes oversampling the multi-level error signal to provide sampled symbols, wherein a first level of the multi-level error signal indicates no error, and second and third levels of the multi-level error signal indicate first and second error conditions, respectively. The sampled signals are de-serialized to provide sets of symbols. A start of a symbol period is determined in response to detecting that a given sample is different from a prior sample, and the prior sample indicates no error. The sets of symbols are filtered to provide corresponding output symbols based on the start.