Codeword Detection Using Complementary Memory Cells and Time-Domain Reads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory technologies face challenges in reliably differentiating between high resistive states (HRS) and low resistive states (LRS) due to small read windows, which are further complicated by aging and temperature effects, making it difficult to accurately determine code words in k-out-of-n codes.
Innovation Solution
The use of complementary memory cells, where each data bit is represented by two physical memory cells with complementary states, and a method to read and determine states in the time domain, allowing for early detection of 0-states and subsequent assignment of 1-states, thereby improving code word detection efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If states are determined in the time domain using complementary memory cells, then code word detection reliability is improved, but device complexity increases
Solution Approach 1:
The memory system is segmented into complementary memory cell pairs, where each pair represents a single data bit through two physical cells with complementary states. This segmentation enables more reliable state differentiation by comparing the relative timing of state transitions between paired cells, thereby improving code word detection reliability despite increased device complexity
Solution Approach 2:
The patent transitions from traditional amplitude-based or voltage-based state detection to time-domain detection by measuring the relative timing of state transitions. This dimensional change from voltage amplitude to time measurement provides a new basis for differentiation that is less susceptible to aging and temperature effects, improving reliability while requiring additional timing circuitry
2Measurement precision
If waiting for all 1-states to be determined before completing code word detection, then measurement precision is maintained, but loss of time increases
Solution Approach 1:
The system performs preliminary determination of 0-states first, using their faster transition characteristics to establish a reference timing. This preliminary action allows the system to begin processing and validation before all 1-states have completed their transitions, reducing overall detection time while maintaining precision through the time-domain comparison framework
Solution Approach 2:
The detection process is made dynamic by allowing early termination of the measurement sequence. Instead of a static wait-for-all-approach, the system dynamically determines when sufficient 0-states have been detected to confidently identify the code word, adapting the measurement duration to the actual data characteristics and reducing unnecessary waiting time
Data Source
AI summary
A method for detecting a code word is proposed, wherein the code word is a code word of one of at least two codes, wherein n states are read from memory cells of a memory, respectively. The n states are determined in a time domain for each of the at least two codes, wherein additionally n states are read from further memory cells and at least one reference value is determined therefrom and wherein the at least one reference value is taken as a basis for determining which of the at least two codes is the correct code. A corresponding device is furthermore specified.