Adaptive Data Bus Inversion for Parallel Channel Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data transmission methods over parallel channels face issues with inter-symbol interference (ISI), crosstalk, and simultaneous switching noise (SSN), and the minimum transitions algorithm fails to optimize signal quality and power consumption due to lack of correlation between data packets and neglecting binary state considerations.
Innovation Solution
Implementing multiple Data Bus Inversion (DBI) algorithms, such as minimum transitions, minimum zeros, and minimum ones, to encode data bits across parallel channels, using a separate DBI bit to identify and restore inverted data bits, thereby reducing ISI, crosstalk, and SSN, and improving power efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data bits are transmitted over parallel channels without encoding, then transmission speed is maintained, but inter-symbol interference (ISI), crosstalk, and simultaneous switching noise (SSN) degrade signal quality
Solution Approach 1:
The patent applies parameter changes by inverting data bits based on transition counting. The encoder monitors the number of transitions in incoming data bits and inverts the entire data bus when transitions exceed a threshold, thereby reducing simultaneous switching noise and inter-symbol interference while maintaining transmission speed
Solution Approach 2:
The core DBI technique inverts data bits to reduce noise and interference. By inverting the data bus when transition counts exceed a threshold, the patent transforms harmful high-activity patterns into lower-activity patterns, reducing crosstalk and SSN while preserving data integrity through the inversion indicator bit
2Loss of energy
If minimum transitions algorithm is used to reduce transitions, then power consumption decreases, but signal quality deteriorates due to lack of packet correlation and binary state considerations
Solution Approach 1:
The patent implements dynamic encoding by switching between different DBI algorithms based on packet boundaries and data characteristics. The encoder dynamically selects among minimum transitions, minimum zeros, and minimum ones algorithms to adapt to varying data patterns, thereby optimizing both power consumption and signal quality
Solution Approach 2:
The patent changes encoding parameters by selecting different algorithms based on binary state considerations. Instead of solely minimizing transitions, the encoder adjusts its strategy to account for packet boundaries and binary distributions, improving signal quality while maintaining power efficiency
3Reliability
If multiple DBI algorithms are implemented, then signal quality and power consumption are optimized, but device complexity increases
Solution Approach 1:
The patent segments the encoding process into distinct algorithms (minimum transitions, minimum zeros, minimum ones) that are selectively applied based on packet boundaries and data characteristics. This segmentation allows optimization of specific data patterns without requiring complex hybrid algorithms, balancing performance with implementability
Solution Approach 2:
The patent implements dynamic algorithm selection based on packet boundaries and data characteristics. The encoder transitions between different DBI algorithms depending on the data pattern, optimizing performance for each segment while maintaining overall system efficiency
Data Source
AI summary
Apparatus, systems, and methods are disclosed such as those that operate to encode data bits transmitted on a plurality of channels according to at least one of multiple Data Bus Inversion (DBI) algorithms. Additional apparatus, systems, and methods are disclosed.