Hamming Weight Data Encoding for Low-Power Interconnect Transfer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Energy consumption in computing devices is high due to data transfer across interconnects, particularly because transferring data as '1' values consumes power in terminated interconnects and signal level transitions consume power in unterminated interconnects.
Innovation Solution
Implementing minimum or maximum Hamming weight encoding to reduce the number of '1' or '0' values in data, depending on the memory interface, thereby minimizing energy consumption during data transfer by encoding data to have fewer '1' values or '0' values, which reduces power consumption based on the type of memory interface used.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If data is transferred using conventional encoding, then data transfer functionality is maintained, but energy consumption is high due to numerous '1' values and signal level transitions
Solution Approach 1:
The patent applies parameter changes by transforming the data representation from conventional binary encoding to Hamming weight-encoded data. This changes the fundamental parameter of data encoding to minimize either the number of '1' values (minimum Hamming weight) or '0' values (maximum Hamming weight), directly reducing energy consumption in terminated interconnects without requiring complex additional hardware beyond standard encoders and decoders
Solution Approach 2:
The patent implements preliminary action by encoding data before transfer using Hamming weight encoding. The encoder processes the data in advance to create an encoded representation that minimizes energy-consuming patterns, and the decoder reverses this transformation at the destination, allowing energy reduction to be achieved through pre-processing rather than during the actual transfer
2Loss of energy
If minimum Hamming weight encoding is used to reduce '1' values, then energy consumption in terminated interconnects is reduced, but the encoding and decoding process adds system complexity
Solution Approach 1:
The patent changes the data encoding parameter from standard binary to Hamming weight-encoded format, where the number of '1' values is minimized. This parameter transformation reduces power consumption in terminated interconnects because fewer '1' values mean fewer signal level transitions and less capacitive charging/discharging, while the encoding/decoding logic remains relatively simple and can be implemented with standard digital circuits
3Productivity
If data transfer volume is increased to meet processing demands, then computational capability is improved, but energy consumption from data transfer increases proportionally
Solution Approach 1:
The patent applies parameter changes by transforming the data encoding to minimize energy-consuming patterns. By using Hamming weight encoding, each bit transferred consumes less energy, allowing the system to increase data transfer volume to meet processing demands while the energy cost per unit of data transferred remains reduced, thus improving productivity without proportional energy increase
Data Source
AI summary
The energy consumed by data transfer in a computing device may be reduced by transferring data that has been encoded in a manner that reduces the number of one “1” data values, the number of signal level transitions, or both. A data destination component of the computing device may receive data encoded in such a manner from a data source component of the computing device over a data communication interconnect, such as an off-chip interconnect. The data may be encoded using minimum Hamming weight encoding, which reduces the number of one “1” data values. The received data may be decoded using minimum Hamming weight decoding. For other computing devices, the data may be encoded using maximum Hamming weight encoding, which increases the number of one “1” data values while reducing the number of zero “0” values, if reducing the number of zero values reduces energy consumption.