Semiconductor Chip Data Encoding for Bandwidth Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing communication protocols between semiconductor chips result in a decrease in effective bandwidth due to the use of identification codes, which act as communication overhead, thereby reducing the efficiency of data transmission.
Innovation Solution
A data encoding and decoding method that stores and processes data sequences in units of character and identification code pairs, converting control characters into internal control characters with reference numbers, allowing for the generation of data blocks that omit identification codes, thereby increasing the effective channel bandwidth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If identification codes are attached to each character for communication between semiconductor chips, then data transmission reliability is improved, but effective bandwidth is decreased
Solution Approach 1:
The patent extracts and removes identification codes from the data transmission stream. Instead of attaching identification codes to each character, the system transmits only data characters, extracting the identification function entirely from the transmission protocol, thereby eliminating the bandwidth overhead while maintaining transmission reliability through alternative error detection mechanisms
Solution Approach 2:
The patent makes the communication protocol universal by designing it to handle both data transmission and error detection without requiring separate identification codes. The protocol uses a unified character set where all characters are treated equally, and error detection is achieved through checksum or parity mechanisms that work across all character types, eliminating the need for specialized identification markers
2Measurement precision
If identification codes are transmitted with each character, then character classification accuracy is improved, but communication overhead increases
Solution Approach 1:
The patent extracts the identification function from the transmission data by removing identification codes entirely. The system classifies characters based on their inherent data structure and context rather than requiring separate identification markers, thereby achieving accurate character classification without communication overhead
Solution Approach 2:
The patent introduces an intermediary decoding mechanism that reconstructs character information from compressed or encoded formats without requiring identification codes. The decoder uses context awareness and pattern recognition to accurately classify characters based on their position, structure, and relationship to surrounding data, eliminating the need for explicit identification markers
Data Source
AI summary
In a data encoding method for a communication between semiconductor chips, a data sequence is received. The data sequence includes multiple characters and multiple identification codes coupled to the characters, respectively. The data sequence is stored in a buffer in a unit of a predetermined number of pairs of characters and identification codes. Whether at least one control character is stored in the buffer is determined based on the identification codes stored in the buffer. When at least one control character is stored in the buffer, the at least one control character is converted to at least one internal control character, respectively. When multiple control characters are stored in the buffer, at least one internal control character includes a reference number representing a location of a sequentially next control character in the buffer. A first type data block is generated. The first type data block includes data characters stored in the buffer and the at least one internal control character.