Incremental Data Transmission with Joint Source-Channel Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wireless communications, particularly in systems like LTE and NR, suffer from large air interface overhead due to separate source and channel coding, which is exacerbated by real-time data transmission requirements in applications involving sensing or artificial intelligence, necessitating a reduction in overhead to accommodate future wireless communication scenarios.
Innovation Solution
A method and device for incremental data transmission that involves channel encoding and decoding, with feedback-based retransmission or alternative data transmission based on decoding outcomes and performance metrics, allowing for adaptive source coding to reduce air interface overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If separate source coding and channel coding are performed with lossless compression, then data transmission reliability is improved, but air interface overhead increases
Solution Approach 1:
The patent combines source coding and channel coding into a unified joint source-channel coding framework. The encoder simultaneously performs source compression and channel encoding, while the decoder jointly performs channel decoding and source reconstruction, eliminating the need for separate coding stages and reducing overall overhead while maintaining reliability through integrated error correction and compression.
Solution Approach 2:
The system dynamically adjusts coding parameters such as compression ratio, code rate, and block size based on channel conditions and application requirements. By changing these parameters adaptively, the system can optimize the trade-off between reliability and overhead, selecting higher compression ratios when channel conditions are good and lower ratios when reliability is more critical.
2Loss of information
If lossless compression is applied to source coding, then data integrity is improved, but transmission block length increases
Solution Approach 1:
The patent segments the source data into multiple smaller blocks that are processed independently through the joint source-channel coding framework. Each segment is compressed and protected with appropriate channel coding, allowing parallel processing and reducing the effective block length while maintaining overall data integrity through systematic reconstruction of all segments at the decoder.
Solution Approach 2:
The system applies partial lossless compression to only the most critical portions of the data while using more aggressive compression for less critical portions. This selective approach maintains data integrity for essential information while reducing overall block length, achieving a practical balance between completeness and efficiency.
3Reliability
If HARQ retransmission is performed for error correction, then transmission reliability is improved, but transmission time increases
Solution Approach 1:
The joint source-channel coding framework incorporates forward error correction capabilities directly into the encoding process, providing built-in protection against transmission errors. This beforehand cushioning through integrated FEC reduces the need for reactive HARQ retransmissions, as many errors can be corrected at the decoder without requiring additional transmission rounds, thereby reducing transmission time while maintaining reliability.
Data Source
AI summary
In current wireless communications, air interface overhead resulting from data transmission between devices is too large, especially when the traffic involves sensing or artificial intelligence (AI) tasks. To try to reduce the air interface overhead, the present disclosure provides methods and devices for incremental data transmission with channel and source coding. According to an aspect of the present disclosure, the method may include transmitting, in a first transmission, a first subset of source data that is channel encoded. The method may further include receiving feedback indicating that a further transmission related to the source data is required. In response to receiving the feedback, the method may further include transmitting a second transmission that comprises either a retransmission of the first subset of the source data or a second subset of the source data that is different from the first subset of the source data.