HARQ Soft-Bit Compression for Non-Terrestrial Network Memory Limits
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In non-terrestrial networks, the increased round trip delay leads to a larger size of the hybrid automatic repeat request (HARQ) memory, which complicates implementation and design due to the proportional increase in HARQ processes and memory size.
Innovation Solution
A method for compressing soft bits in the HARQ memory using operations such as transform coding, thresholding, zigzag or diagonal read schemes, and encoding algorithms like run-length encoding and Lempel-Ziv-Welch, to reduce the memory size required for storing compressed soft bits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the round trip delay increases in non-terrestrial networks, then the number of HARQ processes increases, but the HARQ memory size becomes excessively large
Solution Approach 1:
The patent extracts only the essential information from soft bits by separating them into information values and sign values, then selectively processing only the information values through compression. This extraction approach retains the necessary data for HARQ reliability while removing redundant information that contributes to memory size growth.
Solution Approach 2:
The patent changes the parameter representation of soft bits by transforming them through mapping to subblocks, applying transform coding, and performing thresholding operations. These parameter transformations convert the original soft bit format into a compressed representation that requires less memory space while preserving the essential information needed for HARQ processes.
2Quantity of substance
If the HARQ memory size increases to accommodate more processes, then more soft bits can be stored, but implementation and design become complicated
Solution Approach 1:
The patent segments the soft bits into distinct components (information values and sign values) and further divides information values into subsets that are mapped to specific subblocks. This segmentation allows for targeted compression operations on only the necessary portions of the data, reducing overall memory requirements while simplifying the management of HARQ processes.
Solution Approach 2:
The patent applies compression operations selectively rather than to all soft bits uniformly. By performing transform coding and thresholding only on information values and using run-length encoding specifically on compressed results, the system achieves sufficient compression without the overhead of applying complex algorithms to every bit, thereby reducing implementation complexity.
3Volume of stationary object
If soft bits are compressed using transform coding and encoding operations, then memory size is reduced, but processing latency increases
Solution Approach 1:
The patent applies thresholding operations periodically during the compression process, where values below a certain threshold are set to zero. This periodic thresholding creates a compressed representation more efficiently than continuous compression algorithms, reducing both memory size and processing time by eliminating the need for complex continuous transformations.
Solution Approach 2:
The patent uses run-length encoding on compressed information values, which is a computationally inexpensive compression technique. By employing this simpler encoding method rather than more complex algorithms, the system achieves adequate compression ratios with minimal processing overhead, thus reducing latency while still reducing memory size.
Data Source
AI summary
Methods and apparatuses for HARQ in a non-terrestrial network are disclosed. An operation method of a first node may comprise receiving a transport block (TB) from a second node; classifying total soft bits for the TB into information values and sign values; configuring the information values of the total soft bits into one or more subsets; performing a compression operation on each of the one or more subsets; and performing a compression operation on the sign values. Therefore, performance of the communication system can be improved.