HARQ Buffer LLR Compression for Lower Memory Chip Area
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-performance HARQ operations in wireless communication systems require significant memory space for storing received packets, which is costly in terms of chip area, especially with the massive growth in traffic volume due to 5G wireless communications.
Innovation Solution
The apparatus implements a method that reduces memory usage by compressing and decompressing log likelihood ratio (LLR) values using a compression codebook based on the row weight of the parity check matrix and signal quality, allowing for efficient storage and retransmission of packets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If memory space is increased to store received packets for high-performance HARQ operations, then error control reliability is improved, but chip area cost increases
Solution Approach 1:
The patent extracts only the essential information from received packets for storage in the HARQ buffer. Instead of storing complete packets, it stores extracted features and parameters that are sufficient for error control decisions, thereby reducing memory requirements while maintaining reliability.
Solution Approach 2:
The patent inverts the conventional approach by not storing the actual packet data but rather storing compressed representations and using the original packets when needed for soft combining. This inversion allows the system to maintain high error control performance without requiring large memory capacity.
2Reliability
If complete packets are stored in HARQ buffer for soft combining, then decoding performance is improved, but memory usage increases
Solution Approach 1:
The patent creates compressed copies of packet information that can be stored in limited memory space. These copies contain sufficient statistical information (such as LLR distributions, mean values, and variances) to enable effective soft combining without requiring storage of the complete original packets.
Solution Approach 2:
The patent transforms the storage requirement from complete packet data to compressed parameter representations. By changing the form of stored information from raw packet bits to compressed statistical parameters, the system achieves the same decoding performance with significantly reduced memory usage.
3Area of stationary object
If memory size is reduced to lower chip area cost, then chip area is optimized, but HARQ operation performance degrades
Solution Approach 1:
The patent introduces an intermediary compression mechanism that bridges the gap between limited memory capacity and high-performance HARQ operations. This intermediary layer compresses packet information into compact representations that can be efficiently stored and later decompressed for soft combining, maintaining performance while optimizing chip area.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method reducing amount of memory utilized by hybrid automatic repeat request (HARQ) operations is disclosed. When an apparatus employing the method receives an LDPC encoded packet, the apparatus firstly estimates a signal quality value of the received packet, and performs an LDPC decoding algorithm on the received packet. If the received packet is not successfully decoded by the apparatus, the received and decoding-failed packet is compressed in order to reduce bit-width, and the compressed packet is stored in a memory device for later HARQ operations. The apparatus employing the method is also disclosed.