HARQ Buffer Subset Selection for Memory Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
HARQ buffer size requirements in communication systems are excessive, particularly in high-throughput systems like UMTS LTE, due to the need to store soft bit values, which increases hardware size and power consumption, and existing methods to limit buffer size, such as Limited-buffer Rate Matching and Two-stage Rate Matching, come at the expense of increased effective code rate.
Innovation Solution
A method where only a subset of soft symbol values is stored in the HARQ buffer, determined dynamically based on criteria such as target signal-to-noise ratio, block error rate, and available buffer size, reducing the memory needed for HARQ processing without significant loss in performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all soft bit values are stored in the HARQ buffer, then decoding reliability is improved, but buffer size and hardware complexity increase significantly
Solution Approach 1:
The patent extracts and stores only a subset of the most important soft bit values (those corresponding to systematic bits or bits with highest reliability) in the HARQ buffer, rather than storing all soft bit values. This selective extraction maintains essential decoding information while significantly reducing buffer memory requirements.
Solution Approach 2:
The patent applies different storage quality levels to different soft bit values based on their importance. Critical soft bits (systematic bits) are stored with full precision, while less critical bits are either stored with reduced precision or not stored at all, optimizing the trade-off between reliability and buffer size.
2Measurement precision
If high resolution is used for each stored soft bit value, then decoding accuracy is improved, but memory space requirements increase
Solution Approach 1:
The patent applies different resolution levels to different soft bit values based on their significance. Systematic bits are stored with full resolution to maintain decoding accuracy, while parity bits or less critical soft bits are stored with reduced resolution or truncated, thereby reducing overall memory space requirements.
Solution Approach 2:
The patent changes the resolution parameter of soft bit storage dynamically or selectively, using full precision for critical bits and reduced precision for non-critical bits, optimizing the balance between decoding accuracy and memory consumption.
3Productivity
If multiple simultaneous HARQ processes are supported, then system throughput is improved, but HARQ buffer size increases
Solution Approach 1:
The patent extracts and stores only the essential subset of soft bits for each HARQ process, enabling multiple simultaneous processes to be supported with reduced buffer size per process. The selective storage of critical soft bits allows parallel HARQ operations without proportionally increasing total buffer requirements.
Solution Approach 2:
The patent designs a universal HARQ buffer structure that can efficiently handle multiple simultaneous HARQ processes using the same subset selection and storage mechanism, allowing the buffer to serve multiple functions and processes without requiring separate full-size buffers for each process.
4Volume of stationary object
If Limited-buffer Rate Matching or Two-stage Rate Matching is used, then buffer size is reduced, but effective code rate increases
Solution Approach 1:
The patent extracts and stores the most critical soft bits (systematic bits) in the HARQ buffer, which contain the essential information for decoding. This selective extraction maintains better effective code rate performance compared to rate matching methods, while still achieving buffer size reduction, because the stored systematic bits provide the foundation for successful decoding without requiring additional redundancy.
Data Source
AI summary
A method for an electronic receiver of processing a packet of a hybrid automatic repeat request (HARQ) system is disclosed. The method comprises receiving a first transmission of the packet, wherein the first transmission comprises a first plurality of soft symbol values and determining whether the first plurality of soft symbol values meets a stop criterion. If it is determined that the first plurality of soft symbol values does not meet the stop criterion, a subject of the first plurality of soft symbol values is determined, wherein the subset comprises a number of soft symbol values of the first plurality of soft symbol values, the number being greater than zero and less than the first plurality. The subset of the first plurality of soft symbol values is stored in a HARQ buffer. A second plurality of soft symbol values, comprised in a second transmission of the packet received by the electronic receiver, is combined with the stored subset of the first plurality of soft symbol values to produce a third plurality of combined soft symbol values. Corresponding arrangement, receiver, device, and computer program product are also disclosed.