Dynamic FEC Decoder Pool for Satellite Return Links
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In satellite communications systems, the dedicated FEC decoders for codewords of varying sizes lead to high latency and inefficient resource utilization due to unpredictable decoding times and bursty data transmissions, particularly in TDMA systems where burst assignments are not deterministic.
Innovation Solution
The method involves dynamically allocating a total number of decoders among decoder queues based on the current offered load of codewords of different sizes, using a proportional allocation and correction factors to manage queuing statistics, thereby optimizing decoder utilization and reducing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If dedicated FEC decoders are assigned to each return link channel, then channel-specific decoding can be performed, but latency increases and resource utilization becomes inefficient
Solution Approach 1:
The patent merges dedicated decoders into a shared decoder pool that serves multiple return link channels. Instead of having separate decoders for each channel, a pool of decoders is created that can be dynamically allocated to any channel needing decoding service, thereby improving resource utilization and reducing latency while maintaining decoding accuracy through the shared pool architecture
Solution Approach 2:
The patent implements dynamic allocation of decoders to channels based on real-time traffic conditions and burst characteristics. The system continuously monitors channel utilization and adjusts decoder assignment dynamically, allowing decoders to be reassigned from idle channels to active channels, thus reducing waiting time and improving overall system efficiency
2Productivity
If more FEC decoders are deployed in the PHY processing subsystem, then decoding capacity increases, but hardware resources and system complexity increase
Solution Approach 1:
The patent creates a universal decoder pool where each decoder can serve multiple return link channels rather than being dedicated to a single channel. This multi-functional approach allows the same hardware resources to be reused across different channels, increasing overall decoding capacity without proportionally increasing the number of physical decoders required
Solution Approach 2:
The patent changes the operational parameters of the decoder system by introducing dynamic allocation mechanisms and statistical multiplexing. Instead of static decoder-channel mappings, the system adjusts decoder assignment based on traffic patterns and burst characteristics, thereby optimizing the use of existing hardware resources to achieve higher effective decoding capacity
3Reliability
If iterative decoding is used for all codewords, then decoding accuracy is maintained, but processing time varies unpredictably causing jitter
Solution Approach 1:
The patent applies partial iterative decoding by limiting the number of iterations for certain codeword types or channel conditions. Instead of always performing full iterative decoding, the system uses a predetermined maximum iteration count or early termination criteria for codewords that can be decoded with fewer iterations, thereby reducing processing time variability and jitter while maintaining adequate decoding accuracy for most cases
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for dividing a total number of decoders among decoder queues of codewords of different sizes, the codewords transmitted on return communication links from data terminals to a gateway of a satellite communications system, includes for each of K groups, allocating a respective number of decoders dedicated to decoding codewords of a particular size, where K is a number of different sizes of codewords, and the respective number of decoders is allocated from the total number of decoders and allocated in proportion to current offered load of codewords of the particular size.