Parallel Channel Packet Partitioning for Faster FEC Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data transmission systems over multiple parallel channels face challenges in efficiently scheduling packets to ensure reliable and timely recovery of data segments due to varying channel performance, leading to redundant data transmission and delays.
Innovation Solution
The system partitions the data segment into equal-length packets, generates channel-encoded copies using forward error correction techniques, and schedules initial packets across channels based on their expected transmission speeds, allowing for early termination of packet transmission when sufficient packets are received, thereby reducing redundancy and transmission time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is transmitted over multiple parallel channels using FEC, then reliability of data recovery is improved, but transmission time increases due to scheduling delays and redundant data transmission
Solution Approach 1:
The system performs preliminary scheduling of packets across multiple channels based on expected transmission speeds before actual transmission. This pre-scheduling optimizes the distribution of systematic and encoded packets, ensuring that sufficient packets are delivered to the receiver in minimum time while maintaining recovery reliability.
Solution Approach 2:
The scheduling system dynamically adapts to varying channel performance by adjusting packet distribution in real-time. The scheduler monitors channel conditions and reallocates packets to optimize transmission efficiency, allowing the system to maintain reliability while minimizing transmission time under different network conditions.
2Reliability
If redundant encoded packets are transmitted across all channels, then data recovery reliability is improved, but network resource utilization deteriorates
Solution Approach 1:
The system extracts and transmits only the necessary number of encoded packets required for reliable data recovery, rather than transmitting all possible encoded packets across all channels. The scheduler determines the optimal number of encoded packets needed and allocates transmission resources accordingly, eliminating redundant transmissions and improving network efficiency.
Solution Approach 2:
The system applies partial action by transmitting a carefully calculated subset of encoded packets that is sufficient for recovery but not excessive. This optimized approach provides just enough redundancy for reliable recovery while avoiding the waste of transmitting unnecessary packets, thus improving network resource utilization.
3Device complexity
If packets are scheduled without considering channel performance variations, then scheduling complexity is reduced, but transmission efficiency deteriorates due to delays
Solution Approach 1:
The scheduling system incorporates channel performance parameters (such as expected transmission speeds) into the packet scheduling decisions. By adjusting scheduling parameters based on channel conditions, the system optimizes transmission efficiency without requiring complex real-time control mechanisms, achieving a balance between simplicity and performance.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Transmit data over multiple communication channels in parallel with forward error correction. An optimized number is determined to partition a data segment of a given size into the optimized number of original packets of the same size, by reducing the cost of transmitting dummy data added to the original packets due to the partition, the data fields added to communication packets to support decoding, and redundant packets that are expected to be transmitted via a plurality of parallel channels before the termination of the transmission, as well as the computation cost that increases as a function of the number of original packets. Copies of packets are generated by distributing the original packets to the copies as initial packets and generating each subsequent channel-encoded packet by rejecting useless channel-encoded packets in view of packets assumed to have been received prior to the transmission of the subsequent channel-encoded packet.