Packet Encoding with Network-Coded Sub-Packets for Lower BLER
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication systems face challenges in reducing the Packet Error Rate (BLER) for large packets divided into multiple encoding blocks, especially in systems without ARQ support, leading to inefficiencies in signal-to-noise ratio and increased complexity.
Innovation Solution
A packet processing method that involves code block segmentation, error correction encoding, network encoding, and bit selection to form encoded packets, using techniques like LDPC, Turbo, or convolutional encoding, while ensuring a high signal-to-noise ratio and reducing the number of retransmissions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If packets are divided into many code blocks for transmission, then the channel encoding complexity is reduced, but the Packet Error Rate increases due to more blocks needing to be error-free
Solution Approach 1:
The patent divides a large packet into multiple sub-packets, and each sub-packet is further divided into code blocks for independent encoding. This segmentation allows the system to manage encoding complexity at the code block level while maintaining packet-level reliability through sub-packet-level error detection and indication mechanisms.
Solution Approach 2:
The patent introduces sub-packets as an intermediary layer between the original packet and code blocks. Each sub-packet contains multiple code blocks and has its own error detection capability (through CRC bits). This intermediary structure allows error identification at the sub-packet level, enabling selective retransmission or forward error correction without requiring all code blocks to be perfectly decoded, thus resolving the contradiction between fine-grained encoding and packet-level reliability.
2Reliability
If the Packet Error Rate is reduced to ensure reliable transmission, then transmission reliability improves, but the signal-to-noise ratio requirement increases leading to lower system efficiency
Solution Approach 1:
Instead of requiring all code blocks in a packet to be perfectly decoded (excessive action), the patent allows some sub-packets to contain errors while still considering the overall packet successfully received if enough sub-packets are correctly decoded. This partial success approach reduces the stringent SNR requirements needed for perfect packet-level error-free transmission.
Solution Approach 2:
The patent changes the evaluation parameter from binary packet-level success/failure to a more nuanced metric that considers the number of correctly received sub-packets. This parameter change allows the system to operate at lower SNR ratios by accepting a certain threshold of errors while maintaining adequate overall transmission reliability.
3Reliability
If ARQ mechanism is used to ensure reliable transmission, then transmission reliability improves, but retransmission delay increases
Solution Approach 1:
The patent performs error detection (CRC) on each sub-packet during the initial encoding and transmission phase. By preliminarily identifying which sub-packets contain errors, the system can prepare for potential retransmission of only those specific sub-packets rather than waiting for complete packet-level ACK/NACK feedback, thereby reducing retransmission delay.
Solution Approach 2:
By segmenting the packet into multiple sub-packets with independent error detection, the patent enables selective retransmission of only the erroneous sub-packets rather than retransmitting the entire packet. This segmentation-based selective retransmission significantly reduces the average retransmission delay compared to traditional ARQ mechanisms that operate at the packet level.
Data Source
AI summary
A packet processing method and device are disclosed. The method includes: performing code block segmentation on a source packet to obtain a sub-packets; performing error correction encoding on each sub-packet respectively, or performing error correction encoding after respectively adding a CRC sequence to each sub-packet to obtain a error correction encoding sub-packets; performing network encoding on the a error correction encoding sub-packet to obtain b check sub-packets; and performing bit selection operation on the a error correction encoding sub-packets and the b check sub-packets respectively to collectively form an encoded packet; herein a and b are integers greater than 0.