Binary-Clustered FEC Encoding for Multi-Packet Loss Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional XOR-based forward error correction (FEC) techniques in multimedia streaming often fail to provide adequate redundancy for packet recovery, especially when multiple packets are lost within the same frame, and may require excessive computation and transmission of duplicate packets.
Innovation Solution
The binary clustered FEC encoding scheme selects media packets into clusters based on their binary representation, allowing each packet to contribute to multiple FEC packets, thereby enhancing redundancy and recovery rates without transmitting duplicates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional XOR-based FEC techniques are used to compute FEC packets, then the encoding process is simple, but the packet recovery rate is inadequate especially when multiple packets are lost in the same frame
Solution Approach 1:
The patent segments media packets into clusters based on their binary representation patterns. Each cluster is identified by a specific bit position in the binary representation, creating multiple clusters that collectively cover all media packets. This segmentation allows the system to generate multiple FEC packets, each protecting a specific cluster, thereby improving overall packet recovery rate without requiring complex encoding operations beyond XOR.
Solution Approach 2:
The patent introduces a new dimension to FEC encoding by utilizing the binary representation structure of packet indices. Instead of treating all packets uniformly, the system exploits the bit patterns in binary representations to create structured clusters. This dimensional approach (using bit positions as cluster identifiers) enables more efficient redundancy distribution and improves recovery capabilities while maintaining encoding simplicity.
2Reliability
If more FEC packets are computed and transmitted to improve redundancy, then packet recovery capability improves, but network bandwidth consumption increases
Solution Approach 1:
The patent applies partial action by generating FEC packets for only specific clusters of media packets rather than creating redundant copies of all packets. Each FEC packet protects a particular cluster identified by a bit position, providing targeted redundancy. This approach achieves adequate protection with fewer FEC packets compared to conventional methods that would require more extensive redundancy to ensure recovery of any arbitrary set of lost packets.
3Reliability
If duplicate media packets are transmitted to improve recovery, then packet recovery improves, but transmission efficiency decreases
Solution Approach 1:
Instead of transmitting duplicate media packets, the patent creates compressed representations (FEC packets) that are XOR combinations of multiple media packets from specific clusters. These FEC packets serve as compact copies that can recover original packets when needed. This copying approach is more efficient than transmitting actual duplicates because multiple pieces of information are encoded into single FEC packets, reducing overall transmission volume while maintaining recovery capability.
Data Source
AI summary
Embodiments of the present disclosure relate to a binary clustered forward error correction encoding scheme. Systems and methods are disclosed that define binary clustered encodings of the media packets from which forward error correction (FEC) packets are computed. The different encodings specify which media packets in a frame are used to compute each FEC packet (a frame includes M media packets). The different encodings may be defined based on the quantity of media packets in a frame, M≤floor(2N), where each bit of the binary representation of N is associated with a different cluster pair encoding of the media packets. Each cluster pair includes a cluster for which the bit=0 and a cluster for which the bit=1. Computing FEC packets using at least two cluster pair encodings provides redundancy for each media packet, thereby improving media packet recovery rates.