Variable-Code FEC for Real-Time Video Transmission

The patent applies dynamics by making the error correction rate variable rather than constant. The transmitting terminal dynamically adjusts the error correction rate based on real-time network conditions (packet loss rate, delay, jitter) detected by the receiving terminal and fed back through RTCP packets. This allows the system to adapt the redundancy level to actual network needs, reducing unnecessary redundant load when network conditions are good while maintaining high reliability when conditions deteriorate.

The patent implements parameter changes by modifying the error correction rate parameter according to network conditions. Different error correction rates (e.g., 1/2, 2/3, 3/4, 7/8) are selected based on the current network state, allowing the system to optimize the balance between reliability and bandwidth utilization by changing this critical parameter dynamically rather than using a fixed value.

The system dynamically adjusts the error correction rate based on network conditions and packet importance. For time-sensitive packets (e.g., key video frames with small remaining lifetime), the error correction rate is reduced or omitted to minimize processing delay. For less time-sensitive packets, higher error correction rates are applied. This dynamic approach balances reliability improvement with delay minimization by adapting the correction intensity to each packet's urgency.

The patent applies local quality by differentiating error correction treatment based on packet characteristics. Different error correction rates are applied to different types of packets (e.g., I-frames vs. P-frames, packets with different remaining lifetimes). This selective approach ensures that critical packets receive adequate protection while less critical packets undergo minimal processing, thereby maintaining system reliability without uniformly increasing transmission delay for all packets.

The patent implements preliminary action by proactively adding forward error correction codes to packets before transmission, rather than waiting for packet loss to occur and then requesting retransmission. This pre-prepared redundancy allows the receiving terminal to immediately recover lost packets without initiating a retransmission cycle, thereby reducing the delay inherent in ARQ protocols while still maintaining efficient bandwidth utilization.

The system uses feedback from the receiving terminal (through RTCP packets containing network condition information) to dynamically adjust the error correction rate. This feedback mechanism allows the transmitting terminal to optimize the balance between redundancy and delay by adapting to actual network performance, reducing unnecessary retransmissions while maintaining reliable delivery.

The patent applies dynamics by transitioning from a static error correction rate to a dynamic one. The receiving terminal monitors network conditions (packet loss rate, delay, jitter) and feeds this information back to the transmitting terminal, which then dynamically adjusts the error correction rate accordingly. This dynamic adaptation significantly improves network condition responsiveness while the underlying FEC mechanism remains relatively simple, managing complexity through intelligent control rather than complex algorithms.

The system implements feedback by having the receiving terminal continuously monitor network conditions and communicate this information back to the transmitting terminal using RTCP packets. This feedback loop enables the transmitting terminal to adapt the error correction rate to current network conditions, significantly improving versatility and adaptability while maintaining manageable system complexity through the use of standard feedback protocols.