Adaptive Packet-Level FEC Algorithm for Improving the Video Quality over IEEE 802.11 Networks

Wireless video transmission suffers from degrading the quality of video service due to dynamic change of the wireless network environment. Packet-level forward error correction (FEC) enables video to be transmitted robustly as adding the redundant packets. This paper proposes two cross-layer strategies to improve the quality of video using the adaptive packetlevel FEC over IEEE 802.11 networks. These strategies adaptively produce redundant packets according to the types of video frames encoded by H.264. Among the frame types, I frame (Intra frame) is the most important because it is referenced by other frames as well as decoded by itself without the prediction of any other frames. One approach, named AFEC_PI (Adaptive FEC with Priority to Intra frames), generates FEC packets only for Intra frames. The other, named AFEC_BI (Adaptive FEC with Both Intra and Inter frames), makes FEC packets for all frames, but applies differential level according to the types of frames. By using the two approaches, the received video quality is estimated based on the various error rates in wireless networks. Experimental results show that AFEC_PI improves video quality as generating fewer FEC packets than the traditional packet-level FEC and AFEC_BI increases PSNR more than AFEC_PI at a low error rate.