Trading Delay for Distortion in One-Way Video Communication Over the Internet

Abstract

We study the problem of one-way video communication in single-source, multiple-destination scenario over the lossy Internet. Forward Error Correction (FEC) coding is commonly adopted for data protection in implementing loss-resilient video transmission systems. However, the burst packet losses over the Internet frequently degrade the FEC performance and induce video quality deteriorations. To address the challenging problem, we propose a novel transmission scheme dubbed Trading dELay FOR Distortion (TELFORD) that includes three components: 1) adaptive multi-destination status estimation; 2) delay-constrained transmission rate assignment; 3) differentiated FEC packets spreading. We analytically formulate and solve the problem of FEC packets allocation and scheduling to minimize the endto-end video distortion. The proposed TELFORD is able to cope with multiple-destination scheduling separately. We conduct the performance evaluation through semi-physical emulations in Exata using real-time H.264 video streaming. Experimental results show that TELFORD outperforms existing error control transmission schemes in improving the video PSNR (Peak Signalto-Noise Ratio) and mitigating packet transmission impairments.

TCP-Oriented Raptor Coding for High-Frame-Rate Video Transmission Over Wireless Networks

“…As shown in Figure 6, each router in the core networks is connected to four different traffic generators, which are used to generate background traffic and emulate network dynamics. The packet sizes of the Pareto traffic generator are adapted to mimic the real traces collected on the Internet: 43% are large (> 1400 Bytes), 17% are medium [(144, 1400) Bytes], and 40% are small (< 144 Bytes) [41]. The other cross traffic generators include FTP (File Transfer Protocol), CBR (Constant Bit Rate), VBR (Variable Bit Rate).…”

Delay-Constrained High Definition Video Transmission in Heterogeneous Wireless Networks with Multi-Homed Terminals

“…The GoP structure is IP...P. The test video sequences are HD content blue sky, riverbed, and mobcal. Each of the HD test sequences features a different pattern of temporal motion and spatial characteristics [43]. We concatenate the test sequences several times to be 3000 frame-long in order to obtain statistically meaningful results.…”

“…We concatenate the test sequences several times to be 3000 frame-long in order to obtain statistically meaningful results. The delay constraint (T) for each video frame is set to be 50 ms to guarantee the seamlessrate video streaming perceived by end users [43].…”

Enabling Adaptive High-Frame-Rate Video Streaming in Mobile Cloud Gaming Applications

“…If the destination receives at least k out of the n data packets, it can recover all the source packets. If the number of received packets is less than k, the received packets can still be used for the decoding process due to the systematic property of the FEC code [53]. The soft-decision decoding algorithm (e.g., the Koetter-Vardy (KV) decoder [49][50] [51]) is implemented at the receiver side to achieve such goal.…”

“…As the gaming application is imposed with stringent delay constraint, the value of θ should be small enough (e.g., less than 0.5 ms) to respect the delay constraint [53]. Then, we can have the expression of P(c) as…”