video streaming quality
O.I. Sheluhin, U.A. Ivanov
Streaming traffic is the kind of traffic which is characterized by processing data in real time as it received by user equipment. Transfer of streaming services over different media (wireless systems, Internet) becomes more and more popular. This rapid spread defines a new challenge to maintain service quality for each flow. However, transmission errors can seriously affect the quality of streaming video because the compressed data is very sensitive to these errors. This paper investigates the impact of errors on the quality of streaming H.264/AVC video in broadband wireless access systems such as WiMAX.
Wireless channels are characterized by randomly distributed and independent bit errors. Therefore, the model of additive white Gaussian noise (AWGN) is often used in simulating the wireless channel, in which certain bits in the sequence are inverted with a given probability. This probability is described by the value of BER. We can calculate the impact of bit errors on the final quality of the video by comparing the original and distorted video. Traditionally, the video quality is evaluated using subjective and objective indicators. Subjective quality assessment is always based on the impression of the viewer and calculated as the average score MOS (Mean Opinion Score). Objective video quality is usually measured by peak signal to noise ratio (PSNR). The simulation transfer of a 30-minute video over a IEEE 802.11 wireless network with random bit errors in the channel was held in order to analyze the effect of bit errors on the resulting quality of the video.
It is shown that BER ≤ 3 * 5.10 does not affect on the received video quality and easily eliminated by implemented in WiMAX methods of protection. BER ≥ 4 * 3.10 leads to the maximum value of loss and to an unacceptable quality of the received video. We can ensure excellent quality of video when the probability of bit error rate less than 1 * 10-4, a good quality in a range of 1 * 10-4 to 4 * 10-4, satisfactory quality in the range of 4 * 10-4 to 8 * 10-4, poor quality in the range of 8 * 10-4 to 1 * 10-3 and worst at BER ≥ 1 * 3.10.
Using variable length codes in WiMax systems in coding of H.264/AVC video leads to disruption of the synchronization of decoded video sequences and the appearance of additional groups of errors. This influence on the decoded video is much stronger than bit errors, because it leads to loss of large segments of the information.