关键词: 可伸缩视频编码, 非均等错误保护, 自适应遗传算法, 优化, 丢包率

Abstract: In order to improve the packet loss resilience of Scalable Video Coding (SVC) over communication networks, an efficient Unequal Error Protection (UEP) algorithm for SVC using adaptive genetic algorithm was proposed. A method to encapsulate network abstract layer units according to the feature of the head information of a packet was introduced. Then the problems of pair codes assignment were transformed into the problems of multi-constraint optimization, which could be transformed into unconstrained objective by exploiting penalty function. Therefore, the adaptive genetic algorithm was employed to obtain globally optimal solution. The simulation results reveal that compared with the typical unequal error protection algorithms, the Peak Signal-to-Noise Ratio (PSNR) is improved by 0.8dB-1.95dB, and the proposed algorithm provides substantial improvement for the decoding speed and received video quality over best effort packet networks.

Key words: Scalable Video Coding (SVC), Unequal Error Protection (UEP), adaptive genetic algorithm, optimization, packet loss ratio