Power efficient and scalable video compression for mobile multimedia devices.

摘要

The energy longevity of mobile devices, especially those which support multimedia, is becoming more important. In this dissertation, two methods are introduced to reduce the energy consumption of mobile video devices, because energy efficiency is critical in prolonging battery life. Also, network bandwidths of these mobile devices vary and are not always guaranteed. First, the work in this dissertation analyzes various lookup table techniques used for variable length coding in video codecs such as H.264/AVC to determine which one is most suitable in a low power environment and which power component should be minimized. The results presented show that a tree lookup table structure is an excellent choice to reduce power consumption that would help prolong battery life of the mobile decoding device. Also, scalability is a major issue for future video systems and their network adaptability and compatibility with legacy devices. Thus, this dissertation also presents a bit- depth scalable video algorithm that can be used to provide both the support for future high bit depth colors as well as reducing mobile power consumption via bandwidth reduction.