Date of this Version
Multimedia communication is expected to be the major application of the next-generation wireless networks. However, providing delay-sensitive, loss-tolerant and resource-guzzling multimedia services over the current resource-limited wireless networks is still a very challenging task. In addition to their heterogeneous nature and resource limitation, wireless networks also suffer from channel variations and environmental changes. In this dissertation, we study the optimization techniques for wireless multimedia communications and address the issues from different theoretical and practical perspectives.
To solve the issues resulting from the increasingly heterogeneous wireless networks, we propose a framework for error resilient source coding and distributed application-layer error control, where the encoder calculates source encoding options, enabling globally optimal rate allocation between source and channel bits for all the hops considered in the end-to-end path.
Further, we propose content-aware video communications to differentiate the region of interest (ROI) from the less important background area so that resources can be classified, prioritized and more efficiently utilized for encoding and transmission. We also apply the theories by proposing a wireless e-healthcare system.
To take into consideration of channel variations and to provide a framework that can dynamically adapt to the instant channel changes, we propose a cross-layer optimized framework which jointly optimizes different functions residing at different network layers, including video encoding, network congestion, retransmission and link adaptation. We also demonstrate how this design can be widely used in wireless environments by presenting our research on the scheduling algorithm of wireless peer-to-peer (P2P) multimedia networks.
At last, we move one step further on dynamic adaptation to many other ``context" information such as user preference, location, weather, etc. By providing the design of an adaptive multimedia communications system based on context-aware services, we study how to retrieve and utilize all the context data to better serve the users with significantly-improved video quality.
The proposed optimization techniques are usually discussed within certain application scenarios to demonstrate the practicality, efficacy and efficiency. However, these research findings can be widely used in other wireless networks and applications.