Power Conservation Methods for Real-time Wireless Video Transfer Applications

Abstract

Cellular and Handheld devices are evolving from simple feature entities into true smart devices, providing all kinds of multimedia applications and high bandwidth ubiquitous access. A key challenge is the power management of these devices, as they are battery powered and data processing and communication required by multimedia applications are extremely power hungry. In this dissertation, the focus is to optimize and reduce power consumption for Real-time Wireless Video Transfer applications on Cellular and Handheld devices. Two important and original contributions have been made by this research work. The first is that a power management framework integrated with modules of Real-time Wireless Video Transfer applications has been proposed. The other is that a set of signaling protocols and algorithms for supporting the proposed Real-time Video Transfer Network Architecture are designed and implemented. To date, no studies have applied such power management strategies that explicitly integrate power control interfaces with wireless video transfer applications in the context of an optimal power management framework, to the author's knowledge.In the first part of this dissertation, various power conservative design techniques from non-network related and network related perspectives have been examined. The shortcomings of current power management of operating systems for Real-time Wireless Video Transfer applications have been explained. Then, the dominant power consuming components on Mobile devices are indentified when Real-time Wireless Video Transfer applications are running. At the same time, different factors which can be used to define the operating points and can be used in developed power control strategies have been discussed in detail. In the second part of this dissertation, the desirable architecture of power management module has been proposed. The main functions of power management module and detailed communications between Real-time Wireless Video Transfer applications and power management strategy are implemented. The performance of power management scheme has been evaluated experimentally and analytically. Our study shows that about 44.3% power usage can be saved with the optimized power management. In the third and final part of this dissertation, a novel integrated multi-hop cellular data network instead traditional infrastructure has been proposed. Design of a low-complexity routing algorithm to relay upstream and downstream data for this new network infrastructure has also been proposed. To further improve the power consumption of the network, a distributed layer scheduling algorithm with load balancing is discussed. In the simulation, it is shown that the proposed algorithm delivers excellent performance across a wide range of data packet parameters and configurations. The results of this research provide useful insight into optimal power management for wireless video transfer applications on mobile and handheld devices.