The functionality of smart mobile devices is improving rapidly but these devices are limited
in terms of practical use because of battery-life. This situation cannot be remedied by simply
installing batteries with higher capacities in the devices. There are strict limitations in the
design of a smartphone, in terms of physical space, that prohibit this “quick-fix” from being
possible. The solution instead lies with the creation of an intelligent, dynamic mechanism for
utilizing the hardware components on a device in an energy-efficient manner, while also
maintaining the Quality of Service (QoS) requirements of the applications running on the
device.
This thesis proposes the following Energy-aware Adaptive Solutions (EASE):
1. BaSe-AMy: the Battery and Stream-aware Adaptive Multimedia Delivery (BaSe-AMy)
algorithm assesses battery-life, network characteristics, video-stream properties and
device hardware information, in order to dynamically reduce the power consumption of
the device while streaming video. The algorithm computes the most efficient strategy for
altering the characteristics of the stream, the playback of the video, and the hardware
utilization of the device, dynamically, while meeting application’s QoS requirements.
2. PowerHop: an algorithm which assesses network conditions, device power consumption,
neighboring node devices and QoS requirements to decide whether to adapt the
transmission power or the number of hops that a device uses for communication.
PowerHop’s ability to dynamically reduce the transmission power of the device’s
Wireless Network Interface Card (WNIC) provides scope for reducing the power
consumption of the device. In this case shorter transmission distances with multiple hops
can be utilized to maintain network range.
3. A comprehensive survey of adaptive energy optimizations in multimedia-centric wireless
devices is also provided.
Additional contributions:
1. A custom video comparison tool was developed to facilitate objective assessment of
streamed videos.
2. A new solution for high-accuracy mobile power logging was designed and implemented.