ENERGY EFFICIENT CODING IN WIRELESS VIDEO COMMUNICATION

Abstract

In scenarios with wireless communications involved, efficient utilization of transmission energy is an important design consideration. As most of the users of the wireless network are mobile, they rely on a battery with a limited energy supply. Minimization of transmission energy thus can extend the lifetime of this battery. With the usage of smallest amount of transmission energy to transmit a message it decreases the likelihood of the interception of that message. Reduction in transmission energy decreases the interference between users sharing a wireless link: thereby increasing the overall network capacity. Energy-efficient wireless communication network design is an important and challenging problem. The importance lies on the fact that the mobile units are operating with batteries having limited energy supply. The challenge lies with the fact that there are many different issues that needs to be dealt with while designing a low-energy wireless communication system (for instance amplifier design, coding, modulation design, resource allocation, and routing strategies) besides, these issues are coupled among one another. For the design of an energy-efficient communications system the understanding of sources of energy consumption in mobile devices is mandatory. Computations, transmission, display, and driving speakers are main sources of energy consumption. Among these main energy consumers are computation and transmission. In the process of computation, energy is consumed for the running of the operating system software, and for encoding and decoding of the audio and video signals. As for transmission, energy is consumed for the transmission and acceptance of the radio frequency (RF) audio and video signals. In this dissertation report the energy efficient coding techniques employed in wireless video communication which can simplify the process of encoding and decoding process are proposed. The proposed techniques use the concept of activity region in the frames individually. The region of activity is coded with more coefficients while transform coding. The determination of activity region is done using the concept of temporal redundancy, spatial redundancy or 3D DCT block energy. Finally a technique which uses 3D DCT and block energy for activity region determination, also it uses dynamic/adaptive blocksize allocation (based on the fact that whether it falls under activity region or not) is proposed