Design of an Adaptive Architecture for Energy Efficient Wireless Image Communication

With the projected signi cant growth in mobile internet and multimedia services, there is a strong demand for next-generation wireless appliances capable of image communication. However, wireless image communication faces signi cant bottlenecks including high energy and bandwidth consumption. Past studies have shown that the bottlenecks to wireless image communication can be overcome by developing adaptive image compression algorithms and dynamically adapting them to current channel conditions and service requirements [1, 2]. In this paper, we present the design of an adaptive hardware/software architecture that enables adaptive wireless image communication. Through intelligent co-design of the proposed architecture and algorithms, we achieve an architecture which enables not only power and performance e cient implementation, but also fast and e cient run-time adaptation of image compression parameters. To achieve e cient image compression and run-time adaptation, we characterized the adaptation needs of an adaptive image compression algorithm in terms of parameters, and implemented an adaptive hardware/software architecture capable of executing JPEG image compression with di erent parameters. We present experimental results demonstrating that the proposed architecture enables low overhead adaptation to current wireless conditions and requirements while implementing a low cost (energy and performance) implementation of adaptive image compression algorithms.