Rapid Refinable Software Defined Radio Design

摘要

A major challenge in design of the Software Defined Radio (SDR), or indeed any real-ime embedded system, is the optimal selection of the candidate algorithms, implementations and platforms. In particular, since the SDR methodology allows for the convergence of seemingly different communication standards into a single realization, the optimal selection is of paramount importance. In this paper, the SDR design flow and platform configuration are modeled as an Integer Linear Programming (ILP) optimization problem. We will show that both the refinement process and configurability can readily be presented via sets of appropriate objective and constraint functions. The analysis of the Bluetooth wireless technology is demonstrated as an example.