This Small Business Innovation Research (SBIR) Phase I project will explore how techniques from coding and signal processing can be used to build better digital to analog converters for RF communications and other applications. A key requirement for such devices in radio frequency hardware is dynamic range, i.e., the ability to represent small and large signals simultaneously. Today, the highest dynamic range converters use a technique called Sigma-Delta modulation. Sigma-Delta modulation allows a high precision signal to be generated from binary quantized data, and so removes the requirement for precise analog levels. Traditionally, Sigma-Delta modulation is explained in terms of a linear feedback system. This research will show that Sigma-Delta modulation is, in fact, a crude optimization algorithm. The work will go on to find far better ways of solving this optimizations problem than the Sigma-Delta approach that is used now. The broader impact/commercial potential of this project will appear wherever digital computation must be interfaced to the physical world, e.g., to get the analog signal which an antenna radiates. Indeed, the digital revolution depends on analog-to-digital and digital-to-analog conversion. Personal CD players, cellular telephones, high definition TV, computer modems, and deep space communication systems all use analog-to-digital and digital-to-analog conversion blocks. Furthermore, A to D and D to A conversion is currently the weak link in many important systems, e.g., software defined radio. Looking at the conversion problem from a new perspective will result in solutions with improved performance and reduced complexity. Considering the number of CD players, cellular telephones, modems, and televisions in existence today, any improvement in performance or cost reduction in the manufacturing of these products will have a major impact on the world in which we live and improve our standard of living
