Ctrl-P to Print

Analog to Information (A-to-I)

Overview

Military signals analysis today is expensive in hardware, time, operator effort, and computation. Are there smarter ways to glean information from the complex soup of broadband electronic signals bouncing through the atmosphere? What are the fundamental limits of our current methods based on analog-to-digital conversion implementing the quantized Shannon representation; i.e. uniform discretized samples at the Nyquist rate or better? It sometimes seems we develop very sophisticated and costly receivers and expend tremendous resources to collect and process enormous volumes of high resolution data containing almost no useful information.

A-to-I will enable accurate extraction of useful information from broad band environments crowded with diverse signals and interference spread over a large dynamic range in order to meet the DoD's requirements for RF applications now and in the future.

Program Goals

A-to-I will establish the fundamental bounds of analog-to-digital conversion with sampling at less than Nyquist rates.

A-to-I will establish fundamental bounds on the performance of alternatives to standard analog-to-digital conversion based on multiples of the Nyquist criteria. A-to-I will determine if additional prior knowledge or hypotheses can be used to create novel data conversion approaches that apply system resources more effectively to capture useful information content in the RF environment. This effort will determine the minimum sampling required to characterize and reproduce any signal of interest with no information loss, and determine practical mechanisms to implement such processing. Shannon sampling will be used to establish a baseline. A-to-I will emphasize finding constructive solutions to the optimization problem of compressed sampling. These solutions will be evaluated against the practical criteria of realistic hardware at reasonable power and cost.