Performance of Opportunistic Distributed RF Operations in Uncertain and Challenged Network Environments

Abstract

Opportunistic operations are functions carried out by leveraging spare capacity on existing devices. We investigate the performance of implementing distributed RF functions, such as spectrum monitoring, as opportunistic operations on groups of radios and RF sensors. We assume that the primary activities of the devices cannot be fully predicted, and that opportunistic operations on a device may be preempted by its primary functions. Therefore there is uncertainty regarding the ability of RF devices to complete assigned opportunistic operations. Moreover, we assume the communications network is intermittent, lossy and low bandwidth. The network limitations create command and control uncertainty. The primary research challenge addressed in this work is to overcome these sources of uncertainty in order to carry out distributed RF operations with high probability of success, while minimizing resource consumption. The approach we have developed is implemented in a middleware layer called Management of RF Network and Tasking Infrastructure (MARTI). We use an analytic model and experiments with MARTI in a wireless testbed to quantify the impact of various strategies for mitigating uncertainty, the tradeoffs between improved communication reliability and its latency and bandwidth costs, and strategies for trading off between result fidelity and network resource consumption.