Army tactical operations frequently occur in congested and contested RF spectral conditions, which can impair a system's ability to transmit information over wireless networks. Whether the source of impairment is due to ordinary congestion in the spectrum, malicious jamming, propagation conditions or natural/man-made interference, identifying the source of the impairment is an important step in addressing the issue in order to take appropriate corrective action. As these impairments may affect limited regions of the spectrum, other portions may be feasible for use. Operators at each individual network node must have situational awareness of the spectrum, so that critical data can be transmitted with a high degree of confidence to other nodes as well as globally to control centers. Traditional spectrum scanners are slow and consume substantial energy, particularly if rapid tracking of fast moving signals is required. Compressive Sampling (CS) coupled with Machine Learning (ML) can create a power efficient system capable of scanning large bandwidths in a short amount of time while providing sufficient data to determine the underlying impairment. In this proposal, Aspen Consulting Group proposes to integrate a CS receiver model and a ML algorithm into a complete end-to-end tactical radio simulation environment to demonstrate adaptable communications.