Broadband Processing for Mine Warfare Sonar

This Phase I proposal presents a unique opportunity to improve active sonar capability for mine detection and classification through the use of robust signal processing of broadband signals. This effort will improve target recognition in environments withstrong reverberation and dense multi-path. The current importance of operating in littoral waters, along with an increased noise background and reverberation conditions, and the high importance of mine avoidance, establishes a high priority for improvedbroadband active sonar performance. Stronger and more complex multipath propagation increases clutter, reduces the probability of detection, impacts target localization, and degrades target classification performance. Often target returns are receivedusing existing signal processing methods, but the target returns are embedded in clutter making target classification difficult. The only way to make a true advance in system performance is not to select a different operating point on a fixed receiveroperator characteristics (ROC) curve, but rather to shift the ROC curve itself. The work proposed here has the potential to improve the target minimum detectable level, to achieve processing gains against clutter, and to provide better estimation ofcontact features by shifting the ROC curve. Completion of this Phase I effort will result in development of technology that provides an improved broadband mine detection and classification capability for the AN/SQQ-32 minehunting sonar system. Withappropriate interface modifications, this same capability could be provided to other U.S. Navy sonar systems and commercial systems that use acoustic energy to explore or characterize the ocean.