An Efficient Direct-Sequence Signal Detector Based on Dempster-Shafer Theory

The detection of direct-sequence spread-spectrum signals is studied. We consider the environment of broadband noise jamming with the assumption that jammer state information is unknown. We show how Dempster-Shafer theory can be applied to this problem; in particular, we treat the unknown jammer state information as uncertainty and use the concept of “confidence discount” and the Dempster’s combining rule to derive a heuristic detector for the problem. We show that the resulting detector has a better performance than the hard-decision detector and the soft-decision detector when the jammer is significant. The degradation relative to the soft-decision detector is smaller than the hard-decision detector when the jammer is low.