Statistical Performance Analysis of MDL Source Enumeration in Array Processing

摘要

In this correspondence, we focus on the performance analysis of thewidely-used minimum description length (MDL) source enumeration technique inarray processing. Unfortunately, available theoretical analysis exhibitdeviation from the simulation results. We present an accurate and insightfulperformance analysis for the probability of missed detection. We also show thatthe statistical performance of the MDL is approximately the same under bothdeterministic and stochastic signal models. Simulation results show thesuperiority of the proposed analysis over available results.