Bootstrap modeling of a class of nonstationary signals

The problem of modeling polynomial-phase signals is considered. Techniques based on the bootstrap and multiple hypotheses tests for optimal model selection of constant amplitude polynomial-phase signals embedded in stationary noise are developed. Phase parameter estimators based on both the least-squares method and the polynomial phase transform are used. The proposed techniques are compared with existing ones, including Akaike’s information criterion and Rissanen’s minimum description length criterion and are shown to outperform these procedures for the considered small sample sizes.