On the detection of a signal in colored noise (Q1118563)

A modeling approach is used in the detection of a random signal in colored noise. The received sequence is modeled as a regressive/autoregressive time series, and the presence or absence of the desired signal is determined through a hypothesis testing procedure. The test is based on the construction of an F-statistic using likelihood functions. The statistic can be easily incorporated into the computation of the probability of a false alarm, such as required in the processing of radar signals. Results based on simulated data and actual radar data are presented.