New parametric method of detecting random signals in unknown noise. 1. Asymptotic theory

The problem of detecting random signals in unwanted clutter in a radar system is discussed. First, the models of the received data and the neighbouring clutter cells are developed. A new detection scheme is then proposed based on the comparison of the complete information of the second-order statistics estimated from the data under test and those estimated from the neighbouring clutter cells. The maximum likelihood ratio test is developed accordingly. The performance of the new detection scheme is analysed using the asymptotic properties of the likelihood ratio and is shown to be superior to the conventional noncoherent detection method