OTHR Sensitivity Improvements through Recovery of Signal Processing Losses

Each signal processing stage in the Over-theHorizon Radar (OTHR) processing chain introduces some signal-to-noise ratio (SNR) loss. These could be due to application of windows for sidelobe control or localised noise estimation in Constant False Alarm Rate (CFAR) detection. A number of techniques have been proposed in the past to recover the signal processing losses in some of the processing stages. In this paper, we summarise those methods presented in the past, and also present new techniques for recovering CFAR losses. Additionally we present a new paradigm of machine learning based detection which alleviates conventional CFAR losses. This paper serves as a review of past efforts to minimise OTHR processing losses while also introducing new techniques for reducing the detection losses.