Consider a Gaussian stationary stochastic vector process with the property that designated pairs of components are conditionally independent given the rest of the components. Such processes can be represented on a graph where the components are nodes and the lack of a connecting link between two nodes signifies conditional independence. This leads to a sparsity pattern in the inverse of the matrix-valued spectral density. Such graphical models find applications in speech, bioinformatics, image processing, econometrics and many other fields, where the problem to fit an autoregressive (AR) model to such a process has been considered. In this paper we take this problem one step further, namely to fit an autoregressive moving-average (ARMA) model to the same data. We develop a theoretical framework and an optimization procedure which also spreads further light on previous approaches and results. This procedure is then applied to the identification problem of estimating the ARMA parameters as well as the topology of the graph from statistical data. / <p>Updated from "Preprint" to "Article" QC 20130627</p>

Radar signal processing techniques aim to suppress clutter to enable target detection. Many clutter suppression techniques have been developed to improve the detection performance in literature. Among these methods, the most widely known is MTI plus coherent integrator, which gives sufficient radar performance in various scenarios. However, when the correlation coefficient of clutter is small or the spectral separation between the target and clutter is small, classical approaches to clutter suppression fall short. In this study, we consider the ARMA spectral estimation performance in sea clutter modelled by compound K-distribution through Monte Carlo simulations. The method is applied for varying conditions of clutter spikiness and auto correlation sequences (ACS) depending on the radar operation. The performance of clutter suppression using ARMA spectral estimator, which will be called ARMA-CS in this work, is analyzed under varying ARMA model orders. To compare the clutter suppression of ARMA-CS with that of conventional methods, we use improvement factor (IF) which is the ratio between the output Signal to Interference Ratio (SIR) and input SIR as performance measure. In all cases, the performance of ARMA-CS method is better than conventional clutter suppression methods when the correlation among clutter samples is small or the spectral separation between target and clutter is small.

Statistics / M. Sc. (Statistics)