Um método empírico para estimação do número de fontes de sinais incidentes em um arranjo de sensores

Abstract

This dissertation proposes an empirical method for estimating the number of sources of signals impinging on multiple sensors. The method is based on the analysis of the norms of vectors whose elements are the normalized eigenvalues of the received signal covariance matrix and the corresponding normalized indexes. It is shown that such norms can be used to classify the eigenvalues in two groups: the largest and the remaining ones, thus allowing for the estimation of the number of sources without the knowledge of any additional parameter. The norm-based method, as it is called here, produces consistent estimates as the number of samples increases, and in some situations has a higher probability of correct estimation when compared to a new and effective method based on the random matrix theory. Is also addressed in this study methods based on information theory which is the Akaike information criterion (AIC), the method of minimum description length (MDL) and the Bayesian information criterion (BIC). Some fundamental concepts are also provided, in order to facilitate the understanding of the above methods.