We propose a simple algorithm for improving the MDL (minimum description length) estimator of the number of sources of signals impinging on multiple sensors. The algorithm is based on the norms of ...vectors whose elements are the normalized and nonlinearly scaled eigenvalues of the received signal covariance matrix and the corresponding normalized indexes. Such norms are used to discriminate the largest eigenvalues from the remaining ones, thus allowing for the estimation of the number of sources. The MDL estimate is used as the input data of the algorithm. Numerical results unveil that the so-called norm-based improved MDL (iMDL) algorithm can achieve performances that are better than those achieved by the MDL estimator alone. Comparisons are also made with the well-known AIC (Akaike information criterion) estimator and with a recently-proposed estimator based on the random matrix theory (RMT). It is shown that our algorithm can also outperform the AIC and the RMT-based estimator in some situations.
In this paper we propose 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. It is shown that, in some situations, our norm-based method produces satisfactory performance when compared to a recently proposed random matrix theory method
Blind source separation and estimation of the number of sources usually demand that the number of sensors should be greater than or equal to that of the sources, which, however, is very difficult to ...satisfy for the complex systems. A new estimating method based on power spectral density (PSD) is presented. When the relation between the number of sensors and that of sources is unknown, the PSD matrix is first obtained by the ratio of PSD of the observation signals, and then the number of source signals can be estimated by clustering the column vectors of PSD matrix. The effectiveness of the proposed method is verified by theoretical analysis and experiments.